Chapter 16
Bacterial Diarrheal Diseases
Layne O. Gentry, M.D., Colonel Kenneth W Hedlund, MC, USA, Colonel Ralph F. Wells, MC, USA (Ret.),
and Brigadier General Andre J. Ognibene, MC USA
Diarrheal diseases have always been of major military importance because of their severe, albeit short-lived, debilitating effects and the speed with which epidemics may occur. Gordon (1965, p.78) provides the following definition of acute diarrheal disease:
Acute diarrheal disease is a clinical syndrome of varied etiology, in large part infective, and having diarrhea and often fever as the common manifestations. It includes specific infectious diseases such as shigellosis, salmonellosis and amebiasis and also infections with enteropathogenic Escherichia coli, enteroviruses, protozoa and helminths. A variety of infectious agents of low pathogenicity, when present in large numbers, also may be etiologically associated. More often than not, no definable infectious agent can be identified. Looseness of the bowels is also due to other than microbial or parasitic causes.
The diarrheas may be classified on the basis of specific etiology or on the basis of their epidemiological pattern. Gordon suggested five epidemiological categories. The more stereotyped classification based on specific etiology will be the system used in this review.
Section I. Shigellosis
Layne O. Gentry, M.D.
HISTORY AND MILITARY SIGNIFICANCE
Historically, shigellosis has had a major impact on confined populations. Numerous epidemics have occurred in homes for the mentally ill, in orphanages, aboard ships, and in prisoner-of-war camps. Reports of severe dysentery among British prisoners of war arriving in Italy from North Africa in 1942 describe classical shigellosis which resulted in many deaths (Bloom 1944). Personnel on naval vessels during World War II suffered from repeated outbreaks of shigellosis. In one outbreak at Pearl Harbor, 9.3 percent of a ship`s company of 905 men were infected, a morbidity rate which made the vessel unfit for sea duty (Philbrook et al. 1948). Reports from the United Nations prisoner-of-war camps in Korea revealed that shigellosis had infected 8 percent of 1,000 prisoners
356
despite specific preventive measures; ineffectiveness of these measures was caused in part by the high endemic level of infection at the time of capture (Garfinkel etal. 1953).
Shigellosis has become such a common problem in the United States that the CDC (Center for Disease Control) has added it to the list of infectious diseases currently under surveillance. The magnitude of the problem is indicated by the number of isolations of shigellae reported by the CDC, which totaled 13,752 in 1972. Of these, 68.3 percent were obtained from children under 10 years of age and 90.9 percent from patients under 30 years of age. Of all Shigella isolates in the United States, 13.2 percent were obtained from either Indian reservations or mental institutions, none of which was involved in an epidemic or outbreak during this period. There was no predominance by sex (CDC-73).
EPIDEMIOLOGY
Shigellosis proved to be the most common enteric infection identified among U.S. personnel in Vietnam. The peak incidence months for diarrhea were April, May, and June, reflecting seasonal increases in shigellosis (USARV-CHR). Shigellosis occurred more often and was more widespread than parasitic infestations, with the possible exception of amebiasis. A comparison of two epidemics substantiates this point. In March 1966, in a single outbreak from a 1st Cavalry Division unit (Vandevelde 1966), 39 men were admitted to the hospital because of severe hookworm gastroenteritis; symptoms included diarrhea in 91 percent, epigastric discomfort in 82 percent, frank abdominal tenderness in 17 percent, and vomiting in 22 percent. In contrast, in April 1968 (USARV-CHR Apr. 68, pp. 3-4), an outbreak of shigellosis occurred, also in the 1st Cavalry Division, involving an estimated 7,000 individuals, including personnel in units attached to the division. Most individuals improved with supportive therapy in 24 to 48 hours, and only 11 cases were hospitalized.
The shigellosis epidemic was described as follows. On 16 April a temporary breakdown had occurred in the chlorinating machines at the water point in the Quang Tri area. On 18 April, an outbreak of gastroenteritis occurred in the 1st Brigade, 1st Cavalry Division, which received its water from this source. Members of the 1st Brigade moved to an area southwest of Camp Evans between 20 and 22 April. Individuals from the unit swam and bathed upstream from the Camp Evans water point, which was not being operated as prescribed. Another outbreak of gastroenteritis began on 23 April, was full blown on the 24th, and peaked between the 24th and 26th, after which the epidemic abated. Shigella flexneri was cultured from stool samples and from the raw water in the stream. By the time cultures could be obtained, water was again being adequately chlorinated.
During 1972, in the United States, S. sonnei was responsible for 79.2 percent of all cases of shigellosis reported, and almost all of the remaining cases were caused by the subtypes of S. flexneri (CDC-73, p.1). A regional epidemic of
357
shigellosis occurred in 1969 and 1970 in Central America and Mexico and was caused by S. dysenteriae subtype 1, which had essentially disappeared from the world by 1920. This epidemic was responsible for fatality rates of 7 to 8 percent and involved thousands of people (Mats et al. 1970; Gangarosa et al. 1970).
ETIOLOGY AND PATHOGENESIS
Shigellosis, or bacillary dysentery, as it was termed by the Japanese bacteriologist Shiga, was first described in the 4th century B.C. Table 63 outlines the Shigella classification by species, serological subgroup, and serological subtype. ???
Shigella was the first group of bacteria to demonstrate the extraordinary genetic instability of intestinal coliform bacteria; in Japan during the 1960`s, the genus was shown to contain a plasmid which carries genes capable of transfer ring multiple antibiotic resistance to other intestinal coliform bacteria (Mitsuhashi 1969). This plasmid, termed "R factor," has had worldwide impact, the most recent complication being the emergence of a markedly drug-resistant shigella in the United States and Mexico (NCDC-ShS).
