Part III
DISEASES TRANSMITTED CHIEFLY THROUGH ALIMENTARY TRACT
CHAPTER XVII
Diarrhea and Dysentery
Frank R. Philbrook, M. D., and John E. Gordon, M. D.
ARMY EXPERIENCE WITH DIARRHEAL DISORDERS BEFORE WORLD WAR II
Montgomery says the Eighth Army won, but Rommel claimed the victory for dysentery * * *. But, as the Germans learned at El Alamein, dysentery can still win battles, when hygiene discipline on one side is slack.-SIR SHELDON F. DUDLEY.
General Considerations
Throughout history, military populations have experienced great morbidity by reason of the diarrheas and dysenteries. As causes of illness and norneffectiveness, these diseases have plagued the United States Army since it was first organized. Until the time of World War I, they were important causes of mortality.
Diarrheal disease was rampant among troops during the Revolution and apparently was responsible for more deaths than were caused by enemy action. Available figures for the Civil War indicate the military significance of this disease in combat troops; diarrheas and dysenteries "occurred with more frequency and produced more sickness and mortality than any other form of disease." More than 1,755,889 patients were admitted to field medical installations and hospitals. For diarrheas and dysenteries, including gastroenteritis, the case rate was 741.2 per annum per 1,000 average strength.Deaths were 46,277, representing a mortality rate of 18 per 1,000; 2.6 percent of the men admitted for these causes died. During the Civil War, more than 1 death in every 4 caused by disease was ascribed to diarrheas and dysenteries. In the Union Army, of every 1,000 men the following numbers succumbed to dysentery or diarrhea during each year of the war; Negro troops fared worse than white troops:1
Year | White troops | Negro troops |
1861-62 | 4.17 | 0 |
1862-63 | 15.99 | 0 |
1863-64 | 15.78 | 43.54 |
1864-65 | 21.29 | 36.29 |
1865-66 | 16.00 | 26.97 |
1 Medical and Surgical History of the War of the Rebellion. Medical History. Washington: Government Printing Office, 1879, vol. 1, pt. II, pp. 1, 6, passim.
320
The dysenteric diseases are at their worst in prisoner-of-war camps.The stockade for Union prisoners of war at Andersonville, Ga., is an outstanding historical example, to such extent that it became known as the most fatal field in the War Between the States. At least 16,772 cases of diarrhea and dysentery occurred at this prison and represented an estimated annual rate of 1,724 per 1,000 strength; 4,529 men died. More than one-half of all fatalities were attributed to diarrhea and dysentery.2
During the Spanish-American War, at the turn of the 20th century, the incidence of diarrheas and dysenteries among United States Army troops was slightly lower than that among troops during the Revolutionary War and the Civil War. However, typhoid fever was distinguished as a disease problem because of a mortality far exceeding the more frequent but less deadly related dysenteric illnesses. Indeed, diarrhea and dysentery seemed in this war to have lost much of their killing power. During the war years of 1898-1901, 204,040 admissions were, reported representing a rate of 426 admissions per 1,000 troops. The number of deaths was 1,595, a rate of 3.3 per 1,000. Case fatality dropped to 0.78 percent in this period. Diarrheal disease occurred in severe epidemic form in United States Army troops in the Philippines shortly after the American occupation in 1899-1900. Much of the excess incidence was doubtless due to bacillary dysentery, but typhoid fever was the more serious problem in terms of resulting deaths. At the time of the Spanish- American War, attention was focused upon typhoid fever with a resultant improvement in diagnosis of this enteric condition. Examination of data of this and other wars suggests that many, fatal intestinal infections of earlier wars were in reality typhoid fever, erroneously diagnosed as infections within the group of diarrheas and dysenteries (including gastroenteritis). Comparison of experience in World War II with that of former wars must take this into account.
