THE 
MILITARY SURGEON 
Vol. 99 
NOVEMBER, 1946 
No. 5 
ORIGINAL ARTICLES 
Authors alone are responsible for opinions expressed in their contributions 
FOREWORD 
Colonel James E. Ash, Medical Corps, U. S. Army, has just retired after thirty years of 
distinguished service. He has made signal contributions to the Medical Corps, to the Army, 
and to the Nation through his guidance of the development and organization of the Army 
Institute of Pathology. Recognized as the leading pathologist in the Army, his diligence, fore- 
sight, professional knowledge, and undeterred devotion to duty have led to the organization 
of the most extensive service in tissue pathology ever known to the world. Under his leadership 
the American Registry of Pathology has expanded greatly, and under his supervision the Army 
Medical Illustration Service was established. The wealth of pathologic and illustrative material 
brought together at the Army Institute of Pathology has given it a unique position for the 
study of diseases in the military age group, and for the training of pathologists. 
The editors of The Military Surgeon have issued this specialjiupiber in honor of Colonel 
Ash. All the contribotions are by members of the Army Institute of Pathology. The pap£f? give 
a cross-section of the productive investigation carried on at the Institute even under the adverse 
conditions of heavy war-time routine. They are a tribute of a loyal and devoted staff to their 
Chief. 
The services of the Army Institute of Pathology under the Directorship of Colonel Ash 
have been of inestimable value to the Medical Department and to the Nation. Colonel Ash 
richly merits the honor bestowed on him by the publication of this volume. 
Norman T. Kirk, 
Major General 
The Surgeon General 
361 Col. James E. Ash, M.C., U. S. Army 
Director, Army Institute of Pathology COLONEL JAMES EARLE ASH 
AN APPRECIATION 
THE men and women of the Army In- 
stitute of Pathology in the exercise of 
their various talents have contributed 
this special issue of The Military Surgeon 
as a means of honoring and of showing their 
affectionate regard for the officer who con- 
ducted the affairs of the Institute during the 
ears of the lately terminated war. 
Colonel Ash who has recently completed 
is thirtieth year in the medical service of the 
Vrmy, has devoted the greater number of 
hese years to laboratory work and studies. 
The University of Pennsylvania in his native 
city gave him his jnedical education, and its 
courses were followed and supplemented by 
several years of interne service in Phila- 
ielphia’s hospitals, including the great Block- 
’ey institution. By the end of this service he 
had fixed upon a career of laboratory medicine 
with his main interest in pathology. 
From his interne service he went to the 
laboratory of the Norristown Hospital for the 
Insane, a State institution. Three years of 
service here and he moved on to the Harvard 
Medical School to work with Professor How- 
ard Thomas Karsner, with whom he had 
been associated in medical school and in his 
hospital work in Philadelphia. It was with 
this background of work in pathology that he 
entered the Medical Corps of the Army in 
1916, with war on in Europe and our military 
forces massed along the Mexican border. 
Competent pathologists have always been far 
too few in the Army to employ them on other 
than their very special work. It follows then 
that the list of Colonel Ash’s stations is not a 
long one, but includes nearly all of the per- 
manent General Hospitals that existed prior 
to our second World War, including the 
Sternberg hospital in Manila. If we add to 
these the Army Medical School and the Army 
Medical Museum the list is about complete. 
He has had two details to the Museum, the 
present one still uncompleted, having covered 
nearly ten years. 
Colonel Ash’s tenure of office in the 
Museum has been marked by two major ac- 
complishments. The American Registry of 
Pathology, which started its first registry in 
1922 was expanded from five to fourteen by 
Colonel Ash’s cooperation with the national 
medical societies. The registries have the 
records of 43,000 cases involving unusual 
pathological material. 
The essential function of the Medical 
Museum, that of pathological diagnosis and 
research, was long obscured by the name 
borne by the institution. Through recom- 
mendations made by Colonel Ash, the estab- 
lishment is now the Army Institute of 
Pathology, which places the emphasis upon the 
vital work and the museum feature in its 
proper secondary place. Needless to say 
Colonel Ash is a master craftsman with all 
the tools and appliances of the pathologist and 
a highly competent judge of the material that 
comes to his hands or that falls under the lens 
of his microscope. 
The widely expanded Institute has been 
manned during the war by a staff of diligent 
workers, whose duties have been made the 
pleasanter by the understanding attitude of its 
Director. Colonel Ash’s quiet supervision is of 
the quality that obtains a loyal following from 
those working with him. 
It would be expected that a quiet, some- 
what introverted person like Colonel Ash 
should have a liking for music, which can be 
muted when desired and which requires little 
or no company for its enjoyment. He is indeed 
devoted to piano-music, in which he has a 
talent and always a refuge from boredom. 
The burdens of his exacting position have 
been aggravated for him by an accident to one 
of his eyes, which has required much treat- 
ment during the past year. As this is written, 
orders for his retirement from active service 
are awaited. 
There is expectation that Colonel Ash, after 
retirement, will take over the direction of the 
American Registry of Pathology. His many 
friends will welcome his continued association 
with the Institute and wish him a world of 
contentment in his new work. 
J. M. P. THE ARMY INSTITUTE OF PATHOLOGY DURING 
WORLD WAR II 
COLONEL BALDUIN LUCKS, Medical Corp 
Foundation. The first World 
War emphasized the need for a central 
laboratory of pathology to serve the 
entire United States Army. During the years 
between the wars such a laboratory was 
organized as a department of the Army Medi- 
cal Museum, which itself had been founded 
during the War between the States. In the 
course of years, and particularly with the ex- 
pansion of the Army during the period of 
mobilization and war, the department of 
pathology assumed much greater medicomili- 
tary importance than the parent organization. 
The enlarged scope of its activities and the 
frequent and serious misunderstandings of its 
function because of the name, Army Medical 
Museum, led the Surgeon General to approve, 
in 1943, a new designation, “The Army Insti- 
tute of Pathology”; the title of the command- 
ing officer was changed from Curator to 
Director. Thus, both the Army Institute of 
Pathology and the Army Medical Museum 
trace their origins to times of national 
emergency. 
Organization and Function. The Army 
Institute of Pathology operates under the 
direct control of the Surgeon General, in con- 
formity with Army Regulations (AR 
40-410). It comprises four departments 
which are administratively co-ordinated: The 
Department of Pathology, The Army Medi- 
cal Illustration Service, The American Regis- 
try of Pathology, and the Army Medical 
Museum. The principal functions of the Insti- 
tute may be summarized as follows: (1) It is 
the central laboratory of pathology for the 
United States Army. (2) It furnishes a con- 
sultation service for the diagnosis of pathologic 
tissue, and it acts as the chief reviewing 
authority on the diagnosis of pathologic ma- 
terial for all Army installations. (3) It pro- 
vides instruction in pathologic anatomy to 
medical, dental and veterinary officers. (4) It 
conducts research in the pathology of diseases 
which have medicomilitary importance. (5) 
It is the headquarters of the Army Illustration 
Service. (6) It houses and acts as custodian 
of the American Registry of Pathology, and 
(7) It maintains the Army Medical Museum 
for the instruction of medical department per- 
sonnel and the education of the public. 
This enumeration does not exhaust the list 
of activities but it suffices to indicate the broad 
plan on which the Army Institute of Pathology 
was developed. 
A brief discussion may now be given of the 
aims and the work of the several departments, 
excepting the American Registry of Pathology 
an account of which may be found elsewhere 
in this issue. 
Defartment of Pathology. Three essential 
functions, diagnosis, teaching, and research, 
are performed by the Department of Patholo- 
gy. Here are referred for diagnosis or review 
of diagnosis, and for permanent file, the per- 
tinent records and the tissues from all im- 
portant surgical operations (such as tumors) 
and from all postmortem examinations on 
military personnel of the entire Army. Some 
of this material is sent directly to the Institute, 
but most of it first passes through other labora- 
tories known as “histopathologic centers.” As 
organized for wartime, one or more of these 
centers were located in each of the nine service 
commands of the Zone of the Interior, and 
others throughout the various Theaters of 
War. Similar centers an d the laboratories of a 
number of general hospitals function on this 
plan, which proved so efficient in the active 
phase of war. They make immediate examina- 
tions, “screen out” relatively unimportant ma- 
terial, and forward to the Institute such data 
and material as may have either scientific or 
administrative significance, from surgical cases 
and all autopsies. Here the material is re- 
examined, and the submitting station is noti- 
fied of the opinion of the staff pathologists. 
Air mail and radiograms enable military in- 
stallations, no matter where they are located, 
to make immediate use of the facilities of the The Army Institute of Pathology During World War II 
365 
institute. This service protects military person- 
nel as well as the Government against possible 
errors in diagnosis, and aids surgeons, internists 
and other medical officers in their work by 
facilitating the diagnosis, treatment, and pre- 
vention of disease. Moreover the centralization 
in one laboratory makes it possible for the 
Office of the Surgeon General to obtain 
promptly accurate information on the causes 
of death, the diagnosis of operative cases, and 
the regional trend of disease. 
The volume of material received at the 
Institute is considerable. During 1945 the 
records and tissues from 18,895 postmortem 
examinations and from 20,539 selected surgi- 
cal operations were submitted, the latter pre- 
senting mainly diagnostic problems. The 
handling of such an amount is an administra- 
tive and technical problem. Each “case” is 
given an accession number under which is 
filed all pertinent material including written 
records, specimens, blocks, slides and photo- 
graphs. Many tissues are prepared for micro- 
scopic examination by special methods. All 
records are analyzed; all diagnoses cross- 
indexed for ready availability. 
The steady growth of the work of the Insti- 
tute from the beginning of the first, to the end 
of the second, World War is shown in 
Figure 1. The polygons represent the cumu- 
lative increase in the number of accessions. The 
curve in the insert shows the number of 
“cases” studied during World War II; the 
dotted line at the terminal of the curve, the 
probable number of accessions for the next two 
years. This estimate is based on the expectation 
that the decrease in the size of the Army will 
be more than balanced by the increase in the 
Veterans’ Administration. 
The second function of the Department of 
Pathology, teaching, was carried on during the 
War along two lines: Medical officers from 
key laboratories were placed on temporary 
duty at the Institute to gain first hand informa- 
tion concerning its facilities. A valuable part 
of the training was attendance at the daily staff 
conference, where important current cases 
were reviewed by members of the staff. Micro- 
scopic slides were projected and the gross 
specimens and x-ray plates were shown when 
pertinent. The presentations were brief but to 
the point, and the discussion, in which every- 
body participated, free and lively. 
Approximately 150 officers received such 
training; the crowded conditions at the Insti- 
tute and the world wide distribution of the 
Army limited the number who could be as- 
signed. Hence extramural instruction was 
organized: study sets exemplifying the lesions 
of important diseases, supplemented by lantern 
slides, clinical records and the results of 
pathologic studies were sent to Army labora- 
tories. Illustrative syllabuses and comprehen- 
sive atlases covering special fields, with appro- 
priate slides were made available on loan to 
keep laboratory officers abreast of new de- 
velopments. During 1945 a total of 1669 
study sets were distributed, approximately 90 
per cent of them to Army installations. Nor 
was the civilian profession neglected; all study 
material that could be spared was loaned. Be- 
cause tropical medicine, a subject much neg- 
lected in time of peace, assumed great im- 
portance during the War, tissues from repre- 
sentative cases, lantern slides, and microscopic 
preparations were furnished to medical schools 
of this country and Canada. 
The third main function, research, was 
vigorously conducted throughout the war. The 
wealth of material flowing into the Institute 
afforded a unique opportunity to gain greatly 
needed information on the fathology of the 
soldier, i.e. on disease processes prevalent in 
the military age-group, 18 to 38 years. The 
results of all definitive studies which were of 
interest to both the Medical Department and 
the profession in general were published 
promptly. The number of publications which 
have already emanated from the Institute and 
their high standard are worthy of note. Among 
the subjects investigated were; coronary 
sclerosis in young soldiers, the pathology of 
fatal epidemic hepatitis, the pathology of tropi- 
cal diseases, infection of the central nervous 
system and meninges after tooth extraction, 
pathology of trench foot, comparative path- 
ology of scrub typhus and other rickettsial dis- 
ease, peripheral nerve injuries, fat embolism, GROWTH OF ARMY INSTITUTE OF PATHOLOGY 
1917- 1946 
CUMULATIVE INCREASE IN CASE ACCESSIONS 
TOTAL NUMBER OF ACCESSIONS AT EACH YEAR-END 
Fig. i 
THOUSANDS OF ACCESSIONS The Army Institute of Pathology During World War II 
367 
gynecomastia, teratoma of the anterior 
mediastinum, and odontogenic tumors. The 
papers in this issue of the Military Surgeon 
are other examples of studies which were 
started during the War. These studies and 
others will form the basis for the Volume on 
Pathology in the Medical History of World 
War II. 
Staff. The Institute was fortunate through- 
out the War to have as its commanding officer 
a man of high professional attainments, who 
was liked and trusted by everyone, who in- 
spired loyal devotion to duty, and who, with- 
out seeming effort, was able to weld the staff 
into a harmonious working unit: Colonel J. E. 
Ash. Perhaps no man knows the Institute, its 
early trials and its ultimate success as does he, 
for he served there from March 1929 to 
September 1931, and again from February 
1937 until his recent retirement. Few medical 
officers have as extensive an acquaintance or 
have kept in as close touch with civilian pa- 
thologists throughout the country. The Insti- 
tute owes much to these attributes. 
Soon after the beginning of hostilities 
Colonel Ash brought together a professional 
staff which was selected on the basis of 
specialized knowledge and competence in the 
various branches of pathology. Thus it was 
possible for the group to render expert opinion 
on difficult or controversial diagnostic problems 
and to conduct essential teaching and research. 
The staff was never large; it averaged fewer 
than fourteen pathologists, but these were 
hand-picked and they remained on duty for 
the major part of the war. Eight were sta- 
tioned at the Institute for more than three 
years, four for more than two. 
An important aid to these officers were the 
Resident Consultants. A total of seventeen 
distinguished civilian pathologists were invited 
to act as Consultants to the Staff, and to 
undertake, or to collaborate in, certain investi- 
gations. Most of them spent approximately one 
month in residence, but several remained for 
as long as three months or returned for addi- 
tional terms of duty. The Consultants were a 
great stimulus to the staff. Their interest did 
not flag when they returned to their own 
laboratories; most of them continued to pursue 
investigations begun at the Institute. Twelve 
of the consultants have made, or have in 
progress, valuable contributions to the research 
program, two examples of which are printed in 
this volume. 
Not the least service was rendered by the 
detachment of approximately thirty enlisted 
men and WACS, and about twice that number 
of civil service personnel. Without their tech- 
nical and clerical help the Institute would not 
have been able to function. 
The Army Medical Illustration Service. 
The Photographic Laboratory, established over 
eighty years ago, and the Museum and Medical 
Arts Service, organized during the War, make 
up the Army Medical Illustration Service. 
The former has long been distinguished for 
the excellence of its photomicrography. Its 
wartime tasks were heavy. For example, dur- 
ing 1945, the photographic laboratory made, 
in round numbers, 10,000 negatives, over 
50,000 prints and enlargements, over 10,000 
lantern slides in color and 5,000 more in 
black and white, over 30,000 photostats, and 
over one million offset prints. 
The Museum and Medical Arts Service 
(recently reorganized under the name of 
Medical Mobile Photographic Units) had its 
headquarters at the Institute and nine units 
in overseas theaters of operation. The units 
consisted of one officer and seven enlisted men, 
all of whom were well trained medical artists 
or photographers. Through their work the 
medical department was furnished with mov- 
ing and still pictures of nearly every phase of 
medical activity in combat areas; these illustra- 
tions were not only invaluable for training 
purposes, but they constitute a record of lasting 
historical value. 
This organization was also charged with 
the supervision of medical photography in the 
Zone of the Interior. To it were submitted all 
photographs and moving pictures taken in 
Army hospitals in this country. During 1945 
it received for review and file over 75,000 
photographs and nearly 100,000 feet of mov- 
ing picture film. 
Another contribution of the MAMAs, as 368 
The Military Surgeon—November, 1946 
they were called, was the production of plastic 
models representing different kinds of wounds 
and of sets of posters illustrating first-aid pro- 
cedure. Many thousands of such models and 
posters were distributed to Army installations 
for purposes of instruction. Both the Photo- 
graphic Laboratory and the Museum and 
Medical Arts Service were headed by able men 
of vision and devotion to the Army. They 
made the Institute the headquarters of medical 
arts for the Medical Department. 
Army Medical Museum. The exigencies of 
war made it necessary to store large portions 
of the collections, so that the space could be 
used for laboratories and offices. The main 
hall, however, was kept open, for even with 
the restrictions in travel, the Museum drew as 
many as 200,000 vistors annually, indicating 
the interest of the public in medical exhibits. 
Future. This brief sketch would be incom- 
plete without mentioning plans for the future. 
The Army Institute of Pathology is unique; 
it is the only institution of its kind in the world. 
It has amply proved its worth to the Army 
and to the Nation. The present quarters of the 
Institute are unsuitable and inadequate. Recog- 
nizing these facts The Surgeon General im- 
mediately after cessation of hostilities requested 
Colonel Ash and the Committee on Pathology 
of the National Research Council to submit 
recommendations “for the most advantageous 
use of the facilities of the Army Institute of 
Pathology, both for the Army Medical De- 
partment and for the medical profession as a 
whole.” Meetings of this Committee with 
Colonel Ash and his staff led to plans for a 
new Institute which have been approved by 
The Surgeon General and the War Depart- 
ment. Thus we may look forward to a new 
Army Institute of Pathology, which, while 
retaining its special traditions, will be enabled 
to render even greater service in a wider field. 
THE AMERICAN REGISTRY OF PATHOLOGY AND ITS RELA- 
TION TO THE ARMY INSTITUTE OF PATHOLOGY 
By HOWARD T. KARSNER, M.D., Resident Consultant, Army Institute of Pathology 
COLLECTION for the sake of mere 
possession is as futile in medicine as in 
any other field. In cultural spheres, 
the art museum serves as a source of education 
to the public, of stimulus to artists, and of re- 
search in history both special and general, and 
the library has much the same purposes. In 
science, however, although the objectives are 
the same, the emphasis often differs in that 
research occupies the foremost position, even 
though education and stimulation are promi- 
nent aims. 
A museum of pathology may be principally 
educational, but it is not fully utilized unless 
the opportunity for research is seized, A regis- 
try of pathology or of any of its special branches 
differs from a museum in several respects. In 
the registry, display of specimens is not a 
principal function; study of the material is 
both the primary and principal objective. 
Whereas records of the source of museum 
specimens is important, in the registry they are 
essential. In the museum, “follow-up” adds 
greatly to the value of the material, but in the 
registry the purpose is defeated if “follow-up” 
is not pursued assiduously. The museum may 
be conducted by a curator not necessarily an 
expert in the field covered, but the registry 
requires highly trained, experienced, com- 
petent scientific personnel at its head and in 
many parts of its organiaztion. A museum 
may be collected locally, but a registry must 
have contributions from widely dispersed 
sources. 
A registry is usually established because of 
the direct interest of a comparatively small 
group of inspired guiding spirits. The location 
of the registry may be determined by the 
institutional situation of an individual or small 
group. However, to be most effective, the The Army Registry of Pathology 
369 
registry must be at least national in scope and 
often, if it is placed in a local institution, 
various factors may operate so that collection 
may not be adequate. As a consequence of 
these elements, the Army Institute of Pa- 
thology, situated in the capital city of this 
country, has been especially favored. This In- 
stitute is unique in the world. Nowhere else 
has there ever been a concentration of patho- 
logical specimens that is comparable. No- 
where else is the pathology of the entire Army 
of a great country so concentrated. Nowhere 
else have the civilian pathologists and other 
interested physicians taken such a great part 
in organization and operation. Nowhere else 
has there been, as continues to be true, such a 
close scientific liaison between medical officer 
and civilian doctor. 
It is only natural that when the American 
Academy of Ophthalmology and Otolaryn- 
gology decided to form a registry it should 
turn to the Army Medical Museum. The 
Curator, George R. Callender, now Brigadier 
General, M.C., opened the Registry of 
Ophthalmic Pathology in 1922. At the end 
of 1945 there was a total of 19,385 accessions 
in this registry alone, a material which has led 
to numerous important studies, especially of 
melanotic tumors of the eye. In 1925, a 
Registry of Tumors of the Lymphatic System 
was established at the behest of the American 
Association of Pathologists and Bacteriologists 
and now has 2,638 accessions. In 1927, the 
American Urological Association sponsored a 
Registry of Urologic Pathology, which has a 
total of 5,940 accessions of which over 5,000 
refer to tumors of the bladder. Beginning in 
1933, and continuing through 1945, II other 
registries have been formed, including Dental 
and Oral Pathology, Otolaryngological Pa- 
thology, General Tumors, Dermal Pathology, 
Renal Tumors, Thoracic Tumors, Neuro- 
pathology, Orthopedic Pathology, Prostatic 
Tumors, Veterinary Pathology, and Geron- 
tology, all sponsored by special national so- 
cieties. Naturally, the collected material varies 
in amount with the registry but the total num- 
ber of accessions has reached 43,442. 
A movement once started may grow of its 
own momentum. The history of the registries, 
however, shows that in the first half of the 
period between 1922 and 1946, 4 registries 
were started and that in the latter half, 10 
were organized, with more to come in the 
near future. The impetus given by Gen. Cal- 
lender was undoubtedly of great significance. 
The subsequent growth, however, was due 
almost entirely to the industry and interest of 
James Earle Ash, now Colonel, M.C. Here 
was a man trained at Pennsylvania, Harvard 
and Vienna, one who had early decided that 
his career was to be devoted to pathology, one 
who throughout his Army life gave selfless 
effort to the development of Army laboratories. 
There could be no question as to his com- 
petence in his field. With this exceptional 
background of general pathology, he could 
turn his attention to special fields with every 
assurance that his work would be broadly based 
and expert. Through his devotion to the regis- 
tries he has brought to the special societies a 
wealth of analysis and information which can- 
not be equalled. Civilian medicine has profited 
enormously by the skillful study of specimens 
and cases. The material was also of great value 
in the training of Medical Officers, who dur- 
ing World War II, conducted laboratories in 
the widely dispersed theaters of operation. 
Colonel Ash’s modesty of demeanor could not 
hide his merit, his sympathy, his competence, 
or his deep interest in the projects. These at- 
tributes, coupled with his fine character, 
straightforwardness and integrity had much 
to do with the confidence reposed in him by 
the members of the sponsoring societies. Medi- 
cine and the military profession as a whole 
owe an eternal debt of gratitude to Colonel 
James Earle Ash as physician, pathologist, 
medical officer, Curator of the Army Medical 
Museum and Director of the Army Institute 
of Pathology, for many reasons, not the least 
of which is his consecration to the origin, de- 
velopment and operation of the American 
Registry of Pathology. THE REGISTRY OF DENTAL AND ORAL PATHOLOGY 
AT THE ARMY INSTITUTE OF PATHOLOGY 
IN 1933, when the American Dental As- 
sociation authorized the establishment of 
the Registry of Dental and Oral Pa- 
thology at the Army Institute of Pathology, 
the first attempt was made by the dental pro- 
fession toward establishing a central laboratory 
of oral pathology. Before this time there was 
but little specialization in dental pathology, and 
the contact between dental pathologists and 
the medical profession in general was slight. 
In an effort to better this situation, the Amer- 
ican Dental Association instituted the Registry 
program, which was designed primarily to 
demonstrate the importance of dental and 
oral pathology and to further the relationship 
between medicine and dentistry. It was essen- 
tial to the success of the Registry that it have 
the support of the practitioners of dentistry, the 
encouragement of the Director of the Insti- 
tute, and the diligent and active efforts of the 
Registry Committee of the Association. For- 
tunately all of thees requirements were met as 
the program developed. Progress was slow but 
continuous, the guiding spirit during the criti- 
cal periods being Colonel James E. Ash,, the 
Director of the Army Institute of Pathology 
from 1929 to 1931 and 1937 to 1946. Be- 
cause of his constant support the Registry was 
able to expand to a point where its importance 
to the progress of dentistry became clearly 
apparent to the profession at large. 
Some of the more important facilities, such 
as the biopsy service, the loan sets and the atlas 
were produced at the suggestion of Colonel 
Ash. His contributions have been continuous, 
not only as plans but also as the work that 
made them a reality. The guidance of one so 
outstanding in the field of pathology has been 
essential in bringing the Registry of Dental 
and Oral Patholoy to the recognized position 
it holds today. Its files now contain over 6,000 
selected cases from which teaching material 
has been prepared for dental schools, study 
clubs, and practitioners. The value of this 
Registry should not only be measured in terms 
of services rendered and material collected, but 
particularly in the changed point of view of 
the practicing dentist toward pathology as an 
adjunct in improved care of his patient. 
We who have been interested in the growth 
of the Registry are convinced that it would not 
occupy its present position except for the sup- 
port it received from the medical profession 
and the Army Medical Department. It was 
through the efforts of Colonel Ash that this 
support was forthcoming, particularly during 
the critical period when the Registry was 
struggling to attain the status enjoyed by 
other more established registries. His diligent 
efforts to procure adequately trained personnel, 
his insistence upon equal representation for the 
dental profession, and his foresight in general 
has insured the attainment of the desired goal. 
The dental profession owes a debt of grati- 
tude to Colonel Ash for his efforts in its be- 
half. Only those who have been actively asso- 
ciated with the Registry during its period of 
growth can realize how much has depended 
on this one individual, how many significant 
advances have been made at his suggestion and 
through his intercession. The most recent of 
them have been the establishment of a Fellow- 
ship in Dental and Oral Pathology at the 
Army Institute of Pathology and the organiza- 
tion of the American Academy of Oral 
Pathology. 
Even though Colonel Ash has come to the 
time when he will relinquish the directorship 
of the Institute, because of retirement from 
the Regular Army, it is the hope of every 
member of the American Dental Association 
that he will remain in close touch with the 
organization which owes so much to his 
enthusiasm and leadership. 
Henry A. Swanson, 
Chairman, Committee on National 
Museum and Dental Registry 
American Dental Association 
Joseph L. Bernier, 
Secretary LOWER NEPHRON NEPHROSIS* 
(THE RENAL LESIONS OF THE CRUSH SYNDROME, OF BURNS, TRANSFUSIONS, AND OTHER 
CONDITIONS AFFECTING THE LOWER SEGMENTS OF THE NEPHRONS) 
By COLONEL BALDUIN LUCKE, Medical Corfs 
(From the Army Institute of Pathology, Washington 25, D.C.) 
(With sixteen illustrations) 
Introduction 
DURING the early stages of the war, 
British pathologists,13"17’30 drew at- 
tention to distinctive lesions of the 
kidney found in fatal cases of the crush 
syndrome, a condition characterized by renal 
insufficiency following crushing injury to 
muscle. The syndrome was not new; it had 
been described during the first World War, 
but in the interim had been all but forgotten.63 
Its rediscovery led to intensive study, and soon 
it became clear that the renal changes are not 
specific for the crush syndrome. Similar or 
identical lesions have been observed in a 
variety of conditions: after any severe trauma 
to muscle,9’14’27’45’71 nontraumatic muscular 
ischemia,12’18’43 burns,46’53 transfusion with in- 
compatible blood,3’25’26’28’40 and heat stroke;58 
also in blackwater fever,40,56 toxemias of preg- 
nancy and uteroplacental damage,37’72’90 alka- 
losis,55 sulfonamide intoxication,41’42’52 and 
poisoning with certain vegetable and chemical 
agents.12’40’48 These conditions all have 
features in common. In each the initiating 
factor consists in destruction of tissue or of 
blood. Clinically, shock is usually an early 
manifestation, followed by renal insufficiency 
which develops according to a definite pattern. 
In all, the renal lesion is essentially the same: 
degeneration, often necrosis, limited to the 
distal segments of the tubules, with brown 
casts of some heme compound in the lower 
nephrons and the collecting tubules. 
We may, therefore, regard the kidney 
lesion in the crush syndrome, in transfusion 
nephrosis, in burns and the other conditions 
mentioned as unique and regularly attended by 
characteristic functional disturbances. There- 
* This communication is based on lectures de- 
livered at the Army Institute of Pathology. 
fore, it is appropriate to designate all cases 
exhibiting these renal disturbances, no matter 
what their etiologic background, by a single 
term. Because of the prominence of the heme 
casts and the primarily degenerative nature of 
the lesions, the collective term “hemoglobinuria 
nephrosis” has been widely used. Even though 
the heme casts are a prominent feature, the 
location of the lesions is so characteristic and 
unique, that the term “lower nephron 
nephrosis” seems more descriptive and has 
been adopted. 
Lower nephron nephrosis has been the most 
frequent form of fatal kidney disorder encoun- 
tered among military personnel during the 
war. For example, Mallory59 observed it in 
io per cent of the first thousand autopsies re- 
viewed by him in the Mediterranean Theater; 
among 427 battle casualties dying in hospital, 
the incidence was 18.6 per cent. Similarly, 
Angevine and Harman2 found these renal 
lesions in 15.2 per cent of 1065 cases in which 
death resulted from battle wounds. 
Material 
This investigation is based upon 538 fatal 
cases, the records and material of which were 
received at the Army Institute of Pathology 
during the war (179 additional cases were 
excluded because pertinent data were incom- 
plete). 
The conditions in which the characteristic 
renal lesion was observed are given in Table I. 
Eleven groups are represented, namely: battle 
wounds, crushing injuries, abdominal opera- 
tions, burns, blood-transfusion reaction, sul- 
fonamide intoxication, heat stroke, falciparum 
malaria (black-water fever), poisoning due to 
a variety of agents, hemolytic anemia, and a 
miscellaneous group containing such unrelated 
conditions as uteroplacental damage, acute 372 
The Military Surgeon—November, 1946 
Table I 
Etiologic Factors in 538 Fatal Cases Having the Characteristic Renal 
Lesions of Lower Nephron Nephrosis 
(Under Most of the Groups Listed Are Given the Number of Cases which Received Transfusions of Blood, 
Sulfonamides, or Both, as Therapeutic Measures) 
Battle Wounds (Gunshot, mine explosion, blast injury, severance of large blood vessel, etc.) 221 
Transfusion and sulfonamide treatment 126 
Transfusion 79 
Sulfonamide 1 x 
No transfusion or sulfonamide 5 
Crushing Injuries 4 6 
Transfusion and sulfonamide treatment 21 
Transfusion 16 
Sulfonamide 4 
No transfusion or sulfonamide 5 
Abdominal Operations (Carcinoma of colon, stomach, pancreas, etc. ruptured ulcer of duodenum or 
stomach; ruptured appendix, etc.) 36 
Transfusion and sulfonamide treatment 18 
Transfusion 11 
Sulfonamide 3 
No transfusion or sulfonamide 4 
Burns 48 
Transfusion and sulfonamide treatment 19 
Transfusion 12 
Sulfonamide o 
No transfusion or sulfonamide 17 
Blood Transfusion Reaction (In cases of trauma, poisoning, infections, etc.) 45 
Sulfonamide Intoxication (In cases of meningitis, pneumonia, other nontraumatic infections, infec- 
tions associated with trauma, etc.) 47 
Heat Stroke 19 
Transfusion and sulfonamide treatment . . . 1 
Transfusion 10 
No transfusion 8 
Malaria (Falciparum); Blackwater Fever 14 
Transfusion and sulfonamide o 
Transfusion 6 
Sulfonamide 2 
No transfusion or sulfonamide 6 
Poisons (Arsenicals, 3; carbon tetrachloride, 3; alkali, 2; carbon monoxide, 15 alcohol (adulterated), 
2; isopropyl alcohol, 1; phenol, 4; photodeveloper, 2; mussel, 1; mushroom, 1) 20 
Hemolytic Anemia (Etiology undetermined) 4 
Miscellaneous (Uteroplacental damage, eclampsia, acute pancreatitis, “shock” from various causes, 
etc.) 38 
pancreatitis, and rickettsial disease. The sig- 
nificance of these conditions as the sole or 
major etiologic factor in the production of the 
renal lesions may be questioned; the table 
shows that in the majority of cases in most of 
the groups, blood-transfusion, sulfonamides, or 
both were employed as therapeutic measures. 
Since transfusion with mismatched blood and 
sulfonamide intoxication may themselves in- 
duce the lesions under discussion, it is not 
possible to decide in every instance if the 
treatment was not in itself a factor in the 
production of the kidney changes. An effort 
has been made, however, to separate all cases 
with frank transfusion reaction or with evi- 
dence of sulfonamide intoxication; in these, 
the treatment has been considered of dominant 
etiologic significance. 
Assuming that in every case the condition 
listed had a major causal relation to the fatal 
syndrome, it is seen that in more than one-half 
(57 per cent) the origin was traumatic: a Lower Nephron Nephrosis 
373 
battle wound, a crushing injury, or a surgical 
operation. In the majority of the remaining 
cases there was probably intravascular he- 
molysis. Occasionally trauma was relatively 
slight or massive hemolysis was not evident. 
In such cases severe and prolonged shock 
seemed to be the outstanding etiologic factor. 
Clinical Manifestations 
The salient feature of lower nephron 
nephrosis, common to all the conditions cited, 
is rapidly progressing renal insufficiency. A 
number of investigators have published excel- 
lent clinical descriptions of the syndrome as it 
develops: after crushing injury,9’13-15 after 
non-crushing injury,27’45’56’71 in nontraumatic 
forms of myoglobinuria,12’18’43’64 after 
burns,46’53 after incompatible tranfusions of 
blood,26 in blackwater fever,40’56’86 in favism,40 
in sulfonamide intoxication,52 in heat stroke,58 
and after uteroplacental damage.72’90 This list 
of references is fragmentary but it may serve 
as a guide to further reading. In all these 
conditions the clinical course and the labora- 
tory findings are essentially similar. Basing our 
observations on the fatal cases studied at the 
Army Institute of Pathology, we shall con- 
sider, in order: certain complications com- 
monly associated with the precipitating condi- 
tion; the urinary manifestations; the dis- 
turbances of the chemical composition of the 
blood; the development of hypertension; of 
edema; and of uremia; and finally, the dura- 
tion of the disease. To illustrate the general 
statements, clinical data and laboratory find- 
ings will be given for 14 representative cases. 
Complications Associated with the 
Precipitating Condition 
Two complications are commonly associated 
with the conditions that lead to the develop- 
ment of lower nephron nephrosis, namely 
shock and vomiting. Shock of some degree 
nearly always follows soon upon the initial 
insult, be it traumatic or nontraumatic. Thus, 
after crushing injuries, battle wounds, other 
forms of extensive trauma or burns, a state of 
shock generally ensues within a few hours. 
Similarly, in blackwater fever, after incom- 
patible transfusions of blood, or after other 
events precipitating intravascular hemolysis, 
shock usually develops promptly. Frequently, 
the signs of shock are unmistakable and com- 
prise significant lowering of the blood pressure, 
coldness of the extremities, pallor or an ashen 
appearance. Many cases, however, have no 
such clear-cut manifestations, but appropriate 
tests may reveal hemoconcentration as an in- 
dication of loss of plasma and of a deficit in the 
volume of the circulating blood.14 In other, 
cases shock cannot be demonstrated by clinical 
criteria, as for example in certain instances of 
sulfonamide intoxication. Furthermore there 
is no agreement as to what the syndrome 
“shock” includes;10’66’88 and, since not all 
cases of lower nephron nephrosis were ade- 
quately observed from the very onset of the 
symptoms, there is no certainty as to whether 
or not shock always occurred. At present, we 
must conclude that in the majority of cases 
shock is a common early complication; 
whether it is invariably present or essential in 
the pathogenesis of the renal disturbances is 
still an open question. It is known that when 
shock does exist it usually responds to thera- 
peutic measures, so that within less than 24 
hours the general state of the patient appears 
satisfactory. 
The other frequent complication is exces- 
sive vomiting. It may set in soon after the 
insult; it may not occur for a day or so, or 
it may be delayed, and appear as one of the 
manifestations of terminal uremia. Vomiting 
was recorded in 9 of the 14 representative 
cases cited. We shall return to the possible 
significance of this complication in the 
discussion. 
Urinary manifestations. Diminution in the 
output of urine occurs so invariably that it is 
one of the main characteristics of the lower 
nephron syndrome. Generally, reduction in 
urinary elimination is noted within the first 24 
hours. Since shock itself usually leads to re- 
duction, oliguria may not at first occasion con- 
cern. But, usually, despite intake of large vol- 
umes of fluid and other measures to restore 
elimination, oliguria persists and becomes more 
marked; the daily amount of urine passed is 374 
The Military Surgeon—November, 1946 
reduced to less than 500 c.c. Frequently 
oliguria progresses into complete anuria. Such 
urine as is passed is generally highly acid in 
reaction, its specific gravity is low and tends 
to become fixed at approximately 1.010 (see, 
for example, case 6). In many cases, the color 
of the urine may be frankly bloody or smoky 
for the first day or two. In practically all cases, 
regardless of its color, the urine gives a positive 
benzidine reaction, indicating the presence of 
a heme pigment. This pigment is myohemo- 
globin (muscle hemoglobin or myoglobin) 
when the causal condition involves massive de- 
struction of muscle.17 But, there usually is an 
admixture of hemoglobin, which may even 
predominate in some cases of trauma.59 The 
heme pigment which is passed when intra- 
vascular hemolysis takes place is hemoglobin. 
Usually the excretion of either heme com- 
ponent is transitory and no longer demon- 
strable on the third or fourth day. 
Chemical examination further reveals pro- 
teinuria of varying degree. This also occurs 
early, and, unlike myoglobinuria or hemo- 
globinuria, generally persists throughout the 
illness. 
Microscopically the sediment contains 
granular and pigmental casts, the latter com- 
posed of heme derivatives,17 In addition, there 
sometimes are red blood cells, or pigmented 
spherules of heme which resemble red blood 
cells. 
Chemical composition of the blood. The 
changes in the chemical composition of the 
blood reflect the renal shut-down. The level 
of total nonprotein nitrogen rises rapidly so 
that by the third day values may exceed 
150 mg. for each 100 c.c. of blood. Azotemia 
was invariably present in all cases of this series 
in which determinations were made and in 
which survival was at least 2 days. 
As is to be expected azotemia is accom- 
panied by other chemical changes in the blood. 
Thus for the crush syndrome a progressive 
increase in potassium and phosphate, and a 
decrease in the alkali reserve has been re- 
ported.14’15 At times there is a lowering of 
the concentration of blood chlorides (as in 
illustrative cases 5 and 13). 
Hypertension. Early but moderate rise in 
blood pressure is one of the cardinal signs of 
the lower nephron syndrome. In our series a 
common sequence was: on the first day, fall 
to shock levels; on the second day, restoration 
to normal (which for soldiers of the age group 
18 to 38 years ranges around 125/80 mm. 
Hg.) ; on the third day,* a rise to approximately 
150/90; and thereafter the maintenance of 
this level or a further increase. 
Edema. The occurrence of edema is 
variable. It is usually slight or moderate; some- 
times it is generalized; more often it is con- 
fined to the lower extremities or to the lungs. 
Edema of the face is uncommon. 
Uremia develops in all cases in which death 
may be attributed primarily to renal failure and 
not to the precipitating condition. Typical 
manifestations usually appear during the last 
two or three days of life. Vomiting is a fre- 
quent feature. The patient becomes irrational 
or drowsy and lapses into a final coma. Convul- 
sions are uncommon. 
Mortality and length of survival. The mor- 
tality rate in lower nephron disease is very 
high. Once the cardinal signs of the syndrome 
—oliguria, excretion of heme pigment, 
azotemia and hypertension—have appeared, 
the mortality is probably in the neighborhood 
of 90 per cent. The course of the disease is 
relatively brief. For example, in fatal blood 
transfusions the survival period usually is from 
3 to 10 days.26 In the crush syndrome the end 
of the first week is said to be the critical period; 
most patients surviving to the eighth or ninth 
day, will recover.14 
The length of the survival period after 
trauma is graphically shown in Fig. 1 for 100 
cases of the present series. The polygons give 
the number of cases (which here is equivalent 
to percentage) ; the abscissae show the length 
of survival. It will be seen that three-fourths of 
the group (74 per cent) died within eight 
days; only 8 per cent survived for more than 
12 days. Whether death in each case was solely 
the result of the renal changes, or to what ex- 
tent the precipitating conditions contributed to 
the fatal outcome, cannot be ascertained with 
certainty. Lower Nephron Nephrosis 
375 
No.of cases 
LENGTH OF SURVIVAL AFTER TRAUMA 
IN IOO CASES OF 
LOWER NEPHRON SYNDROME 
Fig. i. Duration of disease. The polygons represent the number of cases and the abscissae the corresponding 
duration of disease. It is seen that the great majority succumbed before the 9th day. 
Case I (A.I.P. 137402) Crushing Injury 
A white male, 25 years old, was buried up 
to the waist in a cave-in for about 8 hours on 
November 6, 1944. He voided a large amount 
shortly after release. On admission he com- 
Illustrative Cases 
plained of abdominal pains, and later went into 
shock with a blood pressure of 85/68 mm. Hg. 
On November 7 he vomited 6 or 7 times, and 
did not void. On November 9 he voided a few 
cubic centimeters of acid urine, containing al- 
bumin and red blood cells. At this time his 376 
The Military Surgeon—November, 1946 
right thigh was hot, tense, and greatly swollen 
from knee to hip. Blood pressure was now 
124/90 mm. Hg. He did not void but 15 c.c. 
of urine were obtained by catheterization; the 
specific gravity was 1.012, albumin and red 
blood cells were present. The blood nonprotein 
nitrogen was 114 mg. per cent. November 10: 
Blood pressure was 126/80 mm. Hg.; 
vomiting continued. November 12: Blood 
pressure was 132/84; anuria was complete. 
The patient received several transfusions of 
plasma and of glucose solution. November 13: 
Blood pressure was 158/82. He voided 50 c.c. 
of urine; specific gravity was I.OI2; reaction 
was acid; albumin, red blood cells, granular 
casts were present. November 15: Blood pres- 
sure was 152/86; edema of both legs ap- 
peared. The patient died. Duration of illness, 
9 days. 
Kidneys. Both were greatly swollen and 
soft. The cortical zones were pale and con- 
trasted sharply with the deep purplish-brown 
medulla. 
Case 2 (A.LP. 123343). Crushing Injury 
A white male, 24 years old, sustained severe 
crushing injury in an automobile accident on 
mm. Hg. It was noted that the patient had 
eliminated only a small quantity (about 100 
c.c.) of urine, which was very dark and bloody. 
On December 4, the lower extremities were 
moderately edematous. Transfusions of blood 
were given. The urinary output appeared to 
be low, but the exact amount could not be 
determined. December 5: the general condi- 
tion seemed satisfactory, but the urinary out- 
put continued to be low. Blood pressure was 
158/80 mm. Hg. December 6: massive edema 
of the lower extremities, scrotum, and an- 
terior abdominal wall developed. Only 150 c.c. 
of urine was passed during 24 hours. Decem- 
ber 7: blood pressure was 150/80, Decem- 
ber 8: the patient vomited, became irrational, 
and rapidly went into a deep uremic state, 
dying at 9 a.m. Duration of illness, 8 days. 
Laboratory Lata 
Urine. 12/1. Color red, reaction acid, 
albumin 4 plus, microscopically loaded with 
red cells. 
Kidneys. Combined weight 590 gm., 
greatly enlarged. Capsule thin and tense. Cut 
surfaces show pale cortex and prominent 
brownish-red medulla. 
Case 2 
Blood, 
Date 
Chlorides 
NPN 
Urea N. 
RBC 
(mg/per cent) 
(mg/per cent) 
(mg/per cent) 
(millions) 
12/3 
424 
12/4 
480 
3.8 
12/5 
434 
12/6 
383 
100 
32 
4.1 
12/7 
426 
175 
3i 
December 1, 1944. It is probable that his 
thighs and lower abdomen were pinned-down 
under a wheel of the car for several hours. 
When released he was in profound shock; the 
blood pressure was not obtainable. Shock treat- 
ment, including transfusions of plasma, was ad- 
ministered. Later, an exploratory laparatomy 
was performed, disclosing much hemorrhagic 
infiltration of the pelvic tissue, but no rupture 
of viscera. On December 2, the blood pressure 
had risen to 104/76 and on the 3rd to 143/82 
Case 3 (13th Medical General Laboratory 
15-A-1407)* Crushing Injury 
A white male, 19 years old, was buried be- 
neath rubble in the wreckage of a stone house 
when a bomb exploded at 1 a.m. on Novem- 
ber 22. He was extricated at 4 a.m., but could 
not be evacuated until after dark, reaching the 
hospital 13 hours after extrication. He re- 
* The abstract of this history was kindly furnished 
by Lt. Colonel Tracy Mallory. Lower Nephron Nephrosis 
377 
ceived no intravenous fluid in the interval and 
no medication other than morphine. Shortly 
after admission he voided 300 c.c. of dark 
reddish-brown urine. He seemed in relatively 
good condition with a blood pressure of 96/76 
mm. Hg., pulse of 120, good color and no 
respiratory difficulty. Physical examination was 
essentially negative except for the extremities. 
The muscles of the lower left thigh, the left 
forearm, right calf and, to a lesser degree, 
the lower right thigh were tense, hard, and 
swollen. Pain was elicited by pressure over 
some of the involved areas, but there was al- 
most total sensory loss over the lower right 
leg and foot. Pulsations could be felt in the 
left posterior tibial and dorsalis pedis arteries, 
but were absent on the right side. 
Immediate treatment consisted of shock 
position, fluids and sodium bicarbonate by 
mouth, intravenous infusion of 1,000 c.c. of 
5 per cent solution glucose, and penicillin. 
Oliguria developed promptly, the daily outputs 
running 300 c.c. on November 22, 500 c.c. 
on November 23, 250 c.c. on November 24, 
and 100 c.c, on November 25. The urine had 
a bloody appearance, owing to the presence 
of a heme pigment which was found chemically 
to be predominantly myoglobin, though some 
hemoglobin was also present. On November 
24 the blood pressure rose to 164/100 mm. 
Hg. and remained elevated until just before 
death. Progressive nitrogen retention occurred, 
the nonprotein nitrogen level of the blood ris- 
ing to a maximum of 188 mg. per cent. No 
whole blood transfusions and no sulfonamides 
were given. The patient died with severe pul- 
monary edema on the fourth day. 
Case 4 (A.I.P. 90321). Trauma to Muscle; 
Sulfonamide Reaction (?) 
A white male, 22 years old, was admitted 
on November 8, 1942, with a shrapnel wound 
of the left leg. Under chloroform anesthesia 
the wound was debrided and sprinkled with 
sulfanilamide. The patient was given I gm. of 
sulfanilamide every 4 hours. On November 9 
he received 300 c.c. of plasma intravenously. 
On November 11 the dosage of sulfanilamide 
was reduced to 1 gm. twice daily. (It is not 
known for how long the drug was continued). 
On November 18, 20, and 21, secondary 
hemorrhages occurred in the wound; it was 
reopened and much old clotted blood and 
macerated muscle were removed. Between 
November 20 and 22 five intravenous trans- 
fusions, ranging from 125 to 400 c.c., of 
whole blood or plasma, were administered 
without reaction at any time. The patient did 
not void on November 24. The following day 
75 c.c. of clear urine was obtained by 
catheterization. The blood pressure was 
100/50 mm. Hg. One liter of glucose solu- 
tion and 300 c.c. of plasma were given by 
vein. On November 26 the patient voided 
300 c.c. of dark colored urine. Early manifes- 
tations of uremia were noted and became more 
evident during the following day. On Novem- 
ber 28 he sank into deep uremia, with Cheyne- 
Stokes respiration. Death occurred at 10 a.m. 
Duration of illness, uncertain. 
Laboratory Data 
Urine. 11/25: straw-colored, acid, specific 
gravity 1.010, albumin, occasional casts, a few 
red blood cells. 
11/26: dark colored, hazy, acid, specific 
gravity 1.008, albumin, hematin crystals in 
sediment. 
Blood. 11/26: red blood cells 2.2 millions, 
white blood count 18,400, hemoglobin 60 
per cent. 
Kidneys. Combined weight 400 gm. The 
cut surfaces were moist and showed no note- 
worthy changes, except that the tips of the 
papillae were coated with a crystalline deposit 
which microscopically resembled sulfanilamide. 
Case 5 (A.I.P. 63344). Burns 
A white male, 22 years old, was severely 
burned on April 29, 1942, about the face, 
arms, chest, and right thigh by an explosion 
of gasoline. On admission he was treated for 
shock; the burns were sprayed with tannic 
acid solution followed by silver nitrate; blood 
plasma and normal saline were administered. 
Later in the day the patient began to vomit. 
The morning following the accident he was 
slightly febrile; his pulse was rapid but of good 
quality. He received 5 liters of a 5 per cent 378 
The Military Surgeon—November, 1946 
solution of glucose in saline intravenously, and 
later in the day 250 c.c. of blood plasma. 
Anuria developed approximately 48 hours 
after the burn. His blood pressure rose to 
160/70 mm. Hg.; the nonprotein nitrogen 
to 200 mg. per 100 c.c. of blood. He vomited 
frequently and his condition was considered 
serious. Vomiting continued throughout the 
fourth day and on the fifth day the blood 
pressure rose to 180/90; nonprotein nitrogen 
was 200 and blood chloride 300 mg. per cent. 
He voided a small quantity; the specimen con- 
tained much albumin and many leukocytes and 
epithelial cells. On the sixth day he voided 
120 c.c. Both lower extremities had become 
edematous. Nonprotein nitrogen continued at 
200 mg. per cent. Red blood cells in the earlier 
days of his illness averaged 5.5 millions; 
hemoglobin 90 per cent; white blood cells 
22,000; polymorphonuclears 93 per cent. The 
patient received no whole blood, but a total of 
approximately two liters of pooled plasma. His 
condition gradually became worse. He died 
May 9 on the sixth day after the accident. 
Kidneys. Swollen and congested; the cor- 
tices were turbid. 
Case 6 (A.I.P. 06719). Burns 
A 43-year-old white male sustained multi- 
ple, third-degree burns on October 15, 1942, 
from an explosion; approximately one-sixth of 
his body was involved. His blood pressure on 
admission to hospital was 120/80 mm. Hg. 
The burned areas were sprayed with tannic 
acid solution. The next day one liter of 5 per 
cent glucose in normal saline and 500 c.c. of 
pooled plasma were given by vein; the tem- 
perature rose to 102 F., and the blood pressure 
to 140/70. October 18, the patient’s general 
condition was good. The burned areas were 
well tanned. Two liters of glucose solution 
were given intravenously. On October 2 2, the 
patient vomited repeatedly; his skin was 
slightly jaundiced; during the following day 
jaundice deepened. By October 23, the output 
of urine was approximately 600 c.c. per day, 
the intake of fluid 4,000 c.c. On October 26 
his condition remained unchanged. Urinary 
output continued at the same rate on the same 
intake. Blood pressure was 150/50, On 
October 28 the patient died. Duration of ill- 
ness, 9 days. 
Kidneys. Combined weight was 410 gm. 
Laboratory data 
Case 6 
Blood. 
Date 
RBC 
WBC 
Hb. 
Polys 
Lymphocytes 
(millions) 
(per cent) 
(per cent) 
(per cent) 
io/i 6 
5 
22,000 
14 gm. 
92 
8 
10/22 
4.4 
13,000 
84 
16 
10/23 
4.4 
36,000 
15 gm. 
10/26 
4.6 
36,000 
15 gm. 
92 
8 
10/24 Total protein 
5.57 gm. per cent 
; albumin 
2.56, globulin 3.01. 
Date 
Icterus Index 
NPN 
(mg/per cent) 
10/22 
11 7.6 
i33 
10/23 
135 
10/24 
124.1 
10/26 
93 
150 
Urine 
Sp. Gr. 
Albumin 
WBC 
RBC 
10/18 
1 plus 
10/22 
1.009 
1 plus 
20-25 (HPF) 
1 
(HPF) 
10/25 
1.007 
1 plus 
15 (HPF) 
10/26 
1.007 
1 plus 
U> 
Ln 
1 
-f* 
O 
(HPF) 
10/27 
1.008 
1 plus 
15-20 (HPF) 
1 
(HPF) Lower Nephron Nephrosis 
379 
They were swollen and flabby. On section 
the cortices bulged; the markings were 
indistinct. 
Case 7 (A.I.P. 104707). 
Transfusion Reaction 
A white male, 37 years old, was injured in 
an automobile accident on October 28, 1943, 
sustaining concussion of the brain, compound 
fractures of the left arm, and multiple lacera- 
tions. He was admitted in shock with blood 
pressure 80/40 mm. Hg.; treatment included 
transfusion of blood plasma. On October 29 
a transfusion of citrated blood was adminis- 
tered. After approximately 250 c.c. had been 
given, the patient broke out in profuse per- 
spiration, became nauseated and vomited 
1,500 c.c. of greenish liquid. Shortly after- 
wards he had difficulty in breathing, his lips 
became cyanotic and his respiration of the 
Cheyne-Stokes type. About 15 minutes after 
the transfusion he began to have a series of 
very severe chills, and his axillary temperature 
rose to 104 F. The condition of the patient 
improved during the next few days, but his 
output of urine was very small. The urine was 
found to be heavily loaded with blood. On 
November 6 the blood pressure had risen to 
190/90 mm. Hg.; the temperature was nor- 
mal. Oliguria persisted. The patient vomited 
repeatedly. His breath had a definite urinous 
odor. On November 9 he fell into a coma and 
died. Duration of illness 11 days. 
Laboratory Data 
Urine. 10/28 to 11/4 (several examina- 
tions). Specific gravity i.oio to 1.030; bloody, 
acid, albumin 4 plus. 11/6: Specific gravity 
I.OIO, acid, albumin 3 plus, many red blood 
cells. 
Blood 
NPN 
Urea N. 
Date 
(mg. per cent) 
(mg. per cent) 
11/3 
in 
n/8 
225 
76 
Kidneys. Combined weight 560 gm. Great- 
ly enlarged and soft. Capsule stripped readily. 
Cut surface showed a bulging pale cortex, and 
brownish-red medulla with exaggerated 
striations. 
Case 8 (A.P.P. 92146). Sulfonamide 
Reaction 
A 24-year-old white male was admitted to 
the hospital on March 10, 1943 with a “cold” 
of one week’s duration. The day prior to ad- 
mission he had vomited and complained of pain 
in the abdomen and back. Temperature ranged 
around ioi° F., respirations averaged 22 per 
minute. The mucous membranes of the upper 
respiratory passages were moderately con- 
gested; the lungs were clear. 
Approximately one month before admission 
the patient had been exposed to meningitis and 
had received a total of 17 gm. of sulfathiazole. 
As far as could be learned, no symptoms re- 
sulted either from the medication or the ex- 
posure. Because of his upper respiratory infec- 
tion and new exposure to meningitis he was 
again given sulfathiazole, receiving a total of 
13 gm. within 48 hours. His respiratory infec- 
Laboratory data 
Case 
8 
Date 
RBC 
WBC 
Hb. 
Date 
NPN 
(Million) 
(per cent) 
(mg. per cent) 
3/11 
4.6 
10,400 
9° 
3/i5 
160 
3/15 
4-5 
3o.75o 
80 
3/16 
150 
3/i7 
3-°4 
68 
3/i7 
1 70 
U rine 
Date 
Color 
Sp. Gr. 
Reaction 
Albumin 
RBC 
3/13 
Brownish 
1.030 
acid 
plus 
many 
3/16 
1.018 
plus 
many 
3/18 
Brownish 
1.019 
acid 
plus 
many 380 
The Military Surgeon—Novembery 1946 
tion cleared, but nausea, vomiting, and pain 
in back and abdomen recurred. His urinary 
output rapidly declined; the total amount ex- 
creted from March 14 to March 18 was ap- 
proximately 150 c.c. Blood pressure fluctuated 
between 125/75 and 175/95 mm. Hg. Pul- 
monary edema developed rapidly. To relieve 
respiratory embarrassment phlebotomy was 
performed on March 14 and 500 c.c. of blood 
removed; a second phlebotomy was done on 
March 16 and 600 c.c. removed. On 
March 17 he had a large gastric hemorrhage. 
He died March 18 with manifestations of 
uremia. Duration of illness, 8 days. 
Kidneys. Combined weight 410 gm., both 
markedly swollen. On cut section the 
parenchyma bulged; the cortices were widened 
and pale; the cortical markings were slightly 
dimmed. 
Case 9 (A.I.P. 64841). Blackwater Fever 
A 30 year old Colombian was admitted on 
April 3, 1943 with a greatly enlarged spleen 
and a diagnosis of malaria. On April 2 he had 
vomited brownish fluid. The temperature was 
104 F., the blood pressure 94/20 mm. Hg. 
The scleras were jaundiced. He did not void 
between April 3 and April 5, when he passed 
some blood-stained urine. On April 6 he be- 
came irrational; his blood pressure was 92/50 
mm. Hg, On an intake of two liters of fluid 
per day, the urinary output was 105 c.c. On 
April 7 the patient vomited a great deal. The 
urine was no longer bloody. The intake of 
fluid was 1840 c.c., the output of urine 
240 c.c. His temperature had been normal 
since April 4. On April 8, blood pressure was 
56/40. The patient died April 10. Duration 
of illness, 8 days. 
Laboratory Data 
Blood. On 5 examinations erythrocytes 
were found to range from 2.2 to 3 million per 
cu. mm,, the hemoglobin from 55 to 65 per 
cent. 
4/10, NPN, 218 mg. per cent. 
Urine. 4/5: specific gravity 1.013, acid, 
large amount of albumin, no bile, benzidine 
test for blood strongly positive; sediment con- 
tained “degenerated erythrocytes.” 
4/7: specific gravity I.OIO; acid, yellow 
color, trace of albumin, benzidine test nega- 
tive; no red blood cells in sediment. 
4/10: specific gravity 1.010, neutral, straw- 
colored, albumin 2 plus, sediment contained 
pus cells and a few red blood corpuscles. 
Kidneys. Combined weight 540 gm., en- 
larged, cortices thickened and cloudy, mark- 
ings blurred, scattered brownish streaks, and 
pyramids also streaked with brown. Small pale 
patches in boundary zone. 
Case 10 (A.I.P. 79906). Arthritis; 
Hyferfyrexia 
A colored male, 26 years old, had suffered 
from recurrent attacks of arthritis involving 
different large joints of the body. The ad- 
mission to the hospital on December 23, 1942, 
was the fifth for arthritis. On December 28 
he received an intravenous injection of typhoid 
vaccine; this was followed by a moderately 
severe chill, and a rise in temperature to 
103° F. On December 29 the injection was 
repeated. Within an hour a severe chill de- 
veloped and his temperature rose rapidly to 
109° F. Between January 1 and 3 the tem- 
perature averaged about 1 degree above nor- 
mal. The urinary output gradually diminished; 
after January 3, the volume ranged from 90 
to 120 c.c. per day. He was treated by intra- 
venous injections of glucose solution and blood 
transfusions. Uremic symptoms developed and 
he died on January 6. Duration of illness, 
9 days. 
Laboratory Data 
Blood 
NPN 
Date 
(mg. per cent) 
i/3 
75 
1/4 
99 
x/5 
x 09 
Urine. All specimens after January 5, 
1942, contained much albumin, and many red 
and white blood cells. 
Kidneys. Combined weight was 650 gm. 
On section the parenchyma bulged; the corti- 
cal zones were pale and widened; the corti- 
comedullary boundary was indistinct. Lower Nephron Nephrosis 
381 
Case 11 (A.I.P. 89891). Toxemia of 
Pregnancy (?); Uterine Hemorrhage 
A white female, 22 years old, was admitted 
in labor on February 19, 1943. Blood pres- 
sure was 170/110 mm. Hg. During the third 
stage, copious hemorrhage occurred, and the 
patient went into shock; blood pressure fell to 
a non-readable level. She received two trans- 
fusions of plasma, totalling 1,000 c.c., and one 
injection of 300 c.c. of whole blood. The 
uterus was packed, but oozing of blood, and 
shock were not relieved. The uterus was now 
explored and a piece of placental tissue was 
found and removed. Since no suitable blood 
was available a total of 2300 c.c. of a hemo- 
globin solution was injected intravenously dur- 
ing the next 24 hours. Following these trans- 
fusions her blood pressure rose to 150/100. 
Sodium bicarbonate was administered to keep 
the urine alkaline. The following day the pa- 
tient’s condition was satisfactory. On the 
fourth day postpartum the patient voided on 
an average less than 5 c.c. per hour despite 
transfusions of citrated blood and glucose so- 
lutions. On the sixth day postpartum urinary 
output increased to approximately 15 c.c. per 
hour. On the eighth day she became restless, 
confused, and on the 9th day (February 27) 
died suddenly. 
Kidneys. Combined weight was 600 gm. 
The cut surfaces exuded a thin bloody fluid. 
The cortical zones were pale, edematous, and 
sharply demarcated from the congested 
medullas. 
Case 12 (A.I.P. 18444). Mushroom 
Poisoning 
A white male, 27 years old, on March 13, 
1942, developed severe epigastric pain, vomit- 
ing, and diarrhea approximately 12 hours 
after having eaten some “mushrooms.” The 
“mushrooms” were subsequently identified as 
Amanita Phalloides. The urinary output 
rapidly decreased, and during the 12 days 
prior to death was alleged to total only 330 c.c. 
The specimens contained much albumin, 
numerous erythrocytes, leukocytes and casts. 
The nonprotein nitrogen of the blood grad- 
ually rose to 265 mg. per 100 c.c. Death 
occurred on March 25. Duration of illness, 12 
days. 
Kidneys. Combined weight 470 gm.; 
swollen, edematous, cortical zones widened. 
Case 13 (A.I.P. 87329). Alcohol Poisoning 
A white male, 37 years old, was admitted 
on December 6, 1942 after he had been 
treated for 3 days by a civilian physician for 
“gastritis,” possibly resulting from drinking 
adulterated alcohol. On admission he was 
quiet, but soon became restless, excited, con- 
fused, and disoriented; the clinical impression 
was delirium tremens. On physical examina- 
tion the throat was found to be reddened, 
there were moist rales over the posterior bases 
of the lungs. Blood pressure was 136/84; 
pulse 62. Electrocardiogram indicated marked 
myocardial damage and right bundle block. 
The patient continued to be disoriented, and 
hallucinations developed. On December 8 
scleras and skin became jaundiced; on the 
next.day jaundice was more marked. Output 
of urine steadily diminished to approximately 
250 c.c. per day; the specimens had to be ob- 
tained by catheter, and were dark colored and 
bile stained. One liter of 10 per cent glucose 
solution in saline was given 3 times daily by 
vein. Because of upper respiratory infection, 
sulfanilamide was employed; the amount prob- 
ably was small as the postmortem blood con- 
centration was only 1 mg. per 100 c.c. During 
the morning of December 10, the blood pres- 
sure fell to 82/60. Patient died on this day. 
Duration of illness, 7 days. 
Laboratory Data (12/7 to 12/9) 
Urine contained many red cells and granular casts. 
Blood 
CO2 combining power 77 volumes per cent. 
Chlorides 160 mg. per cent. 
Sugar 141 mg. per cent. 
NPN 150 mg. per cent. 
Urea N 33 mg. per cent. 
White blood cells 10,000, polymorphonuclear 
leukocytes 95 per cent. 
Kidneys. Swollen; cortices widened, pale 
and turbid; medullas grossly normal. 382 
The Military Surgeon—November, 1946 
Grams 
WEIGHT OF KIDNEYS 
Days 
LENGTH OF SURVIVAL 
• Following trauma 
O Not associated with trauma 
Fig. 2. Weight of kidneys in lower ne-phron nephrosis. The combined weight of the kidneys has been 
plotted against the length of survival for 89 cases, of which 49 followed trauma and 40 were not associated 
with trauma. In all but 10 instances the weight of the kidneys exceeds the normal, 300 gm.; the median 
weight for the group shown is approximately 450 gm. 
Case 14 (A.I.P. 90402). Carbon 
Tetrachloride Poisoning 
A white male, 29 years old, on February 1, 
1943 accidentally drank an unknown quan- 
tity of carbon tetrachloride (the estimated 
amount was 2 or more ounces). Soon after- 
wards he became dizzy and had epigastric 
pains; he took some milk and forced himself 
to vomit. The next day he felt well enough 
to work, but became febrile during that night. 
Beginning February 2 he was nauseated and 
vomited after eating. 
The output of urine steadily diminished. 
Thus, on February 7 the intake of fluid was 
1980 c.c., the output 810; on February 9, 
intake 1800, output 560; on February 10 Lower Nephron Nephrosis 
383 
Fig. 3. Affearance of cut surface of kidney in a representative case of lower nephron nephrosis. (The 
combined weight of the kidneys was 500 gm.) The cortex is pale, and here and there shows whitish stripes 
or patches. The medulla is dark and its striations are accentuated. (Photograph reduced to approximately 
7/10 normal size. A.I.P. Acc. 135173) 
intake 1350, output 480 c.c. The course con- 
tinued downhill. On February 5 he became 
jaundiced. He daily vomited blood stained ma- 
terial. His temperature remained normal until 
two days before death when it rose to 100° F. 
On February 11 pulmonary edema developed; 
the patient became very dyspneic, and died on 
February 14. Duration of illness, 13 days. 
Kidneys. Combined weight 620 gm.; pale 
color; cut surfaces bulge; cortical zones 
turbid; markings indistinct. 
Pathologic Anatomy 
In lower nephron nephrosis organic changes 
other than those incident to the precipitating 
condition are largely confined to the kidney. 
The only other common lesions are edema and 
scanty petechial hemorrhages on serous or 
mucous surfaces. We shall here confine our- 
selves to the changes that may be observed in 
the kidney. 
Gross affearance of the kidney 
The gross appearance of the kidneys is not 
pathognomonic. They are usually swollen and 
their weight is increased. Thus in 89 cases 
chosen at random from the present series, the 
weight in all but 10 exceeded the normal 
(which, for young adults, averages 300 gm.). 
In 64 cases the kidneys weighed from 400 to 
Laboratory Data 
Urine 
Date Sp. gr. Albumin 
Casts 
2/6 1.007 
Trace 
Few granular 
2/12 1.019 
Trace 
Granular 
2/13 1.020 
Trace 
Granular 
Icterus Index 
2/6 
35 
2/7 
50 
Blood 
NPN 
Urea N 
Date (mg. 
per cer 
it) (mg. per cent) 
2/12 
150 
58 
2/13 
240 
45 384 
The Military Surgeon—November, 1946 
Fig. 4 
600 gm., and 13 were more than twice the 
normal weight. In Fig. 2 the weight of kid- 
neys in individual cases is graphically analysed, 
first, with respect to duration of disease, and, 
second, with reference to the main types of 
precipitating conditions—traumatic and hemo- 
lytic. Considerable scattering of the weights is 
evident. For example, in 15 cases with a sur- Lower Nephron Nephrosis 
385 
vi’val period of 7 days it varied from 300 to 
over 800 gm. Hence, no definite correlation 
is found between size of kidney and duration 
of disease, although there is a tendency for 
greater swelling when the survival period is 
prolonged. There is also a suggestion that 
increase in size after trauma or burns is more 
marked than after nontraumatic (hemolytic) 
conditions. 
In a typical case, the organ is flaccid; the 
capsule is easily stripped off; the outer surface 
is smooth and rather pale. The cut surfaces 
ooze clear or slightly bloody fluid. The cortex 
is distinctly widened, bulges perceptibly, is 
moist and pale in sharp contrast to the dusky 
medulla, the striations of which are often ac- 
centuated. In the inner zone of the cortex 
(and sometimes elsewhere) a distinctly whitish 
stripe is sometimes seen (16); a fairly repre- 
sentative cut surface is shown in Fig. 3. 
Microscopic appearance of the kidney 
The histopathologic picture of lower 
nephron disease has four distinctive features; 
1. Degeneration or actual necrosis which 
involves selectively focal portions of the lower 
segments of the nephrons, i.e. the thick tubules 
of Henle and the distal convoluted tubules. 
2. Edema and cellular reaction which de- 
Fig. 4. Diagram of the structure of the normal nephron and of the site of the lesions in lower nephron nephrosis. 
Each nephron comprises a vascular complex, the renal corpuscle and an attached, unbranching tubule. The latter is 
differentiated into .three principal segments, the proximal (upper), intermediate, and lower. The remaining tubules of 
the kidney, the collecting tubules, are excretory ducts. Each of the different components has specific functions, character- 
istic structure, and may sustain selective damage. 
Renal corpuscle: The structure and function of the glomerulus and its capsule are too well known to require much 
comment. Suffice It to recall that the glomerules are filters which daily separate about 150 liters of fluid from the blood. 
For the filtration of this relatively large amount of fluid it is necessary that both circulation and pressure within the 
glomerular capillaries be kept at an adequate level. 
Proximal segment (upper segment; proximal convoluted tubule). These segments constitute the bulk of the cortical 
parenchyma. Emerging from the glomerular capsule each forms a compactly coiled mass. Normally the lining cells are 
large, many have a protoplasmic “hump”; their free surface is covered by a hair-like brush border; their lateral bound- 
aries are indistinct; the cytoplasm is definitely eosinophilic. In its terminal or spiral portion this segment uncoils and, 
without change in structure, extends into the boundary zone of the medulla. In the proximal segment the glomerular 
fluid is concentrated by reabsorption of approximately 80 per cent of water; glucose is completely resorbed, and various 
other modifications take place.74,87 
Intermediate segment. (Thin limb of Henle.) The transition from the proximal to the intermediate segment is 
abrupt. The diameter narrows from approximately 60 to 20 microns. The epithelium becomes flat and so thin that Its 
nuclei may bulge into the lumen. In general appearance the intermediate segment resembles a capillary. It runs a 
straight course of greatly variable length. Some segments are short and form a hairpin loop in the boundary zone; 
others are relatively long and, before looping, extend nearly to the tips of the renal pyramids. The length of the 
segments depend upon the position of their corresponding renal corpuscles; the nearer the latter are located to the 
corticomedullary junction, the longer are these tubules. The precise function of the intermediate segment is not known: 
it may further concentrate the glomerular filtrate; mechanically it has been likened to a pressure trap which slows the 
flow of fluid within the upper segment, thus prolonging contact with its resorbing surface. 
Lower segment. (Thick limb, of Henle and distal convoluted tubule.) These two portions are regarded as a histologic 
and functional unit, i.e. the lower segment. The transition from thin to thick limb may take place in the descending 
portion, in the loop, or in the ascending portion of these tubules. Usually the change is sudden, the diameter Increasing 
to approximately 40 microns. The thick limbs run a straight course in the boundary zone and ascend into the medullary 
rays of the cortex, where they turn and approach the vascular pole of the corresponding renal corpuscle. Here, without 
preceptible change, they become the distal convoluted tubules. The latter run a zig-zag course between coiled masses. 
Because of their angulations, several cut sections of the same lower segment usually lie as a group within a field of 
the microscope. Compared to the proximal segment, the cells of the lower segment are smaller, more numerous, and 
lower; they lack a brush border; their cytoplasm is but faintly eosinophilic, and, in sections stained with hematoxylin- 
eosin, the cytoplasm may even have a slightly bluish tint. In the lower segment, the intertubular fluid becomes acidified 
and undergoes its final modifications before being emptied as urine into the collecting tubules.87 
It Is in the lower segment, and in the adjacent stroma and its thin-walled veins, that the principal lesions of lower 
nephron nephrosis are located. As indicated in the diagram the segment Is usually not Involved throughout its length, 
but in a patchy, focal manner. 
Collecting tubule. The collecting tubules are not part of the nephrons but excretory ducts that convey the now 
completely modified glomerular filtrate, i.e. the urine, to the renal pelvis. They begin in the cortex, run a straight 
course through the rays, the boundary zone and the medulla, to the apex of the papillae, their caliber gradually 
increasing. Their epithelium is clear, the cell boundaries are well defined, the nuclei sharply outlined, the cytoplasm 
basophilic. The collecting tubules are rarely damaged to a significant degree, but their lumens may be packed with 
heme casts, 386 
The Military Surgeon—November, 1946 
velop in the stroma around the more severely 
damaged or disintegrated portions of the 
tubules; these changes are commonly associated 
with thrombosis of the adjacent veins. 
3. Casts of a heme compound which lie 
within the lumens of variable numbers of the 
lower segments and of the collecting tubules. 
4. Relatively slight or no structural changes 
in the upper parts of the nephrons—renal 
corpuscles, proximal and intermediate seg- 
ments. 
It is clear that the damage in this disease 
is essentially confined to a particular com- 
ponent of the nephron. The identification of 
the several components is not difficult; their 
appearance and location are shown diagram- 
matically in Fig. 4. Here also is indicated the 
site of the tubular lesions. We may now con- 
sider the histopathologic picture in greater de- 
tail. 
The renal corpuscles. The glomeruli are 
normal in size and cellularity; their coils 
are patent, but commonly their blood con- 
tent is poor, suggesting inadequate circula- 
tion.28’47’56’90 The capsular spaces usually are 
of normal width;28 dilatation, such as occurs 
when tubules are completely obstructed, is 
rare. Nearly always, the spaces contain an 
abundant eosin-staining, granular, foamy, or 
globular protein precipitate,16’30’90 indicating an 
increase in glomerular permeability. Cuboidal 
swelling of the normally flat epithelium that 
lines the capsule is sometimes observed, 
especially near the mouth of the tubule. The 
cellular swelling may be related to the leakage 
of proteins through the tufts, sine? renal 
epithelium is to some extent capable of absorb- 
ing proteins. 
An interesting change has lately been de- 
scribed in the juxtaglomerular apparatus;47 in 
cases of crush injury marked hypertrophy and 
increased granularity of the cells were noted. 
Systematic examination of our own material 
has not as yet been carried far enough to 
determine whether similar changes occurred. 
Proximal segment. The appearance of the 
lining cells of the proximal segment is often 
quite normal; even their delicate brush border 
is well preserved. At times, however, the cells 
are slightly swollen and somewhat more gran- 
ular, that is to say they exhibit “cloudy swell- 
ing.” Occasionally fatty deposits are disclosed 
by appropriate staining; frank degeneration or 
actual necrosis is very rare. The lumens of the 
tubules sometimes seem slightly dilated, and 
commonly contain a protein precipitate similar 
to that in the capsular spaces, whence it prob- 
ably has been washed down (Fig. 5, 6, and 
11). Heme casts are very rarely observed in 
this segment. 
The comparative absence of significant 
damage in the proximal tubule is of interest. 
Most renal poisons—mercury, uranium, 
oxalates, tartrates and cantharidin—damage 
this segment almost or quite exclusively.31’32’33 
Intermediate segment. Definite changes can 
seldom be found in the intermediate segment; 
there is neither obvious cellular injury nor 
precipitation of heme compounds. 
Lower segment. It is widely agreed that 
the lower segment is damaged selectively in 
the various conditions mentioned in the intro- 
duction. For example, such selective damage 
has been reported in cases of crush injury,16’30 
in burns,46’53 in blood transfusion fatalities,3’26 
in uteroplacental damage,90 in sulfonamide 
intoxication,42’52 and after excessive vomit- 
ing.55 
In summarizing the microscopic appear- 
ance, it is convenient to describe, first, the 
changes in tubular structure, and then the casts 
within the tubules and the reactions in the 
neighboring stroma and the vessels. 
I. Tubular alterations. The damage pri- 
marily involves the tubular cells. It varies 
widely in degree from slight degeneration to 
frank necrosis and complete disintegration. In 
general, when the survival period is less than 
three days the regressive changes are relatively 
inconspicuous and are indicated chiefly by 
increased granularity and by minute droplets 
of fat. More definite degeneration or frank 
necrosis may be seen by the third or fourth 
day. In distribution, the lesions are character- 
istically focal, affecting small areas rather than 
whole segments, many of which appear to 
escape damage entirely. At times the most pro- 
nounced lesions are in the boundary zone, or Fig. 
5 
Fig. 
6 
Fig. 
7 
Fig. 
8 
Figure 5. Pigment casts in distal tubules. Area of tubular destruction and interstitial inflammatory 
reaction. The glomerulus and the proximal tubules show little alteration. Case of transfusion (AIP 
Acc. 79906). 
Figure 6. Pigment casts in lower tubular segments adjacent to a thin-walled vein. Protein precipitate 
in glomerular space and in proximal tubules, which otherwise are normal. Case of burns (AIP 
Acc. 83444). 
Figure 7. Interstitial edema. Several distal segments have disappeared completely; other segments 
are degenerating or contain pigment casts. Case of mushroom poisoning (AIP Acc. 81444). 
Figure 8. Parietal thrombus in thin-walled vein, probably resulting from rupture of necrotic tubule 
into vein. Case of crush injury (AIP Acc. 81070 A). Fig. 
9 
Fig. 
io 
Fig. 
12 
Fig. 
i i 
Figure 9. Interstitial edema. Disintegration and early regeneration of the lining of several distal 
segments. Dense hyaline cast in a dilated tubule. Case of crush injury (AIP Acc. 81070 A). 
Figure io. Area of tubular disintegration and inflammatory reaction. Pigment casts, and one large 
“mixed” cast. Case of burns (AIP Acc. 83444). 
Figure ii. Area of tubular disintegration and interstitial cell reaction. Hyaline cast in a disinte- 
grating segment. Normal proximal tubules, with protein precipitates in lumens. Case of burns 
(AIP Acc. 83444)- 
Figure 12. Area of tubular disintegration and inflammatory reaction. Pigment casts. Case of sulfa- 
thiazole intoxication (AIP Acc. 92146). Lower Nephron Nephrosis 
387 
at points where the lower segments lie in close 
proximity to veins.16’30 Obvious dilatation of 
the lumens is relatively uncommon, but local 
outward bulging (herniation) or actual rup- 
ture of necrotic portions is frequent. It is 
interesting to note that diverticula are com- 
mon in the lower segment; excellent photo- 
graphs of them are given in a recent paper by 
Oliver.67 From about the fifth day onward, 
the remnants of necrotic portions of tubules 
may undergo complete distintegration and dis- 
appear (Fig. 5, 7, 10, 11, 12). 
Where the destructive process is not so 
intense, and when the survival period has ex- 
ceeded 3 or 4 days, regeneration is evident. 
Through proliferation of epithelium that has 
escaped irreversible injury, new cells creep be- 
neath the dead lining. The latter becomes de- 
tached and is cast off into the lumen where it 
further disintegrates. The new cells differ 
from the pre-existing epithelium; at first they 
are thin and flat, resembling endothelium; 
later they become cuboidal in shape. In the 
early stages their cytoplasm is distinctly 
basophilic, and the nuclei are richly chromatic. 
The healing process proceeds rapidly; within 
less than 10 days most damaged areas are 
completely relined. 
2. Casts. Aside from cellular debris two 
kinds of casts are common within the lumens. 
The most conspicuous casts are pigmented 
masses of a heme compound; they are the most 
prominent, though probably not the most sig- 
nificant feature in the histopathologic picture 
(Fig. 5-7, and 10-16). After destruction of 
muscle, the pigmented masses are composed 
predominantly of myohemoglobin or deriva- 
tives thereof;17 in hemolytic conditions they 
are formed by hemoglobin and its degradation 
products. Both types of heme casts are micro- 
scopically alike, and their staining properties 
are similar. In unstained preparations they 
have a reddish hue; in sections stained with 
hematoxylin and eosin they usually are 
brownish or copper colored. Some are smooth 
and solid in texture (Fig. 13), others granu- 
lar (Fig. 13, 16); frequently they look like 
strands of coarsely beaded wire or frayed rib- 
bons (Fig. 14) ; at times they have the form of 
spherules which may be mistaken for erythro- 
cytes (Fig. 15). The casts do not give a re- 
action for free iron, but in their earlier stages 
they have the tinctorial reactions of hemo- 
globin. 
These heme masses are relatively scanty in 
the beginning straight part of the lower 
nephron; they become more numerous in the 
convoluted portion. They are often particu- 
larly prominent in tubules immediately ad- 
jacent to thin-walled veins (Fig. 6). 
The number of segments containing casts 
varies greatly in different cases; rarely they 
are entirely absent, in some cases they must 
be searched for, in others they are conspicous 
in every microscopic field. They are infre- 
quently found when the survival period is less 
than 2 days. However, unlike the pigment ex- 
cretion in the urine, which lasts but a day or 
so, the casts in the tubules usually persist. 
The second kind of cast is nonpigmented, 
hyalin in texture, stains faintly with eosin, 
and resembles a dense coagulum of protein 
(Fig. 9, 11). It is much less common than the 
heme cast, and tends to occur only in portions 
of tubules that have been severely injured. 
It lies more often in the straight than in the 
convoluted portions of the lower segment. The 
lumens of affected tubules are completely ob- 
structed, and upstream from the point of 
blockage may be definitely dilated. Through 
rupture or distintegration of portions of the 
tubule these casts are extruded into the stroma. 
3. Stroma. The interstitial tissue around 
foci of tubular disintegration is edematous and 
exhibits an inflammatory reaction (Fig. 5, 7, 
9-12). At first, the participating cells are pre- 
dominantly lymphocytes and histiocytes; 
granulocytes are usually scanty and giant cells 
occasional. At later stages, fibroblasts make 
their appearance, and when survival exceeds 
a week, the destroyed parenchyma is replaced 
by recent scars. 
The areas of edema and cellular reaction in 
the stroma are as a rule small and scattered, 
although they may be numerous. They con- 
form with sites of tubular necrosis; the foci 
may be especially prominent in the boundary 
zone, in the cortex near glomeruli, around 388 
The Military Surgeon—November, 1946 
venous channels, and around extruded casts. 
4. Involvement of veins. Some of the large 
venous channels of the kidney, especially those 
in the boundary zone, have a unique structure: 
they are so exceedingly thin walled that nor- 
mally the adjacent renal tubules bulge into 
their lumens. When tubules become necrotic, 
they may rupture into a vein, spilling their 
contents and inducing thrombosis.3’16’30’52’55 
The thrombi usually are parietal (Fig. 8), 
rarely occlusive. Remnants of epithelium are 
commonly found embedded in the clot. The 
walls of the veins are often infiltrated with 
inflammatory cells. Some tubulovenous lesions 
occur in the majority of cases having a survival 
period of more than five days. 
Collecting tubule. Conspicuous degenera- 
tive changes rarely involve the collecting 
tubules. The majority appear normal, only an 
occasional tubule is found partly stripped of 
its epithelium or relined by regenerated cells, 
which may become heaped up.16 Casts of heme 
compounds, however, are usually more con- 
spicuous in these tubules than in the lower 
nephron (Fig. 14, 16); they are larger, and 
fill longer stretches. At times they become 
coated by epithelial cells and attract leuko- 
cytes; heme casts in the lower nephron seldom 
exhibit either of these phenomena. It seems 
probable, therefore, that the casts in the col- 
lecting tubules represent an older and more 
advanced stage of degeneration of the heme 
compounds. The number of collecting tubules 
involved and the completeness of obstruction 
vary greatly. 
Pathogenesis and Functional 
Disturbances 
From the many problems concerning 
pathogenesis of lesions, and mechanism of func- 
tional disturbances we have singled out several 
for brief discussion: What part is played by 
heme compounds, by products of tissue break- 
down, by disturbances in the physicochemical 
composition of the blood, by shock and renal 
vasoconstriction in the production of the spe- 
cific renal changes? What is the mechanism 
of oliguria; does it involve extrarenal factors? 
None of these questions can as yet be an- 
swered, rather a discussion of them may point 
the way to further investigation. Moreover, 
the several questions are difficult to separate, 
for the factors concerned are interrelated. It 
will be seen that no single hypothesis can satis- 
factorily account for the renal disturb- 
ances;34’40 perhaps several factors act con- 
jointly, their combination depending upon the 
precipitating condition. 
Role of heme compounds 
Characteristic of the renal lesions is the 
almost constant presence of heme casts. As has 
already been stated, two kinds of hemoglobin 
compounds are involved, myohemoglobin, and 
hemoglobin. These two compounds differ in 
various respects,62 but it is fair to assume that 
the role they play in regard to the renal dis- 
turbances is similar. We shall consider, first, 
how these compounds are excreted by the kid- 
neys; second, why they are precipitated in the 
lower segments, and, third, whether they or 
any of their degradation products are toxic. 
Excretion of heme compounds of kidneys, 
(a) hemoglobin. When destruction of red 
cells is great, as in extensive intravascular 
hemolysis, not all the liberated hemoglobin 
can promptly be metabolized into bile pig- 
ments, and much of it passes through the 
glomerular filters.38’91 The manner in which 
this passage takes place is a matter of dispute. 
The glomerular capillaries may be regarded 
as thin membranes, containing liquid-filled 
pores of various size through which a filtrate 
can be expressed by a “blind physical force,” 
i.e., by hydrostatic pressure.74 It is generally 
accepted that the pores of the membranes are 
smaller than the diameter of the molecule of 
serum albumin, the molecular weight of which 
is approximately 70,000, and which normally 
does not escape through the glomerulus. Since 
the hemoglobin molecule has very nearly the 
same weight, several possible explanations have 
been advanced for its passage; (1) that a 
small proportion, perhaps 3 per cent, of the 
capillary pores are sufficiently large to permit 
penetration of hemoglobin; (2) that hemo- 
globin may dissociate into molecules of smaller 
dimensions which can pass through most of the Fig. 
13 
Fic 
14 
Fig. 
15 
Fig 
16 
Figure 13. Smooth (recent) and granular (older) hemoglobin casts in distal tubules; the older casts 
lie in segments adjacent to the glomeruli. Case of eclampsia and hemoglobin transfusion (AIP 
Acc. 89891). 
Figure 14. Copperwire-like heme casts in collecting tubules. Case of transfusion (AIP Acc. 79906). 
Figure 15. Heme casts composed of spherules which resemble erythrocytes. Case of heat stroke 
(AIP Acc. 120161). 
Figure 16. Dense granular heme casts in collecting tubules. Case of crush injury (AIP Acc. 81070 B). Lower Nephron Nephrosis 
389 
pores of the normal glomerular capillaries; and 
(3) that glomerular permeability is increased, 
i.e., that all of the pores are widened. For de- 
tailed discussions of these opposing views the 
papers of Yuile,91 Botts and Richards,11 and 
Foy, Altmann, Barnes and Kondi41 should be 
consulted. 
Here we shall consider only the hypothesis 
that glomerular permeability is altered. It is a 
known fact that permeability of living cells 
may be made to increase considerably without 
inducing detectable changes in structure; 
within a relatively wide range such an increase 
is reversible. Now, it has been demonstrated 
that hemoglobin, when its concentration in 
the plasma exceeds a certain level, exerts a 
specific, transient vasoconstrictor action on 
renal arterioles.61 It has also been found that 
renal vasoconstriction, produced experimental- 
ly by several means in man and in animals, 
leads to transient albuminuria through in- 
crease in permeability of the glomerular capil- 
laries,80 no structural alterations are evident 
and the effect on permeability is reversible. It 
has further been shown in human experiments, 
that albuminuria appears in all cases in which 
sufficient hemoglobin is injected intravenously 
to induce hemoglobinuria.44 These experi- 
ments support the hypothesis that hemoglobin 
is excreted by the kidney, in large quantities at 
least, only when the permeability of the 
glomerular filters is increased. It is very prob- 
able that this change is effected primarily by 
arteriolar vasoconstriction, which subsequently 
through oxygen lack leads to alterations in 
capillary permeability. Renal vasoconstriction, 
as we shall see, is a factor of prime importance 
in the production of the structural and func- 
tional disturbances in the kidney. Hence, the 
mechanism of hemoglobin escape is not only of 
interest in itself, but also in that it focuses 
attention on vascular constriction in the kidney 
and on changes in glomerular permeability. 
(b) Myohemoglobin. The molecular 
weight of myohemoglobin is 17,500, which is 
sufficiently small to permit its ready passage 
through unaltered glomerular capillaries. But 
it is probable that the conditions which lead 
to liberation of myoglobin from injured muscle 
cells also bring about renal vasoconstriction and 
an increase in permeability of the glomerular 
filters. That permeability is definitely in- 
creased is demonstrated clinically by the al- 
most invariable occurrence of proteinuria, and 
histopathologically by the protein precipitates 
seen in sections. 
Precifitation of heme compound in the lower 
segments of the nefhon 
It is known of hemoglobin, and it is prob- 
ably true of myoglobin, that when it has passed 
through the glomerular filter, a relatively 
small portion is reabsorbed by the cells of the 
proximal segment, through a process akin to 
phagocytosis.91 The larger portion of the pig- 
ment remains in solution until the lower seg- 
ment is reached. Here some of the heme com- 
pound may be removed from solution by pre- 
cipitation caused by factors only • partially 
known. 
According to one hypothesis, that of Baker 
and Dodds,4 hemoglobin is thrown out of solu- 
tion, probably as hematin, when the intra- 
tubular fluid becomes sufficiently acid (below 
pH 6.0), and when simultaneously through 
absorption of water the concentration of sodium 
chloride is increased to about 1 per cent. These 
requirements are met in the lower nephron87 
and in the collecting tubules. In analogy with 
these experiments on hemoglobin, it has been 
shown by Bywaters and Stead19 that intra- 
venous injections of myohemoglobin may lead 
to renal failure with pigment retention when 
the acidity of the urine reaches levels of pH 
4.5 to 6.1. 
The hypothesis that acidity of the urine is 
of prime importance in the genesis of renal 
damage has led to widespread therapeutic use 
of alkalinization in cases of severe muscle 
trauma or intravenous hemolysis. But there is 
some evidence that hemoglobin and its 
products, as well as other solutes, are more 
readily secreted in an acid than in an alkaline 
urine.28’40 It has also been pointed out that 
blood transfusion fatalities may occur when 
the reaction of the urine is persistently alka- 
line;28’40’91 and, similarly, that in blackwater 
fever, anuria is no more common in patients 390 
The Military Surgeon—November, 1946 
with acid than with alkaline urine.40 It is 
obvious that further investigation of this im- 
portant question is demanded. 
Another hypothesis relates the precipitation 
of heme compound to cellular injury.40’90 
Some support of this view has been obtained 
in recent experiments. It was shown by Yuile, 
Gold and Hind that heme pigment is precipi- 
tated in tubules previously damaged by 
ischemia or by chemical poison.92 Cellular in- 
jury in the lower nephron, like precipitation 
of heme compounds, has been attributed to the 
acid reaction of the intratubular fluid in the 
segment.30’31’54 
It has also been suggested that the inade- 
quate flushing of the tubules due to lowered 
filtration pressure may be a factor leading to 
accumulation and retention of pigment masses 
within the tubules.40 
In connection with these unsolved questions, 
certain points made by Oliver in his recent 
Harvey Lecture67 are pertinent. When 
nephrons are dissected out in their entirety, 
it may be seen more clearly than in sections 
that coagula and casts are not found in all 
parts of the tubular system, but that they 
occur particularly in the lower half, i.e., in 
the distal and the collecting tubules. Evidently 
conditions there are favorable for coagulation 
of protein-containing fluids. Referring to 
older work on the so-called x body, a sub- 
stance normally present in urine and a factor 
in coagulation, he reports experiments which 
lead him to the conclusion that proteins attain 
the isoelectric point necessary for coagulation 
only when the glomerular filtrate reaches the 
distal tubules; here, since the concentration of 
the x body is high, cast formation takes place. 
Are heme compounds toxic? It is generally 
accepted that solutions of pure hemoglobin or 
myohemoglobin as such are not nephro- 
toxic.1’4’8’28’40 Discrepancies in results of ex- 
periments are probably explainable on the 
basis of variation in the preparation of solu- 
tions and on species differences. As Foy40 
points out, hemoglobin metabolism in man is 
not comparable with that in dogs, cats, or 
rabbits and the results of hemoglobin injection 
into lower animals should not be unqualifiedly 
referred to man. 
When nephrotoxic effects follow injections 
of either hemoglobin or myoglobin, they are 
generally attributed to abnormal breakdown 
products or derivatives. Some investigators hold 
that such toxic derivatives are most apt to 
form when the reaction of the urine is 
acid.4’19’22 The exact nature of these substances 
is doubtful, but it seems to be experimentally 
established that hematin (sodium ferrihemate) 
exerts definite injurious effects, and pro- 
duces lesions which resemble those of hu- 
man lower nephron disease.1’21’22 This sub- 
stance is released from heme-globin linkage 
during the degradation of either hemoglobin 
or myohemoglobin. It causes intense renal 
vasoconstriction and tubular damage.22 The 
nephrotoxic effect of methemoglobin8 in dogs 
which have been rendered acidotic is perhaps 
also due to release of hematin. 
Whether heme derivatives act directly on 
tubular cells or whether the intratubular heme 
casts are injurious is uncertain. Some observers 
believe that heme casts are not necessary for 
the development of the renal lesions,3’90 and 
that, as already stated, the casts are the conse- 
quence not the cause of the tubular dam- 
age.40’92 But heme precipitates, no matter how 
induced, will probably aggravate cellular in- 
jury caused by other factors.91 
In sum, information concerning the part 
played by heme compounds in the genesis of 
renal disturbance is still incomplete. But it 
seems to be established that these compounds 
are not the sole agents of tubular damage. 
Nefhrotoxic substances arising from 
injured tissue 
Many observers maintain that in cases of 
crush injury and other forms of muscle 
trauma, a toxic substance, or substances, may 
be liberated from the damaged tissue into the 
blood stream, effecting renal damage in the 
course of its excretion.21’23’30’45’72’90 It is a 
known fact that some products of tissue break- 
down have toxic properties, but the particular 
agent responsible for the renal damage is as 
yet unknown. One of the agents suggested on 
the basis of experiments is adenosine tri- 
phosphate.6’7’81 It has been shown that an ex- 
tract can be obtained from muscle which on Lower Nephron Nephrosis 
391 
injection will produce nephrotoxic signs and 
symptoms. The effect is accentuated by simul- 
taneous injection of myoglobin. The greater 
part of the activity of the extract has been 
attributed to adenosine compounds, and an 
increased concentration of these compounds 
has been demonstrated in the blood returning 
from ischemic muscle.81 By other experiments 
it has been demonstrated that a substance toxic 
to the kidney can be extracted from ischemic 
muscle but not from normal muscle.34’35 The 
toxic factor may be some early breakdown 
product of large organic molecules (perhaps 
proteins) formed only under prolonged 
anaerobic conditions. The interesting sugges- 
tion has been made that if this substance is 
liberated slowly into the bloodstream, the liver 
is able to detoxify it, so that the kidney is pro- 
tected from damage. If this hypothesis proves 
to be correct, it may account for the great var- 
iation in the occurrence of the lower nephron 
syndrome after ischemic muscle trauma, de- 
pending on (a) the degree and duration of 
muscle ischemia and (b) the functional ca- 
pacity of the liver. 
Dunn and his coworkers,30 while not ques- 
tioning that other nephrotoxic substances ob- 
tainable from damaged muscle may be impli- 
cated, suggest that uric and phosphoric acid 
possibly play a part in the produciton of the 
renal lesions. Both of these acids are liberated 
in considerable amounts from severely injured 
muscle; and both, under experimental condi- 
tions, may produce lesions in the lower 
nephron.32’54 
In a recent paper65 some proteolytic 
enzyme, set free or activated from damaged 
tissue, has been held responsible for the renal 
disturbances. In support of this hypothesis it 
was demonstrated that intraperitoneal injec- 
tions of trypsin cause kidney lesions which are 
said to resemble strikingly those in human 
cases of crush injury or burns. 
The mode of action of these various 
nephrotoxic agents released from damaged 
tissue is difficult to assess. Some may affect 
tubular cells directly; others may bring about 
primary renal vasoconstriction,21’47 and, in 
consequence, a state of anoxemia which is in- 
jurious to cells. 
Part flayed by physicochemical alterations 
of blood 
There is evidence that profound physico- 
chemical alterations of the blood occur in some 
conditions that lead to lower nephron 
nephrosis. These alterations are mainly the 
result of destruction of tissue, and of excessive 
vomiting with dehydration. As to the 
former, considerable quantities of acid me- 
tabolites, such as lactate and phosphate, may 
be released. Absorption of these substances by 
the blood reduces the alkali reserve, causes an 
elevation of inorganic phosphate, and leads to 
acidification of the urine.16 
Destruction of tissue or of blood will also 
set free intracellular potassium which has been 
locked up within cells. It is possible that the 
sudden liberation of excessive amounts may 
exert toxic effects.9’13 
Persistent vomiting leads to loss of chlorides 
and of water and to consequent upset of the 
physicochemical balance of blood and body 
fluids. This, in turn, causes diminution of kid- 
ney function. For example, it has been demon- 
strated that hypochloremia in man and ex- 
perimental animals is commonly attended by 
a fall in glomerular filtration rate and extreme 
oliguria, despite normal fluid intake.50’89 The 
renal lesions observed in human cases of alka- 
losis (resulting from excessive vomiting or 
the taking of large quantities of sodium bicar- 
bonate) are very similiar to those in cases of 
the crush syndrome, excepting for the absence 
of heme casts.55 Two such cases are included 
in the present series. 
The mechanism by which physicochemical 
imbalance of body fluid brings about renal 
disturbances is conjectural. 
Disturbances of renal blood flow. The role of 
shock 
A cardinal prerequisite for normal function 
of the kidney is adequate circulation. Any 
measure which diminishes blood flow in the 
glomerular capillaries and which reduces pres- 
sure in these capillaries will lead to decreased 
filtration;74 it will also lead to a diminished 
peritubular circulation and to reduction in 
essential oxygen supply.40 There is a growing 
belief that renal ischemia, and consequently 392 
The Military Surgeon—N ov ember, 1946 
anoxia, is of fundamental importance in the 
pathogenesis of the kidney disturbances with 
which we are concerned.20’23’49’57’69’70’77’82’83 
The famous dictum of Haldane may aptly be 
applied here: “Anoxia not only stops the ma- 
chinery but wrecks the machine.” 
The deleterious effects of inadequate circu- 
lation upon the kidney have been extensively 
studied in relation to shock. Although the 
hypotheses concerning shock are controversial, 
it seems to be established that deficit in the 
volume of circulating blood is an essential fea- 
ture.84’88 It brings about regional vascular 
constriction as a compensatory mechanism. 
Now it has been demonstrated that blood flow 
in the kidney can vary independently of the 
general circulation.51’75’76’84 For example, 
when in shock the total circulating volume 
decreases to approximately one-half the normal 
value, the flow through the kidney decreases 
to one-tenth, one-twentieth, or even less.75 
The immediate effect is marked depression of 
urinary output, i.e. oliguria or anuria.51’68 The 
duration of ischemia is an important factor in 
determining renal injury.21’16,77,84 Thus van 
Slyke and his associates divide the effects of 
ischemia into three stages, depending on its 
duration: (I ) reduced kidney function without 
damage to nephrons; (2) reversible damage to 
the nephrons, and (3) irreversible damage, 
with subsequent death from uremia. All three 
stages may occur in man.84 
The main factor held responsible for injury 
to the nephrons is renal anoxia.57’77’82 
Mechanism of oliguria 
Three principal hypotheses have been ad- 
vanced to account for the diminution in uri- 
nary output characteristic of lower nephron 
nephrosis: (I) Shut-down in renal circulation; 
(2) obstruction of tubules, and (3) unselec- 
tive reabsorption of glomerular filtrate. The 
first of these hypotheses has already been re- 
ferred to in the preceding sections; hence dis- 
cussions will be confined to the two remaining. 
Obstruction of tubules. That mechanical 
blockage of tubules with debris and heme 
masses is the deciding factor for oliguria and 
renal failure is the view put forth by Baker 
and Dodds4 and accepted by others.67’86 Most 
investigators, however, do not share this 
hypothesis.3'12’16’26’28’30’40-46-56 Histologically, in 
many, perhaps in most, cases the number of 
tubules obstructed seems inadequate to cause 
renal impairment. To be sure, kidneys of 
lower nephron disease have not been studied in 
serial sections, and no quantitative determina- 
tion of the extent of blockage has been made. 
Oliver67 has recently stated that blockage be- 
comes more evident when nephrons are dis- 
sected out in their entirety than in the usual 
sections. Nevertheless, on morphologic grounds 
the blockage hypothesis of oliguria is opposed 
(a) by finding little if any dilatation of the 
glomerular spaces and upper segments such as 
is to be expected when obstruction is com- 
plete; (b) by the fact that in many cases heme 
casts are scanty, and that they may even be 
absent.28 On functional grounds, the criticism 
has been made that if blockage were the only 
factor concerned in the mechanism of oliguria, 
such urine as is passed should be of approxi- 
mately normal composition, since it would be 
elaborated by unobstructed and presumably 
normal nephrons.15’16 But, usually analysis has 
shown that the urine has a low specific gravity, 
indicative of inadequate concentration of the 
glomerular fluid, and an abnormal composi- 
tion. Some investigators regard tubular ob- 
struction not as the cause but as the conse- 
quence of diminished renal function.12’40 It 
seems fair to conclude that mechanical block- 
age, while it may be a contributing factor, is 
in most cases not the primary cause of oliguria. 
Uns elective reabsorftion of glomerular fil- 
trate. Normally the renal tubules concentrate 
and otherwise modify daily about 150 liters of 
glomerular filtrate to 1.5 liters of urine 
through a process of selective reabsorption of 
water and of certain solutes. When the tubules 
become injured beyond a certain degree, their 
permeability is altered, and reabsorption be- 
comes unselective. In other words, injury or 
death of tubular epithelium allows some of the 
unabsorbed glomerular filtrate to diffuse back 
into the blood of the peritubular capillaries. 
Any urine passed is relatively small in amount, 
of low specific gravity, and of abnormal 
composition. 
Dunn and his coworkers were perhaps the Lower Nephron Nephrosis 
393 
first to account for experimental oliguria on 
the grounds of tubular damage.31’32 Richards 
and his associates,73 by direct microscopic ob- 
servation of the living frog’s kidney, confirmed 
Dunn’s views. After poisoning frogs with a 
suitable dose of bichloride of mercury, and 
then examining the exposed kidneys of the 
living animals, it was found that glomerular 
circulation and filtration were extraordinarily 
active, and that the composition of the filtrate 
was approximately normal. But despite filtra- 
tion of excessive amounts of fluid, no urine 
issued from the ureters: there was complete 
anuria. The only possible explanation is that 
because of loss of permeability (of the dead 
cells in the proximal tubules), the osmotic 
pressure of the blood in the peritubular capil- 
laries is able to draw back all or nearly all of 
the glomerular filtrate. These experiments 
strengthen the hypothesis that tubular injury 
causes oliguria or anuria in lower nephron 
nephrosis. It will be recalled that portions of 
the lower segments show structural evidence 
of injury which varies from barely perceptible 
degeneration to frank necrosis* There can bf 
little doubt that in the necrotic portion selec- 
tive permeability of the tubular wall i; 
abolished. It seems proper to assume that in 
earlier and less severe stages of the retrograde 
changes there also are alterations in cellular 
permeability. Unfortunately, the methods of 
the present-day do not permit us to determine 
such alterations. Perhaps it requires less exten- 
sive damage in the lower than in the upper 
segment to bring about oliguria, for when the 
glomerular filtrate reaches the lower segment 
it has already been considerably concentrated 
and modified. 
Dunn and his associates30 and others ascribe 
the oliguria of lower nephron disease to leakage 
of glomerular filtrate through the damaged 
tubular walls. As Bywaters and Dible16 state, 
this hypothesis fits best with histologic evidence. 
With these views we agree. 
The role of extrarenal factors 
It is generally believed that extrarenal fac- 
tors may bring about a form of renal insuf- 
ficiency which cannot be accounted for by 
morphologic changes in the kidneys. The main 
clinical manifestations are, oliguria (or 
anuria) and azotemia. Among conditions in 
which extrarenal factors are held to operate 
are: excessive vomiting, dehydration, burns, 
profuse hemorrhage, trauma (including opera- 
tive trauma) and shock.5’39 But these are pre- 
cisely the conditions which may lead to the 
lower nephron syndrome. The histopathologic 
changes characteristic of this syndrome were 
until recent years unfamiliar. In discussing 
the pathogenesis it has been shown that several 
of the causal factors affect the kidney directly, 
either through vascular constriction with re- 
sulting ischemia and selective anoxia, or 
through primary injury of the lower segments. 
It may well be that physicochemical im- 
balance of body fluids is of importance in 
bringing about renal failure.40 But it seems 
necessary critically to reappraise the sig- 
nificance of various “extrarenal” factors in 
their relation to disturbances of kidney func- 
tions. In many cases, at least, renal insuf- 
ficiency may be directly traced to faults within 
the kidney itself.57 
Summary 
( i ) A syndrome characterized by oliguria 
(or anuria), heme pigment excretion, 
azotemia, hypertension, and uremia may de- 
velop in a variety of conditions associated with 
destruction of tissue (especially muscle) or 
intravascular hemolysis. Conditions in which 
this syndrome has been observed include: 
crushing injury and other forms of trauma to 
muscle, nontraumatic muscular ischemia, 
uteroplacental damage (and eclampsia), burns, 
transfusion with incompatible blood, black- 
water fever, and other types of intravascular 
hemolysis, heat stroke, sulfonamide intoxica- 
tion, alkalosis, mushroom poisoning, and some 
other kinds of poisoning with vegetable and 
chemical agents. The mortality is high; death 
usually occurs within 10 days. 
This syndrome has been the most frequent 
form of fatal kidney disorder encountered 
among military personnel during the war. 
The present communication is based upon the 
study of records and material from 538 fatal 
cases received at the Army Institute of 
Pathology. The Military Surgeon—November, 1946 
(2) Specific lesions occur in the kidney. 
The essential changes are selectively restricted 
to the lower segments of the nephrons, and 
comprise: focal degeneration or necrosis, 
presence of heme casts, secondary inflamma- 
tory reactions in the surrounding stroma, and 
thrombosis of thin walled veins. The term 
lower nephron nephrosis is suggested as de- 
scriptive of the location and nature of the 
morphologic changes. 
(3) The pathogenesis of the lesions has 
not yet been established. It is probable that 
several factors are concerned in combination. 
Among those implicated are: degradation 
products of myoglobin and hemoglobin, 
products of tissue breakdown, physiochemical 
alteration of blood and body fluids, shock and 
disturbances of renal blood flow resulting in 
ischemia of the kidney and anuria. 
(4) The relation of symptoms to lesions 
is likewise incompletely known. It is probable 
that resorption of glomerular filtrate in in- 
jured segments plays a part in the development 
of renal insufficiency. But to a varying degree, 
at least, three other factors may also be of 
importance: mechanical blockage of the 
tubular lumens by casts, inadequate glomerular 
filtration, and disturbances of electrolyte bal- 
ance of body fluids. 
(5) It is apparent that present day informa- 
tion concerning this important renal syndrome 
is inadequate. It offers a promising field for 
clinical, morphological and experimental 
investigation. 
4 Baker, S. L., and Dodds, E. C.: Obstruction of 
the renal tubules during excretion of haemoglobin. 
Brit. J. Exper. Path. 6:247-260, 1925. 
5 Bell, E. T.; Renal Diseases. Lea and Febiger, 
Philadelphia, 1946. 
4 Green, H. N.: Shock producing factor(s) from 
striated muscle; I. Isolation and biological properties. 
Lancet, 2:147-153, 1943- 
7 Bielschowsky, M., and Green, H. N.: Shock 
producing factor(s) from striated muscle; II. Frac- 
tionation, chemical properties and effective doses. 
Lancet, 2:153-155, 1943- 
1 Bing, R. J.: The effect of hemoglobin and re- 
lated pigments on renal functions of the normal and 
acidotic dog. Bull. Johns Hopkins Hosp. 74:161-176, 
1944. 
9 Birnbaum, W.: Crushing injuries with renal 
failure. California & West. Med. 64:15-18, 1946. 
10 Blalock, A., and Duncan, G. W.; Traumatic 
shock, consideration of several types of injury. Surg. 
Gynec. & Obst. 75:401-409, 1942. 
“Bott, P. A., and Richards, A. N.; Passage of 
protein molecules through glomerular membranes. J. 
Biol. Chem, 141:291-310, 1941- 
“Brass, K.: t)ber ein charakteristisches Syndrom 
bei akuter schwerer Myolyse. Frankfurt. Ztschr. f. 
Path. 58:387-442, 1944. 
1S Bywaters, E. G. L.: War medicine series; crush- 
ing injury. Brit. M. J. 2:643-646, 1942. 
14 Bywaters, E. G. L., and others.; Discussion on 
effects on kidney of trauma to parts other than 
urinary tract, including crush syndrome. Proc. Roy. 
Soc. Med. 35:321-339, 1942. 
15 Bywaters, E. G. L., and Beall, D.; Crush in- 
juries with impairment of renal function. Brit. M. J. 
1:427-432, 1941. 
14 Bywaters, E. G. L., and Dible, J. H.: Renal 
lesion in traumatic anuria. J. Path. & Bact. 54:111- 
120, 1942. 
17 Bywaters, E. G. L., Delory, G. E., Rimington, 
C., and Smiles, J.: Myohaemoglobin in urine of air 
raid casualties with crushing injury. Biochem. J. 
35:1164-1168, 1941. 
14 By waters, E. G. L., and Dible, J. H.: Acute 
paralytic myohaemoglobinuria in man. J. Path. & 
Bact. 55;7-i5> W43- 
19 Bywaters, E. G. L., and Stead, J. K.: Production 
of renal failure following injection of solutions con- 
taining myohaemoglobin. Quart. J. Exper. Physiol. 
33 :53"7°> 1944- 
90 Corcoran, A. C., and Page, I. H.: Effects of 
hypotension due to hemorrhage and of blood trans- 
fusion on renal function in dogs, J. Exper. Med. 
78:205-224, 1943. 
S1 Corcoran, A. C., and Page, I. H.: Post-traumatic 
renal injury; summary of experimental observations. 
Arch. Surg. 51:93-101, 1945- 
22 Corcoran, A. C., and Page, I. H.: Renal damage 
from ferroheme pigments, myoglobin, hemoglobin, 
Acknowledgment 
I wish to express grateful appreciation to 
the staff of the Reference Division, Army 
Medical Library, for bibliographic assistance. 
References 
1 Anderson, W. A. D., Morrison, D. B., and 
Williams, E. F.: Pathologic changes following in- 
jections of ferrihemate (hematin) in dogs. Arch. 
Path. 33:589-602, 1942. 
* Angevine, D. M., and Harman, J. W.: An 
analysis of a series of autopsies with hemoglobinuric 
nephrosis (Personal communication). 
* Ayer, G. D., and Gauld, A. G.: Uremia following 
blood transfusion, nature and significance of renal 
changes. Arch. Path. 33 =513~53*» *94*- Lower Nephron Nephrosis 
395 
hematin. Texas Rep. Biol. & Med. 3:528-544, 1945. 
23 Corcoran, A. C., and Page, I. H.: Genesis of 
crush syndrome. J. Lab. & Clin. Med. 30:351-352, 
1945. 
24 Damon, S. R.: Food infections and Food Intoxi- 
cations. Baltimore, Williams and Wilkins Co., 1928. 
25 Daniels, W. B., Leonard, B. W., and Holtzman, 
S.: Renal insufficiency following transfusion; report 
of 13 cases. J. A. M. A. 116:1208-1215, 1941. 
28 DeGowin, E. L., Warner, E. D., and Randall, 
W. L.: Renal insufficiency from blood transfusion, 
anatomic changes in man compared with those in 
dogs with experimental hemoglobinuria. Arch. Int. 
Med. 61:609-630, 1938. 