During World War I, the Army benefited from the great improvements in environmental sanitation of the preceding two decades. During the period from April 1917 to December 1919, there occurred in the overseas Army 48,202 admissions for diarrheal disease, including gastroenteritis, a rate of 28.9 admissions per 1,000 troops. There were only 208 deaths (0.13 deaths per 1,000 troops) which represents the best record up to that time in any American war. Case fatality was 0.43 percent. The improvement in respect to typhoid fever was even more startling (table 82).This favorable downward trend of cases and deaths of typhoid fever continued throughout World War II, but the rate of 40 cases of reported dysenteries and diarrheas (excluding gastroenteritis) per 1,000 strength in total troops overseas certainly shows no appreciable improvement over World War I experience. However, a comparison of the Army in continental United States and in the European Theater
2 Medicaland SurgicalHistory of the War of the Rebellion. Medical History. Washington:Government PrintingOffice, 1885, vol. 1, pt. 111, p. 35.
321
of Operations, with troops of World War I, is more justifiable andreveals that deaths fromdiarrheas and dysenteries followed a downward trend in World War II,reasonably reflecting theimproved sanitation of America and Europe. During World War II, ascontrasted with WorldWar I, a greater proportion of combat troops were sent to tropical orsemitropical regions withHigh endemicity for dysenteric diseases and a low standard ofsanitation. The rate of 18 casesper 1,000 strength for the Zone of Interior in World War I is to becompared with 9 cases per1,000 for World War II (table 54). The enteric disease rate amongoverseas troops in World WarI, approximately 29 cases per 1,000 troops per annum, may properly becompared with the rateamong troops in the European theater in World War 11, 14 cases per1,000 troops per annum. The rates for World War II are about one-halfthose of World War I. By the same standards, thecase rates for typhoid fever dropped in World War If to less thanone-tenth of the rates for thisdisease in World War I for the total Army, and to approximatelyone-twentieth of the World WarI rates for this disease among United States Army troops in Europe.Typhoid case rates amongthe United States Army at home became an almost insignificant 0.006case per 1,000 troops perannum. Marked improvement also occurred in respect to the paratyphoidfevers.
During World War II, deaths ascribed to diarrheas and dysenteries (including also gastroenteritis, ileitis, and ulcerative colitis) totaled 116. Using the 523,331 cases reported on WD MD Form 86ab, Statistical Health Report, (not including cases diagnosed as gastroenteritis) the deaths represent a case fatality of 0.022 percent. The death rate of 0.005 per 1,000 strength is considerably less than the rate in World War I.
In comparing current experience with past Army history of these diseases, due consideration must be given to the fact that earlier data often were based on hospital admissions alone, whereas World War II morbidity statistics include all patients, whether sick in quarters or admitted to hospital. Although reporting is still far from complete, the general level of reporting is better than in former wars. World War II data for common diarrheas are considerably increased by these factors, while the more severe illnesses such as the dysenteries (when diagnosed as such) and typhoid fever are probably little affected, since most patients with these diseases would have been admitted to hospital even in former wars. As a result, more mild cases of diarrheal disease probably have crept into statistical records of recent wars as compared with earlier wars. This factor must be considered in evaluating both incidence of disease and effectiveness of medical care. With the reporting of a larger number of mild cases and better medical care, case fatality rates fell to almost insignificant levels, despite the high morbidity reported in several theaters of operations in World War II. Nevertheless, control of the diarrhea's and dysenteries has not kept pace with that accomplished for typhoid fever.
322
323
324
Classification
For the purposes of this analysis, the diarrheas and dysenteries are classified as common diarrheas and dysenteries (protozoal, bacillary, and unclassified) according to the diagnostic. titles set forth in statistical health reports.
The common diarrheas (as defined in Army Regulations No. 40-1080) include a variety of infections of the gastrointestinal tract of unknown cause, such as colitis, enterocolitis, enteritis, diarrhea, mucous colitis, intestinal indigestion, sigmoiditis, and intestinal toxemia (when associated with diarrhea). Grouping these infections under this single inclusive title has an advantage over the practice of former years when each clinical entity had individual consideration irrespective of admittedly inexact anatomic or symptomatic criteria for diagnosis. All these infections are now grouped within the single category of the common diarrheas. Too little is known about these several conditions, and the criteria for differentiation are too inexact to warrant continued separation in the course of attempts to describe the diarrheal disorders of an army or another population.