27 Darmady, E. M., and others.: Traumatic 
uraemia, reports on 8 cases. Lancet, 2:809-812, 1944. 
28 DeNavasquez, S.: Excretion of haemoglobin, 
with special reference to “transfusion” kidney, J. 
Path. & Bact. 51:413-425, 1940. 
20 Douglas, J. W. B.: Incidence of signs of renal 
injury following prolonged burial under debris in 
an unselected series of 764 air-raid casualties admitted 
to hospital. Brit. J. Urol. 17:142-147, 1945. 
“Dunn, J. S., Gillespie, M., and Niven, J. S. F.: 
Renal lesions in 2 cases of crush syndrome. Lancet, 
2 :549'552> i94i. 
31 Dunn, J. S., Haworth, A.,'and Jones, N. A.: 
Pathology of oxalate nephritis. J. Path. & Bact. 
27 :299-31 8, 1924. 
32 Dunn, J. S., and Poison, C. J.: Experimental 
uric acid nephritis. J. Path. & Bact. 29: 337-352, 
1926. 
33 Edwards, J. G.: The renal tubule (nephron) as 
affected by mecury. Am. J. Path. 18:10x1-1027, 
1942. 
34 Eggleton, M. G., Richardson, K. C., Schild, H. 
O., and Winton, F. R.: Renal damage due to crush 
injury and ischaemia of limbs of anaesthetized dog. 
Quart. J. Exper. Physiol. 32:89-106, 1943. 
35 Eggleton, M.: Crush kidney syndrome in the 
cat. Lancet, 2:208-210, 1944. 
38 Erb, I. H., Morgan, E. M., and Farmer, A. W.; 
Pathology of burns, pathologic picture as revealed 
at autopsy in series of 61 fatal cases treated at 
Hospital for Sick Children, Toronto, Canada. Ann. 
Surg. 117:234-255, 1943. 
37 Fahr, Th. in Henke, F., and Lubarsch, O.: 
Handb. d. spez. path. Anat. u. Hist., Berlin, J. 
Springer, v. 6, pt. 1 & 2, 1925, 1934. 
33 Fairley, N. H.: Fate of extracorpuscular circu- 
lating haemoglobin. Brit. M. J. 2:213-217, 1940. 
39 Fishberg, A. M.: Hypertension and nephritis. 
Philadelphia, 4th Ed. Lea and Febiger, 1939. 
40 Foy, H., Altmann, A., Barnes, H. D., and 
Kondi, A.: Anuria, with special reference to renal 
failure in blackwater fever, incompatible transfu- 
sions, and crush injuries. Tr. Roy. Soc. Trop. Med. 
& Hyg, 36:197-238, 1943- 
41 Foy, H., Gluckman, J., and Kondi, A.; Pigment 
metabolism and renal failure in acute sulphonamide 
haemolysis resembling blackwater fever. Tr. Roy. 
Soc. Med. & Hyg. 37:303-319, 1944. 
42 French, A. J.: Hypersensitivity in the patho- 
genesis of the histopathologic changes associated with 
sulfonamide chemotherapy. Am. J. Path. 22:679-701, 
1946. 
43 Gilligan, D. R., and Blumgart, H. L.; March 
hemoglobinuria; studies of clinical characteristics, 
blood metabolism and mechanism, with observations 
on 3 new cases, and review of literature. Medicine. 
20:341-395, 1941, 
44 Gilligan, D. R., Altschule, M. D., and Katersky, 
E. M.: Studies of hemoglobinemia and hemoglo- 
binuria produced in man by intravenous injection of 
hemoglobin solutions. J. Clin. Investigation. 20:177- 
187, 1941. 
45Gollan, K. R., Lemberg, R., and Money, R. A.: 
Observations upon “crush injury” syndrome and 
Volkmann’s contracture associated with severe brain 
injury and hyperthermia. Med. J. Australia, 2:212- 
214, 1945. 
46 Goodpastor, W. E., Levenson, S. M., and others.; 
Clinical and pathologic study of kidney in patients 
with thermal bums. Surg. Gynec. & Obst. 82:652- 
670, 1946. 
41 Goormaghtigh, N.: Vascular and circulatory 
changes in renal cortex in anuric crush-syndrome. 
Proc. Soc. Exper. Biol. & Med. 59:303-305, 1945. 
44 Gruber, G. B. in Henke, F., and Lubarsch, O. 
Handb. d. spez. path. Anat. u. Hist., Berlin, J. 
Springer, v. 6, pt. 2, 1934 (refs, to hemoglobinuria 
in poisoning). 
49 Keele, C. A., and Slome, D.: Renal blood-flow 
in experimental “crush syndrome.” Brit. J. Exper. 
Path. 26:151-159, 1945. 
80 Landis, E. M., Elsom, K. A., Bott, P. A., and 
Shiels, E.: Observations on sodium chloride restric- 
tion and urea clearance in renal insufficiency. J. 
Clin. Investigation. 14:525-541, 1935. 
11 Lauson, H. D., Bradley, S. E., and Cournand, 
A.: Renal circulation in shock. J. Clin. Investigation. 
23:381-402, 1944. 
59 Luetscher, J. A., Jr., and Blackman, S. S., Jr.: 
Severe injury to kidneys and brain following 
sulfathiazole administration; high serum sodium and 
chloride levels and persistent cerebral damage. Ann. 
Int. Med. 18:741-756, 1943. 
58 Lund, C. C., Green, R. W., Taylor, F. H. L., and 
Levenson, S. M.: Burns, collective review. Surg. 
Gynec. & Obst. (Intern, Abstr. of Surg.) 82:443- 
478, 1946. 
54 McFarlane, D.: Experimental phosphate 
nephritis in rat. J. Path. & Bact. 52:17-24, 1941. 
85 McLetchie, N. G. B.: Renal lesions in case of 
excessive vomiting. J. Path. & Bact. 55:17-22, 1943. 
88 Maegraith, B. G., and Findlay, G, M.; Oliguria 
in blackwater fever. Lancet. 2:403-404, 1944. 
w Maegraith, B. G., Havard, R. E,, and Parsons, 396 
The Military Surgeon—November, 1946 
D. S.: Renal syndrome of wide distribution induced 
possibly by renal anoxia. Lancet. 2:293-296, 1945. 
68 Malamud, N., Haymaker, W., and Custer, R. 
P.: Heat stroke. Military Surgeon. 5:397-449, 
1946. 
59 Mallory, T.: Personal communication. 
60 Mann, F.: Peripheral origin of surgical shock. 
Johns Hopkins Hosp. Bull. 25:205-212, 1914. 
61 Mason, J. B., and Mann, F. C.: Effect of hemo- 
globin on volume of kidney. Am. J. Physiol. 98:181- 
185, 1931. 
62 Millikan, G. A.; Muscle hemoglobin. Physiol. 
Rev. 19:503-523, 1939. 
93 Minami, S.: Uber Nierenveranderungen nach 
Verschuttung. Virchows Arch. f. path. Anat. 245: 
247-267, 1923. 
63 Minett, F. C.: Haemoglobinurias and myoglo- 
binurias of animals. Proc. Roy. Soc. of Med. 28: 
672-678, 1935. 
85 Mirsky, I. A., and Freis, E. D.; Renal and 
hepatic injury in trypsin “shock.” Proc. Soc. Exper. 
Biol. & Med. 57:278-279, 1944. 
68 Moon, V. H.: Shock: Its Dynamics, Occurrence 
and Management. Philadelphia, Lea and Febiger, 
1942. 
87 Oliver, J.: New directions in renal morphology: 
a method, its results and it future. Harvey Lectures, 
40:102-155, 1945. 
88 Olson, W. H., Walker, L., and Necheles, H.; 
Study of anuria in experimental shock. Proc. Soc. 
Exper. Biol, and Med. 56:64-67, 1944. 
89 Page, I. H.: Occurrence of vasoconstrictor sub- 
stance in blood during shock induced by trauma, 
hemorrhage and burns. Am. J. Physiol. 139:386- 
398, 1943- 
70 Page, I. H., and Abell, R. G.: State of vessels of 
mesentery in shock produced by constricting limbs 
and behavior of vessels following hemorrhage. J. 
Exper. Med. 27:215-232, 1943. 
Parsons, C. G.: Traumatic uraemia. Brit. M. J. 
1 ;i 80-182, 1945. 
72 Paxson, N. F., Golub, L. J., and Hunter, R. 
M.: Crush syndrome in obstetrics and gynecology. 
J. A. M. A. 131:500-504, 1946. 
73 Richards, A. N.; Direct observations of change in 
function of the renal tubule caused by certain poisons. 
Tr. A. Am. Physicians. 44:64-67, 1929. 
4 Richards, A. N.; Croonian lecture: processes of 
urine formation. Proc. Roy. Soc., London, s. B, 126: 
398-432, 1938. 
75 Richards, D. W., Jr.: Circulation in traumatic 
shock in man. Harvey lecture, February 17, 1944. 
Bull. New York Acad. Med. 20:363-393, 1944. 
78 Scarff, R. W., and Keele, C. A.: Effects of 
temporary occlusion of renal circulation in rabbit. 
Brit. J. Exper. Path. 24:147-149, 1943. 
77 Selkurt, E. E.: Changes in renal clearances 
following complete ischemia of the kidney. Federa- 
tion Proc. 4; 63, 1945. 
78 Shorr, E., Zweifach, B. W., and Furchgott, R. 
F.: On occurrence, sites and modes of origin and 
destruction, of principles affecting compensatory 
vascular mechanisms in experimental shock. Science. 
102:489-498, 1945. 
79 Sobin, S. S., Aronberg, L. M., and Rolnick, H. 
C.: Nature of renal lesion with sulfonamides and 
its prevention with urea. Am. J. Path. 19:211-223, 
1943- 
89 Starr, I. Jr.: Production of albuminuria by renal 
vasoconstriction in animals and in man. J. Exper. 
Med. 43:31-51, 1926. 
81 Stoner, H. B., and Green, H. N.: Adenosine 
compounds and phosphates in blood of shocked 
rabbits. J. Path. & Bact. 66:343-354, 1944. 
82 Tomb, J. W.: Anuria and anoxia. J. Trop. Med. 
45:125, 1942. 
83 Tomb, J. W.; Crush injury and anoxia. New 
Zealand M. J. 42:75-77, 1943. 
84 Van Slyke, D. D., Phillips, R. A., Hamilton, 
P. B., Archibald, R. M., Dole, V. P., and Emerson, 
K., Jr.: Effect of shock on the kidney. Tr. A. Am. 
Physicians. 58:119-128, 1944. 
85 Vander Veer, J. B., and Farley, D. L.; Mush- 
room poisoning (mycetismus) ; report of 4 cases. 
Arch. Int. Med. 55:773-791, 1935. 
86 Wakeman, A. M., and others.: Metabolism and 
treatment of blackwater fever. Am. J. Trop. Med. 
12:407-439, 1932. 
8' Walker, A. M., Bott, P. A., Oliver, J., and 
MacDowell, M. C.: Collection and analysis of fluid 
from single nephrons of mammalian kidney. Am. J. 
Physiol. 134:580-595, 1941. 
88 Wiggers, C. J.: Present status of shock problem. 
Physiol. Rev. 22:74-123, 1942. 
89 Wilkinson, B. M., and McCance, R. A.; Secre- 
tion of urine in rabbits during experimental salt de- 
ficiency. Quart. J. Exper. Physiol. 30:249-261, 1940. 
"Young, J.; Renal failure after utero-placental 
damage. Brit. M. J. 2:715-718, 1942. 
91 Yuile, C. L.: Hemoglobinuria. Physiol. Rev. 
22 :i9-3i, 1942. 
92 Yuile, C. L., Gold, M. A., and Hinds, E. G.; 
Hemoglobin precipitation in renal tubules; a study of 
its causes and effects. J. Exper. Med. 82:361-374, 
1945. HEAT STROKE 
A CLINICO-PATHOLOGIC STUDY OF 125 FATAL CASES*f 
By NATHAN MALAMUD, M.D., MAJOR WEBB HAYMAKER, Medical Corps and 
LIEUTENANT COLONEL R. PHILIP CUSTER, Medical Corps 
(With fifty-eight illustrations) 
Introduction 
ushered in by premonitory symptoms. Its dis- 
tinguishing features are defective sweating, 
delirium or coma, convulsive seizures, cir- 
culatory collapse, and a temperature of io6°F. 
or higher. When the temperature is between 
103 and io6°F. the condition is spoken of as 
borderline hyperpyrexia. Although each of 
these categories may justifiably be regarded as 
a separate clinical entity, it is recognized that 
heat cramps may merge into heat exhaustion, 
and heat exhaustion into heat stroke. 
“Sunstroke ” a diagnosis rendered on some 
of the cases described in this paper, has been re- 
garded by certain authors (e.g., Manson32) 
as a distinct entity. Sunlight may contribute 
a heat load of considerable magnitude 
(Blum10’11), but it is extremely doubtful that 
it plays a specific role in causing “stroke.” 
Contrary to a rather prevalent belief, sunlight 
penetrates only a short distance into the body 
and does not directly affect the brain or 
meninges (Blum10). The experiments of 
Aron5 on monkeys show that heat stroke occurs 
not when sunlight impinges on the head alone, 
but when the whole body is exposed, indicating 
that the precipitating cause of the “stroke” is 
the total heat load, not the specific action of 
sunlight on the head. 
This paper, based on a study of material 
received by the Army Institute of Pathology 
from military hospitals throughout the United 
States, is concerned primarily with heat stroke. 
Of the 125 fatal cases with which it deals 
there were 18 in which the temperature on 
admission to hospital was below io6°F., indi- 
cating that they belonged in the categories of 
heat exhaustion and borderline hyperpyrexia. 
In 9 of these, however, the temperature sub- 
sequently rose to hyperthermic heights, some- 
times terminally. In 7 others a temperature of 
from 104 to I05°F. was recorded, and in 
the other 2, in which the course was of approxi- 
Increased attention has been directed to 
the harmful effects of heat during the 
recent war because of the numbers of 
troops suffering illness or death from exposure 
to unusual climatic factors.11 Heat disability 
falls into three categories: heat cramps, heat 
exhaustion, and heat stroke. Heat cramps, 
brought about mainly by a loss of body salt 
through sweating, is characterized by painful 
contractures of voluntary muscles, especially 
those of the extremities and the abdominal 
wall; the temperature is normal or slightly 
subnormal. Heat exhaustion covers a wide 
range of signs and symptoms. The temperature 
may be normal or subnormal but usually is 
slightly raised, and may reach ioi°F. Ladell, 
Waterlow and Hudson,31 whose studies were 
carried on in Southern Iraq, observed that 
there were two types of heat exhaustion. One, 
occurring in the hottest weeks, was manifested 
chiefly by giddiness, anorexia, headache, 
cramps, vomiting, sweating, syncope on stand- 
ing, evidences of dehydration, and oliguria. 
The other, occurring in the second half of the 
summer, was characterized by dizziness, ex- 
haustion, anorexia, insomnia, dyspnea, de- 
creased sweating, and polyuria, and frequently 
by “prickly heat.” The first type they ascribed 
chiefly to salt depletion, the second to “a break- 
down of the defense mechanism against heat,” 
especially the mechanism of sweating. Heat 
stroke (or heat hyperpyrexia) usually comes 
on with dramatic suddenness but sometimes is 
* From the Army Institute of Pathology, Wash- 
ington, D.C. 
f Read before the meeting of the American Asso- 
ciation of Neuropathologists, San Francisco, June 26, 
1946. 
f An outline of the historical aspects of heat stroke 
is to found in the work of Wakefield and Hall.58 398 
The Military Surgeon—Novembery 1946 
mately 12 hours’ duration, the temperatures 
were 97 and ioi°F. respectively. The entire 
group has been analyzed as a single unit. It 
should be emphasized that none of our cases 
were characterized by dehydration and hemo- 
concentration. On the contrary, the patients 
appeared to be normal in these respects and 
some were actually hydremic. 
Material and Methods 
The material consisted of clinical records 
and autopsy protocols, together with fixed 
tissues and stained sections from 190 cases of 
heat stroke. Forty-three were not utilized for 
the Prussian Army between 1895 and 1904 
there were 449 cases with a mortality of 11.6 
per cent; Wakefield and Hall,58 who cite 
2,049 cases with a mortality of 0.07 per cent 
in the United States Navy between the years 
1911 and 1926; and Wallace,59 who during; 
one summer at Keesler Field, Mississippi, ob- 
served 99 cases of which 4 were severe. The 
present series of 125 apparently comprises the 
largest number of instances of fatal heat stroke 
studied, but its ratio to the group in which re- 
covery followed heat disability in the United 
States Army is unknown to us. 
In civilian life heat stroke has, as a rule, been 
reported in individuals of advanced age, who 
frequently had cardiovascular and other dis- 
orders. Our material, on the other hand, is 
from the military age group (Table I), and 
includes only those cases in which post-mortem 
examination revealed no physical abnormality 
which could not be attributed to heat stroke. 
Of the 125 soldiers who sustained fatal heat 
stroke, 120 were of the white race, 4 Negro 
and 1 Indian. They ranged in military rank 
from privates to captains, the ratio of enlisted 
men to officers being approximately 8:1. 
The disorder was most often reported in 
Age Incidence of 125 Fatal Cases of Heat Stroke 
Table I 
Age 18-25 
26-30 
31-38 
39-41 
Unknown 
No. of Cases. 51 
32 
35 
4 
3 
the study because of inadequate clinical in- 
formation, and 22 were rejected because of 
significant intercurrent disease, particularly 
coronary sclerosis. The post-mortem examina- 
tions were performed by Army medical officers 
whose cooperation has made this study possible. 
Since the histories in some protocols were un- 
duly brief, the statistical value of clinical data 
is to be regarded as only approximate. 
Paraffin sections of all tissues stained with 
hematoxylin and eosin were available. Addi- 
tional slides were prepared at the Army Insti- 
tute of Pathology. Sections of heart from 70 
cases were stained by the Bodian method, and 
for purposes of control a number were pre- 
pared with iron hematoxylin and fat stains. 
For the study of the central nervous system 
paraffin sections were stained routinely by 
cresyl violet, and in certain instances frozen 
sections were stained for fat by scarlet red, for 
myelin by the Spielmeyer method, and for axis 
cylinders by the Bodian method. Celloidin 
embedding was done in selected cases. 
Incidence 
A general idea of the incidence of heat 
disability in armed forces may be gained from 
three sources: Steinhausen,62 who states that in 
Table II 
Location of Camps Where Fatal Heat Stroke 
Occurred, and the Respective Incidence 
State 
No. of Cases 
T exas 
28 
Georgia 
Mississippi 
14 
Florida 
12 
Virginia 
11 
Louisiana 
California 
9 
Alabama 
7 
Missouri 
5 
North Carolina 
Kentucky 
2 
Arkansas 
2 
Kansas 
Tennessee 
1 
Oklahoma 
South Carolina 
Total 
125 Heat Stroke 
399 
and around military camps in the southern 
United States. The incidence in relation to 
location of the camps is shown in Table II. 
While the distribution may be proportional to 
the number of troops in the respective states, it 
probably also bears a relationship to climatic 
conditions in these geographic areas. Heat 
stroke always took place from May to Sep- 
tember inclusive, with the peak incidence in 
July. The highest number of cases seen in a 
year was in 1943 when, presumably, the 
largest number of troops were undergoing in- 
tensive training (Table III). 
Etiologic Factors 
Heat as the direct etiologic factor in heat 
stroke has been established through clinical 
studies (Adolph,1 Bazett et al./ Bean and 
Eichna,7 Castellan!,12 Weiner02) and has been 
confirmed experimentally in animals (Hall and 
Wakefield,23 Hartman and Major,24 Marsh33). 
The exact environmental conditions of tem- 
perature, humidity, wind velocity, and move- 
ment of air, as well as accessory factors such 
as strenuous muscular exercise and unsuitable 
clothing are well known. In 118 of our 125 
cases there was a history of exposure to sun 
under conditions of high environmental tem- 
perature; in 3 heat stroke occurred in pursu- 
ance of kitchen or ward duty, while in the 
remaining 4 cases no information bearing on 
environmental factors is available. The great 
majority of the patients were engaged in some 
form of military exercise, such as long marches, 
Table III 
Incidence of Fatal Heat Stroke by 
Month and Year 
Month 
No. of 
Cases 
Year 
No. of 
Cases 
May 
5 
1941 
1 
June 
• • 23 
1942 
■ • 33 
July 
59 
1943 
.. 71 
August 
29 
1944 
20 
September... 
9 
— 
Total 
.. 125 
Total.... 
.. 125 
drill, or target practice. Usually the “stroke” 
occurred during activity but sometimes it set 
in hours later. Data on the weather conditions 
at the camps at which our cases occurred and 
the conclusions as to the etiologic significance 
of dry and wet bulb temperatures, wind veloc- 
ity, and other factors are being presented in 
a separate paper (Schickele48). 
Lack of acclimatization has long been re- 
garded as one of the chief factors leading to 
heat stroke. In our series it seems to have 
played an important role (Table IV). Approxi- 
mately one-fourth of the men had been in their 
respective camps for less than 2 weeks and 
about one-half for less than 2 months when 
heat stroke occurred. The majority had not 
previously been in any other camp in the 
South, nor more than 2 months in the Army. 
It should be emphasized that susceptibility to 
•heat may have been as decisive a factor as lack 
of acclimatization in the production of heat 
Table IV 
Data Dealing with the Factor of Acclimatization in 125 Fatal Cases of Heat Stroke 
Length of Duty 
in Camp at 
which Death 
Occurred 
No. 
of 
Cases 
Length of Continuous Duty in the South 
Total Length of Time in Service 
1-2 
2-4 
1-2 
2-6 
6-12 
1-2 
2-5 
1-2 
2-4 
1-2 
2-6 
6-12 
1-2 
2-5 
wks. 
wks. 
mths. 
mths. 
mths. 
yrs. 
yrs. 
wks. 
wks. 
mths. 
mths. 
mths. 
yrs. 
yrs. 
1-2 wks. 
46 
34 
4 
0 
3 
3 
1 
1 
11 
24 
0 
I 
2 
4 
4 
2-4 wks. 
22 
0 
16 
1 
3 
1 
1 
0 
0 
9 
8 
3 
1 
1 
0 
1-2 mths. 
16 
0 
0 
12 
1 
1 
2 
0 
0 
0 
10 
1 
3 
2 
0 
2-6 mths. 
33 
0 
0 
0 
*5 
6 
1 
1 
0 
0 
0 
20 
7 
3 
3 
6-12 mths. 
6 
0 
0 
0 
0 
5 
1 
0 
0 
0 
0 
0 
6 
0 
0 
1-2 yrs. 
2 
0 
0 
0 
0 
0 
1 
1 
0 
0 
0 
0 
0 
0 
2 400 
The Military Surgeon—November, 1946 
Table V 
Distribution of Cases According to the 
Duration of Illness 
per cent), and c) insidious, with premonitory 
signs and symptoms lasting sometimes for 
several days (8 per cent). The manifestations 
during the prodromal period, in the order of 
their frequency, were faintness, staggering 
gait, dizziness, headaches, nausea and vomit- 
ing, purposeless movements, muscle cramps, 
choking, difficulty in swallowing or speaking, 
numbness of extremities, drowsiness, restless- 
ness, mental confusion, dryness of the mouth, 
excessive thirst, anorexia, and diarrhea. 
Data on the duration of illness reveal that 
death occurred in less than 24 hours in ap- 
proximately 70 per cent of the cases, and in 
from 1 to 12 days in the remaining 30 per 
cent (Table V). The statement frequently 
made in the literature, that if patients with 
heat stroke survive the first 24 hours they 
usually go on to complete recovery, obviously 
needs modification. 
Variation in the clinical course was related 
to the duration of the illness (Table VI). 
When death occurred in less than 48 hours 
early coma or delirium persisted until death. 
When the illness lasted longer than 48 hours, 
the course was characterized either by early 
Duration 
No. of Cases 
Less than 24 hours 
From 24 to 48 hours 
From 2 to 12 days 
63 
12) 
15] 
(70%) 
> (30%) 
No. of Cases of Known Duration... . 
90 
No. of Cases of Unknown Duration.. 
35 
Total 
125 
stroke; however, there was a history of pre- 
vious intolerance to heat in only 5 instances. 
It is significant that most of the patients were 
somewhat overweight or actually obese. There 
are too few data to allow an evaluation of 
alcohol consumption as a predisposing factor. 
Clinical Features 
Course. There were three types of onset: 
a) acute, without apparent warning (71 per 
cent), b) relatively acute, with brief pro- 
dromal symptoms lasting minutes to hours (21 
Table VI 
Types of Clinical Syndrome and the Duration of Illness in 90 Fatal Cases of Heat Stroke 
Survival Time 
Syndrome 
Less than 
24 hrs. 
24 to 
48 hrs. 
2 to 
12 days 
Acute Onset with Early Persistent Coma or Delirium 
63 
9 
0 
Acute Onset with Early Coma or Delirium, but with Remission and Late Re- 
lapse 
O 
3 
8 
Insidious Onset with Progressive Course and Late Development of Coma... 
0 
0 
7 
Table VII 
Temperature Levels, Pulse Rates and Respiratory Rates on Admission to Hospital 
Temperature 
No. of 
Cases 
Pulse Rate 
No. of 
Cases 
Resp. Rate 
No. of 
Cases 
97 to 99 
3 
100 to 130 
•... 13 
10 to 20 
1 
99 to 103 
3 
130 to 160 
•••• 23 
20 to 30. 
4 
103 to 106 
12 
160 to 200 
.... 24 
30 to 60 
... 31 
106 to III 
Total 
... 125 
Total 
Total 
... 36 Heat Stroke 
401 
coma which tended to clear up partially until 
a terminal relapse supervened, or by slow 
progression and late coma. 
Temperature, Pulse, Respiration and 
Blood Pressure. The temperatures on ad- 
mission to hospital ranged from 99 to 111 °F. 
in all except 3 cases, the elevation being 
1060 or higher in 85 per cent (Table VII). 
(Most of the temperatures were rectal or 
axillary.) In the cases in which subnormal or 
slightly elevated temperatures were recorded, 
there was usually a subsequent rise to hyper- 
thermic levels (Fig. lA). Initially high tem- 
peratures were often reduced by ice packs and 
other cooling measures, sometimes to subnor- 
mal levels, but as a rule there was a secondary 
rise (Fig, iB) and, in cases of long duration, 
several rises (Fig. 2). Peripheral vasoconstric- 
tion produced by cold, hindering the escape of 
heat through the skin, may account for the 
initial secondary rise in temperature, but the 
subsequent fluctuations suggest a persistent dis- 
turbance in thermoregulation. 
Elevations in pulse and respiratory rates 
were usually of the same relative degree, and 
corresponded to the rise in temperature. The 
pulse rates varied from 100 to 200 per minute, 
being above 130 in approximately 80 per cent 
of the cases. The pulse was more often rapid 
Fig. i. Temperature, pulse, and respiration curves 
in two fatal cases of heat stroke. In A (AIP Acc. 
102705) the temperature fell to a subnormal level 
before hyperthermia set in; in B (AIP Acc. 116302) 
the high temperature on admission was reduced by 
therapeutic measures, but hyperthermia again de- 
veloped. 
and thready than full and bounding. The 
respiratory rate was usually increased, being 
Fig. 2. Temperature, pulse, and respiration curves in a case in which hyperthermia persisted despite treat- 
ment (AIP Acc. 120161). Frequent fluctuation of the curves was a characteristic finding in the more pro- 
tracted cases of heat stroke. 402 
The Military Surgeon—November, 1946 
Table VIII 
Blood and Pulse Pressures on Admission to Hospital in 55 Fatal Cases of Heat Stroke 
Systolic 
Diastolic 
Pulse Pressure 
Mm. Hg. 
No. of 
Cases 
Mm. Hg. 
No. of 
Cases 
Mm. Hg. 
No. of 
Cases 
40 to 80 
.. 19 
0 to 30 
.. 26 
90 to 130 
10 
80 to 100 
.. 14 
30 to 60 
.. 10 
50 to 90 
27 
100 to 130 
.. 18 
60 to 90 
.. 16 
30 to 50 
15 
130 to 150 
4 
90 to 100 
•• 3 
20 to 30 
3 
above 30 per minute in 86 per cent of the 
cases; breathing was often of the Kussmaul 
type, except in terminal stages when it gen- 
erally took on a Cheyne-Stokes character. 
In 55 cases in which the blood pressure 
readings on admission to the hospital were 
available (Table VIII) the systolic pressure 
ranged from 40 to 100 mm. Hg. in 60 per 
cent, the diastolic pressure from o to 60 mm. 
in 65 per cent, and the pulse pressure from 50 
to 130 mm. Hg. Thirty-five per cent of the 
55 had normal or slightly elevated blood 
pressure readings. 
Symptoms. A survey of the symptoms 
(Table IX) reveals that virtually all systems 
were affected. Three fundamental types of 
disturbances were evident: 1) symptoms re- 
garded as directly related to the hyperthermia, 
therefore primary, 2) symptoms characteristic 
of shock, and 3) complications arising from 
chemical and pathologic changes. 
Among the primary symptoms, those re- 
lating to the nervous system were outstanding. 
Disturbances in consciousness in the form of 
coma, stupor, or delirium were almost con- 
stant features, while convulsions, either gen- 
eralized or Jacksonian in type, occurred in 
approximately 60 per cent of the cases. Pupil- 
lary changes and exaggeration or suppression 
of tendon reflexes were common but were not 
tabulated, since in most instances they were re- 
garded as concomitants of stupor or coma. 
Damage to the central nervous system was 
manifest from the onset and persisted to the 
end; in the cases of longer duration dementia, 
aphasia, or hemiplegia indicated that the effect 
on the nervous system was probably lasting 
and irreversible. A direct relationship between 
the nervous manifestations and the degree and 
duration of hyperthermia was always evident. 
Disturbances in sweating likewise were inti- 
mately related to the hyperthermia, most 
commonly in the form of suppression; in a few 
instances, however, there was hyperhidrosis. 
Rarely was there the cold clammy skin ob- 
served in shock. In none of the cases was 
Table IX 
Signs and Symptoms in 125 Fatal Cases of Heat 
Stroke, in Order of Frequency 
Signs & Symptoms 
No. of Cases 
Coma 
... 116 
Cyanosis 
... 90 
Disturbances of Sweating 
.. . 88 
Suppression 
... 70 
Excess 
“Clammy Skin” 
... 7 
Convulsions 
... 77 
Rales 
... 58 
Incontinence 
... 48 
Vomiting 
... 47 
Bleeding (Skin & mucous membranes). 
,... 46 
Delirium or Stupor 
•• • 33 
Disturbances of Muscle Tone 
... 32 
Rigidity 
Flaccidity 
.. . 8 
“Cramps” 
4 
Pos. Kernig & Brudzinski 
16 
Pyramidal signs 
.. . 12 
Opisthotonos 
7 
Tetany 
... 7 
Diarrhea 
... 5 
Anuria 
4 
Hematuria 
••• 3 
Jaundice 
••• 3 
Decerebrate Rigidity 
2 Heat Stroke 
403 
“prickly heat” described. Defective sweating 
has long been recognized as an early mani- 
festation of heat stroke (Willcox65). Accord- 
ing to Ladell, Waterlow and Hudson,31 whose 
work was done on nonfatal cases in Southern 
Iraq, virtually all patients with heat stroke had 
a suppression of sweating, whereas in heat ex- 
haustion occurring in the hottest weeks the 
great majority (87 per cent) exhibited pro- 
fuse sweating. The suddenness of the suppres- 
sion of sweating in heat stroke and its prompt 
restoration on treatment, together with the 
absence of “prickly heat” or other condition of 
the skin which could be held accountable, led 
Ladell and his associates to believe that the first 
breakdown in heat stroke is in the sweating 
mechanism, in that part of the mechanism 
which is located in the central nervous system. 
Certain symptoms were present which in all 
respects were similar to those seen in shock. 
Foremost among these were pallor or slight 
cyanosis, vomiting, fall in blood pressure, a 
rapid and thready pulse, and shallow and sigh- 
ing respiration. In contrast to the symptoms 
resulting from hyperthermia, those of shock 
were inconstant: they occurred at any stage 
of the disorder, often were reversible, and their 
severity generally did not correspond to the 
degree of temperature elevation. Other symp- 
toms which may have been related to shock 
were those of diarrhea and oliguria, the latter 
doubtless being more common than the records 
indicated. The degree of fall in blood pressure 
was taken as the most satisfactory, although 
admittedly an undependable, criterion of the 
presence of shock (Moon,3'’38 Wright and 
Devine69). (The pulse pressures were too vari- 
able to serve as a reliable index of shock.) In 
our cases the presence or absence of shock 
rather than the degree of hyperthermia, was 
the best prognostic index since it was on this 
factor that the ultimate outcome usually de- 
pended. Thus, in cases in which survival was 
longest, shock was either successfully controlled 
at the very outset or was not appreciable until 
the terminal stages. 
Among the complications during the course 
of heat stroke were muscle cramps and tetany, 
which may be ascribed to hypochloremia and 
alkalosis respectively. Pneumonia, uremia, 
hematuria, and jaundice were observed, espe- 
cially in cases of longer duration. 
Laboratory Studies 
Blood Cells. Blood counts made during 
the early stages of heat stroke revealed in 22 
of 32 cases a leukocytosis ranging from 10,000 
to 28,950 cells per cmm.; leukocyte values in 
cases of longer duration were not listed be- 
cause of the prevalence of complicating factors 
such as pneumonia (Table X). As a rule the 
Table X 
The Blood Leukocyte Values in Cases of Heat 
Stroke Fatal in Less Than 48 Hours. Only Un- 
complicated Cases Are Included 
Duration of Illness 
W.B.C. (per cmm.) 
5 to 
10,000 
10 to 
18,000 
18 to 
28,950 
Less than 24 hrs 
10 
9 
8 
From 24 to 48 hrs.. . . 
0 
3 
2 
Total 
10 
12 
IO 
differential count was within normal limits 
when the total was not greatly altered. Leu- 
kocytosis was due largely to an increase in 
neutrophils. These figures confirm the observa- 
tions of Gauss and Meyer19 in a group of 25 
cases of heat stroke. 
Red blood cells and hemoglobin were within 
normal limits except for a few cases in which 
a decrease of both was observed, and several 
others in which the erythrocytes rose to 
polycythemic levels (Table XI). In only I of 
15 cases was there an elevation in the hemat- 
ocrit value and this was slight. Various in- 
vestigators (Heilman and Montgomery,20 
Chakravarti and Tyagi,13 Wilcocks64) have 
reported increases in viscosity, specific gravity, 
red and white blood counts and percentage of 
hemoglobin—findings which suggest that 
hemoconcentration occurs in heat stroke. How- 
ever, with the exception of leukocytosis, which 
is probably due simply to accelerated circula- 
tion, we have been unable to confirm the ex- 
istence of hemoconcentration as a regular 
feature. 404 
The Military Surgeon—TVovember, 1946 
Table XI 
Red Blood Cell, Hemoglobin and Hematocrit Values in Fatal Heat Stroke 
R.B.C. (in 
(Normal; 4 
Values 
millions) 
6 to 6.2) 
No. of 
Cases 
Hemoglobin (in %) 
(Normal; 90 to no) 
Values No. of 
Cases 
Hematocrit (in 
(Normal 
Values 
% packed cells) 
45 to 47) 
No. of 
Cases 
3.4a to 4.00. . 
3 
45 to 75 1 
35 to 45 
9 
4.00 to 4.60. . 
15 
75 to 90 15 
45 to 47 
5 
4.60 to 6.20. . 
15 
90 to 110 12 
50 
6.20 to 6.89. . 
2 
no to 130 3 
— 
— 
— 
Total 
15 
Total 
35 
Total 31 
Table XII 
Prothrombin Time, Platelet Count and Bleeding Time in 10 Fatal Cases of Heat Stroke 
Prothrombin Time (Sec.) 
Platelet Count 
Bleeding Time (Min.) 
AIP 
Duration 
111 
(Control; 14-19 Sec.) 
(Normal: 200,000-500,000) 
(Normal: 1-5 Min.) 
Acc. 
ness 
On Ad- 
Subsequent 
On Ad- 
Subsequent 
On Ad- 
Subsequent 
(Hrs.) 
mission 
Times 
mission 
Counts 
mission 
Times 
113894* 
c 
15 
— 
— 
— 
— 
— 
112746* 
6 
30 
— 
— 
— 
— 
— 
117740* 
16 
17 (4th hr.) 
120,000 
— 
— 
— 
120158 
8 c 
— 
— 
56,000 
— 
— 
— 
118077* 
12 
16 
20 (6th hr.) 
104,000 
40,000 (6th hr.) 
— 
— 
107487 
20 
— 
— 
— 
— 
3-5 
— 
107486 
22 
— 
— 
22,300 
— 
3-5 
— 
97556 
47 
— 
11 (8th hr.) 
— 
— 
— 
— 
94751 
5i 
49 (2d day) 
— 
— 
— 
— 
108 (2d day)f 
90,000 (2d day)f 
10 (2d day)f 
72 (3d day) 
70,000 (3d day) 
— . 
118078* 
16 
20 (6th hr.) 
108,000 
92,000 (6th hr.) 
— 
4.8 (2d day) 
18 (2d day) 
44,000 (2d day) 
— 
14 (3 day)§ 
31,000 (3d day)§ 
3.0 (3d day)§ 
— 
78,000 (4th day) 
— 
— 
66,000 (5th day)If 
6.0 (5th day)^f 
— 
168,000 (6th day) 
6.2 (6th day) 
— 
94,000 (7th day) 
— 
130,000 (8th day) 
6.0-8.0 
— 
178,000 (9th day) 
— 
140,000 (10th day) 
16 (12th day) 
260,000 (12th day) 
* The prothrombin and platelet studies done on these cases were reported by Wright, Reppert and Cuttino(see 
References). 
f These determinations were made six hours after parenteral administration of 5.0 mg. of vitamin K. 
§ Values shortly after transfusion of 400 cc. of whole blood. 
T[ Subsequently on the same day a transfusion of 500 cc. of whole blood was given. Heat Stroke 
405 
Coagulation Mechanism, Platelet 
counts were performed in 6 cases (Table 
XII) and all indicated a reduction in number. 
Generally the fall was progressive, but in one 
instance (Case 118078) in which survival 
was unusually long there was a return to a 
normal value before death. The prothrombin 
time, on the other hand, was within normal 
range except in one (Case 94751) in which 
it was much prolonged. The bleeding time 
reduced in number, in 3 the prothrombin time 
was normal on admission but was definitely 
prolonged during the subsequent course, and 
in 3 the bleeding time also was prolonged in 
the later stages.* Prolongation of prothrombin 
time and reduction in platelets have been ob- 
served also by Wilson and Doan67 in patients 
subjected to artificially induced fever. 
Blood Chemistry. Elevation in nonpro- 
tein nitrogen of the blood was moderate in 
Table XIII 
Non protein and Urea Nitrogen Values in 21 Fatal Cases of Heat Stroke 
AIP 
Acc. 
Duration 
of Illness 
(Hrs.) 
Blood Pressure 
Blood or 
Plasma 
Transfusion 
Nonprotein 
Nitrogen 
(Normal: 15 to 
40 mg. %) 
Urea Nitrogen 
(Normal: 
8 to 15 
mg- %) 
120160 
3 
70/0, 90/50 
+ 
— 
25-3 
101829 
7-5 
80/40, 100/60, 96/50 
0 
24.6 
14.2 
98083 
8 
150/90 
0 
53-4 
18.0 
120158 
8.5 
120/0, 110/40, 60/0 
+ 
— 
19.0 
99511 
9 
52/0, 96/50, 105/85 
+ 
40.0 
— 
98258 
11 
60/20, 110/70, 60/40 
0 
52-3 
— 
116302 
*5 
100/40 
+ 
— 
14.1 
98086 
*5 
60/20, 110/30, 90/0 
0 
57.0 
22.0 
99261 
16 
84/50, 80/60 
0 
60.0 
— 
120159 
18 
80/0, 140/80, 80/60 
0 
— 
*3-9 
114215 
*9 
90/70, 90/50, 120/70 
0 
45.0 
— 
99112 
21 
130/80 
+ 
66.0 
28.0 
97555 
26 
60/40, 110/70, 60/40 
+ 
45.0 
— 
86617 
32 
80/60 
0 
— 
17.0 
97554 
34 
136/80, 86/40 
+ 
48.0 
— 
97*48 
35 
150/110, 60/40 
0 
58.0 
— 
115686 
39 
135/65. *30/90 
0 
49.0 
— 
97556 
47 
80/50, 100/70, 135/85 
+ 
43-o 
— 
94751 
60 
125/70,108/60 
+ 
51.0 
20.0 
115309 
132 
142/30, 82/52, 100/80 
0 
48.0 
8.6 
120161 
249 
100/60, 114/60, 110/90 
+ 
130.0-229.0 
25.9-46.8 
was found to be slightly or moderately in- 
creased. The clotting time in 2 instances 
(Cases 107486 and 107487) was within nor- 
mal limits, and in one (Case 120158) was 
found to be 45 minutes; in this case it was 
noted that for many minutes after a trans- 
fusion had been given blood continued to ooze 
from the puncture wound. Most of these ob- 
servations are in accord with those of Wright, 
Reppert and Cuttino70 in 6 nonfatal cases of 
heat stroke; in all instances the platelets were 
all except one of 21 samples, with values rang- 
ing from 40 to 229 mg. per cent (Table 
XIII). Blood urea nitrogen, on the other 
hand, was less elevated, the levels being 17 to 
46.8 mg. per cent. These findings were just 
* The figures of Wright and his associates showed 
a maximum deviation from normal varying from 
30,000 to 85,000 per cmm. for platelets, from 17 to 
38 sec. for prothrombin time, and from 3 to 12 min. 
for bleeding time. 406 
The Military Surgeon—November, 1946 
as evident in cases of a few hours’ standing 
as in those of longer duration, and just as 
marked with or without blood transfusion. 
Similar findings have been reported in clinical 
cases (Chakravarti and Tyagi,13 Wallace,59 
Wilcocks64) as well as in experimental heat 
stroke in animals (Hall and Wakefield,23 
Marsh33). The changes have been attributed 
by Hall and Wakefield to rapid destruction of 
proteins resulting from hyperthermia; by 
Marsh, to renal damage, and by Chakravarti 
and Tyagi, to both. That destruction of pro- 
teins may be a contributing factor in azotemia 
is suggested by the discrepancy in nonprotein 
and urea nitrogen values. 
The validity of an actual increase in the urea 
nitrogen is open to some question, however, in- 
asmuch as Ladell, Waterlow and Hudson31 
found the blood urea of fit soldiers in Southern 
Iraq to average 47.5 mg. per cent (normal: 
20-30 mg.). In view of the fact that the daily 
temperature was more than 115°F. during the 
week their study was made, it seems unlikely 
that the nonprotein and urea nitrogen levels 
could have been that high in the normal soldier 
in southern U.S.A. It is also of interest that, 
according to their norm, Ladell, Waterlow and 
Hudson found no elevation of blood urea in 
cases of nonfatal heat stroke, but a substantial 
rise (average of 103 mg. per cent) in non- 
fatal heat exhaustion. 
As to the factor of renal damage, micro- 
scopic examination frequently revealed lower 
nephron nephrosis in early stages of heat 
stroke (see section on pathology). It would 
seem therefore that altered functional activity 
of the kidneys may also play a role in the ac- 
cumulation of nitrogenous products in the 
blood. In cases of longer duration in which 
renal damage was more severe, the increase 
in both nonprotein and urea nitrogen was much 
greater and was associated with symptoms of 
uremia. 
The C02 combining power was definitely 
reduced in the 16 cases in which the values are 
known, ranging from 19.6 to 46.0 volumes 
per cent, with an average of 33. In the absence 
of a record of the pH of the blood this change 
is difficult to evaluate; it is in keeping, how- 
ever, with reports in the literature which state 
that high blood lactic acid and low C02 com- 
bining power are characteristic findings in 
heat stroke. The fall in C02 combining power 
of the blood has been regarded as evidence 
of alkalosis due to hyperventilation, an inter- 
pretation which seems plausible in light of the 
development of symptoms of tetany in some of 
our cases, in one of which the C02 combining 
power was 19.8 volumes per cent. 
The blood chlorides ranged from 231 to 
804 mg. per cent (as NaCl), being lower 
than 600 mg. per cent in 27 of 31 cases. 
These figures are not open to evaluation, how- 
ever, since the majority of the patients re- 
ceived saline solution intravenously. In a study 
of 20 nonfatal cases of heat disability occurring 
among troops in the desert of Arizona, Rosen- 
baum4*5 was unable to demonstrate a definite 
hypochloremia but the possibility of a relative 
salt deficiency in the tissues could not be ex- 
cluded. 
The nature of these complex chemical 
changes is obscure. Similar findings have been 
reported in shock. Moon3''3*5 has emphasized 
the frequency with which electrolytic disturb- 
ances, elevation of nonprotein nitrogen, and 
reduction of alkali reserve of the blood occur in 
shock. Hemoconcentration also has been re- 
garded by him as a characteristic feature of 
shock, but other investigators have shown that 
either hemoconcentration or hemodilution may 
obtain, depending on the cause of the shock 
(Cournand et al.,14 and Richards45). Accord- 
ing to Ladell, Waterlow and Hudson,31 hemo- 
dilution, as manifested by low hemoglobin and 
diminution of chloride content of the blood, is 
usual in the later stages of nonfatal heat stroke, 
because water accumulates in the body upon 
the suppression of sweating. A tendency to- 
ward hemodilution was noted also in our cases. 
Urine. Urinalyses were rarely done, so 
that our information is limited to the fact that 
albuminuria, hematuria and casts were found 
in cases of longer duration. In every one of 
25 cases of heat stroke studied by Gauss and 
Meyer19 the urine contained hyaline and gran- 
ular casts, in 5, albumin, and in 20, pus cells. 
Spinal Fluid. The pressure was increased Heat Stroke 
407 
in 2 cases of the 24 in which spinal fluid was 
examined, and the fluid was blood-tinged in 
7- 
Electrocardiography. In the few cases 
in which electrocardiographic determinations 
were made the records suggested “toxic” 
damage to the myocardium (Fig. 3). Repeated 
July 20, 1943. During that day he had partici- 
pated in routine military exercises, and appar- 
ently was well until 5:45 p.m. when he sud- 
denly collapsed and became unconscious. In 
hospital 15 minutes later he became delirious. 
The temperature was I09°F., pulse 128, 
respiration 36, and blood pressure 80/0 mm. 
Fig. 3. Electrocardiogram of a comatose patient with a temperature of io9°F. (AIP Acc. 98083). There 
is a regular rhythm with changing pace maker. P-2 and P-3 deviate from upright to inverted; QRS com- 
plexes vary somewhat in height, and T-i is diphasic while T-z and T-3 are inverted to upright. The 
changes suggest an advanced toxic process in the myocardium. 
electrocardiograms in one case indicated that 
the abnormalities were transient. 
Illustrative Cases 
I. Acute Onset of Early and Persistent Coma 
or Delirium. 
AIP Acc. 97599. A white male, aged 37, 
was admitted to a hospital in Louisiana on 
Hg. The skin was hot and dry, and there were 
scattered petechial hemorrhages over the chest 
and abdomen. Coma soon set in. Repeated 
generalized convulsions occurred. The treat- 
ment consisted of ice packs, cold enemas, and 
exposure to an electric fan. Death occurred at 
7:53 P.M., approximately 2 hours after the 
onset. 408 
The Military Surgeon—Nov 1946 
AIP Acc, 99112. A white male, aged 22, 
was admitted to a hospital in North Carolina 
at 6:25 P.M. on June 17, 1943. During a 
four-mile hike in the afternoon, he had com- 
plained of excessive thirst, and about 4:30 p.m. 
he collapsed, became delirious and incontinent 
of urine and feces. He was in deep coma, the 
pupils were dilated and nonreactive, the skin 
hot and dry, the muscles hypertonic, and the 
knee jerks unobtainable. The temperature was 
io9°F,, pulse 120, blood pressure 120/80 
mm. Hg. Blood studies yielded the following 
results: red blood cells, 4,900,000; hemo- 
globin, 90 per cent; white blood cells, 28,050; 
nonprotein nitrogen, 66 mg., urea nitrogen, 
28 mg., chlorides, 410 mg. per cent; and C02 
combining power, 31.9 volumes per cent. 
The temperature was reduced to I02.6°F. 
by means of an ice water spray and exposure 
to electric fans. The patient also received 
1000 cc. of 5 per cent glucose in saline intra- 
venously. On being placed in an oxygen tent 
he seemed to respond slightly, the pulse being 
of good volume and respirations numbering 
24 per minute. Soon, however, he lapsed into 
a wild delirium which abated when paralde- 
hyde was administered by rectum. Intravenous 
saline solution again was given, but coma 
supervened and death occurred at 1:41 P.M., 
approximately 21 hours after onset. 
AIP Acc. 97554. A white male, aged 19, 
was admitted to a hospital in Georgia at 1:20 
P.M, on July 23, 1943. He had been on a 
march that morning when at 11 :oo a.m. he 
suddenly staggered and fell. Shortly thereafter 
he became unconscious and incontinent of 
urine and feces. His axillary temperature was 
I07°F. On examination the patient was 
comatose, the rectal temperature was I09°F., 
the blood pressure 136/80 mm. Hg., the skin 
hot and dry, the pupils pin-point and poorly 
reactive, and the tendon reflexes unobtain- 
able. Treatment consisted of ice water baths 
and enemas, alcohol sponges, and infusion with 
5 per cent glucose in saline. The temperature 
fell to 98°F. but soon rose to ioi°F. The 
limbs became spastic, and the Babinski sign 
was elicited bilaterally. Convulsive seizures oc- 
curred. The breathing was stertorous, and 
there was bloody expectoration. The patient 
then received 20 cc. of 50 per cent glucose 
and 10 cc. of calcium gluconate intravenously. 
Shortly afterward he was given a transfusion 
of 1500 cc. of plasma, but he remained coma- 
tose. Cardiac stimulants were to no avail. The 
temperature fluctuated between 101 and 
i04°F., the blood pressure fell to 86/64 mm- 
Hg., there was increasing cyanosis and the 
pulse became rapid and thready. 
Blood studies yielded the following results: 
red blood cells, 4,320,000; hemoglobin, 85 
per cent; white blood cells, 11,500 (neutro- 
phils, 66, lymphocytes, 26, monocytes, 7, 
eosinophils, 1) ; nonprotein nitrogen, 48 mg., 
chlorides, 495 mg. per cent; and C02 com- 
bining power, 45 volumes per cent. The urine 
contained 100 mg. per cent of albumin. The 
spinal fluid was within normal range. 
Death occurred at 9:05 p.m. on July 24, 
approximately 34 hours after the initial at- 
tack. 
II. Acute Onset of Early Coma or Delirium, 
with Subsequent Partial Remissiony a Pro- 
longed Coursey and Terminal Relapse 
AIP Acc. 118078. An obese white male 
aged 25, was admitted to a hospital in Florida 
on September 6, 1944. While on a march 
that day he had suddenly collapsed and be- 
come unconscious. On admission he was in 
deep coma; the skin was hot and dry, the 
rectal temperature iio°F., the pulse 140 and 
of poor quality, the respiration 40 and stertor- 
ous, and the blood pressure 80/40 mm. Hg. 
Rhonchi were heard throughout the chest. 
There was pupillary constriction, and the ten- 
don reflexes could not be elicited. Shortly 
after admission several generalized convul- 
sions occurred. 
During the first day of the illness the pa- 
tient frequently vomited bloody fluid and was 
incontinent of urine and feces; the coma lifted 
slightly, the pupils began to react and the 
tendon reflexes reappeared. However, it soon 
was evident that there was rigidity of the neck, 
mild opisthotonos, and spasticity of the right 
arm and leg. On the second day the patient 
was restless, nuchal rigidity persisted, the eyes Heat Stroke 
409 
deviated upward and to the left, and the head 
was held tilted to the left. Cutaneous petechiae 
were observed. On the fourth day coma per- 
sisted; there were rales in the chest. The tem- 
perature during the previous three days had 
fluctuated between 101 and I04°F. Treat- 
ment consisted of sponging, plasma transfu- 
sions, intravenous fluids, oxygen, cardiac 
stimulants, and sulfonamides. On the fifth day 
the patient’s condition seemed to improve, but 
on the sixth his temperature rose to io8°F., 
the pulse to 160 and respirations to 60. From 
the eighth to the eleventh day the patient 
rallied and the temperature fell to ioi°F. 
At this time he was mentally confused and 
unable to comprehend or speak. The blood 
pressure gradually rose to 130/78 mm, Hg. 
Oliguria became more pronounced. Electro- 
cardiograms, initially abnormal, now were 
normal. 
Examinations of the blood revealed the 
following: red blood cells, 4,050,000 to 
5,210,000; hemoglobin, 76 to 102 per cent; 
white blood cells, 5,300 (neutrophils, 51, 
lymphocytes, 49) to 23,250 (neutrophils, 90, 
lymphocytes, 10); hematocrit, 36 to 58 per 
cent; chlorides, 412 to 449 mg. per cent; 
C02 combining power, 38 volumes per cent; 
total serum protein, 6.54 per cent (alb. glob, 
ratio: 4.4:2.14); and icterus index, 15 to 16. 
(The platelet count and the prothrombin and 
bleeding times are given in Table XII; the 
clotting time varied from 3F2 to 4F2 min- 
utes.) The urine contained 2-f- albumin and 
occasional casts and pus cells. The spinal fluid 
was within normal limits. 
On the twelfth day the temperature rose 
to I03°F., bronchopneumonia became evident, 
and the patient lapsed into coma and died. 
AIP Acc. 120161. A white male, aged 18, 
was admitted to a hospital in Texas on July 
28, 1944. The day before he had complained 
of abdominal distress and of feeling hot, faint, 
and mentally confused. On admission he was 
delirious, being restless, resistive, apprehensive, 
suspicious, and disoriented. The skin was hot, 
dry and moderately cyanotic; the rectal tem- 
perature was io8°F., pulse 116, respirations 
50, blood pressure 100/60 mm. Hg. (Fig. 2). 
There were stiffness and cramping of leg 
muscles and hyperextension of the fingers. 
Extrasystoles and a soft apical systolic 
murmur were heard, but the electrocardio- 
gram was normal. Therapy consisted of cold 
water sprays, massage, oxygen, intravenous 5 
per cent glucose in saline, and vitamins C and 
Ba. 
By the second day the temperature had 
fallen to ioi°F, and the blood pressure had 
increased to 120/70 mm. Hg., but the de- 
lirium and cardiac irregularity persisted. The 
patient began to pass dark urine which showed 
a specific gravity of 1.013, was acid, and con- 
tained blood and 3-f- albumin. Blood studies 
revealed the following: red blood cells, 
4,500,000; hemoglobin, 105 per cent; white 
blood cells, 10,250 (neutrophils, 75, lympho- 
cytes, 24); hematocrit, 46 per cent; and 
urea nitrogen, 25.9 mg. per cent. Following 
a transfusion of 500 cc. of plasma the tempera- 
ture rose to io8°F. On the third day the 
temperature was again reduced by cooling 
measures and there was a slight improvement 
in the mental condition. On the sixth day 
there was twitching of the fingers, which was 
regarded as a manifestation of uremia. 
From the third to the eleventh day of illness 
the patient showed progressive oliguria and 
azotemia. The nonprotein nitrogen increased 
from 133 to 231 mg., the urea nitrogen 
from 26.7 to 37.5 mg., creatinine from 5.5 
to 10.6 mg,, and uric acid from 5.9 to 6.1 
mg. per cent. During the course of the illness 
the blood chlorides varied from 561 to 396 
mg. per cent, the C02 combining power 
from 60.6 to 35.6 volumes per cent, and the 
blood leukocytes from 10,250 to 14,350. The 
blood pressure gradually rose to 180/110 
mm. Hg. (Fig. 2). 
The patient became more and more evi- 
dently uremic and died on August 8, approxi- 
mately 11 days after the onset. 
III. Insidious Onsety Slowly Progressive 
Coursey and Late Development of Coma 
AIP Acc. 102705. A white male, aged 22, 
was admitted to a hospital in Texas on August 
14, 1943. For 4 days previously there had 410 
The Military Surgeon—November, 1946 
been increasing weakness, nausea and vomit- 
ing, and tachycardia. Examination revealed 
dryness of the skin and tongue; the tempera- 
ture was 99.4 °F., pulse 128, respiration 24 
and blood pressure 100/70 mm. Hg. (Fig. 
IA). The patient appeared somewhat 
lethargic. He was placed on a “heat exhaus- 
tion regime.” Blood studies revealed the fol- 
lowing: red blood cells, 5,350,000; hemo- 
The blood pressure was unobtainable, and the 
skin was cold, clammy, and cyanotic. The 
patient received a transfusion of 500 cc. of 
plasma and 1 ooo cc. of saline solution intra- 
venously. Signs of circulatory failure became 
progressively more marked, death occurring 
on August 16. The entire illness lasted 6 days, 
but hyperthermia was present for only 18 
hours. 
Fig. 4. Duration of illness, 11 hours. The cerebral cortex shows edema, congestion and severe cell 
disintegration. Cresyl violet stain. X450. AIP Acc. 114682. 
globin, 85 per cent; white blood cells, 17,350 
(neutrophils, 75, lymphocytes, 17, monocytes, 
8); and urea nitrogen, 35.7 mg. per cent. 
The urine had a specific gravity of 1.019, 
and contained 2-f- albumin, a moderate num- 
ber of red blood cells and a few granular casts. 
The spinal fluid was slightly blood-tinged. On 
the second day following admission the pa- 
tient became comatose; the temperature, after 
falling to 97.6°F., gradually rose to io6°F., 
the pulse became rapid and virtually imper- 
ceptible, and the respirations rapid and shallow. 
Pathologic Findings 
The literature on the pathologic anatomy 
of heat stroke is relatively meager. It is agreed 
that changes occur both in the central nervous 
system and in the viscera, but most authors 
have regarded these changes as neither strik- 
ing nor specific. 
Central Nervous System 
In approximately one-third of the 125 cases 
the entire brain was available for study. In- 
formation in regard to the others was obtained Heat Stroke 
411 
from the protocols. In the great majority of 
cases the autopsies were performed within 15 
hours of death, often in an hour or two; 
therefore post-mortem autolysis was minor or 
nonexistent. 
The weight of the brain was usually in- 
creased, often by several hundred grams, and 
the average was 1493 gm- Most cases pre- 
sented distinct edema of the leptomeninges 
well defined, the blood vessels thin-walled and 
the ventricles of normal size. 
Parenchymal Changes. Examination of 
analogous areas of the brain in representative 
cases of varying duration revealed parenchy- 
mal changes which were either slowly or 
rapidly progressive. 
Cerebral Cortex. Edema and congestion, as 
well as degenerative changes in the neurons 
Fig. 5. Duration of illness, 72 hours, with 18 hours of hyperthermia. The cells of the cerebral cortex 
are considerably shrunken and hyperchromatic. The changes do not exceed those of the usual 18-hour case 
of heat stroke. Cresyl violet stain. X 450. AIP Acc. 101118. 
and the brain, a flattening of convolutions, 
and a cerebellar pressure cone. In approxi- 
mately half of the cases the leptomeninges were 
diffusely congested, but the underlying brain 
showed only a patchy congestion, especially 
intense in the white matter and in the vicinity 
of the ventricles. No massive hemorrhages 
were observed; petechiae, however, were com- 
mon in the walls of the third ventricle and 
the floor of the fourth ventricle, but were 
scant in the leptomeninges. There was no other 
gross change, the gray and white matter being 
(Fig. 4) were found in a section of the 
frontal cortex from a representative case of 
11 hours’ duration. Most of the nerve cells 
and their dendrites were swollen, their 
cytoplasm chromatolytic or the seat of vacuolar 
disintegration, and their nuclei pyknotic. Some 
of the nerve cells had been transformed into 
“ghosts.” At this stage there was no apparent 
reaction on the part of the glia. A somewhat 
later change is illustrated in a case of 18 hours’ 
duration; the neurons were shrunken, had 
wire-like dendrites and hyperchromatic cyto- 412 
The Military Surgeon—Novemher, 1946 
plasm and nuclei (Fig. 5). Pericellular edema 
was evident, but glial reaction had not yet 
appeared. In cases of more than 24 hours’ 
duration, the number of neurons was distinctly 
reduced and glia were beginning to proliferate. 
These changes were more pronounced the 
longer the illness. The most severe damage 
was seen in a case in which survival was 12 
days (Fig. 6 and 7). The cyto-architectural 
moderate increase in glial elements; no dis- 
tinct demyelination was found, but fat stains 
revealed a slight increase of lipoids in nerve 
cells and occasional fat-laden gitter cells in 
perivascular spaces. 
Basal Ganglia. The corpus striatum and 
thalamus had undergone changes less severe 
but similar to those in the cortex. In cases 
of brief duration, the nerve cells, particularly 
Fig. 6. Duration of illness, 276 hours. The cyto-architectural pattern of the cerebral cortex is considerably 
altered because of dropping- out and degeneration of neurons and diffuse proliferation of microglia. Cresyl 
violet stain. X300. AIP Acc. 118078. 
pattern was profoundly altered because of the 
disappearance of numerous nerve cells and the 
disintegration or hyperchromatosis of many 
of those that remained; there was also an 
extensive and diffuse proliferation of glia. 
Most of these cells were microglia, belonging 
mainly to the category of “rod” cells, but an 
increase of macroglia was also observed. The 
upper layers of the cortex were generally more 
affected than the lower. The white matter 
was relatively spared, there being only a 
the large ones of the caudate nucleus and puta- 
men, were diffusely damaged. In cases of 
longer standing a proliferation of glial ele- 
ments had also taken place: microglia focally, 
especially around degenerating neurons, and 
macroglia diffusely. In the thalamus were fo- 
cal collections of glia, most numerous in in- 
tercalated areas between the chief nuclei (Fig. 
8). The globus pallidus was the least affected 
of the basal ganglia. No glial proliferation was 
noted in the periventricular system. Heat Stroke 
413 
Cerebellum. Changes in the cerebellum 
were more striking, more consistent, and more 
rapid in development than in any other part 
of the brain. When death occurred in less 
than 24 hours, edema of the Purkinje layer 
was marked and the number of Purkinje cells 
was reduced, those remaining being swollen, 
pyknotic, or disintegrated. The molecular and 
granular layers, on the other hand, were not 
layer, where only the remnants of Purkinje 
cells remained, but there was only moderate 
increase in the microglial cells of the molecular 
layer (Fig. 11). In another case of 3 days’ 
standing, glial elements had proliferated to 
about equal degree in the Bergmann and mo- 
lecular layers (Fig. 12), and under high mag- 
nification the few remaining Purkinje cells 
were found to be in a state of coagulation ne- 
Fig. 7. Duration of illness, 276 hours. This is another field from the case illustrated in Figure 6, showing 
the proliferation of microglia and of some astrocytes in greater detail. Cresyl violet stain. X 600. AIP Acc. 
11 8078. 
altered except for a moderate proliferation of 
satellite oligodendroglia (Fig. 9 and 10). 
When survival was more than 24 hours there 
was almost complete degeneration of the 
Purkinje layer and gliosis of corresponding 
severity. The glial reaction was first evident 
in the Bergmann layer, and soon afterward 
in the molecular layer. By the end of 3 days, 
one or the other or both of these alterations 
were pronounced. Thus, in one case of 3 days’ 
duration a marked proliferation of the astro- 
cyte-like forms was noted in the Bergmann 
crosis. The most advanced changes were noted 
in a case of 12 days’ duration: the molecular 
and Bergmann layers were hyperplastic, the 
granular layer rarefied (Fig. 13), and re- 
maining Purkinje cells all but consumed by 
macrophages (Fig. 14). The changes in the 
cerebellum were equally pronounced in 
hemispheres and vermis. The dentate nucleus 
was similarly affected. In a case with a survival 
period of 11 hours most of the neurons of the 
denate nucleus were hyperchromatic and be- 
ginning to disintegrate (Fig, 15), Damage to 414 
The Military Surgeon—November, 1946 
Fig. 8 
Fig. io Heat Stroke 
415 
this nucleus was so great in a case of 12 days’ 
standing that it could hardly be identified since 
most of the neurons had been replaced by glia 
(Fig. 16). High magnification of gliotic re- 
gions revealed a predominance of microglia 
over macroglia and abundant neuronophagia. 
Hypothalamus. The hypothalamus was 
carefully investigated in view of its well recog- 
nized role in the regulation of body tempera- 
ture. Like other portions of the brain, the 
hypothalamic nuclei were subject to edema 
during the early stages of the disorder. Later, 
however, no significant alteration was ob- 
served. Occasionally slight general depopula- 
tion of neurons and a mild increase in glial 
cells seemed apparent but the changes were 
so equivocal that they could not be regarded 
as significant. The hypothalamus in cases of 
different duration is illustrated in Figures 17 
to 20 inclusive. The lack of demonstrable 
damage here contrasts with the condition in 
other portions of the brain. 
Midbrain, Ponsy Medulla Oblongata and 
Spinal Cord. Nerve cells and glia were gen- 
erally normal in number and appearance. Mild 
damage of nerve cells and slight gliosis were 
observed only in the quadrigeminal region, 
the inferior olivary nuclei, and the reticular 
formation. 
Hemorrhages. Hemorrhages were pres- 
ent in the brain in 65, and in the leptomen- 
inges in 20 of the cases (Table XV). The 
cerebral hemmorrhages usually were confined 
to perivascular spaces and were not conspicu- 
ous. Leptomeningeal hemorrhages, on the 
other hand, tended to be diffuse (Fig. 21) 
and were generally most severe in cases of 
brief duration. Regardless of the length of ill- 
ness, the red blood cells in the hemorrhagic 
areas were intact, suggesting that the seepage 
was relatively recent. 
Although hemorrhages varied in location 
from case to case they were most pronounced 
in the walls of the rostral part of the third 
ventricle. Thus, they were found in the peri- 
ventricular system (Fig, 17), the paraventric- 
ular nucleus (Fig. 22), the supraoptic 
nucleus (Fig. 23), the more medial parts of 
the ventromedial and dorsomedial hypothala- 
mic nuclei, somewhat less often in the peri- 
fornical and septal regions and the medial 
portion of the thalamus. The caudal part of 
the hypothalamus was less affected, the hemor- 
rhages being mainly in the posterior hypo- 
thalamic nucleus and the adjacent periventric- 
ular system. No hemorrhages were observed 
in the mamillary body. In the midbrain, the 
periaqueductal region and the oculomotor 
nuclei were the sites of predilection (Fig. 24). 
Hemorrhages of the pons and medulla ob- 
longata were restricted to the floor of the 
fourth ventricle, most frequently in and near 
the dorsal efferent nucleus of the vagus (Fig. 
25). 
Hemorrhages were encountered occasion- 
ally in the cerebral cortex, white matter, 
striatum, pallidum, nucleus basalis, subthala- 
mus, cerebellum, and the tegmental and 
ventral portions of the lower brain stem. In 
two cases, unique in our series, the cerebral 
white matter was extensively affected, with 
widespread and well developed “brain pur- 
pura” in one (Fig. 26), and countless peri- 
vascular foci of rarefaction in the other (Fig. 
27). 
Discussion of the Pathologic 
Changes in the Brain. The authors of 
reports on some of the large series of heat 
stroke, McKenzie and Le Count30 and others, 
emphasize the frequency with which cerebral 
edema and congestion occur. Hemorrhages 
also have been commonly noted. In certain 
Fig. 8. Duration of illness, 276 hours. The thalamus displays a reduction in neurons and proliferation of 
glia, especially in intercalated areas (IA) between chief nuclei (N). Cresyl violet stain. Xi44- AIP Acc. 
118078. 
Fig. 9. Duration of illness, 5 hours. The Purkinje cells of the cerebellum are swollen and disintegrated, 
and are surrounded by oligodendroglial satellites. Cresyl violet stain. X450. AIP Acc. 85367. 
Fig. 10. Duration of illness, hours. In the cerebellum there is ‘dropping out’ and disintegration of 
Purkinje cells as well as edema of the Bergmann layer. X135. AIP Acc. 99088. The Military Surgeon—November, 1946 
416 
Fig. h. Duration of illness, 72 hours. The Purkinje cells of the cerebellum have almost completely dis- 
appeared and there is proliferation of glia in the Bergmann layer. Cresyl violet stain. X 130. AIP Acc. 96554. 
Fig. 12. Duration of illness, 72 hours. The glial cells of the molecular and Bergmann layers of the cere- 
bellum are hyperplastic and the Purkinje cells have largely disappeared. Cresyl violet stain. X65. AIP Acc. 
101158. Figure 13. Duration of illness, 276 hours. There is marked proliferation of glia in the molecular 
layer, disappearance of Purkinje cells, and rarefaction of the granular layer. Cresyl violet stain. 
X90, AIP Acc. 116078. Fig. 
36 
Fig. 
37 
Figure 36. Duration of illness, 96 hours. The photograph illustrates that breakdown of intrinsic 
muscle structure may begin in the portion of a segment adjacent to an intercalated disk. Some of 
the segments of muscle fibers are completely amorphous. Bodian stain. A1P Acc. 95093. 
Figure 37. Duration of illness, to hours. Considerable fat is present in degenerated muscle seg- 
ments. Sudan III. AIP Acc. 99987. Heat Stroke 
417 
Fig. 14. Duration of illness, 276 hours. A degenerated Purkinje cell and its apical dendrite are in the 
process of phagocytosis by glial elements. Cresyl violet stain. X450. AIP Acc. 118078. 
cases, hemorrhage constituted the most con- 
spicuous finding: for instance, severe “brain 
purpura” was described by Schwab,49 hemor- 
rhage into the putamen by Schwartz,50 and 
widespread hemorrhages into the leptomen- 
inges by Fleck and Hiickel.1' Encephalo- 
malacia (Stern,->! M’Kendrick'54), perivascular 
rarefaction (Rosenblath4'), and presumed in- 
ternal capsular hemorrhage associated with 
hemiplegia (Messiter46) also have been ob- 
served but are exceedingly rare. 
Degeneration of nerve cells, either in the 
form of acute swelling or shrinkage and 
pyknosis, has been regarded as a significant 
finding by some authors (Hartman and 
Major24), but others have believed it to be 
a post-mortem artifact. Omorokow41 has ex- 
pressed the opinion that high temperature 
causes necrosis by inducing coagulation of 
neuroglobulin. 
In our opinion the cellular changes in the 
brain are essentially the result of hyperthermia 
whereas the hemorrhages, congestion, and 
edema are chiefly secondary phenomena co- 
incident with shock; hemorrhage is probably 
exaggerated by a clotting defect resulting from 
the thrombocytopenia already mentioned. 
Degenerative neuronal change was a constant 
feature in all our cases of heat stroke, and 
could be traced through successive stages from 
acute and chronic cell alteration to disappear- 
ance of neurons and their replacement by glia. 
Moreover, it was observed that the changes 
corresponded to the length of survival after 
the onset of hyperthermia, not to the duration 
of the illness in the event that the hyperthermia 
was only terminal. Thus, in a case of “heat 
exhaustion” of 72 hours’ duration (Case 
101118), in which hyperthermia developed 
in the last 18 hours of life, there was con- 
siderable damage to neurons but neither cel- 
lular disintegration nor glial proliferation (Fig. 
5). We regard the irreversible damage to 
the central nervous system as seen in our cases 
as adequate pathologic basis for the cerebellar 
and mental sequelae which have been re- 418 
The Military Surgeon—November, 1946 
Fig. 15 
Fig. 16 Heat Stroke 
419 
ported in nonfatal heat stroke (Antheaume 
and Mignot,4 Freeman and Dumoff,18 
Goebel,22 Shepherd,51 Stewart,54 Wakefield 
and Hall,58 Weisenburg63). 
Edema, congestion, and hemorrhage were 
inconstant and were not related to the degree 
of hyperthermia. A direct relationship existed, 
however, between the severity of hemorrhage 
the same as in numerous other disorders 
terminating in shock but not associated with 
hyperthermia. 
It is recognized that a relative anoxia exists 
in shock (Kopp and Solomon30) and that the 
oxygen partial pressure of arterial blood may 
fall as much as 25 per cent ( a reduction com- 
parable to that resulting from an ascent to an 
Table XIV 
A Correlation of Blood Pressure Readings (as Index to the Severity of Shock) with the 
Degree of Cerebral Hemorrhage in 12 Cases of Heat Stroke 
AIP 
Dura- 
tion of 
Adm. 
Temp. 
(° F.) 
Blood Pressures 
Pulse 
Degree of 
Cerebral 
Hemorrhage 
Chief Sites of Hemorrhage 
Acc. 
Illness 
(Hrs.) 
Pressures 
Severe 
to 
Mod. 
Mild 
120160 
3 
IIO° 
70/0, 60/0, 90/50 
70, 60, 40 
+ 
Septal, perifornical & ant. periventricu- 
lar regions, paraventricular nucleus, 
& floor of I Vth ventricle. 
112744 
4 
ni° 
80/55 
25 
+ 
Ant. periventricular & medial hypo- 
thalamic regions. 
99612 
8 
109° 
60/0, 110/60 
60, 50 
+ 
Ant. periventricular & suprachiasmic 
regions. 
118077 
12 
I IO° 
100/60, 120/80, 108/56, 
70/50 
40, 40, 52, 20 
+ 
Septal, preoptic & subcommissural re- 
gions, & paraventricular nucleus. 
99261 
16 
O 
O 
84/50,^80/60 
34, 20 
+ 
Periventricular region (from ant. com- 
missure to corpus Luysi). 
120159 
18 
IIO° 
80/0, 100/50, 140/80, 
82/64, 80/60 
80, 50,60,18, 20 
+ 
Thalamus, medial hypothalamus (ant. 
part), paraventricular nucleus, & re- 
gion of nucleus basalis. 
115914 
2i 
1080 
150/90 
60 
+ 
Ant. periventricular region. 
117740 
7 
I IO° 
120/90 
3° 
+ 
Thalamus & septal region. 
114682 
11 
109° 
130/70, 100/60 
60, 40 
+ 
Floor of IVth ventricle & subcommis- 
sural & post, periventricular regions. 
98176 
19 
109° 
122/70 
52 
+ 
Periventricular region (lat. & Hid ven- 
tricle) & supraoptic nucleus. 
115309 
120 
109° 
142/30, 82/52, 96/60, 
120/80 
112, 30, 36, 40 
+ 
Medial hypothalamus (ant. part). 
120161 
240 
1080 
100/60, 110/70, 114/66, 
106/70,120/80,110/90 
40, 40, 48, 36, 
40, 20 
+ 
Medial hypothalamus (ant. & post.). 
and the degree of shock. Study of a number 
of cases picked at random indicated that when 
shock, as manifested by a lowering of blood 
pressure, was profound, the hemorrhages in 
the brain tended to be pronounced, but when 
the blood pressure was normal or slightly re- 
duced they were usually minimal (Table 
XIV). Moreover, the site, distribution, and 
appearance of the hemorrhages were much 
elevation of 17,500 feet) in therapeutic 
hyperthermia of I05°F., even though oxygen 
saturation of arterial blood and the arterio- 
venous oxygen difference are normal (Cullen, 
Weir and Cooke15). Although this degree of 
deficit in oxygen partial pressure may lead to 
apprehension and other mental symptoms, it 
is improbable that anoxia is a significant cause 
of changes in the brain in heat stroke. Con- 
Fig. 15. Duration of illness, 11 hours. Neurons of the dentate nucleus are hyperchromatic, and there is 
moderate capillary engorgement. Cresyl violet stain. X230. AIP Acc. 114682. 
FlG. 16. Duration of illness, 276 hours. The dentate nucleus has undergone extensive gliosis. Only a few 
shrunken neurons remain. Cresyl violet stain. X 75. AIP Acc. 118078. 420 
The Military Surgeon—Novembery 1946 
Fig. 17 
Fig, x8 Heat Stroke 
421 
siderably greater degrees of anoxia may have 
no deleterious effect. Ward and Olsen6' have 
related the case of an aviator who, owing to 
depletion of his oxygen supply, was without 
supplemental oxygen at altitudes between 
25,000 and 20,000 feet for 39 minutes, and 
at altitudes between 20,000 and 12,000 feet 
for an additional 16 minutes; he was un- 
conscious for 8 hours and semiconscious for 
12 more hours and yet recovered completely. 
The lack of neurologic residua following depri- 
vation of oxygen to the point of unconscious- 
ness has been noted also by Horvath, Dill and 
Corwin,26 and von Tavel.5° On the other 
hand, Titrud and Haymaker56 have reported 
profound mental changes in aviators subjected 
to sudden and severe anoxic insult: in one 
instance in which the patient was exposed to 
an atmosphere at 24,000 feet for 10 minutes 
there was complete mental incapacitation dur- 
ing the ensuing 3 weeks until death occurred 
as the result of massive hemorrhage from an 
esophageal ulcer; in another instance, in which 
the patient was deprived of supplemental 
oxygen during a forced descent from 27,000 
to 7,000 feet, much the same clinical picture 
was observed for three weeks, after which the 
patient was lost sight of. Pathologically, the 
findings differed considerably from those ob- 
served in heat stroke: thus, in the case of 
anoxia of three weeks’ standing, there was 
severe degeneration of the lower laminae of 
the cerebral cortex, massive necrosis of the 
striatum, focal necrosis of the globus pallidus, 
and relative sparing of the cerebellum; where- 
as in heat stroke the cerebral cortex was dif- 
fusely affected, the striatum and globus 
pallidus were almost untouched, and the cere- 
bellum was seriously damaged. Lesser involve- 
ment of the central nervous system in anoxia 
has been reported by Gildea and Cobb21 and 
van Bogaert ct al.:17 The pathologic changes 
reported by Windle, Becker and Weil6S in 
guinea pigs temporarily asphyxiated at birth 
also differed from those of heat stroke. In these 
animals a variable degree of neuronal necrosis 
in the central nervous system was observed, and 
there were swarms of microglia which had 
advanced from the subependymal cell plate of 
the fourth and lateral ventricles into adjacent 
structures, ultimately permeating most of the 
brain. In our cases of heat stroke no cellular 
activity was noted in the region of the 
ependyma, the microglia seeming to proliferate 
only in damaged areas. 
The next question that may be raised is 
whether or not the hyperthermia in heat stroke 
may be an outcome of shock. It is well known 
that a moderate elevation of rectal temperature 
is the rule in the early phases of shock, what- 
ever its cause, since the escape of body heat is 
prevented by peripheral vasoconstriction 
(Wright and Devine,69 Blalock and Price9). 
The excessive subsequent rise may be accounted 
for I) by a progressive accumulation of heat in 
the body owing to peripheral vasoconstriction 
and 2) a greater production of body heat 
brought about by increased metabolism. Kopp29 
and Kopp and Solomon'56 have found in fever 
therapy that for each degree of rise in body 
temperature there is an elevation of from 5 
to 14 pet cent in the metabolic rate, and that 
at levels of io6°F. and above the metabolic 
rate per degree of fever tends to increase 
even more sharply. Accelerated output of ad- 
renalin doubtless contributes to this elevation in 
metabolism. It would seem logical to conclude 
that hyperthermia develops in shock because 
the central mechanism for heat dissipation is 
incapacitated and that instrumental in heat 
conservation is overactive, a view substantially 
the same as that held by Hyndman and 
Fig. 17. Duration of illness, 4 hours. Admission T. m°F.; B. P. 80/55. A transverse section of the 
brain at the level of the subcommissural and preoptic regions and the anterior portion of the optic tract 
(OT) discloses scattered petechial hemorrhages beneath the anterior commissure (AC) adjacent to the third 
ventricle (3V) and in the supraoptic nucleus (SO). The medial and lateral preoptic nuclei (MP and LP) 
and the nucleus basalis (NB) have a normal appearance. X20. AIP Acc. 112744. 
Fig. 18. Duration of illness, 8 hours. Admission T. io9°F.; B. P. 132/70. The section is through the 
anterior portion of the hypothalamus, showing the paraventricular (PV), supraoptic (SO) and ventrome- 
dial (VM) nuclei. No significant changes are visible. Cresyl violet stain. X20. AIP Acc. 98175. 422 
The Military Surgeon—November y 1946 
Fig. 19 
Fig. zo Heat Stroke 
423 
Wolkin.27 The same conclusions would apply 
to heat exhaustion terminating in hyperthermia, 
an example of which (Case 102705) is de- 
scribed under Illustrative Cases. 
In attempting to frame a concept as to the 
cause of heat stroke one has to take into con- 
sideration not only the mechanism by which 
body heat is dissipated but also the conditions 
of the external environment. Body heat is lost 
entirely through physical means, but may be 
reduced by a lowering of basal metabolism. In 
an individual doing light muscular work at 
ordinary room temperature, radiation, con- 
vection, and conduction account for 70 per 
cent of heat loss, evaporation from the skin 
14.5 per cent, vaporization of water from the 
lungs 8.0 per cent, the warming of inspired 
air 2.5 per cent, the loss of heat from the urine 
and feces 1.5 per cent, and by other means 
3.5 per cent (Gemmill20). In a warm at- 
mosphere, the loss of body heat is accelerated 
in a number of ways, of which the most sig- 
nificant are sweating and a redistribution of 
blood so that a greater amount reaches the skin. 
It is apparent, however, that the temperature- 
regulating mechanism of the body is capable 
of adequate function only within certain limits 
of external environment. If the temperature of 
the air and of surrounding objects is higher 
than body temperature, the body will gain heat 
by radiation, convection, and conduction, 
leaving evaporation from the skin and vapori- 
zation of water from the lungs the only sig- 
nificant cooling mediums; thus, evaporation 
and vaporization, instead of accounting for the 
customary 22.5 per cent of heat loss from the 
body, now become responsible for nearly 100 
per cent of heat loss. High vapor pressure (ab- 
solute humidity) will vitiate these last natural 
means of maintaining normal body tempera- 
ture, whereupon the increase may reach such 
a point that the temperature-regulating mech- 
anism is damaged and a vicious cycle estab- 
lished. In this connection it should be pointed 
out that despite impending collapse, the men 
whose histories are represented in our series 
often had expended every effort to continue 
their march or other activity, thus increasing 
further the body heat-load at a time when the 
capacity for heat dissipation was critically im- 
paired by the high temperature and high vapor 
pressure of the external environment 
(Shickele48). Taking these factors into con- 
sideration, the following sequence may 
be postulated; 1) the precipitating cause 
of heat stroke is excessive body heat; 2) the 
excess heat incapacitates the central heat-dis- 
sipating mechanism; 3) as a result sweating 
ceases, causing a greater accumulation of heat 
in the body; 4) with the rise of body tem- 
perature the increasing metabolism adds further 
fuel to the flames; 5) sooner or later shock 
usually ensues and, owing to peripheral 
vasoconstriction, further augments the body 
temperature. The shock may be brought under 
control, but the damage to the brain induced 
by the hyperthermia has been done This hy- 
pothesis is at variance with that of Adolph and 
Fulton2 who believed heat stroke to be the 
result of circulatory failure occasioned by the 
following series of events: 1) stimulation of the 
skin by heat, leading to reflex sweating and 
sometimes to a progressive deepening and ac- 
celeration of respiratory movements; 2) a re- 
sultant loss of carbon dioxide from the blood, 
particularly in the sweat; 3) a decrease in re- 
sistance of peripheral blood vessels, followed by 
a significant reduction in the volume of cir- 
culating blood, the heart compensating for the 
lack of venous return of the blood by beating 
Fig. 19. Duration of illness, 130 hours. Admission T, io9°F.; B. P. 140/30, 95/60, 120/80. The section 
of the hypothalamus is at the level of the ventromedial (VM) and tuberomamillary (TM) nuclei. The 
fornix (F), optic tract (OT) and third ventricle (3V) are indicated. No changes are observed. Cresyl violet 
stain. X20. AIPAcc. 115309. 
Fig. 20. Duration of illness, 14 hours. Admission T. io8°F.; B. P. not recorded. A transverse section of 
the hypothalamus at the level of the mamillary body shows a normal appearance of the medial (M) and 
lateral (L) mamillary, the tuberomamillary (TM) and the posterior hypothalamic (P) nuclei. The fornix 
(F), mamillothalamic tract (MT) and third ventricle (3V) are designated. Cresyl violet stain. X20. 
AIP Acc. 98163. 424 
The Military Surgeon—November, 1946 
Fig. 21 
Fig. 22 Heat Stroke 
425 
faster, and leading ultimately to “a type of 
circulatory failure which would have to be 
included under the general category of shock.” 
It is possible that the cases studied by Adolph 
and Fulton may fall into a different category 
than the great majority of our series, for theirs 
are apparently heat exhaustion complicated by 
shock and hyperthermia, while ours are in- 
stances of primary heat stroke. 
Our attempt to demonstrate structural 
changes in the portions of the hypothalamus 
concerned with temperature regulation was 
unsuccessful. The work of Ranson and his 
associates,42 Keller,28 and others, established 
the location of the central heat-dissipation 
mechanism in experimental animals in the 
preoptic and supraoptic regions, and heat- 
conservation and heat-production mechanism 
in the more caudolateral portion of the hypo- 
thalamus. The same general plan applies to 
man (Alpers,3 Beaton and Herrmann,8 Davi- 
son,16 Zimmerman71). In an intensive study 
of the hypothalamus in 13 cases of heat stroke 
in relatively old persons afflicted with various 
chronic diseases, Morgan and Vonderahe40 
found that acute cell alterations were mainly 
in cells of the tuberomamillary nucleus, and 
chronic changes, ascribed to pre-existing dis- 
ease, in the nuclei tuberis laterales and the para- 
ventricular nucleus. These authors attributed 
the hyperthermia to stimulative action induced 
by the acute cellular changes. They intimated 
that a predisposition to heat stroke existed by 
virtue of previous damage to nuclei concerned 
with heat dissipation. This is in accord with 
the observations of Morgan39 on dogs and 
rabbits in which moderate hyperthermia (a 
rise of 1.2 to 5.4°F. in body temperature) 
was induced by the injection of typhoid or 
bronchisepticus toxin; chromatolytic changes 
were found in 60 per cent of cells of the 
tuberomamillary nucleus, in 15.8 per cent of 
those of the paraventricular nucleus, and in 
12.6 per cent of those of the supraoptic 
nucleus. 
Although we have no reason to doubt that 
the thermoregulatory mechanism in our cases 
was severely damaged functionally, we were 
unable to find significant anatomic change in 
the hypothalamus other than the hemorrhages. 
Other Tissues of the Body 
The pathologic changes in the thoracic and 
abdominal viscera and elsewhere in the body 
were numerous and diverse. Evidences of acute 
circulatory failure, such as hemorrhage, edema, 
and vascular engorgement, were observed in 
virtually all cases regardless of the duration 
of illness. Parenchymal damage, on the other 
hand, was present only in certain cases, but 
tended to be more severe the longer the sur- 
vival. 
Hemorrhages. Hemorrhages constituted 
one of the most striking features at au- 
topsy. They were petechial or of greater 
scope regardless of their location. The approxi- 
mate incidence of hemorrhage in the 125 cases 
is shown in Table XV. The longer the pa- 
tients’ survival, the wider the dissemination 
of hemorrhage. Thus, when death occurred 
within 12 hours after the onset, the different 
locations of hemorrhage averaged 4.4 whereas 
when death occurred after 48 hours or more 
the average was 5.3. No correlation could be 
made between dissemination of hemorrhages 
and blood pressure levels. Hemorrhage was as 
frequent in cases in which the diastolic pres- 
sure was below 50 as when it was above. In 
our collection of 125 cases several which 
were parallel clinically showed the widest 
variation in the number of sites of hemor- 
rhage. 
Hemorrhages were absent in only 3 of our 
cases. In each of these the only known com- 
mon factor was that of comparatively low 
temperature. Thus, in 2 (Cases 100855 anc^ 
Fig. 21. Duration of illness, 5 hours. Admission T. io9°F. ; B. P. not recorded. Diffuse hemorrhage is 
present in the leptomeninges. X 75. AIP Acc. 88639. 
Fig. 22. Duration of illness, 4 hours. Admission T. iii°F. ; B. P. 80/55. Coalescing petechial hemor- 
rhages predominate in the paraventricular nucleus of the hypothalamus. (F) fornix; (3V) third ventricle. 
Cresyl violet stain. X40. AIP Acc. 112744. 426 
The Military Surgeon—November, 1946 
Fig. 23 
Fig. 24 
Fig. 25 Heat Stroke 
427 
101159), of less than 24 hours’ duration, the 
temperatures were ioi° and 97°F. respec- 
tively, while in the other (Case 98160), of 
several days’ standing, the temperature re- 
mained below io4°F. until terminally. The 
blood pressures in the first 2 cases were too 
low to read whereas that of the third is not 
known. 
Serous Cavities. Transudates into one or 
more serous cavities occurred in 33 of the 
series, being equally frequent in acute and 
chronic cases. Pericardial fluid in excess was 
observed in 20 cases; the fluid was blood- 
tinged in 4, bloody in 2, and straw-colored in 
the remainder, the amount ranging between 
25 and 75 cc. with an average of 40 cc. In 3 
other instances there was a slight amount of 
bloody fluid in the pericardial sac. Where 
pericardial blood existed there were always 
numerous petechial or ecchymotic hemorrhages 
of the serous membrane. The pleural cavities 
were second only to the pericardium as sites of 
transudate. The fluid, usually about equal in 
amount on the two sides, was serous in 9 
instances, blood-tinged in 7, and bloody in 3. 
The range was between 50 and 2100 cc. with 
an average of 330 cc. Least often affected was 
the peritoneal cavity, which contained serous 
fluid in 3 cases and blood-tinged in 2, the 
amount ranging between 40 and 300 cc. with 
an average of 185 cc. In 2 instances in which 
large subserosal hematomas (6 and 10 cm. in 
diameter) were found there was no free blood 
in the peritoneal cavity. 
Heart. No disproportion in the relative 
sizes of the heart chambers could be seen in 
the majority of cases. In 23 the right side, 
especially the auricle, was dilated, and in 2 the 
left ventricle. The heart was in complete 
systole in 4 cases. Frequently the muscle was 
described as flabby. 
Hemorrhages were common (Table XVI). 
They were observed as often in cases of short 
as of long duration. Those in the sub- 
epicardial tissue, usually petechial (Fig. 28), 
Table XV 
The Incidence of Hemorrhage in 125 Fatal 
Cases of Heat Stroke 
Skin 
38 
Conjunctivae 
Meninges 
Brain 
65 
Pleurae 
33 
Lungs 
73 
Epicardium 
76 
Endocardium 
65 
Heart Muscle 
24 
Peritoneum 
73 
Gastro-intestinal Tract 
49 
Spleen 
15 
Pancreas 
8 
Peri-adrenal 
20 
Renal Pelvis 
IO 
Urinary Bladder 
5 
Others* 
14 
* These include hemorrhages into subcutaneous tis- 
sue, skeletal muscle, kidney parenchyma, ureters, mu- 
cous membranes of nose and mouth, thymus, liver, 
neurohypophysis, choroid plexus, wall of gall bladder, 
and peritracheal, peribronchial, perithyroid and peri- 
aortic tissues. 
tended to be concentrated at the base of the 
heart, especially in the vicinity of the anterior 
longitudinal sulcus, and they sometimes in- 
volved the parietal pericardium as well. Sev- 
eral relatively large subepicardial extravasations 
of blood were found, the largest measuring 
4x3 cm. Occasionally, confluent hemor- 
rhages covered all surfaces of the heart. The 
Fig. 23. Duration of illness, 8 hours. Admission T. io9°F.; B. P. 60/0. The supraoptic nucleus (SO) 
and vicinity exhibit perivascular and diffuse hemorrhage. (OT) optic tract. Cresyl violet stain. X40. AIP 
Acc. 99612. 
Fig. 24. Duration of illness, 35 hours. Admission T. io6°F.; B. P. 100/70. There are petechial hemor- 
rhages and vascular engorgement of the periaqueductal gray matter and the oculomotor nuclei (OM). 
Cresyl violet stain. X40. AIP Acc. 110465. 
Fig. 25. Duration of illness, 5 hours. Admission T. io9°F. ; B. P. not recorded. Petechial hemorrhages 
present in the floor of the fourth ventricle (4V) are in the vicinity of the dorsal efferent nucleus of the 
vagus (V). Cresyl violet stain. X40. AIP Acc. 88639. 428 
The Military Surgeon—November, 1946 
Fig. 26 
Fig. 27 Heat Stroke 
429 
Table XVI 
Data on Nineteen Fatal Cases of Heat Stroke in which Lower Nephron Nephrosis Was Observeo 
AIP 
Acc. 
Duration 
of Illness 
(Hrs.) 
Degree of 
Hb. Nephrosis 
Blood Pressure 
NPN 
(mg. %) 
Transfusion 
Centro- 
lobular 
Necrosis 
of Liver 
Plas- 
ma 
Blood 
Mild 
Mod. 
Severe 
88640 
8* 
+ 
0 
0 
90/0, 118/70 
— 
+ 
0 
0 
85875 
+ 
0 
0 
— 
— 
+ 
0 
0 
101019 
135 
+ 
0 
0 
100/60 
— 
0 
+ 
0 
83038 
155 
+ 
0 
0 
— 
— 
0 
0 
0 
101442 
18 
0 
+ 
0 
80/20, 120/60, 80/50 
— 
+ 
0 
0 
98176 
19 
0 
+ 
0 
122/70 
— 
0 
0 
0 
98213 
235 
+ 
0 
0 
110/60 
— 
0 
0 
0 
102099 
24 
0 
+ 
0 
80/30, 60/40 
— 
0 
0 
0 
99625* 
24 
0 
+ 
0 
104/40 
— 
+ 
0 
0 
97555 
26 
0 
+ 
0 
60/40, 110/70, 60/40 
45 
+ 
0 
0 
84111 
26 
+ 
0 
0 
100/20, 120/60 
— 
0 
0 
0 
97554 
34 
0 
+ 
0 
136/80, 86/40 
48 
+ 
0 
0 
97U8 
35 
0 
0 
+ 
150/110, 60/40 
58 
0 
0 
+ 
113428 
36 
0 
0 
+ 
78/50,105/65 
— 
0 
0 
+ 
115686 
39 
0 
0 
+ 
135/65. 130/90 
49 
0 
0 
0 
98544 
72 
0 
0 
+ 
98/60, 100/30, 120/60 
— 
0 
0 
+ 
96554 
72 
0 
0 
+ 
102/64 
— 
+ 
0 
+ 
95093 
96 
0 
0 
+ 
50/0, 102/68, 50/20 
— 
+ 
+ 
+ 
120161 
249 
0 
0 
+ 
100/60, 114/66, 110/90 
133, 231, 
229, 2I3f 
+ 
0 
0 
* This is the only case of the series in which sulfonamides were administered, 
f Urea nitrogen values were also elevated: 25.9, 37.5, 66.0 and 46.8 mg. %. 
subendocardial tissue was somewhat less fre- 
quently the seat of hemorrhage. The site of 
predilection was the left side of the inter- 
ventricular septum, but in numerous instances 
the adjoining left ventricular wall, including 
the papillary muscles, was involved. Leaflets 
of the aortic, mitral and tricuspid valves con- 
tained petechiae occasionally, and larger 
hemorrhages were present in the subendo- 
cardium (Fig. 29) in 10 cases. In one case 
extravasation of blood had undermined almost 
all of the endocardium of the left ventricle, 
and in another, hemorrhage reddened an entire 
tricuspid leaflet, but most of the ecchymoses 
were in the region of the bundle of His. This 
concentration of endocardial hemorrhages has 
previously been emphasized by Wilson.GiJ 
Hemorrhage into cardiac muscle, especially 
that of the left ventricle, occurred in 24 
instances. Occasionally a massive extravasation 
of blood was found in the wall of the heart 
(Fig- 30)- 
Sections of heart stained by the Bodian 
method revealed degenerative changes in 13 
of the 27 cases in which the illness lasted more 
than 24 hours; the lesions were minor in 6 and 
major in 7. Degeneration was also seen in 
approximately one-third of the cases of shorter 
duration, and was minor in all except 2. The 
lesions were either focal and well circum- 
scribed, or diffuse and patchy. Usually they 
were small (Fig. 31), occasionally large 
(Fig. 32). Focal necrosis was evident in cases 
in which death occurred as early as 6 hours 
Fig. 26. Duration of illness, 4 days. T. 97, io7°F.; B. P. 40/0. Ring- hemorrhages surround thrombosed 
vessels in the cerebral white matter (‘brain purpura’). Cresyl violet stain. X 70. AIP Acc. 101661. 
Fig. 27. Duration of illness, 12 hours. Perivascular foci of rarefaction are scattered throughout the cerebral 
white matter. Hematoxylin and eosin stain. X 70. AIP Acc. 95285. 430 
The Military Surgeon—Novemhery 1946 
Fic. 28 
Fig. 29 
Fig. 30 Heat Stroke 
431 
after onset (Fig. 33). The degenerative 
change was often decidedly irregular, im- 
parting to the muscle fibers a moth-eaten ap- 
pearance (Fig. 34). In some of the chronic 
cases, and in an occasional acute one, several 
contiguous segments of muscle fibers were 
shrunken, devoid of nuclei and entirely 
amorphous (Fig. 35). Sometimes only a part 
of a muscle segment adjoining an intercalated 
disk was affected (Fig. 36). Fragmentation 
or rupture of muscle fibers (Fig. 38) was 
conspicuous in well over one-fourth of the 
cases. 
When sections of the heart from repre- 
sentative cases were stained for fat a somewhat 
greater incidence of degenerative changes was 
disclosed than by the Bodian method. Sig- 
nificant fatty change in cardiac muscle was 
evident in cases in which the span of illness 
was as short as hours. A section of the 
heart from a case of 10 hours’ duration is 
shown in Figure 37. 
Lungs. The lungs were the seat of hemor- 
rhage in 73 of the series. Gross observation 
revealed hemorrhages of three sorts: petechial, 
which were widespread and usually associated 
with pleural petechiae; ecchymotic, which 
were wedge-shaped and discrete and fre- 
quently subpleural in location; and massive. 
In most of these cases the lungs were literally 
saturated with blood. The hemorrhage was 
usually limited to the lower lobes of both 
lungs, but frequently it involved the lower 
portions of the middle and upper lobes as well. 
That edema was almost invariably associated 
with hemorrhage was apparent from the 
abundant frothy hemorrhagic fluid which the 
lungs and bronchial tree yielded on section. 
The high degree of hemorrhage and edema 
was reflected in the weights of the lungs: that 
of the two together varied from 640 to 2360 
gm., the average being 1315 gm., and in 
no case was within normal limits. 
Microscopic examination confirmed the 
gross findings. In virtually all cases there was 
intense vascular congestion. Where hemor- 
rhage existed, it usually filled or partly filled 
individual alveoli (Fig. 39). Wedge-shaped, 
neatly circumscribed ecchymoses were usually 
found to be hemorrhagic infarcts (Fig. 40) 
which apparently were secondary to circula- 
tory stasis with venous thrombosis. Sometimes 
there was hemorrhage without edema (Fig. 
41). Perivascular hemorrhage was rare. In 
the portions of the lung free from hemorrhage 
and edema the air sacs were considerably dis- 
tended; series of them were often ruptured 
(Fig. 40). 
Lobular pneumonia was found in 16 cases 
in which survival was less than 48 hours and 
in 15 cases of longer duration. The overall 
incidence was about 25 per cent, in contrast 
to the 10 per cent reported by Reid.43 Pneu- 
monia in our group was observed as early as 
8 hours after the onset of hyperthermia (Case 
99825) and generally was restricted to hemor- 
rhagic areas of the lung. It was severe in only 
4, and was regarded as the chief contributor 
to death in 2 (Cases 94751 and 101118). 
Kidney. The kidney usually was hyperemic 
no matter what the duration of the illness. In 
about half of the cases congestion was intense, 
involving medullary, cortical, and glomerular 
vessels alike. The glomerular tufts in many of 
the remaining cases were relatively less en- 
gorged than the capillary bed elsewhere. 
Ischemia of glomeruli, usually associated with 
a general decrease in the volume of blood in 
the kidney, was observed in 7 cases, in 3 of 
which there was hemoglobinuric nephrosis 
(Cases 97148, 97554 and 120161). The 
weights of the kidneys were above the recog- 
nized normal, the average combined weight 
being 360 gm., with a range between 200 
Fig. 28. Duration of illness, 47 hours. The section in through the base of the left ventricle, the mitral 
valve, the left auricle and the main left coronary artery. The ventricular and auricular subepicardium 
contains numerous petechial hemorrhages, mostly perivascular. X 7. AIP Acc. 97556. 
Fig. 29. Duration of illness, 144 hours. A large ecchymosis is present in the subendocardial tissue of the 
left ventricle. X20. AIP Acc. 102705. 
Fig. 30. Duration of illness, 19 hours. A widespread extravasation of blood in the myocardium disrupts 
many of the muscle fibers. X 190. AIP Acc. 86244. 432 
'Phe Military Surgeon—November, 1946 
Fig. 31 
Fig. 32 Heat Stroke 
433 
and 600 gm. In three-fourths of the cases 
the weights were more than 300 gm. 
Hemorrhages, usually petechial but occa- 
sionally ecchymotic, were present in 20 cases: 
in pelvic tissue in 10, subcapsular in 9, and 
in interstitial tissues or tubules of the kidney 
in 3, Hemorrhages dotted the mucosa of the 
ureter in one, and were of greater scope in 
another (Fig. 42). The bladder was the seat 
of petechial or ecchymotic hemorrhage in 5 
instances. 
Parenchymal damage was scanty when the 
illness ran an acute course, but was more 
pronounced when the duration was longer. 
In approximately one-third of the cases in 
which the course was 24 hours or less a few 
pigmented casts were found in distal con- 
voluted tubules, and in about one-ninth 
epithelial cells of distal convoluted tubules were 
in a state of early degeneration. A combina- 
tion of the two was sometimes observed (Fig. 
43). Changes sufficiently pronounced to war- 
rant the diagnosis of lower nephron nephrosis 
(“hemoglobinuric nephrosis”) were observed 
in 9 of this group of early cases (Table XVI). 
In most of these there were pigmented casts 
in distal segments of the nephron; in a few, 
foci of inflammatory cells were seen in the 
interstitial tissue, and occasionally necrosis of 
distal convoluted tubules dominated the picture 
(Fig. 44). 
In cases of longer standing, the incidence 
of lower nephron nephrosis was considerably 
higher. Thus, of 27 cases in which survival 
was longer than twenty-four hours 10 
exhibited this complication (Table XVI). Pig- 
mented casts and disintegration of lining 
epithelium in the lower nephron (Fig. 45) 
were, in general, more widespread than in the 
cases of shorter duration; hyaline casts fre- 
quently were in profusion, collections of in- 
flammatory cells (mostly lymphocytes) were 
sometimes observed in the interstitial tissue 
(Fig. 46), and there was generally a pro- 
liferation of interstitial cells and an inter- 
tubular edema in cases of relative longstanding. 
A survey of certain available clinical and 
laboratory data from the cases of lower 
nephron nephrosis (Table XVI) reveals that 
reduction in blood pressure and elevation of 
blood nonprotein nitrogen were usual. In a 
little less than half of the cases neither plasma 
nor whole blood transfusions were given. 
None of the patients received sulfonamides. 
Lobular necrosis of the liver was an associated 
finding in 5 of the more chronic cases. 
Liver. Like other organs, the liver usually 
was congested and weighed more than normal. 
The average weight was 1790 gm., the range 
being between 1200 and 3260 gm. Perisinus- 
oidal edema was frequent, and on occasion 
severe (Fig. 47). In cases in which survival 
was less than 30 hours there was no evidence 
of damage to parenchymal cells. Thirty-one 
hours after onset was the soonest centrolobular 
necrosis was detected. In the 12 cases in which 
necrosis was present (Table XVII) the degree 
of damage ranged from moderate to severe 
(Figs. 48 and 49). Frequently the liver was 
relatively ischemic. Dissociation of liver cords 
was observed in several of the chronic cases and 
in a few of the acute ones. Kupffer cells re- 
mained unaltered. 
Reference to clinical and laboratory data 
in the 12 cases in which the liver underwent 
necrosis (Table XVII) reveals that lower 
nephron nephrosis was present in 8. Trans- 
fusion of plasma or blood had been given in 6. 
Jaundice became manifest in some cases of 
longer standing, and the icterus index was 
elevated in 2 cases in which jaundice was not 
detected. 
Adrenal. In most of the cases of brief 
duration there was engorgement of sinusoids, 
while in those of longer duration a relative 
ischemia was frequent. Hemorrhage was rare, 
Fig. 31. Duration of illness, 72 hours. A number of miliary infarcts are visible in heart muscle, the fibers 
in these foci being- uniformly disintegrated. Bodian stain. X 70. AIP Acc. 9854.4. 
Fig. 32. Duration of illness, 130 hours. Section through the left ventricle of the heart discloses a rela- 
tively large area of necrosis associated with moderate extravasation of blood. Bodian stain. X no. AIP Acc. 
115309. 434 
The Military Surgeon—November, 1946 
Fig. 33. Duration of illness, 6 hours. In focal areas of cardiac degeneration many of the cross-striations 
have lost their identity, and those that remain consist of dispersed granules. Capillary engorgement is notable. 
Bodian stain. X 500. AIP Acc. 98519. 
Fig. 34. Duration of illness, 168 hours. The degenerative changes in the cardiac muscle are irregular, 
giving a moth-eaten effect. Bodian stain. X500. AIP Acc. 98160. 
Fig. 35. Duration of illness, 96 hours. Disintegration is present over several continguous segments of 
cardiac muscle. Such areas were numerous in this case. Bodian stain. X350. AIP Acc. 95093. Heat Stroke 
435 
Fig. 38. Duration of illness, 9 hours. There is frank rupture of degenerated muscle fibers. 
Bodian stain. X850. AIP Acc. 110273. 
being minimal and confined to the junctional 
zone of medulla and cortex in only two. 
Pericapsular hemorrhage, on the other hand, 
occurred in 20 instances (Table XV). Peri- 
sinusoidal edema was infrequent. The lipoid 
content of cells of the adrenal cortex was 
within normal limits in most of the cases in 
which survival was less than 24 hours, but 
generally was depleted in cases of longer dura- 
tion. In some of the more chronic cases the 
cytoplasm of the cells of the adrenal cortex 
had a compactness similar to that of liver cells. 
Degenerative changes in the adrenal cortex 
were of relatively minor degree. In about 12 
of the series, especially in those in which the 
illness ran an acute course, the cords of the 
Table XVII 
Data on 12 Fatal Cases of Heat Stroke in which Centrolobular Necrosis of the Liver Was Observed 
AIP 
Acc. 
Duration 
of Illness 
(Hrs.) 
Degree of 
Necrosis 
Blood Pressure 
Jaun- 
dice 
Icterus 
Index 
Transfusion* 
Lower 
Nephon 
Nephro- 
sis 
Plas- 
ma 
Blood 
Mild 
Mod. 
Severe 
95887 
3i 
0 
+ 
0 
— 
? 
— 
0 
0 
0 
97H8 
35 
0 
+ 
0 
150/110, 60/40 
? 
— 
0 
0 
+ 
110465 
35 
0 
0 
+ 
110/70 
0 
— 
+ 
0 
+ 
113428 
36 
+ 
0 
0 
78/50, 105/65 
0 
31.8 
0 
0 
+ 
947 51 
60 
+ 
0 
0 
125/70, 108/60 
+ 
— 
0 
+ 
0 
98544 
72 
0 
0 
+ 
98/60, 100/30, 120/60 
0 
— 
0 
0 
+ 
96554 
72 
0 
0 
+ 
102/64 
0 
60.0 
+ 
0 
+ 
101158 
72 
0 
0 
+ 
110/70 
+ 
— 
+ 
0 
+ 
95093 
96 
0 
0 
+ 
50/0, 102/68, 50/20 
0 
— 
+ 
+ 
+ 
132693 
104 
0 
+ 
0 
82/60 
+ 
— 
0 
0 
0 
115309 
130 
0 
0 
+ 
142/30, 82/52,120/80 
+ 
— 
0 
0 
+ 
118078 
276 
0 
+ 
0 
80/40, 98/60,130/78 
? 
16.0 
+ 
0 
0 
* None of the patients in this series received sulfonamides, 436 
The Military Surgeon—November, 1946 
Fig. 39 
Fig. 40 
Fig. 41 Heat Stroke 
437 
zona fasciculata and sometimes the ovoid 
groups of cells of the zona glomerulosa showed 
the “tubular degeneration” originally described 
by Rich44 in fulminating infectious diseases: it 
was as though intercellular adhesion had been 
lost, causing apposing columns or groups of 
cells to spring apart, converting them into 
“tubules” (Fig. 50). Actual necrosis of cells of 
the adrenal cortex was observed only in cases 
in which survival was more than 24 hours. 
Altogether 12 of the 28 cases in the “chronic” 
group showed this change. Usually the necrosis 
was focal and was limited to the zona fascicu- 
lata, as has been noted also in hyperthermia by 
Kopp and Solomon.30 In some instances the 
cells undergoing disintegration were greatly 
shrunken, distorted, and disarranged, and 
were surrounded by cells having the appear- 
ance of macrophages (Fig. 51). Relatively 
large areas of degeneration of the zona 
fasciculata were present in only one case (Fig. 
52). Occasionally the degenerate cells had 
disappeared from the fibrous stroma (Fig. 53). 
Except for congestion, which varied in degree, 
the adrenal medulla appeared normal in every 
case. 
Bone Marrow. Histologic study of bone 
marrow was limited to 15 cases. Congestion 
was a prominent feature in all. Of the various 
elements of the bone marrow, the megakaryo- 
cytes suffered by far the most. In only 3 cases 
were the megakaryocytes undamaged; in 2 of 
these the survival period was exceedingly short 
(2 hours), and in the other the patient’s tem- 
perature was subnormal (Table XVIII). The 
changes in megakaryocytes consisted of severe 
pyknosis of nuclei, karyorrhexis, and disappear- 
ance of nuclei (Fig. 54 and 55). Severe 
degenerative changes in megakaryocytes were 
evident as early as 6 hours after onset of hyper- 
thermia. In approximately half of the 15 cases 
there was a reduction in the number of mega- 
karyocytes. Regeneration was apparent in 3 
instances, numerous megakaryoblasts being in 
evidence (Fig. 56). Depletion of cells of the 
granulocytic and erythrocytic series was ob- 
served when survival was 35 hours or more, 
but in only one was the loss of cells severe 
(Fig. 57). Inasmuch as bone marrow changes 
of this severity were not seen in the controls, 
which included cases of anoxic anoxia (Titrud 
and Haymaker56), they were believed to be due 
essentially to excessive heat. 
Spleen. The only significant finding in the 
spleen was that of congestion, and in 15 cases 
small hemorrhages. The weight of the spleen 
was somewhat increased in approximately one- 
seventh of the cases, the average of this group 
being 290 gm. 
Gastrointestinal Tract. The vessels 
throughout the gastro-intestinal tract were en- 
gorged in virtually all the cases, with hemor- 
rhages usually punctate and confined to the 
mucosa in 49 (Fig. 58). Edema of the sub- 
mucosa was a common finding. Ulceration of 
the stomach and duodenum was observed only 
once (Case 115309), the tarry black mucoid 
material in the entire gastro-intestinal tract at 
autopsy indicating that the ulceration had been 
present during life. 
Remaining Tissues. No significant le- 
sions were observed except as listed in 
Table XV. 
Discussion of the Pathologic 
Changes in Tissues Other Than the 
Brain. Most of the lesions apart from those 
in the brain can be attributed to the anoxia 
and circulatory collapse incident to shock. 
Thus, hemorrhages, serous transudates, focal 
myocardial degeneration, pulmonary infarcts 
Fig. 39. Duration of illness, 18 hours. Hemorrhage or edema fluid, or both, fill virtually all alveolar 
sacs of the lung. The partially adherent interlobar cleft is edematous, and persistent mesothelial-lined spaces 
are filled with transudate. X 30. AIP Acc. 101442. 
Fig. 40. Duration of illness, 12 hours. The section presents a hemorrhagic infarct of the lung which 
apparently is secondary to stasis venous thrombosis. A beginning lobular pneumonia is present. Uninvolved 
portions are emphysematous. X25. AIP Acc. 118077. 
Fig. 41. Duration of illness, 2 hours. Part of the section of the lung presents a diffuse hemorrhage whereas 
the remainder is relatively normal. X 10. AIP Acc. 149848. Fig. 42 
Fig. 43 
Fig. 44 Heat Stroke 
439 
of the stasis type, centrolobular necrosis of the 
liver, tubular degeneration and necrosis in the 
adrenal cortex, and lower nephron nephrosis 
have all been observed in shock developing 
from a variety of causes. It is likely that the 
liver damage can be attributed in part to hyper- 
stroke. We gained the impression that the 
extent of hemorrhage was greater than that 
ordinarily seen in shock due to other causes. 
Furthermore, although there seemed to be 
some correlation between hemorrhage in the 
brain and low blood pressure indicative of 
Table XVIII 
Findings in the Bone Marrow in 15 Fatal Cases of Heat Stroke 
AIP 
Dura- 
tion of 
Maxi- 
mum 
Blood Pressure 
Megakaryocytes 
Platelet 
Depletion 
of Cells 
other than 
Megakary- 
ocytes 
Degree 
of 
Acc. 
Illness 
(Hrs.) 
Temp. 
(° F.) 
Nor- 
mal 
Degen- 
eration* 
Regen- 
eration* 
Count 
Hemor- 
rhage f 
94560 
2 
108.8 
50/0 
+ 
0 
0 
— 
0 
+ + + 
97599 
2 
109 
80/0 
+ 
0 
0 
— 
0 
+ + 
99445 
6 
106.6 
— 
0 
+ + + 
0 
— 
0 
+ 
117740 
7 
no 
120/90 
0 
+ + 
0 
120,000 
0 
+ + + 
101829 
ih 
108 
80/40, 100/60, 60/40 
0 
+ + + 
0 
— 
0 
+ + + 
101159 
12 
97 
90/0 
0 
0 
0 
— 
0 
0 
118077 
12 
no 
100/60, 108/56, 70/50 
0 
+ 
0 
40,000 
0 
+ + + 
97555 
26 
109.2 
60/40, n0/70, 60/40 
0 
+ 
+ + 
— 
0 
+ + 
86617 
32 
106.4 
80/60 
0 
+ + + 
+ 
— 
0 
+ + + 
97!48 
35 
107.2 
150/110, 60/40 
0 
+ + 
0 
— 
+ 
+ + 
110465 
35 
106 
110/70 
0 
+ + 
0 
— 
+ 
+ + 
97556 
47 
109 
80/50, 100/70, 135/85 
0 
+ 
+ + + 
— 
+ + 
+ 
96554 
72 
108 
102/64 
0 
+ + + 
0 
— 
+ 
+ + + 
98544 
74 
107.8 
98/66, 100/30, 120/60 
0 
+ + 
0 
— 
+ + + 
+ + + 
118078 
276 
no 
80/40, 98/60, 130/78 
0 
+ + 
0 
92.000 
31.000 
66.000 
130.000 
260.000 
4- 
+ + + + 
* +, mild; ++, moderate; + ++, severe. 
f The degree of hemorrhage is based on the number of sites of hemorrhage: +, 1 to 2 sites; + +, 3 to 4; + + + 
5 to 7; + + ++, more than 7. 
thermia inasmuch as disturbances in liver 
function occur regularly after fever therapy 
(Wallace and Bushby60). 
It would appear that some factor other than 
shock also contributes to hemorrhage in heat 
shock, the same could not be said for the 
tissues at large. We believe that the other 
factor concerned is that of impaired coagula- 
tion of the blood. Wilson and Doan'” noted a 
reduction in platelets and prothrombin time in 
Fig. 42. Duration of illness, 18 hours. The mucosa of the ureter is the seat of widespread hemorrhage, 
and the wall is very edematous. X15.AIPACC. 118655. 
Fig. 43. Duration of illness, 12 hours. Haem casts (C) occupy two of the distal convoluted tubules of 
the kidney. The cytoplasm of the epithelial cells is fragmented and some of the nuclei are pyknotic. The 
capillaries are not engorged. X550. AIP Acc. 101159. 
Fig. 44. Duration of illness, 24 hours. Autopsy was performed 10 hours after death. A number of the 
distal convoluted tubules are necrotiq. Proximal convoluted tubules are dilated and the epithelial cells are 
flattened but well preserved. Congestion is of lesser degree in glomerular tufts than in intertubular vessels. 
X 280. AIP Acc. 99625. 440 
The Military Surgeon—November} 1946 
Fig. 45. Duration of illness, 34 hours. Autopsy was performed 12 hours after death. Two of the distal 
convoluted tubules contain well formed pigmented casts while in others there is considerable pigmented 
debris. Degenerative changes are present in the epithelial cells of these tubules. Below the glomerulus, two 
of the distal convoluted tubules are undergoing disintegration. Intertubular capillaries are somewhat en- 
gorged whereas the glomerulus is relatively ischemic. X300. AIP Acc. 97554. 
patients during artificially induced fever, and 
the few cases in our series in which such data 
were available showed low platelet counts and 
prolonged bleeding times. The degenerative 
changes in megakaryocytes we regard as a 
direct effect of heat. It may be of some sig- 
nificance that of the 3 cases in our series which 
did not display hemorrhage even though the 
blood pressure in 2 was too low to read (not 
known in the third), the temperature in all 
was relatively low. 
of sufficient acclimatization seems to have been 
a predisposing factor in many of the cases. 
An analysis of the clinical data establishes 
three categories of signs and symptoms: i) 
those due primarily to hyperthermia, especially 
central nervous system manifestations; 2) 
those due secondarily to shock; and 3) those 
due to complications, such as lower nephron 
nephrosis and bronchopneumonia, arising dur- 
ing the course of the illness. 
The clinical course was determined by the 
severity and duration of the primary symptoms, 
and was influenced largely by the factors of 
shock and complicating visceral disorders. 
Three types of course were observed: 1) an 
acute onset with early persistent coma or 
delirium; 2) an acute onset of early coma or 
delirium, with remission and late relapse; and 
3) an insidious onset with a progressive course 
and late development of coma. 
The duration of the disorder varied from 
This study is based on 125 fatal cases of 
heat stroke which occurred in the United 
States Army during the summer months of 
1941 to 1944. All were from military instal- 
lations in southern states of this country. In 
virtually all instances the soldiers were under- 
going strenuous muscular exercise under condi- 
tions of high environmental temperature. Lack 
Summary Figure 46. Duration of illness, 35 hours. There is well developed lower nephron nephrosis (hemo- 
globinuria nephrosis). The interstitial tissue is edematous, shows early stromal proliferation, and 
presents a sprinkling of inflammatory cells, mainly lymphocytes. Distal convoluted tubules and an 
occasional ascending limb contain casts: some are heme casts while others apparently are stained 
with bile pigment. (The patient had moderately severe centrolobular necrosis of the liver, but no 
evident jaundice.) The kidney as a whole was relatively ischemic. AIP Acc. 97148. Figure 55. Duration of illness, 32 hours. There appears to be a distinct increase in the number of 
megakaryocytes in the marrow. Most have lost their nuclei; the few nuclei remaining are greatly 
shrunken. AIP Acc. 86617. Fig. 47. Duration of illness, 6 hours. Owing- to advanced perisinusoidal edema the liver cords are widely- 
separated. A moderate dissociation of liver cords is visible. The sinusoids proper are dilated. The liver 
weighed 1750 gm, Xioo. AIP Acc. 98519. 
Fig. 48. Duration of illness, 276 hours. Liver cells in the central portion of the lobules are extensively 
degenerated and there is considerable fatty change. Liver cords of the lobular periphery show a moderate 
degree of dissociation. The icterus index was 16 and clinically there was a suggestion of jaundice. The 
Prussian blue reaction for iron was negative. The liver weighed 2350 gm. X no. AIP Acc. 118078. 442 
The Military Surgeon—November, 1946 
Fig. 49. Duration of illness, 96 hours. Autopsy was performed hours after death. Necrosis of lobules 
is far advanced. No jaundice was observed. X50. AIP Acc. 95093. 
Fig. 50. Duration of illness, 48 hours. The adrenal cortex is the seat of so-called tubular degeneration. 
Cells of the zona fasciculata, because of separation in the longitudinal axis of the columns, have taken on a 
tubular arrangement. A few of the cells seem to be undergoing disintegration. There is considerable peri- 
sinuco:dal edema. X350. AIP Acc. 95888. Heat Stroke 
443 
Fig. 51. Duration of illness, 35 hours. Numbers of necrotic cells of the mid-portion of the zona fasciculata 
are surrounded by smaller cells regarded as macrophages. The sinusoids are engorged. X500. AIP Acc. 
97148. 
Fig. 52. Duration of illness, 72 hours. Autopsy was performed hours after death. A large area of 
the adrenal cortex displays early necrosis. The intact cells are deficient in lipoids. X160. AIP Acc. 96554. The Military Surgeon—November, 1946 
444 
Fig. 53. Duration of illness, 130 hours. The adrenal cortex is depleted of lipoids and is the seat of focal 
necrosis. The cells in these foci have almost completely disappeared, leaving the empty stromal framework. 
X 130. AIP Acc. 115309. 
Fig. 54. Duration of illness, 6 hours. The bone marrow is greatly congested. The megakaryocytes show' 
advanced degenerative changes; two of them (A) have lost their nuclei, while in four others (B) only 
remnants of the nuclei remain. One young megakaryocyte (C) is visible. X650. AIP Acc. 99445. Heat Stroke 
445 
Fig. 56. Duration of illness, 26 hours. In this bone marrow section, four cells of the megakaryocytic 
series are present. The cell indicated by A is a megakaryoblast, by C a fully matured megakaryocyte, by B 
an intermediate developmental form, and by D a degenerating megakaryocyte. X1670. AIP Acc. 97555. 
less than an hour to twelve days; in 30 per 
cent it was more than 24 hours. 
Among the more frequent laboratory find- 
ings were the following: early-developing 
blood leukocytosis; normal hemoconcentration 
with a tendency to hydremia; a moderate rise 
in nonprotein nitrogen content of the blood 
and a lesser elevation of urea nitrogen, both 
apparent in the earlier stages of the disorder, 
and a decrease in C02 combining power. Too 
few figures were available to permit an assess- 
ment of the status of the blood chlorides. The 
changes appeared to be identical with those 
associated with shock. Platelets were con- 
sistently decreased in number early in the 
course of the disorder, and prothrombin and 
bleeding times were frequently increased later. 
The defect in blood coagulation was probably 
due to the thrombocytopenia rather than hypo- 
prothrombinemia, as the platelets sometimes 
reached critical levels whereas prothrombin 
time did not. This was regarded as a con- 
sequence of hyperthermia. Other laboratory 
findings consisted of elevated icterus index, and 
albumin, casts, and red blood cells in the urine 
in some cases of longer standing. 
Pathologic changes in the central nervous 
system were most conspicuous, and consisted 
of I) progressive degeneration of neurons and 
replacement by glia, especially in the cere- 
bellum, cerebral cortex, and basal ganglia, but 
not in the hypothalamus or the rest of the 
brain stem, the severity of the changes cor- 
responding to the length of survival after the 
occurrence of hyperthermia; 2) congestion, 
edema, and petechial hemorrhages, most com- 
monly in the region of the third ventricle, the 
aqueduct, and the fourth ventricle, all of which 
were inconstant and regarded as terminal. In 
our opinion the cellular changes were caused 
by excessive heat, and the hemorrhages by 
shock. The changes in thoracic and abdominal 
viscera also may be divided into two categories: 
hemorrhages and parenchymal lesions. Hemor- 446 
The Military Surgeon—November, 1946 
Fig. 57 
Fig. 58 Heat Stroke 
447 
rhages occurred in a wide variety of structures 
regardless of the duration of the illness, 
especially in the lungs, subepicardial and sub- 
endocardial tissues, and in the region of the 
bundle of His. Parenchymal lesions were pre- 
dominant in cases in which survival was more 
than 24 hours, consisting, in order of their 
frequency, of degeneration and subsequent re- 
generation of megakaryocytes, necrosis of 
heart muscle, lobular pneumonia, lower 
nephron nephrosis, centrolobular necrosis of 
the liver, and degenerative changes in the 
adrenal cortex. In some cases there were vary- 
ing combinations of these disorders. 
The clinical, laboratory and pathologic 
findings indicate that two factors are operative 
in heat stroke: hyperthermia and shock. It 
would seem that increased body temperature 
imparts to the disorder a specific character, 
affecting tissues of the organism to varying 
degrees. Our data yield no definite informa- 
tion on the mechanism underlying the tem- 
perature disturbance but the hypothesis is 
advanced that heat irreparably impairs the 
thermostatic function of the hypothalamus and 
that as a consequence the autonomic nervous 
system is no longer capable of reestablishing 
sweating or adequate peripheral circulation. 
Where anoxic anoxia or other conditions com- 
bined with severe shock leads to the same 
sequence of clinical events it is believed that 
the same mechanism is at fault. But although 
shock undoubtedly plays a significant role in 
the course of heat stroke, including an aug- 
mentation of body temperature owing to 
peripheral vasoconstriction, it is regarded as a 
secondary manifestation and therefore non- 
specific and unessential to the fundamental 
pathogenesis of the disorder. 
exposure to high temperatures upon the circulation 
in man, Am. J. Physiol. 67:573-588, 1924. 
3 Alpers, B. J.: Hyperthermia due to lesions in the 
hypothalamus, Arch. Neurol. & Psychiat. 35:30-42, 
1936. 
4 Antheaume, A., and Mignot, R.; Insolation et 
paralysie generale. Quelques particularity cliniques, 
Encephale 3:493-591, 1908. 
■'Aron, H.; Investigation on the action of the 
tropical sun on men and animals, Philippine J. Sc., 
B., Med. Sc. 6:101-132, 1911. 
0 Bazett, H. C., Sunderman, F. W., Doupe, J., and 
Scott, J. C.; Climatic effects on volume and com- 
position of blood in man, Am. J. Physiol. 129:69-83, 
1940. 
7 Bean, W. B., and Eichna, L. W.; Performance in 
x’elation to environmental temperature. Reactions of 
normal young men to simulated desert environment, 
Federation Proc. 2:144-158, 1943. 
8 Beaton, L. E., and Herrmann, J. D.: Hyper- 
thermia following injury of the preoptic region, 
Arch. Neurol. & Psychiat. 53:150-151, 1945. 
0 Blalock, A., and Price, P. B.: Panel discussions} 
traumatic shock—early signs, prevention and treat- 
ment, Bull. Am. Coll. Surgeons 27:102-105, 1942. 
10 Blum, H. F.: The physiological effects of sun- 
light on man, Physiol. Rev. 25:483-530, 1945. 
11 Blum, H. F.: The solar heat load} its relation- 
ship to total heat load and its relative importance 
in the design of clothing, J. Clin. Investigation 24: 
712-721, 1945. 
12 Castellani, A.; Climate and Acclimatization, ed. 
2, London, Bale & Curnow, 1938. 
13 Chakravarti, D. N., and Tyagi, N.: Studies in 
effects of heat, Ind. J. Med. Res. 25:791-827, 1938. 
11 Cournand, A., Riley, R. L., Bradley, S. E., 
Breed, E. S., Noble, R. P., Lauson, H. D., Gregersen, 
M. I., and Richards, D. W.: Studies of the circulation 
in clinical shock, Surgery 13:964-995, 1943. 
15 Cullen, S. C., Weir, E. F., and Cook, E.: The 
rationale of oxygen therapy during fever therapy, 
Arch. Phys. Therap. 23:529-535, 1942. 
16 Davison, C.: Disturbances of temperature regula- 
tion in man, A. Research Nerv. & Ment. Dis., Proc. 
20:774-823, 1940. 
11 Fleck, U., and Hiickel, R.: Zur Klinik und 
Pathologic des Hitzschlages, Deutsche Ztschr. f. 
Nervenh. 117:113-137, 1931. 
18 Freeman, W., and Dumoff, E.; Cerebellar syn- 
drome following heat stroke, Arch. Neurol. & 
Psychiat. 51:67-72, 1944. 
13 Gauss, H., and Meyer, K. A.: Heat stroke: re- 
1 Adolph, E. F.: Physiological fitness for the desert, 
Federation Proc. 2:158-165, 1943. 
"Adolph, E. F., and Fulton, W. B.: The effects of 
References 
Fig. 57. Duration of illness, 74 hours. Between fat cells of the marrow interstices there is marked depletion 
of hematopoietic cells of all categories. The section was taken from a rib, the marrow of which should be 
nearly 50 per cent cellular in this age group. X 650. AIP Acc. 98544. 
Fig. 58. Duration of illness, hours. The mucosa of the stomach is the seat of deep punctate and 
superficial diffuse hemorrhage. The submucosa is congested and edematous. X30. AIP Acc. 101019. 448 
The Military Surgeon—November, 1946 
port of one hundred and fifty-eight cases from Cook 
County Hospital, Chicago, Am. J. M. Sc. 154:554- 
564, 1917. 
20 Gemmill, C. L.: Physiology in Aviation, Spring- 
field, Illinois, Charles C Thomas, 1943. 
21 Gildea, E. F., and Cobb, S.: The effects of 
anemia on the cerebral cortex of the cat, Arch. 
Neurol. & Psychiat. 23:876-903, 1930. 
22 Goebel, A. J. L.: Ueber die Nachkrankheiten 
des Hitzschlages, Berlin, O. Franke, 1905. 
23 Hall, W. W., and Wakefield, F. G.: A study of 
experimental heatstroke, J. A. M. A. 89:177-182, 
1927. 
24 Hartman, F. W., and Major, R. C.: Pathological 
changes resulting from accurately controlled artificial 
fever, Am. J. Clin. Path. 5:392-410, 1935. 
25 Heilman, M. W., and Montgomery, E. S.: Heat 
disease, J. Indust. Hyg. & Toxicol. 18:651-667, 
1936. 
26 Horvath, S. M., Dill, D. B., and Corwin, W.; 
Effects on man of severe oxygen lack, Am. J. Physiol. 
138:659-668, 1943. 
27 Hyndman, O. R., and Wolkin, J.: The automatic 
mechanism of heat conservation and dissipation, Am. 
Heart J. 289-304, 1941. 
28 Keller, A. D.; Separation in the brain stem of the 
mechanisms of heat loss from those of heat produc- 
tion, J. Neurophysiol. i:543-557> 1938. 
29 Kopp, I.; Metabolic rates in therapeutic fever, 
Am. J. M. Sc. 190:491-501, 1935. 
30 Kopp, I., and Solomon, H. C.; Shock syndrome 
in therapeutic hyperpyrexia, Arch. Int. Med. 60: 
597-622, 1937. 
31 Ladell, W. S. S., Waterlow, J. C., and Hudson, 
M. F.: Desert climate. Physiological and clinical 
observations, Lancet 2:491-497, 527-5 31, 1944. 
32 Manson, P.: Manson’s Tropical Diseases. A Man- 
ual of the Diseases of Warm Climates. Edited by 
P. H. Manson-Bahr. Baltimore, Williams & Wilkins, 
1940. 
33 Marsh, F.: The etiology of heat stroke and sun 
traumatism, Tr. Roy. Soc. Trop. Med. & Hyg. 
24:257-288, 1930. 
34 M’Kendrick, J. G.: A case of meningo-cerebritis, 
caused probably by exposure to the sun, Edinburgh 
M. J. 14:517-522, 1868. 
35 McKenzie, P., and LeCount, E. R.; Heat stroke, 
J. A. M. A. 71:260-263, 1918. 
38 Messiter, A. F.: A case of insolation accom- 
panied by hemiplegia, Lancet 1:1741-1742, 1897. 
37 Moon, V. H.: The vascular and cellular dy- 
namics of shock, Am. J. M. Sc. 203:1-18, 1942. 
38 Moon, V. H.; Shock. Its Dynamics, Occurrence 
and Management. Philadelphia, Lea & Febiger, 1942. 
39 Morgan, L. O.: Cell changes in some of the 
hypothalamic nuclei in experimental fever, J. Neuro- 
physiol. 1:281-285, 1938. 
40 Morgan, L. O., and Yonderahe, A. R.: The 
hypothalamic nuclei in heat stroke, Arch. Neurol. & 
Psychiat. 42:83-91, 1939. 
41 Omorokow, L.: Ueber den Einfluss holier Tem- 
peraturen auf das Zentralnervensystem des Kanin- 
chens, Histol. u. Histopath. Arbeiten, Nissl-Alzheimer 
6:1-32, 1913. 
4‘ Ranson, S. W.: Regulation of body temperature, 
A. Research Nerv. & Ment. Dis., Proc. 20:342-399, 
1940. 
Reid, W. D.; Pneumonia not a rare complica- 
tion of heat prostration, Boston M. & S. J. 167:217- 
219, 1912. 
Rich, A. R.: Peculiar type of adrenal cortical 
damage associated with acute infections, and its 
possible relation to circulatory collapse, Bull. Johns 
Hopkins Hosp. 74:1-15, 1944. 
45 Richards, D. W., Jr.; The circulation in trau- 
matic shock in man, Bull. N. Y. Acad. Med. 20:363- 
3935 1944- 
46 Rosenbaum, L.: Significance of salt (NaCl) in 
torrid temperatures, Military Surgeon 98:43-47, 
1946. 
47 Rosenblath: Zur Pathologic der Encephalitis 
acuta, Deutsche Ztschr. f. Nervenh. 50:342-405, 
I9I4- 
48 Schickele, E.: Environment in fatal heat stroke. 
An analysis of 157 cases occurring in the Armed 
Forces in the United States during World War II, 
Military Surgeon, Dec. 1946. 
40 Schwab, W.: Ueber Hirnveranderungen bei 
Sonnenstich, Schweiz, med. Wchnschr. 55:33-38, 
1925. 
50 Schwartz, P.: Die Arten der Schlaganfalle des 
Gehirns und ihre Entstehung, Berlin, J. Springer, 
1930. 
01 Shepherd, H. M. D.: Effects of heat in Iraq, J. 
Roy. Army Med. Corps 84:1-8, 1945. 
52 Steinhausen, F. A.; Nervensystem und Insola- 
tion, Berlin, A. Hirschwald, 1910. 
M Stern, K.: l)ber Kreislaufstorungen im Gehirn 
bei Wandeinrissen in extracerebralen Arterien, Ztschr. 
f. d. ges. Neurol, u. Psychiat. 148:55-82, 1933. 
54 Stewart, R. M.: On the occurrence of a cerebellar 
syndrome following heat stroke, Rev. Neurol. & 
Psychiat. 16:78-93, 1918. 
56 Tavel, F. von: Die Auswirkungen des Sauer- 
stoffmangels auf den menschlichen Organismus bei 
kurzfristigem Aufenthalt in grosser Hohe. Ein Bei- 
trag zur Frage der Leistungsfahigkeit 5m Hohenflug, 
Helv. Physiol. Pharmacol. Acta (supp. 1), pp. 1-128, 
1943- 
58 Titrud, L. D., and Haymaker, W.: Cerebral 
anoxia from high altitude asphyxiation, Arch. Neurol. 
& Psychiat. (in Press). 
57 van Bogaert, L., Dalleiuagne, M. J., and Wegria, 
R.: Recherches sur le besoin d’oxygene chronique et 
aigii chez Macacus rhesus. Absence de lesions ex- 
perimentales des centres nerveux apres intoxication Heat Stroke 
449 
par Poxyde de carbone, le nitrite de soude et Papauv- 
rissement de Pair en oxygene, Arch, internat. de Med. 
exper. 13:335*378, 1938. 
68 Wakefield, E. G., and Hall, W. W.: Heat in- 
juries. A preparatory study for experimental heat- 
stroke, J.A.M.A. 89:92-95, 1927. 
A. W.: Heat exhaustion, Military 
Surgeon 93:140-146, 1943. 
60 Wallace, J., and Bushby, S. R. M.: Hazards of 
hypertherm treatment, Lancet 2:459-464, 1944. 
61 Ward, R. L., and Olson, O. C.: Report of a case 
of severe anoxic anoxia with recovery, J. Aviation 
Med. 14:360-365, 1943. 
62 Weiner, J. S.: Experimental study of heat col- 
lapse, J. Indust. Hyg. & Toxicol. 20:389-400, 1938. 
83 Weisenburg, T. H.: Nervous symptoms follow- 
ing sunstroke, J.A.M.A. 58:2015-2017, 1912. 
64 Wilcocks, C.: Heat stroke and allied conditions, 
Practitioner 147:463-468, 1941. 
4,5 Willcox, W. H.: The nature, prevention, and 
treatment of heat hyperpyrexia; the clinical aspect, 
Brit. M. J. 1:392-397, 1920. 
06 Wilson, G.: The cardiopathology of heat-stroke, 
J.A.M.A. 114:557-558, 1940. 
6‘ Wilson, S. J., and Doan, C. A.; The pathogenesis 
of hemorrhage in artificially induced fever, Ann. Int. 
Med. 13:1214-1229, 1939. 
68 Windle, W. F., and Becker, R. F., and Weil, A.: 
Alterations in brain structure after asphyxiation at 
birth, J. Neuropath. & Exper. Neurol. 3:224-238, 
1944. 
°J Wright, R. D., and Devine, J.: Body tempera- 
tures in shock, M. J. Australia 1:2i-2j, 1944. 
70 Wright, D. O., Reppert, L. B., and Cuttino, 
J. T.: The purpuric manifestations of heat stroke. 
A report of prothrombin and platelet studies in 
twelve cases, (in preparation). 
71 Zimmerman, H. M.; Temperature disturbances 
and the hypothalamus, A. Research Nerv. & Ment. 
Dis., Proc. 20:824-840, 1940. FATAL PULMONARY EMBOLISM IN 100 BATTLE CASUALTIES* 
By CAPTAIN TOM R. HAMILTON,** and COLONEL D. MURRAY ANGEVINE,f 
Medical Corps, Army of the United States 
(With two illustrations) 
NOT long after D day in Normandy 
an apparent disproportionate number 
of deaths from pulmonary embolism 
following battle wounds aroused interest at the 
where the majority of these cases 
were first reviewed. Because pulmonary em- 
bolism seemed a more frequent cause of death 
than accumulated experience would have sug- 
gested, a survey was made of 1065 autopsies 
on cases in which battle wounds had recently 
been sustained. It was found that in 66, or 
6.19 per cent of these, death was due to 
pulmonary embolism. These cases, augmented 
by 34 others available for study at the Army 
Institute of Pathology, form the basis for this 
paper. 
It is necessary to have it clearly understood 
at the outset that the term “fatal pulmonary 
embolism” has been employed advisedly in 
this report which is based on the analysis of 
autopsy protocols. This usage is consistent with 
the recent interpretation of Murnaghan, Mc- 
Ginn and White1 (1943), who write, “It is 
emphasized that ‘pulmonary embolism’ and 
‘acute cor pulmonale’ are not synonymous 
terms.” Acute cor pulmonale was described 
originally by McGinn and White in 1935P 
The first report of pulmonary embolism 
was made by Virchow5 in 1846. Trauma, 
childbirth, and disease were the predisposing 
conditions before the days of modern surgery. 
The literature has recently been thoroughly 
reviewed by McCartney,4 who also presented 
the results of an analysis of 25,771 necropsies 
in which the incidence of fatal pulmonary em- 
bolism was 2.6 per cent. In an exhaustive study 
of fatal pulmonary embolism following trauma 
published in 1935, McCartney5 reported an 
incidence of 4.0 per cent in post-traumatic 
cases in contrast to 2.6 per cent in cases without 
trauma. He explained the disparity in sex inci- 
dence in this series (2.5 per cent in males and 
8.6 per cent in females) on the grounds of 
the greater prevalence of pre-existing vari- 
cosities in women. 
The largest series of cases of post-traumatic 
fatal pulmonary embolism previously reported 
are those of McCartney4 and Vance.6 Their 
observations present certain basic differences 
from, and few analogies with, ours. In Mc- 
Cartney’s series, 33 of 61 were in men, only 
3 of whom were under 40 years of age. In 
Vance’s series, 23 of the 60 were males; 14 
were less than 50, and only 4 less than 40 
years of age. Our series, drawn from combat 
troops, was comprised entirely of men, 1 of 
whom was over 40. 
A possible virtue of an analysis of battle 
casualties is the absence of such usually 
plaguing variables as age and previous or in- 
tercurrent disease since all had presumably 
been physically fit soldiers. Sex and color as 
influencing factors are eliminated, because all 
of these soldiers happened to be white. Study 
of the relationship of trauma to thrombosis and 
embolism is also possible, since all emboli fol- 
lowed injury. 
The preliminary observation that pulmonary 
embolism was more frequent when wounds 
were in the region of the leg and hip was 
borne out by subsequent study. The percentage 
of leg and hip wounds in the entire battle 
casualty group was 29 per cent, whereas it 
was 53 per cent (calculation based upon site 
of single wounds and fractures) in cases of 
fatal pulmonary embolism. Figures from 
World War I,7 indicate a high mortality fol- 
lowing wounds of the lower extremities. Of 
all the fatalities resulting from battle wounds 
* From Army Institute of Pathology, Washington, 
D.C. 
** Department of Pathology, The University of 
Kansas, Kansas City, Kansas. 
t Professor of Pathology, The University of Wis- 
consin, Madison, Wisconsin. 
If First Medical General Laboratory in England 
and France. Fatal Pulmonary Embolism in 100 Battle Casualties 
451 
among the American troops in Europe from 
1917 to 1919, 42.2 per cent followed injury 
to the lower extremities; of course gas gan- 
grene was responsible for a majority of these. 
An analysis of the material in this series 
has been attempted for the purpose of deter- 
mining the role of certain factors in fatal 
pulmonary embolism. These include: (1) age; 
(2) type of case; (3) distribution of wounds; 
(4) site of fractures; (5) interval between 
injury and death; (6) associated factors pos- 
sibly precipitating attacks; (7) symptoms; (8) 
duration of pulmonary embolism; (9) inci- 
dence and location of pulmonary infarcts; 
(10) site of massive emboli; (11) alteration 
of the heart; (12) location of phlebothrom- 
bosis, and (13) relationship of anatomic con- 
siderations to thrombosis and embolism. 
Age. The limits of variation in age of 89 
cases in which it was recorded are 17 and 41 
years, and, if the 6 prisoners of war are ex- 
cluded, the range is from 19 to 38 years. The 
mean calculated from either set of figures is 
26 years (Table I). All except 7 per cent of 
these men were in their twenties or thirties 
and were without known physical disabilities 
at the time they were wounded, making this 
series unlike others reported, since it com- 
prised young healthy men. 
Type of Case. To give a proper perspective 
on the problem the cases may be classified 
from the point of view of the clinician. There 
are three major groups of patients for whom 
immobilization is a necessary concomitant of 
treatment; those (1) in casts, (2) with post- 
operative laparotomies, and (3) with pene- 
trating wounds of the chest. 
The significance of immobilization and lack 
of exercise as underlying factors has been 
clearly indicated by the clincial study of Potts 
and Smith.8 Although many protocols omit 
specific designation of treatment, in general 
the data from our series support their findings. 
At least one-fourth of the patients with fatal 
pulmonary embolism had been immobilized 
by casts, usually on the lower extremities, 
body, or both. Two had casts on the arms. 
Traction was being applied for at least two 
fractures of the leg at the time of death. 
Laparatomy had been performed in 24 cases. 
In 23 a serious consideration was penetrating 
wounds of the chest. 
Attention is focused again on the role of 
trauma of the lower extremities by the 14 
deaths from pulmonary embolism following 
amputation. Shock was prominent early in 
most cases, particularly in those of traumatic 
amputation. Other possible predisposing factors 
Table I 
Decade in Which Fatal Pulmonary 
Embolism Occurred 
Decade 2nd 
3rd 
4th 
5th 
Number of Deaths 6 
54 
28 
1 
were extensive local operative procedures, in- 
cluding those on blood vessels. Four patients 
were in serious condition because of penetrat- 
ing wounds of the bead, and 3 because of 
damage to the spinal cord. 
In 9 of the cases it was stated that the 
patient had been in severe shock. Transfusions 
are known to have been given in 37 per cent, 
but there is a lack of recorded information in 
regard to blood and plasma, which almost 
routinely have been given such wounded 
patients. 
Overlapping conditions placed 14 cases in 
more than one category. In only 3 was there 
no evident predisposing factor for thrombosis 
such as venous stasis or damage to tissue or 
blood vessels. 
It is significant that 91 of the wounds were 
either penetrating or perforating. The type of 
wound and extent of damage to soft tissues 
and bones are largely dependent on the nature 
of the missiles producing them. Data on the 
missiles were given in 75 of the protocols and 
in general they could be grouped as shell frag- 
ments or bullets. In 58 cases there were shell 
fragments; high explosive from such sources 
as 88 mm. guns, cannons, German bazookas, 
aerial bombs, and also shrapnel and mines. 
Bullet wounds were recorded in 14 cases, 8 
from machine gun and 6 from rifle; 3 were 
classified simply as gunshot wounds. 
Distribution of Wounds. Although evalua- 452 
The Military Surgeon—N ov ember, 1946 
tion of the relationship of local trauma to 
thrombosis and embolism is highly desirable, 
the multiplicity of wounds in individual cases 
complicated such an analysis (Table II). 
Eighty-two of 211 wounds in the 100 cases 
were in the lower extremities, 13 in the but- 
tock, 8 in the pelvis, and 3 in the hip. The 
upper extremities were involved in 20 in- 
stances. Chest wounds, most of which were 
penetrating, were present in 28, while the 
of fracture in this region. Pulmonary embolism 
occurred subsequent to 22.2 per cent of the 
157 fractures in the lower extremities in Mc- 
Cartney’s5 post-traumatic cases. He observed 
that pulmonary embolism was rare after frac- 
tures in other locations. Age was a contributing 
factor in his cases. The prolonged rest in bed 
and strict immobilization attendant on treat- 
ment of fractures of the lower extremities tend 
to promote stasis which is of primary im- 
portance in the production of phlebothrom- 
bosis. 
In our series there were fractures in 65 
cases, some of them multiple since there were 
81 in all. There were 37 fractures of the 
lower extremity, preponderantly of the thigh 
(24). Twenty-three of the fractures of the 
lower extremity were on the right side, in 
contrast to 14 on the left. In 55 there was a 
single site of fracture, so that in this group a 
correlation of injury with local vascular throm- 
bosis might be attempted. 
It is to be expected that in cases of fractured 
vertebrae (13), pelvis (8), and skull (4) 
venous stasis might have developed or vascular 
damage been sustained at the time of injury. 
With the penetrating wounds of the chest, 
the fractures were of the ribs, scapulae, and 
vertebrae. Broken ribs may seem trivial unless 
one remembers that among the most grave 
are cases in which the missile passed through 
the thoracic cage in its course to deeper struc- 
tures. 
Interval from Injury until Death. The time 
interval from injury until death was 2 to 91 
days; the mean was 18 and the average 20.7 
days, both of which fall in the third week 
(Table III). The length of these intervals 
was the only clue to the time required for 
thrombus formation in 90 per cent of cases 
in which there was no evidence of embolism 
before the fatal attack. It may appear dubious 
to suggest stasis as the cause of phlebothrom- 
bosis in the short periods (e.g. 3 days) noted 
in 3 cases. Certain fatalities incident to this 
war, however, emphasize the lethal effects of 
stasis. A number of ambulatory but elderly 
persons have been victims of fatal pulmonary 
embolism, the result of stasis attributed to 
Table II 
Distribution of Wounds and Fractures 
Site of Wounds 
Number 
Site 
of 
Fractures 
I. Head and Neck 
13 
4 
Skull, penetrating wounds 
( 4) 
(4) 
2. Chest 
28 
12 
3. Abdomen 
20 
4. Back (including all verte- 
brae) 
24 
13 
5. Pelvis, Buttock and Hip 
24 
8 
Pelvis 
( B) 
( 8) 
Buttock and Hip 
(16) 
— 
6. Lower Extremity 
82 
37 
Right 
(44) 
(23) 
Left 
(37) 
(14) 
Thigh 
(50 
(24) 
Leg and Foot 
(30 
(13) 
7. Upper Extremity 
20 
8 
211 
82 
abdomen was the site of 20, and the back of 
24. There were 8 wounds of the head and 5 
of the neck. 
A single area only was involved in 50 cases. 
In this group there is an opportunity for cor- 
relation of wounds with vascular lesions; how- 
ever, the relatively small number of cases 
limits the significance of such a correlation. 
Wounds of the lower extremities were ob- 
served in one-half (26 cases); penetrating 
wounds of the chest in 7; wounds of the ab- 
domen in 3, and of the pelvic region in 4. 
Site of Fractures. Twelve deaths following 
fractures have been reported by Lister9 in a 
series of 281 cases of fatal pulmonary em- 
bolism; 9 followed fractures of the femur, 
giving an incidence of 11 per cent in 80 cases Fatal Pulmonary Embolism in 100 Battle Casualties 
453 
pressure on the backs of their legs from wooden 
crossbars in their deck chairs, while in air raid 
shelters for relatively long intervals.10 
Associated Factors, Possibly Precipitating 
Attacks. In connection with the attacks in 
about 30 per cent of cases in our series, there 
were incidents which may be interpreted as 
associated, or possibly precipitating, factors. 
Some of these appear trivial and may be coinci- 
dental as they constitute parts of the daily 
routine of hospitalization. They are presented 
as they were noted by the attending Medical 
patently well on the road to recovery, was 
building a fire; one was eating, and another 
had just finished a meal. 
There were still other variations of the in- 
cidents which immediately preceded the fatal 
pulmonary embolism. One soldier had had a 
massage because of a cramp in his leg; another 
with pneumonia had been coughing and vomit- 
ing; a third had received a transfusion. 
Symptoms. Clinical evidence of pulmonary 
embolism indicated by the symptomatology 
has been analyzed in the 78 cases in which 
Table III 
Time of Death in Relation to Date of Injury 
Time in Weeks after 
Injury 
ist 2nd 
3rd 4th 
5th 6th 
7th 
8th 9th 10th 11 th 12th 13th 
Number of cases 
11 26 
26 17 
10 2 
2 
0 0 1 0 2 1 
Officer. 
Bowel movements were apparently related 
to 6 of the episodes; in 4, the fatal embolism 
occurred while the patient was on the bed 
pan, in 2, the patients had just returned from 
the latrine. One fatality took place after blad- 
der irrigation. 
Manipulation of a cast had been carried out 
in 6 cases a few minutes or hours before the 
accident. The attack took place after the cast 
had been changed in 4 instances; after it was 
cut in I, and while it was being cut in an- 
other. 
Passive activity was associated with fatal 
pulmonary embolism in 4 instances. One pa- 
tient was being lifted; another fell out of bed; 
a third had just been transferred by ambulance 
from one hospital to another; and a fourth 
had been moved from a private room to a 
ward. In 3 cases the patient had returned 
from the operating room a few minutes or 
hours before he died of pulmonary embolism, 
but it was not possible to determine whether 
local manipulation or passive activity was the 
precipitating factor. 
Exertion was associated with the attack in 
4 cases. Two patients were out of bed, one 
was up in a wheel-chair, while a third, ap- 
the history was detailed. Dyspnea dominated 
the picture, being present in about three- 
fourths of the cases. Although pain was a 
symptom at some time in approximately two- 
thirds, pain in the chest was observed in only 
one-third (33) during the terminal attack. 
Pain in the abdomen or leg was a conspicuous 
feature in approximately 10 per cent during 
the fatal episode. Abdominal pain was ob- 
served in 12 cases; in 5 it occurred in the 
absence of pain in the chest; in 11 without 
pain in the legs. In 5 cases the abdominal pain 
preceded the terminal attack, and in 3 it re- 
curred during the fatal episode. The signifi- 
cance of abdominal pain has been discussed 
by Middleton11 who interpreted it as based 
upon the vagus reflex from the pulmonary 
artery to the gastro-intestinal tract. 
Manifestations of shock were observed in 
38 and cyanosis in 36 instances. 
The features of hemoptysis and cough (7 
and 5 respectively) were evident in less than 
10 per cent of the attacks and, with one excep- 
tion, were limited to the minority of cases of 
longer standing or of repeated pulmonary em- 
bolism in which actual infarction of the lung 
was present. 
Duration of Pulmonary Embolism. Attacks 454 
The Military Surgeon—November, 1946 
of pulmonary embolism persist for a matter of 
minutes, hours, or days, and are usually fatal 
if the embolus is large. It may be said to be 
the rule that the larger the embolus or mass of 
emboli, the shorter and more likely fatal the 
attack. 
In this series, the duration of the terminal 
attack may be tabulated by time intervals in 
93 of the 100 cases. In 91, death occurred in 
chest previous to the terminal attack had 
pulmonary infarcts, but 5 of these had no 
hemoptysis. Pulmonary infarction was present 
in 10 and hemoptysis in only 2 of the 24 cases 
in which pain in the chest occurred only in 
the terminal attack. 
Site of the Embolus. The terminal massive 
pulmonary embolisms viewed at autopsy were 
without the definite pattern of distribution 
within the major pulmonary vessels which 
was characteristic of infarcts when embolism 
had occurred earlier. (See figures 1 and 2.) 
The Heart. An evaluation of a possible role 
of cardiac failure or enlargement offers many 
difficulties. Records in 57 cases show great 
variation in weight of the heart. One weighed 
620 gm.; 8 weighed between 450 and 500 
gm.; 8 between 400 and 450 gm.; 10 be- 
tween 350 and 400 gm.; 21 between 300 
and 350 gm., or nearest to the weight of 316 
gm., which is accepted as normal; while only 
9 weighed less than 300 gm. The mean 
weight was 340 gm. Body weights were re- 
corded in only some of the protocols. 
To determine whether strenuous training 
for endurance and other conditions peculiar to 
this conflict may have affected the cardio- 
vascular system is not within the province of 
Table IV 
Duration of Fatal Attack 
Time 
No. of 
Cases 
Less than 5 minutes 
33 
From 5-30 minutes 
29 
From 30-60 minutes 
From 1-24 hours 
19 
Total 
91 
less than 24 hours; the majority (72) did not 
survive over one hour (Table IV). Two 
patients had typical attacks and then went 
downhill gradually, one dying 8, the other 31 
days later. In 10 cases there was evidence of 
pulmonary embolism prior to the terminal at- 
tack (Table V). 
Incidence and Location of Pulmonary In- 
farcts. The incidence of pulmonary infarction 
in this series, as seen at autopsy, was 33 per 
cent. The location of the infarcts has been 
analyzed insofar as information allowed. They 
were limited to single lobes in 19 of the 33 
cases, and were multiple in 14. 'Die right 
lung was the site of infarction in 24 cases, 
while the left lung was involved in only 9. 
Localization of infarction in the lower lobe 
was demonstrated in all except 1 case. 
Twelve cases with pulmonary infarcts were 
without history of pain in any location, while 
15 were without that of pain in the chest. 
Pulmonary infarction was demonstrated in 
only 18 of the 33 cases in which chest pain 
was a symptom, so in 15 there appeared to 
be no relationship between pain in the chest 
and infarction. 
Table V 
Interval Between Primary and 
Terminal Attacks 
Time 
No. of 
Cases 
Less than 24 hours 
2 
From 1- 5 days 
2 
From 6-10 days 
2 
Over 11 days 
4 
Total 
10 
this paper. It is possible, however, that the 
figures might be appreciably different from 
what would be found in a similar group of the 
same age and physical standard without mili- 
tary training. 
Dilatation of the right ventricle was de- 
scribed in 27 cases; however, such an ob- 
Eight of the 9 patients with pain in the Fatal Pulmonary Embolism in 100 Battle Casualties 
455 
Fig. 1, Emboli in pulmonary arteries, demonstrated in situ in gross specimen 
Photograph AIP Acc. 97602 
servation is all too frequently made by prosec- 
tors without adequate evidence. Measurements 
of tricuspid and pulmonary valves suggest some 
increase in size; but the records are again 
incomplete, with figures for only 37 and 33 
respectively. Considerable variation in meas- 
urements are inevitable because of the numer- 
ous pathologists involved. It is our opinion 
that it would be ill-advised to attempt to draw 
conclusions from such incomplete and incon- 
sistent data. 
Location of the Phlebothrombosis. Venous 
thrombosis, which may be regarded as the 
source of the embolism, was demonstrated in 
62 cases, in which 113 separate thrombi were 
observed. In 46 the femoral vein or its tribu- 
taries were involved, and in 46 the iliacs, so 
that together they represent 81.4 per cent of 
the entire number. 
The greater prevalence of thrombosis in 
the lower extremity has been reported rather 
consistently and has some foundation on ex- 
perimental evidence.8 It has been demonstrated 
that throttling the blood flow by double partial 
ligations produced thrombosis in 13 of 2 2 
experiments on the femoral vein in dogs, while 
only I small thrombus was produced in 14 
instances of similar interference with circula- 456 
The Military Surgeon—November, 1946 
Fig. 2. Embolus coiled in pulmonary artery demonstrates eccentric pattern of lamination under 
low power. Photomicrograph AIP Acc. 6122. 
tion through the external jugular vein. It was 
believed that impossibility of exercising the legs 
in the cramped quarters of the cage was re- 
sponsible for femoral thrombosis; however, as 
the dogs were able to move their heads freely, 
the jugular veins were almost free of thrombi. 
In our series the left side predominated as 
the location of the phlebothrombosis even 
though there were more wounds and fractures 
on the right side. Most significant is the strik- 
ing localization of thrombosis of veins on the 
left in cases of single penetrating wounds of 
the chest indicated by 5 instances on that side, 
with none on the right except once when 
thrombi were bilateral. The splinting of the 
diaphragm and stasis in the circulation of the 
lower extremity are the primary factors in the 
absence of localized trauma in these cases. 
It appears that thrombosis is as likely to 
occur on the left as on the right in cases of 
wounds of the right lower extremity, but in 
the small number of single wounds on the left, 
thrombosis was found on the right only once 
and then it was bilateral. One may observe 
that there is a tendency for thrombosis to occur 
in the more central vessels in cases of abdomi- Fatal Pulmonary Embolism in 100 Battle Casualties 
457 
nal wounds. Perhaps this is due directly to 
trauma or to infection when there are lacer- 
ations of the bowel. 
In only about 10 per cent of the cases had 
there been forewarning of possible throm- 
bophlebitis, and in these the venous thrombosis 
follows the pattern described by Ochsner and 
DeBakey12 and appears to be consistent with 
the conception of Homans.13’ 14 
Relationship of Anatomic Considerations to 
the Thromhosis. The predominance of throm- 
bosis on the left in veins of the lower extremi- 
ing the stream, and explains the well-known 
frequency of thrombosis in the left lower 
limb.” A further agreement between the find- 
ings in the present study and his observations 
lies in the striking predominance of left-sided 
thrombosis at the level of the iliacs (internal 
and external). Aschoff continues, “The 
thrombosis on the right side extends up to 
Poupart’s ligament, whereas on the left side 
it extends up to the point of compression of 
the left iliac vein by the right iliac artery.” 
The analysis by Dietrick16 of thrombosis fol- 
Table VI 
Location of Thrombi* 
(Relationship to Crossing of Common Iliac Vein Behind Artery) 
Right 
Not Stated 
Left 
Iliacs (Portion not stated) 
4\TT 
4\ 
s\.. 
Common Iliac Vein 
(11 
7 \ 
o/4 
6 r 
Internal Iliac Vein 
a\ 
lol 
External Iliac Vein 
ij3 
«P 
7/'7 
Femoral Vein 
it) , 
o\ 
I9' 
Tributaries of Femoral Vein 
si16 
fO 
°J 
h/30 
Total 
30 
6 
58 
* (The figures are derived from total distribution data.) 
ties in contrast to the greater frequency of 
wounds on the right is demonstrated by the 
findings in our series (Table VI). The an- 
atomic basis for this striking disparity is funda- 
mental and is found in the relationships and 
structure of the common iliac vein on the left. 
This vessel crosses behind the artery, passes 
over the sacrum to the right side of the verte- 
bral column, thus running a longer and more 
oblique course than its fellow on the right. 
Aschoff10 in 1924 stressed conditions favor- 
ing thrombosis, including an anatomic basis 
for localization. Our observations are in agree- 
ment with the concept he has expressed; 
“When lying on the back the increased com- 
pression of the left iliac vein by the arterial 
trunk (right iliac, middle sacral and left hypo- 
gastric arteries) has a direct influence in slow- 
lowing battle casualties is of interest. It indi- 
cated the dominant role that trauma plays as 
a causative factor, whereas in peace time the 
factors of stasis and infection are the more 
prominent. The considerations of Aschoff and 
Dietrick apply to the general problem of 
thrombosis and are not limited, as in the 
present study, to cases of fatal pulmonary em- 
bolism. It cannot be concluded that post- 
traumatic infection has been absent in all cases 
of our series but it was not a notable feature. 
It should be remembered that stasis as a 
factor in thrombosis is not limited to the aged 
and infirm, but is also incident to conditions 
of combat. 
Summary, (i) One hundred cases of fatal 
pulmonary embolism in battle casualties are 
reported. They occurred in young, previously 458 
The Military Surgeon—TVovember, 1946 
healthy men and thus permitted observations 
without the usual variables and complicating 
factors of age and vascular disease. 
(2) Venous stasis, caused most frequently 
by immobilization, appears to be the most im- 
portant factor in the production of phlebo- 
thrombosis. 
(3) Pulmonary infarcts were located pre- 
dominantly on the right side and in the lower 
lobes. 
(4) The prevalence of venous thrombosis 
on the left side of the body did not have a 
direct relationship with trauma which occurred 
more frequently on the right side. 
(5) The distribution of the venous throm- 
bosis seems to be based upon the anatomic re- 
lationships at the level of the common iliac 
vein. 
4 McCartney, J. S.; Postoperative pulmonary em- 
bolism, Surgery, 17:191, 1945. 
J McCartney, J. S.: Pulmonary embolism following 
trauma, Surg. Gynec. & Obst., 61:369-379, 1935. 
6 Vance, B. M.; Thrombosis of veins of the lower 
extremity and pulmonary embolism as complication 
of trauma, Am. J. Surg. 26:19-26, 1934. 
7 The Medical Department of the U.S. in the 
World War, vol. XV, Statistics, Part 2, Medical and 
Casualty Statistics, War Department, 1925. 
8 Potts, W. M. and Smith, S.: Pulmonary em- 
bolism, clinical and experimental study, Arch. Surg. 
46:27-39, 1943. 
J Lister, W. A.: Causation of pulmonary embolism 
following operation, Lancet, 1:111-116, 1927. 
I Simpson, K.: Shelter deaths from pulmonary 
embolism, Lancet 2:744, 1940. 
II Middleton, W. S.: Abdominal pain in pulmonary 
thrombosis, Ann. Int. Med., 18:345-349, 1943. 
12 Ochsner, A. and DeBakey, M.: Thrombophlebitis 
and phlebothrombosis. C. Jeff Miller lecture, South. 
Surgeon, 8:269-290, 1939. 
13 Homans, J.; Thrombosis of the deep veins of 
the lower leg, causing pulmonary embolism, New 
England, J. Med., 211:993-997, 1934. 
14 Homans, J.; Circulatory Diseases of the Extremi- 
ties, New York, MacMillan, 1939. 
0 Aschoff, L.: Lectures on Pathology, New York, 
Paul B. Hoeber, 1924. (p. 264). 
10 Dietrick, A.: Die Thrombose nach Kriegsverlet- 
zungen. Veroff. a.d. Geb. d. Kriegs u. Konstitution- 
pathologie., 1920, Heft 3. 
References 
1 Murnaghan, D., McGinn, S. and White, P. D.: 
Pulmonary embolism with and without acute cor 
pulmonale with especial reference to the electro- 
cardiogram, Am. Heart J., 25:573-597, 1943. 
'McGinn, S. and White, P. D.; Acute cor pul- 
monale resulting from pulmonary embolism. 
J.A.M.A. 104:1473-1480, 1935. 
3 Virchow, R.: Die Verstopfung der Lungenarterie 
und ihre Folgen. Beitr. z. exper. Path. u. Phys. H., 
2:1-90, 1846. INTRA-OCULAR TUMORS IN SOLDIERS, WORLD WAR II 
By HELENOR CAMPBELL WILDER, Pathologist to the Registry of Ophthalmic Pathology, 
Army Institute of Pathology 
(With sixty-six illustrations) 
TRUE intra-ocular neoplasms and other 
tumors which may be regarded as at 
least potentially neoplastic were found 
in 42 (1.08 per cent) of 3882 eyes of soldiers 
studied during World War II at the Army 
Institute of Pathology. The tumors were of 
particular interest because they occurred in a 
relatively homogeneous population, that is, in 
presumably healthy men 18 to 38 years of age. 
This paper deals primarily with the clinical 
manifestations and the histopathologic features 
of these neoplasms. The types encountered are 
shown in Table I, and they will be considered 
in the order in which they are listed. 
to have caused ectropion uveae (Fig. i), and 
in a case of bilateral metastatic malignant 
melanoma, reported later in this paper, benign 
melanomas of the iris (Fig. 41 and 42) were 
bilateral as well as larger and more deeply 
pigmented than in the other cases. Reese1 re- 
garded such growths, when associated with 
malignant melanomas of the uveal tract, as 
separate benign tumors rather than seedings 
from the malignant tumor. 
Benign Melanoma of the Iris 
Benign melanoma of the iris, not associated 
with other tumors, was present in two cases. 
In both of these, a pigmented lesion had been 
observed for 15 years and iridectomy was per- 
formed because of recent active growth. The 
tumors extended from the anterior surface 
deep into the posterior portion of the iris, and 
by slit lamp examination as well as in un- 
bleached sections, had the appearance of pos- 
sible malignancy (Fig. 2). However, in the 
bleached sections (Fig. 3) the sparse cellularity 
of the tumors indicated their benign character, 
and the eyes were not enucleated. 
Malignant Melanoma of the Iris 
The oldest patient with malignant 
melanoma of the iris was 34, 4 others were 
in their twenties. The tumors had been ob- 
served for from one and a half to 5 years; 
2 interfered with vision. Melanoma was diag- 
nosed clinically in 4, and in I in which the 
growth was lightly pigmented the clinical diag- 
nosis was leiomyoma. In this and one other 
instance iridectomy was performed before 
enucleation. In the first case (Fig. 4 and 5) 
microscopic sections revealed the nevoid type 
of cell and the nest-like arrangement character- 
istic of skin nevi. This is often seen in malig- 
nant melanomas of the iris, although it is 
rarely encountered in those of the choroid, 
and it does not fall into Callender’s classifica- 
tion.2’3’4’5 In the second case the cells, al- 
Tumor Number of 
Cases 
Benign melanoma of the iris associated with 
malignant melanoma of the choroid* 
7 
Benign melanoma of the iris 
2 
Malignant melanoma of the iris 
5 
Benign melanoma of the choroid 
Malignant melanoma of the ciliary body 
6 
and choroid 
25 
Metastatic melanoma in the eye 
i 
Metastatic carcinoma in the choroid 
Pseudo-adenomatous hyperplasia with 
i 
dyskeratosis of an epithelial implant 
i 
von Hippel’s hemangiomatosis 
i 
Table I 
* Observed in 7 of the 25 eyes with malignant 
melanoma here listed. 
Benign Melanoma of the Iris 
Associated with Malignant 
Melanoma of the Choroid 
Small benign melanomas or pigment 
freckles of the iris were found in 7 of the 
eyes enucleated for malignant melanoma of 
the choroid. Whether the growths are true 
tumors, or whether they are malformations 
that may become neoplastic is still an open 
question. In most instances the cellular ac- 
cumulations were very small and similar to 
those often seen in eyes without other tumors. 
In one instance, however, the growth appeared 460 
The Military Surgeon—November, 1946 
Fig. i. Benign melanoma (pigment freckle) of the iris, causing ectropion uveae. Xioo. AIP Neg. 
96124. 
Fig. 2. Benign melanoma of the iris, removed by iridectomy. X50. AIP Neg. 95871. 
Fig. 3. Bleached section, demonstrating the sparse cellularity of the tumor shown in Fig. 2. X650. AIP 
Neg. 959J4- 
though they retained their nest-like arrange- 
ment, were predominantly of the spindle B 
type of Callender (Fig. 6). Both tumors had 
extensively invaded the stroma of the iris. In 
the first case the eye was not submitted for 
microscopic examination. In the second case 
examination showed a residual tumor in the 
filtration angle on the side opposite that of 
operation (Fig. 7). 
In 4 instances the filtration angle and 
anterior ciliary body were involved (Fig. 8), 
with tumor cells in the spaces of Fontana, canal 
of Schlemm, and in or around the vessels of 
the intrascleral plexus (Fig. 9). Occlusion of 
the filtration angle by tumor led in one in- 
stance to secondary glaucoma with cupping of 
the optic disc, but in the other eyes the tension 
was normal. In none of these tumors was Intra-ocular Tumors in Soldiers, World War II 
461 
Fig. 4. Malignant melanoma of the iris removed by iridectomy. The tumor is densely cellular and in- 
vasive. X 60. AIP Neg. 95881. 
Fig. 5. Nevoid type of cell and arrangement in tumor shown in Fig. 4. X450. AIP Neg. 95882. 462 
The Military Surgeon—Novemher, 1946 
Fig. 6. Malignant melanoma of the iris removed by iridectomy. Nevoid arrangement of spindle-shaped 
cells. X400. AIP Neg. 95886. 
Fig. 7. Tumor cells in the filtration angle on the side opposite that of iridectomy. The eye was enucleated 
following microscopic examination of the section shown in Fig. 6. X35. AIP Neg. 95888. Intra-ocular Tumors in Soldiers, World War II 
463 
Malignant Melanoma of the Iris 
Fig. 8. Involvement of the anterior ciliary body and the filtration angle. AIP Neg. 95887. 
Fig. 9. Tumor cells on the anterior surface of the iris, in the filtration angle, and around the canal of 
Schlemn and vessels of the intrascleral plexus. X90. AIP Neg. 95866. 
Fig. 10. Mixed cell type: spindle cells, subtype B, and an occasional large epithelioid cell. X650. AIP 
Neg. 95880. 464 
The Military Surgeon—Novembery 1946 
Fig. ii. Incidental finding in an eye removed 6 months after penetrating injury. X50. AIP Neg. 95868. 
Fig. 12. Incidental finding in an eye removed 3 months after penetrating injury. X50. AIP Neg. 95869. 
Fig. 13. Incidental finding in phthisical eye enucleated 21 years after penetrating injury. Calcified drusen 
on Bruch’s membrane. X50. AIP Neg. 95867. 
Benign Melanoma of the Choroid 
extrabulbar extension seen. Microscopically, 
two of the tumors combined nevoid and spindle 
B type cells, and three were of the mixed cell 
type with spindle B and epithelioid cells pre- 
dominating (Fig. 10). 
the eyes were enucleated in from less than 24 
hours to 6 months, but in the sixth case the 
injury had been incurred 21 years before 
enucleation of the phthisical eye. Flistologically, 
the tumors appeared as deeply pigmented 
thickened areas in the posterior choroid (Fig. 
11, 12 and 13). In bleached sections they 
exhibited different degrees of cellularity, some 
being sparsely (Fig. 14) and others more 
densely cellular. Although the tumor shown 
in Fig. 15 was much more cellular than that 
Benign Melanoma of the Choroid 
The 6 benign melanomas of the choroid 
were all incidental findings in eyes removed 
because of penetrating wounds. Five of the 
injuries had occurred in action or training and Intra-ocular Tumors in Soldiers, World War II 
465 
FlG. 14. Benign melanoma of the choroid, sparsely cellular. Bleached section. X650. AIP Neg. 95896. 
Fig. 15. Benign melanoma of the choroid with nevoid cells. No invasion of the choriocapillaris. Bleached 
section. X650. AIP Neg, 95872. 
in Fig. 14, it was limited anteriorly by the 
choriocapillaris and there was no invasion of 
that layer. The cells were either nevoid with 
irregular nuclei, or spindle shaped with long 
narrow nuclei which were usually homo- 
geneous in appearance. Rarely an ill-defined 
nucleolus was seen. 
Malignant Melanoma of the Ciliary 
Body and Choroid 
The malignant character of 24 of these 25 
malignant melanomas was fully developed; in 
only one case was the tumor regarded as of 
low-grade malignancy. This small melanoma 
was first observed on ophthalmoscopic exami- 466 
The Military Surgeon—November, 1946 
Melanoma of the Choroid. Early or Low Grade Malignancy 
Fig. i6. Tumor adjacent to optic nerve head. Bruch’s membrane intact. X14. AIP Neg. 95894. 
Fig. 17. Spindle cell, subtype A. X200. AIP Neg. 96221. 
Fig. 18. Argyrophil fiber content heavy, except in an area beneath Bruch’s membrane. Wilder reticulum 
stain. X200. AIP Neg. 95891. 
nation following contusion of the eye; enuclea- 
tion was performed 23 days after the injury. 
The irregularly pigmented tumor (Fig. 16) 
in the choroid adjacent to the optic disc was 
composed of closely packed spindle shaped 
cells, sometimes arranged in bundles (Fig. 17). 
The nuclei of the cells were seldom nucleo- 
lated, and little nuclear detail was seen. The 
tumor was much more cellular, however, 
than the benign melanoma shown in Fig. 15 
and it involved all the layers of the choroid 
except Bruch’s membrane which appeared in- Intra-ocular Tumors in Soldiers, World War II 
467 
Fig. 19. Malignant melanoma of the choroid. Bruch’s membrane intact. AIP Neg. 95873. 
Fig. 20, Malignant melanoma of the choroid which has broken through Bruch’s membrane and invaded 
the overlying retina. There is serous detachment of the adjacent retina. AIP Neg. 95874. 
Fig. 21. Malignant melanoma of the choroid. Bruch’s membrane intact. AIP Neg. 95876. 
Fig. 22. Malignant melanoma of the ciliary body and choroid. Intra-ocular hemorrhage followed con- 
tusion 6 months before enucleation. AIP Neg. 95877. 
tact. Argyrophil fibers6 were abundant 
throughout a large part of the tumor, indicat- 
ing a low grade of malignancy (Fig, 2, 3, 4, 
5); in the inner and apparently more active 
portion, where the protective mechanism of 
reticulum formation had been unable to keep 
pace with the growth of the tumor, there were 
nonfibered areas (Fig. 18). The retina over 
the tumor was atrophic and its rods and cones 
had disappeared but there was no exudate be- 
neath it. Detachment, apparent in Fig. 16, 
occurred during the technical procedures. 
Symptoms in the 25 cases were noted from 
3 weeks to 11 years before enucleation; in 2 
instances the tumors were found on routine 
examination. Loss or disturbance of vision was 
the usual symptom first to manifest itself and 
was recorded in all but 4 instances. Pain was 
recorded in 4 cases in 2 of which there was 
glaucoma, in the other 2 intra-ocular hemor- 
rhage. Two patients had multiple moles, one 
on the axillae and buttock, the other on the 
back. The urine was negative for melanin in 
the 3 patients on which this test was recorded. 468 
The Military Surgeon—Novembery 1946 
Fig. 23. Malignant melanoma of the choroid. Extension of the tumor through the sclera along the canal 
of a long posterior ciliary nerve and artery. X30. AIP Neg. 95911. 
Fig. 24. Malignant melanoma of the choroid associated with a long posterior ciliary nerve in the scleral 
canal. X150. AIP Neg. 96125. 
Fig. 25. Malignant melanoma of the choroid invading the nerve head and extending into the optic nerve. 
X 100. AIP Neg, 96123. Intra-ocular Tumors in Soldiers, World War II 
469 
A clinical diagnosis of tumor was made in 
24 cases and, more specifically, of malignant 
melanoma or sarcoma in 20 in which the 
tumor was visualized on ophthalmoscopic 
examination. In 5 instances cataract, present 
in 2 eyes, or intra-ocular hemorrhage, present 
in 7 eyes, was of sufficient density and propor- 
tion to obscure the fundus (Fig. 22). In one 
of these the tumor was believed to be von 
Hippel’s hemangiomatosis and, in the single 
instance in which no tumor was suspected, the 
eye was enucleated because of intra-ocular 
hemorrhage following contusion. Posterior 
sclerotomy, performed on both of these eyes, 
was followed by endophthalmitis; in one other 
case an operation for retinal detachment was 
performed with a subsequent diagnosis of 
tumor. Secondary glaucoma was observed in 
4 eyes. Detachment of the retina (Fig. 20) 
was present in 24 eyes; in only one was de- 
tachment, at least around the tumor, absent. 
Of the 25 malignant melanomas, the small- 
est (Fig. 16) measured 2.5 x 2.5 x 1.0 mm. 
and the largest 17 x 15 x 15 mm. Sixteen 
involved the posterior choroid (Fig. 19 and 
20), 2 the choroid at the equator, extending 
both anteriorly and posteriorly (Fig. 21), and 
6 the ciliary body and choroid (Fig. 22). 
Three had extended outside the eye along the 
posterior ciliary nerves and vessels (Fig. 23). 
The close association of these tumors with the 
ciliary nerves is shown in Fig. 24, lending 
support to Dvorak-Theobald’s theory7 of their 
origin from the sheaths of Schwann of the 
ciliary nerves. Four had extended into, but 
apparently not through, the sclera. One had 
invaded the nerve head and one the optic 
nerve beyond the lamina cribrosa (Fig. 25), 
but not beyond the line of excision. In 2 in- 
stances tumor cells were present in the 
sclerotomy wounds; in 9 the tumor had in- 
vaded the overlying retina (Fig. 20), and in 
6 more it had broken through Bruch’s mem- 
brane but the retina was uninvolved (Fig. 21). 
Some of the tumors had two or more modes 
of extension; only 6 were confined to the 
uveal tract (Fig. 19). 
The group of malignant melanomas have 
been classified according to cell type,2’3’4’5 
fiber content,3’4’5 and pigment content.5 
Table II shows the distribution of the tumors 
by Callender’s classification based on cell types 
(Fig. 26, 27, 28, 29). 
Table II 
Cell Type 
Number 
Spindle cell subtype A 
2 
Spindle cell subtype B 
14 
Fascicular 
1 
Mixed cell type 
6 
Too necrotic to classify 
2 
Epithelioid 
0 
Total 
25 
The preponderance of spindle cell types 
over those containing: the less differentiated 
epithelioid cells does not follow the pattern of 
our earlier survey5 in which the more malig- 
nant mixed cell type predominates. Twenty- 
five cases is not a sufficient number on which 
to base conclusions, however, and similar 
studies were made on the larger civilian groups 
to determine whether or not there was any 
foundation for this apparent relation between 
age and type. Table III indicates that, al- 
though the preponderance of the less malig- 
nant types does not exist in the larger group, 
there is still a higher percentage of them at 
the military age than of tumors removed in the 
later decades. 
Table IV shows the distribution of tumors 
of the Army series according to fiber content 
(Fig. 31, 32, 33, 34). 
Although there is no preponderance of the 
more heavily fibered and apparently less malig- 
Table III 
Source 
Ag-e 
Spindle cell 
Total 
Per Cent of Spindle 
Subtype A and B 
Cell Tumors 
Army 
18-38 
l6l 
351 
64.01 
>120 
V 256 
[46.8 
Civilian 
OO 
1 
00 
104.) 
231) 
45-°) 
Civilian 
Over 38 
518 
1491 
34-7 470 
The Military Surgeon—November, 1946 
nant tumors over those with a lower fiber 
content it is seen in Table V that a higher per- 
centage of the less malignant types is present 
in the earlier than in the later decades. 
statement that the average age for tumors of 
the iris is lower than for those of the ciliary 
body and choroid. This observation and the 
fact that melanomas of the iris behave less 
malignantly than do those of the choroid ap- 
pear to be explainable on the basis of easy visi- 
bility and consequent early recognition of the 
iris tumors. On the contrary, unsuspected be- 
nign melanomas are found in the choroid of 
eyes enucleated for other causes. Malignant 
melanomas of the choroid have been noticed 
for the first time on routine examination or 
called to the patients’ attention by injury. They 
have been known to exist for years, causing 
only minor visual disturbances. It appears that 
such tumors may persist for a long time in 
their less malignant spindle cell forms; sud- 
denly, however, they show increased mitotic 
activity and formation of epithelioid cells 
Fiber Content 
Number 
Heavy 
i 
Marked 
7 
Medium 
5 
Light 
11 
Absent 
i 
Total 
25 
Table IV 
Melanomas of the uveal tract were found 
in white soldiers only. This was to be expected 
because of the rarity of these tumors among 
Negroes as exemplified in our previous series 
Discussion 
Source 
Age 
Fiber Content Heavy 
Total 
Per Cent of Heavily 
and Marked 
Fibered Tumors 
Army 
18-38 
8] 
25] 
29.0] 
U56 
J-3°-4 
Civilian 
18-38 
70 J 
231J 
30.3 j 
Civilian 
Over 38 
258 
1491 
17-3 
Table V 
of 16005 malignant intra-ocular melanomas 
of which only 8 were in Negroes. In the 
present group the youngest soldier with 
melanoma was 19, the oldest 38, 23 were be- 
tween 18 and 29, and 15 between 30 and 38 
years of age. In the series of 1600 malignant 
melanomas just referred to, 240 or 15 per 
cent occurred in patients between the ages of 
18 and 38. The addition of 370 received from 
civilian sources since June 1942 and 30 from 
Army sources now brings the total of 
melanomas of the eye on file in the American 
Registry of Pathology to 2,000. Three hun- 
dred fifteen or 15.75 per cent of these were in 
patients 18 to 38 years of age. The increased 
percentage of younger individuals in the more 
recent series, which includes Army cases, is so 
slight as to appear insignificant. 
In a series of 137 malignant melanomas of 
the iris, including the 5 from the Army, 52 or 
37.0 per cent were from patients under 40 
years of age, which agrees with Duke-Elder’s8 
(Fig. 30). Studies, the results of which are 
presented in this paper, indicate that a higher 
percentage of the less malignant spindle cell 
and heavily fibered tumors is present in the 
earlier than in the later decades, and that 
epithelioid cells and tumors with a lower fiber 
content are more often encountered in eyes 
enucleated from patients over 40 years of age. 
The foregoing observations are based on 
investigations of the collection of intra-ocular 
melanomas at the Army Institute of Pathol- 
ogy, and for the most part corroborate those 
of Duke-Elder in his masterly chapter on 
“Tumors of the Uveal Tract.”8 These studies 
appear to justify the hypothesis that melanomas 
of the ciliary body and choroid which produce 
symptoms and metastasize are tumors pre- 
dominantly of the higher decades, although, 
like the melanomas of the iris, they may exist 
in younger individuals as benign, quiescent, or 
slow growing forms more often than the clini- 
cal history suggests. Intra-ocular Tumors in Soldiers, World War II 
471 
Fig. 26. Spindle cell, subtype A. X650. AIP Neg, 95879. 
Fig. 27. Spindle cell, subtype B. X 650. AIP Neg. 95878. 
Fig. 28. Fascicular type. X200. AIP Neg. 95892. 
Fig. 29. Epithelioid cell type. X 650. AIP Neg. 95884. 
Malignant Melanoma of the Choroid 
Metastatic Melanoma in the Eye 
Melanoma, metastatic to both eyes from a 
primary site in the skin, represents such a rare 
condition that the one case in the present 
series deserves detailed consideration. The 
growth occurred in a 34 year old white soldier 
who had been through the North African, 
Sicilian and Italian campaigns. The first symp- 
toms, which led to the diagnosis of battle 
fatigue, were those of psychosis evidenced in 
his reaction to either friendly or enemy ar- 
tillery fire. Physical examination, however, 
revealed several melanomas, having the ap- 
pearance of buck shot, under the skin of the 472 
The Military Surgeon—November, 1946 
abdomen and shoulders and on the left side 
of the face. There were also enlarged supra- 
clavicular, axillary, epitrochlear, and inguinal 
nodes. Ophthalmoscopic examination disclosed 
fresh retinal hemorrhages but no intra-ocular 
tumor (Fig. 35, 36). The patient died with 
uremia during a blood transfusion a little over 
two months after the onset of symptoms. 
At autopsy a malignant melanoma was 
found which involved the skin, bone marrow, 
39). However, Bruch’s membrane appeared 
intact. Tumor cells also packed veins in the 
ciliary body (Fig. 40) and in the iris (Fig. 
41, 42). Another interesting feature in the 
iris was the presence of bilateral benign mel- 
anomas or pigment freckles (Fig. 41, 42) at 
the anterior surface of the pupillary zone. 
Tumor cells were present in and around the 
vessels of the intrascleral plexus (Fig. 45), 
in small veins in the conjunctiva, in veins of 
Fig. 30. Malignant melanoma of the choroid, spindle cell, subtype B, with a mitotic figure and an epithelioid 
cell indicating recently increased activity. X 650. AIP Neg. 95885. 
meninges, cerebrum, pericardium, heart, 
lungs, liver, spleen, pancreas, kidneys, mesen- 
tery and mesenteric lymph nodes. There was 
hematogenous seeding of the entire body. 
Veins plugged with tumor cells in the retina 
had given rise to the hemorrhages (Fig. 43, 
44) which were noted on the ophthalmoscopic 
examination. Close to the macula, in the nerve 
fiber layer of the right eye, was a group of 
cytoid bodies near a hemorrhage. 
Most striking was the histologic appearance 
of the choroid in both eyes where large veins 
in the vascular layer and capillaries in the 
choriocapillaris were engorged with tumor 
cells (Fig. 37, 38). Tumor cells were rarely 
seen in the choroidal arteries (Fig. 38). For 
the most part they were confined within blood 
vessels, but in the peripheral choroid of both 
eyes they were present in the stroma (Fig. 
the episclera, occasionally in the orbit (Fig. 
46) and around small orbital arteries (Fig. 
47) 
Only one of the skin tumors (Fig. 48) was 
removed for sectioning. The tumor cells were 
in close association with nerve bundles deep 
in the dermis (Fig. 49) but no connection 
with the surface epithelium was seen. This 
tumor was not regarded as the primary tumor, 
although the clinical history and the distribu- 
tion leave little doubt that the primary growth 
was in the skin. 
The tumor cells in the skin, the eyes, and 
in all the metastases were similar. They were, 
for the most part, round or polygonal in shape, 
with abundant cytoplasm filled with finely 
granular brown pigment. Few spindle shaped 
cells were present. Irregularly shaped nuclei, 
rich in chromatin but with ill-defined nuclear Intra-ocular Tumors in Soldiers, World War II 
473 
Fig. 31. Heavy argyrophil fiber content. Wilder reticulum stain. X650. AIP Neg. 95862. 
Fig. 32. Marked fiber content. In only a few areas are the tumor cells not surrounded by argyrophil 
fibers. Wilder reticulum stain. X650. AIP Neg. 95861. 
Fig. 33. Marked fiber content but with small bundles of tumor cells not penetrated by argyrophil fibers. 
Wilder reticulum stain. X650. AIP Neg. 95863. 
Fig. 34. Argyrophil fibers only in the interlobular stroma, none surrounding individual tumor cells. Wilder 
reticulum stain. X 650. AIP Neg. 95864. 
Malignant Melanoma of the Choroid 
detail, gave the cells a nevoid appearance. 
Well-defined nuclei were rare, but occasional 
vacuolated nuclei and mitoses were seen. 
Large macrophages containing coarse pigment 
granules were observed in most of the metas- 
tases as shown in photomicrographs of the 
liver (Fig, 50) and the bone marrow (Fig. 
51). In the brain (Fig. 52) as in the retina, 
hemorrhages resulted from tumor emboli. 
Both Fuchs9 and Parsons10 doubted that 
malignant melanomas, primary elsewhere, 
metastasize to the uveal tract. However, 10 474 
The Military Surgeon—November, 1946 
cases in the literature are widely accepted as 
actual or possible instances of this condition. 
The cases of Bromser,11 Schiess-Gemuseus and 
Roth,12 Pfiiiger,13 Boente,14 Cordes and 
Horner,15 Corrado,16 Fry17 and Uhler18 ap- 
pear to have been primary in the skin, al- 
though in Bromser’s case this tumor was not 
examined microscopically, and in Pfliiger’s 
the eye was not enucleated, autopsy was re- 
to be primary in the conjunctiva or lid of the 
left eye with metastases to the skin, brain and 
right ciliary body. 
Another case of probable malignant mel- 
anoma metastatic in the eye, not reported 
formally but mentioned by Dr. Luther Peter 
in a discussion of Fry’s case, was one in which 
multiple metastases followed a primary growth 
in the skin of the upper arm. The eye was 
Figs. 35 (O.D.) and 36 (O.S.). Malignant melanoma metastatic in the eye, bilateral, from the skin. 
No gross tumor appears in either eye. AIP Neg. 96126. 
fused, and none of the lesions was subjected 
to microscopic examination. 
The uveal tract was the site of metastasis 
in all but two instances: the case of Schiess- 
Gemuseus in which the nerve head and ad- 
jacent retina were involved and that of Uhler 
in which multiple metastases were present in 
the retina but none in the optic nerve. In 
Boente’s case the tumor involved the choroid, 
retina, and optic nerve, and, as in ours, Bruch’s 
membrane was intact. Tumor cells were noted 
in retinal veins in Boente’s case and in a 
choroidal capillary in Fry’s. Only in Cordes 
and Horner’s case were the metastases to the 
eyes bilateral. There was however, bilateral 
involvement in ten Doesschate’s20 case in which 
the tumor is believed to have been primary in 
the right eye with metastases to the skin, 
the brain, and the choroid of the left eye. In 
the case of Adamiik19 the tumor was believed 
invaded only during the last 2 weeks of the 
patient’s life but then growth was rapid. The 
same type cell was present in all lesions. The 
diagnosis was “sarcoma of the choroid.” 
Metastatic Carcinoma in the Choroid 
One other tumor, metastatic in the choroid 
bilaterally, less rare than the preceding one, 
was a papillary adenocarcinoma in a 33-year- 
old white soldier. The case is interesting in 
that the first symptom of disease was dimness 
of vision in the right eye. Soon after this a 
productive cough, night sweats, and a low 
grade fever developed for which the patient 
was hospitalized. A tumor of the right eye 
was discovered on examination; roentgeno- 
grams revealed lesions of the lungs, scapula, 
pelvis and spine. Because the first symptoms 
appeared in the right eye it was believed that 
this was the site of the primary tumor and Intra-ocular Tumors in Soldiers, World War II 
475 
Fig. 37. O.S. Choroidal veins and capillaries engorged with tumor cells. X250. AIP Neg. 92872. 
Fig. 38. O.D. Choroidal veins and capillaries and one artery engorged with tumor cells. X200. AIP 
Neg. 92870. 
Intra-ocular Metastases from Malignant Melanoma of the Skin 476 
The Military Surgeon—November, 1946 
Intra-ocular Metastases from Malignant Melanoma of the Skin 
Fig. 39. Tumor cells in the peripheral choroidal stroma. Cystoid degeneration of the retina at the ora 
serrata. X150. AIP Neg. 92887. 
Fig. 40. Tumor cells in the veins of the ciliary body. X 150. AIP Neg. 92883. 
that the generalized lesions were metastases of 
a malignant melanoma of the choroid. How- 
ever, 1 o months after the appearance of 
symptoms in the right eye, fundus changes 
were noted in the left eye. After this the 
course of the disease was complicated by 
Cheyne-Stokes respiration and left facial and 
right trochlear palsy. The patient died one 
year after the onset of symptoms. 
At autopsy, tumor of both lungs, particu- 
larly of the right lower lobe, was found with 
neoplastic invasion of the spleen, liver, adrenals, 
kidneys, brain, mesenteric, lumbar, and in- 
guinal lymph nodes, the left 5th rib and the 
lumbar spine. Sections showed adenomatous 
neoplasm in the posterior choroid of both eyes Intra-ocular Tumors in Soldiers, World War II 
477 
Fig. 41, Tumor cells in iris veins. Benign melanoma (pigment freckle) on the anterior surface of the 
iris. X 115. AIP Neg. 92884. 
Fig. 42. O.S. Tumor cells in a vein in the region of the sphincter pupillae. Benign melanoma on the 
anterior surface of the iris. X115. AIP Neg. 92882. 
Intra-ocular Metastases from Malignant Melanoma of the Skin 
(Fig. 53). Papillae projected into the alveoli 
which were lined by columnar epithelium 
(Fig. 54). A moderate number of mitoses was 
present (Fig. 55). The tumor in the other 
organs (Fig. 56, 57 and 58) was similar in 
appearance. 
Histologically, both lung and gastrointes- 
tinal tract must be given consideration as sites 
of the primary tumor. However, multiple 
metastases, particularly to the brain, mesenteric 
lymph nodes, and bone, indicate that the pri- 
mary site was more probably the lung. 
The American Registry of Pathology now 
has a civilian series of 40 carcinomas meta- 
static in the uveal tract in which the age, sex 
and race are known: all were from white pa- 478 
The Military Surgeon—November, 1946 
Intra-ocular Metastases from Malignant Melanoma of the Skin 
Fig. 43. O.S. Tumor embolus in a retinal vein causing hemorrhage in the nerve fiber layer. X115. AIP 
Neg. 92881. 
Fig. 44. O.D. Tumor cells in a retinal vein. Subretinal hemorrhage. X250. AIP Neg. 92875. 
tients, 28 from females and 12 from males. 
Only one patient was 33 years old; 5 others 
were in the middle or late thirties, and the rest 
were over 40. Breast, lung, gastro-intestinal 
tract, kidney, prostate, ovary and pancreas 
are all represented as primary sites of the 
tumors, but next in frequency to the breast, 
the lung predominates as it did in Usher’s21 
series of no cases. In 4 of our 40 cases 
carcinoma of the eye was found before the Intra-ocular Tumors in Soldiers, World War II 
479 
Ocular Metastasis from Malignant Melanoma of the Skin 
Fig 45. Tumor cells within the spaces of Fontana and in a vein near the canal of Schlemm. X330. AIP 
Neg-. 92880. 
Fig. 46. Tumor cells in an orbital vein. X230. AIP Neg. 92876. 
Fig. 47. Tumor cells around orbital arteries. X330. AIP Neg-. 92879. 480 
The Military Surgeon—November, 1946 
Fig. 48. Malignant melanoma of the skin with ocular metastases. This skin nodule 
was probably not the primary lesion. X 50. AIP Neg. 95899. 
primary tumor was suspected. In 2 of these 
it was in the lung, in 1 in the gastro-intestinal 
tract and in 1 in the breast. In 6 cases of 
carcinoma of the breast, metastases were first 
manifested in the eye. These were noted 12 
years, 10 years, 9 years, and in 3 cases, less 
than 2 years after removal of the primary 
tumor. In 4 cases ocular metastases are known 
to have been bilateral. 
Considering the age and sex incidence as 
well as the comparative rarity of this tumor, 
it is not surprising that only one should have 
been found among the eyes from Army 
sources. There is no feature of this case, how- 
ever, age, sex, source, duration, symptoms, 
dissemination, or histopathology, that is not 
duplicated in our series of 40 from civilian 
sources. 
Pseudo-Adenomatous Hyperplasia with 
Dyskeratosis of an Epithelial Implant 
Although hyperplasia with dyskeratosis of 
an epithelial implant might not be considered 
true tumor, its behavior was so like that of a 
neoplasm that its inclusion in this series seems 
warranted. The eye came from a 22-year-old 
white soldier wounded in action; a magnetic 
foreign body was retained in the eye until 
enucleation, 16 days later. The foreign body, 
which measured 14 X 8 X 5 mm., was 
lodged in the posterior segment and protruded 
through the sclera (Fig. 59). The globe was 
partially eviscerated, collapsed, and its remain- 
ing structures disorganized. Organizing hem- 
orrhage and inflammatory tissue filled the 
globe and subacute endophthalmitis was evi- 
denced by the presence of lymphocytes, plasma Intra-ocular Tumors in Soldiers, World War II 
481 
cells, and polymorphonuclear leukocytes. The 
blocked areas in Fig, 59 were the sites of im- 
planted surface epithelium, probably conjunc- 
tival rather than corneal. One (A) was in 
the ciliary region and the other (B) in the 
posterior segment near the site of the foreign 
interesting to speculate if frank malignancy 
might have resulted had the eye not been 
enucleated. It is more probable that retrogres- 
sion would have occurred or, as scar tissue 
developed, the growth would have been choked 
off. 
Fig. 49. Malignant melanoma of the skin with ocular metastases. Tumor cells in association with a nerve 
bundle in the dermis. X 450. AIP Neg. 92874. 
body. The implants showed hyperplasia and 
acanthosis (Fig. 60, 61) with abundant in- 
tracellular kerato-hyalin and the formation of 
epithelial pearls. Intercellular bridges were 
apparent (Fig. 61). Basal palisading was al- 
most entirely absent (Fig. 62), and there ap- 
peared to be actual invasion where nests of 
epithelial cells were present beneath and sepa- 
rated from the larger mass. 
The condition has been classified as a 
pseudo-adenomatous hyperplasia of epithelial 
implants, but in view of the dyskeratosis it is 
The behavior of the implanted epithelium 
in this case is reminiscent of Lucke and 
Schlumberger’s22 experiments on autotrans- 
plantation of epitheliomas of the lip and mouth 
of catfish into the anterior chamber of the 
eye. In 10 of 14 instances the transplanted 
tumor grew and vascularized in its new loca- 
tion, sometimes with actual invasion of the 
iris; however, after 2 months it became sta- 
tionary and eventually regressed. Greene23 
and other investigators of tumor immunity 
have also found the anterior chamber a fertile 482 
The Military Surgeon—TVovember, 1946 
Generalized Melanomatosis from Malignant Melanoma of the Skin 
Fig. 50. Metastasis to the liver. The large dark cells are pigment-laden macrophages. X140. AIP Neg, 
92873. 
Fig. 51. Metastasis to the rib. Pigment-laden macrophages are also apparent here. X 160. AIP Neg, 92885. 
Fig. 52. Metastasis to the brain. Cerebral hemorrhages resulting from tumor emboli. X75. AIP Neg, 
92886. Intra-ocular Tumors in Soldiers, World War 11 
483 
Papillary Adenocarcinoma in the Choroid, Metastatic from the Lung 
Fig. 53. Metastasis in the posterior choroid. Xxo. AIP Neg. 95903. 
Fig. 54. Columnar epithelium lining- alveoli. X200. AIP Neg. 95906. 
Fig. 55. Columnar epithelium with mitoses, covering a papilla. X 500. AIP Neg. 95905. 484 
The Military Surgeon—November, 1946 
Fig. 56. Carcinoma of lung, primary tumor which metastasized to both eyes. X30. AIP Neg. 95910. 
Fig. 57. Primary papillary adenocarcinoma of the lung, similar in appearance to the metastases in the 
eyes and other organs. X230. AIP Neg. 95908. 
Fig. 58. Metastasis in the brain from the papillary adenocarcinoma of the lung. X70. AIP Neg. 95909. Intra-ocular Tumors in Soldiers, World War 11 
485 
Fig. 59. Intra-ocular epithelial implants following penetrating wound. A, implant in the ciliary region 
B, implant in the posterior segment near the site of a foreign body. X5. AIP Neg. 95860. 
Fig. 60. Hyperplasia of an intra-ocular epithelial implant (Fig. 59B). X 60. AIP Neg. 95897. 
field for tumor implants. 
Much has been written on the subject of 
intra-ocular implants following trauma and 
surgery (Perera24) but this is the first time, 
so far as I know, that changes have been ob- 
served in them which might be interpreted 
as malignant. 
disc diameters in size, was visualized project- 
ing forward from the disc and retina, below. 
Roentgenograms revealed questionable lesions 
of the sella turcica and mediastinum. All 
other examinations were negative. Although a 
diagnosis of probable von Hippel’s disease was 
given, the eye was enucleated because of the 
possibility of malignant neoplasm. The eye was 
opened in the vertical plane and a vascular 
mass measuring 4 X 3 X 2.5 mm. was seen 
at the nerve head. 
Microscopic examination revealed a mass 
involving the nerve head and adjacent retina 
with cystoid degeneration of the outer plexi- 
form layer of the retina near the tumor (Fig. 
63). Gliosis was present throughout most of 
von Hippel’s Hemangiomatosis 
The last growth of the series represents a 
malformation rather than a true tumor. It oc- 
curred in a 28 year old white soldier who, 2 
years before the eye was enucleated, noticed 
a blurring of vision which increased rapidly in 
the last 5 months. On ophthalmoscopic ex- 
amination a highly vascularized mass, about 3 486 
The Military Surgeon—November, 1946 
Fig. 6i. Prickle cells with prominent intercellular bridges and an epithelial pearl in an intra-ocular im- 
plant (Fig. 59B). X 550. AIP Neg. 95900. 
Fig. 62. Pearl formation in an epithelial implant (Fig. 59A). Nests of epithelial cells are seen beneath 
the large mass, suggesting invasion. X200. AIP Neg. 95898. Intra-ocular Tumors in Soldiers, World War II 
487 
Von Hippel’s Hemangiomatosis 
Fig. 63. Involvement of the nerve head and adjacent retina. Cystoid degeneration of the outer plexiform 
layer of the retina near the tumor. X 14. AIP Neg. 95902. 
Fig. 64. Glial proliferation in the tumor. X250. AIP Neg. 96222. 488 
The Military Surgeon—November, 1946 
Von Hippel’s Hemangiomatosis 
Fig. 65. Proliferated capillaries and glial cells in the superficial portion of the tumor. X500. AIP 
Neg. 96223. 
Fig. 66. Thin-walled blood channels, some containing red blood cells, in the deep portion of the tumor. 
X500. AIP Neg. 95901. Intra-ocular Tumors in Soldiers, World War 11 
489 
the mass but was particularly marked at its 
periphery where it involved the retina (Fig. 
64). Here the proliferated glial cells and their 
fibers were arranged in whorls and bundles. 
Elsewhere the vascular nature of the tumor 
was apparent. Proliferation of thin-walled 
blood vessels with intervening glial prolifera- 
tion was present in the inner portion of the 
tumor (Fig. 65). In the deeper portion, nu- 
merous small channels usually lined by a single 
layer of endothelial cells frequently contained 
red blood cells (Fig. 66); occasionally better 
formed vessels were seen. Histologically the 
tumor was characteristic of hemangioglioma- 
tosis or von Hippel’s disease, and, with the 
history of possible lesions of the sella turcica 
and mediastinum, it must be considered as a 
probable component of Lindau’s syndrome. 
other growths were represented by single cases 
and may be looked upon as rare. Metastatic 
carcinoma of the choroid from the lung is 
relatively infrequent, particularly in this age 
group. The case of intra-ocular metastases of 
malignant melanoma is the eleventh to be re- 
ported. Von Hippel’s angiomatosis, although 
a fairly large number of cases has been studied 
pathologically since the first was reported by 
Panas and Remy in 1879,25 is not often seen 
clinically and less often comes to enucleation. 
In the case reported here, there was clinical 
evidence of Lindau’s syndrome. No reference 
has been found in the literature to malignant 
dyskeratosis of an intra-ocular epithelial im- 
plant and it is believed that the condition is 
reported here for the first time. 
It is inevitable that rare conditions should 
appear in such a large group subjected to com- 
pulsory physical examination and with readily 
available medical facilities. It is, however, sur- 
prising that under these circumstances the 
percentage of the more usual intra-ocular 
tumors should remain so low. 
Summary 
Only 1.08 per cent of eyes enucleated from 
soldiers 18 to 38 years of age between Pearl 
Harbor and VJ Day and submitted to the 
Army Institute of Pathology contained intra- 
ocular neoplasms. These, however, included a 
variety of tumors both primary and metastatic. 
Represented in the group were: benign and 
malignant melanomas of the uveal tract; bi- 
lateral intra-ocular metastases from carcinoma 
of the lung and from malignant melanoma 
of the skin; malignant dyskeratosis in a trau- 
matic intra-ocular epithelial implant, and von 
Hippel’s disease. Malignant melanoma of the 
uveal tract was the most common tumor and 
benign melanoma was next in frequency. 
Those benign melanomas which occurred in 
the choroid were incidental findings in eyes 
enucleated following trauma. One was in an 
eye which had suffered a preinduction injury 
resulting in phthisis bulbi, but the other five, 
unless they had showed malignant change, 
would probably have gone unnoticed in civilian 
life. 
Additional information concerning mel- 
anoma, relative to histologic malignancy and 
age incidence, indicate that the less malignant 
histologic types are more often found in tumors 
removed from patients less than 40 years of 
age than from those in the later decades. The 
References 
1 Reese, A. B.: Pigment freckles of the iris (benign 
melanomas) : their significance In relation to 
malignant melanoma of the uvea. Am. J. Ophth. 
27:217-22 6, 1944. 
2 Callender, G. R.; Malignant melanotic tumors of 
the eye: a study of histologic types in 111 cases. 
Tr. Am. Acad. Ophth. 131-142, 1931. 
3 Callender, G. R., and Wilder, H. C.: Melanoma 
of the choroid; the prognostic significance of 
argyrophil fibers. Am. J. Cancer. 25:251-258, 1935. 
4 Wilder, H. C., and Callender, G. R.: Malignant 
melanoma of the choroid: further studies on prognosis 
by histologic type and fiber content. Am. J. Ophth. 
22:851-855, 1939. 
5 Callender, G. R., Wilder, H. C., and Ash, J. E.: 
Five hundred melanomas of the choroid and ciliary 
body followed five years or longer. Am. J. Ophth. 
25:926-967, 1942. 
Wilder, H. C.: An improved technique for silver 
impregnation of reticulum fibers. Am. J. Path. 11: 
817-819, 1935. 
7 Theobald, G. D.; Neurogenic origin of choroidal 
sarcoma. Arch. Ophth. 18:971-997, 1937. 
8 Duke-Elder, W. S.: Textbook of Ophthalmology. 
The C. V. Mosby Co., St. Louis, Missouri. Vol. 3, 
2446-2536, 1941. 
9 Fuchs, E.: Das Sarcom des Uvealtractus. Wien, 
W. Braumuller, 1882. 490 
The Military Surgeon—Novembery 1946 
10 Parsons, J. H.: Pathology of the Eye, London, 
Hodder and Stoughton. Vol. 2, pt. 2, p. 533, 1905. 
11 Bromser, G.; Ueber einen Fall von secundarem 
Melanom der Chorioidea. Berlin, G. Lange. Pt. 1, 
45 PP-j 1870. 
12 Schiess-Gemuseus and Roth, M.: Metastatisches 
Sarcom der Papille und augrenzenden Retina. Arch, 
f. Ophth. 25:177-192, 1879. 
13 Pfluger, E.: Metastatisches Sarcom der Cho- 
rioidea. Arch. f. Augenh. 14:129-132, 1884-5. 
1 Boente: Metastatische Melanoblastome in der 
Retina. Klin. Monatsbl. f. Augenh. 82:732-740, 
1929. 
15 Cordes, F. C., and Horner, W. D.: Metastatic 
melanoma of both eyes. Case Report. J.A.M.A. 
95:655-658, 1930, 
18 Corrado, A.: Sul sarcoma metastatico del 
corpociliare. Arch. d. Ophth. 38:508, Sept., 523, 
Oct., 1931. 
17 Fry, W. E.: Metastatic sarcoma of the choroid. 
Am. J. Ophth. 9:248-255, 1933. 
18 Uhler, E. M.: Metastatic malignant melanoma of 
the retina. Am. J. Ophth, 23:158-162, 1940, 
19 Adamuk, V.: Ein Fall von metastatischen 
melanosarkomder Uvea. Zeitschr. f. Augenh. 21: 
505-509, 1909. 
20 ten Doesschate, G.: Over metastatisch sarkom in 
het 00g. Nederl. tijdschr. v. geneesk. 2:1432-1436, 
I9I9- 
21 Usher, C. H.: Cases of metastatic carcinoma of 
the choroid and iris. Brit. J. Ophth. 7:10-54, 1923. 
22 Lucke, B., and Schlumberger, H.: Transplanta- 
tion epitheliomas of the lip and mouth of catfish. 
J. Exper. Med. 74:397-408, 1941. 
23 Greene, H. S. N.: Participation of anterior 
chamber of eye in resistance phenomena related 
to tumor growth. Cancer Research. 2:669-674, 1942. 
24 Perera, C. A.; Epithelium in the anterior 
chamber of the eye after operation and injury. Am. 
J. Ophth. 21:605-617, 1938. 
20 Panas, F., and Remy, A.; Anatomic pathologique 
de Poeil. Paris, V. Adrien Delahaye et Cfe, Libraires- 
Ediceurs. 88, 1879. THE OCCURRENCE AND ANATOMIC CHARACTERISTICS OF 
FATAL TUBERCULOSIS IN THE U. S. ARMY 
DURING WORLD WAR II* 
By COLONEL JOSEPH D. ARONSON, Medical Cor-ps, Army of the United States 
(With one chart) 
DURING World War II, in the 
United States and its possessions, the 
enrollment, mobilization, and induc- 
tion of great numbers of men, and the entrance 
into the military service of a large group of 
women made it possible, as never before, to 
obtain precise data on the incidence of tuber- 
culosis in the general population, as well as on 
the occurrence and evolution of tuberculosis 
among Army personnel. 
The attempt made during World War I 
to exclude men with manifest tuberculosis 
from the military service was only partially 
successful due to the great dependence on 
physical signs and the limited use of roentgeno- 
logic examination. Improvements in roent- 
genologic equipment and technique and the 
establishment of diagnostic standards since 
World War I, have made possible routine 
roentgenologic examination of large numbers 
of persons. 
Although the number of fatal cases of tuber- 
culosis received at the Army Institute of 
Pathology must not be construed as the true 
index of the mortality of the disease in the 
Army, comparison with preliminary official 
data of the Surgeon General’s Office shows 
close agreement. In view of this fact and of 
the further consideration that the differences 
between the mortality from tuberculosis in 
World War II and in previous comparable 
periods are so great that they could not be 
appreciably affected by slight revisions in the 
data, it is believed appropriate to present the 
following comparisons. 
A striking decline in mortality from tuber- 
culosis has occurred between World Wars I 
and II in both the military and general popula- 
tions of the United States. Table I and Figure 
1 present the comparison between civilian and 
Army experience by standard age groups at 
approximately the same time (1920 and 
1940). It will be noted that because of the 
existence of hospital facilities in the Veterans 
Administration during World War II, it has 
been necessary to add to the figures of the 
Surgeon General’s Office for World War II 
Table i 
Age—Specific Mortality from Tuberculosis in 
the Military and Civilian Population, 
World Wars I and II 
(Deaths per 100,000) 
Age in 
years 
World 
War I* 
Civilian 
Popula- 
tion 
I92of 
World 
War IIJ 
Civilian 
Popula- 
tion 
I94°t 
18-19 
43-8 
107.5 
4.2 
26.0 
20-24 
60.4 
151.0 
7-3 
40.4 
25-29 
67.6 
162.6 
5.8 
51.0 
30-34 
87.4 
163.7 
6.1 
59.8 
35-39 
122.2 
170.0 
9-4 
70.0 
40-44 
97-7 
172.0 
13-4 
78.2 
45 plus 
89.6 
174.0 
J7-1 
86.0 
Mean 
all ages§ 
65.9 
151.4 
6.7 
47-7 
* Med. Dept, of the U. S. Army in the World War, 
Vol. XV. 
f Tuberculosis in the United States, prepared by 
National Tuberculosis Association and U. S. Public 
Health Service, 1943. 
J Estimated from Army Institute of Pathology data 
and from Veterans’ Administration for period from 
Dec, 7,1941 to Dec. 1945, inclusive. 
§ Civilian mean rate for all ages standardized by use 
of respective Army age distribution for June 30, 1945, 
based on A.G.O. sample data applied to total strength 
from Dec. 1941 to Dec. 1945. 
those service-connected deaths from tubercu- 
losis of veterans of this war in hospitals of the 
Veteran’s Administration. 
Average annual mortality from tuberculosis, 
to date, in the Army of World War II has 
* From the Army Institute of Pathology, Washing- 
ton, D.C. 492 
The Military Surgeon—TVovember, 1946 
TUBERCULOSIS MORTALITY ( all forms) 
Fig. i Fatal Tuberculosis in the U. S. Army During World War 11 
493 
been approximately but one-seventh of that 
of the civilian population in the year 1940, 
the rate among Army personnel being about 
7 per 100,000, as compared to the civilian 
rate of 48, standardized with respect to the 
Army age distribution. In addition, between 
World Wars I and II the mortality rates of 
tuberculosis in the Army population dropped 
90 per cent, while that in the civilian popula- 
lated to the occurrence of fatal tuberculosis 
in United States Army personnel and with 
the distribution and character of the lesions. 
It is based on an analysis of the available clini- 
cal records and autopsy protocols of 579 males 
and 2 females who died in the United States 
or abroad between December 7, 1941 and 
December 1945, inclusive, in which the 
primary cause of death was tuberculosis. This 
Table 2 
Number of Deaths from Tuberculosis in the U. S. Army by 
Year of Occurrence and Race 
Race 
1942 
Year of Death 
1943 1944 
1945 
Total 
Percent by 
Race 
White 
37 
81 
95 
84 
297* 
51.1 
Negro 
30 
82 
87 
53 
252 
43-4 
Indian 
3 
3 
3 
3 
12 
2.7 
Filipino 
2 
2 
1 
4 
9 
i-5 
Hawaiian 
1 
0 
0 
O 
1 
0.1 
Puerto Rican 
0 
0 
2 
2 
4 
0.6 
Chinese 
0 
1 
1 
I 
3 
0.5 
Japanese 
1 
0 
2 
O 
3 
0.5 
Total 
74 
169 
191 
H7 
581 
* Two females included. 
tion of the same age composition fell 69 per 
cent. Preinduction roentgenologic examination 
carried out in World War II is probably the 
most significant factor responsible for the ex- 
tremely low mortality from tuberculosis in 
the Army during this war as compared with 
that of the civilian population and of the army 
in World War I. 
The pre-induction roentgenologic examina- 
tion of the chest of large numbers of persons 
from the general population has proved of 
great value in the control of tuberculosis, since 
numerous previously undiagnosed cases of this 
disease have been detected and referred for 
observation and treatment. Dempsey1 esti- 
mated in 1943 that within that year approxi- 
mately 12,000,000 persons in the general 
population would have been examined roent- 
genologically and it was estimated that about 
25,000 of those examined would require 
hospitalization for tuberculosis. 
The present study deals with factors re- 
study does not include the 805 service-con- 
nected deaths from tuberculosis of those mem- 
bers of the Army who died in the hospitals of 
the Veterans’ Administration from July 1942 
to July 1945. The number of deaths from 
tuberculosis exclusively in Army hospitals is 
shown in reference to year and race in Table 
2. 
Effect of Race, Age, and Length of 
Service on Occurrence of Tuberculosis 
in the Army 
The interpretation of many studies dealing 
with variations in the morbidity, the evolution, 
and the mortality from tuberculosis among 
racial groups is complicated by economic and 
environmental factors which differ widely 
among these groups. Experimental studies 
disclose wide variations in the resistance 
of laboratory animals to infection with M. 
tuberculosis. Wright and Lewis2 showed 
that in five families of guinea pigs inbred by 494 
The Military Surgeon—November, 1946 
brother and sister mating over a long period 
of time there existed marked differences in the 
resistance to infection by a virulent strain of 
M. tuberculosis. Lewis and Loomis3 using 
the same families of guinea pigs found that 
the local lesions following the intracutaneous 
injection of a virulent strain of M. tuberculosis 
was unlike in the different families. In the 
most resistant family the ulcer tended to heal 
and did not involve, grossly, the regional 
lymph nodes. In the least resistant family not 
only did the ulcer fail to heal but the regional 
nodes became swollen, matted, and ulcerated. 
Kuster and Kroning4 have confirmed the 
observations of Wright and Lewis2 as to 
the variation in resistance of different inbred 
families of guinea pigs to tuberculosis and have 
concluded that resistance is a hereditary factor. 
Lurie5 exposed six inbred families of rabbits 
to respiratory tuberculous infection. He found 
that in the most resistant family the pulmonary 
lesions tended to be localized and ulcerative 
in character, resembling those of reinfection 
in man, and that there were no gross lesions 
in the regional tracheobronchial nodes. On 
the other hand, in the most susceptible family 
the lesions became confluent giving rise to a 
tuberculous bronchopneumonia with extensive 
involvement of the draining tracheobronchial 
nodes and a tendency to hematogenous spread. 
Aronson, Parr and Saylor6 have noted marked 
variation in the local ulceration following the 
intracutaneous injection of a viable, attenuated, 
bovine type of M. tuberculosis (Calmette- 
Guerin strain) in Indian children. In one 
group of children and young adults the result- 
ing ulceration tended to be saucerlike and 
punched out, with the surface covered with 
healthy granulations, while in another group 
the ulcer had a small crater-like opening with 
undermined edges, typical of a tuberculous 
ulcer, which healed slowly. 
It has long been recognized that definite 
differences exist in the incidence, evolution, 
mortality, and anatomic characteristics of 
tuberculosis in the white and the colored. 
Thus the death rates per 100,000 from tuber- 
culosis in the United States for 1940 was 36.5 
for the white race and 127.6 for the colored. 
Studies of Israel and Payne7 indicate that 
the high rate and the rapid course of tubercu- 
losis among the colored is to be attributed to 
the fact that the tuberculous lesion is more 
acute and rapid in its evolution. The precise 
studies of Opie,8 and the investigations of 
Everett,9 Pinner and Kasper10 indicate 
that the anatomic character of the tuber- 
culous process in the colored is different from 
that in the white. The interpretation of data 
on resistance to tuberculosis among different 
racial groups is complicated by economic and 
environmental factors in civilian life. In the 
Army the influence of these factors is gen- 
erally absent, since for all military personnel 
patterns of housing, clothing, feeding, and 
remuneration are uniform while the duties 
incidental to military life are not affected by 
racial lines. As a result, a study of the occur- 
rence and evolution of tuberculosis in the 
Army is of great significance in evaluating 
racial susceptibility and resistance. Despite the 
uniform social and economic conditions Table 
2 shows that Negroes, who make up approxi- 
mately 10 per cent of the Army population, 
contribute 43.4 per cent of the total deaths 
from tuberculosis. Among the American 
Indians who numbered approximately 25,000, 
or 0.3 per cent of the Army population, there 
occurred ten, or 2.7 per cent, of all deaths 
from tuberculosis. These figures suggest that 
the high rate of tuberculosis among Negroes 
and Indians may be due in part to an innate 
lower degree of resistance to the disease rather 
than to poor economic and environmental 
factors. 
The mechanism involved in the selective 
role that age plays in the development of 
tuberculosis and the resulting mortality is not 
clearly understood. Thus for the year 194011 
in the general white population the rela- 
tively high rate of tuberculosis among those 
under 5 years of age was followed by a greatly 
reduced rate for the age group of 5 to 14 years. 
A sharp rise in the rate for the 15-to-19-year 
and the 20-to-24-year groups was succeeded 
by a more gradual increase which reached its 
peak in the 6 5-to-7 4-year group. Among the 
colored a sharp rise occurred in the io-to-14 Fatal Tuberculosis in the U. S. Army During World War II 
495 
and the 15-to-19-year groups, reaching the 
peak of mortality in the 20-to-24~year group. 
This predilection for certain age periods in 
different races cannot be explained solely on 
the basis of exposure to tuberculosis and is 
probably modified by other factors. The dis- 
tribution by age and racial groups of fatal 
tuberculosis in Army hospitals during World 
War II is presented in Table 3. The per- 
the transfer of large numbers of men from 
their homes to the crowded barracks of Army 
camps and the change from civilian occupa- 
tion to duties involving physical stress and ex- 
posure might lead to an increase in tuberculosis. 
Many men came from rural areas where they 
had more often escaped tuberculous infection. 
The relationship between the length of service 
and the occurrence of fatal tuberculosis by 
Table 3 
Mortality from Tuberculosis by Age and Race* 
Age in 
years 
No. 
White 
Per cent 
No. 
Negro 
Per cent 
No. 
All races 
Per cent 
Percentage 
distribution 
in Army 
18-19 
25 
7-5 
11 
4.4 
36 
6.2 
9-i 
20-24 
125 
38-3 
116 
46.4 
241 
41.7 
37-6 
25-29 
87 
26.5 
63 
25.1 
150 
26.0 
29.8 
30-34 
54 
16.5 
30 
11.9 
84 
14.4 
16.2 
35-39 
23 
7.0 
26 
10.3 
49 
8.5 
5.8 
40 plus 
14 
4.2 
5 
1-9 
19 
3-2 
i-5 
Total 
328 
100.0 
251 
100.0 
579 
100.0 
100.0 
* Based on Army age distribution for June 1945 as shown in A.G.O. sample data and representative of 
age distribution throughout period of World War II. 
centage of deaths observed in the various age 
groups is compared with the per cent dis- 
tribution of these age groups in the Army as 
a whole. 
It will be observed from Table 3, that the 
largest percentage of deaths occurred among 
those in the age group of 20 to 24 which in- 
cludes the largest percentage of soldiers, and 
that there was no sharp racial variation in the 
percentage of deaths at different ages. Com- 
parison of the distribution of the age at which 
deaths from tuberculosis occurs with the age 
distribution of the Army as a whole indicates 
that mortality from tuberculosis increases with 
age, particularly in men above 34 when the 
rate becomes almost 50 per cent greater than 
that in men below this age. It should be noted, 
however, that men above 35 in the Army are 
much less representative of all men in that age 
group than are those in the younger groups. 
The relationship of the length of active 
Army service to the occurrence of fatal tuber- 
culosis was analyzed. The possibility exists that 
year and race is presented in Table 4. 
It will be noted from Table 4 that the per- 
centage of deaths from tuberculosis, all races, 
was largest during 1942 and 1943 among 
those who had the shortest length of service. 
In 1944 and 1945 the peak of deaths was 
among those who had a year or more of Army 
service. These differences might be explained 
on the basis of the rate of expansion of the 
Army. In 1942 approximately 3,500,000 
were inducted and in 1943 approximately 
2,200,000 were inducted, while in 1944 and 
1945 only 600,000 and 150,000 respectively 
entered the Army. While this is the most 
likely explanation for these differences, it is 
also possible that the larger number of deaths 
in the early days of the war may have been due 
to the induction of a number of men not ex- 
amined roentgenologically, who may have had 
a tuberculous process which gave rise neither 
to subjective symptoms nor to physical signs. 