Diagnostic and microbiologic interrelationship.-One or another of these terms, and also that of gastroenteritis, is often used provisionally for infections suspected to be of specific bacterial origin. Bacteriologic diagnosis ultimately determines a number of such infections as being bacillary dysentery, amebic dysentery, or instances of salmonellosis. As for the remainder, much of what is now grouped under the common diarrheas is in all probability dysentery of unknown etiology, unknown because a search for an etiologic agent either was not made or was reported negative.
The proportion of common diarrheas actually representative of protozoal or bacillary dysentery can only be estimated from limited data in a few theaters where laboratory surveys were made. The important consideration is that no practical analysis of dysenteric disease nor specific consideration of bacillary dysentery is either reasonable or feasible unless the common diarrheas are included. This opinion is held despite recognition that diagnostic titles included in the common diarrheas do not in their entirety represent enteric infections or intoxications due to known specific microbial agents. Through unofficial statements of former Army medical officers, it is known, for example, that acute alcoholism was recorded occasionally under one or another of the diagnostic titles included in the common diarrheas, or as gastroenteritis, because of the stigma attached to the straightforward diagnosis. The important fact remains that, when careful laboratory studies were made, a major part of the common diarrheas as seen in World War II could be demonstrated as due to Shigella, Salmonella, or Endamoeba histolytica. Many epidemic peaks of common diarrhea coincide in time with recognized outbreaks of shigellosis.
Clinical interrelationship.-The entities included in the mixed group of anatomic and symptomatic diagnostic titles, known collectively as the common diarrheas, seemingly have much in common in their clinical manifestations with
325
diseases distinguished etiologically as either bacillary or amebicdysentery. Diarrhea is thetypical symptom, often of sudden onset: and brief duration, and isaccompanied by varyingdegrees of tenesmus and abdominal pain and followed by rapid recovery.The common diarrheasare characterized by an extremely low fatality. The few deathsattributed to common diarrhea arepresumably the result of failure to recognize a more serious infectionsuch as bacillary or amebicdysentery. Specific etiologic agents are by definition undetermined.
Epidemiologic interrelationships.-Epidemiologically, the behavior of common diarrhea is suggestive of an infection, and transmission in the great, preponderance of cases is by the same general meats, through ingestion of food or drink. The occurrence of most of diem reflects directly or indirectly the general level of environmental sanitation under which different army units existed or of the countries or specific locations in which troops were deployed. Occurrence in close correlation with known specific, infectious diarrheal diseases suggests similar epidemiologic mechanisms. Many epidemic peaks coincide time with recognized outbreaks of shigellosis. Incidence was high in theaters of operations when the environment was poorly sanitated and bacillary and amebic dysenteries were common diseases. Incidence was low in theaters or environmental situations where known specific agents of diarrheal diseases were infrequent. The rigid application of sanitary principles, especially in relation to food and drink, to feces disposal, and to fly control reduced the risk of the common diarrheas in equal degree as it lowered the incidence of the dysenteries.
Titles not included.-Some 10 other diagnostic titles or diseases are variously included in classifications of diarrheal disease, none of which are represented in this analysis. Brief consideration is given to possible effects of their exclusion on comparisons of the present data with diarrheal disease in former wars and in some instances on comparisons between theaters of World War II.
The diagnostic title of gastroenteritis is not included among diarrheas and dysenteries as presented in table 54. Statistics of former wars often included gastroenteritis, an addition materially affecting comparisons. Separate, data on the incidence of gastroenteritis are riot available for the first 2 years of World War II, but it is estimated that approximately 350,000 cases occurred during the 4 war years. If these cases had been included among diarrheas and dysenteries of World War II, the total would be more than 850,000, and incidence would exceed 34 per 1,000 per annum, a rate somewhat greater than the 29 for the United States Army in Europe during World War I.