The high death rate of the group who had 
served but a short time in the Army may be 496 
The Military Surgeon—November, 1946 
Table 4 
Mortality from Tuberculosis in Relation to Active Army Service, All Races 
(per cent distribution) 
Months of 
Active 
Service 
1942 
1943 
Year 
1944 
1945 
Total 
o-6 
42.9 
43-6 
i3-i 
6.1 
23-9 
7-12 
18.6 
24.2 
17.0 
12.1 
18.1 
13-18 
25.7 
10.7 
24.4 
8-3 
16.7 
19-24 
8.6 
8.7 
19-3 
22.7 
15.8 
25-30 
0.0 
7-4 
13.6 
14.4 
10.2 
31-36 
0.0 
2.0 
5-7 
22.0 
7-9 
37-42 
0.0 
2.0 
6.8 
6.8 
2.8 
43-48 
4.2 
1.4 
0.0 
7.6 
4-5 
Total 
100.0 
100.0 
100.0 
100.0 
100.0 
Deaths 
70 
149 
176 
132 
527 
explained in still other ways. Possibly persons 
who had escaped previous tuberculous infection 
were overcome by their primary infection or 
others who had little resistance to tuberculous 
infection may have been infected and had 
rapidly progressive tuberculosis. 
Table 5 compares the frequency of fatal 
tuberculosis among white and Negro with 
reference to length of active Army service. It 
is clear that there was no significant trend to 
indicate that members of either group con- 
tracted tuberculosis relatively earlier or later 
in their period of service. 
It is possible that morbidity and mortality 
from tuberculosis may be influenced not only 
by the length of service in the Army but also 
by the place of service. Thus, service in foreign 
lands, especially during and immediately 
following war, might result in heightened 
susceptibility due to fatigue, or increased ex- 
posure resulting from unhygienic surround- 
ings, from billets in heavily infected com- 
munities and homes, or from intimate contact 
with tuberculous civilians, whose number in- 
creases rapidly with privation. The per cent 
distribution of deaths from tuberculosis in re- 
lation to foreign or domestic service by race 
and in relation to the distribution of troops is 
presented in Table 6. It will be observed from 
this table that during 1942 and 1943 the per- 
centage of deaths from tuberculosis paralleled 
closely the distribution of Army personnel; 
Table 5 
Mortality from Tuberculosis in Relation to Race and Length 
of Active Army Service 
Months 
White 
Negro 
Total all Races 
of Service 
Number 
Per cent 
Number 
Per cent 
Number 
Per cent 
o-6 
65 
22.4 
61 
25.9 
126 
23-9 
7-12 
46 
IS.8 
49 
20.7 
95 
18.1 
13-18 
52 
17.9 
36 
15.3 
88 
16.7 
19-24 
39 
13-4 
44 
18.7 
83 
15.8 
25-30 
35 
12.1 
19 
8.1 
54 
10.2 
31-36 
27 
9.2 
15 
6-3 
42 
7-9 
37-42 
18 
6.2 
6 
2-5 
24 
2.8 
43-48 
9 
3-8 
6 
2-5 
15 
4-5 
Total 
291 
100.0 
236 
100.0 
527 
100.0 Fatal Tuberculosis in the U. S. Army During World War II 
497 
Table 6 
Mortality from Tuberculosis in U. S. Army in Relation to Foreign 
or Domestic Service by Age and Race 
White 
Negro 
All Races 
Distribution 
of Army 
Foreign 
Domestic 
Foreign 
Domestic 
Foreign 
Domestic 
Foreign 
Domestic 
Year 
Per cent 
Per cent 
Per cent 
Per cent 
Per cent 
Per cent 
Per cent 
Per cent 
194a 
13-7 
86.3 
10.0 
90.0 
12.2 
87.8 
18 
82 
1943 
24.2 
75.8 
18.3 
81.7 
21.3 
78.7 
25 
75 
1944 
61.5 
38.5 
61.0 
39-o 
61.3 
38.7 
So 
So 
1945 
86.3 
13-7 
73-5 
26.5 
81.5 
18.5 
62 
38 
however, during 1944 and 1945 the percent- 
age was conspicuously higher among those 
serving in foreign lands. To a certain degree 
this rise is to be attributed to tuberculosis 
among Americans and Filipinos who were 
prisoners in the infamous enemy prison camps. 
The relative differences in the percentages for 
racial groups were constant in all theaters, 
except in 1945, when the white group over- 
seas showed a sharp increase, probably result- 
ing from this high incidence among released 
prisoners of war. 
among the colored than among the white. 
Rogers12 found that the mean duration 
of tuberculosis was 13 months for colored and 
18 months for white. Opie13 observed that 
among Negroes living in Jamacia, B. W. I., 
tuberculosis for those between 15 and 30 
years averaged 9 months, while for a group 
of white patients of comparable age in 
Philadelphia it averaged 28 months. Pinner 
and Kasper10 found that tuberculosis aver- 
aged 995 days for 96 white and 324 days 
for 47 colored. It has been assumed that the 
shorter duration of tuberculosis in the colored is 
to be attributed to failure to seek medical care 
until the disease has progressed. Israel and 
Payne,7 however, found that the relatively 
Duration of Tuberculous Process 
Accumulated evidence indicates that the 
course of the tuberculous process is more rapid 
Table 7 
Duration of Symptoms, by Month, for White and Negro 
Months 
No. 
Total White 
Per cent 
Total Negro 
No. Per cent 
Total All Races 
No. Per cent 
o-i 
36 
11 -3 
24 
9.8 
60 
10.7 
1-2 
46 
14.4 
23 
9-4 
69 
12.2 
2-3 
37 
11.6 
21 
8.6 
58 
10.3 
3-4 
25 
7.8 
23 
9-4 
48 
8.5 
4-5 
27 
8.5 
21 
8.6 
48 
VO 
00 
5-6 
30 
9-4 
26 
10.5 
56 
9-9 
6-7 
24 
7-5 
3i 
12.7 
55 
9-7 
7-8 
15 
4-7 
18 
7-3 
33 
5.8 
8-9 
11 
3-4 
11 
4-5 
22 
3-9 
9-10 
9 
2.8 
12 
4-9 
21 
3-7 
10-11 
12 
3-7 
7 
2.8 
19 
3-3 
11-12 
8 
2-5 
3 
1.2 
11 
1.9 
12 and over 
39 
12.2 
25 
10.2 
64 
11.2 
Total 
319 
100.0 
245 
100.0 
564 
100.0 498 
The Military Surgeon—November, 1946 
short course of the disease in the Negro was 
due not to the neglect of early symptoms but 
to the acute evolution of the tuberculous lesion, 
which was frequently exudative in character. 
Their cases were selective insofar as they were 
of a more or less chronic nature, but the 
autopsy material forming the basis for the 
considerably from the data previously cited, 
in which the course of tuberculosis was ap- 
preciably longer in the white. The principal 
reason for this discrepancy probably lies in the 
fact that the cases studied in this survey are 
acute cases, while those in the literature were 
primarily chronic cases of tuberculosis. 
Table 8 
Duration of Symptoms in Relation to Tuberculous Meningitis* 
White 
Duration Total deaths Deaths from 
in from Tuberculous 
Mont s Tuberculosis Meningitis 
Negro 
Total deaths Deaths from 
from Tuberculous 
Tuberculosis Meningitis 
All Races 
Deaths from 
Tuberculous 
Meningitis 
Number 
Per cent 
Number 
Per cent 
Per cent 
of total 
of total 
of total 
o-2 inch in 
63-9 
58 
42.6 
56.1 
3-5 79 
41.5 
69 
20.0 
31.6 
6-8 46 
36.0 
57 
25.0 
30.0 
9-11 27 
34-5 
25 
9-i 
23-5 
12 and over 39 
17.9 
22 
12.0 
15.6 
Over all per cent 
45-5 
25.7 
36.9 
Average Duration in months 
Tuberculous Meningitis 
4.2 
4-5 
4-3 
Average Duration in months 
Tuberculosis without Meningitis 
7.0 
6.6 
6.8 
Average Duration in months 
All cases 
5-7 
6.1 
5.8 
* Including tuberculous cases in which meningitis was a terminal complication. 
present study is also selective since it represents 
the more acute cases in Army hospitals. 
The duration of tuberculosis as determined 
from subjective symptoms, confirmed by 
roentgenologic or bacteriologic examination or 
both, was investigated. The distribution of the 
duration of the disease in months for 319 white 
and 245 colored soldiers is presented in Table 
7- 
This analysis indicates that, in general, the 
duration of symptoms was not conspicuously 
different for any or all the years among the 
white and colored groups, the average for the 
white being 5.7 months, while that for the 
Negro was 6.1 months. This result differs 
One of the more frequent forms of acute 
tuberculosis observed in this study is tubercu- 
lous meningitis. Forty-five and five tenths per 
cent of the cases of tuberculosis among the 
white soldiers took this form, whereas it oc- 
curred in only 25.7 per cent of the Negro 
patients. In addition, it was found that the 
average duration of tuberculous meningitis was 
4.2 months in the white and 4.6 months in 
the Negroes, which is appreciably below the 
average duration of 7.0 and 6.6 months re- 
spectively for tuberculosis without meningitis. 
The relatively higher proportion of tubercu- 
lous meningitis among the white patients 
operates to reduce the average duration of Fatal Tuberculosis in the U. S. Army During World War II 
499 
symptoms in that group absolutely, and also 
relatively to that of the Negroes. Table 8 pre- 
sents the data suggesting this view. 
Pathologic Anatomy 
The pathologic material on which this study 
is based is from autopsies performed in Army 
The tabulation presented in Table 9 indi- 
cates that in this series of cases no outstanding 
racial differences were noted in the frequency 
with which tuberculous lesions appeared in any 
of the organs. It will be seen, however, that 
brain, meninges, seminal vesicles and prostate 
were more often involved in the white, and 
Occurrence of Tuberculosis in Different Organs by Race 
Table 9 
V 
White 
Negro 
All Races 
k 
Organs 
No. 
Per cent 
No. 
Per cent 
No. 
Per cent 
Lungs 
305 
92.0 
233 
92.1 
538 
92.5 
Tracheobronchial nodes 
274 
»3-4 
215 
85.5 
489 
84-3 
Brain 
45 
13-7 
25 
9-9 
70 
12.1 
* 
Meninges 
T47 
44.8 
65 
25.8 
212 
36-3 
Intestine 
122 
37-2 
83 
33-o 
205 
35-3 
A 
Peritoneum 
45 
13-7 
5i 
20.2 
96 
16.5 
Liver 
224 
68.2 
158 
62.5 
382 
65.8 
Pancreas 
24 
7-3 
21 
8-3 
45 
7-7 
Stomach 
2 
0.6 
2 
0.8 
4 
0.7 
Heart 
4 
x .2 
4 
1.6 
8 
1-4 
Pericardium 
5 
i-5 
7 
2.8 
12 
2.1 
Kidneys 
133 
40.4 
96 
38.1 
229 
39-4 
* 
Ureter 
6 
1.8 
1 
0.4 
7 
1.2 
Urinary bladder 
10 
3-2 
4 
1.6 
14 
2.4 
| 
Seminal vesicle 
32 
9-7 
8 
3-2 
40 
6.9 
Prostate 
69 
21.0 
22 
8-3 
90 
15-5 
Epididymis 
14 
4.2 
IO 
3-9 
24 
4.1 
) 
Testes 
13 
3-9 
8 
3-2 
21 
3-6 
Spleen 
228 
69.2 
158 
62.5 
386 
66.4 
Adrenal 
93 
28.4 
55 
21.8 
H7 
25-3 
I 
Thyroid 
13 
3-9 
10 
3-9 
23 
3-9 
' 
Vertebrae 
9 
2.7 
22 
8.7 
31 
5-3 
Ear 
3 
0.9 
2 
0.8 
5 
0.8 
> 
i 
Eye 
2 
0.6 
0 
— 
2 
0.3 
Ribs 
1 
0.3 
1 
0.4 
2 
0-3 
hospitals in cases of death from tuberculosis. 
It is necessarily incomplete because of the 
transfer of the more chronic cases to hospitals 
of the Veterans Administration, nor has it uni- 
formity since the autopsies were performed 
by a number of prosectors whose interests, ob- 
servations, and descriptions varied. The proto- 
/ cols have been analyzed to determine the dis- 
tribution and the nature of the anatomic 
lesions. The frequency with which tubercu- 
lous processes were found in the various organs 
of the 329 white and 252 colored soldiers is 
presented in Table 9. 
hilar nodes, peritoneum and vertebrae in the 
colored. 
The studies of Opie,8 Everett,9 and of 
Pinner and Kasper10 indicate that the 
pulmonary lesions of tuberculosis in the 
Negro differ in their anatomic characteristics 
from those observed among the white. A com- 
posite description of the differences, according 
to the authors mentioned, indicates that in 
the Negro the tuberculous pneumonic lesion 
is less apt to involve the apex, tends to be 
exudative and ulcerative, while fibrosis is rare. 
The infection is usually spread by lymphogen- 500 
The Military Surgeon—November, 1946 
ous and hematogenous routes and produces 
extensive involvement and massive caseation of 
the draining tracheobronchial nodes. In the 
white, on the other hand, the apex is more 
frequently involved and the formation of 
cavities is usual. There is a greater tendency 
toward localization and fibroconnective tissue 
formation. Spread of the tuberculous process 
by lymphogenous or hematogenous routes is 
less frequent than in the Negro, and the drain- 
ing tracheobronchial nodes are more apt to 
undergo fibrosis and calcification. 
and confluent. In only a few instances did the 
prosectors report massive involvement of the 
hilar nodes among the colored. 
The character of the draining tracheo- 
bronchial nodes was noted in 274 white and 
215 colored. In 50, or 18.3 per cent, of the 
white and in 21, or 9.8 per cent, of the colored 
the draining nodes were calcified or fibrous. 
In 36, or 13 per cent, of the white and in 13, 
or 6 per cent, of the colored the nodes showed 
no gross lesions of tuberculosis. 
In one instance miliary coccidioidomycosis 
was also present in a white male age 30, who 
reacted to both coccidioidin and tuberculin in 
dilutions of I to lOO and I to IOOO, respec- 
tively, and in whose sputum both acid-fast 
Respiratory System 
The distribution and character of the pul- 
monary lesions and their relation to associated 
Table 10 
Occurrence of Pulmonary and Associated Tuberculous 
Lesions in White and Negro 
Lung and 
associated organ 
No. 
White 
Per cent 
No. 
Negro 
Per cent 
All Races 
No. Per cent 
Lung only 
26 
8.5 
30 
12.9 
56 
10.4 
Miliary generalized 
230 
75.0 
178 
76.0 
408 
75.6 
Meninges and brain 
22 
7.2 
5 
2.1 
27 
5.® 
Gastro-intestinal 
13 
4.2 
9 
3-9 
22 
4.0 
Liver 
8 
2.6 
0 
8 
i-5 
Peritoneum 
0 
— 
3 
1 -3 
3 
0.5 
Spleen 
4 
1 -3 
5 
2.1 
9 
i-7 
Vertebrae 
2 
0.6 
3 
1 -3 
5 
0.9 
tuberculous processes, as well as the character- 
istics of the draining regional tracheobronchial 
nodes, were studied in 306 white and 233 
colored persons. The occurrence of pulmonary 
and associated lesions is presented in Table 10. 
When the pathologic characteristics of the 
pulmonary lesions in the white and colored 
were compared, cavities were found in the 
lungs of 86.6 per cent of the white and 78.5 
per cent of the colored, while generalized 
miliary tuberculosis (Table 10) was equally 
frequent in both races. Evidence of healed 
primary tuberculosis, as indicated by fibrous 
scars or calcified pulmonary nodules or both, 
was noted in 43, or 14.1 per cent, of the white 
and in 20, or 8.5 per cent, of the colored. In 
general, the tuberculous pneumonic lesions 
among the colored were more often ulcerative 
bacilli and the spore form of C. immitis were 
found. From the cavity of the upper lobe, right 
lung, and from the pulmonary nodules, spores 
of C. immitis were demonstrated, while in the 
caseous material, acid-fast bacilli were noted. 
In the miliary lesions of the spleen, liver, and 
kidneys spores of C. immitis were found but 
no acid-fast bacilli. 
In a white male aged 26 an extensive bi- 
lateral adenocarcinoma of the lung was associ- 
ated with extensive tuberculous cavitation. 
Acid-fast bacilli in large numbers were found 
in the cavity, and the tracheobronchial nodes 
showed evidence of tuberculosis. 
Tuberculosis of the meninges, brain, or both 
was noted in 154, or 47 per cent, of the white 
Brain and Meninges Fatal Tuberculosis in the U. S. Army During World War II 
501 
and in 70, or 28.2 per cent, of the colored. 
The process was primary in 9.7 per cent of 
the colored and in 7.1 per cent of the white, 
while it was associated with a generalized 
miliary tuberculosis of other organs in 79.5 
per cent of the white and in 81.5 per cent of 
the colored. There was an associated involve- 
ment of the lung but of no other organ in 9.7 
per cent of the white and 7.1 per cent of the 
colored, while in both racial groups the 
adrenal only was involved along with the 
meninges in 1 white and in 1 colored person. 
Meningitis was associated with tuberculous 
spondylitis in 1 white and 2 colored persons. 
The meningitis was manifested by varying 
amounts of exudate, most marked at the base 
of the brain, extending through the pia- 
arachnoid. Definite tubercles, occasionally 
observed, were located either at the base of 
the brain or along the vessels. They did not 
differ from the usual tubercle except caseation 
was infrequent and slight. The vessel walls 
were often necrotic while the exudate consisted 
of large numbers of mononuclear cells and 
varying numbers of acid-fast bacteria. The 
paucity of definite tubercles in the meninges 
and the brain was probably due to the rapid 
course of the disease. 
Tuberculomas of the brain were reported 
as occurring in 8 white persons and in 3 
colored. It is probable that more would have 
been found had a particular search been made 
for them. 
although the parenchyma of the organs was 
relatively free of tubercles. Tuberculosis of the 
intestinal tract was most frequently noted in 
the ileum, cecum, colon and appendix. Tuber- 
culous ulcers were present in the gastric sub- 
mucosa of 2 white and 2 colored persons. The 
mesenteric nodes showed extensive caseation 
in most instances but calcification was infre- 
quent. Tuberculosis of the pancreas occurred 
only as a part of generalized miliary tuber- 
culosis. The lesions in the pancreas were 
usually discrete, small, and few in number. 
The liver was tuberculous in 68 per cent 
of the white and 62 per cent of the colored, 
the extent and character of the lesions varying 
within wide limits, ranging from lymphoid 
proliferation to fibrosis, and occasionally to 
calcification. 
Heart and Pericardium 
Tuberculosis of the myocardium was ob- 
served in 4 white and 4 colored, all with gen- 
eralized miliary tuberculosis. The tubercles 
were few, focal and miliary in character and 
showed little tendency to fibrosis. In 5 white 
and 7 colored the pericardium showed exten- 
sive caseonodose lesions, associated with other 
similar ones in the contiguous pleura. 
Genito-Urinary Tract 
In both white and colored, renal tuber- 
culosis was found to be associated in every 
instance with generalized miliary involvement 
of other organs. There were 133 cases of 
renal tuberculosis among the white; in 91 
miliary tubercles were present in the kidney 
and in 42 the tuberculous process was con- 
glomerate and extensive, destroying large 
areas of the kidney. Miliary tubercles were 
noted in 75 of 96 cases among the colored 
while in the remaining 21 the lesions tended 
to be conglomerate and extensive. The lesions 
in the ureter and bladder were located in the 
mucosa and submucosa. At times the process 
was diffuse, especially in the urinary bladder. 
Tuberculosis of the prostate and seminal 
vesicle in both white and colored was asso- 
ciated with generalized miliary tuberculosis in 
all cases but one. This exceptional case was 
Gastro-Intestinal Tract 
Tuberculous lesions of the gastro-intestinal 
tract differed neither in frequency nor char- 
acter in relation to racial factors, except that 
tuberculous peritonitis was noted somewhat 
more often among the colored. The tubercu- 
lous process manifested itself by ulceration or 
lymphoid proliferation of the follicles in the 
submucosa and less frequently by ulceration 
or diffuse infiltration of the other layers of the 
intestine. Tuberculous peritonitis was associ- 
ated with extensive involvement of the serosa 
of the intestines which at times penetrated the 
muscularis. The tuberculous process also 
affected the capsule of the liver and spleen, 502 
The Military Surgeon—November, 1946 
that of a white soldier in whom the prostatic 
lesion was associated with tuberculous menin- 
gitis. The tuberculous process in the prostate 
was extensive, having destroyed the greater 
part of the gland in 61 of the 69 cases noted 
among white soldiers, while in 8 the lesions 
were miliary in character. Of the 22 cases 
among the colored, extensive destruction of 
the prostate was noted in 14 and miliary le- 
sions in the remaining 8. 
The extensive, destructive tuberculous le- 
sions in the kidney, prostate, seminal vesicle, 
and testes are of interest and have been noted 
by other observers. The mechanism under- 
lying the selective and extensive destruction 
of certain organs, especially those of the genito- 
urinary tract and the adrenal, is not clearly 
understood. 
in the white. Extensive, conglomerate lesions 
were noted among the colored in the remain- 
ing 23 cases. It is evident that there are no 
racial differences in the character or type of 
the lesion in the adrenal. 
Thyroid 
The incidence of tuberculous involvement 
of the thyroid is the same in the two races. 
Tuberculosis of the thyroid was associated with 
generalized miliary involvement. The lesions 
in the gland were miliary in nature and few in 
number. 
Osseous System 
The bones were involved in 9 instances or 
2.7 per cent among the white personnel. In 7 
cases there was a tuberculous spondylitis, which 
in 2 cases was associated with a tuberculous 
pneumonic lesion, in the remaining 5 cases 
with generalized miliary tuberculosis. In 2 
instances the dorsal vertebrae were involved, 
in 3 the thoracic, and in 2 the lumbar. Tuber- 
culosis of the femur and ilium was associated 
in one case with miliary tuberculosis of the 
spleen and liver and pulmonary apical scar. 
Among the colored the vertebrae were in- 
volved in 19, or 7,5 per cent, of cases: the 
process was in the thoracic and lumbar region 
in 8 each, in the dorsal vertebrae in 2, and in 
the cervical vertebrae in I. In 3 instances 
the tuberculous spondylitis was associated with 
pulmonary tuberculosis, in 2 there was asso- 
ciated meningitis with no active lesion in other 
organs, while in 14 there was an associated 
miliary tuberculosis. In 3 instances there was a 
tuberculous involvement of the second and 
third ribs. 
Spleen 
Tuberculosis of the spleen manifested itself, 
in most instances, by miliary tubercles in vary- 
ing stages of development, and occasionally 
by conglomerate tubercles. Fibrosed tubercles 
were noted in 8 cases, calcified nodules in 1, 
and amyloid degeneration in 2 among the 
white. In general, among the colored there 
was a greater tendency for the tubercles to 
caseate and to become conglomerate. Calcified 
nodules in the spleen were observed in 5 in- 
stances, amyloid degeneration in 1, and 
fihrotic tubercles in 1. 
Adrenal 
In the white the adrenal was involved in 
93 cases. In 53 the tuberculous lesions were 
miliary and varied widely in number and stage 
of development. In 40 the adrenal showed 
large conglomerate tubercles which, in some 
cases, had destroyed the greater part of the 
organ. In one case of tuberculous meningitis 
the adrenal was the only other organ contain- 
ing a tuberculous lesion. In another case the 
lesions in the adrenal were associated with 
tuberculosis of the liver and with a healed 
primary pulmonary lesion. Among the col- 
ored, 32 of the 55, or 58 per cent, showed 
miliary tubercles of varying size and number, 
a percentage which almost coincided with 57 
Ears 
The middle and internal ear were involved 
in 3 white and 2 colored. Among the white 
persons the petrous portion of the bone had 
eroded but there was no meningitis in 2 cases 
although it was present in the third. In the 
colored both lesions were associated with gen- 
eralized miliary tuberculosis and 1 with 
meningitis. Fatal Tuberculosis in the U. S. Army During World War II 
503 
Eye 
nodes, peritoneum, and vertebrae. 
7. Evidence of healed primary tuberculosis 
in the lungs and in the tracheobronchial nodes 
was seen more often among whites. Among 
the colored, pulmonary lesions were more fre- 
quently ulcerative in character. 
The eyes were involved in 2 white persons. 
In one instance miliary tubercles were noted 
in the retina, while in the second case rare 
epithelioid tubercles were found in the 
choroid. 
Summary 
Bibliography 
1. During World War II the annual 
mortality from tuberculosis in the U. S. Army, 
per 100,000, was 6.7, while among the 
civilian population during 1940 the rate was 
47.7. In World War I the incidence among 
military personnel was 65.9, and in the 
civilian population in 1920 it was 151.4 per 
100,000. 
2. Autopsy records of 329 white and 252 
colored, who died from tuberculosis in Army 
hospitals during the period from December 
1941 to December 1945, were analyzed. 
3. Despite uniformity of living conditions, 
income and military duties, the colored, repre- 
senting about 10 per cent of the Army popula- 
tion, contributed 43.4 per cent of the deaths 
from tuberculosis. 
4. The distribution of fatal tuberculosis was 
in general comparable among the whites and 
colored for the same age groups, same periods 
of service, same places of service. The excep- 
tion is the higher rate among white Army 
personnel in 1945 in foreign service, due in 
part to the return of tuberculous prisoners of 
war. 
5. The short duration of the tuberculous 
process in both races, but especially in the 
white, is attributable to the frequency of tuber- 
culous meningitis. 
6. In general, no significant racial differ- 
ence was noted in the occurrence of tuber- 
culosis in many of the organs, but among the 
white the central nervous system, the seminal 
vesicles, and prostate were more commonly 
affected, and among the colored, the hilar 
1 Dempsey, M.: Statistical data; new cases of tuber- 
culosis discovered by case-finding surveys. Am. Rev. 
Tuberc. 48: 58-63, 1943. 
2 Wright, S., and Lewis, P. A.: Factors in the resist- 
ance of guinea pigs to tuberculosis, with especial re- 
gard to inbreeding and heredity. The American 
Naturalist. 55; 20-50, 1921. 
3 Lewis, P. A., and Loomis, D.; Ulcerative types as 
determined by inheritance and as related to natural 
resistance against tuberculosis; experimental study on 
inbred guinea pigs. J. Exper. Med. 47:449-468, 1928. 
4 Kiister, E., and Kroning, F.: Der Einfluss des 
Genotyps und der Einfluss ausserer Faktoren auf die 
Tuberkuloseresistenz beim Meerschweinchen. Arb. a. 
d. Staatsinst. f. exper. Therap. 35: 38-68, 1938. 
5 Lurie, M.; Heredity, constitution and tubercu- 
losis; experimental study. Am. Rev. Tuberc. (suppl.) 
44: 1-125, 1941. 
* Aronson, J. D., Parr, E. L, and Saylor, R. M.; 
BCG vaccine; its preparation and local reaction to its 
injection. Am. Rev. Tuberc. 42: 651-666, 1940. 
7 Israel, H. L., and Payne, H. M.: Tuberculosis in 
negro; clinical and roentgenological characteristics. 
Am. Rev. Tuberc. 41: 188-209, 1940. 
8 Opie, E. L.: Active and latent tuberculosis in negro 
race. Am. Rev. Tuberc. 10: 265-274, 1924. 
9 Everett, F. R.: Pathological anatomy of pulmo- 
nary tuberculosis in American negro and in white 
race. Am. Rev. Tuberc. 27: 411-464, 1933. 
10 Pinner, M., and Kasper, J. A.: Pathological pe- 
culiarities of tuberculosis in American negro. Am. 
Rev. Tuberc. 26: 463-491, 1932. 
11 Tuberculosis In the United States. National Tu- 
berculosis Association and U. S. Public Health Service, 
1943. 
12 Rogers, J. B.; Comparison of gross tuberculous 
lesions in whites and negroes. Am. Rev. Tuberc. 4: 
669-675, 1920. 
13 Opie, E. L.: Anatomical characteristics of tuber- 
culosis in Jamaica. Am. Rev. Tuberc. 22: 613-625, 
1930. FIBROUS DYSPLASIA OF SINGLE BONES (MONOSTOTIC 
FIBROUS DYSPLASIA)* 
By MAJOR HANS G. SCHLUMBERGER, Medical Corps, Army of the United States 
(With thirty-four illustrations) 
Introduction 
described by Schmorl.1 These two diseases are 
further characterized by changes in blood 
chemistry. In von Recklinghausen’s disease 
hyperfunction of the parathyroids leads to an 
increase in the serum calcium and phosphatase, 
a decrease of inorganic phosphorus, and ex- 
cessive excretion of calcium and phosphorus in 
the urine. In Paget’s disease serum calcium and 
phosphorus levels are essentially normal but 
there is an increase in the serum phosphatase. 
During the past decade another disease, 
different from Paget’s and von Reckling- 
hausen’s, but associated with fibrosis and de- 
formity of bones, has been recognized. This 
was perhaps first reported by Weil2 in 1922, 
but not until 1937 was it generally accepted 
as a distinct entity. In that year McCune and 
Bruch3 published a detailed account of what 
they called a case of osteodystrophia fibrosa 
combined with precocious puberty and patho- 
logic pigmentation of the skin in a young girl. 
In the same year Albright, Butler, Hampton 
and Smith4 described a “syndrome charac- 
terized by osteitis fibrosa disseminata, areas of 
pigmentation and endocrine dysfunction, with 
precocious puberty in females.” Their clinical 
findings are now frequently referred to as 
“Albright’s syndrome.” The following year 
Lichtenstein5 reported 8 cases, and suggested 
the term “fibrous dysplasia of bone” to identify 
the disease. In 1942 Lichtenstein and Jaffe6 
described 15 additional cases with a compre- 
hensive study of the gross and microscopic 
changes and a review of the literature. These 
authors were able to collect 75 previously re- 
ported cases which, with their own, brought 
the total to 90. The sex of the patient was 
known in 86 of these; 51 were female and 35 
male. In only 32 of the 90 cases was any de- 
gree of cutaneous pigmentation noted. Only 20 
—all females—showed clinical evidence of en- 
docrine dysfunction, characterized by pre- 
cocious puberty, early skeletal maturation, and 
SIXTY-SEVEN cases of fibrous dysplasia 
involving a single bone, which were 
studied at the Army Institute of 
Pathology, provide the basis for this paper. 
It considers first, the clinical manifestations, 
roentgenologic findings, pathologic anatomy, 
and location of the lesions; second, the relation 
of monostotic fibrous dysplasia to certain other 
conditions involving bones; and finally, the 
part played by trauma in its production. 
The proliferation of connective tissue is one 
of the basic responses of the body to injury; 
usually it is an integral part of the process of 
repair. When it occurs in the medullary 
cavity of bone it is frequently given the distinc- 
tive name “osteitis fibrosa,” although in the 
bones, as in other organs of the body, connec- 
tive tissue overgrowth is a nonspecific reac- 
tion. It occurs, for example, in chronic 
osteomyelitis, in osteomalacia and rickets, fol- 
lowing prolonged acidosis, and in the vicinity 
of primary and metastatic tumors of bone. 
Such connective tissue replacement of the 
marrow cavities and adjacent cancellous bone 
is also a characteristic feature of certain sys- 
temic diseases, namely von Recklinghausen’s 
disease (generalized osteitis fibrosa cystica) and 
Paget’s disease (osteitis deformans). In the 
former the fibrosis is secondary to an increase 
in the rate of bone resorption following en- 
hanced activity of the osteoclasts, and decalci- 
fication of the bone trabeculae due to an 
associated hyperparathyroidism. In Paget’s dis- 
ease osteoclastic destruction of bone is less in- 
tense and new bone formation is accelerated. 
The irregular, partially calcified trabeculae do 
not form symmetrical Haversian systems but 
are united in a haphazard fashion by broad 
cement lines to provide the “mosaic pattern” 
* From the Army Institute of Pathology, Wash- 
ington 25, D.C. Fibrous Dysplasia of Single Bones 
505 
extensive bone lesions. Serum calcium, phos- 
phorus, and phosphatase levels were within 
normal limits. In 15 of 87 cases only a single 
bone was affected; however, Jaffe and Lich- 
tenstein pointed out that monostotic involve- 
ment is probably much more common than 
these figures would indicate. Their suggestion 
is supported by the cases which provide the 
material for this paper. 
there were no symptoms; the abnormality was 
discovered in a roentgenogram of the chest 
taken for other reasons. Such an asymptomatic 
lesion was also found once in an ulna and again 
in a tibia. Extension of the process was usually 
slow. Thus 5 lesions were observed for ap- 
proximately 18 months, and i for 4 years, 
during which time they showed no change. 
A lesion in the greater trochanter of a femur 
remained stationary for 13 months. On the 
other hand, one in a rib doubled in size within 
2 years. 
In contrast to many of the reported cases 
of polyostotic fibrous dysplasia none of the 
cases of the monostotic form showed areas of 
abnormal skin pigmentation or evidence of 
endocrine disturbances. No congenital anoma- 
lies were encountered. In 18 cases in which 
serum calcium, phosphorus, and phosphatase 
determinations were reported the values were 
within the normal range. 
When the lesion was confined to one of the 
ribs, simple excision of the affected portion 
gave wholly satisfactory results. In the long 
bones curettement usually was performed and 
the defect then filled with bone chips; the 
implanted bone was slowly resorbed and re- 
placed by healthy callus. It is interesting to 
note that fracture followed curettement in 4 
cases; twice this occurred in the femur, once 
in the fibula, and once in the humerus. In I 
case the fracture occurred as late as 6 months 
after operation. The fractures all healed 
promptly. 
The prognosis in monostotic fibrous 
dysplasia is uniformly good. The cases com- 
prising the series here reported have been fol- 
lowed for less than 3 years after operation, but 
during this time none have shown signs of 
recurrence. There is no evidence in our ma- 
terial that the monostotic form of fibrous 
dysplasia may progress to the polyostotic 
variety. 
Roentgenologic findings. The appearance 
of monostotic fibrous dysplasia in the roent- 
genogram offers little that is characteristic. In 
the long bones the lesions were found princi- 
pally in the metaphyses (Fig. 1, 2 and 3), 
though occasionally they occupied the middle 
Material Studied 
During the war material and records from 
69 cases of fibrous dysplasia of bone were 
studied at the Army Institute of Pathology. 
The lesion was confined to a single bone in 67, 
involved both the right femur and tibia in 
one and was polyostotic in another. The almost 
complete absence of multiple bone involvement 
in this series may be due to recognition of the 
condition upon examination at induction 
centers with consequent disqualification for 
military service. In 2 cases, monostotic lesions 
visible in roentgenograms of the chest made 
at the time of induction went unrecognized, 
with the result that the inductees were accepted 
for general military service. Only 5 of the pa- 
tients were females, but this is of no statistical 
significance in view of the selective nature of 
the group examined. Two of the men were 
Negroes, the others were white. The average 
age at the time of the onset of symptoms was 
26 years. Again, this figure has little meaning 
since the study was restricted to members of the 
military age group, that is, between 18 and 
38 years. 
Clinical Manifestations and 
Roentgenologic Findings 
The first sign of disease was usually a local 
swelling, particularly if the affected bone was 
superficial, e.g. the skull, ribs, and tibia. Local 
tenderness was sometimes associated with the 
swelling, and occasionally, when one of the 
long bones was involved, pain of an arthritic 
character was referred to the nearest joint. 
The lesion was not suspected in 4 cases until 
the patient suffered a pathologic fracture; in 
3 it was in the femur, in I in the tibia. In 
12 of the 29 cases in which a rib was involved 506 
The Military Surgeon—November, 1946 
Fig. i. Case 51. Fibrous dysplasia of the subtrochanteric region of the left femur. 
Fig. 2. Case 54. Fibrous dysplasia of the proximal metaphysis of the left femur. Note the unusually broad 
area of bone condensation about the lesion. 
Fig. 3. Case 63. Fibrous dysplasia of the proximal metaphysis of the right tibia. The eccentric position 
of the lesion is noteworthy, for this localization is frequently regarded as characteristic of so-called non- 
osteogenic fibroma of bone. 
Fig. 4. Case 67. Fibrous dysplasia in the diaphysis of the right fibula. 
of the shaft (Fig. 4). No site of predilection 
was noted in the ribs. When the skull was 
affected the maxilla was most often involved. 
In the roentgenogram the area of fibrous 
dysplasia is radiolucent, sometimes traversed 
by delicate trabeculae of bone (Fig. 5 and 6). 
It is usually central in position and produces 
thinning and expansion of the cortex, particu- 
larly marked in the ribs and fibula (Fig. 6 and 
4) and in the bones of the calvarium; at times 
there is a narrow margin of condensed bone 
at the periphery (Fig. 7 and 2). 
The nonspecificity of the roentgenogram 
in this disease is emphasized by the fact that in 
not one of our cases was the possibility of 
fibrous dysplasia entertained by the roentgen- Fibrous Dysplasia of Single Bones 
507 
ologist. The differential diagnoses offered were 
in order of frequency: bone cyst 16, giant cell 
tumor 8, osteochondroma 8, tumor 7, enchon- 
droma 4, chondroma, fibroma, ossifying 
fibroma, osteitis fibrosa cystica, sarcoma 2, and 
myeloma 2. Osteoma, chondromyxoma, non- 
osteogenic fibroma, adamantinoma, Ewing’s 
tumor, eosinophilic granuloma of bone, Paget’s 
resected ribs. The lesions appeared as sym- 
metrically fusiform or almost spherical swell- 
ings ranging up to 6 cm. in greatest diameter. 
The surface consisted of thin but intact cortical 
bone. On longitudinal section the cancellous 
bone and marrow of the rib were found to be 
replaced by a firm, resilient, yellow-white 
tissue containing occasional small cysts, usually 
Fig. 5. Case 33. Fibrous dysplasia of the left 7th rib at its junction with the vertebra. Remnants of bone 
trabeculae give the lesion a multiloculated appearance. 
Fig. 6. Case 44. Fibrous dysplasia of the right 10th rib. 
Fig. 7. Case 35. Fibrous dysplasia of the left 7th rib. Note condensation of the cortical bone. 
disease, osteomyelitis, and callus each appeared 
once. The fact that the diagnosis of fibrous 
dysplasia is not found among these 21 diverse 
opinions would indicate that many roentgen- 
ologists are still unaware of the importance of 
considering this disease in the differential 
diagnosis of a solitary bone lesion. 
filled with amber fluid (Fig. 8). The transition 
from the pathologic to the normal bone mar- 
row was often abrupt (Fig. 9). The abnormal 
tissue had a gritty character due to the presence 
of innumerable minute spicules of bone. 
When skull or long bones were involved, 
only fragments obtained by curettement were 
available for examination; grossly these re- 
sembled the lesions in the ribs. 
Histologically the primary component of the 
bone lesion is connective tissue, fairly well 
The gross appearance of the involved bones 
in the present series has been studied in 18 
Pathologic Anatomy 508 
The Military Surgeon—Novembery 1946 
Fig. 8. Case 21. Longitudinal section through the posterior half of the 3rd rib. The cortex is thin, the 
bone expanded by firm, white, somewhat gritty fibrous tissue. The two cysts were filled with amber blood- 
tinged fluid; such relatively large cysts are uncommon. 
Fig. 9. Case 43. Junction of normal marrow and an area of fibrous dysplasia in a rib. The pathologic 
tissue occupies the right half of the figure. Note that the pre-existing cancellous bone appears to inhibit 
spread of the connective tissue. Spicules of metaplastic bone occupy its periphery; they are not remnants 
of the original bone trabeculae. Hematoxylin and eosin stain. X15. 
vascularized and frequently arranged in inter- 
lacing bundles and whorls (Fig. 10). Within 
the connective tissue, and most abundant at its 
periphery, are trabeculae of partly calcified 
newly formed bone (Fig. 9 and 11). That this 
bone is formed by direct metaplasia of the 
connective tissue is clearly evident from a study 
of the sections. The sequence of events appears 
to be as follows: the connective tissue cells 
round up, the nuclei become vesicular, and the 
intercellular fibrils thicken and stain deeply 
with eosin (Fig. 15). Subsequently the inter- 
cellular substance increases in amount and ulti- 
mately becomes calcified. The incarcerated 
connective tissue cells are then indistinguishable 
from osteocytes (Fig. 16). Even the more ma- 
ture trabeculae are deeply penetrated by 
bundles of collagenous fibers (Fig. 17). Not 
infrequently the trabeculae of new bone form 
an arch or complete circle about an area of 
edematous connective tissue (Fig. 12). This is 
usually found in lesions of the long bones, and 
is uncommon in those of the ribs. It may be 
linked to the probability that the lesions in the Fig. 
ii 
Fig, 
12 
Figure ii. Case 43. Trabeculae of metaplastic bone are developing within fairly dense connective 
tissue. The periphery of the bone spicules is not calcified. This is the histologic appearance most 
characteristic of fibrous dysplasia. Hematoxylin and eosin stain. X150. 
Figure 12. AIP Acc. 125831. Section from a biopsy of the right trochanter in a case of the monomelic 
type. The long curved spicules of bone appear to be developing at the junction of loose edematous 
connective tissue and a more dense fibrous tissue. The spaces are chiefly artefact, rarely they are 
vascular channels. The bone trabeculae are more mature than those in Fig. 11 and are sometimes 
found in lesions of the long bones but are uncommon in the ribs. Phosphotungstic acid hema- 
toxylin stain. X60. Fig, 
13 
Fig. 
14 
Figure 13. Case 34. Cyst formation following osteoclastic destruction of bone and its replacement 
by tissue fluids rather than by proliferation of fibrous tissue. A remnant of metaplastic bone is 
being destroyed by osteoclasts within the cyst. Phosphotungstic acid hematoxylin stain. X145. 
Figure 14. Case 12. Histologic appearance of a so-called ossifying fibroma of the right maxilla; see 
also Figs. 32 and 33. This is merely a variant of the more typical lesion of fibrous dysplasia shown 
in Fig. 11. Hematoxylin and eosin stain. X230. Fibrous Dysplasia of Single Bones 
509 
bones of the extremities are somewhat older 
when discovered, and hence the trabeculae 
are larger. The production of this fibrous bone 
under normal and various pothologic condi- 
tions has been exhaustively studied by Weiden- 
reich7 and Leriche and Policard.8 
Occasionally the connective tissue cells, after 
rounding up, apply themselves to the periphery 
of the metaplastic bone as osteoblasts. Associ- 
ated with these may be large multinucleate 
cells, osteoclasts, which appear to break down 
the newly formed bone (Fig. 18). Occa- 
were not observed in our material except in 
conjunction with healing pathologic fractures 
(Fig. 20). Such fractures were found in 3 
of the 18 ribs examined grossly; there was 
evidence of periosteal bone formation as well 
as endosteal callus (Fig. 21). 
The diagnoses offered by the pathologists 
who submitted the bones to the Army Institute 
of Pathology were varied: the lesion was 
diagnosed as fibrous dysplasia in 19 cases, 
osteitis fibrosa in 13, giant cell tumor in 7, 
ossifying fibroma in 7, fibrosarcoma in 5, 
Fig. 10. Case 42. Characteristic whorled character of the connective tissue element in fibrous dysplasia. 
Hematoxylin and eosin stain. X 230. 
sionally this osteoclastic removal of bone leads 
to the formation of small cysts (Fig. 13). 
While metaplastic bone formation is a 
prominent feature of fibrous dysplasia, there 
are sometimes large areas in which no such 
osseous transformation of connective tissue is 
evident (Fig. 19). No histologic differences 
are apparent between this connective tissue 
which exhibits no tendency to form bone, and 
that which does. It is noteworthy, however, 
that osseous metaplasia is more evident at the 
periphery of the lesions, that is to say, in regions 
adjacent to pre-existing bone. 
Islets of cartilage, reported by some authors, 
fibroma in 3, bone repair in 3, osteodystrophy 
in 2, bone cyst in 2 cases; hyperostosis, 
fibromyxoma, osteoid osteoma, and chronic 
inflammation were each reported once. From 
the number of correct diagnoses it would ap- 
pear that pathologists are becoming increas- 
ingly aware of fibrous dysplasia as a distinct 
entity, although a considerable number are 
still classifying it under the nondescript term 
“osteitis fibrosa.” However, the fact that the 
diagnosis of fibrosarcoma was sometimes made 
indicates that there is still work to be done in 
familiarizing pathologists with this lesion and 
in stressing its benign character. The Military Surgeon—Novemhery 1946 
Case AIP* 
No. Acc. No 
Site of Lesion 
Age** 
History of 
Trauma 
Duration of 
Symptoms 
Local Swelling 
Local Tenderness 
Blood Ca, 
P, Phosphatase 
Curettement—C 
Excision—E 
X-Ray Diag. 
Path. Diag. 
Remarks 
i. 
97^5 
Left frontal bone 
21 
— 
7 mos. 
+ 
— 
C 
Fibroma 
Osteofibroma, 
Left eye was displaced downward; 
hyperostosis 
intermittent headaches. 
2. 
168845 
Right frontal bone 
23 
- 
6 yrs. 
+ 
- 
C 
Fibromyxoma 
3- 
120904 
Left parietal bone 
21 
- 
7 yrs- 
+ 
- 
Normal 
c 
Tumor 
Osteitis fibrosa 
4- 
130727 
Left mastoid 
25 
- 
12 yrs. 
+ 
— 
Normal 
c 
Paget’s disease, 
Fibrous dysplasia 
Frequent earache and discharge 
osteoma 
from left ear until 12 years old. 
5- 
122552 
Occipital bone, center 25 
- 
1 yr. 
+ 
+ 
E 
Osteochondroma 
Fibrous dysplasia 
Pain and limitation of motion in 
back of neck. 
6. 
94650 
Left maxilla 
23 
— 
4 mos. 
+ 
- 
C 
Ossifying fibroma 
7- 
116477 
Left maxilla 
27 
- 
4 mos. 
+ 
- 
C 
Periosteal fibrosar- 
coma 
8. 
164352 
Left maxilla 
22 
- 
10 yrs. 
+ 
- 
C 
Reactive bone pro- 
liferation 
9- 
94803 
Right maxilla 
15 
— 
9 mos. 
+ 
— 
C 
Osteofibroma 
Onset of symptoms followed a 
toothache. 
IO. 
102554 
Right maxilla 
26 
- 
2 mos. 
+ 
C 
Ossifying fibroma 
Ossifying fibroma 
ii . 
110804 
Right maxilla 
24 
— 
6 mos. 
+ 
- 
C 
Fibro-osteoma 
Fibro-osteoma 
Recent onset of diplopia. 
12. 
135364 
Right maxilla 
34 
— 
3h yrs. 
+ 
- 
c 
Neoplasm 
Ossifying fibroma 
Growth recurred after initial re- 
moval. 
13- 
108696 
Right mandible 
22C 
- 
12 yrs. 
+ 
- 
c 
Cyst 
Fibrous dysplasia 
T4 • 
126502 
Right mandible 
37 
- 
6 mos. 
+ 
— 
c 
Adamantinoma 
Ossifying fibroma 
r5- 
99474 
4th cervical vertebra 
20 
+ 
11 mos. 
— 
+ 
c 
Giant cell tumor, 
Giant cell tumor 
Fractured 5th cervical vertebra 7 
sarcoma 
months before onset of symptoms. 
16. 
85995 
Right ilium 
47 
- 
? 
- 
- 
Normal 
c 
Osteodystrophia 
“Rheumatic” pains for many 
years. 
* 
Army Institute of Pathology. 
** 
The race is white, the sex male. If otherwise, it is so indicated. 
Table I 
Bones Involved: Skull, Vertebra, Pelvis Fibrous Dysplasia of Single Bones 
511 
Distribution of Bone Lesions 
Axial Skeleton: Skull} Vertebraey mid 
Pelvis: In the axial skeleton the bones of the 
skull were most often the site of monostotic 
fibrous dysplasia, with the maxilla the preferred 
location (Table 1). In 2 cases the lesion was 
present in the frontal bone; in 1 of these there 
was a history of intermittent headaches for 7 
months. Physical examination revealed that 
The overall distribution of the 67 isolated 
bone lesions has been graphically represented 
in Figure 22. The number of times that the 
various bones were involved is as follows: ribs 
29, femur 9, tibia 8, maxilla 7, calvarium 5, 
mandible 2, humerus 2, ulna 2, vertebra 1, 
pelvis 1, fibula I. It is interesting to note that 
Fig. 15. Case 34. A very early stage in metaplastic fibrous bone formation. The connective tissue cells 
have rounded up; the fibrils in the intercellular matrix are more numerous and have an altered index of 
refraction. Phosphotungstic acid hematoxylin stain. X 300. 
in our series the small bones of the hands and 
feet were not affected. From Figure 22 it will 
be seen that although the distribution of the 
lesions is similar to that of many bone tumors 
and tumor-like conditions, involvement of the 
ribs is much more frequent than in other bone 
diseases. The cases in this series are presented 
according to their anatomic location, taking 
up, in order, those involving the axial skeleton, 
the ribs, the upper and lower extremities. For 
each group representative case histories are 
given. 
the left eye had been displaced downward by 
thickening of the supra-orbital ridge and hori- 
zontal plate of the left frontal bone (Fig. 23). 
There was no loss of visual acuity. The patient 
stated that the bony malformations and as- 
sociated malposition of the left eye had been 
present as long as he could remember. 
The 'parietal bone (Case 3, Fig. 25) and 
occipital bone (Case 5, Fig. 24) were each 
once the site of fibrous dysplasia in the present 
series. The roentgenograms in these cases show 
quite clearly that the outer table was more 512 
The Military Surgeon—N ov ember, 1946 
extensively thinned and expanded than the 
inner. This is true in the majority of cases 
with involvement of the calvarium, whether of 
the polyostotic or monostotic variety, reported 
in the literature. 
It may be noted that most examples of the 
deforming disease of the skull known as 
tion the content of the mastoid process was 
found to be extremely vascular and gritty, 
with the consistency of hard, dried-out cheese. 
The maxillary bone was affected 7 times in 
the present series. A representative history is 
presented by Case 11. The patient was a 24- 
year-old white male who first became aware of 
Fig. 16. Case 34. A more advanced stage in the development of metaplastic bone. The intercellular matrix 
is abundant and the incarcerated connective tissue cells have the appearance of osteocytes. Calcification has 
not yet occurred, the tissue may be identified as osteoid. Phosphotungstic acid hematoxylin stain. X300. 
Fig. 17. Case 38. Stout connective tissue bundles pass from the fibrous tissue into a more mature trabecula 
of metaplastic bone, revealing the fibrous origin of the bone. Phosphotungstic acid hematoxylin stain. X 350. 
leontiasis ossea are regarded by Pugh9 and 
Falconer10 as fibrous dysplasia. 
There was a single example of involvement 
of the mastoid process of the temporal bone 
(Case 4). The patient had suffered frequent 
earaches and aural discharge until he was 
twelve years old. Shortly thereafter his mother 
noticed a swelling behind the left ear, which 
slowly increased in size and was associated 
with intermittent pain in the area. At opera- 
slight asymmetry of his face 6 months before 
reporting for medical examination. About 
three and a half months later his dentist dis- 
covered an enlargement of the maxilla above 
the right bicuspid. Subjectively this soldier 
noticed disturbed tactile sensation over the 
right infra-orbital region and impaired vision 
in the right eye, lately associated with diplopia. 
The mandible was involved in 2 cases. In I, 
that of a 22-year-old colored soldier (Case Fibrous Dysplasia of Single Bones 
513 
Fig. 18. Case 34. A spicule of metaplastic bone being destroyed by osteoclasts. This breakdown of newly 
formed bone accompanied the osteogenesis shown in figures 15 and 16. Phosphotungstic acid hematoxylin 
stain. X300. 
Fig. 19. Case 56. The rather abrupt transition from actively osteogenic connective tissue to one which 
is nonosteogenic is well shown. The physicochemical factors responsible for this difference have not been 
determined. Hematoxylin and eosin stain. X 60. 
Fig. 20. Case 36. Cartilage at the site of a healing pathologic fracture through an area of fibrous dysplasia. 
Hematoxylin and eosin stain. X115. 
13), the lesion grew slowly over a period of 
12 years. An unsuccessful attempt was made 
to remove the growth, which at that time was 
diagnosed bone cyst. It remained painless, but 
reached such proportions (Fig. 26) that 
another operation was undertaken. In the 
second, the case of a 37-year-old white male 
(Case 14), the lesion was manifest for only 
about 6 months during which it grew rapidly. 
A clinical diagnosis of adamantinoma was 
made. 
The vertebral column was involved only 
once in our series (Case 15); the bone af- 
fected was the 4th cervical vertebra. Symp- 
toms of local tenderness and pain radiating 
down both arms began 7 months after the 514 
The Military Surgeon—November, 1946 
CO 
U 
O 
t a 
*•4- 
O CO 
.2 2 
bo 
.S 
ID 
% 
cn 
<L) 
c 
U 
V 
G 
<U 
H 
V 
CO 
aS 
CS 
ri 
U So 
Q 
i 
S w 
s 
<u c 
t! 0 
Case 
AIP* 
Site of Lesion 
Age** 
S § 
fci s 
13 
13 
§ 
X-Ray Diag. 
Path. Diag. 
Remarks 
No. 
Acc. No. 
S H 
Q w' 
0 
0 
m oT 
3 X 
U W 
17- 
122550 
1st, right, posterior f 
27 
- 
3 yrs* 
+ 
- 
E 
Giant cell tumor, 
enchondroma 
Fibrous dysplasia 
Six years ago sudden pain over 
upper chest, felt something snap. 
18. 
123791 
2nd, right, anterior 
29 
+ 
16 mos. 
+ 
+ 
Normal 
E 
Osteochondroma 
Bone cyst 
Possible fracture while wrestling 12 
years before. 
19. 
132100 
2nd, right, axillary line 
30C 
— 
E 
Intramedullary fi- 
broma 
Incidental finding on chest film 
20. 
161586 
2nd, anterior f 
21 
— 
— 
— 
E 
Bone tumor 
Malignant giant 
cell tumor 
Incidental finding on chest film. 
21. 
157355 
3rd, posterior half 
31 
- 
- 
- 
Normal 
E 
Giant cell tumor 
Deferred 
Incidental finding on chest film. 
22. 
97204 
4th, left, anterior 
25 
+ 
1 yrs. 
+ 
+ 
E 
Myleoma, giant 
cell tumor 
Fibrous dysplasia 
Struck in region of clavicle 6 years 
ago, tender for 3 weeks. 
23- 
108976 
4th, left, anterior 
20 
35 yrs. 
+ 
+ 
E 
1940 old frac- 
ture 1944 chondro- 
myxoma 
Fibrous dysplasia 
24. 
106133 
4th, right, posterior \ 
24 
8 mos. 
+ 
Normal 
E 
Chondroma, giant 
cell tumor 
Fibrous dysplasia 
25* 
93572 
4th 
23 
- 
2 yrs. 
- 
+ 
E 
Sarcoma ? 
Osteitis fibrosa 
26. 
119366 
5th, left, posterior 
22 
+ 
Normal 
C 
Metastatic malig- 
nancy 
Fibrosarcoma 
Six years ago fractured several ribs 
on left. Compression fracture of 
4th and 5 th D vertebrae. 
27- 
157045 
5th, left, axillary line 
37 
+ 
2J yrs. 
+ 
Normal 
E 
Bone tumor 
Fibrous dysplasia 
Two-and-one-half years previously 
injury to left chest, 2 months later 
onset of pain. 
Table II 
Bones Involved; Ribs Fibrous Dysplasia of Single Bones 
515 
28. 
107092 
6th, left, axillary line 
34 
- 
8 yrs. 
- 
+ 
E 
Bone tumor 
Fibrous dysplasia 
29. 
114907 
6th, left, posterior 
22 
- 
— 
— 
E 
Fibroma 
Incidental finding on chest film. 
30. 
116461 
6th, left, posterior 
25 
- 
- 
- 
E 
Os teochondroma 
Fibrous dysplasia 
Incidental finding on chest film. 
3" 
94924 
6th, right, posterior 
28 
- 
- 
- 
Normal 
E 
Enchondroma, 
Giant cell tumor 
Incidental finding on chest film. 
giant cell tumor 
32. 
"3477 
6th, right, axillary line 
19 
- 
- 
- 
E 
Bone tumor 
Osteitis fibrosa 
Incidental finding on chest film. 
33- 
78123 
7th, left, posterior 
26 
- 
7 mos. 
+ 
+ 
Normal 
E 
Osteochondroma, 
Osteodystrophia 
Pathological fracture. 
myeloma, giant 
fibrosa 
cell tumor 
34- 
111311 
7th, left 
23 
- 
2 mos. 
- 
+ 
E 
Intramedullary 
fibrosarcoma 
35- 
113482 
7th, left, axillary line 
37 
- 
- 
- 
E 
Osteitis fibrosa 
Incidental finding on chest film. 
36- 
117404 
7th, left, axillary line 
22 
+ 
1 yr. 
- 
+ 
E 
Os teochondroma 
Fibrous dysplasia 
Has been boxing for 4 years. 
37- 
134904 
8th, right, posterior 
26 
- 
1 yr. 
+ 
- 
E 
Giant cell tumor 
Giant cell tumor 
Deep 
x-ray 
38. 
92648 
9th, right, axillary line 
36 
- 
- 
- 
E 
Neoplasm 
Fibro-osteochon- 
Incidental finding on x-ray film. 
drosarcoma 
39- 
106268 
9th, right, posterior 
33 
+ 
4 mos. 
- 
+ 
Normal 
E 
Bone cyst 
Medullary fibrosis 
Struck right chest 1 year ago. 
40. 
130731 
9th, right, posterior 
30 
— 
2-| yrs. 
+ 
+ 
E 
Osteoid osteoma 
41. 
156374 
9th, right, posterior 
37 9 
— 
- 
- 
E 
Enchondroma 
Fibrous dysplasia 
Incidental finding on x-ray film. 
42. 
106908 
10th, right, anterior 
29C 
+ 
8 yrs. 
— 
+ 
E 
Osteochondroma 
Fibrosarcoma 
Crushed ribs on right side 8 years 
43- 
107100 
10th, right, anterior, 
30 
+ 
6 yrs. 
+ 
+ 
E 
Osteochondroma 
Fibrous dysplasia 
ago. 
Fractured rib 3 times: 1931, ’35, 
medial 
’36. 
44. 
146302 
10th, light, middle 
27 
- 
8 mos. 
+ 
+ 
C 
Osteitis fibrosa 
Deferred 
cystica 
45- 
154409 
nth, right, posterior 
34 
— 
— 
— 
E 
Osteitis fibrosa 
Incidental finding on x-ray film. 
*Army Institute of Pathology. 
♦* 
The race is 
white, the sex male. If otherwise, it is so indicated. 516 
The Military Surgeon—November, 1946 
patient fractured his 5th cervical vertebra while 
driving a truck over an obstacle course; 4 
months later another soldier struck him a blow 
on the neck. Biopsy of the vertebra showed the 
typical histologic pattern of fibrous dysplasia. 
A lesion in the right ilium (Case 16) is the 
single instance of involvement of the pelvis 
(Fig. 27). It occurred in the only patient in 
this series who was above the 38-year limit 
for the military age group (47 years). He gave 
a history of repeated attacks of “rheumatism” 
characteristic pattern of connective tissue and 
fibrous bone trabeculae (Fig. 4), Evidence of 
fracture was present in several cases; however, 
the fractures probably were incurred after the 
fibrous dysplasia was well advanced. At the 
site of fracture it was quite impossible to dis- 
tinguish normal callus from the pathologic 
fibrous dysplasia (Fig. 21). 
The relation of trauma to fibrous dysplasia 
of bone is suggested by certain cases in this 
series: Case 17. A 27-year-old white male 
Fig. 21, Case 29. Periosteal new bone formation at the site of a pathologic fracture near the junction of 
normal and abnormal bone. Hematoxylin and eosin stain. X 15. 
for 25 years preceding discovery of the pelvic 
lesion. The role played by the latter in pro- 
ducing these symptoms, which involved his 
feet, knees, and hips, is questionable. 
Ribs: The ribs were involved in 43 per 
cent of the cases in this series (Table 2). The 
fact that in 12 of the 29 cases the lesions of 
the rib were chance findings in roentgeno- 
grams of the chest taken for other purposes 
offers a partial explanation for this preponder- 
ance. The rib lesions usually showed sym- 
metrical expansion and thinning of the cortex 
(Fig. 6 and 28); less often the cortical bone 
was condensed (Fig. 7). Occasionally the 
affected area assumed an almost spherical 
shape and was traversed by more or less stout 
trabeculae of bone (Fig. 26). 
Histologically the rib lesions presented the 
experienced sudden pain in the right shoulder 
and felt something snap while exercising. 
Within 3 weeks he was free of the resulting 
symptoms. Roentgenologic examination of the 
chest at an induction center 3 years later re- 
vealed a tumor-like deformity of the 1st rib. 
The patient was rejected at that time, but 
when he was inducted a year later the condi- 
tion in the rib was unchanged. 
Case 22. A 25-year-old white male was 
struck in the left upper chest while playing 
football in 1936. The area remained tender 
for about 3 weeks, after which symptoms sub- 
sided without specific treatment. He was in- 
ducted into the Army in June, 1942 and a 
routine roentenogram of the chest disclosed 
a deformity of the left 4th rib which was 
interpreted as a congenital anomaly. A year Fibrous Dysplasia of Single Bones 
517 
Fig. 22. Regional distribution of the lesions of monostotic fibrous dysplasia. Each dot on the skeleton 
represents a single instance. 518 
The Military Surgeon—November, 1946 
Fig. 23. Case 1. Roentgenogram of the skull showing diffuse thickening of the left half of the frontal 
bone, with disappearance of the normal orbital outline. 
Fig. 24. Case 5. Involvement of the occipital bone, with expansion of the diploe and thinning of the 
outer table. 
Fig. 25. Case 3. Focal area of fibrous dysplasia in the left parietal bone. Fibrous Dysplasia of Single Bones 
519 
bD 
C/J 
c/5 
OJ 
C 
<L) 
O 
O 
U-. 
O cn 
c £ 
a 
£ 
<u 
T3 
C 
<u 
rt 
_r -2 
“ Oh 
c W 
a> 
E i 
Case 
AIP* 
Site of Lesion 
Age** 
2 § 
.2 2 
’S a 
CO 
H 
O tn 
V -2 
2 § 
OJ 
X-Ray Diag. 
Path. Diag. 
Remarks 
No. 
Acc. No. 
X H 
Q £ 
0 
0 
H-l 
« Oh" 
3 X 
u w 
46. 
105267 
Humerus, left, proxi- 
mal 
22 
5 mos. 
- 
+ 
Normal 
E 
Focal infection 
Brcdie’s abscess 
47- 
161218 
Humerus, left, distal 
37? 
6 mos. 
— 
+ 
Normal 
C 
Bone cyst 
Fibrous dysplasia 
Pathclogic fracture 
treatment. 
while under 
48. 
97435 
Ulna, left, proximal 
25 
— 
— 
C 
Bone cyst 
Fibroma 
Incidental finding 
elbow. 
in x-ray of 
49- 
105039 
Ulna, left, distal 
22 
+ 
2 mos. 
+ 
+ 
C 
Ewing’s tumor 
Giant cell tumor 
Injured arm in fall 
onset of symptoms. 
just prior to 
* Army Institute of Pathology 
** The race 
is white, 
the sex 
male. If otherwise, it is so indicated. 
later, while attempting to leap over a tennis 
net the soldier fell, striking his left chest. 
Roentgenologic examination at that time led 
to a tentative diagnosis of myeloma or giant 
cell tumor. Histologic study of the excised 
tissue revealed that the lesion was fibrous 
dysplasia. 
Case 27. A 37-year-old white soldier 
sustained an injury to the left anterior chest 
wall when he was thrown from his truck in 
March, 1943 during a blackout in Persia. He 
was free of symptoms until 2 months after the 
accident, when he suffered occasional pain, 
aggravated by deep breathing, over the left 
anterior aspect of his chest. In May, 1945, 
the pain became more severe and the soldier 
was hospitalized. At that time roentgenograms 
revealed “an extensive bone tumor involving 
almost the entire left 5th rib.” 
Upper Extremity: Humerus and Ulna: 
The upper extremity was involved 4 times; 
the humerus and the ulna each twice (Table 
3). Duration of symptoms, which consisted 
primarily of local tenderness, ranged from 2 
to 6 months. In Case 48 the lesion was found 
in the ulna following roentgenologic examina- 
tion to discover the cause of pain in the elbow. 
A similar arthritic onset was present in Case 
47. Early in 1945 a 39-year-old corporal in 
the WAC complained of arthritic pain in the 
left arm and was hospitalized with a provi- 
sional diagnosis of psychoneurosis. Roentgeno- 
logic examination revealed a “bone cyst” in 
the distal portion of the left humerus (Fig. 
29). In April 1945 the lesion was curetted 
and the defect bridged by a bone graft. In 
November of that year the patient suffered a 
pathologic fracture (Fig. 30) and another 
bone graft was inserted. On February 19, 
1946 roentgenologic examination revealed that 
the bone graft was well encapsulated and 
almost absorbed; there was satisfactory bony 
union. No fibrous dysplasia was demonstrated 
at this time. . . 
Lower Extremity: Femur, Tibia, and 
Fibula: In the 9 cases in which the femur 
was involved, the lesion was found in the 
proximal end 7 times and in the distal end 
once; in the remaining case the location was 
Table 111 
Bones Involved; Humerus and Ulna The Military Surgeon—November, 1946 
Case 
No. 
AIP* 
Acc. No. 
Site of Lesion 
> 
OQ 
O 
* 
* 
History of 
Trauma 
Duration of 
Symptoms 
Local Swelling 1 
Local Tenderness 
1 
Blood Ca, 
P, Phosphatase 
Curettement—C 
Excision—E 
X-Ray Diag. 
Path. Diag. 
Remarks 
5°. 
99152 
Femur, left, distal 
19 
1 mo. 
— 
+ 
C 
Eosinophilic 
Osteitis fibrosa 
Arthritic pain in left knee. 
granuloma 
localisata 
51* 
101108 
Femur, left, subtro- 
21 
- 
- 
C 
Bone cyst 
Fibrous dysplasia 
Onset with pathologic fracture. 
chanteric 
52. 
104959 
Femur, left, neck 
22 
3 mos. 
- 
+ 
Normal 
c 
Osteitis fibrosa 
Osteitis fibrosa 
Pathologic fracture 6 months after 
cystica 
cystica 
curettement. 
S3- 
128826 
Femur, left, proximal 
25 
- 
- 
c 
Bone cyst 
Fibrous dysplasia 
Incidental finding on x-ray film. 
54- 
129463 
Femur, left, proximal 
19 
1 yr. 
- 
+ 
c 
Bone cyst 
Osteitis fibrosa 
55- 
158202 
Femur, left 
279 
- 
- 
c 
Bone cyst 
Giant cell tumor 
Onset with pathologic fracture. 
56. 
98059 
Femur, right, proxi- 
18 
- 
- 
c 
Bone cyst 
Repair of bone cyst 
Onset with pathologic fracture, 
mal 
refract ired 4 months later. 
57- 
130472 
Femur, right, greater 
37 + 
2 yrs. 
- 
+ 
Normal 
c 
Bone cyst 
Fibrous dysplasia 
23 years ago dislocated head and 
trochanter 
fractured neck of right femur. 
58. 
I3I°33 
Femur, right, tro- 
20 9 
1 yr. 
— 
+ 
c 
Cystic tumor 
Osteitis fibrosa 
Symptoms began in 1st trimester 
chanter 
of pregnancy. 
59- 
105123 
Tibia, left, middle 
21 
4 yrs. 
+ 
— 
E 
Bone cyst 
Giant cell tumor 
60. 
127820 
Tibia, left, proximal 
33 
6 mos. 
— 
+ 
Normal 
C 
Bone cyst 
Bone cyst 
61. 
157068 
Tibia, left, middle 
24 
Many yrs. 
+ 
+ 
Normal 
C 
Osteochondroma 
Fibrous dysplasia 
62, 
95795 
Tibia, right, middle 
21 
3 mos. 
+ 
+ 
C 
Bone cyst 
Osteitis fibrosa 
Bone cyst removed from right fe- 
mur 7 years previously. 
63- 
100637 
Tibia, right, proximal 
30 
— 
— 
c 
Bone cyst 
Osteitis fibrosa 
Incidental finding in x-ray after 
knee injury. 
64. 
101320 
Tibia, right, proximal 
23 + 
3§ yrs* 
+ 
+ 
c 
Chondroma, 
Osteitis fibrosa 
Onset of symptoms after striking 
nonosteogenic 
right knee. 
fibroma 
65. 
105196 
Tibia, right, middle 
20 
— 
— 
c 
Osteitis fibrosa 
Onset with pathologic fracture. 
66. 
128805 
Tibia, right, middle 
20 
3 mos. 
- 
+ 
c 
Bone cyst 
Fibrous dysplasia 
67. 
139888 
Fibula, right, middle 
24 
10 mos. 
- 
+ 
Normal 
c 
Enchondroma, 
Osteitis fibrosa 
Pathologic fracture while under 
fibroma 
observation. 
* Army Instiuute of Pathology 
** The race is white, the 
sex male. If otherwise, it is so indicated. 
TABLE IV 
Bones Involved: Femur, Tibia, Fibula Fibrous Dysplasia of Single Bones 
521 
not given (Table 4). None of the cases 
showed local swelling of the soft tissues. The 
roentgenologic appearance varied; occasionally 
the femur showed a local symmetrical expan- 
sion and thinning of the cortex; in 4 there 
was an associated pathologic fracture (Fig. 
31). In some instances the bone was not ex- 
panded and the radiolucent area was sur- 
of the skeleton was normal. A tentative diag- 
nosis of eosinophilic granuloma of bone was 
offered, after which the diseased area was 
thoroughly curetted. Histologically the lesion 
was typical of fibrous dysplasia. 
In the 8 cases involving the tibia the middle 
of the bone was more frequently affected than 
either end; in the latter region the position 
Fig. 26. Case 13. Area of fibrous dysplasia in the right half of the mandible. 
Fig. 27. Case 16. A large radiolucent area in the right ilium. Biopsy had the appearance characteristic of 
fibrous dysplasia. 
Fig. 28. Case 42. Fibrous dysplasia of the anterior posterior of the right 10th rib. 
rounded by a layer of condensed bone (Fig. 
1 and 2). 
Case 50 gives a representative history. A 
19-year-old white soldier first noticed pain and 
limitation of motion in his right knee. During 
the following month these symptoms became 
more severe and the patient was hospitalized. 
Physical examination was negative except for 
the presence of a diffuse tender bony mass 
in the left popliteal space and impaired flexion 
of the knee. The clinical diagnosis was chronic 
osteomyelitis of the distal end of the femur. 
The roentgenograms showed several “cystic” 
areas and subperiosteal new bone formation 
in the distal end of the femur. The remainder 
was often eccentric (Fig. 3). Local swelling 
of the leg was present in 4 cases; a representa- 
tive history follows. 
Case 64. Three and a half years ago, when 
the patient was 20 years old, he fell and struck 
his knee against a board. Shortly thereafter 
he noticed a swelling at the upper end of the 
right tibia associated with an aching pain. 
Physical examination revealed a firm, slightly 
tender mass measuring 5.0 x 3.7 cm. just 
below the anterior aspect of the knee. A 
roentgenogram showed a large irregular area 
of decreased density lateral to the midline, in 
the proximal metaphysis of the tibia. There 
was thinning and expansion of the cortex as 522 
The Military Surgeon—Novembery 1946 
Fig. 29. Case 47. Fibrous dysplasia of the distal half of the left humerus. The cortex is greatly thinned 
and expanded. 
Fig. 30. Case 47. Pathologic fracture of the humerus which was incurred two weeks after taking the 
roentgenograms in figure 29. 
h ig. 31. Case 56. Fibrous dysplasia in the shaft of the right femur. Onset of symptoms with pathologic 
fracture. Fibrous Dysplasia of Single Bones 
523 
well as condensation of the adjacent cancellous 
bone. The rest of the skeleton was normal. 
The roentgen diagnosis was chondroma, gaint 
cell tumor, or nonosteogenic fibroma. 
Only a single instance (Case 67, Table 
4) of involvement of the fibula (Fig. 4) ap- 
peared in the series. The condition was diag- 
nosed on roentgenologic evidence as an en- 
chondroma. A pathologic fracture occurred 
while the patient was under observation; 
histologically the lesions presented no unusual 
features. 
so diagnosed are really variants of fibrous 
dysplasia, as first suggested by Lichtenstein.5 
He pointed out the identity of the histo- 
logic picture of ossifying fibroma of the maxilla 
with that of fibrous dysplasia in other bones. 
Mallory13 has accepted this interpretation, 
which is also supported by clinicoroentgeno- 
logic evidence. Nevertheless, the recognition 
that these conditions are identical is not wide- 
spread14 and needs to be emphasized. It 
is true that at first glance the spherical islets 
of osteoid or new bone frequently found in the 
maxillary lesions (Fig. 14) may appear quite 
different from those seen, for example, in 
lesions of the ribs (Fig. 11). However, such 
spherical structures are often associated with 
the elongate “typical” bone trabeculae of 
fibrous dysplasia in other bones. Furthermore, 
lesions of the maxilla, that clinically and 
on gross pathologic examination are identical 
with what has been called ossifying fibroma, 
may on histologic examination reveal a struc- 
ture indistinguishable from that of fibrous 
dysplasia (Fig. 32). Stages intermediate be- 
tween the trabecular and spherule type of bone 
formation are also encountered (Fig. 33). 
Careful comparison of the histopathologic 
picture of so-called ossifying fibroma with that 
of fibrous dysplasia has convinced us that the 
former is a variant of the latter, and not a 
separate disease entity. 
N0nosteogenic Fibroma: In the files of the 
Army Institute of Pathology are 12 cases 
classified as nonosteogenic fibroma of bone. 
The lesion was located in the distal metaphysis 
of the femur 7 times, in that of the tibia 3 
times, and once in the proximal metaphysis of 
the tibia and fibula respectively. Roentgenolog- 
ically these lesions could not be distinguished 
from monostotic fibrous dysplasia, since the 
characteristic eccentric position of the former, 
stressed by Jaffe and Lichtenstein,15 was also 
found in proved cases of fibrous dysplasia. On 
the other hand, several of the so-called non- 
osteogenic fibromas were centrally located. 
According to Jaffe and Lichtenstein the 
essential histologic characteristic of nonosteo- 
genic fibroma is the absence of metaplastic 
bone. However, the sections from three of the 
Discussion 
We shall now consider, first, the relation- 
ship of fibrous dysplasia to certain other condi- 
tions, namely, von Recklinghausen’s neuro- 
fibromatosis, ossifying fibroma of bone, and 
nonosteogenic fibroma of bone; second, the 
part played by trauma in the production of 
fibrous dysplasia, and third, the connection 
between the monostotic and polyostotic forms 
of fibrous dysplasia. 
Von Recklinghausen's Neurofibromatosis: 
Thannhauser11 believes that the connective 
tissue pattern of fibrous dysplasia resembles 
that of neurofibroma and would therefore 
identify the lesions, particularly the polyostotic 
form, with von Recklinghausen’s neuro- 
fibromatosis. However, the Bodian stain, em- 
ployed on many of our sections, failed to reveal 
any nerve fibers. The significance of this find- 
ing in casting doubt on the nervous origin of 
the connective tissue is heightened by the 
results obtained by McNairy and Montgom- 
ery12 who studied neurofibromas from 15 
typical cases of von Recklinghausen’s neuro- 
fibromatosis employing the Bodian stain. They 
were able to demonstrate non-medullated 
nerve fibers in 12. Furthermore, the great 
variability in the connective tissue pattern 
found in different regions of the same section 
is unfavorable to Thannhauser’s hypothesis. 
Ossifying Fibroma: Several cases of fibrous 
dysplasia of the maxilla in our series were 
diagnosed by the pathologists who submitted 
them as ossifying fibromas. However, it is 
becoming increasingly clear that the lesions 
of the maxilla and mandible which have been 524 
The Military Surgeon—November, 1946 
12 cases at the Institute showed trabeculae of 
osteoid and new bone (Fig. 34). In all cases 
the connective tissue pattern was identical 
with that of fibrous dysplasia. Lipoid cells, 
another distinguishing though not constant 
feature of nonosteogenic fibroma, were present 
in several typical cases of fibrous dysplasia in 
the present series. Similarly giant cells are 
readily found in both conditions. Since, there- 
bone cysts, one is reluctant to suggest the pos- 
sibility of injury as a factor in the production 
of monostotic fibrous dysplasia. Nevertheless 
there is some evidence in its favor. The high 
incidence of rib involvement may be linked 
to the exposed position of these bones, which 
makes them particularly liable to injury, A 
history of trauma is present in almost one 
third of our cases with rib lesions. However, 
Fig. 32. Case 6. Section of a mass in the left maxilla previously identified as an ossifying- fibroma. The 
characteristic histologic appearance of fibrous dysplasia is apparent. Compare with figure 11. Masson’s 
trichrome stain. X230. 
fore, we find no definite clinical, roentgeno- 
logic, or morphologic criteria by which non- 
osteogenic fibroma of bone and monostotic 
fibrous dysplasia may be distinguished, we re- 
gard the former as a variant of fibrous dys- 
plasia. 
Relation of Trauma to Fibrous Dysflasia: 
After the century old debate centering about 
the relationship of trauma to bone tumors, 
particularly of the giant cell variety, and to 
in none is there proof that the fibrous dyplasia 
did not exist prior to the injury. 
Such evidence, however, is presented in a 
case reported by Zenker,16 a roentgenogram of 
the chest having been taken at the time of 
injury. The patient was a 37-year-old white 
male who suffered a fracture of the right 8th 
and 9th ribs in the posterior axillary line while 
engaged in felling trees. The fractures were 
demonstrated by roentgenologic examination. Fibrous Dysplasia of Single Bones 
525 
Fig. 33. Case 9. Irregular islets of partly calcified osteoid in loose connective tissue from a growth in the 
right maxilla. This histologic appearance is intermediate between the “typical” fibrous dysplasia of the 
preceding case and that shown in figure 14. All three, however, are examples of fibrous dysplasia. Hema- 
toxylin and eosin stain. X230. 
Fig. 34. A.I.P. Acc. 132175. Section of a lesion in the distal end of the left femur. The clinical and 
roentgenologic findings were compatible with nonosteogenic fibroma of bone. Microscopically lipoid cells 
and hemosiderin laden macrophages were abundant, however, widely scattered spicules of metaplastic bone 
were also present. In this case no sharp distinction can be made between fibrous dysplasia and nonosteogenic 
fibroma of bone. 
Following treatment the patient was able to 
resume work 6 weeks after the accident. How- 
ever, he complained of intermittent “rheumatic 
pains” in the right chest which ultimately led 
to another examination of the chest 10 years 
after the initial injury. At this time evidence 
of a healed fracture of the 8th rib was seen 
in roentgenogram. Below the site of the old 
fracture the 9th rib was irregularly expanded 
by a “cystic tumor mass” which extended 
posteriorly from the axillary line for a dis- 
tance of 15 cm. The roentgenogram and 
microphotographs leave no room for doubt 
that the lesions were an example of monostotic 
fibrous dysplasia. 
In recent years it has been recognized that 526 
The Military Surgeon—November, 1946 
fracture of the ribs may occur without external 
trauma. These fractures may be the result of 
opposing muscular tensions associated with 
cough, as in the 18 cases found by Harvey17 
among 500 soldiers ill with virus pneu- 
monia, or the 21 cases among approximately 
3000 tuberculous patients reported by 
Oechsli.18 Fatigue fractures of the ribs follow- 
ing prolonged unaccustomed muscular activity 
such as that to which Army recruits may be 
subjected were described by Matthes and 
Thelen.19 
The manner in which complete or partial 
fracture of a bone may act to produce fibrous 
dysplasia is admittedly unclear. Since this 
lesion often very closely resembles normal 
callus, one might compare it with the keloid 
formation that sometimes follows healing of 
a superficial injury. One of the most ardent 
proponents of the causative relationship be- 
tween trauma and proliferative lesions of bone 
is Konjetzny. In a recent article20 he 
considered this in relation to what he called 
localized osteitis fibrosa and expressed the be- 
lief that a correlation did exist. Only twice in 
our series was there a history of trauma as- 
sociated with monostotic fibrous dysplasia in 
the bones of the lower extremity: once the 
lesion occurred in a femur (Case 57) and once 
in a tibia (Case 64). However, those who 
would seek for a pre-existing injury to ac- 
count for the appearance of fibrous dys- 
plasia in a bone may call upon the occurrence 
of fatigue (march) fractures in these bones. 
The sites of fibrous dysplasia in the tibia and 
femur correspond well with those of fatigue 
fractures in these bones as recently reported by 
Leveton.21 An obstacle to the acceptance 
of this thesis, however, is the well known 
fact that the overwhelming number of fatigue 
fractures occur in the metatarsal bones;22 
whereas, these were never the seat of fibrous 
dysplasia in the series here reported. 
The possible relation of injury to local 
fibrosis in the metaphysis of the long bones is 
supported in a report by Hatcher.23 This 
author believes that the lesion identified by 
Jaffe and Lichtenstein15 as nonosteogenic 
fibroma of bone is not a true tumor. Hatcher 
frequently found it in association with an 
epiphysial disorder due to altered blood supply, 
and suggests that the fibrous lesions of the 
metaphysis may be due to a similar vascular 
derangement. 
In sum, there is a considerable body of 
evidence which links monostotic fibrous 
dysplasia with previous trauma, but the sig- 
nificance of this factor cannot as yet be evalu- 
ated with certainty. 
Relation of the Monostotic to the Polyostotic 
Form of Fibrous Dysplasia: In the 67 cases of 
monostotic fibrous dysplasia comprising the 
present series, there were no recognizable 
congenital anomalies in the bones or other 
organs, and excessive cutaneous pigmentation 
was absent—findings which are characteristic 
of the polyostotic form. Our cases therefore 
do not support the belief that polyostotic, and 
monostotic fibrous dysplasia are different stages 
of the same disease entity. Rather, it may be 
that the bone lesions of both varieties are 
secondary to the action of widely different 
etiologic factors. Thus the polyostotic lesions 
found in young girls affected with precocious 
puberty may be related to the associated high 
estrogen content of the blood acting through 
the parathyroids, as suggested by Bremer-24 
and recently supported by Scholder.25 Differ- 
ent endocrine disturbances may be responsible 
for the multiple bone lesions when they occur 
in men or in women with normal secondary 
sexual characteristics. 
A variety of etiologic factors may produce 
the solitary form of fibrous dysplasia. From a 
study of the cases in this series it seems prob- 
able that many represent an abnormal re- 
sponse to injury. Nothing in the history of 
our cases suggests that the lesions are mani- 
festations of a congenital fault. There is no 
support for the identification of monostotic 
fibrous dysplasia as a hamartoma, a term 
employed by certain investigators of this dis- 
ease.6 It should be remembered that this 
term was introduced by Albrecht26 and 
defined as a tumor-like structure in which 
there is evidence only of an abnormal mixture 
of the normal tissue elements of the organ in 
which it occurs. Since the metaplastic forma- Fibrous Dysplasia of Single Bones 
527 
tion of bone and cartilage by the medullary 
connective tissue is a normal occurrence fol- 
lowing bone fracture, and because this same 
transformation of connective tissue into bone 
is frequently found elsewhere in the body, the 
identity of fibrous dysplasia with hamartoma 
seems unlikely. 
8 Leriche, R., and Policard, A.: The Normal and 
Pathological Physiology of Bone. St. Louis, Mosby, 
1928. 
8 Pugh, D. G.: Fibrous dysplasia of the skull: a 
probable explanation for leontiasis ossea. Radiology. 
44: 548-555, 1945- 
10 Falconer, M. A., Cope, C. L. and Robb-Smith, 
A. H. T.; Fibrous dysplasia of bone with endocrine 
disorders and cutaneous pigmentation. (Albright’s 
Disease). Quart. J. Med. 11: 121-154, 1942. 
11 Thannhauser, S. J.: Neurofibromatosis (von 
Recklinghausen) and osteitis fibrosa cystica localisata 
et disseminata (von Recklinghausen), Medicine. 23: 
105-149, 1944. 
12 McNairy, J. D., and Montgomery, H.: Cutaneous 
tumors of von Recklinghausen’s disease (neurofibro- 
matosis). Report of a histologic study, with special 
reference to nerve fibers and the Bodian stain. Arch. 
Dermat. & Syph. 51:384-390, 1945. 
13 Mallory, T. B.: Pathology: Diseases of bone. 
New England J. Med. 227:955-960, 1942. 
14 Kara, H. J.; Ossifying fibroma of the superior 
maxilla. Arch. Otolaryng. 40; 180-188, 1944. 
15 Jaffe, H. L., and Lichtenstein, L.; Non-osteogenic 
fibroma of bone. Am. J. Path. 18: 205-221, 1942. 
16 Zenker, R.: Zur Frage der traumatischen Ent- 
stehung der Ostitis fibrosa lokalisata. Beitr. z. path. 
Anat. u. allg. Path. 103; 451-56, 1939. 
17 Harvey, R. M.: Rib fractures in atypical pneu- 
monia. Am. J. Roent. & Rad. Ther, 52:487-493, 
1944. 
18 Oechsli, W. R.; Rib fractures from cough. J. 
Thoracic Surg. 5:530-534, 1935“3<5. 
19Matthes, H. G., and Thelen, A.; Ermudungs- 
briiche der Rippen mit typischer Lokalisation. 
Chirurg. 11:537-542, 1939. 
“Konjetzny, G. E.: Sogenannte lokalisierte Ostitis 
fibrosa und Trauma. Monatschrift f. Unfallh. 46: 
572-577, 1939- 
21 Leveton, A. L.: March (fatigue) fractures of the 
long bones of the lower extremity and pelvis. Am. J. 
Surg. 71: 222-232, 1946. 
22 Hartley, J. B.: “Stress” or “fatigue” fractures 
of bone. Brit. J. Radiol. 16: 255-262, 1943. 
23 Hatcher, C. H.; The pathogenesis of localized 
fibrous lesions in the metaphyses of long bones. Ann. 
Surg. 122; 1016-1030, 1945. 
24 Bremer, J, L.: Osteitis fibrosa localisata; an ex- 
perimental study. Arch. Path. 32; 200-210, 1941. 
25 Scholder, B. M.: The syndrome of precocious 
puberty, fibrocystic bone disease, and pigmentation of 
the skin: 11 years observation of a case. Ann. Int. 
Med. 22; 105-118, 1945. 
28 Albrecht, E.: Ueber Hamartome. Verhandl. d. 
deutsch. path. Gessell. 7: 153-157, 1904. 
Summary 
1. During the war the Army Institute of 
Pathology received tissue and records of 69 
cases of fibrous dysplasia that had occurred in 
Army personnel. Of these 67 were monostotic 
and in only 2 were several bones affected. 
2. Remarkable in this series was the fre- 
quency of rib involvement. Of the 29 rib 
lesions, 12 were incidental findings on routine 
chest films. 
3. Evidence is presented which suggests 
that so-called ossifying fibroma and non- 
osteogenic fibroma of bone are variants of 
fibrous dysplasia. 
4. The monostotic form of fibrous dysplasia 
is not a congenital anomaly, and etiologically 
has probably nothing in common with the 
form of polyostotic fibrous dysplasia found in 
Albright’s syndrome. It may represent a dis- 
turbance of the normal reparative processes 
following any of a variety of bone injuries. 
1 Schmorl, G.: Ueber Ostitis deformans, Paget. 
Virch. Arch. f. path. Anat. 283: 694-751, 1932. 
2 Weil, A.: Pubertas praecox und Knochenbriichig- 
keit. Klin. Wchnschrft. Pt. 2: 2114-2115, 1922. 
3 McCune, D. S., and Bruch, H.; Osteodystrophia 
fibrosa. Am. J. Dis. Child. 54: 806-848, 1937. 
4 Albright, F., Butler, A. M., Hampton, A. O., 
and Smith, P.: Syndrome characterized by osteitis 
fibrosa disseminata, areas of pigmentation and endo- 
crine dysfunction, with precocious puberty in females. 
New England J. Med. 216:727-746, 1937. 
5 Lichtenstein, L.: Polyostotic fibrous dysplasia. 
Arch. Surg. 36: 874-898, 1938. 
6 Lichtenstein, L., and Jaffe, H. L.; Fibrous dys- 
plasia of bone. Arch. Path. 33: 777-816, 1942. 
7 Weidenreich, F.: Das Knochengewebe. Handb. d. 
tnikr. Anat. d. Mensch. 2 pt 2, 391-520. Berlin, 
Julius Springer, 1930. 
References BRONCHIAL ADENOMAS* 
By MAJOR SION W. HOLLEY, Medical Corps, Army of the United Statesf 
(With eighteen illustrations) 
Bronchial Adenoma 
noma,13 and anlage and rest tumor.14 As 
pointed out by Mallory,15 the term bronchial 
adenoma is not accurately descriptive, for the 
most common type rarely forms glands, and 
its origin is still in doubt; yet the name is 
so widely used and its implications are so well 
understood that its retention seems advisable. 
THE antemortem recognition of bron- 
chial adenoma has increased steadily 
since the reports of Kramer1 in 1930 
and Wessler and Rabin2 in 1932. Earlier 
diagnosis of a larger number of cases has 
resulted from a growing awareness that cough, 
hemoptysis and other pulmonary symptoms 
often demand more extensive investigation 
than roentgenologic examination of the chest. 
Bronchoscopy and biopsy are indespensable ad- 
juncts. 
Bronchial adenomas constitute approxi- 
mately 8 to 10 per cent of all primary 
bronchopulmonary neoplasms. They usually 
occur in the third and fourth decades, that is, 
in the age group which includes the majority 
of the military personnel; hence they have 
been encountered with relative frequency in the 
larger Army hospitals. The favorable outcome 
to be anticipated in these cases is in decided 
contrast to the grave prognosis offered victims 
of bronchogenic carcinoma, but early differen- 
tial diagnosis is essential for their proper treat- 
ment. 
This study is based upon records and speci- 
mens from 39 cases at the Army Institute of 
Pathology. It includes an account of the 
clinical features, the morphology of the neo- 
plasms, and the results of treatment. Fourteen 
of the tumors were found in civilians and 25 
in military personnel. The relevant data have 
been summarized in Table 1. Three cases 
have been reported previously.3’4’5 
Since the recognition of this tumor as an 
entity, it has been given many names: poly- 
poid adenoma,6 basal cell carcinoma,7 adenom- 
atous polyp,8 benign glandular bronchogenic 
tumor,9 vascular adenoma,10 carcinoid and 
cylindroma,11 mixed tumor,12 malignant ade- 
Literature 
Many excellent reviews of the literature 
relating to the various aspects of bronchial 
adenoma have appeared within recent years. 
The reports of Womack and Graham,12 Gold- 
man and Stephens,16 Brunn and Goldman,1' 
Foster-Carter,18 Anderson,19 Stout,20 Laff and 
Neubuerger,21 and Riordan and Richards22 
have included most of the papers on the sub- 
ject published prior to 1944. The most re- 
cent contributions are those of Nager,23 Jack- 
son, Konzelmann and Norris,24 Graham and 
Womack,25 Chamberlain and Gordon,26 Har- 
ris,27 and Moersch, Tinney and Mc- 
Donald.28’29 In all, over 250 cases have been 
recorded. 
Representative Cases 
Seven of the 39 cases have been selected 
to illustrate various clinical and pathologic 
features of bronchial adenoma. 
Case g. A IP. Acc. No. 84528: 
A white male, 27 years old, had an acute 
respiratory infection with pleurisy and slight 
effusion in 1941. In June 1942 he had an- 
other attack characterized by malaise, fever, 
cough, expectoration, dyspnea, and pain in the 
lower part of the right chest. Roentgenograms 
showed elevation of the right diaphragm and 
diffuse mottling at the base of the right lung. 
These changes persisted, although the acute 
illness subsided in 12 days. Bronchoscopy re- 
vealed a granular tumor which bled freely 
on contact and which had produced a narrow- 
ing of the right lower lobe bronchus. The 
* From the Army Institute of Pathology, 
t Nassau County Sanatorium, Farmingdale, New 
York. Bronchial Adenomas 
529 
patient died during bronchoscopy I year and 
5 months after the onset of symptoms. Au- 
topsy was not performed, the histologic de- 
scription being based on biopsy material. 
Pathologic Findings: The bronchial epi- 
thelium which covered the tumor had under- 
gone squamous metaplasia and was separated 
from it by a narrow band of connective tissue. 
The tumor cells were epithelial, medium- 
sized, usually polygonal, with pale, finely gran- 
ular cytoplasm, and round or oval, slightly 
hyperchromatic nuclei. Mitoses were not seen. 
The cells were arranged in cords and nests 
that simulated the pattern and morphology of 
appendiceal carcinoids. The stroma consisted 
largely of bone, though in one area there 
were hyalinized bundles of connective tissue 
that resembled the matrix of a mixed tumor. 
The final diagnosis was bronchial adenoma, 
carcinoid type. 
Case 10. AIP. Acc. No. 84794: 
A white male, 35 years of age, had had 
pneumonia in 1936. An acute pulmonary in- 
fection in June 1942 was accompanied by 
cough and expectoration of dark blood. Symp- 
toms subsided after 20 days, although the 
cough persisted with a small amount of sputum 
but no blood. Roentgenograms showed a round 
mass, 4 cm. in diameter, in the left pulmonary 
hilus, and an area of increased density extend- 
ing from the hilus to the left base. These 
changes were interpreted as a tumor of the 
left lower lobe with atelectasis. The diagnosis 
made from bronchoscopic biopsy specimens 
was bronchial adenoma. Lobectomy was ad- 
vised but refused. 
The patient continued to work for almost 
2 years, but he tired easily, had a chronic 
cough, and occasionally expectorated blood- 
streaked sputum. Periodic roentgenograms of 
the chest showed gradual enlargement of the 
tumor. Left lower lobectomy was performed 
in October 1944, and the patient was dis- 
charged from the hospital in February 1945 
after the subsidence of postoperative pleural 
effusion. Subsequent follow-up data were not 
available. 
Pathologic Findings: A soft, well-circum- 
scribed neoplasm measuring 5 x 5 x 3.5 cm. 
was found embedded in the wall of a large 
bronchus. A pedunculated extension of the 
tumor filled the bronchus for a distance of 
5 cm. The distal bronchi were dilated. 
At microscopic examination it was seen 
that the portion of the tumor which protruded 
into the bronchus was covered by cuboidal and 
respiratory epithelium. Beneath this was a thin 
layer of compressed fibrous tissue, infiltrated 
by tumor cells. The bulk of the intramural 
tumor formed a fairly well circumscribed lesion 
that surrounded the bronchial cartilages and 
displaced the remaining normal structures. At 
the periphery of the main mass were small 
nests of infiltrating tumor cells. 
The cells in some regions of the neoplasm 
were arranged in cords reminiscent of ap- 
pendiceal carcinoids. Here and elsewhere in 
the tumor they were epithelial in character, 
with scanty polychromatic and finely granular 
cytoplasm. The nuclei, though usually uni- 
form, were occasionally variable in size and 
chromatin content; nucleoli were rarely seen; 
mitoses were absent. The Fontana stain 
showed silver-positive intracellular granules in 
the biopsy specimen, although the same stain 
failed to demonstrate them in the main por- 
tion of the tumor. The connective tissue 
stroma was scanty and moderately vascular. 
The lung parenchyma was partially atelectatic 
but not significantly inflamed. The final diag- 
nosis was bronchial adenoma, carcinoid type. 
Case 12. AIP. Acc. No. 86741: 
A white male, 38 years of age, entered a 
hospital in November 1942 with a history of 
persistent cough and hemoptysis following 
pneumonia. Roentgenograms showed atelec- 
tasis of the right lower lobe, where a tumor 
was subsequently demonstrated by bronchos- 
copy. The first diagnosis on the tissue from 
biopsy was bronchogenic carcinoma, but after 
further study this was changed to bronchial 
adenoma. A right lower lobectomy was per- 
formed in December 1942. The patient was 
transferred to another hospital from which he 
was discharged in September 1943. There 
was no evidence of recurrence at that time. 530 
The Military Surgeon—November, 1946 
Pathologic Findings: At the hilus the ori- 
fice of the lower lobe bronchus was filled 
with an encapsulated tumor, 4 cm. in diameter, 
that had its origin in the bronchial wall. Its 
cut surface was tan and homogeneous. 
Microscopic examination showed that the 
tumor was separated from the lumen of the 
bronchus by respiratory epithelium and asso- 
ciated, partly calcified, connective tissue. In- 
corporated in the tumor were nerves, bron- 
chial cartilages, glands, and ducts; peripheral- 
ly, tumor cells infiltrated the bronchial wall. 
Some of the incorporated cartilages were 
undergoing ossification. Athough a few spicules 
of bone were found free in the tumor, they 
were always near bronchial cartilage. Some 
of the bone contained normal marrow, but 
elsewhere neoplastic cells filled the spaces be- 
tween the spicules. The tumor cells were epi- 
thelial in character with pale, finely granular 
cytoplasm and round to oval, medium-sized 
nuclei containing a moderate amount of 
chromatin. Mitoses were not observed. The 
cells were arranged in cords and nests sepa- 
rated by delicate fibrous strands or dense col- 
lagenous bundles. Some of the latter contained 
calcium granules. The pattern was indistin- 
guishable from that of the appendiceal carci- 
noids. There were a few pseudoglands, the 
central spaces of which were formed by small 
capillaries and their surrounding connective 
tissue. Nests of neoplastic cells were found in 
the capsule of a peribronchial lymph node. 
The final diagnosis was bronchial adenoma, 
carcinoid type. 
Case 75. AIP. Acc. No. 94639: 
A white male, 28 years old, first had 
hemoptysis accompanied by cough and expec- 
toration in 1930. Between then and 1937 
he had three attacks of pneumonia. His cough 
increased; he lost weight and became weak. 
In 1937 roentgenograms revealed atelectasis 
of the left lung; tissue removed from the left 
main bronchus was diagnosed adenoma. Addi- 
tional specimens examined during the next 4 
years were similarly classified. The patient 
was fairly well until February 1943 when 
he had scarlet fever complicated by renal in- 
sufficiency. At that time roentgenograms 
showed the trachea and mediastinum shifted 
to the left and the left lung still atelectatic. 
The patient died in uremia 28 days after the 
onset of the scarlet fever and 13 years after 
the first symptoms of the bronchial adenoma. 
Pathologic Findings: At autopsy the heart 
and mediastinum were displaced to the left; 
the left pleural space was obliterated by dense 
fibrous adhesions. The left lung weighed 850 
gm., showed extensive fibrosis, and in its 
lower lobe was bronchiectatic. The bronchi 
were patent and presented no evidence of 
tumor. In the left upper lobe was a thick- 
walled cavity, 3 cm. in diameter. One hilus 
lymph node measured 3x2x2 cm. and had 
a “fish flesh” appearance. The spleen, liver and 
kidneys were enlarged due to the deposition 
of amyloid. 
At microscopic examination it was found 
that the hilus lymph node was almost wholly 
replaced by a metastatic tumor. The neoplastic 
cells were arranged in interlacing cords sepa- 
rated by moderately vascular connective tissue. 
The structure and arrangement of the 
cells were similar to those observed in the 
biopsy tissue obtained in 1937, 1938 and 1941, 
which resembled appendiceal carcinoid. Final 
diagnosis was bronchial adenoma, carcinoid 
type, metastatic in a hilus lymph node. 
Case 37. AIP. Acc. No. 80340: 
A white female, 37 years of age, first had 
fever, cough and signs of right middle and 
upper lobe pneumonia in 1933. Roentgeno- 
grams disclosed a slowly resolving infiltra- 
tion in the right upper lobe and a nodular 
shadow in the right hilus. Subsequently he- 
moptysis occurred with menstruation several 
times. After X-ray therapy the nodule ap- 
peared to regress but became more prominent 
m 1942, although no abnormality was noted 
at bronchoscopy. X-ray therapy was adminis- 
tered again in November 1942. In August 
1943, the patient complained of cough and 
occasional hemoptysis. Two months later a 
right middle and upper lobectomy was per- 
formed. The pathologic diagnosis was ade- 
noma of the right upper lobe bronchus with Bronchial Adenomas 
531 
questionable malignancy. The patient was well 
until November 1945 when a laparotomy for 
suspected ectopic pregnancy revealed several 
tumor nodules, 5 mm. to 5 cm. in diameter 
in the liver, but no ectopic pregnancy. The 
appendix was removed and also the uterus 
which contained fibroleiomyomas. 
Pathologic Findings: Microscopic examina- 
tion of tissue from the lobe removed at opera- 
tion showed the tumor to be separated from 
the bronchial lumen by squamous epithelium 
and an underlying zone of collagenous fibrous 
tissue. It was composed of epithelial cells ar- 
ranged in cords and irregular masses separated 
by a moderate amount of connective tissue. 
The cytoplasm of the tumor cells was faintly 
polychromatic and granular; the nuclei were 
round or oval and moderately hyperchromatic. 
Mitoses were not seen. A few spicules of 
bone in the periphery of the tumor represented 
ossified bronchial cartilages. 
The neoplastic tissue from the liver was 
made up of the same type of cells, but the 
cords and masses in which they were arranged 
were less well defined than in the primary 
tumor. The cell structure and arrangement 
of both the bronchogenic tumor and the he- 
patic metastasis were consistent with those of 
carcinoid. There was no evidence of carcinoid 
in the appendix. 
The final diagnosis was bronchial adenoma, 
carcinoid type, metastatic in the liver. 
These five cases represent the same type of 
bronchial adenoma. In Case 10, the unusual 
nuclear pleomorphism was considered insuffi- 
cient for an unequivocal diagnosis of malig- 
nancy. Case 25 (Table 1) showed a similar 
variation but was classified as an adenoma. 
Case 35. A IP. Acc. No. 73373: 
A white woman, 37 years old in 1940, 
had suffered from mild thoracic pain for the 
previous year. A routine roentgenogram of 
the chest made in August 1935 had been 
normal; another in 1938 had shown a rounded 
density, 1 cm. in diameter, in the lower lobe 
of the left lung near the hilus; this shadow 
appeared much larger in June 1940. A few 
months later a left lower lobectomy was per- 
formed. There was evidence of recurrence 5 
years after operation. 
Pathologic Findings: A well circumscribed 
4.5 x 4 x 4 cm. tumor was present 2 cm. 
from the hilus and 0.5 cm. beneath the pleura. 
Its cut surface was grayish white, the center 
streaked with yellow. 
Microscopically it was seen that the tumor 
arose in the bronchial wall, surrounded the 
cartilages, and filled several pulmonary alveoli. 
It was composed of irregular masses of small, 
deeply stained, spindle-shaped cells with scant 
cytoplasm. The nuclei were small and hyper- 
chromatic. Throughout the lesion were nests 
of concentrically arranged cells that were par- 
tially keratinized or calcified and bore a re- 
semblance to Hassall’s corpuscles. All stages of 
transition from the typical tumor cells to the 
keratinized nodules were demonstrable. The 
stroma was scant in some areas, whereas else- 
where there were prominent bundles of col- 
lagenous connective tissue. 
Frequently the periphery of the cell nests 
showed palisading similar to that observed in 
basal cell carcinomas of the skin. In other 
areas there were small spaces containing an 
amorphous precipitate which failed to stain 
with mucicarmine. These portions of the tumor 
were like some of the mixed tumors of salivary 
glands and also resembled epithelioma ade- 
noides cysticum. 
Originally the tumor was classified as an 
epidermoid carcinoma, basal cell type. The 
final diagnosis was bronchial adenoma, mixed 
tumor type. 
Case 57. A IP. Acc. No. 84802: 
A white male, 28 years old, had pneumonia 
in March and again in April of 1942; these 
episodes were followed by a productive cough. 
Roentgenograms showed atelectasis of the 
right lower lobe, and bronchoscopic examina- 
tions in July 1942 revealed a polypoid tumor 
in the right lower lobe bronchus. The diag- 
nosis following biopsy was bronchial adenoma. 
Additional specimens obtained in August 1942 
were called low grade adenocarcinoma in one 
instance and adenofibroma in another. The 
right lung was removed during the following The Military Surgeon—Novembery 1946 
Case 
No. & 
Accession 
No. 
Sex 
Age 
Predominant 
Symptoms & 
Signs 
Duration 
Roentgenographic 
Changes 
Location 
of 
Tumor 
Biopsy 
Diagnosis 
Method of 
Removal 
Result after Final 
operation Classification 
*i 
60845 
F 
44 
Cough, sputum, chills, 
fever 
1 yr. 
Atelectasis, left lung 
Left lower lobe 
Adenoma, 
Adenomacar- 
cinoma 
Bronchoscopy 
Well, 4 yrs. Carcinoid type 
*2 
74460 
F 
27 
Pleurisy with effusion 
10 mos. 
Atelectasis, right lower lobe 
Right lower lobe 
Adenoma 
Pneumonectomy 
Well, 5 yrs. 8 mos. Carcinoid type 
3 
74517 
F 
37 
Fatigue, cough, fever 2 yrs. 5 mos. 
Inflammatory infiltration. 
Lt. upper lobe 
Left upper lobe 
Adenoma 
Bronchoscopy 
Well, 4 yrs. 8 mos. Carcinoid type 
4 
81971 
M 
23 
Cough, expectoration 
Hemoptysis 
3 y^. 
4 mos. 
Inflammatory infiltration, 
Lt. lower lobe 
Left main bron- 
chus 
Carcinoma 
Pneumonectomy 
Dyspnea, 3 yrs. 1 Carcinoid type 
mo. 
OO 
O CTi 
Oj 
F 
32 
Cough hemoptysis 
6 mos. 
Atelectasis 
Rt. lower lobe; mass, rt. 
hilus 
Right lower lobe 
Carcinoma 
Pneumonectomy 
Well, 9 yrs. 11 mos. Carcinoid type 
6 
82037 
F 
45 
Cough 
Hemoptysis 
4 yrs. 
3 mos. 
Atelectasis, left upper lobe 
Left upper lobe 
Adenoma 
Pneumonectomy 
Well, 6 yrs. 8 mos. Carcinoid type 
*7 
82038 
F 
37 
Hemoptysis associ- 
ated with menstrua- 
tion 
8 yrs. 
Atelectasis left lung 
Left main bron- 
chus 
Adenoma 
Pneumonectomy 
Well, 4 yrs. n mos. Carcinoid type 
8 
82042 
M 
48 
Recurrent respiratory 
infection 
4 yrs. 3 mos. 
Atelectasis, left lung 
Left upper lobe 
None 
Pneumonectomy 
Well, 3 yrs. Carcinoid type 
9 
84528 
M 
27 
Recurrent pleurisy 
1 yr. 
Atelectasis, right lower lobe 
Right lower lobe 
Endobronchial Bronchoscopy 
sarcoma 
Expired at bron- Carcinoid type 
choscopy 
10 
84794 
M 
35 
Cough 
Hemoptysis 
3 yrs. 
4 mos. 
Mass near left hilus 
Left lower lobe 
Adenoma 
Lobectomy 
Well, 7 mos. Carcinoid type 
11 
85956 
M 
21 
Pain in right chest, 
cough 
“several 
years” 
Atelectasis, right lower lobe 
Right lower lobe 
Carcinoma 
Adenoma 
Bronchoscopy 
No follow-up Carcinoid type 
12 
86741 
M 
38 
Hemoptysis, pneumo- 
nia 
Not known 
Atelectasis, right lower lobe 
Right lower lobe 
Carcinoma 
Adenoma 
Lobectomy 
No follow-up Carcinoid type 
90469 
M 
21 
Cough, dyspnea 
8 mos. 
Inflammatory infiltration 
Lt. Iw. lobe 
Left lower lobe 
Adenoma 
Bronchoscopy 
Well, 7 mos. Carcinoid type 
Table I Bronchial Adenomas 
14 
9°99S 
M 
24 
Hemoptysis 
5 mos. 
None 
Left main bron- 
chus 
Adenoma 
Bronchoscopy 
Well, 2 yrs. 
Carcinoid type 
15 
94639 
M 
28 
Recurrent broncho- 
pneumonia hemoptysis 
7 yrs- 
Atelectasis left lung 
Left main bron- 
chus 
Adenoma 
Bronchoscopy 
Expired 2 yrs. af- 
ter last broncho- 
scopy 
Carcinoid type 
16 
100862 
F 
3° 
Recurrent respiratory 
infection 
Hemoptysis 
9 yrs- 
5 mos. 
Mass rt. hilus. Inflamma- 
tory infiltration, rt. base 
Right lower lobe 
None 
Lobectomy 
Well, 1 yr. 9 mos. 
Carcinoid type 
I? 
100866 
M 
!9 
“Atypical pneumonia” 
4 mos. 
Inflammatory infiltration, 
rt. lower lobe 
Right lower lobe 
Carcinoma 
Pneumonectomy 
No follow-up 
Carcinoid type 
18 
M 
23 
Not known 
Not known 
Not known 
Right lower lobe 
Adenoma 
Bronchoscopy 
No follow-up 
Carcinoid type 
102721 
J9 
F 
43 
Hemoptysis 
3 wks. 
Atelectasis, left lung 
Left lower lobe 
Carcinoma 
Pneumonectomy 
Well, 1 yr. 10 mos. 
Carcinoid type 
103405 
20 
103522 
F 
73 
Tumor was incidental 
finding at autopsy af- 
ter death from hyper- 
tension 
Left upper lobe 
Carcinoid type 
21 
M 
23 
Recurrent fever, cough 
4 mos. 
Atelectasis, left upper lobe 
Left upper lobe 
Adenoma 
Bronchoscopy 
Well, 2 months 
Carcinoid type 
105012 
22 
M 
35 
Recurrent pneumonia 
1 yr. 8 mos. 
Atelectasis, rt. middle lobe 
Right middle lobe 
None 
Lobectomy 
Well, 1 yr. 9 mos. 
Carcinoid type 
106993 
23 
109382 
M 
29 
Recurrent cough, fe- 
ver, wheeze 
1 yr. 
Inflammatory infiltration 
with lung abscess, It. Iw. 
lobe 
Left lower lobe 
Carcinoma 
Adenoma 
Pneumonectomy 
Well, 1 yr. 
Carcinoid type 
24 
116096 
M 
37 
Cough, 
Hemoptysis 
11 yrs. 
2 yrs. 
Atelectasis right lower lobe 
Right lower lobe 
Adenoma 
Lobectomy 
Well, 3 mos. 
Carcinoid type 
OO 
M 
38 
Cough, expectoration, 
Hemoptysis 
5 yrs. 
2 yrs. 
Inflammatory infiltration, 
rt. Iw. lobe 
Right middle lobe 
Carcinoma 
Pneumonectomy 
Expired after 28 
days from em- 
pyema 
Carcinoid type 
26 
118167 
M 
24 
Recurrent respiratory 
infection 
11 mos. 
Atelectasis, left lower lobe 
Left upper lobe 
Adenoma 
Pneumonectomy 
Dyspnea, I yr. 3 
mos. 
Carcinoid type 
00 
c* 00 
F 
34 
Hemoptysis 
11 mos. 
Atelectasis, inflammatory 
infiltration rt. Iw. lobe 
Right lower lobe 
Adenoma 
Lobectomy 
Well, 4 mos. 
Carcinoid type 
28 
I2059I 
M 
3i 
Cough, pain in right 
chest 
Not known 
Mass near right hilus 
Right lower lobe 
None 
Lobectomy 
Well, 5 mos. 
Carcinoid type Sr 
Co 
s 
o 
s 
4 
o 
V# 
VD 
-K 
0\ 
Case 
No. & 
Accession 
No. 