Typhoid and paratyphoid fevers and bacterial food poisoning if they were reported as such are not here included. They are considered elsewhere in this volume. Army units were instructed in October 1943 to record bacterial food poisoning as a separate item, and the diagnostic title was added to the Statistical Health Report (WD) MD Form 86ab of 22 August 1940) when the report was revised in 1944. Before October 1943, bacterial food poisoning had been included under common diarrhea, and undoubtedly a considerable pro-
326
portion of cases continued to be so reported after this date.3 The extent to which the new system of reporting was followed in the several theaters of operations and the promptness with which it was adopted were doubtless subject to much variation. Better results were undoubtedly obtained in areas where epidemiologic investigation and laboratory diagnosis were readily available services. Current statistical information on typhoid and paratyphoid fevers can be accepted as reliable, since diagnosis depends almost universally upon demonstration of the etiologic agent. That this reliability factor was not always present must be taken into consideration when comparing incidence (of either these specific diseases or the diarrheas and dysenteries) in World War II with that of earlier wars.
Also excluded from specific and separate consideration here are such entities as diarrhea and dysentery of presumed viral etiology such as epidemic viral gastroenteritis or acute infectious gastroenteritis or winter vomiting disease or vomiting and diarrhea syndrome, as described by numerous authors during and since World War II.
There conditions doubtless played an important role in the World War II Army and will in the Army of the future. As more becomes known about them, they may become subject to separate statistical analysis, but during World War II their recognition and reporting was not specifically accomplished. Viral etiology was suspected in several theaters, and some attempts that were made to recover these agents are described in this chapter. Such entities during World War II were merely given one or another of the common diarrhea diagnoses or were considered to be cases of gastroenteritis or unclassified dysentery.
Factors Affecting Reporting ofSpecific Dysenteries
By regulation, the diagnosis of bacillary and amebic dysentery during World War II was contingent on recovery or identification of the bacillus or ameba, a practice that did not hold in former wars. Instances of clinical dysentery associated with pus, mucus, and possibly blood in the stools were designated "dysentery, unclassified," in the absence of microbiologic confirmation. In general, the frequency of etiologic diagnosis of bacillary or protozoal dysentery was a reflection of the amount and quality of laboratory work.
The techniques necessary for identification of E. histolytica are relatively simple, compared with those of determinative bacteriology. As a consequence, the search for protozoal agents of diarrheal disease was usually more exhaustive than that for bacillary agents and probably accounted for the greater frequency with which amebic dysentery was recognized. A significant proportion of many populations are carriers of E. histolytica, estimated by Mackie
3 Gordon, J. E.: AHistory of Preventive Medicine in the European Theater of Operations,U. S. Army, 1941-45.[Official record.]
327
and others 4 as 9.8 percent for theUnited States andin excess of 50 percent for poorly sanitatedregions of the Tropics and subtropics. An even larger number arecarriers of protozoa closelyresembling E. histolytica. The diagnosis of amebicdysentery was in all probability frequentlymade when the true infectious agent was an unsought Shigella. Personswithout clinicalsymptoms are known to have been treated with a view to eliminating the Endamoeba,and somewere given the diagnosis of amebic dysentery. For these reasons, theincorrect diagnosis ofcommon diarrhea was made less frequently for cases of amebic dysenterythan for cases ofbacillary dysentery. Small numbers of patients with common diarrhealdiseases and shigellosiswere in all likelihood erroneously designated as suffering from amebicdysentery. The tendencywas to over report amebic dysentery and to under report bacillaryinfections.
The biologic gradient of clinical bacillary dysentery ranges from severe infections (in the diagnosis of which the physician is prompted to seek laboratory aid, or which he may diagnose as unclassified dysentery, enteritis, gastroenteritis, or enterocolitis) to instances of lesser severity, with symptoms largely absent or attended by no more than mild diarrhea. The tendency is to give the diagnosis of common diarrhea to these mild infections or, because of vague symptomatology, to call them gastroenteritis, enteritis, or colitis. For reasons already presented, this probably applies to amebic dysentery in lesser degree. Other features presumably account for varying frequencies of confirmed dysentery.