Sex 
Age 
Predominant 
Symptoms & 
Signs 
Duration 
Roentgenographic 
Changes 
Location 
of 
Tumor 
Biopsy 
Diagnosis 
Method of 
Removal 
Result after 
operation 
Final 
Classification 
29 
121943 
M 
25 
Not known 
Not known 
Not known 
Right lower lobe 
Adenoma 
Bronchoscopy 
No follow-up 
Carcinoid type 
3° 
13271:9 
M 
48 
Trauma, pain, left 
chest 
Hemoptysis 
4 mos. 
1 mo. 
None 
Left upper lobe 
Carcinoma 
Pneumonectomy 
Well, 5 mos. 
Carcinoid type 
3i 
80340 
F 
37 
Hemoptysis with men- 
struation 
9yrs. 
Atelectasis, right middle & 
upper lobes 
Right middle lobe 
none 
Lobectomy 
Liver metastases 
after 2 yrs. 
Carcinoid type 
32 
155988 
M 
22 
None. Found at rou- 
tine examination 
Not known 
Right lower lobe 
Adenoma 
Lobectomy 
Well, 7 mos. 
Carcinoid type 
33 
155198 
F 
35 
Cough 
Acute respiratory in- 
fection 
6 mos. 
3mos. 
Atelectasis, right lower lobe 
Right lower lobe 
Carcinoma 
Pneumonectomy 
Well, 6 mos. 
Carcinoid type 
34 
156546 
M 
24 
Acute respiratory in- 
fection 
9 mos. 
Atelectasis, right lower lobe 
Right lower lobe 
Adenoma 
Lobectomy 
Well, 1 mo. 
Carcinoid type 
35 
73373 
F 
37 
Found at routine ex- 
amination 
2 yrs. (be- 
fore opera- 
tion) 
1 yr. 
Mass near left hilus 
Left lower lobe 
None 
Lobectomy 
Recurred at 5 yrs. 
Mixed tumor type 
Pain in left chest 
36 
81914 
F 
35 
Cough, dyspnea 
Hemoptysis 
3 y«. 
2 yrs. 
Masses, right hilus 
Right lower lobe 
Carcinoma 
Bronchoscopy 
Bloody pleural ef- 
fusion; pleural no- 
dules 3 yrs. 
Mixed tumor type 
37 
84802 
M 
28 
Chest pain 
Recurrent pneumonia 
1 yr. 
6 mos. 
Atelectasis, inflammatory 
infiltration rt. Iw. lobe 
Right main bron- 
chus 
Adenoma 
Carcinoma 
Adenofibroma 
Pneumonectomy 
Well, 3 yrs. 6 mos. 
Mixed tumor type 
38 
95277 
M 
38 
Acute respiratory in- 
fection and slight 
weight loss 
5 mos. 
Inflammatory infiltration 
It. Iw. lobe 
Left main bron- 
chus 
Carcinoma 
Pneumonectomy 
No follow-up 
Mixed tumor type 
39 
112112 
M 
26 
Cough, expectoration 
fever 
11 yrs. 
5 mos. 
Atelectasis, right lower lobe 
Right lower lobe 
Cylindroma 
Bronchoscopy 
Bronchiectasis 10 
mos. 
Mixed tumor type 
* Case i reported by Schwartz (4); Case 2 by Tyson & Millikcn (5); Case 7 by Overholt (3). 
Table I—{continued) Bronchial Adenomas 
535 
October; convalescence was uneventful. Two 
years and three months after pneumonectomy 
there was no evidence of recurrence, and the 
patient was asymptomatic. 
Pathologic Findings: A segment of the right 
main bronchus bore a smooth fusiform mass, 
measuring 1.7 x 1.3 x 1.2 cm., that bulged 
into the lumen and narrowed it to one half its 
normal diameter. The cut surface was tan 
after formalin fixation; several embedded 
bronchial cartilages were visible. The tumor 
had almost penetrated the bronchial wall. The 
type of bronchial adenoma, for the cell pat- 
tern closely resembles that of certain mixed 
tumors of the salivary glands and epithelium 
adenoides cysticum, and is wholly different 
from the more frequently observed carcinoid 
variety. 
Clinical and Pathologic Analyses 
Fourteen (36 per cent) of the tumors were 
in females, and 25 (64 per cent) were in 
males (Table 2). The neoplasm was dis- 
covered during life in all cases except one 
Table 2 
Sex and Age Distribution by Years 
Total 
Under 
20 
20-29 
30-39 
40-49 
50-59 
60-69 
70 
Female 
14 
0 
1 
9 
3 
0 
0 
1 
Male 
25 
1 
14 
8 
2 
0 
0 
0 
Total 
39 
1 
15 
I? 
5 
0 
0 
1 
greater part of the lung was atelectatic; the 
bronchi were not dilated. 
Microscopic examination showed partial des- 
quamation of the epithelium of the bronchial 
mucosa, but where present it was of squamous 
type and separated from the tumor by a zone 
of chronically inflamed fibrous tissue. The 
neoplastic cells extensively infiltrated the outer 
portion of the bronchial wall. In many regions 
the cells formed tortuous branching tubules; 
elsewhere they appeared as sharply outlined 
nests containing small spaces, some of which 
were filled with amorphous material, which 
did not stain with mucicarmine. The tumor 
cells were polygonal, had a scant amount of 
cytoplasm, and contained small, round or oval, 
hyperchromatic nuclei. Mitoses were not 
found. The connective tissue stroma presented 
no unusual features. The pattern assumed by 
the tumor cells was similar to that found in 
some mixed tumors of salivary glands and in 
epithelioma adenoides cysticum. The final 
diagnosis was bronchial adenoma, mixed- 
tumor type. 
The last two cases represent a less common 
(Case 20), in which a 6 mm. nodule of the 
carcinoid type was found after death from 
hypertensive heart disease. Seventy-one per 
cent of the females and 88 per cent of the 
males were between the ages of 20 and 40 
years. The average age at onset of symptoms 
was 32 years in the women and 25 years in 
the men. The longest duration of symptoms 
among the women was 9 years and 5 months 
with an average of 3 years and 2 months; 
among the men it was 11 years and 5 months 
with an average of 2 years and 7 months. 
Cough was a symptom in 66 per cent of 
cases; hemoptysis (including blood-streaking 
as well as severe hemorrhage) in 44 per cent; 
expectoration in 40 per cent; some form of 
chest pain, usually mild, in 36 per cent; fever 
in 36 per cent; dyspnea in 20 per cent; weight 
loss in 11 per cent, and wheeze in 9 per cent, 
(Table 3). Hemoptysis was concurrent with 
menstruation in two cases. Thirty-three per 
cent of the patients had had one or more at- 
tacks of acute pulmonary infection, apparently 
secondary to bronchial obstruction. 
Clinical laboratory studies had no specific. 536 
The Military Surgeon—November, 1946 
diagnostic significance. 
Roentgenographically the tumor was de- 
monstrable in 8 cases, atelectasis in 16, and 
inflammatory infiltration in 9 others. Figures 
I, 2 and 3 illustrate these and other changes. 
The affected bronchus was partially or 
completely obstructed in 19 cases, (Fig. 1 a). 
Bronchoscopic biopsy was performed in 34 
cases resulting in immediate diagnosis of 
At the last follow-up examinations of the 
remaining patients there was evidence of local 
recurrence in one (Case 35); hepatic metas- 
tases were found at laparatomy in another 
(Case 31). In case 36, with a mixed-tumor 
type of bronchial adenoma, no further attempt 
at removal was made after the original 
bronchoscopy. This patient had had partial col- 
lapse of the right lung and bloody pleural 
Table 3 
Summary of Symptoms 
Hemop- 
Pleural Weight 
Cough 
tysis Sputum Fever 
Pain Dyspnea Effusion Loss 
Wheeze 
Female 
7 
8 
2 4 
1 
2 
3 1 
0 
Male 
18 
8 
12 9 
12 
5 
3 3 
3 
Total 
25 
16 
H 13 
13 
7 
6 4 
3 
Table 4 
Location of Tumors 
Right Lung 
Left Lung 
Upper 
Middle 
Lower 
Upper Lower 
Main Lobe 
Lobe 
Lobe 
Main 
Lobe Lobe 
Bronchus Bronchus 
Bronchus 
Bronchus 
Total 
Bronchus 
Bronchus Bronchus 
Total 
I 2 
2 
15 
20 
5 
7 7 
19 
bronchial adenoma in 18; the same diagnosis 
after additional biopsy or review of the orig- 
inal sections in 5, carcinoma in 14, cylin- 
droma in 1, and endobronchial sarcoma in 1. 
The tumor was removed bronchoscopically 
in 9 cases. There were 11 lobectomies, with 
2 lobes of the right lung removed in 2 in- 
stances. Pneumonectomy was performed in 15 
cases. 
Three of the 38 patients in whom the tumor 
was diagnosed during life have since died. 
One of these (Case 9) died during bronchos- 
copy 4 months after diagnosis. Another (Case 
15) died of scarlet fever, amyloidosis, and 
subacute interstitial nephritis 2 years after the 
last bronchoscopic removal of tumor. The third 
death (Case 25) resulted from empyema 28 
days after pneumonectomy; autopsy did not 
reveal any residual tumor. 
effusion for 3 years. The last roentgenogram 
showed rounded pleural densities which might 
be interpreted as tumor nodules. The longest 
period of observation after operation was 10 
years (Case 5). Nothing is known of 3 pa- 
tients after the immediate postoperative period. 
The tumor was located in the right lung 
in 20 cases and in the left in 19 (Table 4). 
The most frequent site was in the right lower 
lobe bronchus (15 cases). All of the tumors 
were located at or near the pulmonary hilus. 
They measured from 5 mm. in diameter to 
5 x 5 x 3.5 cm. and were round and often 
polypoid with broad attachments to the bron- 
chial walls (Figs. 4, 5 and 6). Some were 
soft; others were firm; their surfaces were 
smooth and pink to red. All of them were 
well demarcated grossly from the adjacent 
non-neoplastic tissue. Microscopic examina- Bronchial Adenomas 
537 
Fig. i. a. Case 26. Blockage of the left lower lobe bronchus with atelectasis and bronchiectasis secondary 
to carcinoid type of bronchial adenoma. Neg. 84296. b. Case 37. Atelectasis of the right lower lobe due to 
obstruction by a mixed-tumor type of adenoma in the right main bronchus. Neg. 84171. 
Fig. 2. a. Case 12. Slight medial atelectasis and inflammatory infiltrate in the right lower lobe. The 
adenoma was 3 cm. in diameter, of carcinoid type, and was located in the lower lobe (Fig. 5b) although 
that is not evident in the roentgenogram. Neg. 84775. b. Case 23. Inflammatory infiltrate and abscess in the 
left lower lobe secondary to carcinoid type of adenoma situated in the lower lobe bronchus. Neg. 83464. 
tion showed the tumors usually covered by 
epithelium of either squamous or respiratory 
type or both with an underlying layer of con- 
nective tissue. (Figs. 7a and b.) Of the 23 
cases in which sections included sufficient bron- 
chial wall for interpretation, neoplastic cells 
had infiltrated peripherally in 15 (Fig. 14). 
In one case they were in the pulmonary alveoli 
(Figure 14c). Perivascular permeation of 
tumor cells was noted in at least 2 cases (Figs. 538 
The Military Surgeon—November, 1946 
Fig. 3. a. Case 10. Mass (carcinoid type of adenoma) and inflammatory infiltration in the lower lobe of 
the left lung-. Neg. 74472. h. Case 35. Mass in lower lobe of left lung (mixed-tumor type of adenoma). 
Neg. 92475. 
Fig. 4. a. Case 34. Polypoid adenoma (carcinoid type) at the bifurcation of the right middle and lower 
lobe bronchi, Neg. 92476. b. Case 38. Polypoid adenoma (mixed-tumor type) in the left main bronchus. 
Photographs show actual size. Neg. 91530. 
I4d and e), and in another the cells were 
found in blood and lymphatic vessels (Fig. 
15). There were two instances of metastasis 
to the regional lymph nodes (Figs. 16 and 
17), and once tumor cells were found in the 
capsule of a peribronchial node (Fig. 14b). 
Hepatic metastases were demonstrated in one 
case (Fig. 18). 
Bone was present in the tumor in 8 cases 
(Fig. 11), usually near or contiguous with 
bronchial cartilages. In 3 cases, however, it 
seemed to have arisen in the connective tissue Bronchial Adenomas 
539 
of the tumor. In 2 cases there were calcium 
deposits in the stroma, and in one a part of 
the subepithelial fibrous layer was calcified 
(Fig. yb). Where cartilage was found within 
a tumor it was mature, of hyaline type, and 
was so placed with respect to definite bronchial 
cartilages that its non-neoplastic nature could 
not be doubted. Bronchial glands were often 
separated by tumor, and occasionally nerves 
and smooth muscle bundles were isolated by it. 
Fat was in close association with tumor cells in 
2 cases (Fig. 13c), but did not appear to be 
inflammation was found in 18 of the lobectomy 
and pneumonectomy specimens. The changes 
were in the form of chronic bronchitis, bron- 
chiectasis, interstitial pneumonitis and pleu- 
ritis. Suppuration was conspicuous in only a few 
cases. 
Discussion 
The greater number of men in this 
series of bronchial adenomas must be related 
to the fact that the Army Institute of Path- 
ology draws most of its material from the 
Fig. 5. a. Case 26. Pedunculated adenoma (carcinoid type) blocking- the left main bronchus, although 
arising in the orifice of the upper lobe bronchus. Neg. 90523. b. Case 12. Well demarcated adenoma (carcinoid 
type) at hilus of the lower lobe of the right lung. Involved bronchus is not known. Photographs show actual 
size. Neg. 84270. 
neoplastic. Vascularity was a prominent fea- 
ture of several of the tumors, but was not a 
constant finding; necrosis was rare. 
The descriptions of the tumor cells and 
their characteristic arrangements were included 
in the accounts of the representative cases and 
need not be repeated. Emphasis should be 
placed on the fact that mitoses were not seen 
except in one case (Case 35). Special stains 
revealed nothing distinctive in cell morphology 
except in the one case in which silver-positive 
granules were demonstrated by the Fontana 
stain (Fig. 10b). Fat was not found in the 
tumor cells. 
Gross or microscopic evidence of chronic 
medical service of the Army. Therefore, the 
figures relating to sex incidence should be 
disregarded in favor of those derived from 
series reported from the civilian population, 
which range from 54 per cent,30 to 80 per 
cent in women,24 the average from a number 
of reports being 62 per cent.18 
Types: Although the clinical features of 
these neoplasms were remarkably uniform, the 
tumors could be readily separated into two his- 
tologic types on the basis of distinct differ- 
ences in cell pattern and structure. 
The larger group consisting of 34 cases 
has been called the carcinoid type of bronchial 
adenoma because the cells, their arrangement, 540 
The Military Surgeon—November, 1946 
and the growth characteristics of the tumor 
were similar to, and in several instances identi- 
cal with, appendiceal carcinoids. It is this 
type of tumor which is usually referred to as 
bronchial adenoma (Figs. 7, 8, 9, 10, 11). 
The other type, represented by 5 cases 
(Figs. 12 and 13), has been classified as 
it is retained because it suggests that the cells 
and cell patterns common to this type of bron- 
chial adenoma resemble those of some of the 
mixed tumors. 
Division of the bronchial adenomas into his- 
tologic types is not new. Hamperl11 classified 
the tumors as bronchial carcinoids and cylin- 
Fig. 6. a. Case 28. Sharply circumscribed bronchial adenoma (carcinoid type) in hilus of the upper 
lobe of the right lung. Neg. 84407. b. The same specimen at a different level. The tumor has filled the 
lumens of two bronchi adjacent to the main mass. Specimen is shown in its actual size. Neg-. 84407. 
bronchial adenoma of mixed-tumor type be- 
cause of its close resemblance to some of the 
mixed tumors of salivary, lacrimal, and other 
glands about the mouth and orbit, as well as 
to some epithelial tumors in the jaws. How- 
ever, in bronchial adenomas the stroma is not 
of the myxomatous, osteoid, or cartilaginous 
type usually associated with mixed tumors. 
To this extent the term is a misnomer, but 
dromas. LafF and Neubuerger21 recognized a 
carcinoid-like pattern and a cylindromatous 
one. The resemblance of this group of tumors 
as a whole to carcinoids has been noted by 
several observers.15’28’31’32 The term bronchial 
carcinoid has been rejected by most authors 
because silver-positive granules found in most 
appendiceal carcinoids could not be demon- 
strated in the cells of the bronchial tumors. Bronchial Adenomas 
541 
Fig. 7. Case 12. Carcinoid type of bronchial adenoma, a. Tumor separated from the bronchial lumen by 
connective tissue and thinned epithelium. Normal respiratory epithelium lies above the bronchial lumen. 
X-X30. Neg. 90987. b. Calcified connective tissue layer between the tumor and the thinned surface epithelium. 
X-130. Neg. 90984. c. Typical carcinoid pattern. X-130. Neg. 90985. d. Two pseudoglands with central 
spaces containing retracted connective tissue. X-6oo. Neg. 90986. Refer to Figs, za, 5b and xia for addi- 
tional photographs in this case. 
It will be recalled that such granules were 
found in the biopsy tissue from one tumor of 
this series (Fig. lob). 
The carcinoid type of bronchial adenoma 
also resembles certain other tumors. For ex- 
ample, a striking histologic likeness between it 
and the islet cell tumors of the pancreas was 
observed by Moore,33 who studied some of the 
cases in this series, as well as by other investi- 
gators.21’28 These observations are noteworthy 
because Boyd34 referred to a similarity between 
pancreatic islet cell tumors and carcinoids of 
the appendix. Moersch, Tinney and McDon- 
ald28 pointed out the close resemblance be- 
tween bronchial adenomas and basal cell carci- 
nomas of the skin and adenomas of the para- 
thyroid gland. 
Womack and Graham12 regarded all bron- 542 
The Military Surgeon—November, 1946 
Fig. 8. Case 25. Carcinoid type of bronchial adenoma, a. Poorly defined cords and a few glands composed 
of tumor cells, the only case of this type in the present study showing distinct gland formation. X-130. Neg. 
91013. b. Glands composed of tumor cells. X-600. Neg. 91015. 
chial adenomas as mixed tumors without dif- 
ferentiation into types. They based their inter- 
pretation upon the presence of fat, cartilage, 
bone and smooth muscle in some of the tumors 
which they studied. Bone, noted rarely by most 
observers, was interpreted by Mallory15 as 
secondary to the tumor and not actually a part 
of it. Moersch, Tinney and McDonald28 be- 
lieved that the bone resulted from ossification 
of bronchial cartilages. Brunn and Goldman17 
were of the opinion that both bone and carti- 
lage probably arose from dedifferentiation of 
adult tumor cells. Although bone was found 
in several of the carcinoid-type tumors in the 
Fig. 9. Case 7. Carcinoid type of bronchial adenoma, a. Tumor arranged in distinct ribbons throughout, 
characteristic of certain carcinoids. Spaces are shrinkage artefacts. X-200. Neg. 92332. b. Ribbons of 
elongated tumor cells with long axes directed transversely; delicate stroma contains blood capillaries. X-600. 
Neg. 92331. Bronchial Adenomas 
543 
present study, it did not give the impression 
of being neoplastic; neither did fat, cartilage, 
smooth muscle, nerves, and blood vessels which 
appeared to have been incorporated in the 
tumors as a result of their expansile and in- 
filtrative properties. Clerf and Bucher30 ap- 
parently accepted bronchial adenomas as 
mixed tumors and described three types based 
on differences in the pattern and appearance 
of the cells; one of their types corresponded to 
some of the adenomas were like carcinoids and 
others like cylindromas, he believed that they 
were essentially the same. McDonald, Moersch 
and Tinney29 regarded the cylindroma as a 
form of bronchial tumor apart from the ade- 
noma, and compared it with what they called 
cylindroma of salivary glands and of skin. 
The close resemblance of the mixed-tumor 
type of bronchial adenoma to some of the 
mixed tumors of salivary glands and to basal 
Fig. io. Case 10. Carcinoid type of bronchial adenoma, a. Closely packed cords of tumor cells about a 
small irregular space, not gland-like, from lobectomy specimen. X-600. Neg. 92312. b. Silver-positive 
granules (Fontana stain) within the cytoplasm of the tumor cells (from biopsy specimen). X-130. Neg. 
92311. 
that which others have referred to as cylin- 
droma. Foster-Carter18 concluded that bron- 
chial adenomas were identical with benign 
salivary gland tumors. 
The similarity of the mixed-tumor type of 
bronchial adenomas in the present series to 
epithelioma adenoides cysticum, to cylindroma, 
and in one case to the ordinary type of basal- 
cell carcinoma of the skin was striking. Similar 
observations have been made in other series. 
Geipel7 called the bronchial adenomas which 
he studied basal cell carcinomas, and Hamperl11 
used the term cylindroma for the mixed-tumor 
type. Stout20 recognized a bronchial mixed 
tumor in addition to the typical adenoma. Al- 
though Anderson19 made the observation that 
cell carcinomas of the skin is interesting, be- 
cause Krompecher, cited by Ewing,35 and Ew- 
ing himself noted the similarity between some 
of the mixed tumors of salivary glands and 
basal cell carcinomas of the skin, particularly 
epithelioma adenoides cysticum. 
It is evident from the descriptions and illus- 
trations appearing in the various papers that 
the 5 tumors classified in the present series 
as mixed-tumor type of bronchial adenoma are 
the same as bronchial tumors referred to by 
others as cylindromas or as simulating cylin- 
dromas. Which term is to be used, and 
whether the tumors are to be regarded as a 
type of bronchial adenoma or as an inde- 
pendent neoplasm is a matter for personal de- The Military Surgeon—November, 1946 
Fig. ii. Bone in carcinoid type of adenomas, a. Case 12. Ossification in the periphery of and about a 
bronchial cartilage. Note tumor in osseous spaces and between spicules near cartilage. X-50. Neg. 90983. 
b. Case 16. Bone spicules in tumor. Note well developed portion of bone in upper corner on the right. X-130. 
Neg. 91011. c. Case 9. Osseous stroma separating sheets of tumor cells. These cells are like those of 
cases showing a more distinct carcinoid pattern. X-200. Neg. 92325. 
termination. This tumor comprised 12.8 per 
cent of the adenomas of this series and 13.6 
per cent of the combined bronchial adenomas 
and cylindromas as classified by McDonald, 
Moersch and Tinney. 
The relationships of bronchial adenomas of 
both types, carcinoids, pancreatic islet cell 
tumors, different varieties of basal cell carci- 
noma of the skin, and certain tumors of the 
salivary glands and other structures about the 
mouth and face remain to be explained. 
Origin: The origin of bronchial adenomas 
although often discussed is still undetermined. 
If they are primarily epithelial tumors, as they 
are believed to be, they should arise from the 
bronchial surface, from the bronchial ducts, 
or from the mucous and serous glands. How- 
ever, most writers agree that origin from the Bronchial Adenomas 
545 
Fig. 12. Case 35. Mixed-tumor type of bronchial adenoma, a. Nests of tumor cells simulating- basal cell 
carcinoma. Note whorl of squamous cells. X-200. Neg. 92322. b. Higher magnification of area similar 
to a. X-600. Neg. 92320. c. More solid portion of tumor, showing pattern of some salivary gland 
mixed tumors. Note close resemblance to epithelioma adenoides cysticum. X-200. Neg. 92402. d. Area 
from portion similar to c. Contrast these cells and their arrangement, also those in b, with those of the 
carcinoid type of adenoma. X-6oo. Neg. 93298. 
surface epithelium is unlikely because the tu- 
mors are separated from it by a connective 
tissue layer. The most frequent suggestion is 
that they arise from the bronchial glands or 
ducts. Reisner,6 Kramer1 and Wessler and 
Rabin2 believed that they arose from bronchial 
ducts, Fried36 and others9’18’37 that they or- 
iginated in the bronchial glands. Brunn and 
Goldman17 suggested that they might arise 
from both ducts and glands, and that the 
fact that the glands are normally of the mixed 
type might account for the variations in pat- 
tern encountered among the tumors. Stout20 
presented evidence in support of Hamperl11 
favoring a special cell, the oncocyte, as the 
probable source. The cell is found, according 546 
The Military Surgeon—November, 1946 
Fig. 13. Case 37. Mixed tumor type of bronchial adenoma, a and b. Tubular portion of tumor of 
adenomatoid pattern. Note homogeneous material in several lumens in a. X-X30. Neg. 91023, 91024. c and 
d. From another portion of the same tumor showing nests of cells adjacent to an artery and in adipose tissue 
in the outer part of the bronchial wall. Observe similarity to mixed tumor and to epithelioma adenoides 
cysticum. X-130, X-600, Neg. 91022,93304. 
to Stout, in the bronchial glands and ducts of 
adult human beings. Similar cells were demon- 
strated in some of the bronchial adenomas 
studied by Stout and in others examined by 
Hamperl. 
Churchill38 perceived a strong similarity be- 
tween bronchial adenomas and the fetal lung 
and suggested that these tumors might be vesti- 
gial pulmonary lobes growing endobronchially. 
Womack and Graham12 likewise regarded 
their structure as like that of fetal lung and 
attributed their origin to undeveloped bron- 
chial buds. In 2 or 3 biopsy specimens in this 
series there was some likeness to the fetal 
pattern, but it may well have been artefact, 
since it never was seen in sections of tumors 
removed by lobectomy or pneumonectomy. 
Harris and Schattenberg14 regarded bronchial Bronchial Adenomas 
547 
Fig. 14. Local tumor cell infiltration in bronchial adenomas, a. Case 34. Infiltration into connective tissue 
just outside margin of tumor (carcinoid type). X-200. Neg. 92336. b. Case 12. Infiltration into capsule of 
peribronchial lymph node by direct or lymphatic extension (carcinoid type). X-130. Neg. 9x010. c. Case 
35. Direct extension into pulmonary alveoli (mixed-tumor type). X-80. Neg. 92321. d. Case 12. Extension 
adjacent to a small blood vessel (carcinoid type). X-235. Neg. 92326. e. Case 34. Similar perivascular exten- 
sion (carcinoid type). X-300. Neg. 93303. 
adenomas as probably developmental, and Har- 
ris39 stated that they consisted in part of cells 
resembling lymphocytes. Because of these cells 
and the frequent vascularity of the tumors, 
Harris suggested that they might be referred 
to as adenomas of “secretory type” and 
“lymphoid or angiolymphoid type.” 
The origin of none of the tumors of this 
present series could be established on the basis 
of cell morphology or pattern. Cells compara- 
ble to the oncocytes described by Hamperl and 
by Stout were not found. The carcinoid type 
of adenoma did not resemble any structure 
found in normal bronchi. Rarely, they con- 
tained a few gland-like structures, which, how- 
ever did not simulate bronchial glands. Only by 
their position and apparent origin within the 
bronchial wall could one speculate that they 
might have begun in bronchial glands or ducts. 
The origin of the mixed-tumor type would 548 
The Military Surgeon—November, 1946 
Fig. 15. Case 23. Vascular invasion by bronchial adenoma (carcinoid type), a. Tumor cells in blood 
or lymphatic vessels; tumor proper is at right, alveolar spaces at left. X-130. Neg. 90989. h. Tumor cells 
in a lymphatic channel. X-435. Neg. 90988. c. Tumor cells in a blood vessel. X-600. Neg. 92327. 
seem to be related to bronchial ducts or glands, 
though the material noted within spaces re- 
tained no specific stain to indicate that it was of 
gland-cell origin. In case 37 there was a re- 
semblance between some of the neoplastic 
tubular structures and the normal ducts, but 
this does not prove that the ducts were the 
source. Stout20 attributes the origin of what he 
calls mixed tumor of the bronchus to the 
bronchial glands. 
Malignant Potentialities: The status of 
bronchial adenomas with regard to malignancy 
is a most important problem. Therapy has not 
been uniform because of lack of agreement Bronchial Adenomas 
549 
Fig. i6. Case 5. Metastasis of bronchial adenoma (carcinoid type), a and b. Pattern and cellular con- 
stituents of the primary bronchial tumor. X-130, X-600. Neg. 90976, 93294. c. Metastatic tumor tissue dis- 
placing and replacing lymphoid tissue and lying within capsule of the node. X-25. Neg. 92304. d and e. 
Portions of metastasis as in c, showing pattern and cell type, X-200, X-600. Neg. 93297, 93296. (Note the 
remarkable similarity in the pattern and cells of the metastasis and those of the primary tumor.) 
regarding their potentialities. Foster-Carter18 
stated that metastasis was unknown but that 
local infiltration might occur. Clerf and 
Bucher30 concluded that they were probably 
benign and non-metastasizing. Only one of 
the 35 cases which they studied presented evi- 
dence of malignancy, and this change did not 
appear until 5 years after discovery of the 
original adenoma. Jackson, Konzelmann and 
Norris24 stated that the adenomas do not 
metastasize and expressed doubt that they ever 
become malignant. 
Wessler and Rabin2 presented 2 cases as 
evidence that bronchial adenomas may become 
malignant. More recently there have been 
several unequivocal reports of extension beyond 550 
The Military Surgeon—November, 1946 
Fig. 17. Case 15. Bronchial adenoma (carcinoid type) with metastasis to lymph node, a and b. From 
bronchial biopsy specimen removed in 1937. X-200, X-600. Neg. 93292, 91006. c and d. From bronchial 
biopsy specimen removed in 1941. X-200, X-600. Neg, 92318, 93293. e and /. Interlacing cords of metas- 
tatic tumor cells in lymph node. (Autopsy-i943.) X-130, X-600. Note similarity of pattern and cells to 
bronchoscopic specimen of 1941 and to the cells of the 1937 specimen. Neg. 91007, 92329. 
the borders of the original tumors. Castleman40 
described a case with metastasis to a regional 
lymph node. Adams, Steiner and Bloch13 pre- 
sented 5 cases as malignant adenomas, because 
there was infiltration of the bronchial wall in 
all, invasion of local lymphatics in 1, metas- 
tasis to the peribronchial lymph nodes in 2, 
and distant metastases in 2: one in the liver, 
the other in a vertebral body. The authors 
did not make any distinction as to pattern types. Bronchial Adenomas 
551 
Fig. i 8. Case 31. Metastasis of a bronchial adenoma to liver, a and b. Typical benign-appearing carcinoid 
type of pattern and cells from the primary bronchial tumor. X-200, X-600. Neg. 93293, 93291. c and d. 
Hepatic metastasis; pattern less distinct than in primary tumor, though cells are similar and carcinoid-like. 
X-200, X-600. Neg. 93288, 93289. 
In Anderson’s case19 with metastasis to the 
liver, the tumor in both the bronchus and the 
liver simulated cylindroma. Laff and Neu- 
buerger21 reported an adenoma with cylin- 
dromatous features, which had metastisized 
extensively to the opposite lung. Moersch, 
Tinney and McDonald28 described invasion 
of the bronchial wall in 13 of 15 of their 
cases, and found tumor tissue in a regional 
lymph node in 2. Extension into lymphatic and 
blood vessels with distant metastasis has been 
described by Lentino.41 
In the present study there were numerous 
examples of local infiltration of tumor cells. 
Lymph node metastasis was found in 2 cases, 
extension into a lymph node capsule in I 
case, vascular involvement in I, and distant 
metastases (liver) in I case. The author has 
recently studied sections from an autopsy per- 
formed at the University of Illinois College 552 
The Military Surgeon—November, 1946 
of Medicine showing a bronchial tumor of long 
standing which had metastasized to a kidney. 
The metastasis reproduced the cell type and 
pattern of the primary lesion accurately and 
was like the mixed-tumor type of bronchial 
adenoma described in this series. In none of 
these tumors with metastasis or local infiltra- 
tion has there been any essential difference be- 
tween the pattern and structure of the cells 
constituting the primary tumor and of those 
which have infiltrated or metastasized. The 
tumor cells have shown little or no anaplasia 
and even in metastases have appeared benign 
morphologically. 
Laff and Neubuerger21 have raised a ques- 
tion as to whether the bronchial adenoma of 
mixed-tumor type (referred to by them as the 
cylindromatous form) may not be more likely 
to metastasize. McDonald, Moersch and Tin- 
ney29 have the impression that the tumors 
which they call bronchial cylindromas are more 
aggressive than the usual adenoma. In the 
cases of Clerf and Bucher30 the clinical be- 
havior of this type was considered to be the 
same as that of the other adenomas. The 
present series of cases does not help particularly 
in answering this question because tumors of 
both types infiltrated locally; three of the 
carcinoid type metastasized; one of the mixed- 
tumor type invaded alveoli; and another 
showed evidence of recurrence after lobec- 
tomy. 
In the light of available information it is 
necessary to state that some bronchial adenomas 
become malignant. The most convincing evi- 
dence is that in doing so they retain their 
original pattern, and apparently continue their 
usual slow rate of growth, as is the rule with 
metastasizing carcinoids of the appendix, 
mixed tumors of the salivary glands, and basal 
cell carcinomas of the skin. Rarely, if ever, 
has death occurred because a previously recog- 
nized typical bronchial adenoma became malig- 
nant through accelerated rate of growth and 
capacity to metastasize widely as do the usual 
bronchogenic carcinomas. That they may com- 
pletely lose their original features and follow 
the course of the usual bronchogenic carci- 
nomas, as Graham and Womack25 believe the 
majority of them do, is difficult to prove or 
disprove, but remains a possibility. One 
bronchogenic carcinoma studied at the Army 
Institute of Pathology suggested such transi- 
tion, but the material available did not per- 
mit examination adequate for conclusions. 
Diagnosis: The diagnosis of the bronchial 
adenomas must depend finally upon tissue 
study, although the age at which they occur, 
the higher incidence in women, hemoptysis, 
cough, expectoration, recurrent acute pulmon- 
ary infection, and the roentgenologic features 
often give definite leads. The ability of bron- 
chologists to identify these tumors in situ has 
contributed to their recognition. One must 
keep in mind that they have often been diag- 
nosed as bronchogenic carcinomas, from 
which they should be distinguished even though 
some of them are malignant in the strict sense 
of the term. This differentiation is difficult at 
times because the cells of the adenomas may 
be elongated and their nuclei spindle-shaped 
and deeply stained, due to the distortion en- 
countered in some biopsy specimens. This often 
gives the impression of extensive infiltration 
and anaplasia. In the past, patients with bron- 
chial adenomas have been treated for months 
or even years for presumed tuberculosis. Bron- 
chiectasis, atelectasis, and empyema have some- 
times existed for long periods under the ob- 
servation of a physician, the true cause of ill- 
ness being overlooked. Bronchoscopic investiga- 
tion is essential in the presence of any unex- 
plained chronic pulmonary symptom or sign. 
Treatment: The methods of therapy for 
bronchial adenomas may be found by referring 
to the papers of recent years. The trend seems 
to be away from bronchoscopic removal, X- 
ray, or local radium ray therapy, toward ex- 
tirpation by lobectomy and pneumonectomy. 
Several factors are involved in this trend: the 
diminishing mortality rate from the more ex- 
tensive thoracic operations; the size of the 
tumor and its peripheral cellular infiltration 
which may prevent endobronchial eradication; 
possible metastasis, and finally, associated bron- 
chiectasis which of itself is justification for 
resection of a lobe or a lung. 
The impression gained from the cases pre- Bronchial Adenomas 
553 
sented is that all bronchial adenomas should 
be removed by lobectomy or pneumonectomy 
as soon after discovery as is feasible, notwith- 
standing their apparently restricted malignant 
potentialities. A possible exception is the occa- 
sional small, pedunculated tumor without evi- 
dence of extrabronchial extension or bronchi- 
ectasis. Even so, there remains the possibility 
of recurrence and chronic pulmonary infec- 
tion with tenacious adhesion of the lung to 
the chest wall, which may make subsequent 
lobectomy or pneumonectomy both difficult 
and dangerous. 
Lobectomy is considered preferable to pneu- 
monectomy if the remainder of the affected 
lung appears healthy and the tumor is so 
located that there is a reasonable chance of its 
extirpation. The procedure is less hazardous, 
and the patient is left a larger amount of func- 
tioning pulmonary tissue. Pneumonectomy is 
advocated if the tumor arises in one of the 
main (stem) bronchi or if it is so large that 
it has obstructed bronchi to all lobes with con- 
sequent chronic infection and loss of function 
of the entire lung. Evaluation of transthoracic 
bronchotomy for the removal of some of the 
smaller tumors must await further trial. 
4. The origin of neither of the types was 
determined, though their intramural location 
in the bronchus suggested bronchial glands 
or ducts as sources. 
5. Several of the tumors exhibited features 
of malignancy. Some had infiltrated locally; 
one had filled several adjacent pulmonary 
alveoli; one had extended into a lymph node 
capsule; two had metastasized to a hilus lymph 
node; one had invaded small vessels, and one 
had metastasized to the liver. 
6. This study indicates that lobectomy or 
pneumonectomy, according to the location of 
the tumor, is the preferred treatment because 
many of the tumors are of such size or have 
infiltrated so that they cannot be entirely re- 
moved through the bronchoscope; because of 
the malignant potentialities of some of the 
tumors; and because of the frequent coexist- 
ence of bronchiectasis and chronic pneumonitis. 
The writer expresses his sincere apprecia- 
tion to the clinicians and pathologists both in 
civilian and military service who submitted 
the cases presented and who secured follow-up 
information wherever possible. He is also grate- 
ful for the consultation photography and other 
help supplied by the professional and technical 
staff of the Army Institute of Pathology. 
Summary 
r. In 38 patients with a bronchial adenoma 
the predominant symptoms and signs were 
cough, hemoptysis, recurrent acute respiratory 
infection, and expectoration. Roentgenographic 
features were those of atelectasis, bronchi- 
ectasis, pulmonary inflammation, bronchial 
blockage, and presence of a mass. All of these 
signs and symptoms were not present in the 
same patient. The tumor was found incidental- 
ly at autopsy in one additional case. 
2. The tumors were located at or near the 
pulmonary hilus, were usually polypoid, and 
in several cases attained rather large size. 
5. They assumed two distinct patterns, one 
which simulated that of appendiceal carcinoid 
and the other that of some of the mixed tumors 
of salivary glands and other structures of the 
mouth and face. They have been designated 
carcinoid type and mixed-tumor type of bron- 
chial adenoma respectively. 
References 
1 Kramer, R.; Adenoma of bronchus, Ann. Otol., 
Rhin. & Laryng. 39:689-695, 1930. 
' Wessler, H., and Rabin, C. B.: Benign tumors of 
the bronchus, Am. J. M. Sc. 183:164-180, 1932. 
3 Overholt, R. H.; Pneumonectomy for malignant 
and suppurative disease of the lung, J. Thoracic 
Surg. 9:17-61, 1939-40. 
4 Schwartz, L.: Adenoma of the trachea and 
bronchus, Arch. Otolaryng. 39:231-242, 1944. 
“Tyson, M. D., and Milliken, N. T.: Total 
pneumonectomy for benign bronchial adenoma, Am. 
J. Surg. 67:111-118, 1945. 
s Reisner, D.: Intrabronchial polypoid adenoma, 
Arch. Surg. 16:1201-1213, 1928. 
7 Geipel, P.; Zur Kenntnis der gutartigen Bron- 
chialtumoren, Frankfurt. Ztschr. f. Path. 42:516-544, 
1931. 
8 Rosenblum, P., and Klein, R. I.: Adenomatous 
polyp of the right main bronchus producing atelec- 
tasis, J. Pediat. 7:791-796, 1935. 
9 Clerf, L. H., and Crawford, B. L.: Benign 
glandular tumors of the bronchus, Tr. Coll. Phys., 
Philadelphia, 4th Series, 4:6-8, 1936. 554 
The Military Surgeon—November, 1946 
“Zamora, A. M., and Schuster, N.: Vascular 
adenoma of the bronchus, J. Laryng. & Otol., 52: 
337-343> *937- 
11 Hamperl, H.: Uber gutartige Bronchialtumoren 
(Cylindrome und Carcinoide), Virchow’s Arch. f. 
path. Anat. 300:46-88, 1937. 
12 Womack, N. A., and Graham, E. A.: Mixed 
tumors of the lung, so-called bronchial or pulmonary 
adenomas, Arch. Path. 26:165-206, 1938. 
13 Adams, W. E., Steiner, P. E., and Block, R. G.: 
Malignant adenoma of the lung, carcinoma-like 
tumors with long clinical course, Surgery 11:503- 
526, 1942. 
14 Harris, W. H., and Schattenberg, H. J.; Anlagen 
and rest tumors of the lung inclusive of “mixed 
tumors” (Womack and Graham). Am. J. Path. 18: 
955-967) 1942. 
16 Mallory, T. B.; Case records of the Massachu- 
setts General Hospital; Case 27511, New England J. 
Med. 225:983-986, 1941. 
16 Goldman, A., and Stephens, H. B.: Polypoid 
bronchial tumors with special reference to bron- 
chial adenomas, J. Thoracic Surg. 10:327-353, 
1941. 
17 Brunn, H., and Goldman, A.: Bronchial ade- 
noma, Am. J. Surg. 54:179-192, 1941. 
8 Foster-Carter, A. F.: Bronchial adenoma, Quart. 
J. Med. 10:139-174, 1941. 
19 Anderson, W. M.: Bronchial adenoma with 
metastasis to the liver, J. Thoracic Surg. 12:351-360, 
1942-43. 
20 Stout, A. P.: Cellular origin of bronchial ade- 
noma, Arch. Path. 35:803-807, 1943. 
21 Laff, H. I., and Neubuerger, K. T.: Bronchial 
adenoma with metastasis, Arch. Otolaryng. 40:487- 
493) 1944- 
22 Riordan, D. C., and Richards, V.: Bronchial 
adenoma, report of an unusual case: review of the 
literature, Stanford Med. Bull. 2:63-71, 1944. 
23 Nager, R.: fiber das sagenannte Bronchialade- 
noma, Praxis 33:84, 1944. 
24 Jackson, C. L., Konzelmann, F. W., and Norris, 
C. M.: Bronchial adenoma, J. Thoracic Surg. 14: 
98-105, 1945. 
25 Graham, E. A., and Womack, N. A.: The prob- 
lem of the so-called bronchial adenoma, J. Thoracic 
Surg. 14:106-119, 1945. 
* Chamberlain, J. M., and Gordon, J.: Bronchial 
adenoma treated by pulmonary resection, J. Thoracic 
Surg. 14:144-159, 1945. 
27 Harris, H. E.: Adenoma of the bronchus, Cleve- 
land Clin. Quart. 12:73-80, 1945. 
28 Moersch, H. J., Tinney, W. S., and McDonald, 
J. R.: Adenoma of the bronchus, Surg. Gynec. & Obst. 
81:551-558, 1945. 
29 McDonald, J. R., Moersch, H. J., and Tinney, 
W. S.: Cylindroma of the bronchus, J. Thoracic Surg. 
14:445-453, 1945. 
30 Clerf, L. H., and Bucher, C. J.: Adenoma (mixed 
tumor) of bronchus: a study of 35 cases, Ann. Otol., 
Rhin. & Laryng. 51:836-850, 1942. 
81 Kernan, J. D.: Treatment of a series of cases of 
so-called carcinoid tumors of the bronchi by dia- 
thermy, report of 10 cases, Ann. Otol. Rhin, & 
Laryng. 44:1167-1191, 1935. 
32 Jackson, C. L., and Konzelmann, F. W.: 
Bronchoscopic aspects of bronchial tumors with 
special reference to so-called bronchial adenoma, J. 
Thoracic Surg. 6:312-329, 1936-37. 
33 Moore, R. A.: Personal communication. 
34 Boyd, W.: Surgical Pathology, ed. 4, Philadel- 
phia, W. B. Saunders Co., p. 341, 1938. 
35 Ewing, J.: Neoplastic Diseases, ed. 4, Philadel- 
phia, W. B. Saunders Co., pp. 790-791, 1940. 
36 Fried, B. M.: Adenoma of bronchial mucous 
glands, Arch. Otolaryng. 20:375-381, 1934. 
37 Brock, R. C.: Pulmonary new-growths, Lancet 
2:1041-1044, 1938. 
38 Churchill, E. D.: Discussion of symposium on 
surgery of the lungs and mediastinum, J. Thoracic 
Surg. 6:335, I936-37- 
39 Harris, W. H.: Histologic analogy of bronchial 
adenoma to late prenatal and early postnatal struc- 
tures, Arch. Path. 35:85-92, 1943. 
110 Castleman, B.: Case records of the Massachusetts 
General Hospital; Case 26171, New England J. Med. 
222:721-724, 1940. 
1 Lentino, A. S.: Tumores mixtos del pulmon (ad- 
enomas bronquiales), Bal. Inst.-CHn. Quir. Buenos 
Aires 21:113-122, 1945. THE PATHOLOGY OF ACUTE FALCIPARUM MALARIA* 
SOPHIE SPITZ, M.D., Contract Surgeon 
(With thirty-two illustrations) 
FALCIPARUM malaria obviously will 
not reach the epidemic proportions in 
the United States that were anticipated 
during World War II chiefly because the sup- 
pressive atabrine therapy used by the Armed 
Forces in all endemic areas has proved of 
greater curative value against Plasmodium 
falciparum than against other malarial para- 
sites. However, this form of the disease is 
still sufficiently common in this country as to 
be of concern; indeed, as the reports by Hel- 
pern1 of epidemics among drug addicts indi- 
cate, falciparum malaria was, on occasion, a 
distinct problem even before the war. Although 
the pathologic anatomy of malaria has been 
previously studied in a most adequate fashion, 
it appears likely that, in the light of current 
investigation of the clinical, physiological and 
immunological aspects of the disease, there 
may be some gain to be derived from a review 
and a re-evaluation of the histologic findings. 
Material 
More than one hundred cases of fatal falci- 
parum malaria were studied at the Army Insti- 
tute of Pathology during World War II. This 
study is based on the first fifty consecutive 
cases in which an autopsy protocol as well as 
sections, at least, of the principal organs were 
available for examination. In many of these 
cases, there was little more than a brief clini- 
cal note; this fact was not considered in select- 
ing the cases since death often ensued relatively 
suddenly without previous illness and without 
sufficient time for clinical information to be- 
come available. However, further selection 
was made in that cases known to have had ma- 
laria previous to the fatal illness, to have been 
treated for malaria, or to have had any other 
chronic illness were not used for this study. 
This series is comprised of soldiers from all 
Theaters of Operation who, presumably, so 
far as is known from the available information, 
died in their first attack of falciparum malaria. 
The duration of illness was from one to four- 
teen days. These men were from 18 to 40 
years of age; all but three were white; one 
was Mongolian (Chinese) and two were Ne- 
gro (American). All but four cases had re- 
ceived either atabrine or quinine, usually only 
one to two doses administered at the time they 
were moribund; in the four cases not receiv- 
ing therapy the cause of death was determined 
at the time of autopsy. 
It is important to note that all but five 
(90 per cent) of the fifty patients were ad- 
mitted to the hospital either in coma or with 
convulsions; three were admitted with diar- 
rhea, and, of these, two were “mentally con- 
fused”; one was admitted with jaundice and 
diarrhea; and one had anuria after an illness 
of eight days. 
Paraffin sections of all available organs 
were studied microscopically. The major por- 
tion of this study was based on the examina- 
tion of hematoxylin-eosin stained paraffin sec- 
tions although many of the tissues were also 
stained with Giemsa stain and by the method 
of Tomlinson and Grocott,2 at times after 
removal of the parasite pigment with saturated 
picric-acid alcohol. Serial sections of the brain 
were made in several cases. Random cases 
were used for the study of the distribution of 
pigment (malarial and iron) by staining with 
potassium ferrocyanide and hydrochloric acid. 
Polaroid film was used to determine the re- 
fractility of pigments. 
Findings 
Brain: Multiple sections, from five to 
twenty in each, were available for study in all 
cases. In 46 cases (92 per cent) there was 
engorgement of the cerebral capillaries. The 
entire capillary network of the brain in these 
cases was distended with erythrocytes (Fig. 3). 
In 41 of these 46 cases, there were large num- 
bers of parasitized erythrocytes in the dis- 
tended vessels whereas in the remaining five 
* Army Institute of Pathology, Washington, D.C. 556 
The Military Surgeon—November, 1946 
cases the erythrocytes were not parasitized. 
Plasmodium falcifarum in all stages of de- 
velopment was present, although in the brain 
there was a predominance of young schizonts. 
The distended vessels were in all cases more 
prominent in the gray matter than in the 
white and they were evenly distributed in all 
locations. Pigmented parasites were found not 
only within erythrocytes but also were seen 
free in the lumen particularly of the larger 
vessels rather than in those of capillary size; 
the arteries were seldom distended. The dis- 
tribution of parasitized cells in the capillaries 
was even and uniform and each vessel ap- 
peared to be equally distended and often oc- 
cluded or “plugged” by the parasitized cells 
(Fig. 2 and 5). In the larger vessels, however, 
parasitized erythrocytes and parasites tended 
to hug the wall of the vessel to produce a 
“marginated” effect and often to form clumps 
at the periphery of the lumen (Fig. 6). The 
endothelial cells lining vessels of all sizes occa- 
sionally appeared hyperchromatic or swollen 
(Fig. 7) but no other reactive changes were 
noted. It was not possible to demonstrate that 
the endothelial cells possessed any phagocytic 
function either for pigment or for parasites, 
and only occasionally a rare pigmented phago- 
cyte could be seen in a lumen. 
In four cases no engorgement of cerebral 
vessels was noted nor was any remarkable num- 
ber of parasites found in the brain or in other 
organs of these cases. One of the four patients 
died immediately following reaction to trans- 
fusion; another died suddenly and was found 
to have a pheochromocytoma of the adrenal; 
the third died after the administration of one 
dose of quinine, and the fourth developed 
anuria and hemogloblinuric nephrosis. The 41 
cases in which the cerebral vessels were en- 
gorged with large numbers of parasitized 
erythrocytes showed a similar degree of para- 
sitization of other viscera, such as heart, lungs 
and intestinal tract; similarly, in those cases 
in which parasites were not present in the 
brain, parasites were not demonstrated in 
other viscera, except, of course, in the spleen 
and liver where parasites and pigment always 
were present. 
In fifteen (30 per cent) of the fifty cases 
multiple hemorrhages were present in the 
brain (Fig. 1); these lesions were associated 
in all cases with engorgement of the cerebral 
capillaries and in only one instance of hemor- 
rhage were parasites not demonstrated in the 
brain. Clinically all the patients of this group 
were in coma on admission to the hospital and 
two had had convulsions. The distribution of 
the hemorrhages was identical in all instances 
although there was numerical variation; in 
the majority (12 cases) there were many 
hemorrhages; in two cases there were only a 
moderate number and in one case only two 
such lesions could be found in several sections 
studied. In all instances, these hemorrhages 
were in the subcortical white matter and not 
in the gray matter; they were also seen com- 
monly in the pons, medulla and cerebellum. 
The spinal cord was not available for study 
in this series. The hemorrhages were identical 
in appearance in all locations and took the 
form of “ring hemorrhages” (Fig. 8) in 
which could usually be discerned a central 
vessel, “plugged” with erythrocytes which in 
all but one case were parasitized, or actually 
thrombosed. The tissue surrounding the cen- 
tral vessel was necrotic and there was de- 
myelinization of fibers (Fig. 11). The hemor- 
rhage formed a round or oval zone im- 
mediately outside the central necrotic zone and 
was always separated from the central vessel 
by the distance of the zone of necrosis. It was 
not possible to demonstrate in any case that 
the origin of the hemorrhage was from the 
central vessel of the ring. Whereas in all but 
one case the number of parasitized erythrocytes 
in the brain was great and the lumens of the 
central vessels of the ring hemorrhages con- 
tained parasites, in no instance did the erythro- 
cytes of the hemorrhage harbor plasmodium. 
In addition to simple ring hemorrhages, 
twelve of the fifteen cases with these lesions 
showed also Durck’s nodules,3 or the so-called 
malarial granulomas (Fig. 10). Every in- 
stance of “granuloma” in this series was as- 
sociated with ring hemorrhages; in fact, the 
“granuloma” appeared actually to be an al- 
tered hemorrhage. These lesions consisted sim- Fig. 
i 
Fig. 
2 
Figure i. Sagittal section of brain of patient whose death was due to acute falciparum malaria. 
Cortex is grayer than normal and the subcortical white matter is studded with petechial hemor- 
rhages. The so-called malarial granulomas are associated with these hemorrhages. 
Figure 2. Capillaries of the brain are practically filled with parasites and parasitized erythro- 
cytes; in the vessels of somewhat larger caliber there is “margination” of the parasites. X700. Fig 
22 
Fig. 
23 
Fig. 
24 
Fig. 
25 
Figure 22. Spleen in acute falciparum malaria; pigment mostly as single grains. X700. 
Figure 23. Spleen in chronic vivax malaria; pigment is coalesced and present intracellularly and 
extracellularly. Cords are fibrotic and sinuses are distended. X700 
Figure 24. Kidney in acute falciparum malaria; pigmented phagocytes are trapped in the glomer- 
ular capillaries. Increased cellularity of the glomerulus. X350. 
Figure 25. Hemoglobinuric nephrosis of acute falciparum malaria. Hemoglobin casts in distal 
convoluted tubules; necrosis and regeneration of tubular epithelium. X230. Pathology of Acute Falciparum Malaria 
557 
Fig. 3. Engorgement of cortical capillaries in acute falciparum malaria. AIP Neg. 89177 X 230 
Fig. 4. Engorgement of cortical capillaries occurring in high altitude anoxia. AIP Neg. 89175 X 230 
Fig. 5. “Plugging” of cortical capillaries by parasitized erythrocytes. AIP Neg. 77034 X 2000 
Fig. 6. Margination of parasitized erythrocytes in larger vessels. AIP Neg. 77351 X 605 
Fig. 7. Thrombosis of cerebral vessel associated with hemorrhage in other parts of the brain. Note the 
“plugging” of the vessel adjacent to the thrombosis. AIP Neg. 93672 X 650 558 
The Military Surgeon—.November} 1946 
Figs. 8-13 Pathology of Acute Falciparum Malaria 
559 
Fig. 14. Myocardium shows engorgement of capillaries equal to that seen in the brain. Interstitial infiltrate 
is composed of mononuclear cells. AIP Neg. 93669 X 650 
Fig. 15. Endocardial infiltrate and verrucae in acute falciparum malaria. AIP Neg. 88922 X 250 
ply of a single layered, or sometimes multiple 
layered, ring of neuroglial cells interposed be- 
tween the hemorrhage and the necrotic zone 
surrounding the central vessel. In this series 
there was no instance of “granuloma” in 
which the necrotic zone was entirely filled in 
with neuroglial cells. It was thought that 
microglia were also present in the granuloma, 
but silver stains were of no benefit in the 
identification of the cells since the tissues sub- 
mitted often had been fixed in formalin too 
long for satisfactory staining. Although the 
hemorrhages were generally of the same size 
and presumably of the same age, the granu- 
lomas were not found at the site of every 
hemorrhage. 
In the study of the “plugged,” parasitized, 
vessels of the brain, it appeared that the 
lumens were not obstructed by parasitized 
erythrocytes but simply filled and distended by 
them, as if during life they were part of 
a moving stream. In the cases showing both 
Fig. 8. Ring hemorrhage in brain of acute falciparum malaria. Note central “plugged” vessel, and necrotic 
intermediate zone. Erythrocytes of the hemorrhage are not parasitized. AIP Neg. 78510 X 330. 
Fig. 9. Ring hemorrhage of brain in fat embolism similar to those found in falciparum malaria. AIP Neg. 
89981 X 330. 
Fig. 10. Malarial “granuloma” composed of central thrombosed vessel, necrotic intermediate zone and 
peripheral rows of neuroglial cells mixed with nonparasitized erythrocytes. AIP Neg. 88928 X 500 
Fig. 11. Malarial “granuloma” showing demyelinization in central necrotic zone. AIP Neg. 80929 X 500 
Fig. i2. Microinfarct of Rocky Mountain spotted fever, similar to the malarial “granuloma.” AIP Neg. 
82899 X 4°° 
Fig. 13. Inflammatory nodule of brain in Chagas’ disease is in contrast to the noninflammatory “granuloma” 
of malaria. AIP Neg. 83009 X 450 560 
The Military Surgeon—November, 1946 
Figs. 16-21. Pathology of Acute Falciparum Malaria 
561 
hemorrhage and “granulomas,” thrombi 
could be identified in the plugged vessels (Fig. 
7), often in the central arteriole of the hemor- 
rhage or “granuloma” but also in the capil- 
laries at some distance from and unassociated 
with these lesions. Thrombi were also occa- 
sionally seen in the capillaries of the cortex. 
The thrombi were small, and usually dis- 
tended a small segment of the otherwise 
“plugged” capillary or arteriole. Parasites, 
parasitized erythrocytes and pigment particles 
often completely encircled the thrombus but 
were not included in it. In other words, some 
thrombi were observed which did not com- 
pletely occlude the lumens. 
In addition to these lesions, other non- 
specific findings were noted in the brain. 
There were moderate to severe grades of 
satellitosis and neuronophagia in all cases 
showing parasitization, paralleling in each 
instance the degree of “plugging” of vessels 
and more prominent in those cases showing 
cerebral purpura. A few mononuclear cells 
were present in the edematous meninges and in 
the Robin-Virchow spaces of many cases. Pial 
edema was a prominent feature of all cases in 
which there were parasitization and engorge- 
ment of cerebral capillaries. A number of 
cases originating in the tropics, and therefore 
poorly preserved for purposes of critical study, 
showed certain effects of postmortem decom- 
position; namely, large vocuoles, particularly 
in the white matter of the brain, which often 
contained crystals of formalin pigment; and 
also, an irregular distribution of the Purkinje 
cells of the cerebellum so that for intervals 
the cells seemed to have fallen out or under- 
gone a degenerative change. 
Lungs: The degree to which parasites were 
present in the alveolar capillaries was found 
to be parallel to the concentration of parasites 
in the brain. Although dilatation and hyper- 
emia of the septal capillaries was prominent in 
all lungs examined, in those which showed 
large numbers of parasites, these changes were 
extreme. Pulmonary edema was present in all. 
The parasitized erythrocytes were found in 
vessels of all sizes but were most prominent in 
the septal capillaries which were engorged 
with such cells often to the exclusion of others. 
Occasionally, considerable numbers of large 
pigmented phagocytes were found in the septal 
capillaries and were wedged tightly into the 
lumen. Thrombosis of vessels was not noted 
in the lungs. Small areas of hemorrhage were 
not uncommon in the lungs; the erythrocytes 
of these hemorrhages were, as in the brain, 
quite free of parasites. No infarcts were noted. 
Pneumonia was found in 21 cases (42 per 
cent) and occurred as commonly in those 
cases in which parasites were few as in those 
in which large numbers of parasites were pres- 
ent. Fourteen (28 per cent) of these cases 
showed simple bronchopneumonia and in one 
of these multiple abscesses were present. The 
infiltrate in all but one case was not unusual; 
in the one case, however, in which the parasite 
concentration was high in all organs, the in- 
filtrate was composed, for the most part, of 
polymorphonuclear leukocytes which con- 
tained many pigment granules. Phagocytosis 
of pigment and parasites in other organs (liver 
and spleen )was accomplished by mononuclear 
cells; moreover, in no other instance did the 
polymorphonuclear leukocytes exhibit evidence 
of the capacity for phagocytosis of pigment or 
Fig. i 6. Splenic follicle in falciparum malaria showing hyperchromatism of cells. Note parasitized cells 
in central artery, and phagocytosis of pigment by cells at the periphery of the follicle. AIP Neg. 78198 X 500 
Fig. 17. Almost complete depletion of splenic follicle occurring in a case of overwhelming falciparum 
infection. AIP Neg. 89875 X 230 
Fig. 18. Parasites and parasitized erythrocytes in splenic pulp. Few granules of pigment only in basophilic 
macrophages. AIP Neg. 93678 X 650 
FlG. 19. Marked phagocytic activity in macrophages of spleen. Each pigment granule is small, round and 
discrete. AIP Neg. 78195 X 650 
Fig. 20. Agglutination of pigment occurs within the macrophages in older cases. AIP Neg. 93679 X 650 
Fig. 21. Spleen of chronic recurrent vivax malaria showing distended sinusoids, fibrosis of cords and large 
amounts of agglutinated pigment. AIP Neg, 84312 X 250 562 
The Military Surgeon—November, 1946 
parasites. Only one case showed lobar pneu- 
monia. In 6 cases (12 per cent) there was 
interstitial pneumonitis, of the type similar to 
that of “virus pneumonia,” scrub typhus and 
several other diseases. This lesion was mild in 
four of the cases, consisting merely of septa 
thickened by mononuclear cells, a few of 
which were pigmented, and moderate num- 
bers of these cells in the edema fluid of the 
alveoli. Two cases showed a more intense 
reaction that was accompanied by purulent 
bronchiolitis. These two cases were among 
those that exhibited renal lesions (hemo- 
globinuric nephrosis). The occurrence of in- 
terstitial pneumonitis in 12 per cent of this 
series is somewhat higher than that reported 
in Panama4 (3.7 per cent in military person- 
nel). 
In addition to these lesions, five of the 
cases showed a prominent hyaline membrane 
lining the alveoli. In none of these cases was 
there pneumonia; on the other hand, in all 
of them renal lesions (hemoglobinuric 
nephrosis) were present. 
Heart: The concentration of parasites 
again was parallel to that seen in the brain 
and lungs. The myocardial capillaries were in 
41 cases distended with parasitized erythro- 
cytes. In these cases parasites were found also 
in larger vessels, particularly veins, hugging 
the walls of these vessels and forming mounds 
of parasitized erythrocytes. It was particularly 
noticeable in the heart that these mounds were 
often present along the same side of the walls 
of vessels rather than around the entire cir- 
cumference. A similar finding was noted in 
the brain but was less conspicuous, and this 
uniform, eccentric settling of parasitized cells 
suggested that at least some of the clumping 
observed in the vessels might represent post- 
mortem phenomenon. In no case was there 
evidence of thrombosis of vessels of the heart, 
nor was there any case in which large numbers 
of parasitized erythrocytes were present in the 
large epicardial branches of the coronary 
arteries. One case showed moderate athero- 
matous changes in the intima of the left 
descending coronary artery and this was the 
only case in which fibrosis of the myocardium 
occurred. 
In forty-two of the cases (84 per cent) there 
was pronounced interstitial edema of the myo- 
cardium; the remainder were among those 
that showed few if any parasites. Edema was 
not localized to any one part of the heart. 
The muscle fibers were often widely separated 
by fluid but necrosis of the muscle fibers was 
not noted. In one case only was there myo- 
cardial fibrosis, and even in this instance the 
fibrosis was minimal. This was the case, noted 
above, associated with coronary sclerosis. In- 
asmuch as the fibrosis was obviously of long 
standing, it was considered to be pathogeneti- 
cally unrelated to the acute infection with 
falciparum malaria. Three of the cases showed 
small, irregular subendocardial hemorrhages. 
Interstitial myocardial infiltrate was pres- 
ent in 20 cases and was considered moderate 
in 11 cases and slight in 9 cases. The infiltrate 
was present exclusively in cases with high para- 
site counts and was irregular in distribution. 
Most often there were simply a few scattered 
foci of cells in the edematous interstitium. 
Occasionally there was infiltrate throughout 
a section. The cells of the infiltrate were in 
all instances mononuclear consisting of lym- 
phocytes, plasma cells, Anitschkow myocytes 
and large macrophages (Fig. 14). Macro- 
phages laden with pigment were not included 
in the infiltrate. 
The myocardial infiltrate involved the 
endocardium in three of the more severe cases 
and in one case there were several small endo- 
cardial verrucae (Fig, 15). In only one case 
of this series was there pericarditis; this lesion 
was thought to be a complication of the asso- 
ciated lobar pneumonia. 
Spleen: In contrast to other organs, the 
spleen and liver showed striking concentration 
of pigment, parasites, and parasitized erythro- 
cytes. The degree of concentration of para- 
sites in the spleen could not be correlated with 
the number of parasites in other viscera. The 
pigment present in the spleen was distinctly 
more abundant in those cases in which para- 
sites were scarce in the brain; whereas, those Pathology of Acute Falciparum Malaria 
563 
cases in which the parasite count of the brain 
was high showed less pigment, though many 
parasites, in the spleen. 
In those cases with high parasite counts, 
parasitized erythrocytes were found in large 
numbers in the splenic arteries and generally 
in the veins as well although in smaller num- 
bers. In the entire series, only four cases 
showed relatively few parasitized erythrocytes 
in the spleen; in the remainder there were 
large numbers. Parasites, parasitized erythro- 
cytes and phagocytes were located predomi- 
nantly in the red pulp although parasitized 
erythrocytes were also found in small numbers 
in the sinusoids. Both the Billroth cords and 
sinusoids were hyperemic. Fibrosis of the 
cords, so prominent in chronic malaria, was 
not noted in this group of acute cases. 
In approximately half the cases, a few cen- 
tral arteries showed the kind of hyaline swell- 
ing commonly found in spleens at autopsy. 
In almost all the cases, there were subintimal 
collections of cells in the trabecular veins. 
These cells had the morphology of extensions 
of white pulp and were so interpreted rather 
than as representing endophlebitis. This find- 
ing was not associated with thrombosis nor 
with any other unusual finding in the spleen 
and has been noted in routine autopsy material 
with approximately the same frequency. In 
only one was there an infarct which was small 
and fresh; it was not possible in the material 
available to identify thrombosis of an apper- 
taining vessel. 
The splenic follicles were variable in ap- 
pearance. In 17 cases (34 per cent) the fol- 
licles were of normal size or somewhat larger 
than normal; in the remainder (66 per cent) 
they were distinctly smaller than normal. Of 
the 17 cases in the former group, three were 
among those that had the highest concentra- 
tion of parasites in the viscera and relatively 
little phagocytic activity in the spleen; five 
had moderate numbers of parasites, and the 
remaining nine were those in which parasites 
were practically absent from the organs. Five 
of the cases showing follicles of normal or large 
size fell into the group also showing hemo- 
globinuric nephrosis. 
The majority of cases (66 per cent) showed 
distinct diminution in size of follicles (Fig. 16) 
and in three of these no definite structure could 
be identified as follicle (Fig. 17). The edges 
of the follicles were no longer sharp but grad- 
ually merged into the red pulp. The cell com- 
ponents of the follicle were completely altered. 
Practically no small lymphocytes were present; 
instead, there were large cells of varying size 
and with hyperchromatic nuclei, which re- 
sembled lymphoblasts and basophilic macro- 
phages (Fig. 16). Mitoses in these cells were 
frequent. Cells similar to those constituting 
the follicles, were also scattered in the cords 
and occasionally in the sinusoids where they 
frequently had phagocytosed pigment and 
parasites. 
Both phagocytosed and free pigment was 
in all cases most concentrated in the Billroth 
cords. It was distinctly unusual to find phago- 
cytosed pigment in any part of the follicle 
although occasionally a cell at the periphery 
of the follicle would contain a few grains (Fig. 
16), Phagocytosis of pigment occurred in the 
red pulp by large phagocytic cells having both 
acidophilic and basophilic cytoplasm (Fig. 18). 
It was not uncommon to count in a single 
plane of section fifty or more granules of pig- 
ment in the cytoplasm of a single cell. 
In this series, the larger amount of pigment 
was found as single round granules (Fig. 19). 
These granules were dark brown and could 
usually be distinguished from iron pigment, 
which might also be present in the same cells 
in considerable amounts, by the lighter color 
of the iron, and from formalin pigment, which 
was commonly present in the spleen, by the 
irregular, crystalline shape of the formalin. 
It was not possible to differentiate formalin 
pigment from parasite pigment on the basis 
of refractility since both were inconstantly 
refractile. In a moderate number of cases the 
malarial pigment was present also in small, 
black, conglomerate masses (Fig. 20) which 
occurred for the most part in macrophages but 
occasionally free in the red pulp as well. Only 
a relatively small number of pigmented 
macrophages escaped into the dilated sinusoids 
which, however, often contained large num- 564 
The Military Surgeon—November, 1946 
Figs. 26-28. Pathology of Acute Falciparum Malaria 
Fig. 29. Bone marrow of acute falciparum malaria. Phagocytosis of pigment and parasites is slight com- 
pared to spleen. AIP Neg. 78174 X 700 
Fig. 30. Peripheral blood of case of falciparum malaria; monocytes containing phagocytosed pigment are 
occasionally seen. AIP Neg. 85165 X 1360 
hers of parasitized erythrocytes. Infrequently 
a macrophage would seem to be adherent to 
the endothelial lining of the sinusoids but it 
was not possible to demonstrate that the endo- 
thelium possessed any phagocytic function. 
Ten cases picked at random were stained 
both by the Giemsa and Tomlinson-Grocott 
methods for plasmodium. All forms of Plas- 
modium falcifarum were present; segmenting 
schizonts and gametocytes were observed in 
greater numbers in the spleen than in the brain. 
Liver: The changes in the liver, so far as 
phagocytosis of pigment is concerned, were 
parallel to, but less intense than, those in the 
spleen. The pigment was present for the most 
part within the cytoplasm of the Kupffer cells 
which were often so crowded with pigment 
granules that they filled the sinusoids (Fig. 
27). The sinusoids also contained varying 
numbers of parasitized erythrocytes. Pigment 
or parasites were not present in the hepatic 
parenchymal cells. 
In 20 cases of this series there were collec- 
tions of mononuclear cells in the portal areas 
but this finding could not be correlated either 
with clinical evidence of anemia or with the 
duration of the disease. In 17 cases, there was 
edema of the liver (Fig. 26). Fat was not 
prominent in the liver of any case. In 4 cases 
there was moderate central necrosis. Another 
Fig. 26. Edema of liver in acute falciparum malaria. AIP Neg-. 93677 X 260 
Fig. 27. Kupffer cells showing- phagocytosis of large numbers of parasites and parasitized erythrocytes 
which can also be seen free in the sinusoids. AIP Neg, 93667 X 650 
Fig. 28. Early necrosis of liver in acute falciparum malaria as indicated by dark hyaline bodies in cytoplasm 
of hepatic and Kupffer cells. AIP Neg. 86375 X 450 566 
The Military Surgeon—November, 1946 
change of interest observed in 11 cases of the 
series was the occurrence of small, round, 
deeply acidophilic, hyalin-like bodies irregu- 
larly distributed through the liver, present 
both in parenchymal and Kupffer cells (Fig. 
28). This lesions was associated with hepatic 
edema or focal frank necrosis of the cells of 
hepatic cords. Morphologically these bodies re- 
sembled the cytoplasmic inclusions found in 
creased in number; erythroblasts were in- 
creased; megakaryocytes seemed to be more 
numerous than normal. 
Kidney: Parasitized erythrocytes were 
present in the peritubular capillaries of the 
kidneys with the same constancy but in smaller 
numbers than in those of other organs; but 
only rarely were they seen in the glomerular 
capillaries. On the other hand, it was unusual 
Fig. 31. Hemoglobinuria nephrosis in case of falciparum malaria. Pigmented casts in distal convoluted 
tubules associated with necrosis of tubular epithelium. AIP Neg. 93676 X 260 
yellow fever and were interpreted as evidence 
of early degeneration. 
Bone Marrow: Parasitized erythrocytes 
were parallel to other organs in the bone mar- 
row available in 18 cases (Fig. 29). The 
macrophages of the marrow were not as nu- 
merous as they were in the spleen; pigmented 
phagocytes were present but these cells con- 
tained fewer pigment granules and parasitized 
erythrocytes than similar cells in the spleen 
and liver. The marrow (either sternal or 
vertebral) was composed predominantly of 
mononuclear cells; granulocytes were de- 
to find circulating phagocytes except within 
the glomerular capillaries where they seemed 
to be trapped in the lumens of these vessels 
(Fig. 24). In practically all cases the peri- 
tubular venules of the medulla contained large 
numbers of mononuclear cells; this finding 
showed no correlation with the degree of 
interstitial infiltrate or with any other lesion 
of the kidney. 
Although pigmented casts were present in 
the lumens of a few of the distal convoluted 
tubules of 30 cases (60 per cent), hemoglo- 
binuric nephrosis, of the type seen following Pathology of Acute Falciparum Malaria 
567 
transfusion and in blackwater fever, was 
present in only 7 cases of this series (14 per 
cent). In these cases there were deeply pig- 
mented casts in the distal part of the nephron, 
particularly in the distal convoluted and col- 
lecting tubules, associated often with necrosis 
and regeneration of the tubular epithelium 
(Fig. 31). In addition, there was dilatation 
especially of the proximal renal tubules, col- 
organs, and in four there were parasitized 
cells only in the spleen. In five of these cases 
the splenic follicles were large. It is interesting 
that the only two Negroes of this group showed 
hemoglobinuric nephrosis. 
In nine cases of this series (18 per cent), 
of which four also had hemoglobinuric 
nephrosis, there was definite glomerular alter- 
ation consisting of generalized ischemia, en- 
Fig. 32. Acute glomerulonephritis in falciparum malaria: glomerular ischemia, increased cellularity of 
glomerulus, hyperchromatism of endothelial cells, as well as vacuolization of epithelium of tubules. AIP 
Neg. 89176 X 230. 
lection of protein fluid and often hemoglobin 
in Bowman’s spaces, interstitial edema and 
infiltration of lymphocytes, plasma cells, 
macrophages and occasional eosinophils. En- 
dophlebitis was rare. In all of these cases there 
had been clinical evidence of renal insufficiency. 
In the only case in which transfusion had been 
given, the patient was already anuric and 
death occurred on the same day the transfusion 
was administered. Of these seven cases of 
hemoglobinuric nephrosis three showed large 
numbers of parasitized erythrocytes in all 
largement and increased cellularity of the 
glomerulus, hyperchromatism and swelling of 
the endothelium, and occasionally thickening 
of the basement membranes (Fig. 32). Leuko- 
cytes were rarely found. Glomerulonephritis, 
of the type described above occurred in cases 
that had clinical evidence of azotemia; several 
of these cases showed also the hyaline mem- 
brane lining the pulmonary alveoli, the so- 
called uremic pneumonitis.* 
* Personal communication of Dr. Tracy B- Mal- 
lory. 568 
The Military Surgeon—November, 1946 
In addition, in many of the cases, including 
those showing glomerulonephritis, the epi- 
thelium of the proximal convoluted tubules 
was swollen by cytoplasmic vacuoles. Three of 
the cases showed calcification of the epithelium 
of the distal convoluted tubules for no appar- 
ent clinical reason. Three cases showed 
moderate fibrinoid alteration of arterioles. In 
two cases there were a few acetylated sul- 
fathiazole crystals in the lumens of the tubules 
although there was no associated inflammatory 
reaction in the kidney. 
Gastrointestinal Tract: One to several sec- 
tions of the gastrointestinal tract were available 
for study in 22 cases. In fifteen of these there 
was engorgement of the mucosal capillaries 
by parasitized erythrocytes. Hemorrhage into 
the mucosa was noted in only two cases unas- 
sociated with uremia, and submucosal edema 
was seen in three cases showing large numbers 
of parasites. Similar changes were present in 
two cases in which parasites were absent. 
Parasite concentration in the intestine was 
also parallel to that of other organs. It was 
interesting that only four of these cases had 
had diarrhea and that eleven cases showing 
equal numbers of parasites in the intestinal 
tract had not had symptoms referrable to the 
intestine. No ulceration was noted, but one 
case in which sections of stomach were sub- 
mitted showed diffuse gangrenous gastritis 
similar to that of mercury poisoning. The possi- 
bility of action by a poison could not be elimi- 
nated on the information available. 
Pancreas: The pancreas was studied in ap- 
proximately half the cases. Only one change 
was noted aside from the presence of para- 
sitized cells in the capillaries to the same extent 
as in other organs. In six cases there was mild 
cystic dilatation of the pancreatic acini and 
ducts; two of these patients had had diarrhea, 
whereas the other four were in coma on ad- 
mission. 
Adrenals: The adrenals were available in 
all but eight cases. The concentration of para- 
sitized erythrocytes was similar to that in other 
viscera; however, it did appear that there were 
greater numbers of pigmented phagocytes in 
the sinusoids of the adrenals than in other 
organs. A moderate degree of degeneration 
of the fascicular cords was noted in twelve 
cases, all of which were admitted in coma; 
this lesion was marked in only one case in 
which bronchopneumonia was also present. 
Lymfh Nodes: Lymph nodes from ‘un- 
stated regions were studied in twelve cases. In 
general, parasites were found infrequently in 
the nodes even when they were present in 
large numbers in other organs. The marginal 
sinuses of several of these nodes were filled 
with mononuclear cells, similar to the baso- 
philic histiocytes in other locations. 
Other organs: Sections of testis, prostate, 
skeletal muscle and skin were present in some 
of the cases but showed no alterations of note. 
Discussion 
“Localization of Parasites”: From this series 
it appears that the majority of patients dying of 
acute uncomplicated falciparum malaria pre- 
sents overwhelming infection of erythrocytes 
by plasmodium. Many of the cases would be 
classified in the category of “cerebral malaria,” 
described and accepted by many authors, since 
90 per cent of this group presented cerebral 
symptoms and many of these showed striking 
numbers of parasitized erythrocytes in the 
brain. Seyfarth5 has divided fatal falciparum 
malaria into several categories, of which the 
“septicemic” form accounts for 30 per cent 
and the “cerebral” form accounts for 55 per 
cent of the fatalities; the remainder are classi- 
fied according to the “localization” of para- 
sites in various viscera. In this series “localiza- 
tion” or concentration of parasites occurs only 
in the spleen, liver and bone marrow and 
parasitized erythrocytes in other organs, in- 
cluding the brain, have been found to be 
otherwise uniformly distributed. In other 
words, if the infection is heavy, there are equal 
numbers of parasitized cells in the capillaries 
of such organs as the brain, heart, lungs and 
intestinal tract, although the clinical symptoms 
may point to one organ primarily. If the infec- 
tion is light, the distribution of parasitized cells 
is similarly uniform and generalized. The 
theory of selective localization of parasitized Pathology of Acute Falciparum Malaria 
569 
cells, then, cannot account for the varied clini- 
cal phenomena so characteristic of fatal 
falciparum malaria. 
Pigment: Although striking numbers of 
parasitized erythrocytes and parasites are 
found in the vessels of all organs and although 
an occasional circulating pigmented phagocyte 
may be seen in the capillaries of any organ or 
may even appear in the peripheral blood (Fig. 
30), the bone marrow, liver and spleen are 
clearly the depositories of malarial pigment 
(Fig. 14-21, 24-27). Taliaferro6’7 has 
emphasized that the death of parasites is ac- 
complished by the phagocytosis of free para- 
sites and parasitized erythrocytes by existing 
and new macrophages in situations where the 
blood flows slowly and comes into intimate 
contact with such cells. The most intense 
phagocytosis is therefore seen in the spleen, to a 
certain extent in the liver, and least of all in 
the bone marrow. Endothelial cells lining 
vessels do not exhibit the power to phagocytose 
pigment of malarial parasites, notwithstanding 
their capacity to phagocytose granules of 
supravital injected dyes. Macrophages, capable 
of phagocytosing pigment, are already present 
in large numbers in these organs prior to in- 
fection. While these cells multiply during the 
infection (Fig. 16) they are not the only 
source of phagocytes inasmuch as other cells 
are activated to produce new macrophages 
(Fig. 16), particularly in the spleen where, as 
Taliaferro and Cannon8 have shown, the di- 
rect transformation of lymphocytes into 
macrophages occurs. In the suppression of the 
infection, the parasites and parasitized eryth- 
rocytes are held in the Billroth cords of 
the spleen by the newly developed macrophages 
probably by means of a specific agglutinative 
substance which develops at the surfaces of 
these cells (Cannon9). The lymphocyte is of 
great importance in malarial infections not 
only from the standpoint of its evolution into 
macrophages, but also on account of its major 
role in the transportation of antibodies.10 Here, 
too, as in the typhus fevers and in other in- 
fections characterized by a lymphocytosis of the 
peripheral blood, the problem must be further 
clarified by multiple correlations of antibody 
levels, peripheral lymphocytosis, and tissue 
lymphocytosis. 
Both malarial and iron pigments were found 
in the macrophages. Malarial pigment consists 
of small round dark brown granules which in 
the majority of cases are discrete globules of 
somewhat varying size (Fig. 18, 19 and 22). 
The smaller granules which represent the pig- 
ment of the early schizonts appear a lighter 
brown color than the darker, somewhat larger 
granules of the older parasites. In some of the 
cases there is agglutination of the single pig- 
ment granules into irregular masses of pigment 
which are almost black (Fig. 20). This 
agglutination of pigment probably indicates 
longer duration of the infection (Cannon ). 
The smaller pigment granules are usually 
within the cytoplasm of macrophages while the 
agglutinated masses of pigment may be found 
extracellularly. 
Although the sinusoids are dilated in these 
cases, the infection had not existed for periods 
long enough to produce the striking fibrosis 
often noted in cases of chronic vivax malaria 
(Fig. 21 and 23) nor did they show the 
acute congestion of the ruptured spleen of acute 
vivax malaria, of which there were six ex- 
amples available for study at the Army Insti- 
tute of Pathology. 
Pigment that could be stained for iron was 
often present within the same macrophages that 
contained parasite pigment. Rigdon11 sug- 
gests, from experimental evidence, that malarial 
pigment, ferrihemate or hematin (Morrison 
and Anderson12), which can not be stained 
for iron, may be gradually oxidized into an 
iron-containing pigment which may then be 
utilized by the body, thus accounting for its 
gradual disappearance. This evidence is con- 
vincing although some of the iron-containing 
pigment in malaria undoubtedly results from 
the destruction of erythrocytes during the 
cyclic development of the parasite. 
Except for its morphology and failure to 
stain for iron, malarial pigment is difficult to 
distinguish from other pigments. Although 
Cowdry13 makes the interesting observation 
that malarial pigment is doubly refractile, this 
feature was not found to be constant in our 570 
The Military Surgeon—November, 1946 
fixed, stained tissues; and in addition, formalin 
pigment, so often present in these tissues, is 
also occasionally refractile, which further 
diminishes the usefulness of this criterion, 
Consideration of the Pathogenesis of Cere- 
bral Lesions: The most constant finding in 
the brains of this series of acute fatal falciparum 
malaria is the intense engorgement of cerebral 
capillaries (Fig. 3) either in the complete 
absence of or in association with parasitization 
of erythrocytes. The precise mechanism of the 
engorgement is not clear but such factors as 
stagnation of blood in the capillaries, throm- 
boses, and anoxia from other causes are 
probably concerned. Certainly a similar picture 
of extreme engorgement of cerebral capillaries 
is seen following anoxia due to flying in high 
altitudes (Fig. 4). 
In any case it appears obvious that factors 
other than the simple “plugging” of the 
capillaries by parasitized cells must exist to 
explain the cerebral symptoms which so often 
characterize this disease. Rigdon,14 Kean and 
Smith15 and others have suggested that these 
symptoms may be related to shock. The signs 
of shock are noted also in this series. However, 
the explanation of the pathogenesis of shock 
is no less simple in acute falciparum malaria 
than in other infections and conditions. From 
the point of view of the morphologist, the 
stagnation of the capillary blood flow due to the 
presence of parasitized red blood cells of in- 
creased density, and perhaps viscosity, the 
effects of toxemia on capillary tone, the role 
of the adrenal gland (Rich16), the effect of 
involvement of important cerebral centers, 
are all possible single or multiple factors in the 
production of the overall picture of shock. In 
these acute cases, anemia was not present to a 
degree sufficient to account for significant 
anoxia. 
The occurrence of cerebral hemorrhages 
(Fig. 1 and 8) and the so-called granulomas 
(Fig. 10 and 11) in 30 per cent of these 
cases is explained on the basis of thrombosis 
(Fig. 7 and 10). Anderson and Morrison17 
have recorded widespread thromboses in 
monkeys with simian malaria and by the in- 
jection of ferrihemate (parasite pigment) but 
explain these findings on the basis of anoxia 
related to anemia; they also disclaim malarial 
pigment as the specific toxic factor since it is 
not liberated in soluble form from the parasite. 
It is clear that in malaria, the hemorrhages 
and granulomas, which are actually minute 
infarcts, are identical with those same lesions 
found in brain purpuras from a variety of 
other causes, for example fat embolism (Fig. 
9) and spotted fever, (Fig. 12) in which the 
vessels of the brain are occluded either by 
emboli as in the former, or by thrombi as in 
the latter. 
Vance18 has offered a clear and logical ex- 
planation for the localization of these hemor- 
rhagic lesions in the white matter of the brain 
(Fig. 1) rather than in the cortex. He states 
that although emboli (of fat) are present in 
the cortex, lesions can not be demonstrated 
since “the capillary anastomoses are so ex- 
tensive that the nutrition of the tissue is rarely 
impaired. ... In the white matter the situa- 
tion is different. The arterioles are end-vessels 
which branch in a dendritic fashion and which 
communicate by scanty capillary anastomoses. 
The embolus obstructing such a vessel effec- 
tively impairs the nutrition of the tissue it 
supplies and degenerative changes occur.” 
The ring form of the hemorrhage is also 
common to other cerebral purpuras, and it is 
generally conceded that the source of the 
hemorrhage is not from the occluded central 
vessel but from the surrounding capillaries 
which are not occluded. The “granuloma” is 
interpreted as part of the process of repair 
whereby the necrotic zone is more or less re- 
placed with glial cells, rather than as an in- 
flammatory nodule provoked by toxins or para- 
sites such as occurs in toxoplasmosis and 
trypanosomiasis (Fig. 13). The fact that the 
“granulomas” were always found to be as- 
sociated with hemorrhage and that solid nod- 
ules unassociated with hemorrhage, such as 
those described by Dhayagude and Puranare,19 
were not found, is undoubtedly due to the 
short duration of the disease in this series. 
Other Thrombotic Lesions: It has been 
possible in this series of cases to demonstrate 
thrombosis only in the cerebral vessels of small Pathology of Acute Falciparum Malaria 
571 
calibre. Only one infarct of the spleen was 
noted in this group, whereas the subendothelial 
proliferation of the white pulp to which splenic 
infarcts have been attributed20 was found in 
practically all the cases. Therefore, the mere 
intimal location of these cells cannot be re- 
garded as the cause of these infarcts. Likewise, 
the capillaries of the heart also contained large 
numbers of parasitized erythrocytes whereas 
the larger branches of the coronary arteries 
were not similarly “plugged.” The only ex- 
ample of myocardial infarction in this series 
was of long duration and related to coronary 
sclerosis rather than the acute malaria, con- 
trary to the interpretation of recent reports.21 
Hepatic Dysfunction: Changes other than 
concentration of parasites and malarial pigment 
in the liver occur in falciparum malaria. 
Edema was noted regularly (Fig. 26), central 
necrosis occasionally, and in a significant num- 
ber of cases there was evidence of early 
necrosis (Fig. 28). Clinical evidence of hepatic 
dysfunction in malaria has been reported by 
Nursky, von Brecht and Williams22 and similar 
dysfunction may have been present in these 
patients with falciparum malaria. However, it 
does not seem justifiable to conclude that the 
histologic evidence warrants the assumption 
that hepatic dysfunction existed. In addition, 
it would appear that these changes could be 
completely reparable since widespread destruc- 
tion of the parenchyma did not occur. 
Renal Complications: Conflicting state- 
ments are found in the literature regarding 
the incidence of nephritis in malaria. In 
chronic malaria, particularly in the recurrent 
malariae malaria, the incidence of nephritis as- 
sociated with the clinical complex of lipoid 
nephrosis is high (Menon and Annamalai23 
and Giglioli24). Nephritis in acute malaria is 
said to be rare except in the fatal forms 
(Seyfarth5). In this series, glomerulonephritis 
occurred in 18 per cent of cases. This type of 
intracapillary glomerulonephritis (Fig. 32) is 
that which is found in other infections such as 
scarlet fever and the typhus fevers,25 and, as in 
these latter infections, is regarded as a toxic, or, 
more likely, an allergic manifestation. These 
lesions have been produced experimentally in 
dogs26 and monkeys17 by inoculation of 
ferrihemate and by the production of experi- 
mental malaria. Again, as in the typhus fevers 
so with the malarias, much experimental work 
needs to be done from the point of view of 
determining the histologic effect of various 
toxins and allergens of the rickettsiae as well 
as the plasmodia on the kidney and other 
organs. 
Blackwater fever is a well known compli- 
cation of falciparum malaria particularly in 
recurrent infections and after quinine therapy. 
The pathologic findings in blackwater fever 
are identical with those of hemoglobinuric 
nephrosis due to a variety of other causes such 
as the crush syndrome, incompatible transfu- 
sion, and other hemolytic diseases. The oc- 
currence of hemoglobinuric nephrosis in 14 
per cent of the cases of acute falciparum 
malaria of this series is unexpectedly high since 
there were not other factors except the malaria 
itself by which to explain this phenomenon. 
The occurrence of “blackwater fever” in acute 
malaria, as well as after recurrent attacks and 
after therapy, may then be considered as re- 
lated to the hemolysis which may be extreme 
in cases of overwhelming infection with 
Plasmodium falciparum. The precise immuno- 
logic or mechanical basis for this phenomenon 
remains to be clarified. It is of interest that 
four of the cases of hemoglobinuric nephrosis 
were associated with early acute diffuse 
glomerulonephritis. This is to say that in a 
considerable proportion of instances of hemo- 
globinuric nephrosis due to malaria, the 
glomerular dysfunction is an important factor 
in the production of the signs and symptoms of 
renal insufficiency. 
Summary 
1. The pathologic lesions in 50 cases of fatal 
falciparum malaria are presented and corre- 
lated with clinical information available. 
2. Selective localization of parasitized 
erythrocytes does not occur outside the bone 
marrow, liver and spleen; the presence of 
“plugged” parasitized vessels in an organ can 
not be correlated with clinical symptoms. 
3. Cerebral symptoms are considered to be 572 
The Military Surgeon—November} 1946 
related to anoxia since the lesions found in 
anoxia from other causes are present in identi- 
cal form in the brain of falciparum malaria. 
4. Thrombosis of the cerebral vessels is fre- 
quent, causing hemorrhages and the so-called 
granulomas. Thrombotic lesions are not pres- 
ent in other organs of this series. 
5. Renal lesions, both diffuse glomerulo- 
nephritis and hemoglobinuric nephrosis, occur 
with relative frequency and are, together with 
the myocardial infiltrate and adrenal “atro- 
phy,” evidence in favor of circulating toxins 
or allergens, probably released by the Plas- 
modium jalcifarum. 
Demonstration of Antibodies in Lymphocytes. Proc. 
Soc. Exper. Biol. & Med. 57: 295-298, 1945. 
11 Rigdon, R. H.: Malarial Pigment. Am. J. Clin. 
Path. 15: 489-493, 1945, 
12 Morrison, D. B., and Anderson, W. A. D.: The 
Pigment of the Malaria Parasite. Pub. Health Rep. 
57: 90-94, 1942. 
13 Cowdry, E. V., and Scott, G. H.: Etudes cytolo- 
giques sur le paludisme; mitochondries, granules 
colorables, au rouge neutre et appareil de Golgi. Arch. 
Inst. Pasteur de Tunis. 17: 233-252, 1928. 
14 Rigdon, R. H.: A Consideration of the Mecha- 
nism of Death in Acute Plasmodium Falciparum In- 
fection. Am. J. Hyg. 36:269-275, 1942. 
15 Kean, B. H., and Smith, J. A.: Death Due to 
Estivo-autumnal Malaria. Am. J. Trop. Med. 34: 
379-381, 1944. 
18 Rich, A. R.: A Peculiar Type of Adrenal Corti- 
cal Damage Associated with Acute Infections and Its 
Possible Relation to Circulatory Collapse. Bull. Johns 
Hopkins Hosp. 74:1-15, 1944. 
17 Anderson, W. A. D., and Morrison, D. B.: Role 
of Parasite Pigment (Ferrihemic Acid) in the Pro- 
duction of Lesions in Malaria. Arch. Path. 33: 677- 
686, 1942. 
18 Vance, B. M.: The Significance of Fat Embolism. 
Arch. Surg. 23; 426-465, 1931. 
19 Dhayagude, R. G., and Puranare, N. M.: Autopsy 
Study of Cerebral Malaria with Special Reference to 
Malarial Granuloma. Arch. Path. 36: 550-558, 1943. 
20 Rigdon, R. H.: A Consideration of the Mecha- 
nisms of Splenic Infarcts in Malaria. Am. J. Trop. 
Med. 24: 349-354) *945- 
21 Merkel, W. C.: Plasmodium Falciparum Malaria. 
Arch. Path. 41: 290-298, 1946. 
22 Mirsky, I. A., von Brecht, R. and Williams, 
L. D.: Hepatic Dysfunction. Science. 99: 20-21, 1944. 
23 Menon, T. B., and Annamalai, D. R.: Nephrosis 
in Malaria. J. Trop. Med. 36: 379-381, 1933. 
24 Giglioli, G.: Clinical Notes, Autopsy and Histo- 
pathological Findings from Five Cases of Quartan 
Malarial Nephritis from British Guiana. Trans. Roy. 
Soc. Trop. Med. & Hyg. 26: 177-187, 1932. 
25 Allen, A. C., and Spitz, S.: A Comparative Study 
of the Pathology of Scrub Typhus (Tsutsugamushi 
Disease) and other Rickettsial Diseases. Am. J. Path. 
21: 603-681, 1945. 
26 Anderson, W. A. D., Morrison, D. B., and Wil- 
liams, E. F., Jr.: Pathologic Changes Following In- 
jections of Ferrihemate (Hematin) in Dogs. Arch. 
Path. 33: 589-602, 1942. 
References 
1 Helpern, M.: Epidemic of Fatal Estivo-autumnal 
Malaria. Am. J. Surg. 26: m-123, 1934. 
2 Tomlinson, W. J., and Grocott, R. G.: Simple 
Method of staining Malaria Protozoa and Other Para- 
sites in Paraffin Sections. Am. J. Clin. Path. 14; 316- 
326, 1944. 
3 Diirck, H.: Uber die bei Malaria perniciosa coma- 
tosa auftretenden Veranderungen des Zentralnerven 
systems. Arch. f. Schiffs. u. Tropenhyg. 21: 117-132, 
191 7- 
4Applebaum, I. L., and Schrager, J.: Pneumonitis 
Associated with Malaria. Arch. Int. Med. 74: 155- 
162, 1944. 
5 Seyfarth, C.: Die Malaria. In Henke, F., and Lu- 
barch, O. Handbuch der speciellen pathologischen 
Anatomie und Histologie. 1: 178-248. Berlin, 
Springer, 1926. 
8 Taliaferro, W. H., and Mulligan, H. W. The 
Histopathology of Malaria with Special Reference to 
the Function and Origin of the Macrophages in De- 
fense. Indian M. Research Mem. 29: 1-138, 1937. 
7 Taliaferro, W. H.: Immunity in Malaria. Am. J. 
Clin. Path. 14: 593-597, 1944- 
8 Taliaferro, W. H., and Cannon, P. R.: Cellular 
Reactions during Infections and Superinfections of 
Plasmodium brasilianum In Panamanian Monkeys. J. 
Infect. Dis. 59: 72-125, 1936. 
* Cannon, P. R.: Some Pathologic Aspects of Hu- 
man Malaria: A Symposium on Human Malaria. 
Pub. of A.A.A.S. 15:214-220, 1941. 
10 Dougherty, T. F., Chase, J. H., and White, A, TUMORS OF THE TESTIS* 
A Report on 922 Cases 
By MAJOR NATHAN B. FRIEDMAN, Medical Corfs, Army of the United States, and 
ROBERT A. MOORE, M.D., Resident Consultant 
(With thirty-one illustrations) 
THIS paper deals first, with the classifi- 
cation, architecture and morphogenesis 
of testicular tumors, and second, with 
the biologic behavior of the different types, as 
evidenced by the incidence and cellular com- 
position of metastases, the evolution of struc- 
of Pathology between October 1940, when 
selective service began, and May 1946. The 
tumors were obtained from a relatively 
homogeneous group, since most of the patients 
were of military age (18-38) and were serv- 
ing in the Army of the United States. 
Fig. i. Percentage distribution of testicular tumors. 
tural patterns and the mortality rates. The 
data were derived from study of 922 tumors 
of the testis collected at the Army Institute 
The fact that only 4 fundamental struc- 
tural patterns (seminomatous, embryonal car- 
cinomatous, chorioepitheliomatous and tera- 
tomatous) were encountered, alone or in com- 
bination, in the vast majority of the tumors 
permitted use of the following simple classi- 
fication : 
* From the Army Institute of Pathology, Wash- 
ington, D.C., and the Department of Pathology, 
Washington University School of Medicine, St. Louis, 
Missouri. 574 
The Military Surgeon—November, 1946 
Fig. 2. Seminoma, showing- typical clear cytoplasm and sharp borders of cells. X 550. 
Fig. 3. Pseudotubular clusters of cells in seminoma without lymphoid stroma. X 130. 
1. Seminoma (germinoma) 
2. Embryonal carcinoma 
2A. Chorioepithelioma 
3. Teratoma 
4. Teratocarcinoma 
Over 96 per cent of the neoplasms fell into 
one of these categories; of the remainder 1 
per cent were interstitial cell tumors, and 3 
per cent were rare or unclassifiable varieties. 
The percentage distribution of the different Tumors of the Testis 
types is indicated in Figure 1. Chorioepitheli- 
omas were included with embryonal car- 
cinomas. Seminomatous tissue was present in 
about 20 per cent of embryonal carcinomas and 
teratocarcinomas, but the seminomatous com- 
ponents of such neoplasms were ignored in 
classification because their presence did not 
affect the behaviour of the tumors. 
cells were usually arranged in unorganized 
masses of cords divided by trabeculae of con- 
nective tissue. Differentiated glandular epi- 
thelial structures were almost never produced 
by seminoma cells, although occasionally cords 
and columns formed pseudotubular aggregates 
without lumens (Fig. 3). 
Although seminomas comprised a more 
homogeneous group than any other type, cer- 
tain variations in structure did occur. In some 
instances the cytoplasm was darkly stained 
instead of clear. Rare “anaplastic” seminomas 
Seminoma 
One third of the tumors were monocellular 
seminomas. They were made up of rounded 
Fig. 4. Anaplastic seminoma. The nuclear pattern resembles that of embryonal carcinoma. X 450. 
Fig. 5. Junctional zone between seminomatous and embryonal carcinomatous components of a mixed 
neoplasm. X 130. 
polyhedral elements with sharp cell borders; 
the cytoplasm was often clear (Fig. 2). The 
round, centrally placed nuclei were clearly 
outlined by a membrane, and the nucleolus, 
which was usually but not invariably single, 
was also centrally located. The nuclear 
chromatin, apart from the nucleolar masses, 
was evenly dispersed; the beading at the 
interstices of the fine chromatin net imparted a 
stippled pattern to sections examined under 
low magnification. The remarkably uniform 
offered diagnostic difficulties because their 
nuclei were hyperchromatic and irregular and 
bore some resemblance to the pattern seen in 
embryonal carcinomas (Fig. 4), Such un- 
usual tumors, to which the term “seminoid 
carcinoma” might be applied, were exceptions 
to the rule that seminomas and embryonal 
carcinomas are differentiated with ease. In a 
few seminomas foci of anaplasia were seen in 
expanding growth centers. 
In about 4 per cent of the 922 tumors both 576 
The Military Surgeon—November, 1946 
Fig. 6. Seminoma with lymphoid stroma. X 230. 
Fig. 7. Seminoma with abundant granulomatous stroma which contains many epithelioid cells. X 130. 
seminomatous and embryonal carcinomatous 
tissue were encountered. All tumors in this 
group were classified as embryonal carcinomas, 
regardless of the relative distribution of the 
2 types of structure, because the presence of 
even a small focus of carcinoma greatly 
worsens the prognosis for the patient. In dif- 
ferent tumors the 2 elements were separate Tumors of the Testis 
577 
and discrete, contiguous or closely inter- 
mingled. When the 2 types were intimately 
associated the junctional zone usually sug- 
gested that the more malignant carcinomatous 
elements were invading the seminomatous tis- 
sue (Fig. 5). In rare instances it seemed as if 
transitions could be made out and that em- 
bryonal carcinoma cells were developing from 
encountered in ovarian disgerminomas, were 
commonly observed. Sometimes the formation 
of granulomas (Fig. 7) and fibrosis were so 
extreme that only scattered neoplastic cells 
remained; fibrous nodules occasionally formed 
in this way. Furthermore, not only were many 
seminomas devoid of lymphoid cells (Fig. 3), 
but occasional embryonal carcinomas showed 
Fig. 8. Atrophic seminiferous tubule entirely surrounded by seminoma cells. X 185 
seminoma cells. 
Most of the neoplastic giant cells encoun- 
tered contained multiple nuclei of the 
seminoma type, but occasional syncytiotro- 
phoblastic elements were also present. Giant 
cells of the foreign body and Langhans types 
were common in the peculiar stroma of many 
seminomas. 
The lymphocytic infiltration of the trabec- 
ulae of connective tissue in seminomas (Fig. 
6) has led to the use in some laboratories of 
the designation “embryonal carcinoma with 
lymphoid stroma.” Although cells resembling 
lymphocytes were certainly the most common 
type of secondary element, the stroma was 
often infiltrated with eosinophils and plasma 
cells. Epithelioid cells and tuberculoid 
granulomas with giant cells, similar to those 
typical “lymphoid stroma.” The misleading 
term “embryonal carcinoma with lymphoid 
stroma,” which designates neither a clinical 
nor a pathologic entity, was not used in classi- 
fying the neoplasms in this series. Most of the 
tumors to which it could be applied were 
seminomas; a small minority were true em- 
bryonal carcinomas. 
Foci of coagulation necrosis were often 
encountered; the necrotic masses were char- 
acteristically bordered by a zone of palisaded 
macrophages and foam cells. The florid 
hemorrhagic necrosis of chorioepitheliomas 
and embryonal carcinomas was not seen in 
seminomas. As seminomas grow they respect 
tissue planes and adjacent structures and do 
not aggressively invade and destroy as do 
embryonal carcinomas. For example, surviv- 578 
The Military Surgeon—Novembery 1946 
ing testicular tubules (Fig. 8) are often ob- 
served within the neoplastic tissue. Invasion 
of the adnexae and cord is exceptional, while 
it is the rule for embryonal carcinomas. The 
small incidence of regional and distant 
metastasis and the low mortality rate are con- 
sistent with the relatively benign appearance 
of seminomas. 
Pure seminomas occurred in relatively older 
men than did the other types of neoplasms. The 
incidence of seminomas and teratocarcinomas, 
comparably common tumors, among men of 
with seminomas although these tumors differ 
not only in fundamental cell type but in bio- 
logic behavior and prognosis. The neoplasms 
designated “embryonal carcinoma” include, 
among other types, those called “embryonal 
adenocarcinoma” and “papillary adenocar- 
cinoma.” As will be pointed out, even chorio- 
epitheliomas probably belong in this group. 
Embryonal carcinomas showed considerably 
more variation in cellular type from tumor to 
tumor and in different portions of the same 
tumor than did seminomas. The cells were 
Fig. 9. Frequency distribution of seminomas and teratocarcinomas by age group. 
Seminomas occur in older men. 
various ages is given in Figure 9. It should be 
pointed out that the difference (5 years) be- 
tween the two means of 25 and 30 is 25 
per cent of the range into which the ages of 
most of the patients fell. In Figure 10, which 
presents the same data, account is taken of the 
number of men in the Army in each age 
group; it shows that the incidence of terato- 
carcinoma was roughly the same over the en- 
tire range, while that of seminoma continued 
to rise with age. The downward curve after 
45 cannot be properly evaluated because the 
number of men affected and the total Army 
population in this group were too small. 
Embryonal Carcinoma 
Embryonal carcinomas are often confused 
frankly epithelial, often cuboidal or columnar, 
and frequently formed differentiated glandular 
or papillary structures. Differentiation was 
often incomplete, so that portions or all of a 
tumor consisted of unorganized solid epithelial 
sheets. The obviously anaplastic cells were 
larger than those of seminomas (Fig. 11). 
The nuclei, which were large, variable and 
bizarre, showed few traces of the orderly 
seminomatous pattern; the chromatin was 
clumped irregularly into large masses which 
stood out sharply against the clear background. 
Although embryonal carcinomas can be 
divided into a number of varieties on mor- 
phologic grounds, biologic significance and 
prognostic import cannot as yet be ascribed to 
all of them. A few distinctive patterns deserve Tumors of the Testis 
579 
comment because of their frequent occurrence. 
Glandular structures, often strikingly papillary, 
characterized a common neoplasm (Fig. 12); 
the cytoplasm was often clear basally, and 
there was a cuticular or hairy cell border. In 
other tumors the epithelial cords formed 
anastomosing trabecular networks; a variant 
of this picture was seen in the reticular type 
(Fig. 13), in which the anastomosing epi- 
thelial cords intertwined with the stroma, as 
solid sheets and masses. Usually the tumors 
resembled only the cytotrophoblastic com- 
ponents of chorioepitheliomas, but occasionally 
dark smudgy elongated cells appeared to be 
applied against the predominant cells (ap- 
pliqued), as if an attempt were being made to 
reproduce the biphasic cytotrophoblastic and 
syncytiotrophoblastic pattern (Fig. 14). Frank 
syncytial trophoblastic elements were often 
present focally (Fig. 15), and fully differen- 
Fig. io. Incidence of teratocarcinoma and seminoma corrected for Army population in each age group. 
Incidence of teratocarcinoma is practically constant, while that of seminoma rises with age. 
in mixed tumors. 
The striking resemblance to trophoblastic 
tissue of one common variety of embryonal 
carcinoma warrants special emphasis. The cells 
of such neoplasms showed prominent masses 
of chromatin against a clear nucleoplasm. The 
cytoplasm tended to be clear and granular, 
as in well glycogenated liver cells. Loss of 
distinct cell outlines and superimposition of 
nuclei in syncytial cytoplasmic masses was often 
seen, particularly in tumors growing chiefly as 
dated chorioepitheliomatous tissue was en- 
countered in about 6 per cent of embryonal 
carcinomas. The hemorrhagic necrosis of both 
neoplastic and infiltrated tissue and the vas- 
cular invasion which characterize chorio- 
epitheliomas were frequently seen in embryonal 
carcinomas. Recognition of the trophoblastic 
nature of many embryonal carcinomas (and 
the carcinomatous component of teratocar- 
cinomas) makes understandable the occurrence 
of secondary endocrine changes, such as 580 
The Military Surgeon—November, 1946 
Fig. ii. Embryonal carcinoma. Large irregular hyperchromatic nuclei of the anaplastic cells contrast with 
the regular pattern of the elements of the seminoma (see Fig. 2). X 550. 
Fig. 12. Papillary adenocarcinoma, a variety of embryonal carcinoma. X 250. 
Fig. 13. Reticulated type of embryonal carcinoma. X 70. 
gynecomastia, in association with tumors of 
this type. 
The differentiation of frankly teratoid 
structures within an embryonal carcinoma 
automatically placed such a neoplasm in the 
category of teratocardnoma, but certain pat- Tumors of the Testis 
581 
Fig. 14. Trophoblastic embryonal carcinoma. Cells have cytotrophoblastic appearance, and secondary 
cellular components seem to be appliqued against predominating elements. X 315. 
Fig. 15. Syncytial cells in a trophoblastic embryonal carcinoma, X 160. 
terns indicating early or minimal differentia- 
tion were recognizable in tumors classified as 
embryonal carcinomas. For example, conden- 
sation of mesenchyme (Fig. 16) around or 
next to the neoplastic epithelium in some em- 
bryonal carcinomas actually represented the 
beginnings of specialization and differentiation 
although it could easily be mistaken for an 
ordinary desmoplastic reaction. In other neo- 
plasms there were microcystic structures (Fig. 
17) bordered by regularly arranged cellular 
rows and arcs with the general conformation 
of early embryos, other indications of primitive 
differentiation; sometimes a trophoblastic 
cushion (Fig. 18) adjoined the embryonic 
disks. 
Chorioepithelioma 
Chorioepitheliomas have such a character- 
istic appearance that they are easily recognized 
under the microscope, and their nature may 
often be suspected grossly because they are 
strikingly hemorrhagic. These neoplasms typi- 
cally have two cellular components, and their 
arrangement duplicates the architecture of the 
placental villi. The combination of compactly 
grouped cytotrophoblastic cells and giant mul- 
tinucleated syncytial structures, which usually 
are arranged at the borders of the cellular 
masses, makes one of the most striking micro- 
scopic pictures in all pathology (Fig. 19). 
Chorioepitheliomas, partly because of the 
unusual morphologic appearance of classic 582 
The Military Surgeon—Novembery 1946 
Fig. i6. Condensation of mesenchyme about carcinomatous epithelium in embryonal carcinoma. X no. 
Fig. 17, Embryoid body in embryonal carcinoma. X 260. 
Fig. 18. Embryoid body and trophoblastic cushion in embryonal carcinoma. X 70. 
examples, have usually been placed in a 
special category. However, during the course 
of this study it became apparent that frankly 
trophoblastic elements occurred focally in 
many tumors of other types and that even in 
the absence of syncytial elements trophoblastic Tumors of the Testis 
583 
Fig. 19. Chorioepithelioma, showing syncytial cells covering masses of cytotrophoblast. X 145. 
Fig. 20. Chorioepithelioma with atypical reticular pattern. X 175. 
features were evident in neoplasms usually put 
in other categories, the embryonal carcinomas 
in particular. Pure primary chorioepitheliomas 
comprised only a small proportion of the en- 
tire group of testicular tumors (less than 0.4 
per cent). Focal chorioepithelioma was rarely 
observed in seminomas but was encountered 
in 12 per cent of the embryonal carcinomas 
and teratocarcinomas (6.4 per cent of the 
entire series). 584 
The Military Surgeon—Novemhery 1946 
Gynecomastia and related changes in the 
endocrine glands were associated as often with 
embryonal carcinoma or teratocarcinoma as 
with chorioepithelioma. The fact that chorionic 
gonadotropin is often present in the urine of 
patients with “embryonal adenocarcinoma” 
provides physiologic substantiation for the 
morphologic observation that many embryonal 
carcinomas are trophoblastic tumors. 
not setting chorioepitheliomas apart from the 
embryonal carcinomas, there is no objection to 
using the designation for tumors clearly of 
that type. 
The trophoblastic potency of embryonic 
cells is expressed in the chorioepitheliomas and 
other trophoblastic tumors, while the comple- 
mentary potency to form somatic structures is 
expressed in the teratoid tumors. 
Fig. 21. Teratoma. X 8 
Even unequivocal chorioepitheliomas often 
differed in appearance from the classic type. 
For example, the cells were sometimes ar- 
ranged in a loose reticulum or reticulated 
syncytium (Fig. 20), with clumps and strands 
of eosinophilic fibrinoid material in the meshes 
of the network. Since such patterns were rec- 
ognized in both embryonal carcinomas and 
chorioepitheliomas, the concept of chorioepi- 
theliomas as a subvariety of carcinoma rather 
than as a separate entity is further supported. 
Despite the practice followed in this study, of 
T ERATOMA 
The teratomas of the testis, with their mad 
array of ordered and disordered structures, 
have fascinated pathologists for generations. 
The range of differentiated structures, both 
recognizable and unrecognizable, is so great 
and their admixture so complicated that the 
completely realized and the partially unfolded 
patterns are rarely duplicated exactly. If a 
teratoma has no histologically recognizable 
malignant components, the qualified designa- 
tion “adult” is justified; the term “benign” Tumors of the Testis 
585 
should never be used, because metastasis of 
testicular tumors which appeared to be only 
adult teratomas has occurred. 