Adequate laboratory facilities and competent personnel were not uniformly available in all theaters of operations. Furthermore, the interests of medical officers varied. A specific concern about amebiasis by medical officers in one location was followed by a relatively high incidence of that type of infection. In another theater, an epidemiologic team had a specific interest in shigellosis, with the result that a relatively high incidence of bacillary dysentery was uncovered. Each of these circumstances led to the separation of a higher percentage of the entity on which interest was focused from the common diarrheas, gastroenteritis, and unclassified dysentery classifications. For example, one factor accounting for the reported high incidence of amebic dysentery in a part of the Mediterranean theater was said to have been the result of interest of certain medical officers in amebiasis.5 The arrival of several groups studying diarrheal disease and the development of special health record forms for amebiasis in the China-Burma-India theater doubtless stimulated search for cases and carriers.
The true incidence of bacillary dysentery represents a sizeable proportion of the common diarrheas, of unclassified dysentery, and of gastroenteritis together with those cases reported as bacillary dysenteries; hence the practicality and feasibility of presenting combined information for diarrhea and
4 Mackie,Thomas T.,Hunter, George W., III. and Worth, C. Brooke: Manual of TropicalMedicine. Philadelphia:W. B. Saunders Co., 1945, p. 191.
5 Gilmore, H. R.: Final Report of the Preventive Medicine Officer, Office of the Surgeon, Mediterranean Theater of Operations, U. S. Army, 1 January-31 October 1945. [Official record.]
328
dysentery as a class or group. Justification for this procedure isfound in the situation in theMediterranean theater in 1943. The annual rate for diarrheas anddysenteries was 132 cases per1,000 troops (table 54). The rate for bacillary dysentery was only 6.22cases (table 59), despitethe presence that year of a theater wide epidemic ascribed to bacillarydysentery. Surveysrevealed that Shigella could be recovered from a highpercentage of the cases of commondiarrhea in the Fifth U. S. Army.6 During the same year,three other theaters with no recognizedgeneral epidemic of bacillary dysentery actually reported higher ratesfor that disease.Two hadlaboratories especially active in recovering dysentery bacilli fromsuspected cases. In the China-Burma India theater, several hundredconsecutive admissions for diarrhea were examined in ageneral hospital in Assam and in two dispensaries. Dysentery bacilliwere demonstrated byculture of material obtained by a single rectal swab in 24 percent ofthe hospital cases acid in 16percent of the dispensary infections. Multiple cultures doubtless wouldhave increased thosepercentages. "As would be expected, the number of positive findings inthe series of dispensarycases is less than in the hospital series. These cases were mild, andhad cultures riot been madethey would have beers classed as common diarrheas." 7
Because of the differences in diagnostic practice just described and because facilities for laboratory work were so uneven, the various clinical entities among intestinal infections cannot be compared individually with what happened in former wars nor, in some instances, can any reasonable comparison be made between theaters of operations of World War II.
Method, of Presentation of Data
The most useful approach appears to be a study of diarrheas and dysenteries as a single group of diseases and as they occurred in the several theaters of operations. In the following presentation of analyses, the practical purpose of the study-to provide a basis for future experience in the event that United States troops should again be deployed in the areas involved is kept constantly in view. Bacillary dysentery is given special consideration in this presentation: other specific; diarrheal diseases are presented in separate chapters. Individual attention is given the dysenteries so far as the data permit. Proved cases of bacillary dysentery are a part of this report; amebic dysentery, although given some general consideration here, is the subject of a separate chapter.