Teratomas (Fig. 21) are characterized by 
the presence of epithelial masses, glands and 
cysts, many of which are organized in com- 
bination with undifferentiated or specialized 
mesenchymal tissues, such as cartilage, into 
structural units. The components abut on one 
rarely encountered. Squamous epithelial nests 
and cysts, often keratinizing, were frequent, 
but glandular appendages and adult hair were 
rare. Epidermoid cysts outnumbered dermoid 
10 to 1. Nests of noncornifying epithelium 
resembling the stomodeal lining were common, 
and in a few tumors early dental germs were 
identified; fully formed teeth were never seen. 
Masses of neuroepithelium often resembled 
Fig. 22. Masses of neuroepithelium resembling fetal neural tubes in teratoma. X 70, 
Fig. 23. Mixed tumor of salivary gland in teratoma. X no. 
another, intermingle, show transitions and 
generally form a complex of confusing con- 
stellations in which, nevertheless, a semblance 
of organization often can be discerned. At- 
tempts have been made to identify specific 
architectural patterns as homologues of normal 
parts of the body, but it is generally admitted 
that it is easy to force such analogies too far. 
Although primitive and undeveloped “pri- 
mordia” of many kinds were observed, the 
corresponding highly specialized tissues or dif- 
ferentiated derivatives and their organs were 
distorted and complexly fused neural tubes 
(Fig. 22). Occasionally the proliferation of 
neuroepithelium resulted in the formation of 
a neuroblastoma; some neuroblastic metastases 
occurred. Illy formed glial nests were far more 
common than the occasional organized bits of 
neural tissue which included ganglion cells. 
In a single tumor meninges had formed on the 
surface of a brain-like mass. 
Enteric glands and cysts which were lined 
by columnar mucinous epithelium and often 
had organized coats of smooth muscle were 586 
The Military Surgeon—November, 1946 
Fig. 24. Teratocarcinoma, showing intermingling of embryonal carcinoma and differentiated teratomatous 
structures. X 15. 
Fig. 25. Teratocarcinoma composed of discrete teratoma and seminoma. X 10. 
regularly observed. A few auxiliary glandular 
structures were recognized, but differentiated 
secretory elements, such as parietal or zymo- 
genic cells, were almost never present. It is 
worth noting that tissue clearly similar to that 
of mixed tumors of salivary glands (Fig. 23) 
formed part of one and nearly all of a second 
neoplasm. In many regions of another tumor Tumors of the Testis 
587 
there was a papillary epithelial growth com- 
posed of strongly eosinophilic “onkocytes.” 
Frank thyroid or pancreatic tissue was never 
encountered, nor was liver. Respiratory epi- 
thelium was not infrequent; it was sometimes 
intimately associated with lymphoid tissue and 
occasionally with cartilaginous arches, but 
clearly bronchial or pulmonary structures were 
not identified. Similarly, although transitional 
epithelium of the urinary type was often found, 
prostatic acini were seen only once and an 
were not observed. An angioendotheliomatous 
type of proliferation appeared twice. 
The term “teratoma” is loosely used in this 
report to designate any growth in which dif- 
ferentiated “adult” structures were recognized. 
There appeared to be no reason for distin- 
guishing between solid and cystic teratomas. 
For convenience, dermoid and epidermoid 
cysts were included in the group of adult 
teratomas, of which they made up about 16 
per cent. 
Fig. 26. Teratocarcinoma, showing- differentiation of squamous nest and mucinous gland in situ. X 60. 
organized structure resembling vesical wall 
only twice. 
Embryonic mesenchyme was commonly 
organized into special structures surrounding 
or satellite to epithelial tissue and was often 
differentiated into smooth muscle and cartilage. 
Skeletal or cardiac muscle and bone, on the 
other hand, were only occasionally identified. 
Mucinous connective tissue and young fat 
cells, sometimes arranged in lobules, were 
often present; but adult adipose tissue was 
rare. Lymphoid tissue was frequently abun- 
dant, but hematopoietic tissue or lymph nodes 
T ERATOCARCINOMA 
The frequency with which the monocellular 
tumors occur in intimate and complicated as- 
sociation with teratomas is partially responsible 
for the array of complex classifications in the 
literature. A new term, “teratocarcinoma,” is 
proposed for the large group of pleomorphic 
tumors in which both differentiated teratoid 
structures and histologically malignant ele- 
ments were present (Fig. 24). Most terato- 
carcinomas (77.5 per cent) were mixtures of 
teratoma, embryonal carcinoma and/or 
chorioepithelioma; but 5 per cent were com- 588 
The Military Surgeon—November, 1946 
posed of teratoma and seminoma, and 15 per 
cent had seminomatous, embryonal carci- 
nomatous and/or chorioepitheliomatous com- 
ponents. The remaining 2.5 per cent formed 
a heterogeneous group which included the 
unusual combination of teratoma and neuro- 
differentiation occurred more often in terato- 
carcinomas (15 per cent) than in embryonal 
carcinomas (6 per cent). The formation of 
somatic and trophoblastic structures was al- 
most exclusively a property of teratocar- 
cinomas and embryonal carcinomas; such dif- 
ferentiation occurred in rare predominantly 
seminomatous tumors, but the intimate inter- 
mingling of teratoid and carcinomatous ele- 
ments which characterizes most teratocar- 
cinomas was not present. 
The results of this study do not support 
the conventional theory that malignant tu- 
mors of the testis usually arise from teratomas 
but the series did include a few teratoid 
tumors in which carcinomas, rhabdomyo- 
sarcomas or neuroblastomas developed. 
Mortality Rates and Incidence 
of Metastasis 
Two important points must be borne in 
mind in evaluating the significance of the 
mortality rates and incidence of metastasis. 
First, no attempt has been made in this report 
to evaluate the effects of surgical or roentgen- 
ologic therapy. Second, follow-up information 
regarding the presence of metastases and the 
results of autopsy is limited for the most part 
to those patients kept under observation or 
treatment in Army hospials for several months 
after orchiectomy. Few patients were followed 
up for more than a year. Most of the men 
were transferred to the charge of the Veterans 
Administration or to civilian medical care, 
and comprehensive data from these sources 
have not yet been compiled. 
Evidence of metastasis, such as palpable ab- 
dominal masses or roentgenologically demon- 
strable pulmonary nodules, often established 
by histologic examination, was obtained in 
about 10 per cent of the cases; 13 per cent 
have terminated fatally. Regardless of what 
the long-term results may prove to be, certain 
conclusions regarding the immediate prognosis 
for the different types of tumor are inescapable. 
The number of cases in which autopsy was 
performed and metastasis occurred is given 
in table I. 
The disparity between the figures for 
Table 1 
Mortality Rates and Incidence of 
Metastasis of Different Types 
of Testicular Tumors 
Type of 
Tumor 
Total 
no. 
of 
Cases 
No. of 
Deaths 
from 
Tumor 
No. of 
Living 
Patients 
with 
Metastasis 
Embryonal 
carcinoma 
171 
47(27-5%) 
26(15%) 
Seminoma 
319 
8( 2-5%) 
20 (6.5%) 
T eratocarcinoma 
319 
55(17%) 
35(n%) 
Teratoma 
68 
10(15%) 
9(i3%) 
epithelioma and the even rarer association of 
teratoma and sarcoma. The constituents of 
teratocarcinomas were usually intimately in- 
termingled but occasionally were segregated, 
particularly when the tumors were made up 
of teratoma and seminoma (Fig. 25). 
The explanation usually advanced for the 
presence of embryonal carcinoma in terato- 
carcinomas is that malignant transformation 
of an “adult” structural component of a 
teratoma has taken place. If this theory is 
tenable, one would expect to encounter 
squamous cell carcinomas and mucinous 
adenocarcinomas in teratocarcinomas because 
squamous epithelial nests and enteric glands 
and cysts are so frequently present in teratomas. 
Actually, the carcinomatous component was 
usually a typical monocellular embryonal 
carcinoma and was almost never of a special- 
ized epithelial type. 
The glands and other differentiated epi- 
thelial structures in teratocarcinomas were 
intimately associated with both carcinomatous 
epithelium and organized mesenchymal com- 
ponents in such a way that they appeared to 
have differentiated in situ, probably from the 
carcinomatous cells (Fig. 26). 
Focal or disseminated chorioepitheliomatous Tumors of the Testis 
589 
seminomas and embryonal carcinomas is con- 
sistent with the differences in histologic be- 
havior of the 2 tumors. The carcinomas infil- 
trated adjacent tissues and adnexal structures 
aggressively and showed hemorrhage, necrosis 
and vascular invasion. They were of the same 
order of malignancy as chorioepitheliomas. 
The relatively benign microscopic appearance 
of seminomas has already been mentioned. 
The difference in the relation of mortality 
rate to incidence of metastasis for seminomas 
(1:2.5) and for embryonal carcinomas (2:1) 
indicates that seminomas disseminated more 
slowly. Metastasis of seminomas was usually 
restricted to the retroperitoneal tissues or the 
peritoneal linings; distant or parenchymal 
metastases were rare, while they were the rule 
for embryonal carcinomas. The mortality 
rates and incidence of metastasis for terato- 
carcinomas were not as high as for embryonal 
carcinomas but were much higher than for 
seminomas. It is of exceptional interest that 
the death rate and incidence of metastasis 
were practically the same for adult teratomas 
as for teratocarcinomas. 
There are several possible interpretations of 
the malignant behavior of the “benign” adult 
teratoma. First, it is possible that the testicular 
neoplasm was only one of several teratoid 
tumors which developed multicentrically along 
the course of the genitourinary tract and that 
the truly primary malignant neoplasm was 
extragenital. Secondly, it is conceivable that 
adult differentiated teratoid tissue had invaded 
blood vessels and metastasized as such. A third 
and frequently advanced explanation is that 
malignant foci in the primary tumor were 
overlooked, especially if its size was incom- 
patible with study by serial section. However, 
malignant foci were found in no metastasizing 
adult teratomas of this series despite careful 
study by the multiple block technic, which 
consists of sectioning and embedding of the 
entire tumor and examination of slides from 
each block. The fourth, and most satisfactory, 
theory is that a teratocarcinoma matured into 
an adult teratoma after malignant embryonal 
cells had already been disseminated to other 
parts of the body. The similarities in the struc- 
ture of the metastases of teratomas and tera- 
tocarcinomas and the natural history of differ- 
entiating and evolving teratocarcinomas are 
consistent with the suggestion that adult 
teratomas represent matured teratocarcinomas. 
Structure of Metastases 
Comparison of the histologic compositions 
of the primary tumor and the metastases 
yielded considerable information regarding the 
potencies of the various cell types and the 
dynamic balance between undifferentiated ele- 
ments and teratoid structures. Ninety per cent 
Table 2 
Number of Metastases of Testicular Tumors 
Showing Chorioepithelioma 
Metastases 
Pure 
Focal 
No 
Primary 
Chorio- 
Chorio- 
Chorio- 
Tumor 
epi- 
epi- 
epi- 
thelioma 
thelioma 
thelioma 
Pure chorioepithelioma 
4 
o 
o 
Focal chorioepithelioma 
5 
4 
4 
No chorioepithelioma 
n 
ii 
o 
of the embryonal carcinomas metastasized as 
embryonal carcinomas or chorioepitheliomas; 
teratoid structures were rare. Forty per cent 
of the metastases of teratocarcinomas consisted 
of embryonal carcinomas or chorioepitheliomas, 
and in 60 per cent there were frank teratoid 
structures. Half of the secondary deposits of 
adult teratomas were of the monocellular 
variety, and half were teratoid. 
All degrees of differentiation from slight 
condensation (Fig. 27) and “specialization” 
of the stroma about neoplastic epithelium to 
complex teratomatous structures (Fig. 28) 
were encountered. The appearance of muscle 
(Fig. 29) or cartilage (Fig. 30) in con- 
densed mesenchyme and the development of 
specialized glands within a mass of undifferen- 
tiated carcinomatous tissue supported the con- 
cept of differentiation in situ rather than the 
idea that a complex teratoid clump of tissue 
had embolized and then grown. 
In addition to the “unfolding” of teratoid 
potentialities the metastases of embryonal car- 590 
The Military Surgeon—November, 1946 
Fig. 27. Pulmonary metastasis of teratocarcinoma showing- condensation of mesenchymal elements about 
neoplastic epithelium. X 70. 
Fig. 28. Complex teratoid metastasis in intermediate stag-e of differentiation. X 85. 
cinomas, teratocarcinomas and teratomas may 
display a complementary development of 
chorioepitheliomatous tissue. Table 2 shows 
that nearly twice as many of the metastases 
which exhibited chorioepitheliomatous struc- 
tures arose from primary tumors containing Tumors of the Testis 
591 
no chorioepithelioma as from pure chorio- 
epitheliomas or neoplasms containing focal 
chorioepithelioma. Moreover, metastases of 
only 4 chorioepitheliomatous primary tumors 
failed to manifest chorioepitheliomatous struc- 
tures. 
While only 0.4 per cent of the primary 
short time thus far, it is reasonable to assume 
that only the most malignant type of neoplasm 
will already have caused death and that the 
number of deaths attributable to this variety 
of tumor will decrease with the lapse of time. 
Experience supports this assumption. A survey 
completed 4 months before preparation of this 
Fig. 29. Muscle fibers in specialized mesenchyme of teratocarcinomatous metastasis. X 130. 
Fig. 30. Cartilage in specialized mesenchyme of teratocarcinomatous metastasis. X 160. 
testicular tumors were pure chorioepitheliomas 
and only 6.4 per cent showed focal chorio- 
epitheliomatous tissue, 27 per cent of all the 
metastases in cases which terminated fatally 
contained chorioepitheliomatous elements. 
These figures indicate not only the tropho- 
blastic potencies of the teratoid and car- 
cinomatous tumors but the high degree of 
malignancy of chorioepitheliomas. Since the 
patients have been followed up a relatively 
report disclosed that 34 per cent of the 
metastases in cases terminating fatally 
exhibited either pure or focal chorioepithelioma 
as contrasted with the current figure of 27 
per cent. 
Too few patients with seminoma have 
come to autopsy for conclusions to be drawn 
regarding the significance of the different 
types of metastases. Most of the secondary 
growths studied were in resected retro- 592 
The Military Surgeon—November, 1946 
peritoneal nodes; they were largely semi- 
nomatous, but a few showed embryonal car- 
cinoma or teratocarcinoma. Seminomatous 
tissue was rarely encountered in the metastases 
of those teratocarcinomas and embryonal car- 
cinomas which were associated with 
seminomas. 
In general, the metastases in a given case 
were uniform, though they occasionally dis- 
played varying structures. For example, some 
teratocarcinomas metastasized as teratocar- 
tissue is present in some mixed neoplasms (see 
dotted lines in figure 31). They occur in 
relatively older men and run a relatively more 
benign course than do the embryonal car- 
cinomas and teratoid tumors. The observation 
that 80 per cent of the neoplasms in un- 
descended testes encountered in this series 
were seminomas underlines their biologic indi- 
viduality and gives rise to interesting specula- 
tions regarding their genesis. 
The resemblance between seminoma cells 
Fig. 31. Schematic presentation of lines of development of testicular tumors. 
cinomas to one organ and as embryonal car- 
cinomas or chorioepitheliomas to others. Some 
differentiated teratoid metastases also con- 
tained trophoblastic tissue or isolated syncytial 
cells. In other words, the metastases, like the 
primary testicular tumors, showed 4 basic 
neoplastic patterns which occurred singly or in 
combination. 
Morphogenesis 
Seminomas are distinct biologic entities 
which, for the most part, are pure monocellu- 
lar growths devoid of teratoid structures in all 
stages of development, although seminomatous 
and some of the elements in seminiferous 
tubules has been taken as evidence that 
seminomas arise from the cells (spermato- 
gonia) of the testicular tubules themselves. 
Although the appearance of tubules invaded 
and partially replaced by neoplastic cells often 
gave such an impression, careful comparison 
revealed that the similarity between neoplastic 
seminoma cells and spermatogonia was more 
superficial than real and that the apparent 
formation of tubular structures was equally 
illusory. Serial sections of a few small 
seminomas in an early stage of development, 
found incidentally in the hilus of the testicle Tumors of the Testis 
593 
at autopsy, yielded no evidence supporting the 
theory of tubular origin. The term seminoma 
is therefore inappropriate, but it has the ad- 
vantage of common usage. The designation 
“disgerminoma,” which is employed for the 
identical tumor occurring in the ovary, or 
“germinoma” would be more satisfactory than 
“seminoma.” The possibility should be investi- 
gated that such tumors represent neoplasms 
of the so-called primordial germ cells. 
The cell of origin of embryonal carcinomas 
and teratoid tumors has the capacity to form 
both somatic and trophoblastic tissues, as 
might be expected of a primitive embryonic 
cell. The lines along which these neoplasms 
evolve are indicated in figure 31. Differentia- 
tion and specialization may take place focally 
within the primary growth or may be mani- 
fest only in the metastases. The primary and 
secondary tumors may evolve in different 
directions; for example, one may form teratoid 
tissue while the other displays trophoblastic 
tendencies. Embryonal carcinomas and terato- 
carcinomas are dynamic, not static; apprecia- 
tion of the fact that their structural com- 
ponents keep changing and evolving along the 
lines indicated in figure 31 makes it possible 
to explain the complex intermingled patterns 
of some neoplasms. The frequency with which 
careful study discloses teratoid foci in new 
growths in the male gonad is understandable 
in view of the latent potentialities of the neo- 
plastic cells, but it does not mean that all 
tumors of the testicle arise from teratomas. 
The new growths of the male gonad ap- 
pear exceptional in that they do not reproduce 
testicular tissue (unless the theory of the tubu- 
lar nature of seminomas is accepted), while 
most other neoplasms reflect the architecture 
of the tissue in which they originate. They 
express, in distorted fashion, the latent un- 
limited potencies of the embryonic germ cells 
from which they arise, while other tumors can 
make manifest only the limited potencies of 
their somatic cells of origin. It is important 
to determine whether a common oncologic 
principle underlies all new growths or whether 
the testicular neoplasms (and other teratoid 
growths) are truly unique. 
Summary 
Study of 922 tumors of the testis established 
that 96 per cent of such new growths can be 
classified as seminoma, embryonal carcinoma, 
teratoma or teratocarcinoma. 
Seminomas and embryonal carcinomas are 
distinct tumors which differ not only in funda- 
mental cell type but in biologic behavior and 
prognosis. Chorioepitheliomas are considered 
a subvariety of embryonal carcinomas. The 
term teratocarcinoma is applied to those tu- 
mors which show both differentiated teratoid 
structures and malignant elements and are 
thought to result from teratoid differentiation 
in embryonal carcinomas. Monocellular testic- 
ular neoplasms do not ordinarily originate from 
preexisting teratomas. 
Virtually all embryonal carcinomas metasta- 
size as monocellular embryonal carcinomas, 
but choriomatous characteristics may be evi- 
dent in the metastases of embryonal carcinomas 
or teratocarcinomas even when they are not 
manifest in the primary tumor. Roughly half 
of the teratoid neoplasms which metastasize 
give rise to growths with teratocarcinomatous 
structures and half to pure embryonal 
carcinomas. 
Immediate prognosis is bad for embryonal 
carcinomas and chorioepitheliomas and poor 
for teratocarcinomas and adult teratomas, 
which should be regarded as matured terato- 
carcinomas. Seminomas, in comparison with 
the other testicular tumors, have a good im- 
mediate prognosis. 
The architecture of testicular tissue is not 
reproduced in seminomas, and the neoplastic 
cells do not resemble spermatogonia. Semi- 
nomas are probably tumors of primordial germ 
cells and should be called germinomas. Em- 
bryonal carcinomas and teratoid tumors, 
which are composed of evolving and differen- 
tiating somatic and trophoblastic tissues, are 
neoplastic expressions of the unlimited poten- 
cies of embryonic cells. PRIMARY INTRACRANIAL NEOPLASMS IN MILITARY 
AGE GROUP—WORLD WAR II 
MAJOR WARREN A. BENNETT, Medical Corps 
(With one hundred and six illustrations) 
Introduction 
THE great amount of pathologic ma- 
terial sent to the Army Institute of 
Pathology during World War II has 
afforded the unusual opportunity of studying a 
series of intracranial neoplasms in a selected 
group, namely in Army personnel, between the 
ages of 18 and 38 inclusive. 
Because the statistics presented in this paper 
have been gathered in so relatively short a 
period and from a sharply delimited age 
group, correlations with previous reports can- 
not be made, but comparisons with statistics 
from series studied by other authors may be 
of interest. 
A majority of the tumors in this series oc- 
curred in persons examined and declared to be 
in good health upon entering the Army. No 
history of signs or symptoms of intracranial 
neoplasm was obtained at the initial examina- 
tion; relevant symptoms were recorded in 
only a few cases. It is safe to say that most of 
the symptoms became evident during the 
relatively short period in which the person was 
in active service. 
Although intracranial neoplasms in persons 
over 38 years of age have not been included 
in the present study, when a significant number 
appears in any category in the older group 
they are briefly mentioned. 
The material was sent to the Army In- 
stitute of Pathology from hospitals and stations 
all over the world. The study of gross speci- 
mens was necessarily limited, because in some 
instances only blocks of tissue were submitted. 
The microscopic examination was carried out 
by the officers in the Section on Neuropathology 
of the Army Institute of Pathology and only 
cases with verified diagnoses are included in 
this series. The clinical records which accom- 
panied the specimens were, on the whole, ex- 
cellent; some, however, lacked certain data 
desired for our study. 
The primary purpose of this paper is to 
present the available information on intra- 
cranial neoplasms in the military age group. 
It is in no sense a definitive clinicopathologic 
study; but rather a survey to indicate the ex- 
tent of the data and material which has ac- 
cumulated at the Army Institute of Pathology 
during the War, and which is available for 
detailed study of the problems of neuroecto- 
dermal and other intracranial neoplasms. 
The divisions of this series are based on a 
morphologic classification as well as on patho- 
genesis. Only primary intracranial neoplasms 
have been included; all aneurysms, granu- 
lomas, cranial lesions involving the brain, and 
metastatic intracranial tumors have been 
omitted, as well as those of spinal cord and 
peripheral nerves. Reference will be made to 
the classifications of Cushing;7 Baker;4 
Bailey;2 and Elvidge, Penfield and Cone,9 
for purposes of comparison. We have adopted 
Cushing’s classification for our study, since it 
is the one most generally accepted. 
Globus and Kuhlenbeck10’11 have presented 
a variant classification to which frequent refer- 
ence is made in this paper. Their classification 
is based on the histogenesis of the neuroecto- 
dermal elements derived from the medullary 
epithelium and the differentiation along two 
main lines, the spongioblastic and the neuro- 
blastic. The derivation of bipotential mother 
cells from the medullary epithelium is also as- 
sumed. These cells, often called medulloblasts, 
may differentiate into either neuroblasts or 
spongioblasts. “It is the degree of maturity of 
these elements and their evolution into the 
several glial or neuronal cell forms in a varying 
relative ratio which determine the morphologic 
character and other biologic features of a given 
tumor of neuroectodermal origin.” A com- 
parative analysis of Cushing’s terminology and 
that of Globus, Kuhlenbeck and others clarifies 
the points of agreement and divergence. Primary Intracranial Neoplasms in Military Age Group 
595 
Cushing) Bailey, and 
Others 
Globus, Kuhlenbeck, and 
Others 
Glioblastoma multiforme 
Spongioblastoma 
multiforme 
Astrocytoma 
Glioma 
Astroblastoma 
Ependymoblastoma 
Ependymoma 
Transitional glioma 
Spongioblastoma 
ependymale 
Ependymoma cellulare 
Papilloma ependymale 
Papilloma chorioideum 
Oligodendroglioma 
Oligodendroglioma 
Ganglioneuroma 
Ganglioglioma 
Spongioneuroblastoma 
Transitional 
glioneuroma 
Glioneuroma 
Spongioblastoma polare 
Spongioblastoma polare 
Pinealoma 
Pinealoma 
inous, and usually of increased consistency. 
The cut surface varies from gray-pink to 
bright red due to the degree and age of the 
hemorrhage and to the amount of necrosis. 
There is no definite capsule, but the 
edematous brain about the tumor provides an 
indefinite line of demarcation. The cortex is 
invaded causing distortion of the brain (Fig. 
1-4). 
As the name implies this tumor assumes 
cellular patterns which are variants of the 
same process. The morphologic characteristics 
are: (1) pleomorphism of cells, (2) areas of 
necrosis and hemorrhage, (3) endothelial pro- 
liferation of vessels (Fig. 5 and 6), (4) giant 
cells, and (5) frequent palisading around areas 
of necrosis. 
Depending on the amount and character of 
the cytoplasm, the component cells may re- 
semble astrocytes, unipolar and bipolar spongio- 
blasts, and neuroblasts. The nuclei vary in 
size, shape, and chromatin content. Giant cells 
with single or multiple nuclei are characteris- 
tic of the tumor. The pleomorphic cells do 
not usually have a particular type of arrange- 
ment although around areas of necrosis they 
tend to assume a palisade pattern. One must 
remember that various fields in these tumors 
may show different arrangements and 
cells; thus areas of small astrocyte-like cells 
may adjoin others containing numerous neuro- 
blastic elements with many multinucleated 
giant cells. The pleomorphic cells may be 
round, fusiform, or piriform. They do not 
form reticulum or connective tissue (Fig. 7). 
The cytoplasm is scanty and may appear to 
surround an almost naked nucleus or to form 
long processes. The nuclei are said to divide 
both by mitosis and amitosis. Mitotic figures 
are common and many are seen to be atypical. 
The giant cells have numerous nuclei and 
granular cytoplasm (Fig. 8). Spongioblasts 
and astrocytes represent the two cells of origin 
of these giant cells. The perivascular strands 
of connective tissue are prominent and persist 
after necrosis. 
One of the most characteristic features of 
glioblastoma multiforme is vascular prolifera- 
tion. An actual increase in the number of 
In the period from Pearl Harbor to V-J 
Day material from 84,615 cases was received 
at the Army Institute of Pathology; 543 in- 
tracranial tumors made up 0.64 per cent of 
this total. Ninety-seven of the 543 were not 
included because the age of the patient was 
under 18 or over 38, or because the informa- 
tion was inadequate. Thus 446 primary intra- 
cranial neoplasms have been studied. Three 
hundred and eighty-one of these were classified 
as gliomas (See Table 1, p. 647). 
Gliomas 
Glioblastoma Multi]orme: One hundred 
and two intracranial neoplasms classified as 
glioblastoma multiforme were found in the 
series of 446 from the military age group. 
There was a total of 134 when there was 
no limitation of age. It will be seen that glio- 
blastoma multiforme represented 22.9 per 
cent of all tumors in this series and 36.6 per 
cent of the gliomas. Ninety-nine of these 
tumors were in males; 93 were in white pa- 
tients, 8 in Negroes, and I in an Okinawan. 
The ages of the patients averaged 27.6 years. 
Glioblastoma multiforme is a soft, gray-red, 
hemorrhagic, necrotic tumor, usually contain- 
ing numerous small degenerative cysts. It is 
very vascular with ill-defined borders. The 
surrounding brain is gray, glistening, gelat- 596 
The Military Surgeon—November, 1946 
Fig. i. Glioblastoma multiforme extending across the midline within the corpus callosum. AIP Neg. 78764. 
Fig. 2. Glioblastoma multiforme projecting into the lateral ventricle. Note the necrosis and hemorrhage. 
AIP Neg. 89024. Primary Intracranial Neoplasms in Military Age Group 
597 
Fig, 3. Glioblastoma multiforme: Inferior surface of brain showing the tumor at the right of brain stem. 
AIP Neg. 89293. 
Fig, 4. Longitudinal section through brain stem showing extent of lesion in Fig. 3. AIP Neg. 89294. 
Fig, 5. Glioblastoma multiforme: marked vascular proliferation. X 200. AIP Neg. 96243. 
Fig. 6. Intimal proliferation in vessels in glioblastoma multiforme. X 425. AIP Neg. 96240. 598 
The Military Surgeon—November, 1946 
Fig. 7. Glioblastoma multiforme: marked pleomorphism with a few giant cells. X 205, AIP Neg. 96244. 
Fig. 8. Glioblastoma multiforme: numerous giant cells with pleomorphic spongioblasts. X 225. AIP Neg. 
96237. 
Fig. 9. Glioblastoma multiforme with foci of necrosis and pseudopalisading of surrounding pleomorphic 
cells. X 175. AIP Neg. 96236. Primary Intracranial Neoplasms in Military Age Group 
599 
vessels is apparent, and the cells of both ad- 
ventitia and intima undergo proliferation. At 
a later stage the walls of the vessels may be- 
come hyalinized. The rupture of some vessels, 
and obliteration by endothelial proliferation 
with subsequent infarction, accounts for the 
frequent hemorrhages and areas of necrosis 
seen in this tumor (Fig. 9). 
The vascular processes have two distinct 
phases: one, new formation of vessels with 
many thin-walled capillaries, the other, pro- 
liferation of the cells of the intima and 
adventitia. The lumens of the vessels may be 
filled with cells arising from the intima. At 
times formation of papillary masses traversed by 
numerous vessels gives rise to a “glomeruloid” 
appearance. The nature of the vascular 
changes is uncertain, some consider them a re- 
action to the tumor, others part of the neo- 
plastic process (Fig. 6). 
The brain surrounding the tumors under- 
goes a slight gliosis characterized by large 
plump astrocytes and proliferation of the glial 
mesh work. Satellitosis may be marked. 
Elvidge, Penfield and Cone9 have outlined 
the histologic features of glioblastoma multi- 
forme as follows: 
Constant findings: 
1. Type cell, spongioblasts 
2. Plump astrocytes 
3. Tumor giant cells 
4. Areas of necrosis 
5. Proliferation of vascular endothelium 
6. Mitosis 
7. Fibroblastic overgrowth 
Probably constant findings: 
1. Increased vascularity 
2. Adventitial proliferation 
Frequent findings: 
1. Cyst formation, small 
2. Neighborhood glial reaction, satellitosis 
The glioblastomas usually occur in the tel- 
encephalon, involving the cerebral cortex. The 
locations of those in this series are as shown at 
the top of the next column. 
More than one lobe was involved in 10 
cases. Seventy-three of the 102 glioblastomas 
extended into neighboring areas, but the site 
Side Not 
Right 
Left 
Mentioned 
Frontal lobe 
20 
9 
Parietal lobe 
16 
10 
2 
Temporal lobe 
7 
12 
Occipital lobe 
2 
1 
Cerebellum 
1 
1 
1 
3rd ventricle 
4 
4th ventricle 
2 
Pons 
4 
4 
Midbrain 
1 
Quadrigeminal Plate 
i 
Thalamus 
1 
Left cerebrum 
1 
Corpus callosum 
1 
Medulla 
2 
Basal ganglion 
2 
5 
Total: 
65 
42 
3 
of maximum growth is as given above. In 2 
cases implants were present in the ventricle 
some distance from the original growth. In 
one other the white matter of the entire left 
hemisphere was involved. 
The duration from onset of symptoms to 
diagnosis is as follows: 
Number 
Per Cent 
Less than i month 
40 
39-2 
1-2 months 
14 
13.7 
3-5 months 
25 
24.5 
6-12 months 
12 
xi.8 
1-2 years 
9 
8.8 
Unknown 
2 
2.0 
102 
100.0 
In 84 per cent of cases symptoms had been 
present for less than 6 months; in 54 per cent, 
less than 2 months. The duration of symptoms 
of patients from 18 to 20 years of age was 
especially short not exceeding 6 weeks in any 
case. Previous trauma was reported in 10 of 
the 102 cases. 
Papilledema took precedence over headache 
as the most frequent symptom; it occurred in 
90 cases, headache in 81. There was a multi- 
plicity of other symptoms including uncon- 
sciousness in 19 cases; hemiparesis on the 
right in 15, on the left in 6; hemiplegia of the 
right in 12, of the left in 2; nausea and vomit- 
ing in 16; mental changes in 14; diplopia in 
14; failing vision in 12; dizziness in 12, diffi- 600 
The Military Surgeon—November, 1946 
Fig. io. Astrocytoma frontal lobe with no distinct line of demarcation. AIP Neg. 77738. 
Fig. ii. Astrocytoma in 4th ventricle containing hemorrhage. AIP Neg. 96368. 
Fig. 12. Astrocytoma showing small stellate cells with short processes. X 250. AIP Neg. 96224. 
Fig. 13. Astrocytoma, fibrillary type. X 405. AIP Neg. 96229. Primary Intracranial Neoplasms in Military Age Group 
601 
Fig. 14. Astrocytoma in aqueduct of Sylvius. X 25. AIP Neg. 93355. 
Fig. 15. Astrocytoma showing spindle cells. Same as Fig. 14. X 150. AIP Neg. 93353. 
Fig. 16. Astrocytoma characterized by plump “gemistocytic” astrocytes. X 250. AIP Neg. 96280. 
Fig. 17. “Gemistocytic” astrocytoma. Note the persistence of some processes. X 605. AIP Neg. 96279. 602 
The Military Surgeon—Novembery 1946 
culty in speech in 7; nervousness in 6; tingling 
in 5; lethargy in 5; stupor in 5; numbness, 
general weakness, syncope, aphasia, drowsiness, 
and convulsions, each observed in 4; coma in 
3; blindness, ataxia, difficulty in swallowing 
and general tremor in 2 each, and quadriplegia 
in 1. No symptoms were noted in 5 cases, and 
sudden death occurred in 5 others. 
Craniotomy was performed in 76 cases; in 
the remaining 26 there was no operation. 
Ventriculograms were done in 53 and enceph- 
alograms in 10. Only 22 of the patients were 
able to leave the hospital after craniotomy. 
Four deaths occurred after spinal puncture, 6 
after ventriculogram. Sixteen deaths occurred 
the day of the operation; 14 from 2 days to 3 
months after admission without operation. 
The intervals between operation and death 
are listed: 
Negro, and I Chinese; the average age was 
26.5 years. 
The gross appearance of the lesion is that of 
an infiltrating tumor containing foci of de- 
generation with formation of cysts. The solid 
variety of astrocytoma is not sharply demar- 
cated; about two-thirds are of this kind; they 
are gray-white, firm, rubbery and slightly 
elevated. The cystic astrocytoma usually has 
a firm, white, rubbery mural nodule in the 
periphery of the cyst; in spite of its infiltrative 
nature it appears grossly to be well circum- 
scribed (Fig. 10 and 11). The rubbery con- 
sistency of astrocytomas is due to the concen- 
tration of neuroglia fibers. 
Microscopically the astrocytomas are classi- 
fied as fibrillary, protoplasmic, pilocytic (piloid 
or hair-like), gemistocytic, or diffuse. Cour- 
ville6 described only two varieties, fibrillary and 
protoplasmic. The fibrillary astrocytoma is 
composed of widely separated cell bodies with 
long intertwining processes (Fig. 12). The 
protoplasmic variety is made up of cells with 
lack of detail in the intercellular substance. In 
the pilocytic variety the processes are elongated 
and “hair-like;” the cells are widely separated, 
and the meshwork, in which the pilocytic fibrils 
run in parallel bundles, is less dense (Fig. 13- 
15). Cysts are frequently seen with liquefac- 
tion necrosis of fibrils. The gemistocytic, or 
bloated cell variety, contains large plump 
eosinophilic cells (with hematoxylin-eosin 
stain) (Fig. 16 and 17). The fibrils of the 
gemistocytic cells are short; the nucleus is 
large, and the cell body is filled with homo- 
geneous cytoplasm. An occasional cell may 
have multiple nuclei. Gemistocytic cells may 
occur in any form of astrocytoma and are 
frequently seen in reactive gliosis. The diffuse 
astrocytoma is composed of numerous small 
nuclei of equal size scattered diffusely through- 
out the brain tissue. The cells are stellate and 
appear to be in various stages of development 
from the astroblast to the mature astrocyte. 
This tumor has no boundaries (Fig. 18-23). 
Astrocytomas may develop anywhere 
throughout the brain where glial supportive 
elements are present. The locations of 63 as- 
trocytomas were as follows: 
Deaths 
Number 
Per Cent 
On table 
1 
2.0 
Less than i day 
6 
12.0 
1-7 days 
*9 
38.0 
2-4 weeks 
7 
14.0 
2-5 months 
11 
22.0 
6-8 months 
6 
12.0 
5° 
100.0 
Multiple operations were performed in 6 
cases. 
Irradiation therapy was given 10 of the 
23 patients who are alive at this time.* 
The clinical diagnosis varied, as in all other 
intracranial neoplasms, but “brain tumor” was 
specified in 83 cases. Cerebrovascular accident 
was recorded in 5, brain abscess in 3, enceph- 
alitis in 2, metastatic lesion, tuberculous men- 
ingitis, acute sclerosis, psychosis, and alcoholic 
neuritis in one case each. Four cases were un- 
diagnosed. 
The correct localization of the lesion was 
made in 42 cases. 
Astrocytoma: Sixty-three astrocytomas of 
various types made up 14.1 per cent of the 
intracranial neoplasms and 22.6 per cent of 
gliomas. Sixty astrocytomas were in males and 
3 in females; 59 of the patients were white, 3 
* The final revision of this paper was made in 
June, 1946. All statements concerning length of 
survival refer to this date. Primary Intracranial Neoplasms in Military Age Group 
Per cent of 
Location 
Number of Cases Astrocytomas 
Frontal lobe 
22 
(Right 
10, Left 
2) 
34-9 
Temporal lobe 
14 
(Right 
8, Left 6) 
22.2 
Parietal lobe 
9 
(Right 
5, Left 4) 
14.2 
3rd ventricle 
5 
7-9 
Cerebellum 
4 
(Right 
z, Left 
2) 
6.4 
Midline cerebellum 
3 
4.8 
Pons 
2 
3.2 
Pineal region 
2 
3.2 
Unknown 
2 
3-2 
Total: 
63 
100.00 
stupor, syncope, coma, and hemiplegia in 4 
each, hemiparesis in 3, diplopia in 2, blindness 
in 2; sudden death occurred once. Multiple 
symptoms were present in many cases. 
Craniotomy was performed on 54 patients 
and was preceded by ventriculogram or en- 
cephalogram in 35. Two patients died 1 day 
after ventriculogram without further opera- 
tion. Nine deaths occurred less than 1 day 
after operation, 8 after 1 day, 7 in less than 1 
week, and 5 in less than 1 month. The remain- 
ing deaths occurred 3 and 5 months after 
operation. All of the deaths were directly re- 
lated to the surgical procedure. In 5 cases 
death occurred before operation could be per- 
formed. In 25, follow-up studies have not 
been completed, but at the last review of the 
histories no recurrences or deaths were noted. 
In 3 cases subsequent craniotomies were 
necessary to alleviate symptoms, and in I of 
these, 3 operations were performed before 
death occurred. Irradiation treatment was em- 
ployed in 5 cases. 
The diagnosis of brain tumor was made in 
54 cases with localization in 24. Other diag- 
noses were: psychosis with tumor, meningitis, 
meningioma, cerebral hemorrhage, pineal 
tumor, ruptured aneurysm, and epilepsy. No 
history of trauma was mentioned in any of this 
group. 
Medulloblastoma: Forty-five intracranial 
neoplasms were diagnosed as medulloblastomas, 
making up 10.2 per cent of all the tumors, and 
16.2 per cent of the gliomas. Forty occurred 
in white soldiers, 5 in colored. The average 
age in this group was 23.5 years. 
Grossly, medulloblastomas are grayish red, 
soft, friable, solid tumors, usually located in 
the cerebellum in or about the midline. The 
margins of tumor invade the surrounding 
brain and meninges and often extend into 
ventricles with implantation on the spinal cord 
by way of the spinal fluid. The margins are 
grossly distinct though histologically the 
tumors are not well circumscribed (Fig. 28- 
30). If the tumor obstructs the ventricular 
system hydrocephalus results. 
The microscopic appearance of the tumor 
is characterized by small carrot-shaped cells 
Astrocytomas in the military age group show 
no predilection for a particular side of the 
brain, but a striking one for the frontal, parietal 
and temporal lobes. Fourteen of the tumors 
had spread into the surrounding structures in- 
volving more than one lobe of the brain; the 
exact point of origin could not be determined 
because of the size of the lesions, hence they 
were considered to be located in the area where 
the maximum amount of tumor was found. 
One astrocytoma was implanted in the 3rd 
ventricle some distance from the original tumor 
in the temporal lobe. 
The duration of symptoms of the astrocy- 
tomas calculated from onset to diagnosis varied 
from less than I week to 3 years. Details are 
as follows: 
Duration 
Number of Cases 
Per Cent 
Less than i month 
13 
20.6 
1-2 months 
9 
14.3 
3-5 months 
15 
23.8 
6-12 months 
15 
23.8 
Over 1 year 
9 
14.3 
Unknown 
2 
3.2 
T otal: 
63 
100.0 
The duration of 37, or 58.7 per cent of 
the tumors, was under 6 months. One astro- 
cytoma of unknown duration was found in the 
aqueduct of Sylvius in the brain of a suicide 
who had no previous history indicating the 
presence of a brain tumor. 
The most constant symptom at the time 
of diagnosis was headache which occurred in 
47 cases. Next in frequency was papilledema 
in 27, nystagmus in 19, nausea and vomiting 
in 19, convulsions in 10, dizziness in 9, de- 
creased vision in 9, blurring of vision in 8, 604 
The Military Surgeon—November, 1946 
Figs. 18-23 Primary Intracranial Neoplasms in Military Age Group 
605 
each with a distinct elongated nucleus and 
scanty cytoplasm which ends in an indefinite 
tail. A nucleolus is sometimes seen. The cells 
form pseudorosettes, irregular ball-like masses, 
or may assume a perivascular arrangement. 
An occasional mitotic figure may be seen. The 
blood vessels are numerous but do not show 
any specific change. The reticulum network is 
usually prominent; intercellular substance is 
practically absent. The tumor occasionally con- 
tains neuroblastic and spongioblastic elements. 
In our group the latter were seen in 2 tumors 
while neuroblasts were prominent in 3 (Fig. 
31-34). 
Although medulloblastomas have a predilec- 
tion for the cerebellum, they may occur else- 
where. 
Duration 
Number of Cases 
Per Cent 
Less than i month 
7 
15-5 
1-2 months 
I I 
24.4 
3-5 months 
11 
24.4 
6-12 months 
12 
26.8 
Over 1 year 
I 
2.2 
Unknown 
3 
6.7 
T otal: 
45 
xoo.o 
In 36, or 85.7 per cent, of the 42 cases in 
which duration was noted it was less than 7 
months, and in 29, or 69 per cent, less than 
4 months. 
Numerous signs and symptoms, most of 
them related to the cerebellum, were caused 
by medulloblastomas. Again headache was the 
most frequent, being present in 36; staggering 
gait (ataxia) was noticed in 26, as well as 
papilledema in 25. Vertigo was reported in 14, 
vomiting in 14, stiff neck in 11, nystagmus in 
10, nausea in 9, hemiparesis in 7, diplopia, 
pain in the neck, drowsiness, blurring vision in 
5 each, projectile vomiting in 4, hemiplegia, 
tinnitus, syncope, loss of vision, coma, and 
deafness in 2 each. Homonymous hemianopsia, 
areflexia, slurring speech, 6th nerve palsy, per- 
sonality changes, uncinate fits, and exophthal- 
mos were each noted in I case. 
One history recorded the fact that the pa- 
tient’s brother had died of medulloblastoma. 
Hydrocephalus was diagnosed in 15 cases fol- 
lowing encephalogram or ventriculogram. 
Craniotomy was resorted to in 25 of the 45 
cases of medulloblastoma. Ventriculograms in 
20 cases confirmed the diagnosis of a lesion of 
the posterior fossa. 
Of the 27 on whom operation or biopsy was 
performed, 10 died within 24 hours, one 3 
and another 6 days after operation. One 
lived for 5 months, another for 8 months, 
and 3 were still living 3 months after crani- 
otomy. As far as is known, 11 are still alive, 
The locations for the 45 tumors are listed 
below: 
Right cerebellum 
*9 
Left cerebellum 
IO 
Midline cerebellum 
11 
Pons 
3 
Cervical cord 
i 
Right temporal lobe 
i 
45 
In the majority of the cases the spinal cord 
was not removed but an implant on the spinal 
cord was mentioned in one case. The tumors 
have a strong tendency to invade surrounding 
tissues; this was noted in 24 of our 45 cases. 
In several the tumor appeared to be expansile 
and compress the surrounding structures. The 
extent and location of the extension varied 
but it was usually into the 4th ventricle and 
brain stem from either side of the cerebellum. 
The tumor in the temporal lobe, which was 
surgically removed, may represent an implanta- 
tion from a cerebellar lesion. 
The duration of symptoms from onset to 
diagnosis are as follows: 
Fig. i 8. Astrocytoma—fibrillary type. X 30. AIP Neg. 96157. 
Fig. 19. Astrocytoma—protoplasmic type. X 330. AIP Neg. 96277. 
Fig. 20. Astrocytoma—pilocytic type. X 330. AIP Neg. 96228. 
Fig. 21. Astrocytoma—pilocytic type. X 550. AIP Neg. 96156. 
Fig. 22. Astrocytoma—mixed type, with numerous plump astrocytes. X 210. AIP Neg. 96253. 
Fig. 23. Astrocytoma showing processes, some of which are attached to blood vessels. AIP Neg. 96276. 606 
The Military Surgeon—November, 1946 
Fig. 28. Medulloblastoma of cerebellum with extension into brain stem and implantation along- ventricular 
surface. AIP 130099. 
Fig. 29. Medulloblastoma arising in right lobe of cerebellum. AIP Neg. 84635. 
Fig. 30. Medulloblastoma of cerebellum obliterating the 4th ventricle. AIP Neg. 96364. 
and 7 of these have had irradiation treatment. 
Five patients died following ventriculogram 
before a craniotomy was done; 8 died suddenly 
before any surgical measures could be em- 
ployed, I of these during a Metrazol treatment 
for convulsions. 
Multiple operations, up to 3 in number, 
were performed in 3 cases. 
In 15 cases the diagnosis was intracranial 
neoplasm; in 4, tumor of the posterior fossa; 
in 16, cerebellar tumor, with the side in- 
volved mentioned in 6. Psychosis was diag- 
nosed in 7. Cerebral hemorrhage, multiple 
sclerosis, cord tumor, and psychosis due to 
tumor were other diagnoses. Trauma occur- 
ring from 3 months to 11 years previously was 
mentioned in 5 cases. 
Astroblastoma: Six of the 446 intracranial Primary Intracranial Neoplasms in Military Age Group 
607 
Fig. 31. Medulloblastoma showing' fine vascular stroma. The cytoplasm is scanty about the nuclei of many 
cells. X 220. AIP Neg. 96271. 
Fig. 32. Medulloblastoma made up of cells with more granular nuclei than is usually seen. These resemble 
neuroblasts. X 230. AIP Neg. 96267. 
Fig. 33. Medulloblastoma with numerous pseudorosettes and clumps of small hyperchromatic cells. X 250. 
AIP Neg. 96270. 
Fig. 34. Medulloblastoma invading cerebral cortex. Note the nerve cells with typical Nissl’s granules. 
X 175. AIP Neg. 96265. 608 
The Military Surgeon—November, 1946 
Fig. 24. Spongioblastoma polare showing cellularity and a few giant forms. X 310. AIP Neg. 96273. 
Fig. 25. Spongioblastoma polare. Higher magnification emphasizes characteristic cells. X 650. AIP Neg. 
96161. 
Fig. 26. Spongioblastoma polare consisting of unipolar and bipolar spongioblasts; often classified as 
astrocytoma. X 230. AIP Neg. 96x60. 
FlG. 27. Processes of spongioblasts are distinct. (Same as Fig. 16.) X 900. AIP Neg. 96274. Primary Intracranial Neo-plasms in Military Age Group 
609 
neoplasms were classified as astroblastomas, 
representing 1.3 per cent of the series and 
2.2 per cent of gliomas. All these tumors were 
in males, 5 of them white, one Negro. The 
ages of the patients were 21, 22, 23, 24, 25, 
and 37, averaging 25.1 years. 
Grossly the tumors vary greatly and may 
lie either superficially or deep in the brain sub- 
stance. They are usually poorly circumscribed 
and infiltrate the surrounding tissue. Hemor- 
rhage and cystic degeneration are common. 
These tumors are probably transitional be- 
tween astrocytoma and glioblastoma multi- 
forme, differentiating in the astrocytic direc- 
tion. 
The cellular components vary from mature 
astrocytic elements to immature spongioblasts. 
The typical cell is roughly triangular in shape, 
and has an abundant cytoplasm with one or 
more nuclei. There are often well-defined 
processes or “sucker feet” which extend to the 
vessels. These processes appear as single, thick 
expansions which attach to or abut against the 
blood vessel. Multinucleated cells are not un- 
common. Atypical mitotic figures are rarely 
seen. In some of the tumors endothelial pro- 
liferation is marked. The perivascular connec- 
tive tissue is increased in amount; condensa- 
tion of fibrils about the thin-walled vessels is 
characteristic. 
Four of the 6 astroblastomas were located in 
the parietal lobes, the others in the temporal 
and occipital lobes of the cerebral cortex. All 
had invaded the surrounding structures, 3 had 
extended into the ventricles. The largest tumor 
appeared to involve the entire left cerebral 
hemisphere. 
The duration from the onset of symptoms to 
date of diagnosis ranged from 1 day to 5 
years. In 2 the symptoms were present less 
than 3 months, in 1 of these less than I month. 
Headache was present in all cases; other- 
wise the symptoms were varied. In 1 an initial 
sudden attack of unconsciousness was the only 
sign of a brain tumor. Weakness, vertigo, 
ataxia, blurring vision, diplopia, stupor, and 
tremor were most often found at the initial 
examination. One patient had left hemiplegia 
which was of sudden onset. Roentgenograms 
showed calcification of I tumor. Trauma was 
not mentioned in any of the cases. The clinical 
diagnosis of brain tumor was made in 5 of the 
6 cases; in the only one undiagnosed the pa- 
tient was comatose on admission. 
Craniotomy was performed in 4 cases; in 
the other 2 death occurred before operation 
could be performed. Craniotomy had been 
done previously in 2 cases, each having had 
2 operations at an interval of about 3 months. 
One patient died I day after operation while 
no follow-ups are recorded on the other 3. 
Spongioblastoma polare: Thirteen examples 
of spongioblastoma polare were observed in 
the series, representing 2.9 per cent of all the 
tumors and 4.6 per cent of the gliomas. 
Twelve were in males and 1 in a female; 4 
of the patients were Negroes. The average age 
was 25.3 years. 
The gross appearance of this tumor is sub- 
ject to wide variation. It is usually well de- 
marcated, gray and firm, with foci of degen- 
eration, and cyst formation. Spongioblastoma 
polare frequently occurs in the pons or the 
brain stem and a diffuse enlargement of the 
pons results from infiltration by the tumor 
(Fig. 24-27). 
Microscopically, the tumor is made up of 
spindle shaped spongioblasts with either bipolar 
or unipolar processes. The cells are piriform 
to spindle in shape, their processes are long 
and thick, often wavy, and in parallel rows. 
The nuclei are elongated and vesicular. An 
occasional mitotic figure may be seen. In 2 
tumors astrocytes dominated the picture in 
association with large numbers of bipolar 
spongioblasts. Another tumor was primarily a 
spongioblastoma polare with foci of glioblas- 
toma multiforme. 
Spongioblastoma polare, like many others 
of neuroectodermal origin, is usually mixed 
and rarely of pure type; some authors classify 
it as a variety of astrocytoma. 
Spongioblastoma polare may be located any- 
where in the brain but is most frequent in the 
brain stem. The sites of tumor in our series 
are as follows: 610 
The Military Surgeon—Novembery 1946 
Location 
Number 
Per Cent 
Pons 
4 
30.8 
Midbrain 
1 
7-7 
3rd ventricle 
3 
23.1 
Pineal body 
I 
7-7 
Thalamus 
I 
7-7 
Occipital lobe 
I 
7-7 
Temporal lobe 
I 
7-7 
Cerebellum 
I 
7.6 
1 3 
100.0 
9 cases there was hydrocephalus. 
Craniotomy was performed in 7 cases; in 
an eight death took place after the encephalo- 
gram and before craniotomy could be per- 
formed. At the time of this writing 3 patients 
are believed to be still alive although there 
has been no follow-up; 2 died one day post- 
operatively, while 2 others died 5 months after 
operation. Mortality statistics are necessarily 
incomplete so soon after the end of the war. 
The clinical diagnosis was recorded in 12 
cases. The location in the pons was recognized 
clinically in 5 cases and confirmed at post 
mortem in 4. The diagnosis of “encephalitis 
versus tumor” was made in 1 case, and “brain 
tumor” in the others. Sudden coma suggested 
“phenobarbital poisoning” in one case. 
Oligodendroglioma: Fourteen of the tumors 
were classified as oligodendrogliomas, compris- 
ing 5.0 per cent of the group of gliomas, and 
3.2 per cent of intracranial tumors. All oc- 
curred in males, 12 of them white and 2 
Negro, with an average age of 27.9 years. 
The tumor is usually solid, well circum- 
scribed, and gray-red. In sections through the 
brain the surface of the tumor is slightly el- 
evated above the surrounding tissue. Foci of 
calcification are present in 32 per cent of 
oligodendrogliomas of this group (Fig. 35). 
The cells have a distinct limiting mem- 
brane, are usually hexagonal, and when packed 
closely together give a “honey comb” appear- 
ance. The nuclei are small, round, hyper- 
chromatic, and appear to be “boxed” by clear 
cytoplasm. Short processes from the angles of 
cytoplasmic wall are seen with silver carbonate 
stains. An immature type of the tumor, 
oligodendroblastoma, is made up of larger cells 
which are more undifferentiated. Astrocytes 
and ependymal cells in varying numbers may 
be distributed irregularly through the tumor. 
The vascular component is prominent. Necrosis 
and calcification are seen in about one-fourth 
of these tumors (Fig. 36-38). 
Oligodendrogliomas are usually located in 
the cerebrum. Eight of those in this series in- 
volved adjacent portions of the brain by con- 
tiguous extension. No ventricular implants 
were noted. 
The tumor was locally infiltrative and ex- 
tended imperceptibly into surrounding tissues. 
There was great variation in size; tumors in 
the pons were smaller but caused symptoms 
earlier than those in the cerebral cortex. From 
the pons the tumors extended into the medulla 
oblongata, brachium pontis and cervical cord; 
those in the 3rd ventricle infiltrated the chiasm, 
thalamus, and basal nucleus. The tumor in 
the midbrain surrounded the aqueduct of 
Sylvius. 
The interval from onset of symptoms to 
diagnosis varied from a few days to 18 months. 
.Number of Cases 
Per Cent 
Under i week 
3 
23.0 
z-6 months 
2 
15.4 
7-12 months 
5 
38-5 
Over 1 year 
2 
15.4 
Not given 
1 
7-7 
13 
100.0 
It is of interest that in the 4 cases in which 
the interval was shortest the tumor was situated 
in the pons or midbrain, whereas the symp- 
toms caused by tumors in other locations pro- 
gressed more slowly. 
The symptoms varied from a severe head- 
ache to coma. Headache was absent in only 2 
cases; other symptoms in order of frequency 
were: vomiting in 6, papilledema in 5, coma 
in 4, impaired vision in 3, convulsions, diplopia, 
bulbar palsy, hemiparesis, ataxia and hyp- 
esthesia in 2 each and single instances of 
apathy, tremor, areflexia, paraplegia, exoph- 
thalmia, dysarthria, vertigo, nystagmus, and 
tinnitus. 
Roentgenograms showed calcium deposits 
in the pituitary and pineal glands in 2 cases 
and erosion of the sella turcica in another; in Primary Intracranial Neoplasms in Military Age Group 
611 
Fig. 35. Oligodendroglioma arising from the subependymal cell plate of lateral and third ventricle in the 
region of the foramen of Munro. AIP Neg. 85084. 
Fig. 36. Oligodendroglioma with small nuclei surrounded by clear zone, giving rise to a “honeycomb” 
appearance. Note the flecks of calcium. X 100. AIP Neg. 96268. 
Fig. 37. Oligodendroglioma showing so-called “boxing in” of the nucleus. X 400. AIP Neg. 75672. 
Fig. 38. Oligodendroglioma: nuclei and thin cytoplasmic wall surround a clear zone. X 575. AIP Neg. 
96245* 612 
The Military Surgeon—November, 1946 
The locations of the 14 oligodendrogliomas 
are as follows: 
surface is often granular and gritty, because 
of flecks of calcium. 
The morphologic appearance of ependymo- 
mas varies with the cytologic pattern and on 
this basis three varieties have been described; 
papillary, cellular, and epithelial. (Globus and 
Kuhlenbeck10’11 believe that papillomas of the 
choroid plexus should also be classed as epen- 
dymomas.) 
(I ) Papillary ependymomas are character- 
ized by fibrous vascular cores surrounded by 
single or multiple layers of epithelium. The 
cells are cuboidal to columnar with large, oval, 
vesicular nuclei. The stroma usually is slightly 
myxomatous, hence the tumor is often called 
a myxopapillary ependymoma (Fig. 39). The 
epithelium does not contain mucus, unlike that 
of the choroid plexus tumors. So-called “bleph- 
aroplasts” seen in the cytoplasm represent the 
basal attachment of the cilia of the ependymal 
cells. 
(2) The cellular ependymoma is made up 
of closely packed cells with abundant cytoplasm 
and oval nuclei containing large prominent 
chromatin granules. Ependymal spongioblasts 
may be present with the characteristic bleph- 
aroplasts. Such cells are piriform, and have 
heavy broad processes which terminate near 
a vessel, giving the appearance of pseudoro- 
settes (Fig. 40-43)- 
(3) The epithelial form of ependymoma is 
characterized by pseudorosettes about the ves- 
sels. The ependymal cells grow side by side 
like cuboidal epithelium. Mitotic figures are 
very rarely seen (Fig. 44-47). Astrocytes and 
polar spongioblasts are usually scattered 
through ependymomas. 
Although ependymomas are usually located 
near a ventricular surface, they may be found 
anywhere in the central nervous system. One 
of the 4th ventricle ependymomas had ex- 
tended into the right cerebellum. The tumor 
arising from the right lateral ventricle ex- 
tended into the 3rd ventricle. The cerebello- 
pontine angle tumor involved the caudate nu- 
cleus and the floor of the 3rd ventricle. The 
lesions located in the parietal, the temporal, 
and the frontal lobes were in close proximity 
to the lateral ventricles. Two of the cerebellar 
Frontal lobe 
8 
Parietal lobe 
3 
Temporal lobe 
2 
Optic chiasm (near septum pellucidum) 
I 
The duration of the tumor calculated from 
onset of symptoms to time of diagnosis varied 
from less than 1 week to 5 years. 
Time 
Number 
Per Cent 
Less than i week 
I 
7-i 
1-3 weeks 
2 
14.2 
1-5 months 
3 
21.4 
6-11 months 
4 
28.5 
1-2 years 
2 
14.2 
Over 2 years 
1 
7-i 
Unknown 
1 
7-i 
14 
100.0 
The presenting symptoms and signs were 
those of increased intracranial pressure. The 
most frequent symptom was headache with 
or without nausea and vomiting. Papilledema 
was seen in 8 cases. Syncope, lethargy, slurring 
speech, psychosis, convulsion, blurred vision, 
vertigo, loss of memory and coma were present 
separately or in combination in many of the 
cases. 
Operation was performed in 10 cases, and 
death occurred in 4 of these within 3 days. Six 
patients were living at the last follow-up. One 
patient died 2, one 3, and a third 22 hours 
after ventriculogram; 2 expired suddenly be- 
fore any surgical treatment. 
The clinicians diagnosed all of the cases as 
brain tumor, and localized the lesion correctly 
in 5 of the 14. Sudden coma led to the diag- 
nosis of aneurysm of the circle of Willis in 
one case. 
Efendymoma: Fourteen tumors were classi- 
fied as ependymomas which represent 3.2 per 
cent of the entire series and 5.0 per cent of 
gliomas. All occurred in males, 13 of whom 
were white and 1 a Negro, and whose aver- 
age age was 26.6 years. 
Grossly the tumor appears as a solid reddish 
brown mass. It is encapsulated, lobulated, and 
usually is situated near or in conjunction with 
a ventricle, from the surface of which it may 
project. Cysts are frequently seen. The cut Primary Intracranial Neoplasms in Military Age Group 
613 
Fig. 39. Ependymoma—myxopapillary type. Note the formation of canals. X 230. AIP Neg. 96185. 
Fig. 40. Ependymoma—cellular type. X 250. AIP Neg. 96168. 
lesions had extended into the 4th ventricle. 
The two lesions arising in the region of the in- 
fundibulum had encroached on and incorpo- 
rated the optic chiasm. The lesion in the aque- 
duct remained localized. The duration of 
symptoms in cases of ependymoma varied from 
3 weeks to 2 years. Two cases were found for 
each of the following durations: 3 weeks, 4 614 
The Military Surgeon—Novembery 1946 
Fig, 41. Ependymoma showing the pattern of cells about the blood vessels. X 650. AIP Neg. 96184. 
Fig, 42. Ependymoma of cellular type with accumulation of cells about vessels and a relatively clear zone 
between the vessels and the cell bodies. X 305. AIP Neg. 96183. 
Fig. 43. Ependymoma demonstrating a variation in cellular pattern about vessels. X 230. AIP Neg. 96178. Primary Intracranial Neoplasms in Military Age Group 
615 
Fig. 44. Ependymoma forming indistinct canals. X 230. AIP Neg. 96163. 
Fig. 45. Ependymoma in higher magnification showing the distinct nucleus with scanty cytoplasm. X 650. 
AIP Neg. 96179. 
Fig. 46. Ependymoma with an orderly arrangement of cells on a reticulum background. X 230. AIP Neg. 
96182. 
Fig. 47. Ependymoma with elongated cells forming canals and columns. X 500. AIP Neg. 96150. 616 
The Military Surgeon—Nov 1946 
Fig. 48. Pinealoma invading the cerebellum, X 12. AIP Neg. 96x64. 
Fig. 49. Pinealoma showing two cell types; larger cells with granular nuclei and smaller lymphocyte- 
like cells. X 450. AIP Neg. 96165. 
Fig. 50. Pinealoma with small clusters of large cells with distinct cytoplasm and small cells appearing 
as naked nuclei. X 330. AIP Neg. 81380. Primary Intracranial Neoplasms in Military Age Group 
617 
months, 5 months, 6 months, and 9 months. 
In one case the duration was I month, in 
another a year and a half, and in a third 2 
years. No information as to duration was avail- 
able in 3. History of trauma was recorded in 
one case. 
The presenting symptoms of ependymoma 
were as variable as for other intracranial 
tumors. Headache was the most prominent in 
10 cases, vomiting was a symptom in 6, nausea 
in 4, blurring of vision in 4, ataxia in 3, 
vertigo, diplopia and photophobia in 2 each. 
Other symptoms were: sudden onset of un- 
consciousness or coma, tinnitus, loss of vision, 
explosive speech, attacks of syncope, blindness, 
loss of libido, and personality changes. 
Roentgenologic examination showed ero- 
sion of the sella turcica in 5 cases and calcifica- 
tion in one tumor. 
Acute dysentery, generalized tuberculosis, 
and acute pyelonephritis were the causes of 
death in 3 cases in which ependymomas were 
incidental findings. 
Craniotomy was performed in 9 cases. The 
postoperative deaths occurred in from I to 8 
days. In one case in which no operation was 
performed sudden death followed spinal punc- 
ture. 
Eleven of the 14 tumors were recognized 
clinically and the area of involvement was 
properly localized in 8. 
Eight of the ependymomas were classified as 
epithelial, 5 as cellular, and the remaining as 
papillary. 
Efendymoblastoma: One ependymoblasto- 
ma was found in a white male 22 years of age. 
The symptoms of nausea, vomiting, headache, 
and dizziness had been present for three 
months. Sudden coma occurred 1 day prior to 
operation. At craniotomy a tumor in the 4th 
ventricle was found to block the 3rd ventricle. 
No follow-up report is available. 
Pinealoma (Pinealoblastoma and terato- 
ma) ; Thirteen of the tumors were classified 
as pinealomas, representing 2.9 per cent of the 
entire series and 4.7 per cent of gliomas. All 
were in males, 12 white and I Negro, with an 
average age of 25.4 years. 
Grossly, these tumors are well encapsulated, 
often cystic, and frequently calcified. The ex- 
ternal surface is usually smooth and lobulated. 
The tumors are soft, friable, and yellow-gray. 
The cystic spaces are filled with colloid-like 
material. Invasion of surrounding brain is 
common (Fig. 51). 
Microscopically two types of cells predomi- 
nate: (a) Large cells with abundant granular 
cytoplasm and prominent vesicular nuclei, con- 
taining 1 or 2 nucleoli, are arranged in sheets 
or round masses. The large cells are sur- 
rounded by small lymphocyte-like elements, 
and all are enmeshed in dense connective tissue. 
The larger cells resemble the cells of the adult 
pineal parenchyma and are supported by fine 
connective tissue stroma, (b) The other 
characteristic cells are small and round. (Fig. 
49-56). They are particularly prominent in 
the more immature tumors. 
Courville6 divides pinealomas into pineo- 
blastomas, pinealomas, ganglioneuromas, tera- 
tomas and pineal cysts. Nine of the 13 tumors 
in this series were regarded as pinealomas. Of 
the remaining 4, 2 were pineoblastoma, and 
the other 2, teratomas of the pineal gland 
(Fig. 59-63). 
The tumors, as the name implies, are located 
in the region of the pineal body. Ten in our 
series were close to the pineal gland. The 
major portion of the tumor was in the 3rd 
ventricle in two cases; in another the tumor 
appeared to be chiefly in the roof of the 4th 
ventricle (Fig. 48). 
The duration of the disease from onset to 
diagnosis varied from 3 weeks to I year. 
Duration 
Less than i month 2 
1-5 months 5 
7-12 months 4 
Unknown 2 
The presenting symptoms and signs were 
headache in 8, papilledema in 7, blurring of 
vision in 4, weakness, deafness, diplopia, ataxia, 
and tinnitus in 3 each, nausea and vomiting 
in 2, polyuria in 1, and polydipsia in 1. Con- 
vulsions, lethargy, vertigo, hemiparesis, hemi- 
plegia, optic atrophy, and nystagmus developed 
later but were not initial symptoms. 
In our group pinealomas have extended as 618 
The Military Surgeon—November, 1946 
Fig. 51. Pinealoma filling the 3rd ventricle and extending into lateral ventricle. AIP Neg. 89297. 
Fig. 52. Microscopic of the same pinealoma seen in (52). Note the two types of cells with an occasional 
giant cell. X 141. AIP Neg, 93119. Primary Intracranial Neoplasms in Military Age Group 
619 
Fig. 53, Pinealoma with extension into pia arachnoid, X 7. AIP Neg. 93549. 
Fig. 54. Pinealoma showing two types of cells. Note the similarity to seminoma of the testes. X 210. AIP 
Neg. 96164. 
Fig. 55. Microscopic of pinealoma invading the pia arachnoid. Note the predominance of small cells. 
X203. AIP Neg. 93490. 
Fig. 56. Pinealoma composed chiefly of large cells which resemble the cells of the adult pineal. X 700. 
(\IP Neg. 93494- 620 
The Military Surgeon—November, 1946 
Fig. 57. Same as Fig-. 55. Metastasis to the lung with primary tumor in the pineal gland. X 85. AIP Neg. 
93538. 
Fig. 5S. Metastasis to the lung from a pinealoma. X 85. AIP Neg. 95307. Primary Intracranial Neoplasms in Military Age Group 
621 
Fig. 59. Teratoma of pineal distorting- the 3rd ventricle. AIP Neg. Bu 8386. 
Fig. 60. Cross section of teratoma showing cystic cavities. AIP Neg. Bu 8386. 
Fig. 61. Wall of one of the cystic spaces of the teratoma showing tall columnar epithelium and foci of 
stratified squamous epithelium. X 500. AIP Neg. Bu x0-410. 622 
The Military Surgeon—November, 1946 
far as the following structures: midbrain, 
thalamus, cerebral peduncles, cerebellum, 
aqueduct, 3rd ventricle and 4th ventricle. One 
teratoma of the pineal gland metastasized to 
the lung. It furnishes the only example of 
metastasis from an intracranial neoplasm in the 
entire series (Fig. 57 and 58). 
Craniotomy was performed in 11 of the 13 
cases. A diagnosis of pineal tumor was made 
from roentgenograms after a ventriculogram 
in 12 instances. Postoperative irradiation was 
given to one patient. Deaths occurred from 2 
days to 11 months postoperatively. One death 
occurred 6 months after ventriculogram and 
in the absence of other surgery. One patient 
died suddenly while undergoing observation. 
The period of survival after operation was 
as follows: 
the astrocytic elements of the tumor large, 
often multinucleated ganglion cells with 
vesicular nuclei and prominent nucleoli. The 
cytoplasm is basophilic, vacuolated, and con- 
tains Nissl’s granules. These large cells are 
usually surrounded by satellites, and lie in a 
matrix of nerve cell processes and glial fibrils 
(Fig. 64-68). The nerve elements (neuro- 
cytes), generally show various stages of de- 
generation. Kuhlenbeck and Haymaker15 have 
classified these tumors according to the termi- 
nology of Globus as spongioneuroblastomas 
and glioneuromas depending on the degree of 
maturity of the cells involved. Some of the 
nerve cells are hyalinized, or the Nissl sub- 
stance appears to be clumped (Fig. 69). 
Neuronophagia is not uncommon. 
Ganglioneuromas are usually located in the 
cerebral cortex. In our group one arose in the 
left parietal region and extended into the 
thalamus and corpus callosum. Another had 
its origin in the pineal region between the 
superior and inferior colliculus and invaded the 
pons and cerebellum. One tumor, arising in 
the basal ganglia, extended into the sella 
turcica. One was in the pontine region, an- 
other in the right thalamus, and the sixth 
arose in the vicinity of the optic tract and 
chiasm. 
Number Per Cent 
Less than i week 7 63.6 
1-3 weeks o o 
1-5 months 2 18.2 
6-12 months 2 18.2 
11 100.0 
A diagnosis of pinealoma was made in 5 
cases. The diagnoses of pituitary tumor, sub- 
arachnoid hemorrhage, intracranial neoplasm, 
psychosis, and brain tumor were made in the 
remainder. 
The duration from onset of symptoms to 
diagnosis varied. Two patients had symptoms 
for 1 month, one for 5 months, another for 
two and a half years; 2 tumors apparently 
were symptomless and were incidental findings 
at post mortem. 
Ganglioneuroma (Ganglioglioma): Six in- 
tracranial tumors were classified as ganglio- 
neuroma. This represents 1.3 per cent of the 
series, and 2.2 per cent of gliomas. All the 
patients were white, 5 were men and I a 
woman; the average age was 26.1 years. 
Again headache was the most common 
symptom, and was present in 4 cases, blurred 
vision in 2, vomiting, loss of vision to blind- 
ness, hypalgesia of the finger, hemiparesis, 
papilledema, diplopia, and vertigo in I each. 
The gross appearance of ganglioneuroma is 
not distinctive. The tumor is usually rather 
diffuse and appears as a firm, gelatinous, 
slightly elevated mass in the brain substance, 
with little or no line of demarcation. Oc- 
casionally cysts and hemorrhages are seen. 
A preoperative roentgenogram showed 
calcification in the region of the sella turcica 
in one case. 
The term ganglioneuroma implies that 
nerve cells or ganglion cells are neoplastic and 
active components of a glioma. The difficulty 
in making the diagnosis lies in differentiating 
neoplastic nerve cells from preexisting nerve 
cells incorporated in other varieties of tumor. 
Usually, however, one sees scattered through 
Craniotomies were performed on 3 patients. 
One died 2 days and one 4 months following 
craniotomy, the fate of the third is unknown. 
Two other deaths unrelated to brain tumor 
were caused by septicemia and gunshot wound 
of the abdomen. Primary Intracranial Neoplasms in Military Age Group 
623 
Fig. 62. Columnar epithelium of wall with foci of stratified squamous epithelium. AIP Neg. Bu 10-413. 
Fig. 63. Another area of teratoma of pineal gland showing goblet cells and vascular connective tissue 
base. X500. AIP Neg, Bu 10-410. 624 
I'he Military Surgeon—N ov ember, 1946 
Fig. 64. Ganglioneuroma showing spongioblasts and nerve cells. X 160. AIP Neg. 95383. 
Fig. 65. A higher magnification of the proliferating nerve cells in a ganglioneuroma. Note the distinct 
nucleolus and NissPs granules in the cells. X 450. AIP Neg. 95371. 
Fig. 66. Ganglioneuroma or spongioneuroblastoma; neoplastic glial elements and proliferating nerve 
cells. X 175. AIP Neg. 95387. Primary Intracranial Neoplasms in Military Age Group 
625 
Fig. 67. A focus of glioblastoma multiforme In a ganglioneuroma. A mixed variety of tumor. X 175. AIP 
Neg. 95389. 
Fig. 68. A focus of spongioblastoma polare in a ganglioneuroma (spongioneuroblastoma). This tumor 
contained 4 different glial patterns. X 400. AIP Neg. 95404. 
Fig. 69. Proliferating nerve cells in a ganglioneuroma. X 550. AIP Neg. 95369. 626 
The Military Surgeon—TVovember, 1946 
Brain tumor was diagnosed in 3 of the 4 
cases with symptoms and in the other a 
diagnosis of acute disseminated sclerosis was 
made. 
malignant gliomas are believed to be atypical or 
transitional forms of other kinds of gliomas. 
Microscopically, the cells are usually spindle 
shaped and resemble neuroblasts. The picture 
varies from solid masses of lymphoid-like cells 
with scanty cytoplasm to multinucleated cells 
similar to those seen in glioblastoma multi- 
forme. 
N euroefithelioma: One neuroepithelioma 
was seen in the series, and an additional tumor, 
not included in the series, was observed in a 
patient 44 years of age. The patient was a 33- 
year-old white male who had had symptoms of 
headache and impaired vision in the right eye 
for 5 months. A ventriculogram was followed 
by craniotomy, revealing a tumor in the region 
of the sella turcica which could not be re- 
moved. The diagnosis of neuroepithelioma was 
made on examination of the tissue removed at 
operation. 
The location of one tumor was not given, 
another was in the medulla and the third in 
the left frontal lobe with extension into the 
temporal lobe. 
One patient had been operated on 6 years 
before. A craniotomy was performed on one 
who later received irradiation. One patient 
died during insulin shock treatment for battle 
fatigue. 
This rare tumor is formed by cells which 
arise directly from neuroepithelium and which 
resemble primitive spongioblasts. The gross ap- 
pearance is not characteristic. The tumor may 
be located anywhere in the cerebrum. While it 
is uncommon in the central nervous system, it 
occurs more frequently in the retina and spinal 
cord. The microscopic picture resembles that of 
the primitive neural tube or of glioma of the 
retina. The primitive elongated cells are 
arranged in rosettes, with the central space 
surrounded by cells which appear to radiate 
from it. There is a definite internal limiting 
membrane. The inner lining of the cells is 
often ciliated and contains small granular 
bodies, so-called “blepharoplasts.” The rosettes 
are separated by cuboidal or columnar cells 
which are arranged in bands and columns re- 
sembling medulloblasts. Mitotic figures are 
common. Neuroepitheliomas on occasion 
metastasize out of the central nervous system. 
Symptoms were present for 1 month in 1 
case, and for 1 year in 2 cases. The symptoms 
at the time of diagnosis were vomiting, stupor, 
hemalgesia, nystagmus, convulsions, and peri- 
odic unconsciousness. 
The diagnosis of brain tumor was made 
in two cases and the other was called bulbar 
encephalitis. No history of trauma was re- 
corded. 
Twenty-six (5.8 per cent) of the intra- 
cranial neoplasms were classified as pituitary 
tumors. The group is divided into: (1) 
chromophobic adenomas representing 22 or 
4.9 per cent, of all intracranial tumors, (2) 
eosinophilic adenomas, 3, or 0.7 per cent, and 
(3) malignant adenomas, I, or 0.2 per cent. 
Pituitary Adenoma 
Only tumors which produced characteristic 
symptoms were included in this study; lesions 
which were incidental findings were omitted, 
except for 2 eosinophilic adenomas. In 1 case of 
Cushing’s syndrome, pituitary basophilism was 
noted, but no definite adenoma was found. 
Atyfical Malignant Glioma (Unclassi- 
fied) : Only 3 “malignant gliomas” were 
found among the 446 intracranial neoplasms. 
They made up 0.7 per cent of all tumors and 
1.1 per cent of gliomas studied. All occurred 
in males, 2 white and I Negro; the average 
age was 28.6 years. 
Chromofhobe Adenoma: Twenty-two 
tumors, or 4.9 per cent, proved to be chromo- 
phobe adenomas; 20 were in males and 2 in 
females; 16 in whites, 4 in Negroes and I in 
an Hawaiian. The average age of the patients 
was 28.4 years. 
The gross appearance of so-called malignant 
gliomas may simulate that of any other intra- 
cranial tumor, but they do not conform to any 
morphologic type, and hence are usually 
designated as unclassified gliomas. The 
All the tumors were found in the sella 
turcica. Two of the larger had compressed the Primary Intracranial Neoplasms in Military Age Group 
627 
Fig. 70. Chromophobe adenoma of the pituitary. AIP Neg. 82224, 
Fig. 71. Chromophobe adenoma of pituitary. Note the adenomatous arrangement of the cells. AIP Neg. 
85226. 
Fig. 72. Chromophobe adenoma of the pituitary. AIP Neg. 96180. 
Fig. 73. Chromophobe adenoma of pituitary. AIP Neg. 96166. 628 
The Military Surgeon—November, 1946 
pons, 11 had compressed the optic chiasm, giv- 
ing rise to visual symptoms, without invasion 
of the surrounding structures. Grossly, 
chromophobe adenomas are soft, globoid and 
encapsulated; they vary from 3 mm. to 2.5 
cm. in diameter. The cut surface is red-gray 
and often cystic. They cannot be distinguished 
grossly from eosinophilic adenomas. 
The cells are pale, roughly round to poly- 
gonal in shape, with an abundant finely granu- 
lar cytoplasm which does not stain well. Fine 
mitochondria may be seen. The nuclei are oval 
and deeply basophilic. The cells are divided 
into clumps or islands by vascular connective 
tissue strands. (Fig. 70-73). 
Pituitary dystrophy, infantilism, and 
Froehlich’s syndrome are a few of the endo- 
crinologic features of these tumors. Other 
signs and symptoms are related to the location 
of the tumor or to increased intracranial 
pressure. 
The duration of symptoms from onset to 
diagnosis are as follows: 
Pituitary tumor 20 
Rathke pouch tumor 1 
Encephalitis i 
Eosinophilic Adenoma: Three of the 
pituitary tumors were classified as eosinophilic 
adenomas. Two were in white males and I 
in a Negro male; the ages were 22, 23 and 
25 years. 
Two of the 3 patients had had acromegaly 
for 3 and 7 years respectively. Optic atrophy 
was demonstrated by ophthalmoscopic exami- 
nation, and one patient had been totally blind 
for a year. Roentgenograms showed an en- 
larged sella turcica in 2 patients. The third 
patient had been without vision in the left eye 
for 3 months, was obese, and had female distri- 
bution of hair. 
Operation was performed in all 3 cases; 2 of 
the tumors removed were eosinophilic aden- 
omas while the third was mixed, with 
eosinophilic cells predominating. There were 
no postoperative deaths. 
The histologic picture shows oval to poly- 
gonal cells with a moderate amount of cyto- 
plasm containing numerous fine eosinophilic 
granules. The cells were arranged in sheets 
or alveoli, and suspended on a thin scanty 
vascular stroma. The fine eosinophilic granules 
are called the “alpha” granules. The 
eosinophil cells are often mixed with 
chromphobe elements. 
Time Number Per Cent 
Less than 1 month 2 9.1 
2-5 months 5 22.7 
7-11 months 5 22.8 
1-2 years 6 27.2 
Over 2 years 2 9.1 
Unknown 2 9.1 
22 100.0 
The symptoms were predominantly visual, 
as evidenced by failing or blurred vision, com- 
plete blindness, bitemporal hemianopsia and 
homonymous hemianopsia. Other symptoms 
related to pituitary dysfunction were absence 
of body hair, obesity, acromegaly, polyuria, 
dwarfism and female contours. The remain- 
ing symptoms resulted from intracranial pres- 
sure. Trauma was suggested as a predisposing 
factor in 2 cases. 
Malignant Adenoma (Adenocarcinoma of 
Pituitary): One case was diagnosed as ma- 
lignant adenoma. The tumor occurred in a 
white male of unknown age. The symptom 
was progressive loss of vision for 1 year. En- 
largement of the sella turcica was seen in the 
preoperative roentgenogram. Craniotomy was 
performed. The microscopic picture showed 
features of chromophobe adenoma with 
marked hyperchromatism and numerous 
mitotic figures. 
Preoperative roentgenograms showed en- 
larged or eroded sella turcicas in 14 patients. 
Craniotomy was performed in 18 cases; 
postoperative death occurred in 7; on the day 
of operation in I case, 1 day after operation 
in 3, 2 days after operation in 2, and 4 days 
after operation in I. 
Meningioma 
Thirty-nine tumors in the series were classi- 
fied as meningiomas; that is, as dural endothe- 
liomas, meningeal fibroblastomas, lepto- 
meningiomas, and arachnoidal meningeal 
fibroblastomas. These make up 8.7 per cent of 
The clinical diagnoses were as follows: Primary Intracranial Neoplasms in Military Age Group 
629 
this series. Thirty-four occurred in males and 
4 in females; 36 patients were white and 2 
colored. In the period from Pearl Harbor to 
V-J day, 17 additional meningiomas were 
seen in patients over 38 years of age. The ma- 
jority (10) of these were in the fifth decade. 
The average age of the patients was 28.4 
years, and that of the group including the older 
patients was 38.8 years. 
so-called psammoma bodies. The mesenchymal 
variety is distinguished by a type of cell which 
shows more apparent activity which is reflected 
in the rather rapid growth. The distinctive 
cells are bipolar or stellate with large interlac- 
ing cytoplasmic processes. The numerous 
vessels of the tumor are thin walled. Myx- 
omatous degeneration is often seen. The 
angioblastic variety is distinguished by marked 
cellularity. The cells are arranged in irregular 
sheets, with giant cells scattered throughout. 
Small and large irregular endothelial lined 
spaces separate the islands of cells. The back- 
ground of the tumor is a fine reticulum net- 
work. The melanoblastic variety is made up 
of the chromatophores of the leptomeninges 
and is characterized by long spindle shaped 
cells which accumulate around vessels and 
may or may not contain pigment. Mitoses are 
common. The growth of the tumor is rapid 
and progressive. The sarcomatous variety is 
usually a diffuse tumor composed of large oval 
and round cells with eosinophilic cytoplasm 
and eccentrically placed, hyperchromatic 
nuclei. The arrangement of cells around 
vessels has caused the tumor to be referred to 
as peritheUal or perivascular sarcoma. Mitoses 
are common (Fig. 81). The lifomatous 
variety is usually seen on the dorsal surface of 
corpus callosum or in the region of tuber 
cinereum, mammillary bodies, or midbrain. 
The cells are mature fat cells with varying de- 
grees of interstitial connective tissue. The 
osteochrondroblastic variety is composed chiefly 
of mature bone and cartilage with islands of 
cellular connective tissue and fat (Fig. 78). 
The meningiomas are gray to reddish 
brown, well encapsulated, nodular tumors. A 
majority of them at some point are attached to 
the dura. They are usually very vascular and 
commonly contain foci of calcification. 
Meningiomas may invade bone and brain sub- 
stance, but more often compress the surround- 
ing structures (Fig. 74). A reactive exostosis 
of the cranium is sometimes associated. These 
tumors may arise from the leptomeninges 
anywhere in the calvarium, frequently in the 
midline along the parasagittal region. Menin- 
giomas spreading over surface of the brain are 
referred to as endotheliomas or meningiomas 
“en plaque.” The cut surface is often gritty 
because of the calcified particles (psammoma 
bodies) or spicules of bone. 
The meningiomas are classified as angio- 
blastic, chondroblastic, osteogenic, fibro- 
matous, and sarcomatous on the basis of 
predominating morphologic features (Fig. 75- 
81). 
Bailey and Bucy3 have a classification based 
on the family resemblance and histofunctional 
differentiation of cells. They designate a 
fibroblastic variety, characterized by sheets of 
proliferating fibroblasts with no whorl forma- 
tion in which the vessels are usually large and 
often hyalinized. The meningotheliomatous 
variety is made up of sheets of cells with large 
vesicular nuclei and plentiful granular cyto- 
plasm. The masses of cells are separated into 
lobules by vascular connective tissue strands. 
Hyalinized fibrils are common. The fsam- 
momatous variety tends to form whorls com- 
posed of rather plump cells. The whorl may 
consist of a compact nest of cells only or it 
may be arranged around a central blood vessel. 
The center often contains calcified masses, the 
The majority of meningiomas consist of 
plump fusiform cells with vesicular nuclei. The 
cells are arranged in whorls and sheets with 
or without calcium granules. The predominat- 
ing cell is mesenchymal in origin. 
The classification of meningiomas in this 
series is as follows: 
Meningioma (without qualifying classification) 15 
Osteogenic 2 
Psammomatous 4 
Fibroblastic 8 
Fibrous 1 630 
The Military Surgeon—November, 1946 
Fig. 74. Meningioma arising in the lateral ventricle. AIP Neg. 80248. 
Fig. 75. Cellular meningioma with characteristic whorls and strands of cells. X 175. AIP Neg. 96263. 
Fig. 76. Fibrous type of meningioma. X 90. AIP Neg. 96251. 
Fig. 77. Cellular meningioma with few tumor giant cells and multinucleated giant cells. X 230. AIP 
Neg. 96252. Primary Intracranial Neoplasms in Military Age Group 
631 
Fig. 78. A meningioma invading bone and forming islands of cartilage. AIP Neg. 96243. 
Fig. 79. A meningioma with numerous calcified masses, some of which are spicules of bone, the osteo- 
blastic variety. AIP Neg. 96256. 
Fig. 80. A meningioma with atypical cells adjacent to the characteristic whorls. The atypical cells repre- 
sent malignant change. AIP Neg. 96259. 
Fig. 8 i . A malignant meningioma with atypical cells with hyperchromatic nuclei, some of which are giant 
forms. AIP Neg. 96170. 632 
The Military Surgeon—November, 1946 
Meningothelial 1 
Cellular 1 
Angioblastic 2 
Malignant 5 
There is no correlation between the type of 
meningioma and the location of the lesion. 
neck, syncope, nausea, lethargy, generalized 
convulsion, and rectus palsies, each were ob- 
served in 3 cases. Ataxia, stiff neck, loss of 
smell, pain in the arm, and uncinate fits also 
were early symptoms. 
The preoperative roentgenogram showed 
calcification in 5 tumors, erosion of the sella 
turcica in 2, and invasion of the calvarium 
in 3. 
Location of meningiomas 
Right Midline Left 
Cerebellum 1 2 1 
Frontal lobe 104 
Parietal lobe 1 o 3 
Temporal lobe 105 
Occipital lobe 1 o 1 
Sphenoid wedge 102 
Parasaggital 080 
Sella turcica 2 
Lateral ventricle 1 
3rd ventricle 1 
Multiple 1 
Unknown 2 
T otal: 13 10 16 
Craniotomy was performed in 32 cases and 
ventriculograms in 10. Seven patients died 
postoperatively, 1 on the day of operation, 
others I, 2, and 8 days, and I, 3, and 5 months 
later. Three patients died in coma before 
operation could be undertaken. The remaining 
3 deaths were caused by carcinoma of the 
esophagus, subarachnoid hemorrhage, and an 
automobile accident. 
Four cases had recurrences from 3 months 
to 2 years after the initial craniotomies. Three 
craniotomies were performed on one patient 
for recurrences in the same region. 
Four meningiomas had extended into and 
invaded the bone; 4 others invaded the brain. 
Two tumors arising in the pia had extended 
into the dura. One of the meningiomas was 
associated with hyperostosis frontalis. 
A clinical diagnosis of brain tumor was 
made in 33 cases. Pituitary tumor, encepha- 
lopathy, subacute endocarditis, subarchnoid 
hemorrhage, meningocele (nasal), and 
carcinoma of esophagus were the diagnoses 
in the remaining cases. 
The duration from onset of symptoms to 
clinical diagnosis is as follows: 
Number Per Cent 
Less than i month 2 5.1 
1-5 months 14 35.9 
6-11 months 5 12.8 
1-2 years 9 23.x 
Over 2 years 4 10.3 
Incidental 3 7.7 
Unknown 2 5.1 
39 100.0 
Neurofibroma 
This group includes all tumors arising from 
cranial nerves: fifteen of these, or 3.4 per cent, 
are included in the series. All neurofibromas 
occurred in white males, with an average age 
of 26.0 years. 
The tumor is nodular, well encapsulated, 
yellow-gray, and incorporates nerve structures. 
The usual location is in the region of the 8th 
nerve (acoustic) but any cranial nerve may 
be involved. Soft yellow areas of degeneration 
may be scattered throughout the firm gray 
tumor. Neurofibromas are often adherent to 
surrounding brain, meninges, vessels, cere- 
bellum, and bone, but usually can be shelled 
out. The tumor is round or oval with the long 
axis of the mass parallel to the nerve. The 
internal acoustic meatus is often enlarged by 
extension of the tumor along the nerve (Fig. 
82). 
A mass, present in the nose of one patient 
since birth, on removal was diagnosed as a 
cellular meningioma. Trauma was mentioned 
in 4 cases in which it preceded the onset of 
headaches. 
Headache was again the most prominent 
symptom (in 23 cases) and other presenting 
symptoms and signs were papilledema in 15 
cases, impaired vision in 7, facial weakness, 
diplopia, vomiting and vertigo in 5 each, loss 
of vision in 4, and Jacksonian convulsions in 
4. Nystagmus, coma, hemiparesis, pain in the Primary Intracranial Neoplasms in Military Age Grou'p 
633 
Fig. 82. A right acoustic neurofibroma. Note the shifting and compression of pons and brain stem. AIP 
Neg. 84991. 
FlG. 83. A neurofibroma of the 8th nerve with characteristic palisading. X 250. AIP Neg. 96173. 
Fig. 84. A fibrous neurofibroma with palisading about a vascular connective tissue. X 200. AIP Neg. 96188. 
FlG. 85. A cellular neurofibroma with some attempt at palisading. X 230. AIP Neg. 96159. 634 
The Military Surgeon—November, 1946 
The predominating cell is elongated and 
spindle shaped with a vesicular nucleus. In 
places these cells tend to accumulate in a 
parallel fashion forming a palisade. Other 
cells are stellate with homogeneous vacuolated 
cytoplasm and relatively short fibrils. The 
processes formed by the cells are arranged in 
interlacing bundles and whorls. A few 
macrophages containing lipoid material are 
scattered throughout the tumors. The 
reticulum fibers are usually prominent. Mitoses 
are uncommon (Fig. 85). 
The preoperative roentgen examinations 
showed enlargement of the acoustic meatus in 
3, and erosion of the petrous bone in 1 case. 
Fourteen of the 15 patients underwent opera- 
tion, and one half are known to be alive. 
Postoperative death occurred in 7 cases. One 
patient died on the table, 2 others died on the 
day of operation, 2 died 2 days, and the others 
2 and 6 weeks after operation. One patient 
entered the hospital in coma and died before 
surgery could be undertaken. 
In this series the acoustic nerve (8th) was 
involved in 14, the trigeminal (5th) in one. 
Seven of the neurofibromas were regarded as 
neuromas while 4 others were called perineural 
fibroblastomas. 
The clinical diagnoses made on admission 
were as follows: 
Acoustic nerve tumor 8 
Cerebellopontine angle tumor 2 
von Recklinghausen’s disease 1 
Brain tumor 3 
Undetermined 1 
The tumors were located as follows: 
Right Left Bilateral 
Acoustic nerve 662 
Trigeminal 010 
Eight of the 446 intracranial tumors were 
classified as craniopharyngiomas. All of them 
occurred in males; 5 were white, 2 Negroes 
and I unknown; the average age was 28.6 
years. 
Craniopharyngioma (Adamantinoma) 
Three tumors were features of von Reck- 
linghausen’s disease. Multiple intracranial 
neurofibromas were found in I case, and in 
4 the tumor had extended into the petrous 
bone and involved a portion of the cerebellum. 
The duration from the onset of symptoms to 
diagnosis was more prolonged with tumors of 
this type. 
All craniopharyngiomas were located in the 
region of the sella turcica. These tumors arise 
from the remnants of craniopharyngeal ducts; 
they are cystic and have varying amounts of 
calcium deposited in their walls. Their lumens 
are filled with yellow grumose material con- 
taining cholesterin crystals. These tumors are 
usually well encapsulated but may invade the 
surrounding brain substance, and at times ex- 
tend into the 3rd ventricle (Fig. 86-88). 
Number Per Cent 
Less than 1 month 1 6.7 
2- months 2 13.3 
6-12 months 4 26.7 
1 -2 years 5 33.3 
3- years 2 13.3 
Over 5 years 1 6.7 
15 100.0 
Histologically, craniopharyngiomas fall into 
3 groups: (1) Squamous papillary epithelial 
cysts which are characterized by numerous 
papillary projections lined by squamous epi- 
thelium; (2) Adamantinomas which are 
characterized by epithelial cells arranged in 
columns, resembling the pattern of the em- 
bryonic enamel organ. The basilar layer of 
columnar cells (ameloblasts) later becomes 
stratified. Stellate cells are prominent in the 
connective tissue cores of the epithelial columns 
(Fig. 90 and 91); (3) Rathke’s pouch cysts 
which are characterized by a thin fibrous wall 
containing flecks of calcium. The lining is 
composed of ciliated columnar and goblet cells. 
The presenting symptoms and signs were 
related to the nerve, to the side involved, and 
usually to the mechanisms of hearing and 
balance. Ataxia, tinnitus, headache, and deaf- 
ness were most frequently encountered, other 
symptoms were papilledema, nystagmus, cafe 
au lait spots, dimness of vision, diplopia, 
vertigo, vomiting, failing hearing, nausea, 
blindness, lethargy and paralysis of the vocal 
cord. In one case the symptoms of headache 
and gradual progressive deafness had been 
present throughout life. Primary Intracranial Neoplasms in Military Age Group 
635 
Squamous papillary cysts are suprasellar and 
compress overlying structures. Adamantinomas 
tend to invade the sella turcica and destroy the 
hypophysis. All the tumors arise in the region 
of the sella turcica and extend into the chiasm, 
temporal lobe, and sphenoid bone. The inter- 
val from onset to the diagnosis is as follows: 
terized by masses of immature vessels, and 
proliferation of dilated endothelial lined spaces. 
The spindle-shaped endothelial cells composing 
the lining often fuse to form giant cells. The 
capillary endothelium proliferates to form 
irregular cellular buds. A few of the cells may 
show the intracytoplasmic fenestrations or 
channels (Fig. 92-94). Mitotic figures are 
fairly common. The Perdrau stains brings out 
a prominent reticulum network. Foam cells are 
scattered between the vascular spaces. The 
tumors may be classified as capillary, cavern- 
ous, or cellular hemangioblastomas. 
Time Number Per Cent 
Less than 1 week 1 12.5 
1-8 months 3 37.5 
8-12 months 3 37.5 
Over 1 year 1 12.5 
8 100.0 
The most prominent presenting symptoms 
were related to intracranial pressure and in- 
cluded headache, lethargy, nausea, vomiting, 
and coma. In addition, there were visual 
symptoms, such as bitemporal blindness and 
progressive loss of vision; and symptoms of 
dwarfism and obesity related to pituitary 
dysfunction. 
Hemangioblastoma of the cerebellum may 
be associated with angiomas of retina and 
cysts of the pancreas and the kidneys and 
represent the condition known as von Hippel’s 
or Lindau’s disease. On occasion, hyper- 
nephroma and angioma of the liver and spinal 
cord are also seen in these diseases. Another 
condition related to hemangioblastoma is 
Sturge-Weber’s disease characterized by facial 
nevus, glaucoma, and epilepsy. 
The preoperative roentgenograms showed 
enlargement of the sella turcica in 3 cases and 
erosion in 1. 
The hemangioblastomas were located as 
follows: 
Craniotomy was performed in 6 cases. 
Death occurred on the first day after operation 
in I case, the fourth day in another. One 
patient died of pulmonary embolus from 
thrombophlebitis on the seventh day. 
Right Midline Left Unspecified 
Cerebellum 6x61 
Frontal lobe 1 1 
Temporal lobe 1 o 
The clinical diagnoses were 
The tumors of the left frontal lobe extended 
into the parietal and temporal lobes and that 
of the right frontal into the lateral ventricle. 
The duration from onset of symptoms to the 
diagnosis varied from 1 week to 13 years. 
Tumor of chiasm 3 
Suprasellar brain tumor 2 
Pituitary tumor 2 
Hemangioblastoma 
(Hemangioendothelioma) 
i-4 weeks 5 29.4 
1-2 months 2 11.7 
3-5 months 4 23.7 
6 months-i year 3 17.6 
Over 1 year 1 5.9 
Unknown 2 11.7 
17 xoo.o 
Number Per Cent 
Seventeen of the 446 intracranial lesions 
were classified as hemangioblastomas, making 
up 3.8 per cent. Fourteen were found in males 
and 3 in females. All the patients were white, 
and the average age was 30.7 years. 
Hemangioblastomas occur most frequently 
in the cerebellum. They commonly are 
cystic, and often have a mural nodule, 
which nearly always is located on the pial side 
of the cyst. Some are solid or contain only a 
few small cystic areas (Fig. 89). The sur- 
rounding brain is usually compressed but may 
be invaded. The microscopic picture is charac- 
History of trauma incurred 3 and 7 years 
before onset of symptoms was present in 2 
cases. 
The presenting symptoms and signs were 
usually related to the cerebellum, with head- 
ache predominating. Vomiting, ataxia, vertigo, 
nausea, papilledema, stiff neck, nystagmus, 636 
The Military Surgeon—November, 1946 
Fig. 86. A craniopharyngioma lying just posterior to the optic chiasm and projecting into the 3rd ven- 
tricle. Note the gelatinous appearance of the tumor. AIP Neg. 89296. 
Fig. 87. A cross section of same tumor showing gelatinous material and distinct excresences of tumor 
projecting into the cystic cavity. AIP Neg. 89296. 
Fig. 88. Craniopharyngioma: a solid tumor mass projecting into the cystic space formed by the tumor. 
AIP Neg, 77735. 
Fig. 89. A surgical specimen of a mural nodule of a hemangioblastoma of the cerebellum. AIP Neg. 
Br 94-43. 
blurring vision, slurring speech, psychosis, and 
coma were reported in many of the cases. 
Retinal lesions were described in 1 case and 
specified as absent in 1 other; no mention was 
made of the retina in 15 cases. Two cases were 
diagnosed as Lindau’s disease and cysts were 
described in kidneys and pancreas. One case 
was called von Hippel’s disease without qualify- 
ing statements. 
Craniotomy was performed in 14 cases. Primary Intracranial Neoplasms in Military Age Group 
637 
Fig. 90. Craniopharyngioma (adamantinoma). X 160. AIP Neg. 96349. 
Fig. 91. Craniopharyngioma: the characteristic squamous cells are derived from the multilayered basal 
cells, X 160. AIP Neg. 96315. 638 
The Military Surgeon—November, 1946 
Fig. 92. Hemangioblastoma of cerebellum; the immature vascular channels are separated by numerous 
spindle-shaped cells and a fine reticulum. X 210. AIP Neg. 96177. 
Fig. 93. Hemangioblastoma of cerebellum: the cells of tumor appear to be part of formation of vascular 
channels. An occasional mitotic figure is seen. X 210. AIP Neg. 96176. 
Fig. 94., Hemangioblastoma of cerebellum: a higher magnification showing the vascular channels and 
intracytoplasmic channels. X 550. AIP Neg. 96155. Primary Intracranial Neoplasms in Military Age Group 
639 
Two patients died before operation could be 
performed, one of meningitis, the other of 
sagittal sinus thrombosis secondary to otitis 
media. There were 4 postoperative deaths; 2 
on the day of operation, one 2 days later, and 
another after 23 days. 
Sarcoma of undetermined, origin presents a 
variable picture and does not fall into the other 
specific categories (Fig. 96). 
The sarcomas in our series were classified 
as fibrosarcoma (3), perivascular sarcoma 
(1), and primary melanosarcoma (1). 
The clinical diagnoses included: 
A clinical diagnosis of brain tumor was 
made in all 5 cases. 
Brain tumor 6 
Cerebellar tumor 8 
Hypertensive encephalopathy i 
Recurrent hemangioendothelioma (cerebellum) i 
Meningoencephalitis i 
The primary location of sarcomas was diffi- 
cult to ascertain for a majority were rapidly 
growing and very invasive. Three were pre- 
dominantly in the right frontal area and one 
of these had extended into the meninges and 
the lateral ventricle. Another had extended 
across the midline by way of the corpus 
callosum and invaded the opposite frontal lobe. 
A tumor of the 3rd ventricle had invaded the 
brain stem. 
Sarcoma 
Five, or 1.1 per cent, of the intracranial 
neoplasms were classified as sarcomas. All the 
tumors were in white males, 23, 25, 31, 32 
and 33 years of age, making an average of 
28.8 years. 
The interval from onset to diagnosis was 
short, varying from 5 weeks to 4 months; in 
4 of the 5 cases, it was less than 3 months. 
Malignant meningiomas (meningiosar- 
comas) have been included with meningiomas 
and malignant neurofibromas (neurofibrosar- 
comas) with neurofibromas. 
The presenting symptoms and signs were 
as complex as in the other brain tumors, with 
headache and papilledema in every case and 
nystagmus in 2. Dizziness, nausea, vomiting, 
lethargy, coma, stupor, loss of vision, hemiopia, 
and Jacksonian convulsions were additional 
symptoms. 
The solid tumor may be diffuse and spread 
over the surface of the brain or form a more 
compact mass which invades the surrounding 
tissues (Fig. 29). 
This group of tumors includes lesions vari- 
ously called sarcoma, angiosarcoma, gliosar- 
coma, endotheliomatosis, perithelioma, peri- 
endothelioma, and perivascular sarcoma. 
Abbott and Kernohan1 described 3 types of 
primary sarcomas of the brain, namely: (1) 
fibrosarcomas, (2) perivascular sarcomas 
(perithelial or adventitial sarcomas), and (3) 
sarcomas of unknown type. 
Craniotomy was performed in 4 cases and 
ventriculogram in 5. One patient died after 
ventriculography and before craniotomy 
could be performed, because of hemorrhage 
into the tumor. Two patients died, I day after 
operation from shock and surgical trauma, 
another 7 days postoperatively from pneu- 
monia. 
Cysts 
Fibrosarcoma is characterized by spindle or 
plump oval cells lying in a matrix of collagen 
and reticular fibrils. At times, pseudo-alveolar 
arrangement about small blood vessels is ac- 
companied by myxomatous degeneration. 
There were 16 cystic tumors in all, or 3.6 
per cent of the series. “Colloid” cysts of 3M 
ventricle (paraphysial cysts), make up the bulk 
of this group, but cerebral cysts of undeter- 
mined origin and a pituitary stalk cyst are also 
represented. All incidental cysts of the choroid 
plexus were omitted from this study. Twelve 
colloid cysts of the 3rd ventricle comprise 2.7 
per cent of the series under consideration. 
Three (0.7 per cent) cysts were found in the 
cerebral cortex and one in the pituitary stalk. 
Perivascular cellular tumor consists of small 
round and oval cells with abundant eosinophilic 
cytoplasm and hyperchromatic atypical nuclei. 
The vessels, which are increased in number, 
are surrounded by a fine reticulum. The cells 
form perivascular rings or collars (Fig. 
97-99)- 640 
The Military Surgeon—November} 1946 
Fig. 98. Sarcoma: the cells about the vessels have abundant cytoplasm with eccentric nucleus. These cells 
are intimately related to the vessels. X 230. AIP Neg. 96151. 
Fig. 99. The Perdrau stain shows strands of reticulum intimately associated with the cells about the 
vessels. AIP Neg. 96187. Primary Intracranial Neoplasms in Military Age Group 
641 
FlG, 95. Sarcoma: a soft hemorrhagic tumor indistinguishable from glioblastoma multiforme. AIP Neg. 
90288. 
Fig. 96. Sarcoma: the small atypical cells arranged around vessels appear similar to neuroblasts. No 
rosettes are seen. X 250. AIP Neg. 96154. 
Fig. 97. Sarcoma: the elongated cells with abundant cytoplasm are closely adherent to the blood vessels. 
This represents a perivascular sarcoma. X 700. AIP Neg. 93125. 642 
The Military Surgeon—November, 1946 
Fig. ioo. Paraphysial (colloid) cyst of 3rd ventricle. AIP Neg. 89784. 
Fig. i01. “Colloid Cyst”: The wall of cyst is lined by cuboidal epithelium resembling the cells of choroid 
plexus. Note the psammoma body in the wall. AIP Neg. 96281. Primary Intracranial Neoplasms in Military Age Group 
643 
Colloid cysts: (Paraphysial cysts, 3rd ventri- 
cle cysts). Colloid cysts were present only in 
men, 11 white and 1 Negro, with an average 
age of 23.8 years. In this series 9 were in the 
3rd ventricle and 3 in the septum pellucidum; 
3 cysts were multiple. 
Negro; these patients were 22, 23 and 24 
years of age. 
Sudden death occurred in one; in others 
onset was sudden and characterized by severe 
headache with progressive coma and death. 
At the time of examination one patient com- 
plained of headache, nausea, vomiting, and 
dimming of vision which had been present for 
3 months. The preoperative roentgenogram 
revealed a calcified mass in the left temporal 
region, which was removed at craniotomy. 
The cysts are smooth, round, and attached 
to the wall of the ventricle by a delicate fibrous 
stalk or pedicle; their walls are thin; they are 
filled with a homogeneous “colloid” material 
(Fig. 100). The internal lining is smooth and 
glistening. The thin connective tissue wall is 
lined by a layer of cuboidal epithelial cells, and 
is often hyalinized. The nuclei of the cells are 
large and vesicular (Fig. 101). 
The cysts were located in the left temporal 
and right occipital regions. In the cyst walls, 
marked gliosis with some fibrosis could be seen, 
but morphologic examination gave no clue as 
to origin of the cerebral cysts, although it is 
probable that they are porencephalic. 
The colloid cysts probably develop from the 
primordium of the paraphysis and according 
to Moss16 and others should be called para- 
physial cysts. 
The pituitary stalk cyst was found in a 
patient who had had severe headaches for 
about 10 years, and continuous pain in the 
neck for the last 4 years. Just before admission 
a bilateral hemianopsia was discovered. At 
craniotomy a 5 mm. cyst was found in the 
stalk of the pituitary, and hydrocephalus was 
noted. After removal of the cyst, which was 
lined with flattened cuboidal epithelium, the 
patient made an uneventful recovery. 
The duration from onset to diagnosis was 
as follows: 
Time Number Per Cent 
Less than x week 5 41.7 
2-4 weeks 3 25.0 
Over 1 month 3 25.0 
Not given 1 8.3 
12 100.0 
The most frequent symptom present at the 
time of initial examination was periodic head- 
ache in 7 cases, followed by nausea, vomiting 
and papilledema in 5 each, convulsions in 3, 
dizziness, blurring of vision, and stupor in 2 
each, and sudden unconsciousness, diplopia, 
marked rigidity and blindness in 1 case each. 
Hemangioma 
Only those tumors are included in this series 
which had given rise to symptoms or were di- 
rectly responsible for death. Thirty hemangio- 
mas fulfilled these requirements. This repre- 
sents 6.7 per cent of the 446 intracranial 
tumors studied. Twenty-nine occurred in 
males, I in a female; 26 patients were white 
and 4 Negro; their average age was 30.0 
years. 
Craniotomy was performed in 4 cases; I pa- 
tient died I day after operation. Sudden death 
in the absence of surgical procedures occurred 
in 7 cases, coming 1 to 48 hours after sudden 
onset of symptoms. An hour after he had run 
in a foot race one patient suddenly collapsed 
and died. 
The differentiation of hemangiomas and 
telangiectasis is difficult. The hemangiomas 
include true arterial or venous angiomas. Al- 
though these tumors may be considered 
congenital anomalies, they are here classified as 
tumors. 
The clinical diagnoses made were brain 
tumor in 4, hydrocephalus, meningitis, tumor 
of 3rd ventricle, septicemia, malaria, and in 
3 cases no diagnosis was offered. 
Hemangiomas are divided into capillary or 
cavernous, arterial or venous, depending on the 
type of vessel which predominates. These 
tumors may be isolated or diffuse. They are 
Cerebral Cysts: Three cerebral cysts were 
studied. All were in men, 2 white and I 644 
The Military Surgeon—November, 1946 
Fig. 102. Hemangioma; The vascular lesion is located in the right temporal region. Note the dilated 
vessels in the meninges and the close approximation to the ventricle. AIP Neg. 82345. 
Fig. 103. Hemangioma: The vascular spaces are separated by dense hyalinized connective tissue. Note the 
variations in size of channels. X 50. AIP Neg. 92839. 
Fig. 104. Hemangioma: The vascular channels are both arterial and venous. Note the intimal prolifera- 
tion in some channels. A recent thrombosis is also seen. X 60. AIP Neg. 96282. Primary Intracranial Neoplasms in Military Age Group 
usually composed of both arterial and venous 
channels and form wedge-shaped, blood-filled 
masses with the apex pointing toward the 
center of the brain and the base lying along 
the pia arachnoid. On cross section the dilated 
blood channels are seen to be separated by 
varying amounts of connective tissue. Small 
masses of irregularly coiled vessels may extend 
into, or compress, the surrounding structures 
(Fig. 102). Many of the tumors contain 
thrombi and old blood pigment in macro- 
phages. Hemorrhage is frequently seen in the 
brain tissue about the angiomas. The intimal 
surfaces may show proliferation with frag- 
mentation and duplication of the internal 
elastic lamina. The media is usually distorted 
and in some channels is completely absent 
(Fig. 103-104). Calcification is often present 
in any or all of the layers. A marked glial 
reaction often surrounds this tumor, leading 
to mistaken diagnosis of vascular glioma; the 
gliosis, however, is a response to a vascular 
tumor rather than to a primary neoplastic 
process. 
Duration from onset of symptoms to 
diagnosis was as follows: 
Number Per Cent 
Less than i week 14 46.7 
1-4 weeks 2 6.7 
1-2 months 3 10.0 
3-6 months 6 20.0 
6 months-1 year 3 10.0 
Over i year i 3.3 
Unknown x 3.3 
30 100.0 
In the majority of the cases onset was 
sudden with headache or a brief period of 
unconsciousness immediately before death. 
Significant trauma preceded death by a week 
or two in 2 cases; exercise was related to the 
onset of symptoms in 3. Two deaths occurred 
during sleep and one during heat treatment 
for gonorrhea. Sudden unconsciousness in 16 
cases was preceded most often by headache but 
also by convulsions, exercise, sleep, vomiting, 
vertigo and hemiplegia. Other symptoms and 
signs related to increased intracranial pressure 
were recorded in many of the cases. 
The lesions were classified as follows: 
Hemangioma 21 
Venous angioma 2 
Cavernous angioma 4 
Calcified angioma 1 
Racemose angioma 1 
Cystic hemangioma 1 
3° 
Sudden death occurred in 24, and in 21 
was caused by hemorrhage which was cerebral, 
intraventricular, subarachnoid, or subdural. 