Grouping the diarrheas and dysenteries in a combined statistical presentation is judged a practical and feasible approach to the descriptive epidemiology of dysenteric disease. A compilation of the Medical Statistics Division, Office of the Surgeon General, entitled "Morbidity and Mortality in the United States Army, 1940-45," is the source of most of the data now presented (table 54 and charts 28, 30, 35, 39, 40, 42, 45, 46, and 47). These statistics enumerate the diarrheas and dysenteries as they were diagnosed and reported currently and
6 FifthArmy MedicalService History, 1944.
7 Progressand FinalReport, Sub-Commission on Dysentery. Army Epidemiological Board, 20Nov. 1944.
329
regularly on the statistical health report. They may not alwaysreflect the changes in diagnosiswhich occurred during the course of hospitalization. Such datatherefore do not tally identicallywith those derived from statistical tabulations of individual medicalrecords. They do provide anexcellent basis for epidemiologic description of the picture as itunfolded in the several theatersof operations during World War II. For such separate consideration aswill be given to individualdiagnostic entities comprising the group, especially bacillarydysentery (table 59), the majorsource of statistical data is various tabulations of individual medicalrecords as compiled by theOffice of the Surgeon General. In general, admissions data are used forthe common diarrheasand incidence rates for the specific and unclassified dysenteries.
The epidemiologic description of the diarrheas and dysenteries in each major theater of operations or area will be followed by a consideration of available information on bacillary dysentery.
Experience From 1935 to 1941
During the years from 1935 to 1939 inclusive, Army troops stationed within the United States were a relatively stable population, living under sanitary conditions reasonably comparable to those of better sanitated civilian communities. The incidence of diarrheal disease was low, and rates were remarkably uniform from year to year (table 55). For the decade of the 1930's, incidence rates were within the range of 5 or fewer cases per 1,000 men as illustrated in chart 27. During the latter half of the, period, annual
330
incidence varied from 2.5 to 3.6 cases per 1,000 per annum and theaverage was 3 cases.Thehighest rates were in summer, August being the peak month; winter rateswere lowest, with ausual seasonal increase in late spring. Diarrheal disease presented afavorable and stabilizedsituation.
CHART27.-Incidenceof diarrheaand dysentery in the Army in the continental United States, 1930-42
The incidence of diarrheal disease in troops in the continental United States rose rather sharply to 13 cases per 1,000 per annum for the 2 years 1940 and 1941 (chart 27). These were the years of preparation for war, with more than fourfold expansion of the Army. Raw recruits and untrained men were brought together from all parts of the Nation, and they engaged in active field training and maneuvers. In 1940 and 1941, diarrheal disease rates were 7 cases and 15 cases, respectively, per 1,000 troops (table 54). For the same years, rates for the total Army (both overseas and in continental United States) were also 7 and 15, which shows that the bulk of the problem was in the Zone of Interior rather than overseas. Overseas troops had slightly higher rates of 9 and 17, but less than 12 percent of the average strength of this expanding Army was stationed overseas, chiefly in the Philippines with rates of 27 and 28 cases, respectively, per 1,000 per annum and in the Antilles Department (Latin American area) with rates of 34 and 26 cases, respectively, per 1,000 per annum. The well-established program of sanitation and military preventive medicine in the Panama Canal Department proved its worth, for during these 2 years the rate was 3 cases per 1,000 per annum.
That field training contributed heavily to the incidence of diarrheal disorders in the United States is evidenced by the North Carolina maneuvers of August and September 1941. The entire Army had 8,890 cases in those 2 months; a total of 8,297 cases were in the continental United States; a large proportion was related to an outbreak of bacillary dysentery in the course of these maneuvers. Continental rates for these 2 months were 35 and 37 per 1,000 per annum (chart 28), rates higher by far than for any month in the European theater or for the Panama Canal Zone, the North American area, or the Alaskan Department at any time in World War II. However, these
331
monthly rates were exceeded at onetime or another in all tropical or subtropical areas of militaryoperations, with the single exception of the Panama Canal Zone.
CHART28.-Incidenceof diarrheaand dysentery in the U. S. Army, 1940-45