The cause of death in 3 cases was not 
determined. Craniotomy was performed in 10 
cases with one postoperative death on the 
seventh day. 
Intracranial hemorrhage and intracranial 
tumor were the diagnoses in the majority of 
cases; others were tumor with hemorrhage, 
acute poisoning, cardiac failure, and schizo- 
phrenia. Calcification of the tumor was demon- 
strated by roentgen examination in 3. 
Hemangiomas were located as follows: 
Right Midline Left 
Cerebellum 615 
Frontal lobe 3 3 
Parietal lobe 6 2 
Temporal lobe o 2 
Occipital lobe 3 1 
Basal nucleus 1 
Pons 1 
Choroid plexus 1 
Papilloma of the Choroid Plexus 
Three papillomas of the choroid plexus were 
found, representing 0.7 per cent of the series. 
These tumors occurred in white males, one 
19, one 22, and a third 27 years of age. 
Four hemangiomas approximated the border 
of the ventricular systems. Two tumors from 
the periphery of the brain had extended into 
the basal nucleus, one had involved the scalp 
and bone, and a cerebellar lesion had extended 
into the midbrain. 
Papillomas of the choroid have been in- 
cluded among the ependymal tumors by 
Globus and Kuhlenbeck8; however, they arise 
from the choroid plexus and are morphologi- 
cally distinct tumors. 646 
The Military Surgeon—November, 1946 
Fig. 105. The lipoma (A) lies immediately above the corpus callosum and assumes a triangular flat shape. 
AIP Neg, 85152. 
Fig. 106. Epidermoid: Its wall consists of stratified squamous epithelium lying on a thin vascular con- 
nective tissue base. The cysts contain kerato-hyaline material. Note the “daughter” cyst. AIP Neg. 72061. 
The tumors grossly are typical papillomas 
with vascular connective tissue cores; they are 
often cystic. The epithelium, which resembles 
that on the surface of the villi of the choroid 
plexus, arises directly from the medullary 
epithelium. It is not ciliated and does not con- 
tain the so-called “blepharoplasts”* seen in 
* The structures called blepharoplasts by various 
authors correspond to the basal corpuscle of ciliated 
epithelium including ependyma. Primary Intracranial Neoplasms in Military Age Group 
647 
the ependymal epithelium. The cells are 
cuboidal and have large oval nuclei and 
coarsely granular cytoplasm; they are placed 
evenly about a central vascular core of con- 
One patient died 2 months after operation. No 
follow-ups are available on the other 2 cases, 
A diagnosis of brain tumor was made in 2 
cases. 
Table i 
CLASSIFICATION 
S0UPCE 
OF INF CP VATION 
Cushing 1932 
Raker 1943 
A.I.F. 194C 
rlvidge. 
Penfield *r 
'one 1933 
9a i ley 1933 
Case* 
Tumor* 
Glioma* 
C*IM 
% 
Tumor* 
Cl ioma* 
C* * e * 
% 
Tumor* 
% 
Glioma. 
Ca*m» 
Cl ioma* 
C^te* 
t 
Cliomaa 
I. 
Gliomas 
862 
U6.0 
581 
61.8 
281 
62.9 
210 
(W..9) 
UU5 
A. 
Unclassified 
175 
9-3 
6 
1.0 
3 
0.7 
20 
9.5 
67 
If.8 
B. 
Classified 
68? 
56.7 
375 
190 
578 
1. Glioblastoma multiforme 
208 
11.2 
30.5 
102 
16.6 
27-2 
102 
22.9 
56.6 
52 
27.u 
117 
50.9 
2. Astrocytoma 
255 
13.6 
37.1 
166 
26.9 
UU.5 
63 
lU.l 
22.6 
56 
29.5 
156 
35-9 
3. Medulloblastoma 
86 
U.6 
12.5 
25 
U.l 
6.7 
**5 
10.2 
16.2 
28 
1U.7 
55 
U*.6 
U. Astroblastoma 
» 
1.9 
5.1 
16 
2.6 
“*.3 
6 
1.3 
2.2 
15 
6.8 
20 
5.3 
5. Spongioblastoma polare 
32 
1.7 
U.7 
11 
1.8 
2-9 
13 
2.9 
U.7 
11 
5.8 
12 
5-2 
6. Oligodendroglioma 
27 
1.4 
3.9 
12 
1.9 
3.2 
lU 
3-2 
5-0 
8 
U.2 
12 
3.2 
7. Ependymoma 
25 
1.3 
3.6 
32 
5.2 
8.5 
lU 
3.2 
5.0 
19 
10.0 
16 
U.2 
8. Plnealoma 
1U 
0.7 
2.0 
7 
1.1 
1.9 
13 
2-9 
U.7 
2 
1.1 
8 
2.1 
9. Ganglioneuroma 
3 
0.2 
0.5 
6 
1.3 
2.2 
1 
0.5 
10. Neuroepithelioma 
2 
0.1 
0.5 
U 
0.6 
1.0 
1 
0.2 
O.U 
1 
0.5 
1 
0.5 
11. Ependymoblastcma 
1 
0.2 
O.U 
II. 
Pltulta 
ry Adenomas 
560 
19.2 
28 
l*-5 
26 
5.8 
1. Chromophobe 
261* 
lU.l 
22 
l».9 
2. Chromophlllc 
73 
5-9 
5 
0.7 
5. Mixed 
23 
1.2 
1 
0.2 
III. 
Meningioma 
271 
11*.6 
122 
19.8 
39 
8.7 
IV. 
Acoustic Nerve Tumors (Neuroma) 
176 
9-5 
13 
2.1 
15 
3.* 
V. 
Congenital Tumor 
115 
6.0 
22 
3.6 
12 
2.7 
1. Craniopharyngioma 
92 
U.9 
16 
2.6 
8 
1.8 
2. Cholesteatoma and Dermoid 
15 
0.8 
5 
0.8 
U 
0.9 
J. Chondromas and Teratomas 
6 
0.5 
1 
0.2 
VI. 
Blood V 
essel Tumors 
Ul 
2.2 
37 
6.0 
U7 
10.5 
1. Hemanglomas 
23 
3.7 
30 
6.7 
2. Hemangloblastomas 
lU 
2.3 
17 
3-8 
VII. 
Sarcoma 
8 
lU 
0.7 
1.1 
VIII. 
Papillomas, Choroid Plexus 
12 
0.6 
3 
0.7 
IX. 
Miscellaneous 
1. Unclassified Tumors, Brain 
17 
0.9 
13 
2.2 
2 
o.u 
2. Cysts 
6 
0.3 
16 
3-6 
TOTALS 
1,872 
100.0 
616 
100.0 
"6 
100.0 
nective tissue which often is myxomatous. The 
papillomas occasionally may be implanted 
along the ventricular system. 
Lipoma 
Of the 446 tumors 2 were lipomas. Both 
occurred in white men, one 18, the other 26 
years of age. The tumors were incidental find- 
ings in both cases, one in a man killed by a 
stab wound in the heart, the other in a man 
who died from phosphorus burns after a bomb 
explosion. 
The duration of symptoms varied from 3 to 
9 months. Headache and papilledema were 
present in all cases, nystagmus, dimming of 
vision, scotoma, and dizziness each occurred 
once. 
Craniotomy was performed in every case. 
Two of the lesions were located in the 4th 
ventricle and the other in the 3rd ventricle. 
The lipoma in one case was located in the 
meninges, the usual site; the other tumor was 
located on the superior surface of the corpus 648 
The Military Surgeon—Novemhery 1946 
callosum. Baker and Adams5 have reported 
lipomas in the tuber cinereum and midbrain. 
Epidermoid (Cholesteatoma) 
The gross appearance of the elongated, 
encapsulated tumor is that of glistening yellow 
fat (Fig. 105). 
Four tumors classified as epidermoids were 
found in the entire series (0.9 per cent). These 
tumors occurred in white males, 19, 33, 37 
and 38 years of age, with an average of 25.4 
years. 
Lipomas are made up of numerous mature 
fat cells with small areas of embryonic fat and 
a few macrophages containing lipoid material. 
The epidermoid tumor is circumscribed, 
nodular, and cystic, with a smooth external 
Table 2 
AVERAGE AGE OF PATIENTS WITH PRIMARY GLIOMAS, 
IN MILITARY AGE GROUP BY SUB-CLASSES, 
WORLD WARD 
Sub - doss 
NEUROEPITHELIOMA 
MALIGNANT GLIOMA 
OLIGODENDROGLIOMA 
GLIOBLASTOMA MULTIFORME 
EPEN DYMO M A 
ASTROCY TOM A 
GANGLIONEUROMA 
PINEALOMA 
SPONGIOBLASTOMA POLARE 
ASTROBLASTOMA 
MEDULLO BLASTOMA 
EPEN DYMO BLASTOM A 
Tumors of this variety on occasion resemble 
xanthomas. The theories of origin of lipomas 
set forth in the literature are: (i) that the 
component cells arise from lipoid cells already 
in the pia mater; (2) that they are the result 
of fatty transformation of connective tissue; 
(3) that they arise by differentiation of pial 
cells toward embryonic fat (metaplasia), and 
(4) that they arise from embryonal remnants 
of fat. 
surface resembling mother of pearl. The wall 
is made up of dense connective tissue lined by 
stratified squamous epithelium with character- 
istic intercellular bridges and kerato-hyaline 
granules. Its thin fibrous capsule may be calci- 
fied. The cavity of the cyst is filled with 
grayish white, soft, caseous material, composed 
of desquamated epithelial cells and debris con- 
taining cholesterin crystals. Epidermoids may 
be located anywhere in the cerebrum, but are Primary Intracranial Neoplasms in Military Age Group 
649 
usually found at the base of the brain (Fig. 
106). 
age group, 18 to 38 years inclusive. The study 
is based on clinical records and material studied 
during World War II at the Army Institute 
of Pathology. An enumeration of the different 
types of tumors encountered is given in Table 
1, in which comparison is made with other 
series. 
Epidermoid tumors of the choroid plexus 
apparently arise from metaplasia of the choroid 
epithelium, while other epidermoids may arise 
from embryonic epithelial rests. 
The tumors were located in the posterior 
fossa, inferior portion of the occipital lobe, and 
The gliomas represented 62.9 per cent of 
Table 3 
AVERAGE AGE OF PATIENTS WITH PRIMARY INTRA-CRANIAL NEOPLASMS, 
IN MILITARY AGE GROUP, WORLD WAR 31 
Major Groups 
HEMANGIOBLASTOMAS 
HEMANGIOMAS 
SARCOM AS 
CRANIOPHARYNGIOMAS 
PITUITARY ADENOMAS 
MENINGIOMAS 
GLIOMAS 
NEUROFIBROMAS 
CHOLESTEATOMAS 
COLLOID CYSTS 
PAPILLOMAS CHOROID PLEXUS 
LIPOMAS 
the choroid plexus. A tumor of the choroid 
plexus was related to symptoms in I case and 
was the cause of death in the other. The 
interval between onset of symptoms and 
diagnosis varied from 9 weeks to a year and 
a half. The symptom-complex included head- 
ache, blurring vision, papilledema, coma, and 
deafness associated with vertigo and tinnitus. 
the present series, with glioblastoma multi- 
forme (22.9 per cent) the most frequent 
variety and astrocytoma (14.1 per cent) a 
close second. This ratio is reversed in other 
series based on different age groups. Medullo- 
blastomas and pinealomas were more frequent 
and occurred in somewhat older persons in this 
series than in others. Ganglioneuromas were 
slightly more numerous, and the remaining 
gliomas were present in about the same ratio 
as recorded elsewhere. 
In comparing the average age at which 
Summary and Conclusions 
This survey is a summary of data on 446 
intracranial tumors occurring in soldiers of the 650 
The Military Surgeon—November} 1946 
Table 4 
Comparison of Average Ages of Patients 
with Glioma 
series. The average ages are shown in Tables 
2 and 3. It may be seen that the gliomas occur 
in earlier decades in this series than in those 
previously reported by Baker4 and Elvidge, 
Penfield and Cone.9 
Elvidge, Penfield 
Army Institute 
& 
Cone9 
of Pathology 
Astrocytoma 
29.5 
26.5 
Glioblastoma multiforme 
41.2 
27.6 
Medulloblastoma 
19.0 
23-5 
Ependymoma 
29.0 
26.6 
Astroblastoma 
32.0 
25.1 
Spongioblastoma polare 
11-5 
25.3 
Oligodendroglioma 
37-5 
27.9 
Neuroepithelioma 
32.0 
33.0 
Pinealoma 
12.0 
25.4 
Neurofibromas (3.4 per cent) and menin- 
giomas (8.7 per cent) were observed less fre- 
Table 6 
Postoperative Mortality among Patients with 
Primary Gliomas in Military Age Group 
World War II 
(Percentage of death occurring within 
1 month of craniotomy) 
tumors occurred in this series, with that in 
other statistical reports, one must remember 
that a selected age group is represented in this 
Number of 
. Patients 
1 ype or Glioma TT 
Having 
Number of 
Postoperative 
Deaths 
Percent of 
Mortality 
Craniotomy 
Pinealoma 11 
7 
63.6 
Astrocytoma 
54 
29 
53-7 
Medulloblastoma 
27 
I 2 
44-5 
Glioblastoma 
multiforme 
76 
33 
43-4 
Oligodendroglioma 
10 
4 
40.0 
Ganglioneuroma 
3 
I 
33-3 
Spongioblastoma 
polare 
7 
2 
28.6 
Astroblastoma 
4 
X 
25.0 
Ependymoma 
9 
2 
22.2 
Ependymoblastoma 
1 
O 
0.0 
Malignant Glioma 
X 
O 
0.0 
Neuroepithelioma 
1 
O 
0.0 
Total 
204 
91 
44.6 
Table 5 
Race of Patients with Primary Intracranial 
Neoplasms in Military age Group 
World War II 
Wkite Non-white* Total 
Intracranial neoplasms 
Glioma 253 28 281 
Meningioma 36 2 38 
Hemangioma 26 4 30 
Pituitary adenoma 19 6 25 
Hemangioblastoma 17 o 17 
Cyst 14 2 16 
Neurofibroma 15 o 15 
Craniopharyngioma 52 7 
Sarcoma 50 5 
Cholesteatoma 40 4 
Papilloma 30 3 
Lipoma 20 2 
Total Neoplasms 399 44 443 
quently than was expected. This relative 
variation is probably due to the fact that both 
of these tumors tend to develop more often 
in persons of more advanced age. Included in 
this survey were hemangiomas or vascular 
anomalies of hemangiomatous type, which gave 
rise to symptoms or caused death. Forty-seven 
such tumors (10.5 per cent of this series) were 
studied. Pituitary adenomas formed 5.8 per 
cent of the series, with the chromophobe 
variety greatly predominating. The other types 
of tumors in this series were relatively rare; 
they were: craniopharyngiomas (1.8 per 
cent); cholesteatomas (0.9 per cent); sar- 
comas (1.1 per cent); papillomas (0.7 per 
cent) ; lipomas (0.4 per cent), and cysts (3.6 
per cent). 
Gliomas 
Glioblastoma multiforme 93 9 102 
Astrocytoma 59 4 63 
Medulloblastoma 40 5 45 
Oligodendroglioma 12 2 14 
Ependymoma 13 1 14 
Pinealoma 12 1 13 
Spongioblastoma polare 9 4 13 
Astroblastoma 51 6 
Ganglioma 60 6 
Malignant glioma 21 3 
Ependymoblastoma 10 1 
Neuroepithelioma 10 1 
Total Gliomas 253 28 281 
*Note: Non-whites are all Negroes except for 4 
Chinese and 1 Hawaiian. Primary Intracranial Neoplasms in Military Age Group 
651 
Race was specified in 443 cases. Three 
hundred ninety-nine of the tumors occurred in 
whites, and 44 in non-whites (39 Negro, 4 
Chinese, I Hawaiian). The Negro population 
of the Army is approximately 10 per cent, 
which is comparable with the percentage of 
distribution of tumors occurring in this series. 
than in the other major groups. The im- 
mediate mortality was directly related to the 
surgical procedures. Size and location as well 
as type of tumor are the most significant fac- 
tors in the fatal outcome. In over 50 per cent 
of pinealomas and astrocytomas death oc- 
curred less than 1 month after operation. The 
postoperative deaths are analyzed in Table 6. 
Table 7 
Ten per cent of the total deaths occurred 
after spinal puncture, encephalogram or 
ventriculogram, and before other operation 
could be undertaken. In the presence of in- 
creased intracranial pressure such procedures 
should be carried out with extreme caution. 
Other explanations of sudden death occurring 
before operation are: hemorrhage into tumors, 
pressure cone of the medulla oblongata caused 
by changes in hydrodynamics of the ventricu- 
lar fluid by the tumor, edema of the brain 
affecting vital structures. 
Duration of Symptoms Prior to Admission 
among Patients with Primary Intracranial 
Neoplasms in Military Age Group 
World War II 
(Percentage of patients with symptoms present 4 
months and under prior to admission) 
CO 
a 
.2 
C/5 
G 
<D 
C/5 
QJ £ 
c 
C 
<D 
u 
3 
c 
~ hJ 
0 
T ypes 
rt 
Ph 
M-i 
O 
u 
d> 
M 
O 
e 
& 
0 
c 
C/5 
S 
0 
&. 
s 
rt 
Ph 
M-l 
O 
Ih 
1 n 
2 5 
a- „ 
E -c 
§ 
S 
•n 
< 
0 
c 
<u 
U 
~ 
C/2 
-s s 
_o 
Sh 
2 
0 
2; 
• r< u 
& Pi 
<v 
Ph 
Glioma 
Ependymoblastoma 
1 
I 
100.0 
Glioblastoma 
multiforme 
I oo 
74 
74.0 
Medulloblastoma 
42 
29 
69.0 
Astroblastoma 
6 
4 
66.7 
Ganglioneuroma 
4 
2 
50.0 
Oliogodendroglioma 
13 
6 
46.2 
Ependymoma 
I I 
5 
45-5 
Astrocytoma 
65 
29 
44.6 
Pinealoma 
I I 
4 
36.4 
Spongioblastoma 
polare 
12 
4 
33-3 
Total Gliomas 
2 
65 
158 
48.3 
Sarcoma 
5 
5 
100.0 
Hemangioma 
30 
25 
ro 
CO 
00 
Hemangioblastoma 
17 
10 
58.5 
Meningioma 
33 
15 
45-5 
Pituitary adenoma 
20 
7 
35-o 
Craniopharyngioma 
7 
I 
14.3 
Neurofibroma 
i5 
2 
13-3 
Total tumors other 
than gliomas 
127 
65 
51-2 
Correlation between location of tumor and 
symptoms was not attempted, but a tabulation 
of initial symptoms was made in each group. 
A large percentage of patients presented symp- 
toms of increased intracranial pressure. Spe- 
cific localizing symptoms were directly re- 
lated to the region or structures involved. 
The duration of symptoms in each type of 
tumor varied, depending on the location of 
the lesion and the rapidity of growth. The 
shortest course was in the cases of glioblastoma 
multiforme and sarcoma. The duration with 
hemangioma was also brief; in most cases 
death was due to hemorrhage resulting from 
the rupture of the vessels making up the tumor. 
In reference to intracranial neoplasm in gen- 
eral, the immaturity of the lesion was influen- 
tial as well as location in determining the 
development and duration of symptoms. 
This review has brought to light many prob- 
lems regarding intracranial tumors which 
suggest detailed studies to be carried out in 
the future. 
Eighteen, or 4.0 per cent, of tumors were 
in females. This number is so small that no 
conclusions can be drawn as to sex incidence 
of intracranial neoplasms. 
References 
Abbott, K. H., and Kernohan, J. W.: Primary 
sarcomas of brain. Arch. Neurol. & Psychiat., 50:43- 
66, 1943. 
2 Bailey, P.: Intracranial Tumors. Springfield, Il- 
linois, Charles C Thomas, 1933. 
The mortality rate was greater in gliomas 652 
The Military Surgeon—November, 1946 
3 Bailey, P., and Bucy, P. C.; The origin and na- 
ture of meningeal tumors. Am. J. Cancer, 15:15-54, 
1931. 
4 Baker, A. B.: An Outline of Neuropathology. Ed. 
3. St. Louis, John S. Swift Co., Inc., 1943. 
'Baker, A. B. & Adams, J. M.: Lipomatosis of the 
central nervous system. Am. J. Cancer, 34:214-219, 
1938. 
6 Courville, C. B.: Intracranial tumors, their pa- 
thology, symptomatology, diagnosis and prognosis. 
Fiske Fund Prize Essay. Providence, Snow and Farn- 
ham Co., Inc., 1931. 
7 Cushing, H.: Intracranial Tumors. Springfield, 
Illinois, Charles C Thomas, 1932. 
8 Cushing, H., and Eisenhardt, L.: Meningiomas, 
Their Classification, Regional Behavior, Life His- 
tory and Surgical End Results. Springfield, Illinois, 
Charles C Thomas, 1938. 
9 Elvidge, A., Penfield, W., and Cone, W.: The 
gliomas of the central nervous system. A study of two 
hundred and ten verified cases. A. Research Nerv. & 
Ment. Dis., Proc. 16:107-181, 1935. 
10 Globus, J. H., and Kuhlenbeck, H.: The sub- 
ependymal cell plate (matrix) and its relationship to 
brain tumors of the ependymal type, J. Neuropath, 
and Exper. Neurol., 3:1-35, 1944. 
11 Globus, J. H., and Kuhlenbeck, H.: Tumors of 
the stratothalamic and related regions. Tr. Am. Neu- 
rol. A., 67:214-219, 1941. Arch. Path., 34:674-734, 
1942. 
12 Globus, J. H.: Pinealoma. Arch. Path., 31:533- 
568, 1941. 
13 Hassin, G. B.: Histopathology of the Peripheral 
and Central Nervous Systems. Ed. 2. New York, Paul 
B. Hoeber, Inc., 1940. 
34 Horrax, G., and Bailey, P.: Tumors of the pin- 
eal body. Arch. Neurol. & Psychiat., 13:423-470, 
1925. 
13 Kuhlenbeck, H., and Haymaker, W.: Neuroecto- 
dermal tumors containing neoplastic neuronal ele- 
ments: ganglioneuroma, spongioneuroblastoma and 
glioneuroma. (With clinico-pathologic report of 
eleven cases and discussion of their origin and classi- 
fication.) Military Surgeon, 99:273-292, 1946. 
16 Moss, L. D.; Paraphysial Cyst of the Third 
Ventricle. Arch. Path., 39:184-186, 1945. 
17 Weil, A.; Textbook of Neuropathology. Ed. 2, 
New York, Grune and Stratton, 1945. COCCIDIOIDOMYCOSIS: A STUDY OF 95 CASES OF THE 
DISSEMINATED TYPE WITH SPECIAL REFERENCE TO 
THE PATHOGENESIS OF THE DISEASE 
By WILEY D. FORBUS, M.D., Professor of Pathology, Duke University School of Medicine, and 
Resident Consultant of the Army Institute of Pathology, and ANNIE M. BESTEBREURTJE, 
M.D., Instructor in Pathology, Duke University School of Medicine 
(With thirty-seven illustrations) 
CONTENTS 
I. Introduction, Materials, and Methods 654 
II. The History of Coccidioidomycosis 657 
A. The period of discovery, 1892-1913 658 
B. The period of immunologic and epidemiologic development, 1914-1936 659 
C. The period of reorientation, 1937-1942 663 
D. The period of military infection and dissemination of the disease, 1942- 666 
III. Distribution of Cases 667 
A. Origin of the cases 667 
B. Race of the patients 669 
IV. Clinical Features 669 
A. Data from analysis of 45 surviving cases of coccidioidomycosis diagnosed by biopsy 669 
B. Data from 95 cases (50 fatal and 45 surviving) 672 
1. The duration of the disease in 50 fatal cases 672 
2. The initiating signs and symptoms 673 
3. General symptomatology 673 
4. Physical findings 674 
5. Blood findings 677 
6. Sputum findings 678 
7. Spinal fluid findings 678 
V. Pathologic Anatomy as Found in 95 Cases 679 
A. Pulmonary coccidioidomycosis 681 
1. The primary complex 681 
2. Reinfection pulmonary coccidioidomycosis 688 
B. Disseminated coccidioidomycosis 691 
1. Lymph nodes 692 
2. Spleen 694 
3. Skin and subcutaneous tissues 695 
4. Liver and kidneys 697 
5. Bones 698 
6. Adrenals 699 
7. Heart and its coverings 699 
8. Brain and meninges 699 
9. Voluntary muscles 7°3 
10. Gastro-intestinal tract 7°3 
11. Genito-urinary tract 7°4 
VI. The Nature and Genesis of the Typical Lesion of Coccidioidomycosis 704 
A. Chemotactic properties of C. immitis 7°5 
B. Capacity of the organisms to injure tissues 7°7 
C. The sensitization factor in coccidioidal infection 7° 8 
D. Intracellular ferments and other factors 7°8 
E. The proliferative (stimulative) factor 7°9 
F. Correlation of pathogenic factors and the development of the typical lesion 709 
VII. Comment 7IQ 
VIII. Summary and Conclusion 711 
Literature 7*3 654 
The Military Surgeon—November, 1946 
I 
cluded are those cases which had been received 
before the institution of the Army training 
program; interestingly enough, there were 
only two. One of these was contributed by a 
physician in San Antonio, Texas, and the other 
by one in Los Angeles, California. 
Introduction, Materials, and Methods 
PRECEDING the outbreak of hostilities 
in 1941 and early in the war it became 
apparent that the Army would be neces- 
sarily concerned with the problem of coc- 
cidioidomycosis, a rare but disabling and often 
fatal disease. The natural adaptability of the 
terrain and the weather conditions of the 
Southwest led to the selection of this area for 
the concentration of large numbers of troops 
both for flying and for desert warfare training. 
Southern California, Arizona, New Mexico, 
and the western portion of Texas, areas in 
which coccidioidomycosis had been previously 
established as an endemic disease, became the 
site of these activities. 
The occurrence of the first fatal case of 
coccidioidomycosis at a military installation 
was at the Letterman General Hospital, San 
Francisco, February 1941, in a Civilian Con- 
servation Corps worker. This case was filed 
with the Army Institute on June 9, 1941, 
and represents the beginning of an extraordi- 
nary accumulation of material, the sources of 
which are listed in Table I. 
A mere acquaintance with coccidioidomy- 
cosis usually has been considered sufficient for 
physicians outside the endemic areas. That 
this no longer holds true is obvious from the 
fact that this disease has now occurred in 
approximately 6,000 members of the armed 
forces in clinically recognizable form and prob- 
ably in a far greater number as subclinical 
infection. These patients have been scattered 
literally throughout the world, each consti- 
tuting a potential means of transplantation of 
this infection, which formerly was restricted 
to very small areas in North America, South 
America, and perhaps South Europe. These 
facts demand that every physician, especially 
those of the American profession, be fully in- 
formed with regard to this fungus infection. 
Although the primary purpose of this report 
is to present a comprehensive survey of the 
pathologic anatomy of coccidioidomycosis oc- 
curring in the armed forces for the use of the 
Office of the Surgeon General of the Army 
and allied agencies, it its hoped that the study 
of this large accumulation of material may 
at the same time extend our knowledge of 
the general pathology of this little-known 
disease. 
Recognition of the problem that this disease 
might be expected to present to the armed 
forces resulted in the early establishment of a 
carefully organized coccidioidomycosis control 
program. This consisted of (1) careful surveys 
of the areas which were to be used for training 
purposes before the location of Army installa- 
tions, (2) the setting up of a carefully con- 
trolled epidemiologic program within the 
various military establishments, and (3) the 
establishment of a rigidly planned program for 
the study of cases of coccidioidomycosis de- 
veloping in the armed forces in the endemic 
areas. 
One of the important features of the control 
program, the clinical and epidemiologic aspects 
of which have been dealt with in detail in a 
syllabus on coccidioidomycosis prepared by 
Russell V, Lee, Lt. Col., M.C., and Norman 
Nixon, Major, M.C., for the Army Air Forces 
Western Flying Training Command, was the 
careful and complete post-mortem study of all 
persons dying of coccidioidomycosis in the 
Army hospitals and the subsequent accumula- 
tion of all of the data and the materials relating 
thereto at the Army Institute of Pathology for 
the purposes of review and analysis. 
No opportunity comparable to that pre- 
sented by this war-time accumulation of ma- 
terial has previously been offered. The original 
description of the disease by Rixford and Gil- 
christ, although comprehensive and thorough, 
dealt with only two cases. The one report 
which preceded theirs was made by Wernicke 
This report deals with a group of 95 cases 
of disseminated coccidioidomycosis accumulated 
at the Army Institute of Pathology between 
June 9, 1941, and January 31, 1946. In- C occidioidomy costs 
655 
in 1892, and it recorded a single observation. 
Ophuls’ thorough study of the problem in 
1905 was also based on a small number of 
cases and an even smaller number of autopsies. 
According to Bancroft, at this time only 89 
cases of this disease were on record, 49 of 
them fatal. Thus in 1905 a large series of cases 
could not have been studied post mortem. In 
1929 Evans and Ball reported 18 autopsies, 
the largest series on record. At the time of 
this study, 175 cases were known to have been 
seen clinically; but, according to Pulford and 
Larson, only 101 of these had been reported. 
In 1936 Abbott and Cutler were able to find 
only 36 post-mortem studies of coccidioido- 
mycosis among 12,000 autopsies at the Los 
Angeles General Hospital. This yet unpub- 
lished group of cases probably represents the 
largest single accumulation of post-mortem 
material in existence previous to the outbreak 
of the war, since the Los Angeles General 
Hospital is within the area in which the disease 
was first discovered, and also in the area in 
which the largest number of clinical cases has 
occurred. (The Health Department of Cali- 
fornia in 1939 had on record a total of 660 
cases of coccidioidomycosis, the vast majority 
of which had occurred in the San Joaquin 
Valley of south central California. This figure 
is particularly important as an indication of the 
incidence of the disease, which has been report- 
able in California since about 1930.) In 1942 
Benninghoven and Miller stated that only 
125 cases of coccidioidomycosis were on record 
in the Los Angeles County Hospital. It is 
therefore most unlikely that in that endemic 
area more than the 36 autopsies mentioned 
above could be available for study, considering 
the estimated mortality rate of 40 per cent. 
Stiles and Davis in 1942 found that the sum 
total of recorded cases of coccidioidomycosis 
in man did not exceed 790. The total number 
of autopsies to date could therefore not exceed 
approximately 316. 
data concerning these problems as they are 
listed below. 
1. The natural reservoir of the etiologic 
agent and the specific source of human 
infection. 
2. The usual mode of infection of man. 
3. The pathogenesis of the lesion in detail, 
including the mechanism of injury by 
the organism. 
4. The specific factors involved in dis- 
semination of the organisms in the body 
and the time relationship between the 
primary inoculation, the disease which 
follows, and the dissemination of the or- 
ganism to distal points in the body. 
5. The duration and course of C. immkis 
infection and the possibility of more 
accurate prognosis in the disease. 
6. The possibility of spread of the infection 
to areas distant to the endemic areas 
through the migration of patients with 
the disease. 
To the study of these problems as our immedi- 
ate objective may be added the more funda- 
mental objective of our investigations—the 
expansion of our knowledge and understanding 
of that general pathologic process, granulom- 
atous inflammation, which is the basis of 
coccidioidal disease. 
The authors of this report did not partici- 
pate in the actual performance of the autop- 
sies. It therefore seems proper to comment 
briefly on the materials available. We had for 
study pertinent gross material, extensive micro- 
scopic preparations (even in the absence of 
gross material), complete summaries of the 
clincial records of the cases, roentgenograms 
or photographic duplications thereof, photo- 
graphs both of the patient and of the lesions, 
and often complete descriptions of the cultural 
work done for the identification of the etio- 
logic agent. The biopsy materials, with few 
exceptions, were accompanied by the essential 
clinical data and brief descriptions of the 
operative procedures carried out in obtaining 
them. The autopsy protocols were detailed 
and comprehensive, having been prepared 
strictly according to Army regulations. In 
general, the records represent exceptionally 
In reviewing this group of cases and the 
literature our attention has been centered upon 
a number of unsolved problems related to the 
general pathology of coccidioidomycosis. In the 
following discussion we hope to present useful 656 
The Military Surgeon—November, 1946 
Table I 
Hospital 
Location 
Cases 
Santa Ana Air Base 
California 
8 
Beaumont General 
El Paso, Texas 
11 
Brooke General 
Fort Sam Houston, Texas 
6 
Letterman General 
San Francisco, California 
8 
Fitzsimons General 
Denver, Colorado 
4 
Hoff General 
Santa Barbara, California 
5 
Bruns General 
Santa Fe, New Mexico 
2 
Barnes General 
Vancouver, Washington 
1 
Stark General 
Charleston, South Carolina 
1 
Valley Forge General 
Phoenixville, Pennsylvania 
2 
Harmon General 
Longview, Texas 
1 
McCloskey General 
Temple, Texas 
1 
Walter Reed General 
Washington, D. C. 
2 
Newton D. Baker General 
Martinsburg, West Virginia 
1 
Bushnell General 
Brigham City, Utah 
1 
Crile General 
Cleveland, Ohio 
1 
Moore General 
Swannanoa, North Carolina 
1 
Foster General 
Jackson Mississippi 
1 
Lovell General 
Fort Devens, Massachusetts 
1 
Torney General 
Palm Springs, California 
1 
Tilton General 
Fort Dix, New Jersey 
1 
Lawson General 
Atlanta, Georgia 
1 
Tripler General 
Honolulu, Hawaii 
1 
LaGarde General 
New Orleans, Louisiana 
1 
O’Reilly General 
Springfield, Missouri 
1 
Total Interior General 
64 
Station Hospital 
Stoneman, Pittsburg, Calif. 
1 
Camp Van Dorn, Mississippi 
1 
Roswell, New Mexico 
1 
F. F. McDowel 
1 
Camp Ellis, Illinois 
1 
March Field, California 
2 
Camp Roberts, California 
1 
Fort Bragg, North Carolina 
1 
Fort Ord, California 
1 
Receiving Hospital 
AAF, San Antonio, Texas 
1 
Total Interior Station 
11 
34th General 
European Theater 
2 
67 th General 
European Theater 
1 
31st General 
Pacific Area 
1 
93 rd General 
European Theater 
1 
179th Station 
Panama Canal, Caribbean Area 
1 
221st Station 
Panama Canal, Caribbean Area 
1 
Total Foreign General and Station 
7 
7th Field Hospital 
1 
4th Medical Laboratory 
North African and European Theater 
1 
Total Foreign Miscellaneous 
2 
Source (Hospitals) of 95 Cases Coccidioidomycosis 
657 
Table I—(continued) 
Source (Hospitals) of 95 Cases 
Hospital 
Location 
Total Cases 
Civilian Cases: 
Fitzsimons General 
i 
Beaumont General 
2 
Fort Sam Houston Station 
I 
Letterman General 
I 
Walter Reed General 
I 
Mayo Clinic 
I 
Private Hospital 
I 
Total Civilian Cases 
8 
Unknown Sources; 
3 
Total Unknown Sources 
3 
Grand Total 
95 
careful and complete work both on the part 
of those who performed the autopsies and those 
who followed the cases clinically. 
In order to acquire a general familiarity 
with the problems presented by our own ma- 
terials, and thus to determine the possibilities 
of deriving essential contributions from the 
study, about half of our cases were analyzed 
in great detail before any reference was made 
to previous studies in the field. We next turned 
to the literature and studied it completely. 
With the pre-existing picture of coccidioido- 
mycosis thus fixed in our minds, it was then 
possible to return to the material and proceed 
with the detailed analysis of the cases with 
special reference to the less well understood 
features of the disease. Preliminary tabulations 
of data relating to specific problems suggested 
by this review and by the previous studies of 
the disease were made. In the last analysis it 
was found desirable to arrange our materials 
in two major groups, one consisting of cases 
in which biopsies only were performed, and 
the other of cases studied at autopsy, as follows. 
2. Biopsies only . . . . 45 
Total cases 95 
The first group proved to be more useful in 
the formulation of diagnostic criteria for the 
disease, and the second group in the study of 
the general pathologic processes involved in 
this fungus infection. 
II 
The History of Coccidioidomycosis 
By reviewing the entire literature on the 
subject of coccidioidomycosis up to March, 
1945, we have been able to recognize four 
fairly sharply defined periods in the history 
of this disease; they are as follows: 
A. The period of discovery, 1892-1913. 
B. The period of immunologic and epi- 
demiologic development, 1914-1936. 
C. The period of reorientation, 1937- 
1942. 
D. The period of military infection and 
dissemination of the disease, 1942- 
T otal 
Number 
In order that the important events in the 
history of coccidioidomycosis may be seen in 
their proper relationships, the significant con- 
tributions of each of these periods will be 
listed in chronologic order. A complete list of 
all contributions to the subject up to March, 
1. Autopsies: 
a. Biopsies followed by autopsy .... 16 
b. Autopsies without preceding biopsy , 34 
5° 658 
The Military Surgeon—November, 1946 
1945, arranged by years, will be found in the 
bibliography. 
clinical course of the disease, by Ophuls.6 Em- 
phasis in this paper is on the pulmonary portal 
of entry of the organism and on the primary 
pulmonary focus of infection. Ophuls reclassi- 
fied the organism and gave to it the name 
Oidium coccidioides. 
A. The Period of Discovery, 
1892-1913 
1892 
Original description of the disease coc- 
cidioidomycosis by R. Wernicke,1 including 
a brief discussion of the various forms of the 
parasite responsible for it, characterization of 
the lesions as “granulation tissue tumors,” 
and classification of the parasite as a protozoan. 
2. Detailed report by Ophuls' on the 
morphology, the life history, and the patho- 
genicity of C. immitis in dogs, rabbits, and 
guinea pigs. Ophuls thought he recognized a 
sharp difference between the pathogenic effect 
of the immature spherule and that of the 
sporulating spherule; the latter he thought 
responsible for the nodular epithelioid reac- 
tion of the tissues, and former for the suppura- 
tive and necrotizing reaction. 
1896 
Description of the first two cases of coc- 
cidioidomycosis occurring on the North Ameri- 
can continent by Rixford and Gilchrist.2 This 
includes a detailed study of the parasite and of 
the pathologic anatomy of the disease. The 
organism was classified on a preliminary basis 
as a protozoan; on the recommendation of 
Stiles, the name Coccidioides immitis was as- 
signed to it. Dogs and rabbits were successfully 
inoculated with the organism. 
1907 
A complete survey of all reported cases of 
coccidioidomycosis and a comparison with cases 
of blastomycosis, by Hektoen.9 Hektoen recog- 
nized a primary pulmonary and primary 
cutaneous form of coccidioidomycosis and em- 
phasized, first, the predilection of the organism 
for the lymph nodes, and second, the re- 
semblance of the tissue reactions to those in 
tuberculosis. 
1900 
1. Posadas4 published a more comprehensive 
description of the original case presented in 
1892 by Wernicke. Posadas, who originally 
called attention to the disease while working 
in Wernicke’s laboratory in Buenos Aires, 
Argentina, reproduced the infection in 
monkeys and dogs. This is a complete and 
finished report. 
1909 
Description of the fourth case of C. immitis 
meningitis. This report by Evans10 describes 
meningitis unaccompanied by evident infection 
in other parts of the body—presumably 
primary coccidioidal meningitis. All previously 
reported meningeal infections were considered 
secondary to lesions elsewhere. 
2. Preliminary description of the cultural 
characteristics of C. immitis} representing the 
first systematic study of this problem, by 
Ophuls and Moffit.3 
1912 
I9°4 
First mention of coccidioidal infection in the 
German literature. This report by von Wasie- 
lewski11 deals with cultures of the organism 
and is largely a review of the American work 
on the disease. 
Detailed studies of C. immitis, including its 
morphology and cultural characteristics, by 
Wolbach.5 
1905 
1. A summary of all cases of coccidioidomy- 
cosis reported up to this time and a detailed 
description and comprehensive discussion of the 
pathologic anatomy, the histology, and the 
1912 
First report of a case of C. immitis infec- 
tion from Texas. The case was reported as 
one of blastomycosis, but the organism was Coccidioidomycosis 
659 
subsequently identified as C. immitis by Mac- 
Neal and Hjelm.12 
'P'5 
1. The second report dealing with coccid- 
ioidomycosis in the European literature. This 
report by Campiche16 deals with a case origi- 
nating in California. 
B. The Period of Immunologic and Epidemi- 
ologic Development} 1914-1936 
1914 
2. A thorough clinical description of coc- 
cidioidomycosis by Dickson,1' with the addition 
of nine cases to the literature, bringing the 
total reported cases to 40. Dickson cited the 
1. First studies on immunity demonstrat- 
ing a specific precipitin reaction in a patient 
with coccidioidal infection, by Cooke.14 Cooke 
\Macrophagic 
\ Reaction 
Granulomatous 
Reaction 
Maturation.: 
< Externol 
Environment 
\Endospore 
Chlamydospore 
Budding 
Spore 
Mycelium 
Endosporulotior 
Suppurative 
Reaction 
LIFE CYCLE OF COCClDIOIDES IMM1TIS AND RELATED HISTOLOGICAL RESPONSE 
Fig. 1 
also tested her patient for a specific skin reac- 
tion and for complement-fixing antibodies and 
agglutinins, but failed to obtain positive re- 
sults in any of these tests. 
work of Cooke on immunity as the only con- 
tribution to date dealing with this aspect of the 
problem. 
2. Discussion of the 24 cases of coccidioidal 
disease recorded up to this time with an ex- 
cellent pictorial study of the morphology of the 
various developmental forms of the parasite, by 
MacNeal and Taylor.15 
1920 
Lynch18 reported the post-mortem study of 
a case of disseminated coccidioidomycosis in 
South Carolina. This is the first case observed 
east of the Mississippi River. 660 
The Military Surgeon—November, 1946 
complement-fixing antibodies occur in the 
blood of infected patients. 
1923 
1. Hirsch18 described a case of coccidioi- 
domycosis in a Pullman porter who contracted 
the disease in California and migrated to Chi- 
cago, where he died. Emphasis was placed on 
the possibility of spread of the disease by the 
migration of infected persons. 
2. Earliest recorded studies of coccidioi- 
domycosis by x-ray. In this report by Taylor20 
attention is called to the similarity of the 
widespread lesions in the bones and in the 
lungs to the lesions of disseminated tuberculo- 
sis. 
2. DaFonseca and deArea Leao30 demon- 
strated a specific skin reaction to the filtrate of 
a bouillon culture of what was believed to be 
C. immitis in a patient with “coccidioidomy- 
cosis.” The organism under investigation was 
later (1931, 1932, 1934) proved to be a 
new species, Paracoccidioides brasiliensis (later 
shown to be a blastomycete) by deAlmeida. 
(In 1932 deAlmeida also demonstrated that 
the organism “Pseudococcidioides ma%%aiin- 
troduced by daFonseca from Argentina, is 
identical with C. immitis. De Almeida’s work 
on these organisms has been verified by N. F. 
Conant and D. J. Davis.) 
1924 
First case of “coccidioidomycosis” in Brazil 
recorded in a study by Gomes and deAssump- 
sao.21 
3. Description of South American blas- 
tomycosis by deSonza Campos and de- 
Almeida32 produced by Paracoccidioides 
brasiliensis, a genuine blastomycete (later 
named Blastomyces brasiliensis by Conant), 
and distinctly different from C. immitis. This 
report from Sao Paulo in Brazil called atten- 
tion to a form of granuloma that sharply re- 
sembles the granuloma produced by C. immitis. 
!925 
1. Further detailed studies on the general 
biologic character of C. immltis by Bump.22 
2. A comprehensive clinical review of 82 
cases of coccidioidomycosis collected from the 
literature by Riesman and Ahlfeldt.24 Spe- 
cial attention is called in this report to the 
occurrence of the disease in four women and 
eight children. Previously the disease was 
thought to occur almost exclusively in male 
adults. 
3. A report of recovery of C. hnmitis from 
the blood by culture. This first description of 
C. immitis septicemia is by Montenegro.23 
4. Detailed studies of the classification of 
C. immitis in its relation to other pathogenic 
and non-pathogenic fungi by Castellani.28 
5. In association with Benson and d’Andrea, 
Hirsch34’35’36 clearly demonstrated a reliable 
specific skin and testis reaction to C. immitis 
culture filtrate in both man and guinea pig; this 
confirmed earlier work and added further 
emphasis to the possible use of a skin reaction in 
the diagnosis and study of coccidioidomycosis. 
1926 
6. Riesman and Ahlfeldt39 summarized the 
87 cases reported to date and described one 
new case, the significant point in which was its 
origin in New Mexico. 
Ahlfeldt26 demonstrated the possibility of 
infection of guinea pigs by (1) enforced 
aspiration of the organism into the lung, (2) 
feeding infected material, and (3) exposing 
the animal to an atmosphere infected with the 
organism. This is one of the earliest experi- 
mental studies on the route of infection by 
C. immitis. 
1928 
I. Stark and Becker4, presented an excel- 
lent summary of 89 cases of coccidioidomycosis 
with extensive clinical and epidemiologic data. 
7927 
2. Report by Tomlinson and Bancroft48’49 
of the first laboratory infection by C. immitis. 
This infection occurred in a medical student 
working with cultures in Weidman’s labora- 
tory in Philadelphia. 
1. By using the filtrate of a bouillon cul- 
ture as antigen, daFonseca and deArea Leao29 
clearly demonstrated for the first time that C occidioidomy costs 
661 
3. Experimental and clinical investigation 
of a specific allergic cutaneous reaction in 
coccidioidal granuloma by Jacobson.45 Jacob- 
son recommended the use of the skin test as 
a diagnostic procedure. 
5. Publication by Cummins, Smith, and 
Halliday54 of a comprehensive summary of 122 
cases of coccidioidomycosis, including excellent 
tables dealing with incidence, mortality, and 
duration of the disease, age distribution, na- 
tionality and occupation of the patients. 
4. Kogoj46 experimentally demonstrated a 
relation between trauma and the localization of 
C. hnmitis in the skin of animals with general 
infection. In his experiment a dermatomycosis 
developed only where the skin had been 
shaved. 
6. Evans and Ball58 published from Los 
Angeles General Hospital, California, a series 
of 18 autopsies from a total of 50 cases of 
coccidioidomycosis. This is the first and largest 
series of autopsies published. The report is 
especially informative with regard to the rela- 
tion of pulmonary infection to the other aspects 
of the disease. 
1929 
1. Ophiils63 in discussing a paper by Pulford 
and Larson64 called attention to the over- 
whelming evidence in favor of the lung as the 
portal of entry of C. immitis. He cited the fact 
that old primary foci may be found in almost 
all cases of coccidioidal granuloma, and that 
the pulmonary lesions are always of the greatest 
clinical and pathologic significance. He dis- 
counted the idea that human infection is re- 
lated in any way to that of the lower animals, 
stating that animals, like man, get their disease 
from the soil. Absence of intestinal lesions in 
coccidioidal granuloma also was emphasized 
as a characteristic feature of the disease. 
1930 
I. The etiologic organism of Brazilian 
“coccidioidal” granuloma was shown by de- 
Almeida69 to be distinctly different from that 
producing North American coccidioidomycosis. 
This work and that which preceded it by the 
same author and by others established the 
Brazilian parasite as a distinct species to which 
the name Paracoccidioides hrasiliensis has been 
given. The disease in question was shown to be 
a blastomycotic, not coccidioidal, granuloma. 
2. A brief but complete review of all work 
on skin reactions in coccidioidomycosis was 
made by Hirsch and d’Andrea.72 In this paper 
there is described in detail the histologic reac- 
tion occurring in the tissues at the site of the 
skin test. The exudation of fibrin, round cells, 
fluid, large reticulum cells, and a few neu- 
trophilic and eosinophilic polymorphonuclear 
leukocytes, without necrosis, was characteristic 
of the reaction. 
2. First description of coccidioidomycosis in 
six cattle and one sheep slaughtered near 
Bakersville, California, by Beck.52 Beck called 
attention to the observation of Giltener in 
which he mentioned having seen coccidioidal 
granuloma in the bronchial and mediastinal 
nodes of cattle in 1928. 
3. A comparative study of coccidioidal 
granuloma as it occurs in the United States 
and a similar disease in Brazil was made by 
deAlmeida,56 He described in great detail the 
organism responsible for the Brazilian disease 
(Paracoccidioides hrasiliensis, later shown to 
be a blastomycete) and insisted that it is a 
distinct species quite different from that of 
North American coccidioidomycosis. 
3. The first case of coccidioidal meningitis 
involving the spinal cord and presenting the 
clinical picture of spinal-cord tumor was re- 
ported by Rand.'9 
4. First reference to coccidioidal granuloma 
in Italy by Radaelli,78 who stated that two 
cases of coccidioidal granuloma had been pub- 
lished in Italy by Jacono. 
4. Ahlfeldt51 described two different types 
of spores in coccidioidomycosis tissues, one 
staining with osmic acid and the other not. 
This represents the first observation of spores 
in the tissues. Ahlfeldt suggested that the two 
types of spores might represent sexual forms. 
5. It was suggested by Jaffe74 that the large 
giant cells in the reacting tissues in coccidioido- 
mycosis serve as the host of the organism, re- 
maining intact until the organisms mature. 
Following maturation and sporulation the 662 
The Military Surgeon—November, 1946 
giant cells were thought to break up, with the 
escape of the spores into the surrounding tissues. 
Following this, leukocytes were thought to 
invade the area, thus producing a suppurative 
lesion. This paper is a very careful and detailed 
histopathologic study. 
two other cases reported from Texas up to this 
time. 
5. Further studies by deAlmeida92 concern- 
ing distinction between Paracoccidioides bra- 
siliensis and Coccidioides immitis, and proving 
the identity of Pseudo coccidioides mazzai and 
Coccidioides immitis. The former was first 
observed in South America by daFonseca and 
thought to represent a new genus. 
i93i 
1. Beck, Traum, and Harrington81 re- 
view the occurrences of coccidioidal infection 
in animals, reporting altogether 20 observa- 
tions to date involving cattle and sheep. This 
work was the outcome of a survey made in 
southern California. Only lesions in the 
bronchial and mediastinal lymph nodes are 
recorded. 
*933 
First case reported from El Paso, Texas, 
by L. M. Smith.107 
1934 
I. Lutz118 republished a case of “coccidioi- 
dal” granuloma published originally in 1908. 
The observation was made in Brazil. This was 
said to be the first instance of “coccidioidal” 
infection in this area. The case was compared 
carefully with the original case of coccidioido- 
mycosis published by Posadas. This case was 
subsequently proved to be an instance of 
blastomycosis caused by Paracoccidioides 
brasiliensis. 
2. An excellent review of the literature by 
Carter82 up to 1931. The author’s contribu- 
tion relates particularly to x-ray diagnosis of 
pulmonary and bone lesions in coccidioidomy- 
cosis. This is the most extensive and compre- 
hensive study made by x-ray to date. 
3. First Case reported in China. The pa- 
tient was a child. This report by Chou and 
Reiss8'5 records as a positive autogenous in- 
oculation, accomplished by injection of unin- 
volved skin of the patient with a culture of its 
own organisms. 
2. Third laboratory infection, published by 
Tomlinson and Bancroft.125 (The patient 
was infected while working with C. immitis 
at the University of Nebraska.) 
4. Special Bulletin No. 57,84 dealing with 
coccidioidal granuloma, issued by the Cali- 
fornia State Department of Public Health. In 
this work are compiled data relating to the 
occurrence of this disease, not only in Cali- 
fornia, but elsewhere. The bulletin contains an 
abundance of source material. 
3. Ciferri and Radaelli113 succeeded in 
propagating the spherules of C. immitis in the 
absence of hyphae in gelatin cultures. These 
workers suggested the possibility of conjuga- 
tion of two or even three spherules when the 
organism is grown in appropriate cultures. 
4, Third Italian case observed in Naples, 
by Radaelli and Ciferri.121 These workers 
stated that the two other cases were reported 
by Poeri and Jacono. The Italian organisms 
proved to be much less pathogenic for both 
man and animals than the American strain of 
C. immitis. (Conant has studied the Italian 
cultures and states that there is no doubt of 
their identity with C. immitis.') 
1932 
I. First case reported from Mexico, by 
Perrin.97 
2. Case originating in Nicaragua reported 
by Williams in a veteran.100 
3. First instance of isolation of C. immitis 
from the soil. Report by Stewart and Meyer.99 
Special medium composed of acriflavin, potas- 
sium hydrogen phosphate, magnesium sulfate, 
ammonium chloride, and sodium acetate, used 
for this purpose. 
5. A detailed study of fungus infections of 
bone, including infection by C. immitis, was 
published by Carter.111 This is an x-ray study 
supplementing his previous contribution in this 
field. 
4. Report of three cases of coccidioidomy- 
cosis originating in Texas by Caldwell.88 Only 
6. A careful study of the morphology of C occidioidomy costs 
663 
a variety of fungi including C. immitis, with 
special reference to the formation of radiating 
clublike bodies by the capsules of the spherules 
of C. immitis and by other forms of other 
fungi, was made by deAlmeida.117 He sug- 
gested that these bodies are not products of the 
environment of the parasite, but are part of 
the parasite itself. 
sumably by the handling of chicken food made 
from the meat of condemned cattle. The 
primary lesion occurred on the finger. 
C. The Period of Reorientation—Discovery of 
the Etiologic Relationship between San Joaquin 
Valley Fevery with its Characteristic Accom- 
panimenty Erythema Nodosum or the 
“Bumps” and Coccidioidal Granuloma, Thus 
Disclosing the Primary or Early Phase of the 
Entity Coccidioidomycosis, 1937-1942 
7. Publication of an especially good com- 
parative study of Cryptococcus hominis, Blasto- 
myces dermatitidisy and Coccidioides immitis 
by Benham.110 
8. Publication of the first case in Lousiana 
and the second case east of the Mississippi 
River, by McDonald.119 
1927 
1. Report from the California Department 
of Public Health139 summarizing work on 
coccidioidal granuloma in California for the 
years 1934 and 1935, including report by Dr. 
Joe Smith of Bakersfield, California, in which 
is described pulmonary or primary coccidioido- 
mycosis. In this report are summarized 450 
cases of coccidioidal infection in California, 
224 of which were fatal, thus establishing 
the mortality rate at about 46 per cent. 
1925 
1. First case reported from San Antonio, 
Texas. The patient was a Mexican, native to 
Texas. Report by Lehman and Pipkin.128 
2. Fennel12' reported the second case from 
Hawaii, the patient in this case being a native 
Hawaiian. He suggested that the infection may 
have developed from the handling of Cali- 
fornia fruit products. 
2. First report by Dickson141 describing a 
newly recognized entity, pulmonary or pri- 
mary coccidioidomycosis, as he had observed 
it in a laboratory infection. It was in this con- 
nection that Dickson made the suggestion that 
coccidioidal granuloma was most likely merely 
the advanced stage of a primary pulmonary 
infection, and that the lung is the usual portal 
of entry for C. immitis. It was at this time 
also that Dickson140 established the identity of 
“valley fever” of the San Joaquin Valley and 
primary pulmonary coccidioidal infection. 
i936 
I. Final clarification of the whole problem 
of North American and Brazilian “coccidi- 
oidal” granuloma by Jordon and Weid- 
man.134 In this paper a clear distinction is 
drawn between the Brazilian and the North 
American organism, Paracoccidioides bra- 
siliensis and Coccidioides immitis, respectively. 
The name Almeida’s disease is suggested for 
the Brazilian “coccidioidal” granuloma, which 
is now considered to be a form of blastomycosis. 
In 1934 Gifford had recovered C. immitis 
from the lung in one of her cases of “valley 
fever” with erythema nodosum. The record 
of this case was shown to Dickson in 1936 by 
Gifford. Gifford’s observation seems to have 
been the first objective demonstration of the 
etiologic relationship between “valley fever” 
and coccidioidal infection. This observation 
and Dickson’s own study of his laboratory in- 
fection formed the basis for the subsequent 
work which led to unquestioned recognition of 
the identity of coccidioidal granuloma and San 
Joaquin Valley fever. 
2. Description of 14 cases of coccidioidal 
meningitis by Abbott and Cutler.130 This paper 
also contains important statistical data relating 
to coccidioidomycosis as observed in the Los 
Angeles General Hospital, California. It is 
stated that in 12,000 autopsies 36 of coccidi- 
oidomycosis had been studied. This is the 
largest series of autopsies mentioned in the 
literature to this date, but it has not been 
published. 
3. First case published from Kansas, by 
Van Cleve.137 The patient was infected pre- 
3. Demonstration by Miller144 of various 664 
The Military Surgeon—November, 1946 
forms of pulmonary coccidioidomycosis, in- 
cluding (i) the development of caseous 
nodules in the lung, (2) pulmonary cavity 
formation, (3) caseating granulomatous pneu- 
monia with multiple cavities, (4) bronchial 
granulomatous scarring. Miller called atten- 
tion at this time to an observation that old 
fibrous lesions due to coccidioidomycosis may 
persist for years and then become active and 
give rise to dissemination of the organism to 
other parts of the body. 
cificity of the coccidioidin test and its independ- 
ence of the tuberculin reaction. 
2. Demonstration by Cox and Smith158 
of arrested pulmonary coccidioidal granuloma. 
Four cases were described in which old, en- 
capsulated, caseous, partially calcified coccidi- 
oidal lesions harboring typical organisms were 
found. Similar quiescent lesions were demon- 
strated in experimental animals. Sporulating 
organims were found in arrested human 
lesions of from 2 to 15 years’ duration. 
1938 
3. First case reported in a native of Tucson, 
Arizona, by Storts.168 This was a case of fatal 
meningitis, clinically resembling a brain tumor, 
in a child aged four. No other coccidioidal 
lesions were found in this child except a hard 
nodule in the lung, in which typical organisms 
were found. This case confirms the observa- 
tion of Cox and Smith relative to the dis- 
semination of virulent organisms from an old, 
presumably inactive lesion in the lung. 
1. Publication of the biennial report of the 
Kern County, California, Department of Pub- 
lic Health, in which Gifford150 records her 
observation of 1934 that C. immitis is re- 
coverable from cases of “valley fever” with 
erythema nodosum. As noted above, this ob- 
servation had previously been shared with 
Dickson. 
2. Report of the first large-scale study of 
intradermal skin reaction to products of C. 
immitis. This work was carried out in Cali- 
fornia by Hurwitz, Young, and Eddie,151 and 
involved tests of 449 individuals. 
4. Faber, Smith, and Dickson159 made an 
extensive report on the relationship between 
erythema nodosum and primary pulmonary 
coccidioidomycosis or “valley fever.” A de- 
tailed description of cases of “valley fever” 
with erythema nodosum was made, and it was 
shown that the infection with C. immitis ac- 
companied by erythema nodosum gives rise to 
complete immunity to further coccidioidal dis- 
ease. It was observed that generalized granu- 
loma developed in only one of 7 00 cases of 
primary infection with erythema nodosum. 
3. Further studies on the morphology and 
biology of C. immitis demonstrating the for- 
mation of spherules with their maturation and 
endospore formation from chlamydospores in 
in vitro cultures, by Lack.152 
4. Publication of the only recorded instance 
of valvular endocarditis in coccidioidomycosis, 
by Epstein.149 
5. Further studies on the relation between 
erythema nodosum and infection by C. im- 
mitis by Thorner.169 
5. Report of further studies of primary 
coccidioidomycosis by Dickson and Gifford,148 
and by Dickson,147 indicating further the rela- 
tionship between “valley fever” and coccidi- 
oidal granuloma. 
6. Demonstration by Phillips166 of the exist- 
ence of coccidioidal infection in Phoenix, 
Arizona, through the procedure of skin 
testing. 
6. Demonstration by Wooly156 of C. im- 
mitis in the sputum of patients being treated 
for pulmonary tuberculosis in Arizona. All 
cases with positive sputum were proved to be 
instances of coccidioidomycosis, and not tuber- 
culosis. 
7. Confirmation of the existence of coc- 
cidioidal infection in Arizona by Brown,15' 
who demonstrated positive skin reactions in 
cases of erythema nodosum occurring in this 
area. 
1929 
1940 
1. Report of an epidemiologic survey of the 
occurrence of coccidioidomycosis with ery- 
1. Demonstration by Kessel164 of the spe- Coccidioidomycosis 
665 
thema nodosum in Kern and Tulare Counties 
of southern California. In this report Smith175 
adds further support for a specific relationship 
between C. immitis infection and the erythema 
nodosum endemic in that area. 
by Emmons.195 In the area studied there was 
no record of coccidioidomycosis among the in- 
habitants, but Aronson previously had found 
that many of them exhibited positive coccidi- 
oidin skin reactions. Emmons suggested in his 
paper that wild animals may be the natural 
reservoir for C. immitis. 
2. Standardization of preparations of coc- 
cidioidin for use in skin testing for coccidioidal 
infection, by Stewart and Kimura.176 
2. Study of an epidemic of coccidioidomy- 
cosis of the primary pulmonary type in a group 
of students, by Davis, Smith, and Smith.19'1 
This is a careful clinical study of the same 
group of cases studied roentgenologically by 
Powers and Starks, as previously noted. 
3. Confirmation of the existence of coc- 
cidioidal infection, in southern Arizona, by 
Mills and Farness,1 3 who recorded the occur- 
rence of ten or more cases of coccidioidal dis- 
ease in that area. The existence of this focus 
was first indicated by Aronson, who found 
positive coccidioidin skin reactions in Prima 
Indians. 
3. Aronson and Gallagher186 determined 
the sensitivity to coccidioidin of 680 boys in 
an Eastern preparatory school, while studying 
reactions to the tuberculin test. Seventeen posi- 
tive reactors to coccidioidin were found, five 
of whom had calcified nodules in the lung. 
Twelve failed to react to tuberculin. This 
experiment centered attention upon the possi- 
bility that focal calcification in the lung ordi- 
narily attributed to tuberculosis might be the 
result of coccidioidal infection. 
4. Development of methods for the study 
of early primary lesions in the experimental 
animal, by Cronkite and Lack.170 
1941 
I. Report of a careful study of primary 
pulmonary coccidioidomycosis in an epidemic 
involving a group of 14 students who acquired 
the infection simultaneously while on a biology 
field trip in an endemic area. This report by 
Powers and Starks181 is based on x-ray studies, 
and for the first time records the changes that 
occur in the lungs in the course of the develop- 
ment and recession of the pulmonary reactions 
to C. immitis. Correlation of these findings 
with those of previous studies of “valley fever,” 
erythema nodosum, etc., further confirmed the 
identity of primary coccidioidomycosis of the 
lung and these disorders. 
4. Aronson, Saylor, and Parr187 demon- 
strated conclusively that calcified nodules in 
the lung occur in a high percentage of persons 
with positive coccidioidin and negative tuber- 
culin reactions. This confirmed previous ob- 
servations that calcified nodules in the lung 
arise from previous and healed infection with 
C. immitis. 
5. Stiles and Davis206 found that sheep, 
cattle, dogs, and wild rodents not infrequently 
are infected in areas where coccidioidomycosis 
is endemic in man. 
2. Careful x-ray study of cavity formation 
in primary pulmonary coccidioidomycosis by 
Winn.185 
6. Schenken and Palik201 made a careful 
analysis of all the reported cases of coccidioido- 
mycosis and concluded that only 17 cases on 
record could be proved conclusively to have 
occurred independent of contact with Cali- 
fornia. The sources of infection of 12 of the 
17 non-California cases were found to be 
Texas, Arizona, and New Mexico. These 
studies indicate therefore that up to this time 
only five cases of conclusively proved coccidi- 
oidomycosis had occurred outside of California, 
Texas, Arizona, and New Mexico. The cri- 
teria used in this study are extremely rigid. 
3. Demonstration by Thorner184 of a very 
low ratio between the incidence of “valley 
fever” and positive coccidioidin reactions in 
267 school children in Kern County, Cali- 
fornia, indicating that “valley fever” is a very 
benign and perhaps universal infection in that 
area. 
1942 
1. The isolation of C. immitis from the 
soil and from rodents in San Carlos, Arizona, 666 
The Military Surgeon—November, 1946 
7. Winn and Johnson209 in concluding a 
report of a study by x-ray of primary coccidi- 
oidomycosis made the following significant pre- 
diction : “The coming of large numbers of un- 
infected persons, especially military personnel, 
into the inland California valleys and certain 
parts of Arizona and Texas will result in a 
proportional increase in the incidence of coc- 
cidioidomycosis.” The experiences of the 
armed forces during the following four years 
fully justified this prediction. 
2. Report of studies of the chemical com- 
position of a specific polysaccharide found in 
culture filtrates of C. immitis} by Hassid, 
Baker, and McCready.214 This polysaccharide 
was found to give a positive skin reaction and 
to act as specific antigen for the precipitin 
reaction, but not for the complement-fixation 
reaction. 
3. An excellent review of the morphology, 
taxonomy, and distribution of C. immitis} pub- 
lished by Baker, Mrak, and Smith.210 This 
paper also contains an excellent and compre- 
hensive review of all previous work done on 
the general biology of this organism. 
8. Experimental inoculation of mice intra- 
nasally and intraperitoneally with C. immitis 
by Taeger and Liebow.207 This report deals 
especially with the chemotactic properties of 
different forms of C. immitis as they appear 
in the tissues. 
4. The second report on the occurrence of 
coccidioidomycosis in military installations, by 
Goldstein and Louie.213 This paper describes 
an epidemic of 75 cases of coccidioidomycosis, 
which occurred in a body of troops on detail in 
an endemic area. The report is important be- 
cause of its contribution relating to primary 
pulmonary infection. Up to the time of publi- 
cation of the report the granulomatous or dis- 
seminated form of the disease had developed in 
only one of the 75 patients involved. 
9. Futher confirmation of the occurrence 
of coccidioidomycosis in areas of Texas ex- 
tensively used for military purposes. These re- 
ports are by Caldwell,191 Smith,200 Martin,198 
and Schulze.202 
10. Discovery by Ashburn and Emmons188 
of a form of granulomatous disease in wild 
rodents produced by Ha'plosforanglum farvum 
and sometimes associated with coccidioidomy- 
cosis. This observation added a new problem 
to the epidemiology of coccidioidomycosis. 
1944 
1. Report by Colburn220 on the x-ray find- 
ings in the epidemic coccidioidomycosis among 
Army personnel, described by Goldstein and 
Louie, This is the fourth report on the occur- 
rence of coccidioidomycosis in military per- 
sonnel. It is important from the point of view 
of primary pulmonary infection. 
D. The Period of Military Injection and Dis- 
semination of the Disease, 1942- 
1942 
First report from military installations in 
California dealing with coccidioidomycosis. 
This report by Shelton203 records the develop- 
ment of positive coccidioidin reactions in troops 
brought into that area within the first three 
months of residence. Several cases of coccidi- 
oidomycosis developed in Camp Roberts within 
the three-months period, although the area in 
which the camp was situated was not known to 
be one of endemic infection. 
2. A follow-up study by Goldstein and 
McDonald"222 of the 75 cases recorded in the 
epidemic described in the two preceding papers. 
This report contains excellent clinical data 
relating to primary pulmonary infection and 
records ten additional cases in Army personnel. 
In all of these the disease was contracted near 
the California-Arizona border, an area in 
which coccidioidomycosis was not at that time 
known to be endemic. 
1943 
1. Further studies of coccidioidomycosis in 
wild rodents, by Emmons.212 Again he sug- 
gested the possibility of determining the ex- 
tent of endemic areas of coccidioidomycosis 
through the examination of wild animals. 
3. Publication by McKenney, Traum, and 
Bonestell226 of a report on the occurrence of 
coccidioidomycosis in a mountain gorilla in the 
Zoological Museum at San Diego. It is 
recorded in this paper that other anthropoids, C occidioidomy costs 
667 
American monkeys, dying in this same 
museum, had been found to be infected by 
C. immitis. 
coccidioidomycosis from the clinical and roent- 
genologic points of view. 
4. The fifth report of coccidioidomycosis in 
military personnel, by Quill and Burch.229 
This report describes the disease in Texas 
installations and reviews all of the previously 
described cases originating in Texas. They are 
listed by location as follows: 
1945 
I. The eighth report on coccidioidomycosis 
occurring in military personnel and the first 
report published from the Army Institute of 
Pathology dealing with the pathologic anatomy 
of coccidioidomycosis. This report by Schlum- 
berger236 describes 13 cases of coccidioidomy- 
cotic meningitis; these are included in the 
95 cases of this present report. 
San Angelo 4 
Dallas 7 
El Paso 5 
San Antonio (civilian) 6 
Cases 
2. The ninth report of coccidioidomycosis 
in Army personnel, by Bass.231 This is a re- 
port of four cases from the Bruns General 
Hospital, Santa Fe, New Mexico; it deals 
especially with roentgenologic diagnosis of the 
disease. 
5. Brief summary of experiences with 
coccidiodomycosis in the Western Flying 
Training Command of the Army, by Lee.224 
This is the sixth report dealing with coccidio- 
idomycosis in military installations. 
6. The seventh report on coccidioidomy- 
cosis in Army installations, by Cheney and 
Denenholz.218 This report presents a clinical 
study of 14 cases of coccidioidomycosis which 
occurred at Hammond General Hospital in 
California. The material presented is excellent 
from the clinical, immunologic, and bacteri- 
ologic points of view. Of the 14 cases, ten 
were of the primary benign pulmonary type, 
and four of the disseminated form. 
3. The tenth report of coccidioidomycosis 
in Army personnel, by McCracken.235 This 
study deals with four cases of coccidioidomy- 
cosis in soldiers at Fort Bragg, North Carolina. 
Three of the patients had trained in California 
and one in Arizona. The disease appeared in 
the three patients from California while they 
were en route to Fort Bragg. The dissemi- 
nated form of the disease developed in only 
one case, some months after arrival at Fort 
Bragg. 
7. Detailed report of studies on coccidio- 
idomycosis carried out for the Commission on 
Epidemiological Survey, Board for the In- 
vestigation and Control of Influenza and 
Other Epidemic Diseases in the Army, Pre- 
ventive Medicine Division, Office of the Sur- 
geon General,219 This is an unpublished re- 
port covering a comprehensive study of coc- 
cidioidomycosis occurring at Minter, Gardner, 
Lemore, and Merced airfields in California. 
4. The eleventh report of coccidioidomy- 
cosis in Army personnel by Kunstadter and 
Pendergrass,234 from Ashford General Hos- 
pital, West Virginia. In this paper attention is 
called to the occurrence of an arrested form of 
Coccidioidomycosis, characterized by the pres- 
ence of multiple or solitary caseous nodules in 
the lung. In the case reported there was no 
active coccidioidomycosis. The quiescent pul- 
monary lesions were found incidentally in the 
course of an autopsy following death from an 
accident. 
8. Publication of a syllabus on coccidioido- 
mycosis by Lee and Nixon, with extensive 
roentgenograms by Jamison,223 for the use of 
the medical officers of the Army Air Forces 
Western Flying Training Command. This 
syllabus, originally prepared and distributed 
in 1942, describes the coccidioidomycosis con- 
trol program in the Western Flying Training 
Command of the Army Air Forces and also 
presents a brief but complete description of 
Ill 
Distribution of Cases 
A. Origin of the Cases 
The several sources of materials made use of 
in this study are shown in Table I. It is im- 
portant to note that this table gives only in a 668 
The Military Surgeon—November, 1946 
general way the geographic location in which 
the infections occurred, and in some instances 
is no indication whatever of the origin of the 
infection. This is especially true when the cases 
were reported from hospitals situated outside 
of the United States, It will be noted from 
Table I that in nine cases the disease developed 
in apparently normal individuals after they 
had left this country for active duty in foreign 
theaters of war. The table also indicates that 
dissemination of the infection within this coun- 
try may be in progress through the movements 
the clinical records of the Army medical de- 
partment will give a far more accurate picture 
of the degree and extent of this increase in 
coccidioidomycosis, our study of a small group 
of 95 cases, 50 of which were fatal and all 
of which were of the disseminated type, gives 
a clear indication of what occurred. The 
origins of the 95 infections with C. immitis re- 
ported here are summarized in Table II. 
These data do not provide a satisfactory 
basis for discussion but it is rather interesting 
that the number of cases in this series origi- 
nating in Texas is slightly in excess of the 
total number of cases of coccidioidomycosis 
reported as having occurred in that state up to 
October 1942. It will be recalled that the 
disease was not recognized in Texas until 
Caldwell published his paper in 1932 (two of 
the five cases he reported had, according to 
him, been published previously, but he does 
not give a date of publication.) The largest 
group of cases is reported from California, 
which is in harmony with the prevalence of 
the disease in that area. The relatively large 
number from Arizona is interesting in view of 
the fact that the disease was not recognized 
in that state until 1938, when Wooly observed 
the first case in Tucson. This was followed in 
1939 by the demonstration by Phillips of a 
high incidence of positive coccidioidin skin tests 
in the region around Phoenix. Another sig- 
nificant observation made the same year by 
Brown was the occurrence of a number of 
cases of erythema nodosum in this same area. 
Table II 
Geographic Site of Origin 
Total 
Number 
Arizona 
24 
California 
37 
California and/or Arizona 
4 
California and/or Texas 
1 
Texas 
12 
New Mexico 
1 
Colorado 
1 
Utah 
1 
Oregon 
1 
Missouri 
1 (?) 
Unknown origin 
12 
Total 
95 
of subclinically infected persons from one 
Army installation to another. This is shown 
specifically by the occurrence of four cases of 
coccidioidomycosis at Fort Bragg, North 
Carolina, in an area where the disease had 
never before been seen. In this group of cases, 
reported by McCracken, three of the patients 
actually were suffering from active coccidi- 
oidomycosis on arrival at Fort Bragg from Cali- 
fornia. The fourth had been in Arizona, but 
the disease appeared only after several months 
of residence at Fort Bragg. Whether this ex- 
perience is an indication of what may be ex- 
pected on a far greater scale in the future re- 
mains, of course, to be seen. 
The data in this series of 95 cases relating 
to the age and sex incidence of the disease are 
valueless in view of the highly selective type 
of material with which we are concerned. 
Most of the cases were among the age groups 
of greatest numerical strength in the armed 
forces. In only one instance was the infection 
present in a woman, the wife of a soldier. 
The occupational factor, which has appeared 
to be of significance in other series of cases, 
was almost impossible to assess. The data on 
branch of service and military rank, the only 
suggestions of occupational status in the pro- 
tocols, were meaningless. 
It was to be expected that the utilization 
of the Southwest for military training purposes 
would increase the incidence of coccidioidomy- 
cosis in the endemic areas. While, of course, 
The place of birth was unknown in more C occtdiotdomy costs 
669 
than a third of the cases and added no helpful 
information in those in which it was known. 
open lesion by cultural examination or in his- 
tologic preparations of excised tissue. The com- 
pilation of the data from the 45 cases of dis- 
seminated coccidioidomycosis diagnosed by 
biopsy will therefore serve a useful purpose for 
those who are concerned with the diagnosis of 
the disease. 
Exclusive of the geographic sites of origin, 
only the figures on the race of patients with 
disseminated coccidioidomycosis appear signifi- 
cant. 
B. Race of the Patients 
In considering the figures pertaining to this 
group, it is important to keep in mind that all 
the cases presented an advanced stage of the 
coccidioidomycotic process and thus are es- 
Table III summarizes the data relating to 
the factor of race in the development of the 
disseminated form of coccidioidomycosis. 
There is a common belief, indicated by the 
widespread publication of statements to this 
effect, that dark-skinned peoples are more sus- 
ceptible than white to coccidioidal infection, 
particularly to the disseminated form of the 
disease. Our data give some support to the 
general opinion that colored Americans are 
more apt to develop the disseminated form of 
coccidioidomycosis than are other races, but 
the significance of the data is debatable. It 
may be added that insofar as the general 
clinical and pathologic-anatomic character 
of the disease is concerned there are no appre- 
ciable differences between the infections in 
colored and white Americans. 
Table III 
The Race Factor in 95 Cases 
Total 
Number 
White, American 
37 
Colored, American 
48 
Mexican 
2 
Filipino 
1 
American Indian 
1 
Oriental (Chinese) 
1 
Unknown race 
5 
Total 
95 
sentially like those studied at autopsy. They 
represent, nevertheless, what might be re- 
garded as an intermediate stage in the develop- 
ment of coccidioidal infection, and hence de- 
serve special study, since data obtained from 
this group throw light upon the natural course 
of the disease. Furthermore, some of these 
cases have come under observation so early that 
it is reasonable to consider them primary infec- 
tions. The study of these cases helps to deter- 
mine more accurately the portal of entry of 
C. immitis. In contrast to those of short dura- 
tion, there are cases in this group which, be- 
cause of the long duration, represent unques- 
tioned reinfection of internal original. A study 
of the latter group furnishes information on 
the resistance of the individual to coccidioidal 
infection and the process of healing of the foci 
of primary inoculation. 
IV 
A. Data from Analysis of 45 Surviving Cases 
of Coccidioidomycosis Diagnosed by Biofsy 
of Disseminated Lesions 
Clinical Features 
The diagnosis of coccidioidomycosis, even of 
the extensively disseminated type, is not an 
easy matter, because of the resemblance of the 
lesions and of the disease as a whole to a 
variety of chronic granulomatous infections. 
The three most important diseases with which 
it may be confused are tuberculosis, blastomy- 
cosis, and actinomycosis. In areas where coc- 
cidioidomycosis is endemic, its presence is 
always suspected when a patient has widely 
disseminated suppurative lesions involving the 
bone and the subcutaneous tissues. Outside of 
endemic localities such processes rarely indicate 
any one etiologic entity, and the diagnosis is 
therefore not made until the specific organ- 
isms are demonstrated in exudate from the 
In accordance with the general mortality 
statistics, it is to be expected that approximately 
40 per cent of this group of cases in which 
biopsy has been performed will be fatal. 670 
The Military Surgeon—November, 1946 
Table IV 
Diagnosis by Biopsy 
Lesion 
Excised 
Organisms 
Demonstrated 
Lesions in 
Other Organs 
Serological 
Reactions 
Duration 
Days Cases 
No. of 
Cases 
Skin and subcutane- 
In sections 
16 
Lung 
3 
Coccidioidin 
2 
1- 30 
— 
17 
ous tissues 
By culture 
1 
Nodes 
1 
Complement 
1 
31- 60 
4 
By an. inoc. 
3 
Precipitin 
1 
61- 90 
4 
Not demons. 
1 
Sed. Rated- 
2 
91-120 
3 
X21-150 
1 
151-180 
I 
181-2x0 
■— 
301 
I 
Unknown 
3 
Lymph nodes 
In sections 
10 
Lung 
3 
Coccidioidin 
2 
1- 30 
1 
10 
(-f skin: 4) 
By culture 
— 
Skin 
4 
Complement 
1 
31- 60 
2 
By an. inoc. 
— 
Precipitin 
1 
61- 90 
— 
Sed. Rate+ 
1 
91-120 
— 
121-150 
1 
151-180 
1 
181-210 
1 
240 
1 
400 
1 
Unknown 
2 
Bones 
In sections 
10 
Lung 
I 
Coccidioidin 
1 
1- 30 
1 
10 
(+skin; 1) 
By culture 
— 
Skin and Sub- 
Complement 
1 
31- 60 
2 
(+node: 1) 
By an. inoc. 
1 
cut. Tissue 
1 
Precipitin 
— 
61- 90 
2 
(+joint: 1) 
Joint, knee 
1 
Sed. Rated- 
— 
91-120 
121-150 
— 
151-180 
— 
181-210 
2 
211-240 
— 
241-270 
1 
480 
1 
Lung 
In sections 
3 
Coccidioidin 
1 
1- 30 
1 
5 
By culture 
— 
Complement 
— 
31- 60 
— 
By an. inoc. 
— 
Precipitin 
— 
61- 90 
1 
Sed. Rated- 
1 
91-120 
— 
1080 
1 
1175 
1 
3240 
1 
Peritoneum 
In sections 
1 
Retroperito- 
Coccidioidin 
— 
Unknown 
1 
1 
neal nodes 
I 
Complement 
— 
Precipitin 
— 
Sed. Rate+ 
— 
Meninges 
In sections 
1 
Nodes 
I 
Coccidioidin 
— 
360- 
1 
2 
(+node: 1) 
By culture 
1 
Unknown 
1 
By an. inoc. 
1 
In Table IV are summarized the essential cosis on record with survival to the present 
data available on the 45 cases of coccidioidomy- time. C occidioidomy costs 
671 
Table IV furnishes useful information re- 
lating to the diagnosis of the disease by biopsy 
and to the study of the general character of 
disseminated coccidioidomycosis, as summa- 
rized in the following paragraphs. 
brain tumor in one instance and cord neo- 
plasm in the other. 
3. Demonstration of the organisms in ex- 
cised tissues: The finding of typical spherical 
forms of C. immitis in the tissues is the critical 
diagnostic point; without this the diagnosis of 
coccidioidomycosis cannot be made unless the 
fungus is obtained in culture or by inocula- 
tion into the experimental animal. The histo- 
logic character of the lesion varies greatly 
and is not specific for C. immitis. The record 
of demonstration of the organism in the tissues 
by cultures and by inoculation in this series 
of cases is as follows: 
I. Duration of the disease until time of 
biofsy: The shortest period over which symp- 
toms had been observed prior to biopsy was five 
days, the longest 3,240 days. Twenty-nine 
(68 per cent) of the cases had a maximum 
duration of 21 o days or less with an average of 
87 days. In only 8 cases was there definite in- 
formation that symptoms had been present 
longer than 210 days. 
2. Distribution of lesions excised for diag- 
nosis: 
Cases 
Cases 
Organisms demonstrated in section ...41 
Organisms demonstrated by culture 3 
Organisms demonstrated by animal inoculation . 5 
Organisms not demonstrated 1 
Skin and subcutaneous tissues only 17 
Lymph nodes (with skin in 4 cases) 10 
Bones (with skin, joint tissues, and lymph 
nodes in one case) 10 
Lungs (lobectomy or thoracotomy) 5 
Peritoneum (omentum) 1 
Meninges (laminectomy or craniotomy; with 
lymph nodes in one case) 2 
In the last case just cited the diagnosis was 
based on the combination of a specific immune 
reaction, a typical clinical history, and a study 
of the histologic reaction. Ordinarily there 
is no difficulty in the recognition of the organ- 
isms in tissue sections. In the long-standing 
chronic lesion, however, the organisms tend 
to be scarce and often show extensive degen- 
erative alteration which makes their identifi- 
cation uncertain and sometimes impossible. In 
such cases cultures and animal inoculations 
are required. The reliability of the study of 
excised tissues for diagnosis is indicated by 
demonstration of the etiologic agent in the 
tissue in 41 of the 45 cases studied. 
In all 17 of the cases in which the skin and 
subcutaneous tissues were excised the skin 
lesions consisted of ulcerated or non-ulcerated 
nodules or of indurated sinuses leading from 
deeply situated lesions, often in the bone. On 
exploration, lesions in the bone were usually 
found to be large, destructive, and suppurating 
processes involving either the medullary cavity 
or the cortex with the overlying periosteum. 
They were found in a great variety of bones, 
as indicated in Tables IV and XIII. 
The lymph nodes excised for diagnosis were 
sometimes secondarily related to regional 
superficial lesions, but most often they were 
involved in the course of dissemination of the 
organisms by way of the blood. 
4. Distribution of lesions in the biofsied 
cases: Granulomatous lesions were demon- 
strated regularly at exploration for diagnosis 
in the 37 cases with superficial manifestations. 
Lesions were most consistently found in the 
lung, an observation which tends to support 
Ophiils’ very early contention that the lungs 
are the portal of entry in the vast majority of 
cases. The primary lesion in this tissue, for 
many years referred to as “coccidioidal granu- 
loma,” serves as the source for either early 
or late dissemination of the organisms. The 
significance and the importance of the primary 
pulmonary lesion will be discussed in a subse- 
The biopsied lung lesions consisted of soli- 
tary foci of granulomatous tissue, which in 
roentgenograms bore striking resemblance to 
genuine neoplasms of the lung. The lesions 
were excised because of the diagnosis, and it 
was surprising to find them to be old, partially 
healed coccidioidomycotic lesions. The men- 
ingeal biopsies were performed for suspected 672 
The Military Surgeon—November, 1946 
quent section on the pathologic anatomy of this 
disease. 
and particularly of the portal of entry of the 
organisms. In view of the fact that only 
slightly over half of the cases covered in this 
report were studied at autopsy, a careful 
survey of the charcter and distribution of the 
lesions as ascertained by roentgenologic and 
other clinical techniques has been made. The 
roentgenologic findings, particularly those in 
the lungs, have proved most helpful in the 
study of the disease in our cases since repeated 
previous roentgenologic studies have provided 
good leads toward an understanding of the 
pathogenesis and course of development of 
these lesions. The necessarily limited value of 
the clinical findings is enhanced by the correla- 
tion of data obtained in the post-mortem studies 
of the fatal cases. 
5. Immunologic and serologic data: This 
group of data is not sufficiently complete to 
lend itself to discussion. In brief, the coccidioi- 
din skin test was positive in six of the cases; 
complement-fixation was obtained in three; 
the precipitin reaction was positive in signifi- 
cant dilution in two cases; and the sedimenta- 
tion rate was elevated in five. Data of this 
kind are important in the diagnosis when the 
morphologic and cultural identification of 
suspected organisms is uncertain. 
B. Clinical Data from 95 Cases (50 Fatal 
and 45 Surviving) of Disseminated 
C occidioidomycosis 
The clinical data from the entire series of 
95 cases have been treated as a whole because 
all actually belong to the same category, being 
I. The duration of the disease in 50 fatal 
cases of coccidioidomycosis: The shortest course 
of the disease in the 50 fatal cases was found 
to be 25 days, the longest 467 days. Thirty- 
six, or over 70 per cent, ran a course of 210 
days or less; and the average duration was 
104 days (100 days taking the median). If, 
for purposes of studying the character, time, 
and extent of dissemination of the lesions, etc., 
the figures for the biopsied cases and those for 
the fatal cases be combined (an entirely justifi- 
able procedure in view of the disseminative 
character of the disease in both groups of 
cases), the average duration of the disease in 
the combined series of 65 cases running a 
course of 210 days or less is found to be 95 
days. The figures are represented in the fol- 
lowing summary. 
Table V 
Duration of 50 Fatal Cases of Disseminated 
Coccidioidomycosis 
Total 
Number 
1-30 days 
2 
31-60 days 
6 
61-90 days 
9 
91-120 days 
7 
121-150 days 
5 
151-180 days 
5 
181-210 days 
2 
211-240 days 
3 
241-270 days 
0 
271-300 days 
3 
301-330 days 
2 
451-480 days 
2 
Unknown duration 
4 
Total 
5° 
In 29 cases in which biopsy succeeded onset 
within 210 days or less, the average duration 
of the disease processes in surviving patients 
was 87 days. 
In 36 fatal cases in which death occurred in 
210 days or less, the average duration of the 
disease processes in patients dying of the disease 
was 104 days. 
characterized by either a limited or widespread 
dissemination of lesions throughout the body. 
Only the most significant of the clinical mani- 
festations have been selected and catalogued. 
Particular attention has been paid to the initi- 
ating signs and symptoms because they obvi- 
ously contribute to an understanding of the 
farly stages in the pathogenesis of the disease 
In 65 cases of 210 days duration or less, 
the average duration of the disease processes 
was 95 days. 
It is necessary to emphasize that the aver- 
age duration figure of 95 days in the 65 cases 
of 210 days duration or less refers to the dis- C occidioidomy costs 
673 
ease -processes and has no meaning with regard 
to the average length of the course of the 
disease in patients recovering from or dying of 
coccidioidomycosis. The figure is therefore use- 
ful only in connection with a study of the 
development of the pathologic processes con- 
cerned in the disease entity, as suggested 
above. 
2. The initiating signs and symptoms in 95 
cases of disseminated coccidioidomycosis: The 
signs and symptoms most often present at the 
onset of this disease point definitely to a 
lesion in the chest. The patient usually com- 
plained of cough, chest pain, and chest cold. 
Next in frequency were symptoms such as 
chills, fever, malaise, and anorexia. In about a 
third of the cases these two groups of signs 
and symptoms were combined with a swelling 
situated in the skin or subcutaneous tissue. 
These data, especially when derived from the 
cases of short duration, indicate definitely that 
the portal of entry of the organism is the 
respiratory tract. Furthermore, they show also 
that dissemination of the organisms into distal 
portions of the body may occur while the 
primary pulmonary lesion is in its earliest stages 
of development or is rapidly extending to 
involve the lung as a whole. This observation 
is supported by repeated roentgen examinations 
of the chest, especially in the cases of shortest 
duration. The initiating signs and symptoms in 
the 95 cases under consideration and their 
frequency of occurrence are shown in Table 
VI. 
Initiating Signs and Symptoms in 95 Cases (50 Fatal 
and 45 Surviving) of Disseminated 
Coccidioidomycosis 
Table VI 
Total 
Number 
Pain: 
5o 
Chest 
21 
Headache 
ii 
Joints 
5 
Back 
4 
Abdomen 
4 
Generalized ache 
3 
Bones 
2 
Swelling: 
27 
Cutaneous nodule 
II 
Subcutaneous mass 
7 
Bone 
6 
Joint 
2 
Abdomen 
I 
Cough 
30 
Fever 
23 
Cold in chest 
11 
Chills 
8 
Malaise 
6 
Weakness and fatigue 
6 
Hemorrhagic sputum 
5 
Nausea and vomiting 
5 
Hoarseness 
3 
Stiff neck 
3 
Sweating 
3 
Ulcer, chronic (sinus) of skin 
3 
Weight loss 
3 
Anorexia 
2 
Diplopia 
2 
Dyspnea 
2 
Sore throat 
2 
Confusion, mental 
1 
Dizziness 
1 
Drowsiness 
1 
Hemorrhage, pulmonary 
1 
J aundice 
1 
Photophobia 
1 
Staggering gait 
1 
Urinary retention 
1 
3. General symptomatology: The list of 
symptoms and signs making up Table VII in- 
cludes and supplements the data in Table VI. 
Although this tabulation gives some idea of 
the general clinical effect of coccidioidal infec- 
tion, the data are not sufficiently abundant or 
comprehensive to justify analysis. The figures 
in this table are not so significant as those in 
Table VI, the initiating symptoms, which were 
carefully determined in all cases of the series. 
Nevertheless, the general symptomatology, as 
pictured by this accumulation of data, does 
indicate the existence of significant infection 
in the lungs in a majority of the cases. 
The signs and symptoms referable directly 
to the various disseminated lesions, individually 
or in groups, are not included in this table of 
general symptomatology. With the exception 
of those pointing toward disease in the lungs, 
the most prominent group of signs and symp- 
toms indicate intracranial involvement, espe- 
cially of the meninges. This is not surprising in 
view of the relatively large number of cases 674 
The Military Surgeon—November, 1946 
of coccidioidal meningitis encountered in this 
study. 
not present at the time of autopsy, and the 
gross lesions were found only in the lungs 
and the meninges. Only microscopic lesions 
were found elsewhere, and these were re- 
stricted to the spleen. The pulmonary lesion 
was a solitary focus well advanced toward 
healing; the more important and immediately 
fatal process was in the meninges. The dura- 
tion of the illness in this case was 467 days. 
Erythema nodosum developed very early, as 
it usually does. Obviously this case is excep- 
tional, since the development of erythema 
nodosum did not protect this patient against 
further injury by C. immitis. 
Especially worthy of emphasis is the almost 
complete absence of erythema nodosum in 
this group of cases. This cutaneous manifes- 
tation of the disease has been shown by the 
California workers to be frequent in cases of 
Table VII 
General Symptoms'and Signs in 95 Cases (50 Fatal 
and 45 Surviving) of Disseminated 
Coccidioidomycosis 
Total 
Number 
Pain: 
Chest 
30 
Headache 
29 
Joints 
16 
Back 
13 
Abdomen 
2 
Fever 
5° 
Cough 
33 
Weight loss 
32 
Stiff neck 
23 
Weakness, malaise, fatigue 
19 
Drowsiness, coma 
i? 
Nausea and vomiting 
15 
Paralysis and paresis 
15 
Sputum, excessive 
15 
Dyspnea 
12 
Chills 
7 
Sweating 
6 
Sore throat 
6 
Reflexes, hyperactive 
5 
Hemorrhagic sputum 
4 
Anorexia 
2 
Visual disorders 
2 
Speech disorders 
2 
Staggering gait 
I 
Dizziness 
I 
Incontinence 
I 
Erythema nodosum 
I 
4. Physical findings in this series: In the 
physical examination of this group of cases, 
disseminated lesions were found in a great 
variety of locations. In each instance the le- 
sions produced their own peculiar physical dis- 
turbances, and therefore it is neither practical 
nor useful to attempt a detailed analysis of 
these findings. Certain general data relating 
to the state of the individual as a whole are, 
however, worth mentioning and commenting 
upon, as follows: 
a. Loss of weight was a striking feature, 
often amounting to as much as 20 or 30 pounds 
in the course of a few weeks. 
b. Fever occurred in all cases at some time, 
the temperature usually ranging between ioo° 
and 103° F., rarely higher. 
c. The pulse rate was found to range 
around 100 or less, with 130 as the highest 
mark. 
d. No striking increase in respiratory rate 
was noted in any of the cases, the usual rate 
being between 20 and 30 per minute. 
e. Blood pressure was never significantly 
altered. 
primary pulmonary coccidioidomycosis. It has 
been fully established that the disseminated 
form of the disease rarely follows if erythema 
nodosum is present in the early stages; further- 
more, a permanent immunity to subsequent 
C. immitis infection seems to be acquired. It 
is interesting to note, however, that in the only 
fatal case in this series in which erythema 
nodosum did occur the infection was dis- 
seminated. The skin manifestations were 
f. The physical findings in the lungs varied 
enormously from case to case. When present, 
they consisted of (1) rales, (2) muco- 
purulent sputum, sometimes blood-streaked, 
(3) evidences of consolidation of a patchy or 
sometimes lobar distribution, (4) evidences of 
pleural effusion, usually minimal but some- 
times massive, (5) occasional evidences of 
cavity formation. In the absence of physical 
findings, the condition of the lungs was most Coccidioidomycosis 
675 
often revealed only by roentgenograms, which 
demonstrated one or more of the following 
features: (i) hilar and mediastinal lymph 
node enlargement (Figs. 2, 3, and 4), (2) 
prominent bronchial markings, (3) focal or 
lobular consolidation (Figs. 2, 3, and 4), 
(4) homogeneous diffuse lobar consolidation, 
(5) focal, miliary and nodular consolidation, 
(6) solitary areas of consolidation with evi- 
small papular nodules 5 to 10 mm. in 
diameter, superficially situated, (2) ulcerated 
nodules, (3) multiple verrucous lesions, (4) 
pustular furuncular nodules, (5) ulcerations 
representing the opening of sinuses leading 
from the subcutaneous tissues or the more 
deeply situated structures (Fig. 32), (6) 
chronic indurated ulcers, (7) erythema 
nodosum, (8) large subcutaneous suppurative 
Fig. 2. Primary pulmonary coccidioidomycosis, AIP Ac. 83209: Roentgenogram of unilateral diffuse 
confluent bronchopneumonia with cavity formation, accompanied by massive mediastinal lymphadenitis. 
Death on 60th day of the disease with widespread dissemination of lesions. Gross appearance of the lung was 
not unlike that shown in Figs. 7 and 8. 
FlG. 3. Primary pulmonary coccidioidomycosis, AIP Ac. 117769: Roentgenogram of unilateral lobular 
and confluent bronchopneumonia, somewhat wedge-shaped, accompanied by massive bronchial and media- 
stinal lymphadenitis. Death on mth day of the disease, with meningitis and other disseminated lesions. 
dence of encapsulation, (7) solitary areas of 
consolidation with centrally situated thin- 
walled cavities (Fig. 5), (8) multiple foci of 
calcification in the absence of any other change 
in the lung, (9) fibrous thickening of the 
pleura, (10) pleural effusion, either massive 
or minimal, (11) solitary dense tumor-like 
consolidation. 
granulomatous masses or swellings, usually 
communicating with the surface of the skin 
by multiple sinuses and with the more deeply 
situated tissues, especially the bones, by com- 
plex infected and ramifying channels. 
h. The bones clinically observed to be in- 
volved in these cases were as follows: rib, 
scapula, skull, clavicle, patella, vertebra, 
phalanges, tibia, sternum, ilium, sacrum, and 
ischium. As revealed by roentgenograms, the 
g. The lesions discovered in the skin and 
subcutaneous tissues were as follows: (1) 676 
The Military Surgeon—November, 1946 
bone lesions are sharply demarcated, but of a 
destructive character. They may be situated 
in the medullary canal, in the cortex or just 
beneath the periosteum; and they are not con- 
fined to any particular portion of the bones 
(Fig. 29). Complicated sinuses usually form 
communications between the lesions in the 
cular symptoms are usually not referable to 
lesions within the muscles. However, muscular 
lesions were often observed, particularly in the 
psoas muscle, in which case they presented the 
same clinical manifestations as the psoas 
abscess of tuberculosis. 
k. The usual physical signs of meningitis 
Fig. 4. Endogenous reinfection pulmonary coccidioidomycosis, AIP Ac. 150192: Roentgenogram of 
bilateral widespread diffuse and focal consolidation of the bronchopneumonia type, accompanied by bronchial 
and mediastinal lymphadenitis. Death on 274th day with widespread dissemination of lesions. Gross appear- 
ance of lung not unlike that shown in Figs. 6 and 7. 
Fig. 5. Endogenous reinfection pulmonary coccidioidomycosis, AIP Ac.150192: Roentgenogram of same 
lung as in Fig. 4, posed to show one of several cavities situated in the right upper lobe. 
bone and the overlying soft tissues, finally 
discharging their contents through the skin 
(Fig. 30). In some instances the bone lesions 
are tumor-like in their general appearance, 
being readily mistaken for genuine neoplasms. 
were demonstrated in a considerable propor- 
tion of the cases under discussion: diplopia, 
stiffness of the neck, positive Kernig’s sign, 
and other indications of meningeal involve- 
ment. 
i. The joints were not frequently the site 
of the granulomatous process, but in some 
cases lesions were seen in the knee, the elbow, 
the ankle, the sterno-clavicular, the claviculo- 
acromial, and the sacro-iliac junctions. In- 
volvement of the sterno-clavicular joint was 
noticeably frequent. The infected joints 
usually were found to communicate with 
adjacent bone lesions. 
1. Enlarged lymph nodes were found in 39 
of the 95 cases, and usually were those drain- 
ing infected areas. The mediastinal nodes were 
the most frequently involved, as indicated by 
both physical and roentgenologic examination 
of the chest. Sometimes the enlarged nodes 
were found to undergo suppuration, with the 
development of sinuses leading to the over- 
lying skin. 
j, The rigidity, weakness, and other mus- 
m. Enlargement of the liver, suggesting C occidioidomy costs 
677 
abscess formation, was observed in one case. 
n. Typical signs of chronic peritonitis were 
observed in two instances, both of which 
proved to be peritoneal infection with 
C. hnmitis. 
Table IX 
Hematologic Findings in Fatal and Surviving 
Cases of Disseminated Coccidioidomycosis 
Average 
RBC in 38 cases (average of lowest 
counts) 
3.5 million 
Hemoglobin in 32 cases (average of high- 
est counts) 
64 percent 
WBC in 45 cases (average of highest 
counts) 
14,400 
Neutrophils in 45 cases (average of high- 
est counts) 
73.5 percent 
Eosinophils in 22 cases (average of high- 
est counts) 
8.6 percent 
Monocytes in 10 cases 
3.1 percent 
Lymphocytes in 23 cases 
21.9 percent 
Sedimentation rate in 26 cases, elevated 
53.8 percent 
In Table VIII is shown the distribution of 
the more significant diagnostic physical find- 
ings in this series of cases. 
Distribution of the More Significant Positive 
Diagnostic Physical Findings in 95 
Cases of Disseminated 
Coccidioidomycosis 
Table VIII 
Total 
Number 
Skin and subcutaneous tissues 
64 
Lung 
5i 
Lymph nodes; 
39 
Cervical 
23 
Bronchial 
29 
Axillary 
4 
Inguinal 
4 
Retroperitoneal 
3 
Epitrochlear 
3 
Bone 
19 
Meninges 
10 
Joints 
8 
Peritoneum 
3 
Liver 
1 
paratively few cases in this series. In only 6 
were the studies complete, and although there 
are too few upon which to base any conclu- 
sions the results are of some interest. A strict 
parallelism was found between the results of 
the coccidioidin, precipitin, and complement- 
fixation tests in 4 of the cases. Parallelism be- 
tween the precipitin and complement-fixation 
reactions also existed in 4. Parallelism be- 
tween the results of the coccidioidin test and 
complement-fixation reaction was present in 
5 cases. In 5 cases the coccidioidin and pre- 
cipitin reaction ran parallel. From the data it 
appears that there is no relation between the 
complement-fixation by coccidioidin and by 
the syphilitic antigen. The data from the 6 
completely studied cases are presented in 
Table XL 
5. Blood findings: A profound anemia has 
been found characteristic of this group of 
cases. It is usually associated with a moderate 
leukocytosis with a slight increase in the 
neutrophils and a marked eosinophilia. A com- 
posite picture of the blood findings is repre- 
sented in Table IX. In the cases specified in 
Table IX, which were selected to suggest the 
range of the blood counts, the highest eosino- 
phil count was found to be 34 per cent; the 
lowest red cell count, 2.2 million; the highest 
white cell count, 23,700; the highest lympho- 
cyte count, 45 per cent; the lowest hemo- 
globin estimation, 50 per cent. In 53.8 per 
cent of the 26 cases tested the sedimentation 
rate was elevated. 
Tuberculin reactions were tested in only 5 
Table X 
Serologic and Skin Test Findings in Fatal and 
Surviving Cases of Disseminated 
Coccidioidomycosis 
Posi- 
tive 
Nega- 
tive 
In 28 cases, coccidioidin 
16 
12 
In 22 cases, complement-fixation 
20 
2 
In 13 cases, precipitin reaction 
In 22 cases, Wassermann and/or Kahn 
6 
7 
reaction 
4 
18 
The serologic findings in this group of 
cases are summarized in Table X. 
Serologic studies were carried out in com- 678 
The Military Surgeon—November, 1946 
of the 95 cases in this series, and 4 proved to 
be negative. These limited data do not sup- 
port the view held by some of the earlier 
workers that a close correlation exists between 
the coccidioidin and tuberculin reactions in 
cases of coccidioidomycosis. 
6. Sfutum findings: The quantity of 
sputum produced by patients with disseminated 
Organisms were seen in the exudate of 14 of 
17 lesions examined. They were cultured from 
the exudate of 11 of 16 ulcerative lesions. 
Organisms were both seen and cultured in 5 
cases; seen, but not recovered in culture, in 
1 case; cultured, but not seen in the exudate, 
in 1 case. From these data it is obvious that 
the microscopic examination of exudates for 
Serologic and Skin Test Findings in Six Cases of 
Disseminated Coccidioidomycosis 
Table XI 
Coccidioidin 
Precipitin 
Complement- 
Fixation 
Wassermann- 
Kahn 
Case No. 4, AIP Ac. 95409 
Negative 
Negative 
Positive 
0 
Case No. 8, AIP Ac. 108221 
Positive 
0 
Positive 
— 
Case No. 37, AIP Ac. 117401 
Positive 
Positive 
Positive 
Positive 
Case No. 45, AIP Ac. 108114 
Positive 
Positive 
Positive 
— 
Case No. 53, AIP Ac. 117769 
Negative 
Negative 
Negative 
— 
Case No. 79, AIP Ac. 130093 
Positive 
Positive 
Positive 
0 
coccidioidomycosis, of which extensive pul- 
monary involvement is usually a characteristic 
feature, is surprisingly not large. Blood often 
appears in the sputum either as simple blood 
streaking or in considerable quantity. In a few 
instances actual hemorrhage from the lung 
occurred. Bloody sputum was observed in the 
absence of cavities in the lung. In only 3 of 
the 95 cases was the sputum purulent. 
the organisms is a useful diagnostic procedure, 
though not to be relied upon exclusively. In 
some of our cases animal inoculations have 
been made, the animals being guinea pigs, 
white rats, and white mice, all of which are 
highly susceptible to the infection. 
7. Sfinal fluid findings: In this series of 
95 cases there were 8 cases of coccidioidomy- 
cotic meningitis. Examination of the spinal 
fluid from this group of cases revealed the 
findings listed in Table XII. 
Study of the sputum for the presence of 
C. immitis was a common clinical procedure, 
but the results are not reliable because of 
difficulty in recognizing the organisms, 
especially the smaller forms. Sputum examina- 
tion by microscopic preparations was made 
only in the 50 fatal cases. Eleven of these 
were found positive, representing an effective- 
ness of this method of diagnosis of 22 per cent. 
Obviously, this justifies the use of the pro- 
cedure routinely. Sputum cultures revealed 
the organisms in 6 cases, and in 3 of these 
the organisms were seen in the sputum and 
subsequently cultured and identified. 
Organisms were seen in microscopic 
Table XII 
Spinal Fluid Findings in Cases of 
COCCIDIOIDOMYCOTIC MENINGITIS 
Average 
Total cell count, 14 cases 
114 
Lymphocyte count, 11 cases 
59 percent 
PMN cell count, 11 cases 
41 percent 
Globulin, 11 cases 
Elevated 
Pressure in 8 cases 
Elevated (max, 
mm. water) 
600 
Sugar in 6 cases 
Elevated (max. 
mg. %) 
52 
Color of spinal fluid 
Usually cloudy 
rusty 
or 
These data on the finding of organisms in 
the sputum of fatal cases may be compared 
with data relating to the demonstration of 
organisms in exudate from ulcerative lesions. Figure 6. Endogenous reinfection pulmonary coccidioidomycosis, AIP Acc. 133706: Kodachrome 
photograph of the lung to show widespread granulomatous bronchitis and bronchopneumonia, the 
latter often confluent. Roentgenologic appearance like that in Fig. 4, except for greater density of 
the consolidation. See Fig. 19 for microscopic appearance of the lesion. Note the relative absence of 
involvement of the bronchial nodes. Death on 365th day with rapid pulmonary bronchial spread 
and widespread extrapulmonary dissemination. Figure 18. Primary pulmonary coccidioidomycosis, AIP Acc. 96443: Pure granulomatous lobular 
pneumonia. Note that the alveoli are filled with solid granulomatous tissue, while the alveolar walls 
remain intact. In the inflammatory tissue are seen two organisms, one in spherule form, the other 
maturing toward endosporulation. Death in 123 days. Coccidioidomycosis 
679 
preparations of spinal fluid in two cases, cul- 
tured from the spinal fluid in one case, and 
both seen and cultured in one case. 
a negative to a positive coccidioidin skin test. 
However, according to Lee and Nixon, a 
considerable proportion of persons infected 
with C. immitis for the first time do exhibit 
clinical signs and symptoms, and these are 
almost regularly referable to the respiratory 
system. Although definite anatomic alterations 
of the lung occur in this form of the disease, 
Lee and Nixon include this in their “primary” 
stage, rather than in their “intermediate or 
pulmonary” stage, even though they realize 
that in many, if not all of these cases progres- 
sive changes take place in the lungs which 
finally produce their “intermediate or pul- 
monary” stage of the disease. Indeed, the con- 
tinuous progress of some cases results in the 
widely “disseminated” stage of coccidioidomy- 
cosis, usually referred to in the older literature 
as coccidioidal granuloma. (Fig. 6) 
V 
Pathologic Anatomy of 
Coccidioidomycosis 
The distribution of the significant diagnostic 
physical findings in the 95 cases (Table IV 
and VIII) indicates clearly the importance of 
certain organs and anatomic regions in the 
pathogenesis of coccidioidomycosis. Of the 
internal structures, the lungs, the lymph 
nodes, the bones, and the meninges are com- 
monly involved; however, in clinical studies 
of coccidioidomycosis the skin and subcutaneous 
tissues have been found to be most frequently 
the site of granulomatous lesions. Post-mortem 
studies reveal a somewhat different order of 
frequency of involvement of the various tis- 
sues, as will be seen in Table XIII. 
The non-symptomatic phase of the “pri- 
mary stage” of coccidioidomycosis has never 
been studied at autopsy; therefore we have 
no reliable information on the anatomic state 
of the body related to this phase of C. immitis 
infection. Something is known of the anatomic 
state of the body in the symptomatic phase of 
the “primary stage” of coccidioidomycosis 
characterized by chest pain, cough, naso- 
pharyngitis, and general malaise. This infor- 
mation has been gained very largely, though 
not exclusively, from roentgenologic studies, 
especially those of Carter82 and Colburn.220 
The contribution of Colburn is based upon a 
careful study with follow-ups of an epidemic 
of 75 cases; a detailed clinical report of this 
epidemic will be found in the paper by Gold- 
stein and Louie.213 (See also the report of an 
epidemic of coccidioidomycosis occurring in 
14 students by Davis, Smith, and Smith,193 
1942). From these and other clinical studies 
of “primary” coccidioidomycosis of the sympto- 
matic type, it is clear that this phase of the 
infection is accompanied by one or more of the 
following anatomic changes within the lung 
and adjacent structures: 
According to the observations recorded in 
this table, the order of frequency of involve- 
ment in disseminated coccidioidomycosis is as 
follows: lungs, lymph nodes, spleen, skin and 
subcutaneous tissues, liver, kidney, bones, 
meninges, adrenal, myocardium, brain, peri- 
cardium, and pancreas. The remaining tissues 
are involved only incidentally. In the following 
section will be discussed the lesions as found in 
the individual organs, particularly in the lungs, 
since the pathogenesis of this disease is con- 
cerned primarily with the development of the 
pulmonary lesion. 
It has been the custom among those most 
familiar with clinical coccidioidomycosis to 
recognize three fairly distinct stages of de- 
velopment. Lee and Nixon223 list these as 1) 
the primary, 2) the intermediate or pul- 
monary, and 3) the progressive or dissemi- 
nated. According to most workers, what Lee 
and Nixon refer to as the “primary stage” 
of coccidioidomycosis is not characterized by 
any pathologic-anatomic changes in the lung 
or elsewhere; it is, indeed, actually asympto- 
matic. The fact that the patient has the dis- 
ease or has come in contact with the etiologic 
agent is indicated simply by a changeover from 
a. Hilar and mediastinal lymph node en- 
largement 
b. Bronchial inflammatory changes The Military Surgeon—November, 1946 
Table XIII 
Distribution of Lesions in 50 Autopsies, Based on Gross and 
Microscopic Study of the Tissues 
Gross 
Microscopic 
Final 
Lungs 
43 
43 
43 
Lymph nodes: 
Bronchial and mediastinal 
3*1 
Abdominal 
16 
Retroperitoneal 
12 
’36 
30 
36 
Cervical 
7 
Spleen 
29 
35 
35 
Skin and subcutaneous tissue 
3* 
9 
3* 
Kidney 
30 
3° 
3° 
Liver 
i? 
30 
3° 
Pleura 
*5 
5 
*5 
Bones: 
Rib 
lr 
Vertebra 
14 
Skull 
10 
Sternum 
7 
Clavicle 
6 
Ilium 
6 
24 
20 
*4 
Hand 
4 
Tibia 
4 
Scapula 
3 
Fibula 
1 
Meninges (lepto) 
18 
18 
18 
Adrenal 
13 
16 
16 
Myocardium 
5 
14 
14 
Brain 
7 
12 
12 
Bronchi 
3 
10 
IO 
Dura 
8 
7 
8 
Pancreas 
1 
8 
8 
Pericardium 
7 
7 
7 
Retroperitoneal tissues 
5 
— 
5 
Voluntary muscles 
5 
— 
5 
Peritoneum 
4 
3 
4 
Thyroid 
— 
4 
4 
Esophagus 
3 
3 
3 
Joints (sterno-clavicular 2, hand i) 
3 
1 
3 
Prostate 
— 
3 
3 
Stomach 
1 
3 
3 
Bone marrow 
— 
2 
2 
Optic nerve 
2 
2 
2 
Pituitary 
— 
2 
2 
Trachea and larynx 
1 
1 
1 
Nasopharynx 
1 
— 
1 
Diaphragm 
— 
1 
1 
Spinal cord 
— 
1 
1 
c. Focal or lobular consolidation of the 
pulmonary parenchyma, sometimes of a 
distinctly miliary nodular type 
d. Homogeneous, diffuse lobar consolida- 
tion 
e. Solitary areas of consolidation sharply 
defined, some with central cavity forma- 
tion 
f. Pleural effusion, usually small but some- 
times massive Coccidioidomycosis 
681 
g. Multiple, more or less isolated areas of 
consolidation with intervening dissemi- 
nated focal miliary lesions 
The pulmonary changes found have been 
observed to run their course and to clear up 
completely within a period of 3 or 4 months. 
Indeed, in the epidemic of 75 cases mentioned 
above, all of the patients were eventually re- 
2) the generalized or disseminated form. The 
initial pulmonary disease may occur in sympto- 
matic and asymptomatic stages. Indeed, there 
is much to be gained by adopting a classifica- 
tion of the pulmonary form of coccidioidomy- 
cosis analogous to that in tuberculosis, in which 
we recognize simply a primary complex 
(symptomatic and asymptomatic) and a re- 
Fig. 7. Primary pulmonary coccidioidomycosis, AIP Ac. 108095: Duration of disease 120 days. Death 
after widespread dissemination of lesions. The lung lesions are: a) focal granulomatous bronchopneumonia, 
b) necrotizing bronchitis and confluent suppurative and granulomatous pneumonia with early cavity forma- 
tion. See details in Fig. 8. Note the minimal involvement of the bronchial nodes. 
turned to duty, although dissemination to 
distal portions of the body did occur in one 
case. 
infection. Adoption of this view appears to fit 
well the pathologic-anatomic observations made 
in the 50 autopsies under consideration. Our 
anatomic observations accordingly will be pre- 
sented under the headings: A. Pulmonary 
coccidioidomycosis—1. The primary complex, 
and 2. Reinfection pulmonary coccidioidomy- 
cosis; and B. Disseminated coccidioidomycosis. 
The asymptomatic form of “primary” coc- 
cidioidomycosis is recognized solely by skin test 
with coccidioidin, and it is impossible to dif- 
ferentiate clinically between the symptomatic 
phase of “primary” coccidioidomycosis and 
the “intermediate or pulmonary” stage of the 
disease. Thus, it would seem more logical to 
think of coccidioidomycosis as occurring sim- 
ply in two forms, 1) the pulmonary form, and 
A. Pulmonary Coccidioidomycosis 
I. Primary Pulmonary Coccidioidomycosis 
—The Primary Comflex. For reasons given 682 
The Military Surgeon—November, 1946 
above, only the end stages and the healing 
phases of the primary pulmonary complex 
have been available for anatomic and histologic 
study. Despite this handicap it has been possible 
to construct in a fairly satisfactory manner the 
pathologic-anatomic picture that develops in 
the course of the primary pulmonary infection. 
This has been accomplished by selecting fatal 
cases of less than 100 days duration, and sup- 
plementing data from these by the study of 
total duration of the disease was 74 days 
(Figs. 12, 13, and 14). This lesion corre- 
sponds to the focal, fuzzy consolidation of the 
lung so commonly seen in roentgenograms 
during the early stage of the disease. An 
accentuation of the bronchial markings of the 
lung is also apparent, as well as considerable 
enlargement of the bronchial and mediastinal 
lymph nodes. The histologic picture of this 
pneumonia is not at all remarkable except 
Fig. 8. Primary pulmonary coccidioidomycosis, AIP Ac. 108095: Necrotizing- confluent granulomatous 
pneumonia, with cavity formation. Note the multilocular cavity at the center of the pneumonic area and the 
adjacent cavities from which the necrotic sequestrum has not yet been removed. The bronchial lymph node 
at the upper right corner shows minimal involvement, and the adjacent secondary bronchi are normal. Same 
lung as shown in Fig. 7, but different section. 
primary pulmonary lesions excised surgically 
and roentgenologic material from clinical 
cases. 
for the organisms within the pulmonary exu- 
date, which consists largely of polymorphonu- 
clear leukocytes with a scattering of large 
mononuclear wandering cells, and a consider- 
able amount of fluid without much fibrin. The 
lesions relate particularly to the terminal por- 
tions of the bronchial system and the im- 
mediately adjacent atrial and alveolar spaces. 
Other lesions in the same case exhibit begin- 
ning granulomatous transformation. Thus, 
The primary lesions, some of which progress 
and some of which recede to disappear eventu- 
ally, fall within the following categories: 
a. Lobular focal pulmonary consolidation 
strongly resembling the ordinary type of bac- 
terial pneumonia. This we have seen in our 
Case No. 7 (AIP Ac. 74661), in which the Coccidioidomycosis 
683 
different stages of development of the lesion 
may be found in one lung. Almost every- 
where the changes are intra-alveolar, and the 
character of the exudate is such as to make 
possible complete resolution without residual 
structural changes. Quite naturally, some le- 
sions progress too far for complete resolution; 
these heal by the formation of microscopic 
scars, leaving the lung otherwise unaffected. 
tion of cavities like those occasionally seen in 
roentgenologic examination of the living pa- 
tient early in the course of the disease. These 
lesions may persist as chronic abscess-like cavi- 
ties in the lung long after the primary pul- 
monary infection has run its course and the 
patient has recovered from the acute illness. 
Histologically the ulcerative bronchitis has a 
distinctly granulomatous character, but the 
reaction is noteworthy for the large number 
of organisms and the continued presence of 
masses of polymorphonuclear leukocytes. The 
bronchi are usually filled with exudate con- 
sisting largely of leukocytes, fibrin, and 
mononuclear wandering cells (Figs. 16 and 
I?)- 
b. Extensive gelatinous focal or confluent 
lobular consolidation. We have observed this 
lesion in our Case No. 79 (AIP Ac. 130093), 
with a total duration of 87 days (Figs. 7 and 
8). Its striking feature proved to be its wetness 
and in some instances the hemorrhagic char- 
acter of the exudate. Histologically the lesion 
is found to be of an almost suppurative char- 
acter, the exudate consisting largely of poly- 
morphonuclear leukocytes, but with a con- 
siderably greater number of mononuclear 
wandering cells than found in the ordinary 
pneumonic type of consolidation. The lesion 
is patchy in distribution. Between the more 
solid areas of pneumonia there is a profuse 
exudation of fluid without much fibrin, in 
which large mononuclear cells, polymorpho- 
nuclear leukocytes, and a few lymphocytes are 
scattered. The outpouring of fluid and cellular 
exudate is almost exclusively intra-alveolar 
and is associated with the presence of the 
specific organisms in great number and un- 
usual forms. The most impressive feature of 
this form of the reaction is the extensive 
necrosis. It involves both the exudate and the 
adjacent pulmonary structures and is of such 
a character that complete restitution of the 
normal lung architecture would be impossible 
upon healing of the lesion. In both its gross 
and microscopic features this lesion bears a 
striking resemblance to the gelatinous pneu- 
monia of pulmonary tuberculosis with early 
caseation. 
d. Focal lobular, or massive lobar indurated, 
gray, moist but not wet, granulomatous con- 
solidation. This lesion in brilliant form was 
observed in an individual who died after 75 
days illness, our Case No. 27 (AIP Ac. 
85443) (Fig. 9). The histologic appearance 
of this lesion stands in sharp contrast to that 
of the other three types just noted. The reac- 
tion is essentially granulomatous, whether the 
lesion involves only foci or large areas of the 
lung, even to the extent of lobar consolidation. 
The exudate is composed almost exclusively 
of mononuclear cells, many of which have 
been transformed into epithelioid cells or large 
multinucleated giant cells. Masses of exudate 
of this type fill the alveolar spaces but, strangely 
enough, do not tend to cross the alveolar 
boundaries, leaving these structures well pre- 
served. There is much fibrin with this granu- 
lomatous exudate and some, although not con- 
spicuous, fluid. Here and there are microscopic 
areas of suppuration in which the exudate con- 
sists exclusively of polymorphonuclear leuko- 
cytes. These little abscess-like lesions are some- 
times hardly more than the size of an 
alveolus, but they may occupy a whole lobule. 
The distinguishing feature of this lesion is its 
solid character, its absence of necrosis, and the 
seemingly organized arrangement of the cells 
comprising the inflammatory exudate. In 
lesions of this type organisms are always con- 
spicuous except in the areas of suppuration and 
are usually in the intermediate stages of ma- 
c. Necrotizing ulcerative bronchitis and 
bronchiolitis with bronchiectasis and bronchiec- 
tatic cavity formation. This lesion is illustrated 
in our Case No. 24 (AIP Ac. 107640), the 
total duration of which was only 31 days 
(Fig. 10). The infectious process in this case 
was highly destructive, resulting in the forma- 684 
The Military Surgeon—November, 1946 
Fig. 9. Primary pulmonary coccidioidomycosis, AIP. Ac. 85443; Confluent lobular granulomatous pneu- 
monia without conspicuous necrosis and little softening of the tissue. Caked consistency, firm and rubber-like, 
without fluid exudate; great enlargement and granulomatous alteration of the bronchial nodes with coagu- 
lative necrosis. Death on 75 th day with extensive dissemination including meningitis. C occidioidomy costs 
685 
Fig. io. Primary pulmonary coccidioidomycosis, AIP Ac. 107640: Lung from a case with rapid course, 
death on 31st day of disease. Multiple focal suppurative lesions often in clusters where they become con- 
fluent and produce soft suppurating nodules. The lesions grossly and microscopically are distinctly suppura- 
tive and necrotizing like those shown in Figs. x6 and 17. Widely distributed extrapulmonary lesions. 686 
The Military Surgeon—November, 1946 
Fig. ii. Chronic primary pulmonary coccidioidomycosis, AIP Ac. 156611: Lung with huge apical cavity 
with fibrous walls lined by necrotizing granulation tissue, removed by lobectomy. C. immitis cultured from 
pus in cavity. Unidentifiable hyphae and spores demonstrated in cavity exudate in sections. No spherules 
found in the tissues or in the exudate. Duration of pulmonary disease 9 years. This is the only case on record 
in which the mycelial form of the organisms have been found in tissues. Coccidioidomycosis 
687 
turation. They appear both intra- and extra- 
cellularly, being most conspicuous in the large 
multinucleated giant cells. The ultimate fate 
of this lesion, should the patient survive, can 
be only the transformation of the inflamma- 
tory lesion into a mass of hyaline scar tissue 
or eventual necrosis with partial fibrosis and 
encapsulation. Complete resolution of the 
lesion with restoration of the lung tissue to 
its normal state cannot occur. A massive le- 
sion of the type just described is illustrated in 
hyaline scars hardly larger than an alveolus. 
These scars are at times ossified. Completely 
hyaline or hyalinizing tubercles, solitary hard 
hyaline nodules as large as 2 or 3 cm. in 
diameter, usually with necrotic but firm cen- 
ters (Fig. 24), and solitary or multiple cavi- 
ties lined by chronic granulomatous inflam- 
matory tissue (Figs. 25 and 26) are other 
forms observed in our material, chiefly in 
biopsy specimens and usually from patients 
who appear to be almost well. These chronic 
Fig. 12. Primary pulmonary coccidioidomycosis, AIP Ac. 74661: Microscopic appearance of earliest 
lesion encountered. Suppurative bronchitis with adjacent focal lobular pneumonia. Note bronchus filled with 
fibrinopurulent exudate; bronchial epithelium preserved in some small areas; adjacent alveoli filled with 
same exudate. Fibrin is not an outstanding feature. Death in 74 days. (X 60) 
Fig. 9. This form of pulmonary infection, 
like the others of the primary disease, is asso- 
ciated with hilar and mediastinal lymph node 
enlargement. The characteristic lesions in 
these lymph nodes resemble those in the lung. 
or healed lesions have been found in the cases 
of longest duration, ranging from 467 to 
1,175 days (Fig. 11), and represent the heal- 
ing of the primary foci. They are not thought 
to be related to the disseminated form of the 
disease. 
e. Healed or progressively healing lesions. 
These lesions take various forms, although 
they are all essentially alike. Some of these 
consist of solitary or multiple minute focal 
So long as any one of these types of primary 
pulmonary complex exists, the danger of dis- 
semination of the organisms into distal por- 688 
The Military Surgeon—November, 1946 
tions of the body is always imminent. Active 
lesions of any one of the types mentioned may 
be progressive or regressive, which accounts 
for the variability of the clinical and 
roentgenologic picture in any given case. In 
some of the fatal cases of the disseminated 
form of coccidioidomycosis of relatively long 
duration a lack of consistent progressive clini- 
cosis. The recurrence of extensive pulmonary 
disease after the original infection in the lung 
has been observed to recede or even to dis- 
appear constitutes what we have designated as 
reinfection pulmonary coccidioidomycosis. In 
case of complete resolution of the primary com- 
plex the source of the reinfection is from with- 
out the lung and adjacent tissues. When the 
Fig. 13. Primary pulmonary coccidioidomycosis, AIP Ac. 74661: Under high power is seen a small section 
of lesion shown in Fig. 12, demonstrating the presence of organisms. Note the predominantly leukocytic 
character of exudate surrounding the organism. ( X 1000) 
cal development of the pulmonary lesion may 
be the only means of differentiation between 
a primary pulmonary infection and a terminal 
reinfection of the lung. There may be such 
close anatomic resemblance between a pro- 
gressing primary infection and a progressing 
reinfection of the lung acquired by late endog- 
enous pulmonary dissemination, that the 
latter can be recognized positively only when 
pulmonary disease is known to have developed 
after the lung has been observed to clear com- 
pletely from a preceding infection. 
primary infection has only receded, reinfection 
may occur from a destructive lesion in a 
bronchus, perhaps in association with adjacent 
necrotic infected lymph nodes. The result is 
then a fairly certain, wide, massive dissemina- 
tion of infectious material into the lung tissue 
by way of the bronchial system. This 
bronchial spread theoretically might occur also 
after the disseminated form of the disease has 
developed, or indeed after all lesions except 
those of the bronchial and mediastinal lymph 
nodes have subsided. A somewhat similar 
bronchial spread takes place during the course 
2. Reinjection Pulmonary Coccidioidomy- C occidioidomy costs 
689 
of the primary pulmonary infection merely 
through the aspiration of infected bronchial 
contents from one part of the lung to another. 
This is the most obvious explanation of the 
variable roentgenologic picture in the lung in 
the primary pulmonary disease. These bron- 
chial forms of reinfection of the lung stand in 
sharp contrast to the reinfection from the 
will be recognized that here the analogy to 
tuberculosis is close. A good example of re- 
infection of the lung from an extrapulmonary 
focus long after the original pulmonary lesion 
had healed is found in our Case No. 84 (AIP 
Ac. 133706), Figs. 6 and 19. 
It was not easy to determine from the ma- 
terial at our disposal either the portal of entry 
Fig. 14. Primary pulmonary coccidioidomycosis, AIP Ac. 74661: Higher magnification of bronchus in 
Fig. 12. The acute suppurative bronchitis accompanying the early stages of coccidioidomycosis. The exudate 
is made up of abundant polymorphonuclears, some fluid, fibrin, and cellular debris. The bronchial epithelium 
is preserved in a few small areas. ( X 100) 
extrapulmonary disseminated lesions, when the 
organisms are brought to the lung by way of 
the blood stream to produce a widespread 
miliary type of infection. In pulmonary re- 
infection of this kind the bronchial and 
mediastinal lymph nodes do not react as in the 
primary pulmonary infection. They are often 
found to be relatively small in comparison with 
the large, succulent and often necrotic lymph 
nodes of the primary pulmonary complex. It 
of the organisms into the blood stream or the 
foci of lodgement of the circulating organisms 
within the lung. The lesions in the lung have 
the appearance, both in the roentgenogram and 
in sections, of miliary nodules, like those in 
miliary tuberculosis; but close histologic study 
shows that the blood vessels are remarkably 
uninvolved by the infectious process. The le- 
sions are often discrete, but may coalesce, in 
the latter case presenting a picture like that of 690 
The 'Military Surgeon—November, 1946 
a spreading pneumonia with exudation, chiefly 
in the alveolar spaces. The lesion is indistin- 
guishable from the focal lesions which follow 
aspiration of infected material (Fig. 6). These 
lesions most frequently develop about the 
bronchi to produce a gross picture not unlike 
that characteristic of the nodular type of inter- 
stitial pneumonia produced by bacteria, particu- 
larly the influenza bacillus. 
monary terminal involvement produced by 
endogenous pulmonary spread and massive re- 
infection of the lung by way of the blood 
stream from extrapulmonary suppurative foci. 
(In two of our fatal cases there was no ques- 
tion of the origin of the massive pulmonary 
reinfection from a bronchial spread of the 
organism. Suppurative lymph nodes in the 
mediastinum had ulcerated through the wall 
Fig. 15. Primary pulmonary coccidioidomycosis, AIP Ac. 74661: An area of Fig. 12, under higher mag- 
nification, demonstrating the lymphatic involvement as part of the primary lesion. The lymph vessel is 
crowded with monocytes, polymorphonuclears and fluid. The walls are involved in the inflammatory reac- 
tion, making it a real lymphangitis, which will lead to the bronchial and mediastinal lymph node involve- 
ment. (X145) 
The great frequency of massive pulmonary 
disease in all fatal cases of coccidioidomycosis 
(in this series 43 of 50 cases), when viewed 
in the light of the known tendency of the 
primary pulmonary disease to heal, leaves little 
doubt of the development of an extensive 
hematogenous dissemination of the organisms 
to the lung in the terminal phases of the dis- 
ease, particularly in protracted fatal cases. In 
those of short duration it may prove extremely 
difficult to distinguish between massive pul- 
of the bronchus in one instance and the trachea 
in the other, and spilled their infected con- 
tents into the lumen of these structures. The 
material was then aspirated directly into the 
lung. These cases are No. 42, AIP Ac. 95411, 
and No. 73, AIP Ac. 114420.) 
In practically all cases of coccidioidomycosis 
where the lungs are involved there is con- 
current involvement of the pleura. This takes 
a variety of forms; the simplest is a fibrinous 
inflammatory reaction which gives rise to deli- C occidioidomy costs 
691 
cate adhesions between the pleural surfaces. 
When the pleura is heavily infected, fibrino- 
purulent exudate is produced; and eventually 
a fairly thick granulomatous reaction occurs in 
which there may be a considerable amount of 
necrosis. The gross appearance of the pleura 
is then somewhat reminiscent of fibrocaseous 
pleurisy of tuberculosis. Abscess-like pockets 
is impressively large. It was derived from an 
enumeration of all instances where pleural ad- 
hesions were found. It is common for the 
pleura to become secondarily involved when a 
disseminated lesion develops in the wall of the 
chest, particularly in the rib. Complicated 
sinuses may form communications between the 
pleura and the lesions in the chest wall. Active 
Fig. i 6. Primary pulmonary coccidioidomycosis, AIP Ac. 114420: Focal suppurative necrotizing lobular 
pneumonia. In this lesion are found an abundance of organisms. The small abscesses seen will become the 
cavities of early pulmonary coccidioidomycosis. The corresponding gross picture is represented by the necrotic 
white nodules in the lung of Fig. 10. Death occurred in 94 days. (X60) 
are sometimes observed in the earliest stages 
of infection of the lung. If these lesions reach 
the pleural surface, the infection gives rise to 
an immediate serofibrinous exudate, which 
may increase to the extent of 300 or 400 c.c. 
within a relatively short time. Ordinarily, how- 
ever, the amount of fluid within the pleural 
cavity remains small, just enough to obliterate 
the angles above the diaphragm. In Table 
XIII the figure representing the gross in- 
volvement of the pleura in our autopsy cases 
infectious processes were found in the pleura 
in only five of the series of 50 cases. 
B. Disseminated Coccidioidomycosis 
In the advanced and terminal stages of 
disseminated coccidioidomycosis specific lesions 
may be found in virtually any part of the body 
(Table XIII). The general characteristics of 
the lesions and the histologic reactions are 
quite the same regardless of the location; 
therefore only the more important secondary 692 
The Military Surgeon—Novembery 1946 
lesions will be discussed, that is, those which 
are of sufficient importance to be considered 
somewhat in the nature of clinicopathologic 
entities. 
granulomatous fungus infections, particularly 
blastomycosis. 
One of the reasons for the importance of 
involvement of the lymph nodes in this disease 
is that they often become necrotic, ulcerate, 
and discharge their contents into neighboring 
i. The Lymph Nodes. The lymph nodes 
most commonly affected are the bronchial and 
Fig. 17. Primary pulmonary coccidioidomycosis, AIP Ac. 74661: Early granulomatous lobular pneumonia. 
Besides the small abscesses there is already a focal granulomatous response in which some giant cells can be 
seen. Note the abundance of organisms in the abscesses. (X 60) 
mediastinal. These are always enlarged since 
they form a part of the primary infection com- 
plex in coccidioidomycosis. The lymph nodes 
become involved not simply by the passage of 
organisms from infected areas which they 
drain, but also through the lodgement of or- 
ganisms by way of the blood stream in the 
lymphoid tissue. This accounts for the involve- 
ment of lymph nodes throughout the body in 
the long-standing and fatal cases. The fre- 
quent extension of the infection to the nodes 
very early in the disease was emphasized by 
Hektoen,9 who contrasted this feature of 
coccidioidal granuloma with that of other 
tissues. The breakdown of lymph nodes with 
the formation of mediastinal abscesses, which 
in turn discharge their contents into the neigh- 
boring structures, for example, the esophagus 
and the trachea, occurred in two cases of this 
series. In one of these, extensive ramifying 
sinuses led from the mediastinal abscess up 
into the root of the neck, from which the 
accumulating pus was discharged through 
multiple sinuses in the skin. In another of our 
cases the epitrochlear lymph nodes became in- 
volved as a secondary effect in relation to a 
granulomatous suppurating lesion in the 
proximal phalanx of the little finger on the C occidioidomy costs 
693 
Fig. 19. Endogenous reinfection pulmonary coccidioidomycosis, AIP Ac. 133706: Microscopic appearance 
of lung in Fig. 6. Small abscesses are seen in some areas. In other areas is noted a typical granulomatous 
reaction filling the alveoli, whose alveolar walls are intact. Death in 365 days. (X 40) 694 
The Military Surgeon—Novembery 1946 
corresponding side. This case was our No. 81 
(AIP Ac. 131556). 
The enlargement of the lymph nodes is due 
to extensive granulomatous and suppurative 
necrotizing inflammatory lesions developing 
within them. These lesions tend to heal, just as 
ation and death of the parasites. It is important 
to note that lymph nodes remain infected for 
a very long time and undoubtedly form a nidus 
for the propagation of organisms, and thus 
serve as a source for dissemination to other 
parts of the body. 
Fig. 20. Healing pulmonary coccidioidomycosis, AIP Ac. 159164: Adjacent to a bronchus are typical 
tubercles. A spherule can be seen in the tubercle, which is already well walled off by a fibrous capsule. 
Higher magnification of this tubercle is seen in Fig. 23. Death in 467 days. (X65) 
do the lesions elsewhere in the body, and 
leave dense hyaline scars quite like those which 
follow the healing of tuberculous granu- 
lomatous lesions. In the freshly infected lymph 
nodes, where the organisms are particularly 
numerous, extensive necrosis occurs. Some of 
the nodes which we have examined in this 
series were hardly less than a massive culture 
of organisms embedded in highly necrotic 
tissue; others were extensively scarred with 
actual hyalinization of the scar tissue, in the 
presence of an extraordinary number of 
organisms. The organisms seem to persist and 
to proliferate. Some of the most indurated 
nodes, and some that appeared to be least active 
grossly, were found on microscopic examina- 
tion still to harbor great numbers of organisms, 
as well as many forms resulting from degener- 
2. The Spleen. Involvement of the spleen 
is exceeded in frequency only by that of the 
lymph nodes and is the result of both direct 
infection through the blood stream and the 
indirect effect of the infection, arising from 
general intoxication. Thus, there are specific 
coccidioidal granulomas, and, in addition, the 
acute splenic tumor which characterizes so 
many infectious diseases. As a result of this 
combination of lesions, the spleen is often con- 
siderably enlarged, its average weight in 27 
cases being well over 200 Gm., with the 
largest weighing 700 Gm., the smallest only 
78 Gm. This enlargement is chiefly the result 
of the acute splenic tumor. Although the 
specific lesions take a variety of forms, they 
ordinarily are not bulky (Fig. 29). 
The most common lesions are focal miliary Coccidioidomycosis 
695 
nodules produced by the granulomatous in- 
flammatory reaction. These are usually numer- 
ous, fairly soft, and suppurative changes occur 
within some of them. At times the larger 
necrotic nodules break down completely, form- 
ing chronic abscesses. Lesions of this sort may 
be solitary. A striking form of infection in the 
throughout the splenic pulp. In more heavily 
infected individuals, these organisms can be 
found in great abundance, many of them 
being within the splenic sinuses and, thus, well 
on their way toward hematogenous dissemina- 
tion. We have not seen any lesion comparable 
to the calcified nodules or hyaline foci so com- 
Fig. 21. Disseminated coccidioidomycosis, AIP Ac. 127046: This is a disseminated skin lesion under low 
power. Note the intact epithelium and subcutaneous necrosis. Death in 47 days. 
Fig. 22. Disseminated coccidioidomycosis, AIP Ac. 96943: This demonstrates the granulomatous epicarditis. 
Note adhesion. From the epicardium the disease spread to the pericardium. (X512) 
spleen consists of multiple minute abscesses. 
These little foci of suppuration appear by the 
thousands in the splenic substances and, in- 
terestingly enough, organisms are rarely visible 
within them. That organisms are present, how- 
ever, can be demonstrated by the proper stain- 
ing techniques, the ordinary Goodpasture- 
McCallum bacterial stain being satisfactory. 
Another highly interesting histologic detail 
is the widespread dissemination of organisms 
mon in the spleen in disseminated tuberculosis, 
a disease to which disseminated coccidioidomy- 
cosis has so great but so superficial a re- 
semblance. 
3. The Skin and Subcutaneous Tissues. In 
this group of cases lesions in the skin and 
subcutaneous tissue were numerous (Table 
XIII). The cutaneous lesions are of two 
varieties: those related specifically to the in- 
fection with C. immitis} and those which relate FlG. 23. Healing- pulmonary coccidioidomycosis, AIP Ac. 159164: Higher magnification of the tubercle, 
which is well encapsulated, seen in Fig. 20. ( X 60) 
Fig. 24. Healing pulmonary coccidioidomycosis, AIP Ac. 159164: This tubercle has progressed into the 
final stage of necrosis and is surrounded by a fibrous capsule. Eventually the whole lesion will be transformed 
into a dense scar. ( X 60) Coccidioidomycosis 
697 
to the degree of sensitization to products of 
these organisms. The most common lesion of 
the second type is erythema nodosum, and 
occasionally erythema multiforme. This we 
have seen in only one instance, and that oc- 
curred well in advance of the death of the 
patient, that is, in the earlier phases of the 
development of the pulmonary manifestations. 
related directly to infection with C. immitis 
take a variety of forms, as listed on page 671. 
Histologically the lesions in the skin are 
quite like those elsewhere, the reactions be- 
ing typically granulomatous with foci of 
suppuration (Fig. 21). They may occur any- 
where over the body, but in our cases have 
been most frequent on the chest wall, the face, 
Fig. 25. Healing- pulmonary coccidioidomycosis, AIP Ac. 149019: This necrotic lesion shows a central 
cavity in which numerous organisms are still present. The surrounding lung tissue is collapsed. Note the 
patent ulcerated bronchus in the left lower corner. (X 5/^a) 
Few cases of primary pulmonary infection in 
which erythema nodosum occurs progress to 
the disseminated form. The absence, or virtual 
absence, of erythema nodosum and other mani- 
festations of sensitization would seem to indi- 
cate that all of our cases have advanced well 
beyond the stage of primary pulmonary in- 
fection. This, of course, is in harmony with the 
actual findings. In this series we have had no 
opportunity to study the histologic reaction 
of the tissues showing the lesion in the skin. 
and the scalp. The frequency with which the 
skin overlying the sterno-clavicular junction 
has been involved is striking. 
4. The Liver and the Kidneys. Within the 
abdominal cavity the lesions of disseminated 
coccidioidomycosis are more often found in the 
liver and the kidneys than in any other organ 
except the spleen. The changes in the liver are 
similar to those in the spleen, with enlargement 
resulting from the general intoxication in 
addition to the specific lesions of C. immitis. 
The specific lesions in the kidney are like those 
Lesions in the skin and subcutaneous tissues 698 
The Military Surgeon—November, 1946 
occurring in the liver and consist of isolated 
granulomas in various stages of development. 
The lesions are usually quite small, of the 
miliary type, and are scattered through the 
parenchyma without localization in specific 
portions of the lobule. In both of these organs 
it is not unusual to find, along with the typical 
miliary granulomas, abscess-like suppurative 
lesions of about the same size. The lesion 
all. Although the infection may occur in any 
portion of the bone and periosteum, it is per- 
haps most commonly seen at the ends of the 
bones, particularly in relation to the bony 
prominences, such as the tibial tubercle, the 
ankle, the acromial process of the scapula, and 
the medial extremity of the clavicle. When the 
infection develops in the cancellous bone of the 
epiphyses, it may destroy the overlying articular 
Fig. 26. Healing pulmonary coccidioidomycosis, AIP Ac. 149019: A higher magnification of the bronchus 
surrounded by the necrotic infectious material, seen in Fig. 25. From such relatively old lesions organisms 
may disseminate, resulting in pictures as in Figs. 6 and 19. ( X 25) 
usually seen in the liver and the kidney is 
illustrated in color photograph (Fig. 28). In 
one of our cases a portal phlebitis of granu- 
lomatous type developed. In no instance have 
we seen infection of the pelvis of the kidney or 
of the ureter. 
cartilage, and thus involve the joint. Infection 
of the skull is not only one of the most common 
bone lesions, but one of the most important 
because of the extension of the process both 
inward and outward. The former gives rise 
to coccidioidal infection of the pachymeninges, 
which may eventually extend to the subarach- 
noid space, resulting in a usually fatal menin- 
gitis. Outward extension from the skull, as 
from other bones, results in the formation of 
large subcutaneous abscess-like lesions which 
5. The Bones. Coccidioidal granulomatous 
infection of the bones constitutes one of the 
most characteristic features of this disease. 
Usually a number of bones are affected, but 
the type of reaction is essentially the same in Coccidioidomycosis 
699 
drain persistently, when once they rupture 
through the overlying skin. 
superficial portion of the muscles, involving the 
epicardium (Fig. 22). Pericarditis due to C. 
hnmitis infection in 7 of the 50 autopsied cases 
appeared to have resulted from the eruption of 
a superficially situated granulomatous lesion 
located in and just beneath the epicardium. 
This mechanism of infection of the serous 
membranes was observed elsewhere in the 
body, for example, in the peritoneal cavity and 
in the subarachnoid space. In none of the 14 
cases in which the myocardium was involved 
was an endocardial lesion found. Focal 
granulomas sometimes were present just be- 
neath the endocardium of the ventricles, but 
never in any of the valves of the heart. It ap- 
pears that endocarditis of the usual bacterial 
variety does not occur in coccidioidal infection 
(for a possible exception to this, see Ep- 
stein149), an observation which has also been 
made in tuberculosis and in other fungus in- 
fections, such as blastomycosis. Serofibrinous 
pericardial effusion resulting from infection of 
the pericardial surfaces and fibrous adhesive 
pericarditis with foci of necrosis simulating 
caseation were observed in the series. 
One of the most interesting bone lesions 
occurs in the bodies of the vertebrae. Here the 
infection is in the cancellous portion originally, 
but it extends to become subperiosteal. If this 
occurs in the vertebrae of the lumbar region, 
the infection may extend through the attach- 
ments of the muscles, particularly through 
those of the psoas muscles, in which event it 
forms a typical psoas abscess resembling that 
which occurs in tuberculosis of the vertebral 
bone. In a relatively large proportion of de- 
layed disseminations the first manifestation of 
the dissemination was found in a swelling of 
one of the bones, often of the small bones of the 
hand. Involvement of the ribs, like that of 
the vertebrae, is especially important because 
of the relation to adjacent structures, in this 
case the pleura and the overlying skin. Large 
abscess-like lesions within the thoracic wall 
were common in the cases we have studied. 
They usually erupt on the surface, but they 
may produce widespread dissection of the chest 
wall before discharging through the skin. The 
sternal lesions are prone to develop at the 
sterno-clavicular junction, but the reason for 
this is not apparent. Specimens of the gross 
lesions of the bone, except of the ribs, verte- 
brae, sternum, and skull, were not included in 
our material; but our roentgenograms have 
provided a clear picture of the lesions in these 
and the other bones when they were involved. 
8. The Brain and the Meninges. Dissemina- 
tion of C. immitis to the brain results in some- 
times solitary but usually multiple granu- 
lomatous lesions. These may take the form of 
miliary nodules which may be firm, soft, and 
necrotic, or definitely hemorrhagic. Solitary 
granulomas of considerable size (2 to 3 cm.) 
have been observed and are like those that de- 
velop in other tissues. It is important to note, 
however, that these lesions develop both super- 
ficially and deeply in the brain. Our observa- 
tions indicate that infection of the brain sub- 
stance is not nearly so common as in 
tuberculosis; therefore infection of the 
meninges could not often result from the proc- 
ess of extension. Since the organisms are of 
sufficient size to lodge readily within the 
vessels of the subarachnoid space, it is possible 
that the infection of the meninges may be 
brought about through this mechanical factor. 
6. The Adrenals. C. immiiis rather fre- 
quently localizes in the adrenal, to judge from 
our series, in which the gland was involved in 
16 of the 50 cases studied at autopsy. In no 
instance was the adrenal tissue completely 
destroyed, but in a number both adrenals were 
infected. The lesion usually consists of one or 
more granulomatous nodules. No indication 
of adrenal insufficiency has appeared as a result 
of the granulomatous changes in this organ. 
7. The Heart and its Coverings. In 14 
instances the myocardium was found to be the 
seat of disseminated granulomatous lesions, 
most of them miliary, some microscopic, and 
not particularly numerous in either case. In 
some instances the lesions were situated in the 
The meningitis that we have seen in dis- 
seminated coccidioidomycosis is of two essential 
varieties: 1) a firm, plastic type of meningeal 
inflammation, which results in the encasement 700 
The Military Surgeon—Novembery 1946 
Fig. 27. Disseminated coccidioidomycosis, AIP Ac. 117769: The gross appearance of the plastic type of 
basal meningitis. Note the dense exudate covering the brain stem, leading to obstruction of the circulation 
of spinal fluid and internal hydrocephalus. Death after 111 days. 
of the brain substance, particularly the brain 
stem, in a rigid, highly organized, granu- 
lomatous mass of tissue (Fig. 27), and 2) a 
pronounced exudation of fluid with wandering 
cells and polymorphonuclear leukocytes, which 
has undergone only a minimal granulomatous 
transformation. The more typical reaction to 
C. immitis is the former. A characteristic 
feature of the granulomatous reaction is the 
typical tubercle with giant cells which almost C occidioidomy costs 
701 
regularly contain the organism; at the same 
time, many organisms appear free in the 
exudate. Even in the most highly developed 
granulomatous reaction one may see, as in 
other granulomatous lesions, focal areas of 
suppuration in which there are few organisms 
and many leukocytes. The thick, plastic, 
opaque type of alteration of the meninges 
reminds one of tuberculous meningitis; but 
the resemblance is only superficial, since the 
of our cases the development of a local 
meningitis about the cord produced all the 
signs and symptoms of a cord tumor, making 
necessary an exploratory operation for relief 
of spinal cord pressure. Such localized spinal 
leptomeningitis is usually associated with in- 
volvement of the dura, which in turn is the 
result of localization of the infecting agents in 
the bones surrounding the spinal canal from 
the blood stream. 
Fig. 29. Disseminated coccidioidomycosis, AIP Ac. 96943: Gross appearance of the spleen. Great numbers 
of small solitary tubercle-like lesions and great masses of granulomatous tissue. Death after 123 days. This 
is the same case as seen in Figs. 18 and 22. 
degree of thickening and induration of the 
inflammatory exudate is never so prominent in 
tuberculous meningitis as in C. immitis infec- 
tion. Due to the reaction in the meninges about 
the brain stem, the movement of spinal fluid 
from the brain cavities into the subarachnoid 
space is obstructed, resulting in extreme grades 
of hydrocephalus. 
Thirteen of the 18 cases of coccidioidal 
meningitis recorded in this report were previ- 
ously reported by Schlumberger.236 
It is important to note that coccidioidal 
meningitis is often the only evidence of dis- 
semination. In a few cases studied at autopsy 
it has not been possible to locate the primary 
focus of infection representing the point of 
entry of the organisms into the circulation, but 
in most of them the focus of dissemination has 
been in the lungs. In the cases in which 
Granulomatous meningitis of the spinal cord 
may occur either with or independent of the 
infection of the cerebral meninges. In one 702 
The Military Surgeon—November} 1946 
Fig. 30. Disseminated coccidioidomycosis, AIP Ac. 78078: Roentgenogram of lesion of tibia. The bone 
destruction is more pronounced in the medullary portion, but also involves the cortex. Note also the fuzzy 
appearance of portions of periosteum. Death in 140 days. 
FlG. 31. Disseminated coccidioidomycosis, AIP Ac. 94964: Erosion of vertebral bodies by granulomatous 
process. Intravertebral cartilage is also involved. Note the resemblance of the lesion seen in Fig. 32. 
Fig. 32. Disseminated coccidioidomycosis, AIP Ac. 132016: Roentgenogram of destructive lesion in lower 
cervical vertebrae. Note the soft tissue involvement and draining sinus in the neck. Death in 177 days. C occidioidomy costs 
703 
meningitis is only one of the many evidences 
of widespread dissemination, there are many 
foci from which dissemination to the meninges 
might have occurred. Some of the oldest lung 
lesions have been found in cases of fatal menin- 
infection within the bone. Psoas abscesses, al- 
ready described, have been seen in several cases 
of this series. Primary infection of the muscles 
by lodgement of organisms from the blood 
stream rarely if ever occurs. In this series of 
Fig. 33. Disseminated coccidioidomycosis. Multiple draining sinuses on dorsal and palmar aspects of 
the wrist. Note the scars on the dorsum of the hand, where previous sinuses have closed. 
gitis, coming on many months after the 
pulmonary disease has become completely 
asymptomatic. In these old pulmonary lesions, 
as already noted, there are usually still many 
organisms that show active propagation; and 
so the old, presumably healed foci are not in- 
frequently the source of a much-delayed fatal 
dissemination. 
cases we have seen no examples of this form of 
muscle involvement. 
10. The Gastro-intestinal Tract. It has long 
been known that infection of the gastroin- 
testinal tract is most unusual in cases of 
coccidioidomycosis. In our series the small 
intestine was not involved in a single instance; 
however, the peritoneal surfaces of the small 
intestine have been found to share in the in- 
fection of the peritoneal cavity. In one case 
suppurating, masses of lymph nodes ulcerating 
through the stomach wall gave rise to a granu- 
9. The Voluntary Muscles. The voluntary 
muscles adjacent to joints and overlying bony 
prominences are often involved by extension of 
the coccidioidal process from the disseminated 704 
The Military Surgeon—November, 1946 
lomatous type of ulceration in the gastric 
mucosa. In another case a mass of suppurating 
lymph nodes in the mediastinum extended to 
involve the esophagus and the trachea, forming 
a tracheo-esophageal fistula further complicated 
by extension of the sinuses from the medi- 
astinum into the neck. Direct involvement of 
the esophagus or of the buccal cavity and its 
swallowed in such abundance as they are in 
pulmonary tuberculosis. 
II. The Gemto-urinary Tract. Lesions 
have been observed in no other portion of the 
genito-urinary tract in our cases than the 
kidney and the prostate. This is in striking 
contrast to the situation in blastomycosis, in 
which involvement of the bladder and the 
FlG. 34. Forms of C. immitis and related histologic response, AIP Ac. 137706; High power magnification 
of a giant cell containing large organism filled with spores. Note the radial arrangement of the epithelioid 
cells surrounding the giant cell. Same case as Figs. 6, 19, and 28. (Xiooo) 
membranes has not been observed in this study. 
This striking absence of lesions due to C. 
immitis in the gastro-intestinal tract is a matter 
of some interest in connection with the patho- 
genesis of the disease. The extensive pulmonary 
involvement of coccidioidomycosis is much like 
that of tuberculosis, but there is no complica- 
tion in coccidioidomycosis to parallel terminal 
tuberculous enteritis. The difference may very 
likely be attributed to the fact that in coc- 
cidioidal infection of the lung little sputum is 
produced, and hence organisms are never 
ureters and extensive seminal vesiculitis are 
common. The infrequency of infection of the 
lower part of the urinary system found in this 
series of cases seems to be in harmony with 
previously recorded findings. 
VI 
The Nature and Genesis of the 
Typical Lesion of Coccidioidomycosis 
The abundance of histologic material pro- 
vided by this series of cases of coccidioidomy- 
cosis has made it possible to arrive at a C occidioidomy costs 
705 
satisfactory understanding of the pathogenesis 
of the lesion. The character of the lesion, its 
course, and its outcome appear to be de- 
termined by a number of rather specific factors 
which may be discussed individually, as 
follows: 
spherules. From the endospores develop the 
next generation of spherules in the tissues, or, 
in the external environment, the mycelial phase 
of the life cycle of this organism (Fig. I). This 
organism never appears in the tissues in 
mycelial form except when large open cavities 
exist. Hyphae then may be found in a cavity. 
Even this is exceptional and has not been 
described in the literature. In only one of our 
cases were hyphae observed (Fig. n). 
A. The Chemotactic Properties of Coccidioides 
immitis 
During the development of the coccidioidal 
lesion the organism appears in the tissues in 
The chemotactic properties of the chlam- 
Fig. 35. Forms of C. immitis and related histologic response, AIP Ac. 74661: High power magnification 
of a focal abscess found in the heart muscle. The lesion has formed around a cluster of endospores seen 
centrally in the lower pole of the lesion. Same case as Fig. 17. (X 200) 
a variety of forms, each very likely possessing 
its own peculiar chemotactic properties. The 
forms of the organism that appear in the tissues 
are; 1) the chlamydospores, which are de- 
veloped in the hyphae through a fairly simple 
process of sporulation, described in detail by 
Ophuls and Moffit3; 2) the spherules in all 
of their various developmental stages; and 3) 
the endospores resulting from endosporulation, 
which represents the maturation of the 
ydospores have not been accurately determined. 
It is this form of the organism that produces 
the original infection in the lung or in other 
tissues, where it is introduced from the ex- 
ternal environment. All of our cases are far too 
advanced for the material at our disposal to 
give any indication as to the specific cellular 
response to the presence of the chlamydospores. 
It is well known that once the infection is 
established, the chlamydospores are never again 706 
The Military Surgeon—November, 1946 
Fig. 36. Forms of C. immitis and related histologic response, AIP Ac. 46988: Biopsy of ulcer on lower 
lip. Large organism engulfed by a giant cell as seen under oil-immersion. The spherule has just begun to 
form spores. Note the epithelioid reaction surrounding the giant cell. (X 2400) C occidioidomy costs 
707 
seen in the tissues, because the mycelial phase 
of the life cycle of this organism does not 
develop except in the external environment. 
Only in the articles of Cronkite and Lack1,0 
and of Tager and Liebow207 have we found 
observations relating to the pathogenesis of 
the lesions produced by chlamydospores. Since 
these forms, at least morphologically, some- 
what resemble the spores produced by the de- 
veloping spherules in the tissues, it may be that 
the reaction to the chlamydospores will prove 
to be identical with that to the endospores. 
served in the tissues where organisms are 
present in great numbers and maturing 
rapidly. These broken-up membranes may 
sometimes be seen surrounded by macrophages 
of the type that usually phagocytize the intact 
spherules. 
The endospores, as will be seen from the 
accompanying diagram representing the life 
history of this parasite (Fig, i), are about the 
size of an ordinary neutrophilic leukocyte or 
slightly smaller. These very early forms of 
the developing parasite are extruded from the 
mature sporulating spherule into the tissues, 
where they provoke a characteristic cellular 
response. The organisms at this stage are ex- 
tremely difficult to find in the sections, but 
they may be seen when stained properly. They 
possess a high degree of attraction for the 
polymorphonuclear leukocytes of the neutro- 
philic variety; and so, surrounding these forms 
of the parasite, there regularly develops a little 
focus almost exclusively composed of these cells 
(Fig. 35). Many of the endospores, when 
they are just released from the spherules, 
possess no capsule at all or very poorly de- 
veloped ones. The polymorphonuclear leuko- 
cytes attempt to phagocytize the spores, but 
this is not often successful. Either the organ- 
isms are too large or their toxic effects interfere 
with the process of phagocytosis. After the 
spore has grown considerably, particularly after 
the capsule appears to have become well 
formed, the endospores possess also a certain 
attraction for the large mononuclear wander- 
ing cells. It is this form of the organism that 
one sees most often within the cytoplasm of 
the large mononuclear wandering cells. 
The spherules in the intermediate develop- 
mental stages appear in the tissues in varying 
numbers. In some cases they are abundant, 
and in others so scarce as to be very difficult 
to find. Regardless of the number of or- 
ganisms, however, it appears that the cells 
attracted by these forms of the parasites are 
the mononuclear wandering cells (Fig. 18) 
which accumulate in great abundance and 
almost completely replace the polymor- 
phonuclear leukocytes, which dominate the 
histologic field in the earliest stages of the 
development of the lesion (when the organisms 
are of the spore form). These mononuclear 
wandering cells undergo morphologic and 
functional changes which eventually give to 
the mature lesion its characteristic granu- 
lomatous features. 
The mature, sporulating spherules possess 
a combination of properties of the endospore 
and the immature spherule as regards their 
attraction for reacting cells (Fig. 37). How- 
ever, the moment the mature spherule ruptures 
its capsule, thus exposing to the tissues the 
endospores, the polymorphonuclear leukocytes 
gather about the structure in great numbers. It 
is a rather fascinating experience to observe 
the regularity with which the polymorphonu- 
clear leukocytes seem to rush to the endo- 
sporulating organisms and actually invade the 
ruptured capsule of the mature organism the 
moment the endospores become accessible. 
After the sporulating spherule has emptied it- 
self, the polymorphonuclear leukocytes often 
very nearly fill it. The capsular substance 
seems to possess, however, little attraction for 
these cells; empty capsules are'commonly ob- 
B. The Capacity of the Organisms to 
Injure Tissues 
The ability of the organisms in their various 
forms to destroy tissue determines the degree 
of necrosis, as well as the amount of exudation 
of fluid from the blood vessels as the lesion 
develops. Presumably the chlamydospores, and 
certainly the tissue forms of the parasite that 
develop immediately from chlamydospores, are 
capable of injuring the blood vessels as indi- 
cated by the marked extravasation of fluid in 708 
The Military Surgeon—November, 1946 
the earliest reaction to the endospores in the 
tissues. Organisms in this stage of their de- 
velopment also are highly capable of destruc- 
tion of the cells of the inflammatory exudate. 
This is observed in little foci of polymor- 
phonuclear leukocytes that appear about the 
endospores. These cells quickly show evidences 
of injury, and eventually are killed outright. 
This destructive process is indicated by the 
presence of great masses of cellular debris in 
the early lesions, also by the characteristic 
staining reactions of the dying and injured 
leukocytes. Further evidence of the capacity 
of the spores to injure tissue is the hemorrhagic 
character of the exudate in the early stages of 
the development of the infection, especially in 
the lung. 
ing C. immitis may play in the pathogenesis 
of the disease. There is no question that the 
tissues do become sensitized and, when brought 
into contact with culture filtrates of C. immitisy 
do react violently, even to the extent of be- 
coming extensively necrotic. This is a matter, 
however, which requires discriminating ex- 
perimental observation, and cannot be settled 
by clinical or anatomic techniques. In cases 
in our series highly necrotizing lesions have de- 
veloped in individuals giving no skin reaction 
whatever to coccidioidin; and, in contrast, 
very widespread lesions, especially in the lung, 
with virtually no necrosis, have been seen in 
individuals highly sensitive to coccidioidin. In 
this connection we have repeatedly made the 
observation that extensive necrosis of the lesions 
may have taken place in certain areas of the 
body, while lesions in other areas may be 
intact. This would seem to indicate that, if a 
sensitization factor is involved in the develop- 
ment of coccidioidal lesions, it is local, and not 
general. 
In view of the power of the organisms to kill 
living cells, it is remarkable that the fixed 
tissues, with the exception of the blood vessels, 
suffer only a limited degree of injury. This 
explains the complete restoration which so 
often occurs in the primary pulmonary form of 
coccidioidomycosis. In certain exceptional 
cases massive destruction of the fixed tissues 
does take place. Exactly what factors are re- 
sponsible for this necrotizing effect is difficult 
to determine. The actual number of spherules 
in the tissues may be a dominating influence, 
since the most highly necrotizing lesions that 
we have observed are those in which the 
organisms are present in abundance. What is 
probably more significant is that they are in 
the maturing sporulating phase of their de- 
velopment (Figs. 16 and 17). It is not unusual 
to see masses of organisms in the intermediate 
stages of development lying in tissues in which 
virtually no necrotizing effect can be seen 
(Fig. 37). Thus, while the number of organ- 
isms concerned is undoubtedly of importance in 
determining the character and outcome of the 
lesion, the main factor is more likely the stage 
of development of the parasite. 
D. Intracellular Ferments and Other 
Factors 
In lesions which become necrotic, subse- 
quent changes may result either in liquefaction 
of the necrotic material or in its persistence in 
a coagulative form (Figs. 24, 25, and 26). 
According to our observations, liquefaction 
occurs chiefly in those areas where the poly- 
morphonuclear leukocytes have been dominant 
in the inflammatory exudate. It has not been 
a prominent feature in those lesions in which 
the inflammatory exudate is composed chiefly 
of mononuclear wandering cells and their pro- 
ducts of development. 
Lesions characterized by the epithelioid re- 
action tend to remain intact insofar as the 
fundamental structural pattern is concerned, as 
seen so well in some of the very old residual 
lesions in the lung. The necrotic tissue often 
reveals the original pattern of the lung, as is 
so often the case in the gummatous lesions 
of syphilitic infection (Fig. 18). Although 
it is a natural assumption that this difference 
in the outcome of the necrotizing process is 
the result of the character of the ferments in- 
C. The Sensitisation Factor in Coccidioidal 
Injection 
The materials at our disposal have not 
made it possible for us to determine exactly 
what role the sensitization to products of grow- Figure 28. Disseminated coccidioidomycosis, AIP Acc. 133706: The gross appearance of disseminated 
kidney lesions. Note the yellow granulomatous nodules distributed over the outer cut surfaces in- 
volving both cortex and medulla. This is the same case as Figs. 6 and 19. Figure xi. Forms of C. immitis and related histologic response. Section of lung showing great 
number of organisms in different stages of development. Note the macrophage crowded with 
organisms, 4 in spherule form and 1 almost endosporulatmg. I he tissue reaction is of the mixet 
granulomatous type. Coccidioidomycosis 
709 
volved or actually produced by the cells of the 
inflammatory exudate, our own studies in this 
series have not made it possible to substantiate 
such a supposition. Reasoning by analogy, how- 
ever, it seems that we may be reasonably sure 
that the ferments produced by the poly- 
morphonuclear leukocytes are chiefly responsi- 
ble for the liquefactive form of necrosis seen 
in almost all the large disseminated lesions of 
coccidioidomycosis. However, undoubtedly 
there are other factors involved in the process, 
and the matter of judgment must be left open 
for settlement through experimentation. 
Nothing we have seen in our study of this 
series of cases or in reviews of former studies 
gives any indication that products of the organ- 
isms themselves influence in one way or 
another the development of liquefaction fol- 
lowing necrosis of the tissues. 
though it is clear that in the case of C. immitis 
obligate intracellular parasitism does not exist, 
an intracellular location may be most favorable 
for the propagation of these fungi and for the 
development of the characteristic reaction. C. 
immitis may thus be placed among the faculta- 
tive intracellular parasites. It may be that an 
intracellular environment is essential for the 
development of a certain stage in the life cycle 
of this fungus, and that the large mononuclear 
cells serve as hosts during this phase. 
F. Correlation of Pathogenic Factors and the 
Development of the Typical Lesion 
If one gives due consideration to the various 
factors involved in the production of lesions by 
C. immitis, there is no difficulty in under- 
standing the rather complex character of 
coccidioidal lesions. It is customary to describe 
the lesion as a granuloma; and in certain 
phases of its development the reaction is in- 
deed strictly granulomatous, that is, it is a 
lesion resulting from the response of the 
mononuclear cells to the presence of the 
fungus. Unfortunately, however, the reaction 
of the body to C. immitis is not so simple as 
this; instead, it is a combination of reactions, 
both granulomatous and suppurative.* 
E. The Proliferative (Stimulative) Factor 
One of the most striking features of the 
reaction to C. immitis is the proliferation of 
the wandering cells of the exudate and their 
transformation into large epithelioid elements 
and giant cells (Figs. 20, 23, 34, 36, and 37). 
The stimulus for this growth response is 
related directly to the organisms, as indicated 
by the observation that the large mononuclear 
cells begin to grow and reproduce as soon as 
they engulf the organisms. They increase in 
size and the nuclei multiply (Figs. 34 and 
36); the end result is the formation of the 
large giant cells. These changes do not occur 
in those mononuclear cells not in intimate 
contact with the organisms. The growth- 
stimulating influence of the parasite seems to 
affect chiefly the cells of the exudate, almost 
never the cells of the tissues. In the lungs, for 
example, the reaction develops beautifully 
within the alveolar spaces, while the alveolar 
walls remain virtually unchanged (Fig. 18). 
There would, therefore, seem to be some 
peculiar affinity between the parasites and the 
large mononuclear cells. The regularity with 
which the parasites are seen within these cells 
suggests that they may actually grow best in 
an intracellular position, just as do the filtrable 
viruses and certain other disease agents. Al- 
In the cases that we have studied the extra- 
pulmonary disseminated lesions have usually 
been chiefly suppurative; the pulmonary 
lesions, on the contrary, have been predomi- 
nantly granulomatous. It is never possible, 
however, to classify these lesions as strictly of 
one type or another because histologically they 
are all combinations of both. The gross charac- 
ter of the lesions naturally depends upon the 
predominance of one type of reaction; never- 
theless, it is possible to trace a characteristic 
sequence in the development of the “typical” 
lesion that results from C. immitis infection. 
* The distinctions between granulomatous and sup- 
purative inflammatory reactions emphasized here 
have been discussed in detail in the following refer- 
ences: (1) Forbus, Wiley D.: Reaction to Injury, 
Pathology for Students of Disease, Chapters VI and 
XXVI; (2) Forbus, Wiley D.: The Nature and 
General Pathological Significance of Granulomatous 
Inflammation, American Lecture Series, Charles C 
Thomas ( in press). 710 
The Military Surgeon—November, 1946 
The primary lesion in the lung may be taken 
as an example. The introduction of chlamydo- 
spores into the pulmonary alveoli arouses an 
immediate polymorphonuclear leukocyte reac- 
tion, the resulting picture being indistinguish- 
able from pneumonic consolidation of bacterial 
origin (Figs. 12, 13, and 14). Growth of the 
chlamydospores and their conversion to 
spherules, which are the typical forms of the 
parasite in the tissue, are accompanied by an 
influx of great numbers of mononuclear 
wandering cells. These cells surround and 
envelop the tiny organisms to serve as their 
intimate hosts (Fig. 18). The spherule under- 
goes progressive growth and development 
within the cytoplasm of the mononuclear cell 
until it reaches the stage of endosporulation. 
By this time the parasitized cell has been trans- 
formed into a large mononucleated giant cell 
(Figs. 34 and 36). Other macrophages, some 
parasitized and some not, proliferate, thus 
substituting a typical mononuclear or epitheli- 
oid reaction (Figs. 23, 34, and 36) for the 
original polymorphonuclear leukocyte exudate. 
The maturation of the spherules, the develop- 
ment of the endospores, and their discharge 
from the mother spherule result in the destruc- 
tion of the large parasitized giant cell with 
dissemination of the endospores into the tissue 
spaces. Maturation of the spherules takes place 
extracellularly at the same time. The liberated 
endospores immediately attract polymorphonu- 
clear leukocytes to the scene, the result being 
the formation of minute suppurative foci (Fig. 
35). The endospores are very difficult to see in 
the tissues in ordinary sections, but great num- 
bers of spherules in various stages of develop- 
ment are readily visible. This combination of 
suppurative foci and mononuclear and epi- 
thelioid or genuine granulomatous reaction 
constitutes the usual picture met with in the 
histologic study of this disease (Fig. 17). Ulti- 
mately, either the lesion becomes necrotic and 
liquefies (Figs. 24, 25, and 26) or necrosis 
is followed by gradual hyalinization of the 
granulomatous tissue. Scarring results from the 
healing of either type of lesion. In few of the 
lesions which we have seen, even those of the 
longest duration, has significant calcification 
occurred. In conclusion it should be empha- 
sized again that, once the lesion reaches the 
advanced granulomatous state, complete reso- 
lution without residual changes in the tissue 
no longer is possible. Thus, in primary pul- 
monary coccidioidomycosis of the type which 
heals without residual changes, the reactions 
must necessarily have been of the ordinary 
polymorphonuclear or very early mononuclear 
exudative variety. 
VII 
The study of this collection of 95 cases of 
coccidioidomycosis, all of the disseminated type, 
has presented an opportunity for further in- 
vestigation of the clinicopathologic problems 
of the disease in the light of the developments 
of recent years. We have made use of this 
opportunity by attempting to solve some of the 
problems utilizing materials which, up to this 
time, have not been available. We have been 
especially interested in the pathogenesis of the 
primary pulmonary form of this disease, the 
pathologic anatomy of which has remained 
quite unexplored because recovery is the rule 
in coccidioidomycosis restricted to the lungs. 
Although the material at our disposal has come 
from cases of the disseminated form of coc- 
cidioidomycosis, we have been able to con- 
struct, according to well-established pathologic 
principles, the probable pulmonary reactions 
to the primary infection with C. hnmitis. In 
this study, beyond the simple matter of placing 
on record a large series of carefully studied 
cases, we have hoped, among other things, to 
discover data upon which to base a more 
accurate prognosis of the disease, particularly 
with respect to the cases of coccidioidomycosis 
in the armed forces. It is evident from our 
study that disseminated coccidioidomycosis will 
continue to occur among those who have 
served in the armed forces for perhaps as long 
as ten years. These cases will represent re- 
crudescences of quiescent coccidioidal infection 
incurred in endemic areas. Undoubtedly a 
high percentage of these will be in the nature 
of widespread dissemination of coccidioidal 
lesions, and the mortality will be considerable. 
Comment Coccidioidomycosis 
711 
This is a problem for the medical profession as 
a whole, since the men who have been exposed 
to C. immitis infection will be scattered 
throughout the country, and the recognition 
of the disease will be the responsibility of the 
local practitioner. Although the problem in 
the case of the veteran will not be a large one 
and will be definitely restricted as to duration, 
it will be a matter of considerable importance 
to the government. 
for this idea in the 95 cases of our series since 
instances of residual calcification of the lungs 
were rarely observed. Although the problem 
obviously needs further study, our experience 
offers little encouragement to those who would 
attribute to coccidioidal infection the residual 
calcifications so frequently discovered in the 
lungs in the absence of a positive tuberculin 
reaction. 
In this study the nature of the materials 
dealt with has not made it possible to contrib- 
ute materially toward the solution of the 
problem of the mode of infection in man, 
particularly as regards the epidemiologic aspects 
of the problem. What we have learned cer- 
tainly supports the previously held opinion that 
man-to-man infection is of no material im- 
portance. Whether or not man receives his 
infection from a reservoir in nature consisting 
of infected small wild animals, particularly 
rodents, has not been established. The belief 
that man is infected, almost without exception, 
through inhalation of the spores of this fungus 
receives considerable support from our observa- 
tions regarding the primary pulmonary form 
of this disease. Since men do not infect each 
other with C. immitisy it is reasonable to be- 
lieve that there is no great likelihood of the 
spread of this disease into areas where it is 
not endemic. On the other hand, it is con- 
ceivable that infected human beings, of which 
there are now a great number, may be the 
means through which the organisms may be 
spread to places where th» conditions are 
favorable for their propagation in the outer 
environment. From what is known about the 
organism and its life in the outer environment, 
however, it is improbable that this constitutes 
a serious threat. 
Although extensive histopathologic studies 
have not brought forth data which would re- 
move coccidioidal infection from the category 
of granulomatous disease, it has become ap- 
parent that, like certain other fungus infec- 
tions, coccidioidomycosis belongs to a patho- 
genic type of infectious disease that lies 
somewhere between the purer forms of granu- 
lomatous infection represented by tuberculosis 
and the genuinely suppurative infections pro- 
duced by the more common pyogenic bacteria. 
Ample justification for the intermediate classi- 
fication of this disease will be apparent to 
anyone who devotes himself seriously to the 
study of the pathogenesis of the great variety of 
lesions which characterize coccidioidomycosis 
as it develops from the earliest stages of in- 
fection to the widely disseminated and highly 
complicated gross and histologic expressions of 
the terminal process. 
VIII 
1. A study of 95 cases of the disseminated 
form of coccidioidomycosis occurring in the 
armed forces between 1941 and 1946 has 
been made with particular reference to the 
pathogenesis of the disease. Fifty of these were 
fatal cases studied comprehensively clinically 
and at autopsy. Forty-five cases were studied 
clinically and by biopsy. 
2. The analysis of this group of cases appears 
to justify the following statements: 
a. Coccidioidomycosis is primarily and pre- 
dominantly a pulmonary disease, the mortality 
of which is negligible. When death occurs, it 
is always associated with an endogenous rein- 
fection, that is, the disseminated form of the 
disease. 
Summary and Conclusion 
An important aspect of the problem of 
coccidioidomycosis of current interest is the 
confusion of residual coccidioidal lesions in 
man with residual tuberculous lesions. The 
studies of Aronson and associates186’187 and 
others seemed to indicate that a considerable 
number of calcified foci in the lungs believed 
to be of tuberculous origin might be due to 
C. immitis infection. Little support was found 
b. Coccidioidomycosis is a granulomatous 712 
The Military Surgeon—Novembery 1946 
infectious disease with pathogenic and patho-» 
logic-anatomic characteristics which place it 
between the typical granulomatous disease, as 
represented by tuberculosis, and the typical 
acute suppurative infectious disease, as repre- 
sented by the ordinary pyogenic infections. 
i. The ten most frequent sites of the lesions 
resulting from dissemination of C. immitis by 
the blood stream are, in order, as follows: 
(1) the lungs 
(2) the lymph nodes 
(3) the spleen 
(4) the skin and subcutaneous tissues 
(5) the liver 
(6) the kidney 
(7) the bones 
(8) the meninges 
(9) the adrenal 
(10) the myocardium 
c. Coccidioidomycosis occurs in two clinico- 
pathologic forms: 
(1) Primary pulmonary coccidioidomy- 
cosis, occurring in subclinical and 
clinical forms—the primary 
tion complex, and 
(2) Endogenous reinfection coccidioi- 
domycosis, the secondary dissemi- 
nated or generalized infection. 
j. The characteristic reaction of the body 
to C. immitis is purely inflammatory; the re- 
action may be (I) suppurative, the reacting 
cells of the exudate being predominantly poly- 
morphonuclear neutrophilic leukocytes, or 
(2) granulomatous, the reacting cells of the 
exudate being predominantly large mononu- 
clear wandering cells (“reticulo-endothelial 
cells,” “monocytes,” “histiocytes,” “tissue 
macrophages”) and their epithelioid and giant 
cell derivatives, or (3) mixed, in which the 
suppurative and granulomatous reactions are 
about equally distributed. 
d. Primary pulmonary coccidioidomycosis 
in its mildest, often subclinical, forms is repre- 
sented anatomically by essentially non- 
granulomatous inflammation of the lung of 
either a focal or diffuse character, accom-f 
panied by bronchial and mediastinal acute 
lymphadenitis and, in a certain percentage of 
the cases, erythema nodosum. In its severe, 
clinically obvious form genuine suppurative 
and granulomatous lesions develop; these are 
sometimes highly destructive, resulting in the 
formation of pulmonary cavities. 
k. The histologic character of the inflam- 
matory reaction varies greatly in any given 
lesion, and from lesion to lesion in the same 
individual, and appears to be determined 
chiefly by the number and the developmental 
stage of the organisms present in the lesion. 
The factor of sensitization of the individual to 
coccidioidin does not appear to be material. 
e. Primary pulmonary coccidioidomycosis 
usually heals completely without residual puE 
monary lesions, but the rare suppurative and 
granulomatous necrotizing lesions always heal 
by scar formation. Calcification of these le-r 
sions is not common. 
f. Reinfection pulmonary coccidioidomy- 
cosis may arise by intrapulmonary or extra- 
pulmonary bronchial ulceration and aspiration 
of infected exudate, or by vascular dissemina- 
tion of organisms from both intra- and extra- 
pulmonary sources. 
I. Absolute diagnosis of coccidioidomycosis 
can be made solely upon demonstration of the 
organisms in the tissues, exudate, or body 
fluids. This can be done in virtually all cases 
by microscopic examination. Cultures or 
animal inoculations are rarely required when 
tissues can be obtained for study. 
g. Secondary or disseminated coccidioido- 
mycosis may, and frequently does, develop dur- 
ing the active progress of the primary pul- 
monary infection. 
m. It is unlikely that the healing of pul- 
monary coccidioidomycosis is responsible for 
the presence of focal calcification of the lungs 
in any significant proportion of the cases show- 
ing this lesion in the absence of a positive 
tuberculin reaction. 
h. Clinically completely healed but residual 
pulmonary lesions contain vegetative organisms 
in all stages of development. These may serve 
as the source of disseminated infection months 
or years after the pulmonary disease has 
disappeared. 
n. Disseminated endogenous reinfection 
will continue to occur among the members of C occidiotdomy costs 
713 
the armed forces who were trained in the 
endemic areas of the Southwest for approxi- 
mately ten years, and will cause a number of 
deaths. 
mould (formerly described as protozoon “Coccidi- 
oides immitis pyogenes”), Phila. M. J. 1900, 5:1471. 
4 Posadas: Psorospermiose infectante generalisee, 
Rev. de chir., Paris, 1900, 20:275. 
*9°4 
3 Wolbach, S. B.: The life cycle of the organism 
of dermatitis coccidioides, J. Med. Res. 1904, 
13:53-60. 
1905 
8 Ophuls, W.: Coccidioidal granuloma, J.A.M.A. 
1905, 45 :i29i-i296. 
7 Ophuls, W.: Further observations on a patho- 
genic mould formerly described as a protozoon 
(Coccidioides immitis, Coccidioides pyogenes), J. 
Exper. Med. 1905, 6:443. 
1906 
8 Brown, P. K.; Report of fifteenth and sixteenth 
cases of coccidioidal granuloma, California State J. 
M. 1906, 4:324-326. 
I9°7 
9 Hektoen, L.: Systemic blastomycosis and coc- 
cidioidal granuloma, J.A.M.A. 1907, 49:1071-1077. 
1909 
10 Evans, N.: Coccidioidal granuloma and blasto- 
mycosis in the central nervous system, J. Tennessee 
Med. Assn. 1909-1910, 2:166-174. 
191 2 
11 von Wasielewski: Coccidioides immitis, Deutsche 
med. Wchnschr. 1912, 38:2341. 
*9*3 
12 MacNeal, W. J., and Hjelm, C. E.: Note on a 
mold, Coccidioides immitis, found in a case of gen- 
eralized infection in man, J.A.M.A. 1913, 61:2044. 
*9*4 
13 Brown, P. K., and Cummins, W. T.: A differen- 
tial study of coccidioidal granuloma and blastomy- 
cosis; pathology and bacteriology, report of two 
additional cases of coccidioidal disease, Tr. Assn. Am. 
Phys. 1914, 29:628-650. 
14 Cooke, J. V.: Immunity tests in coccidioidal 
granuloma, Proc. Soc. Exper. Biol. & Med. 1914-15, 
12:35- 
15 MacNeal, W. J., and Taylor, R. M.: Coc- 
cidioides immitis and coccidioidal granuloma, J. Med. 
Res. 1914, 30:261-274. 
*9*5 
16 Campiche, P.: The Californian blastomycosis 
(oidium coccidioides), Rev. Med. Suisse Rom. 1915, 
35:769-779. 
17 Dickson, E. C.: Oidiomycosis in California with 
especial reference to coccidioidal granuloma, includ- 
ing nine new cases of coccidioidal granuloma and 
o. It does not appear likely that coccidioido- 
mycosis will spread to non-endemic areas as a 
result of the migration of members of the 
armed forces who were exposed to this disease 
during training in endemic areas. 
Acknowledgement 
This study was carried out through the 
extraordinary facilities of the Army Institute 
of Pathology, made available to us by Colonel 
J. E. Ash, Director. It was suggested by 
Colonel Balduin Lucke, Deputy Director, 
whose support it has continuously had. Without 
the facilities of the Surgeon General’s library, 
whose staff gave of their time and interest quite 
beyond the call of duty, the comprehensive 
survey of the problem presented would have 
been impossible. 
The illustrations are the work of Mr. Roy 
M. Reeve and his staff of the Army Institute 
of Pathology and of Mr. Elon Clark and his 
staff of the Department of Medical Illustra- 
tion, Duke University School of Medicine. 
For editorial assistance generously given under 
trying circumstances, we are indebted to Mrs. 
Helen Knight Steward, Editor of the Army 
Institute of Pathology, and to Miss Jeanne 
Pinaire, Secretary of the Department of 
Pathology, Duke University School of Medi- 
cine. 
This investigation had the combined finan- 
cial support of the Office of the Surgeon Gen- 
eral of the Army and of the Commission on 
Epidemiological Survey, Army Epidemiologi- 
cal Board, 
189 2 
1 Wernicke, R.: Uber einen Protozoenbefund bei 
Mycosis fungoides, Zentralbl. f. Bakt. 1892, 12:859. 
1896 
2 Rixford, E., and Gilchrist, T. C.: Two cases of 
protozoan (coccidioidal) infection of the skin, Re- 
ports Johns Hopkins Hosp. 1896, 1:209. 
1900 
3 Ophuls, W., and MofEt, H. C.: A new pathogenic 
Literature 714 
The Military Surgeon—November, 1946 
one of systemic blastomycosis, Arch. Int. Med. 1915, 
16 :i 028-1044. 
1920 
18 Lynch, K. M.: Coccidioidal granuloma includ- 
ing the first reported case east of the Mississippi, 
South. M. J. 1920, 13:246. 
1923 
10 Hirsch, E. F.: Introduction of coccidioidal 
granuloma into Chicago, J.A.M.A. 1923, 81:375- 
377- 
20 Taylor, R. G.: Coccidioidal granuloma, Am. J. 
Roentgenol. 1923, 10:551. 
1924 
21 Gomes, J. M., and deAssumpsao: En tourno de 
genero coccidioides, Ann. paulist de med. e cir. 1924, 
15:49-61. 
1925 
22 Bump, W. S.: Observations on the growth of 
Coccidioides immitis, J. Infect. Dis. 1925, 36:561- 
565. 
23 Montenegro, J.: Septicemia by Coccidioides im- 
mitis, Brasil-med. 1925, 39:69. 
24 Riesman, D., and Ahlfeldt, F. E.: Coccidioidal 
granuloma, Tr. Assn. Am. Phys. 1925, 40:300-307. 
25 Rocha Lima: Histopathologic der exotischen 
Blastomykosen, Verhandl. d. deutsch. path. Gesellsch. 
1925» 2o:342-353- 
1926 
26 Ahlfeldt, F. E.; Studies in coccidioidal granu- 
loma, mode of infection, Arch. Path. 1926, 2:206- 
216. 
27 Proescher, F., Ryan, F., and Krueger, A. P.: 
Case of coccidioidal granuloma with autopsy find- 
ings, J. Lab. & Clin. Med. 1926, 12:57-70. 
1927 
28 Castellani, A.: Fungi, Arch. Dermat. & Syph. 
1927, 16:383, 571, 714. 
29 daFonseca, O., and deArea Leao, A. E.: Com- 
plement deviation in coccidioidal granuloma; sensi- 
tiveness of filtrate from culture of Coccidioides im- 
mitis used as antigen, Sciencia med. 1927, 5:682-683. 
38 daFonseca, O., and deArea Leao, A. E.: Specific 
cutaneous reaction of filtrate from culture of Coccidi- 
oides immitis, Compt, rend. Soc. de biol. 1927, 
97:i796-i797- 
31 Delore, P., and Rosette, M.: Tuberculose et 
mycoses, J. de med. de Lyon 1927, 8:205, 231. 
32 deSonza Campos, E., and deAlmeida, F. P. : 
Coccidioidal granuloma; 1 o cases at Sao Paulo, Ann, 
Fac. de med. da Univ. de S. Paulo 1927, 2:203-220. 
33 Guy, W. H., and Jacob, F. M.: Granuloma 
coccidioides, Arch. Dermat. & Syph. 1927, 16:308- 
311 • 
34 Hirsch, E. F.: Skin reactions with coccidioidal 
granuloma, Tr. Chicago Path. Soc. 1923-27, 12: 
335- 
35 Hirsch, E. F., and Benson, H.; Specific skin and 
testis reactions with culture filtrates of Coccidioides 
immitis, J. Infect. Dis. 1927, 40:629-633. 
38 Hirsch. E. F., and D’Andrea, D.: Specific sub- 
stance of Coccidioides immitis, J. Infect. Dis. 1927, 
40:634-637. 
37 Jacobson, H. P.: Granuloma coccidioides, appar- 
ently successfully treated with colloidal copper; 2 
cases, California and Western Med. 1927, 27:360- 
364. 
33 Kelton, W.: Coccidioidal granuloma, Northwest. 
Med. 1927, 26:92-93. 
39 Riesman, D., and Ahlfeldt, F. E.: Coccidioidal 
granuloma, review of clinical data with report of a 
Pennsylvania case, Am. J. M. Sc. 1927, 174:151-167. 
40 Shear, C. L.: Mycology in relation to human 
pathology, Am. Naturalist 1927, 61:151-159. 
1928 
41 Castellani, A.: Blastomycosis and some other 
conditions due to yeastlike fungi (budding fungi), 
Am. J. Trop. Med. 1928, 8:379-422. 
42 daFonseca, O., and deArea Leao, E. A.: Coc- 
cidioidal granuloma; forms of development of the 
parasite in the tissues, in most of the lymph nodes, 
and in cultures; systematic position of Coccidioides 
immitis, Compt. rend. Soc. de biol. 1928, 98 ;6i9-62i. 
43 deArea Leao, A. E.: More rapid development of 
Coccidioides immitis in culture media to which fil- 
trates of other cultures of the microorganism have 
been added (vaccinated cultures), Compt. rend. Soc. 
de biol. 1928, 99:883-884. 
44 Jacobson, H. P.: Coccidioides granuloma, Calif. 
& West. Med. 1928, 29:392-396. 
45 Jacobson, H. P.: Coccidioidal granuloma, 
specific, allergic, cutaneous reaction, experimental 
and clinical investigations, Arch. Dermat. & Syph. 
1928, 18:562-567. 
“Kogoj, F.: Hematogene Dermatomykosen nach 
subcutaner und intramuskularer Infektion, Arch. f. 
Dermat. u. Syph. 1928, 154:463-475. 
47 Stark, N. A., and Becker, F. E.; Coccidioides 
granuloma, case, Colorado Med. 1928, 25:196-202. 
“Tomlinson, C. C.: Coccidioides granuloma, M. 
Clin. North America 1928, 12:457-462. 
“Tomlinson, C. C., and Bancroft, P.: Coccidioides 
granuloma, case responding favorably to antimony 
and potassium tartrate, J.A.M.A. 1928, 91:947-951. 
50 Walters, P. R.: Coccidioides granuloma, case, 
California & West. Med. 1928, 29:188-189. 
I979 
81 Ahlfeldt, F. E.: Special observations on mor- 
phology of Coccidioides immitis, J. Infect. Dis. 
1929, 44:277-281. 
52 Beck, M. D.: Occurrence of Coccidioides immitis Coccidioidomycosis 
715 
in lesions of slaughtered animals, Proc. Soc. Exper. 
Biol. & Med. 1928-29, 26:534-536. 
53 Burgess, J. F.: Coccidioidal granuloma, case, 
Brit. J. Dermat. 1929, 41:145-148. 
54 Cummins, W. T., Smith, J. R., and Halliday, 
C. H.: Coccidioidal granuloma; epidemiologic survey 
with report of 24 additional cases, J.A.M.A. 1929, 
93 :i046-1049. 
55 daCosta, A. F., Jr.: Treatment of coccidioidal 
granuloma by gold-salts, case, Brasil-med. 1929, 
43 ;i400-i402. 
56 DeAlmeida, F. P.: Comparative study of coccidi- 
oidal granuloma in United States and Brazil, Ann. 
Fac. de med. de Sao Paulo 1929, 4:91-98. 
57 Dickson, E. C.: Mimicry of tuberculosis by 
coccidioidal granuloma, Tr. Assn. Am. Phys. 1929, 
44:284-294. 
58 Evans, N., and Ball, H. A.: Coccidioidal granu- 
loma, analysis of 50 cases, J.A.M.A. 1929, 93:1881- 
1885. 
59 Jacobson, H. P.; Coccidioidal granuloma; fur- 
ther observations with report of seven additional 
cases, M. J. & Rec. 1929. 130:424-498. 
60 Kadisch, E.; Beitrage zur Lehre von den Derma- 
tomykos<?n; Uber den Verbleib intrakardial injizierten 
Pilzmaterials beim Meerschweinchen, Arch. f. 
Dermat. u. Syph. 1929, 158:349-353. 
61 Kalichman, G. S., and Madsen, L. J.: Coccidi- 
oidal granuloma, case, California & West. Med. 1929, 
31 :i4i-i42. 
82 Lemon, W. S.: Clinical manifestations of coccidi- 
oidal granuloma, Proc. Staff Meet. Mayo Clin. 1929, 
4:305- 
63 Ophuls, W.: Discussion of paper by Pulford and 
Larson, J.A.M.A. 1929, 93:1049. 
94 Pulford, D. S., and Larson, E. E.: Coccidioidal 
granuloma; case treated by intravenous dye, colloidal 
lead, and colloidal copper with autopsy observations, 
J.A.M.A. 1929, 93:1049-1055. 
60 Ragle, H. E.: Coccidioidal granuloma with re- 
port of case, U. S. Nav. M. Bull. 1929, 27:657-661. 
88 Stockton, A. B.: Coccidioidal granuloma, treat- 
ment with thymol, case, California & West. Med. 
1929, 31 :278-28o. 
1930 
87 Clipman, E. D., and Templeton, H. J.; Coc- 
cidioidal granuloma, Arch. Dermat. & Syph. 1930, 
21 :259-278. 
“DaFonseca, O.: Coccidioides immitis and coccidi- 
oidal granuloma, Seuchenbekaempfung 1930, 7:237- 
250. 
89 DeAlmeida, F.: Comparative study of coccidi- 
oidal granuloma in United States and Brazil, estab- 
lishing new genus for Brazilian parasite, Ann. Fac. 
de med. de Sao Paulo, 1930, 5:125-141. 
70 DeAlmeida, F.: Differential features of etiologic 
agent of coccidioidal granuloma of U.S.A. and that 
of Brazil, Compt. rend. Soc. de biol. 1930, 105:315- 
316. 
11 Haberfelt, W.: Atypical sporocytosis of the 
lymph glands (blastomycosis), Arch. f. Schiffs- u. 
Tropen-Hyg. 1930, 34:549-554. 
72 Hirsch, E. E., and D’Andrea, D.: Allergic testis 
reactions in guinea pigs with coccidioidal granu- 
loma, J. Immunol. 1930, 18:121-125. 
73 Jacobson, H. P.: Coccidioidal granuloma, clini- 
cal and experimental review with report of cases, 
Arch. Dermat. & Syph. 1930, 21:790-817. 
74Jaffe, R. M.: Microscopic changes in coccidi- 
oidal granuloma, Virchows Arch. f. path. Anat. 1930, 
278 :42-6i. 
75 Kurotchkin, T. J., and Lim, C. E.: Anaphylaxis 
with water soluble specific substance from yeast-like 
fungi, Proc. Soc. Exper. Biol. & Med. 1930, 28:223. 
78 MacLeod, J. M. H.; Mycosis due to yeast-like 
fungi, Brit. M. J. 1930, 1:1119-1123. 
77 Pulford, D. S.: Coccidioidal granuloma, Tr. A. 
Res. & Ex-Res. Physicians, Mayo Clin. 1930, 10:94- 
100. 
78 Radaelli, P.: Biology of Coccidioides immitis 
and its systematic position, Boll. d. Soc. ital. di biol. 
sper. 1930, 5:752-756. 
79 Rand, C. W.: Coccidioidal granuloma; 2 cases 
simulating tumor of spinal cord, Arch. Neurol. & 
Psychiat. 1930, 25:502-511. 
80 Timpano, M.: Experiments in coccidioidal 
granuloma and Its agent Coccidioides immitis, Boll, 
d. Soc. med.-chir. di Pavie, 1930, 44:679-722. 
1931 
81 Beck, M. D., Traum, J., and Harrington, E. S.: 
Coccidioidal granuloma, occurrence in animals, ref- 
erence to skin tests, J. Am. Vet. M. A. 1931, 78:490- 
499- 
82 Carter, R. A.: Coccidioidal granuloma, roentgen 
diagnosis, Am. J. Roentgenol. 1931, 25:7x5-738. 
83 Chou, C. H., and Reiss, F.: Coccidioidal endo- 
phthalmitis, Nat. M. J. China, 1931, 17:747-754. 
81 Dickson, E. C., and Beck, M. D.: Coccidioidal 
granuloma, California Dept, of Public Health, Spe- 
cial Bull. No. 57, 1931. 
1932 
85 Ball, H. A.: Healing of coccidioidal granuloma 
lesion following medicinal therapy, J.A.M.A. 1932, 
98 :2279-228o. 
86 Basgal, W.; Pulmonary coccidioidal granuloma, 
Rev. de path, comparee 1932, 32:273-299. 
8‘Bruhns, C.: fibersicht liber die neuren Ergeb- 
nisse und Fortschritte auf dem Gebiete der Mykologie, 
Zentralbl. f. Haut- u. Geschlechtskr. 1932, 41:1-27. 
88 Caldwell, G. T.: Coccidioidal granuloma, 3 
cases recognized in Texas, Texas State J. M. 1932, 
23:327-333. 
89 Childrey, J. H., and Gray, P. A.: Coccidioidal 716 
The Military Surgeon—November, 1946 
granuloma, primary in nasopharynx, case, California 
& West. Med. 1932, 37:250-252. 
Cicero, R. E.: Coccidioidal granuloma, case, 
Gac. med. de Mexico 1932, 63:139-157. 
Davis, R. G.: Coccidioidal granuloma, case, 
U. S. Nav. M. Bull. 1932, 30:519-522. 
' DeAlmeida, F. P.: Morphology of Coccidioides 
immitis in tissues, Ann. Fac. de med. de Sao Paulo 
1932, 7:117-123. 
93 DeAlmeida, F. P.: Identity of Coccidioides im- 
mitis and Pseudococcidioides mazzai, Ann. Fac. de 
med. de Sao Paulo 1932, 8:83-91. 
94 Montgomery, D. W.: Historical position of Coc- 
cidioides immitis among pathogenic fungi of skin, 
Ann. M. Hist. 1932, 4:199-202. 
95 Moore, M.: Coccidioidal granuloma; a classifi- 
cation of the causative agent, Coccidioides immitis, 
Ann. Missouri Bot. Garden, 1932, 19:397-428. 
8 Penna de Azevedo, A.; Meningeal plaques found 
at autopsies of cases of blastomycosis caused by 
Coccidioides immitis, Compt. rend. Soc. de biol. 1932, 
109:125-128. 
97 Perrin, T. G.: Coccidioidal granulomatosis, 
demonstration of Coccidioides immitis, Gac. med. de 
Mexico 1932, 65:158-161. 
98 Rixford, E.: Coccidioidal granuloma, Proc. Staff 
Meet. Mayo Clin. 1932, 7:352-355. 
Stewart, R. A., and Meyer, K. F.: Isolation of 
Coccidioides immitis Stiles from soil, Proc. Soc. 
Exper. Biol. & Med. 1932, 29:937-938. 
100 Williams, H. B.; Non-tuberculous lung dis- 
eases, case, M. Bull. Vet. Admin. 1932, 8:435-441. 
101 Zeisler, E. P.: Chronic coccidioidal dermatitis, 
unusual case, Arch. Dermat. & Syph. 1932, 25:52-71. 
*933 
10" Boyers, L. M.: Therapy in 2 cases of infection 
with Coccidioides fungus, apparently effective use 
of creosote and guiacol, M. Herald 1933, 52:61-67. 
103 Imerman, S. W., and Imerman, C. P.: Coccidi- 
oidal granuloma, primary cutaneous lesion, treatment 
with actual cautery, Southwest. Med. 1933, 17:18-21. 
104 Kawatsure, S.: Animal experiments with causa- 
tive agents of so-called American blastomycosis 
(Scopulariopsis americana, Aleurisma tularense, and 
Coccidioides immitis), Arch. f. Dermat. u. Syph. 
193 3) 169:173-199. 
104 Ota, M., and Kawatsure, S.: Zur Aetiologie 
der echten und falschen Blastomykosen, besonders 
der Gilchristschen Krankheit, Arch. f. Dermat. u. 
Syph. 1933, 169:149. 
“"Reaver, D. C., and Furrer, E. D.: Pathogenic 
yeasts and yeast-like organisms, report of case in 
Minnesota simulating coccidioidal granuloma, J. Lab. 
& Clin. Med. 1933, 18:329-348. 
107 Smith, L. M.: Coccidioidal granuloma, case 
originating in West Texas, Arch. Dermat. & Syph. 
W33) 28:175-181, 
108 Takahashi, S.: Experimental infection with 
Coccidioides immitis, Arch. f. Dermat. u. Syph. 1933, 
168 :597~6io. 
109 Walsh, G.; Yeast infections; coccidioidal gran- 
uloma, J. M. A. Alabama 1933, 2:434-437. 
*934 
110 Benham, R. W.: Fungi of blastomycosis and 
coccidioidal granuloma, Arch. Dermat. & Syph. 
1934) 30:385-400. 
111 Carter, R. A.; Infectious granulomas of bones 
and joints with special reference to coccidioidal 
granuloma, Radiol. 1934, 23:1-16. 
112 Castellani, A., and Jacono, I.: Unusual forms 
and aspects of Coccidioides immitis in vivo, Ann. de 
med. nav. e colon. 1934, 40:625-634. 
113 Ciferri, R., and Radaelli, P.: Phenomena of 
conjugation and endosporulation in vitro in Coccidi- 
oides immitis Stiles, Boll. Soc. ital. Biol. sper. 1934, 
9 :6o2-6o4. 
114 Ciferri, R., and Radaelli, P.: La vita saprofitica 
del fungo e ex il suo potere ammonizank, Boll. Soc. 
ital. Biol. sper. 1934, 9:746-748. 
115 Ciferri, R., and Radaelli, P.; Systematic position 
of Coccidioides immitis Stiles, Soc. internaz. de 
microbiol., Boll. d. sez. ital. 1934, 6:373-376. 
118 DeAlmcida, F. P.: Further proof of identity of 
Coccidioides immitis and Pseudococcidioides mazzai, 
Ann. Fac. de med. da Univ. de S. Paulo 1934, 10; 
29-32. 
11 DeAlmeida, F. P.: Formacoes radiadas da mem- 
brana dos cogumelos parasites, Ann. Fac. de med. 
da Univ. de S. Paulo 1934, 10:163-174. 
Lutz, A.: Coccidioidal infection in oral mucosa, 
first case observed in South America, history of hypo- 
blastomycosis americana, Mem. Inst. Oswaldo Cruz 
1934) 28:585-601. 
119 McDonald, C.: Coccidioides immitis, J. Lab. 
& Clin. Med. 1934, 20:47-53. 
120 Penna de Azevedo, A.: Blastomycosis of supra- 
renal gland due to Coccidioides immitis without lym- 
phatic lesions and with foci of fibrosis in lungs, Mem. 
Inst. Oswaldo Cruz 1934, 29:189-193. 
1-1 Radaelli, P., and Ciferri, R.: Studies on Coc- 
cidioides immitis; presence of coccidioidal granuloma 
in Europe, Soc. internaz. di microbiol., Boll. d. sez. 
ital. 1934) 6:255-257. 
1" Ragle, H. E.: Coccidioidal granuloma, Tr. 
Hawaii Terr. M. A. 1934, No. 44, pp. 47-50. 
1-3 Smith, L., and Waite, W. W.: Coccidioidal 
granuloma, fatal case, Southwest. Med. 1934, 18:305. 
124 Stowe, W. P.; Simple technic for finding Coc- 
cidioides immitis, J. Lab. & Clin Med. 1934, 19:1013. 
128 Tomlinson, C. C., and Bancroft, P.: Granuloma 
coccidioides; further observations on use of antimony 
and potassium tartrate and roentgen rays in treat- 
ment; additional case, J.A.M.A. 1934, 102:36-38. 
r935 
128 DeAlmeida, F. P.: Consideracoes en torno dos C occidioidomycosis 
717 
agentes etiologicos das blastomycoses, Ann. paulist. 
de med. e cir. 1935, 29:11-30. 
127 Fennel, E. A.: Coccidioidal granuloma in 
Hawaii, Proc. Clin. Honolulu 1935, 1:1-6. 
128 Lehman, C. F., and Pipkin, J. L.: Coccidioidal 
granuloma (chronic, hypertrophic), Arch. Dermat. 
& Syph. 1935, 31 =587. 
129 Sorsky, E. D., and Nixon, C. E.: Coccidioidal 
granuloma with report of 18 cases with two apparent 
cures, California & West. Med. 1935, 42:98-106. 
1936 
130 Abbott, K. H., and Cutler, O. L: Chronic 
coccidioidal meningitis; review of literature and re- 
port of seven cases, Arch. Path. 1936, 21:320-330. 
131 Courville, C. B.: Primary chronic coccidioidal 
meningitis; case with extensive subarachnoid infiltra- 
tion at base of brain and about spinal cord, Bull. Los 
Angeles Neurol. Soc. 1936, 1:116-119. 
132 DeAlmeida, F. P.: Cardiac inoculation of Coc- 
cidioides immitis, Folia din. et biol. 1936, 8:67-69. 
133 Duckett, T. G., and Fredeen, R. C.; Coccidioidal 
granuloma, J. Kansas M. Soc. 1936, 37:111-114. 
134 Jordon, J. W., and Weidman, F. D.: Blasto- 
mycosis and coccidioidal granuloma; comparison of 
North and South American diseases with special 
reference to Paracoccidioides brasiliensis, Arch. 
Dermat. & Syph. 1936, 33:31-47. 
135 Langeron, M.: Dermatitis from coccidioides, 
Norw. prat. derm. (Darier, Sabouraud, et al) 1936, 
2:457-484. 
138 Sox, H. C., and Dickson, E. C.: Experimental 
therapy in coccidioidal granuloma, J.A.M.A. 1936, 
106:777-779. 
137 Van Cleve, J. V.: Coccidioidal granuloma, J. 
Kansas M. Soc. 1936, 37:54. 
138 Whalen, E. J.; Pathogenic fungi, Arch. 
Otolaryng. 1936, 24:436-454. 
1927 
139 Coccidioidal granuloma in California in 1934- 
1935, California & West. Med. 1937, 46:282. 
140 Dickson, E. C.: “Valley fever” of San Joaquin 
Valley and fungus coccidioides, California & West. 
Med. 1937, 47:151-155. 
141 Dickson, E. C.: Coccidioides infection, Arch. 
Int. Med. 1937, 59:1029. 
142 Gifford, M. A., Buss, W. C., Douds, R. T.: 
Kern County (Calif.) Health Dept. Reports, July 1, 
1936, to June 30, 1937. 
143 Kehoe, E. J.: Coccidioidal infections of lung, 
M. Bull. Vet. Admin. 1937, 13:243-246. 
144 Miller, F. P.: Pulmonary manifestations of 
coccidioidal granuloma, Dis. of Chest 1937, 3:21-24. 
145 Zelman, J.: Disseminated coccidioidal gran- 
uloma (case), California & West. Med. 1937, 47: 
327-329. 
r938 
148 Courville, C. B., and Abbott, K. H.; Pathology 
of coccidioidal granuloma of central nervous system 
and its envelopes, Bull. Los Angeles Neurol. Soc. 
1938, 3:27-41. 
147 Dickson, E. C.: Primary coccidioidomycosis, 
J.A.M.A. 1938, 111:1362-1365. 
148 Dickson, E. C., and Gifford, M. A.: Coccidi- 
oidomycosis, primary type of infection, Arch. Int. 
Med. 1938, 62:853-871. 
149 Epstein, E.: Prognostic significance of cutaneous 
lesions in coccidioidal granuloma, Arch. Dermat. & 
Syph. 1938, 38:752-755- 
150 Gifford, M. A.: Coccidioidomycosis, Appendix, 
Kern County (Calif.) Dept, of Public Health Annual 
Report 1938-39. 
101 Hurwitz, S., Young, J. E., and Eddie, B. U.: 
Coccidioides immitis intradermal skin reaction, pre- 
liminary report of 449 cases, California & West. Med. 
1938, 48:87-89. 
152 Lack, A. R.: Spherule formation and endo- 
sporulation of fungus coccidioides in vitro, Proc. 
Soc. Exper. Biol. & Med. 1938, 38:907-909. 
153 Moore, M.: Blastomycosis, coccidioidal gran- 
uloma, and para-coccidioidal granuloma; compara- 
tive study of North American, South American, and 
European organisms and clinical types, Arch. Dermat. 
& Syph. 1938, 38:163-190. 
334 Nino, F. L.: Microscopic aspects of so-called 
blastomycotic granuloma, Prensa med. argent. 1938, 
25 :2203-22I4. 
753 Stewart, R. A., and Meyer, K. F.: Metabolism 
of Coccidioides immitis (Stiles), J. Infect. Dis. 1938, 
63 :i96-205, 
M. T.; Finding of fungi in sputum, 
J. Lab. & Clin. Med. 1938, 23:553-565. 
*939 
157 Brown, O. H.: Coccidioidomycosis infection in 
Arizona; allergic factors in nodules, Southwest. Med. 
1939> 23:131-132. 
“Cox, A. J., and Smith, C. E.; Arrested pul- 
monary coccidioidal granuloma, Arch. Path. 1939, 
27:717-734. 
1S9 Faber, H. K., Smith, C. E., and Dickson, E. C.: 
Acute coccidioidomycosis with erythema nodosum 
(San Joaquin fever) in children, J. Pediat. 1939, 
15 :i63-i7i. 
100 Farness, O. J., and Mills, C. W.: Primary in- 
fection with coccidioides in lung with cavity forma- 
tion and healing, case, Am. Rev. Tuberc. 1939, 
39 :z66-273. 
161 Gifford, M. A.: Coccidioidomycosis, Kern 
County (Calif.), Session on Fungus Infection, Sixth 
Pacific Science Congress of the Pacific Science Assn., 
Aug. 1939. 
162 Hynes, K. E.: Coccidioidal granuloma (2 cases, 
1 treated with sulfanilamide), Northwest. Med. 
1939, 38:19-21- 
163 Jacobson, H. P.: Immunotherapy for coccidi- 718 
The Military Surgeon—November, 1946 
oidal granuloma, cases, Arch. Dermat. & Syph. 1939, 
40:521-540. 
184 Kessel, J. F.: Coccidioidin skin test, Am. J. 
Trop. Med. 1939, 19:199-204. 
188 McMaster, P. E., and Gilfillan, C.: Coccidioidal 
osteomyelitis, J.A.M.A. 1939, 112-1233-1237. 
168 Phillips, E. W.: Presence of coccidioidal infec- 
tion in Phoenix, Southwest. Med. 1939, 23:48-51. 
187 Ruddock, J. C., and Hope, R. B.: Coccidioidal 
peritonitis, diagnosed by peritoneoscopy, J.A.M.A. 
1939, 113:2054-2055. 
188 Storts, B. P.: Coccidioidal granuloma simulating 
brain tumor in child of four years, J.A.M.A. 1939, 
112:1334-1335. 
189 Thorner, J. E.: Erythema nodosum in childhood 
associated with infection by Oidium coccidioides (7 
cases), Arch. Pediat. 1939, 56:628-638. 
California State Dept, of Public Health 1941, 20: 
X13. 
183 Smith, L. M.: Diagnosis of blastomycosis-like 
infections, J.A.M.A. 1941, 116:200-204. 
181 Thorner, J. E.: Coccidioidomycosis; relative 
values of coccidioidin and tuberculin testing among 
children of San Joaquin Valley, California & West. 
Med. 1941, 54:12-15. 
185 Winn, W. A.; Coccidioidomycosis associated 
with pulmonary cavitation, Arch. Int. Med. 1941, 
68:1179-1214. 
z942 
188 Aronson, J. D., and Gallagher, J. R.; Sensi- 
tivity to coccidioidin (as explanation of pulmonary 
calcification among boys in Eastern preparatory 
school), Am. J. Pub. Health 1942, 32:636-639. 
187 Aronson, J. D., Saylor, R. M., and Parr, E. I.; 
Relationship of coccidioidomycosis to calcified pul- 
monary nodules (in Indian children with negative 
tuberculin reactions), Arch. Path. 1942, 34:31-48. 
188 Ashburn, L. L., and Emmons, C. W.: Spon- 
taneous coccidioidal granuloma in lungs of wild 
rodents, Arch. Path. 1942, 34:791-800. 
3S!'Baker, E. E., and Smith, C. E.: Utilization of 
carbon and nitrogen compounds by Cocciodioides 
immitis (Rixford and Gilchrist 1896), J. Infect. 
Dis. 1942, 70:51. 
100 Benninghoven, C. D., and Miller, E. R.: Coc- 
cidioidomycosis in bone, Radiol. 1942, 38:663-666. 
191 Caldwell, G. T.: Secondary (granulomatous) 
coccidioidomycosis; coccidioidal granuloma, Texas 
State J. M. 1942, 38:376-382. 
192 Carter, R. A.: Coccidioidomycosis, diagnosis of 
fungus infections of lungs, Radiol. 1942, 38:649- 
659. 
193 Davis, B. L., Jr., Smith, R. T., and Smith, C. E.: 
Epidemic coccidioidomycosis, J.A.M.A. 1942, 118: 
1182-1186. 
194 Emmons, C. W.: Coccidioidomycosis, Mycologia 
1942> 34:452- 
195 Emmons, C. W.: Isolation of coccidioides from 
soil and rodents, Pub. Health Rep. 1942, 57:109-111. 
198 Emmons, C. W., and Ashburn, L. L.; Isolation 
of Haplosporangium parvum, non-sporulating, and 
Coccidioides immitis from wild rodents; their rela- 
tionship to coccidioidomycosis, Pub. Health Rep. 
1942, 57:1715-1727. 
197 Kerley, P.: Significance of radiologic manifesta- 
tions (in chest) of erythema nodosum (relation to 
sarcoidosis and coccidioidomycosis), Brit. J. Radiol. 
1942, 15:155-165. 
198 Martin, C. L.; Coccidioidomycosis, roentgen- 
ray findings, Texas State J. M. 1942, 38:385-389. 
199 Peers, R. A., Holman, E., and Smith, C. E.: 
Coccidioidal pulmonary disease, Am. Rev. Tuberc. 
1942, 45:723-740. 
290 Rosenberg, E. F., Dockerty, M. B., and Meyer- 
ding, H. W.; Coccidioidal arthritis; report of case 
with ankle involvement and no results with sulfanil- 
1940 
170 Cronkite, A. E., and Lack, A. R.: Primary 
pulmonary coccidioidomycosis, experimental infection 
with Coccidioides immitis, J. Exper. Med. 1940, 
72 :i 67-174. 
171 Foley, M. P., et al: Coccidioidal granuloma; 
report of case originating in Texas, West. J. Surg. 
1940, 48:738-741. 
172 Henrici, A. T.: Characteristics of fungus dis- 
ease, J. Bacteriol. 1940, 39:113. 
173 Mills, C. W., and Farness, O. J.: Coccidioides 
Immitis infection in Southern Arizona, Tr. Am. 
Climat. & Clin. A. 1940, 56:147. 
174 Moore, M.: The diagnosis of the common 
fungus diseases of the lung, Am. Acad. Tuberc. Phys. 
1940, No. 10. 
170 Smith, C. E.: Epidemiology of acute coccidi- 
oidomycosis with erythema nodosum (San Joaquin 
or valley fever), Am. J. Pub. Health 1940, 30:600- 
611. 
178 Stewart, R. A., and Kimura, F.: Skin test for 
coccidioidal infection; preparation and standardiza- 
tion of coccidioidin, J. Infect. Dis. 1940, 66:212-217. 
177Yegian, D., and Kegel, P.: Coccidioides immitis 
infection of lung; case resembling chronic pulmonary 
tuberculosis, Am. Rev. Tuberc. 1940, 41:393~397. 
*94* 
178 Craig, W. M., and Dockerty, M. B.: Coccidi- 
oidal granuloma; brief review with report of case 
of meningeal involvement, Minnesota Med. 1941, 
24:150-154. 
179 Farness, O. J.: Coccidioidomycosis, J.A.M.A. 
1941, 116:1749-1752. 
180 Kessel, J. F.: Recent observations in coccidi- 
oidomycosis, Am. J. Trop. Med. 1941, 21:447-453. 
181 Powers, R. A., and Starks, D. J.: Acute (pri- 
mary) coccidioidomycosis; roentgen findings in 
group “epidemic,” Radiol. 1941, 37:448-453. 
1S" Smith, C. E., and Baker, E. E.: A summary of 
present status of coccidioidal infection, Weekly Bull. C occidioidomy costs 
719 
amide and roentgen therapy, Arch. Int. Med. 1942, 
69:238-250. 
201 Schenken, J. R., and Palik, E. E.: Coccidioido- 
mycosis in states other than California, with report 
of case in Louisiana, Arch. Path. 1942, 34:484-494. 
202 Schulze, V. E;: Acute coccidioidomycosis in 
western Texas, Texas State J. M., 1942, 38:372-376. 
203 Shelton, R. M.: Survey of coccidioidomycosis 
at Camp Roberts, California, J.A.M.A. 1942, 118; 
1186-1190. 
204 Smith, C. E.: Coccidioidomycosis; parallelism of 
tbc. infections, Radiol. 1942, 38:643-648. 
200 Smith, L. M.; Coccidioidal granuloma in Texas; 
report of 5 cases with dermatological pianifestations, 
Texas State J. M. 1942, 38:383-385. 
208 Stiles, G. W., and Davis, C. L.; Coccidioidal 
granuloma (coccidioidomycosis) ; incidence in man 
and animals; diagnosis in animals, J.A.M.A. 1942, 
119 :765~769. 
207 Tager, M., and Liebow, A. A.: Intranasal and 
intraperitoneal infection of mouse with Coccidioides 
immitis, Yale J. Biol. & Med. 1942, 15:41-59. 
208 Winn, W. A.: Treatment of pulmonary cavi- 
tation due to coccidioidal infection, California & 
West. Med. 1942, 57:45-47. 
209 Winn, W. A., and Johnson, G. H.: Primary 
coccidioidomycosis; roentgenographic study of 40 
cases, Ann. Int. Med. 1942, 17:407-422. 
*943 
210 Baker, E. E., Mrak, E. M., and Smith, C. E.: 
The morphology, taxonomy, and distribution of 
Coccidioides immitis (Rixford and Gilchrist 1896), 
Farlowia 1943, 1:199. 
211 Bush, J. D.: Coccidioidomycosis, J. M. A. Ala- 
bama 1943, 13:159-166. 
212 Emmons, C. W.: Coccidioidomycosis in wild 
rodents; method of determining extent of endemic 
areas, Pub. Health Rep. 1943, 58:1-5. 
213 Goldstein, D. M., and Louie, S.: Primary pul- 
monary coccidioidomycosis; epidemic of 75 cases, 
War Med. 1943,4:299-317. 
214 Hassid, W. J., Baker, E. E., and McCready, 
R. M.: Immunologically active polysaccharide pro- 
duced by Coccidioides immitis, Rixford and Gilchrist, 
J. Biol. Chem. 1943, 149:303-311. 
215 Lindsay, H. C. L.: Acute coccidioidomycosis 
(valley or desert fever), Urol. & Cutan. Rev, 1943, 
47:692-694. 
218 McDougall, T. G., and Kleinman, A. H.: 
Prostatitis due to Coccidioides immitis, J. Urol. 1943, 
49:472-477. 
247 Smith, C. E.: Coccidioidomycosis, M. Clin. 
North America, 1943, 27:790-807, 
*944 
218 Cheney, G., and Denenholz, E. J.: Diagnosis 
and treatment of chronic coccidioidomycosis, Arch. 
Int. Med. 1944, 74:311-330. 
219 Coccidioidomycosis, report of, study in the 
year 1943-1944 for the Commission on Epidemio- 
logical Survey, Board for the Investigation and 
Control of Influenza and Other Epidemic Diseases 
in the Army Preventive Medicine Division, Office of 
the Surgeon General, Service of Supply, War De- 
partment. 
Colburn, J. R.: Roentgenologic types of pul- 
monary lesions in primary coccidioidomycosis, Am. 
J. Roentgenol. 1944, 51:1-8. 
221 Culliner, N. W.: Clinical aspects of coccidioido- 
mycosis, Bull. N. E. Med. Center 1944, 6:238-242. 
"'Goldstein, D. M., and McDonald, J. B.: Pri- 
mary coccidioidomycosis, pulmonary (follow-up of 
75 cases with 10 more cases from new endemic area), 
J.A.M.A. 1944, 124:557-561. 
"3 Lee, R. V., Nixon, N., and Jamison, H. W.: 
Syllabus on coccidioidomycosis, Coccidioidomycosis 
Control Program for the AAF WFTC, 2nd ed. 1944 
(unpublished but issued by the Surgeon, AAF 
WFTC, Hq., Santa Ana, Calif.). 
"l Lee, R. V.; Coccidioidomycosis in Western Fly- 
ing Training Command, California & West. Med. 
1944, 61:133-134. 
Lindsay, H. C. L.: Coccidioidomycosis involving 
brain, history of case, Urol. & Cutan. Rev. 1944, 
48:275-278. 
"8 McKcnney, F. D., Traum, J., and Bonestell, 
A. E.: Acute coccidioidomycosis in mountain gorilla 
(Gorilla beringeri) with anatomic notes, J. Am. 
Vet. M. A. 1944, 104:136-140. 
"7 Marcuse, P., and Karotkin, L.; Review of coc- 
cidioidal granuloma in Texas case, Texas State J. 
M. 1944, 39=534-537- 
"8 Michael, P., McLaughlin, R. F., and Cenac, 
P. L.: Unsuccessful treatment of coccidioidomycosis 
with penicillin, U. S. Nav. M. Bull. 1944, 43:122- 
124. 
229 Quill, L. M., and Burch, J. C.: Surgical mani- 
festations of coccidioidomycosis, Ann. Surg. 1944, 
120 :67o-679. 
1945 
'30 Ashburn, L. L., and Emmons, C. W.; Experi- 
mental Haplosporangium infection (relation to coc- 
cidioidomycosis), Arch. Path. 1945, 39:3-8. 
231 Bass, H. E.: Coccidioidomycosis and tuberculosis, 
diagnostic problem, Tuberculology 1945, 7:72-75. 
"3‘Cheney, G., and Denenholz, E. J.; Coccldioidin 
skin test, Military Surgeon 1945, 96:148-156. 
233 Christie, A., and Peterson, J, C.: Pulmonary 
calcification in negative reactors to tuberculin, Am. 
J. Pub. Health 1945, 35:1131. 
*34 Kunstadter, R. H., and Pendergrass, R. C.: 
Primary coccidioidomycosis; possible pediatric prob- 
lem, J.A.M.A. 1945, 127:624-627. 
235 McCracken, J. P.: Coccidioidomycosis, North 
Carolina M. J. 1945, 6:25-31. 
238 Schlumberger, H. G.: A fatal case of cerebral 
coccidioidomycosis with cultural studies, Am, J. M. 
Sc. 1945, 209:483. Colonel James Earle Ash, An Appreciation 
Col. James M. Phalen, U. S. Army, Ret. 
The Army Institute of Pathology during World War II 
Col. Balduin Lucke, M.C., A.V.S. 
The American Registry of Pathology and Its Relation to the Army 
Institute of Pathology 
Howard T. Karsner, M.D. 
The Registry of Dental and Oral Pathology at the Army 
Institute of Pathology 
Henry A. Swanson, D.D.S. 
Lx. Col. Jos. L. Bernier, D.C., U. S. Army 
Lower Nephron Nephrosis 
Col. Balduin Lucke, M.C., A.U.S. 
Heat Stroke. A Clinico-pathologic Study of One Hundred 
Twenty-five Fatal Cases 
Nathan Malamud, M.D., Major Webb Haymaker and Lt. Col. Richard P. Custer, M.C., A.US. 
Fatal Pulmonary Embolism in One Hundred Battle Casualties 
Capt. Tom R. Hamilton and Col. D. Murray Angevine, M.C., A.U.S. 
Intra-ocular Tumors in Soldiers * 
Helenor Campbell Wilder 
The Occurrence and Anatomic Characteristics of Fatal Tuberculosis 
in the U. S. Army during World War II 
Col. Joseph D. Aronson, M.C., A.U.S. 
Fibrous Dysplasia of Single Bones (Monostotic Fibrous Dysplasia) 
Major Hans G. Schlumberger, M.C., A.U.S. 
Bronchial Adenomas 
Major Sion W. Holley, M.C., A.U.S. 
The Pathology of Acute Falciparum Malaria 
Contract Surgeon Sophie Spitz, V. S. Army 
Tumors of the Testis: Report on Nine Hundred Twenty-two Cases 
Major Nathan B. Friedman, M.C., A.U.S., and Robert A. Moore, M.D. 
Primary Intracranial Neoplasms in Military Age Group—World War II 
Major Warren A. Bennett, M.C., A.U.S. 
Coccidioidomycosis: A Study of Ninety-five Cases of the Disseminated Type 
with Special Reference to the Pathogenesis of the Disease 
Wiley D. Forbus, M.D., and Annie D. Bestebreurtje, M.D. 
Reprinted from The Military Surgeon, Vol. 99, Nov. 1946.