All known species of the genus Shigella are pathogenic for man. Investigations in Russia, Japan, and the United States have established essential features in the pathogenesis of S. flexneri shigellosis (Labrec et al. 1964; Ogawa et al. 1967; Voino-Iasenetskii and Khavkin 1964). Pathogenicity is caused by penetration of colonic mucosal epithelial cells with subsequent intracellular multiplication of bacteria. This gives rise to mucosal ulcerations and to the accumulation of polymorphonuclear leukocytes and fibrin accompanied by unformed stools containing blood and mucus.
More recently, it has been shown that S. dysenteriae, unique among members of the dysentery group because it produces potent neurotoxin and/or enterotoxin, must maintain the ability to penetrate and multiply in the colonic mucosa to produce disease (Gemski et al. 1972). Loss of the ability to penetrate and multiply within cells, despite the capability of producing toxin, resulted in failure to cause classic shigellosis in animals. In volunteer studies (Levine et al. 1973), virulent strains of S. dysenteriae produced disease in doses as low as 101 organisms. This capability undoubtedly influenced the rapid epidemic spread of shigellosis seen during the Central American outbreak.
CLINICAL FEATURES, COURSE, AND COMPLICATIONS
Shigella dysentery is characterized by the sudden onset of abdominal cramps, diarrhea, and fever after approximately 1 to 4 days of incubation. Stools are characterized by mucus containing inflammatory cells and blood. It is rare to see systemic spread of shigellosis; however, bacteremia, osteomyelitis, pneumonia, and meningitis have been described. Infrequently, a persistent colitis which mimics ulcerative colitis proctoscopically and radiographically may be
358
TABLE 63.-Shigella classification
seen. Three such cases studied sequentially in Vietnam showed complete resolution.* None of the patients had had prior inflammatory bowel disease. Fluid and electrolyte loss may cause metabolic disturbances, especially in young children and infants. Specific agglutinins may appear in serum during convalescence; local mucosal antibody of the IgA type is also produced but its role in recovery is not yet clearly defined.
Because persons residing in an area where shigellosis of a specific type is endemic rarely suffer from repeated bouts of disease, there appears to be naturally acquired immunity, although this immunity may be strain specific. Recent investigation attests to this fact; it also clearly demonstrates that reinfection with a known virulent strain, in large doses, produces dysentery. Investigators were able to effectively immunize volunteers against severe dysentery by giving a live attenuated strain of S flexneri by mouth after neutralizing stomach acid with bicarbonate. Most of these volunteers who received the attenuated vaccine had elevated humoral antibody titers (DuPont et al. 1972a; 1972b). No local mucosal IgA antibody studies were done in this group. It is important to note that the attenuated strain used had lost its ability to penetrate intestinal mucosal epithelial cells.
Laboratory Diagnosis
Because Shigella species do not remain viable for long in excreted fecal material despite often being present in large numbers during active disease, it is mandatory to obtain specimens in appropriate holding media or to culture them properly as rapidly as possible (Thomas 1969). Preferably, specimens for culture are obtained either directly from mucosal lesion at proctoscopy or by a rectal swab. Methylene blue or Gram`s stains of specimens under direct vision are most helpful because excessive inflammatory exudate and large numbers of gram negative bacteria strongly suggest bacterial etiology rather than amebic dysentery, which is often mistaken for shigellosis.
Rising humoral antibody titers are often helpful, but the illness usually abates before laboratory test results become available. Another problem with the tests is that they require specific antigens of the Shigella subtypes which are often unavailable in hospital laboratories.
______
1 Col. Ralph F. Wells, MC: Personal communication.
359
Treatment
Because many of the infections caused by Shigella species are acute but self limiting, antibiotic therapy has not been standard or routine. However, severely ill patients obviously require drug therapy, as well as supportive fluid And electrolyte repletion. In general, Shigella species are sensitive to ampicillin, tetracycline, chloramphenicol, kanamycin, and gentamicin. Ampicillin has been considered the drug of choice.
Resistant strains may emerge during therapy; in the recent Central American and Mexican epidemic, the strain of S. dysenteriae was shown to be multiply drug-resistant but fortunately did respond to ampicillin (Mata et al. 1970, p. 175). As early as May 1967, Heggers and Smith (1967) detected a potential problem with tetracycline-resistant shigellae. This was confirmed in a study by Stone (1968) at the 3d Field Hospital, Saigon; he advocated the use of neomycin. Martin and associates (1970) later found widespread resistance to various antibiotics tested but recommended ampicillin, neomycin, or kanamycin therapy (table 64). A study from the 3d Medical Battalion, 3d Marine Division (Chow et al. 1971), recommended ampicillin as the antibiotic of choice, a selection in keeping with the author`s experience.
Because of the ability of intestinal bacteria to transfer drug resistance upon exposure to antibiotics, there has been a trend toward more restrained use of these drugs unless the systemic toxicity suggesting severe shigellosis is present. The USARV (U.S. Army Vietnam) medical consultant in 1969, Lt. Col. (later Brig. Gen.) Andre J. Ognibene, MC, strongly favored only intravenous fluid therapy at division level, with evacuation of nonresponders to hospitals where culture and additional therapeutic effort could be coordinated.* Indiscriminate antibiotic use was believed to produce resistance or delay recover in some cases.
TABLE 64.-Antibiotic resistance of 505 Shigella strains, Vietnam, 1968-69
______
1 Lt. Col. Andre J. Ognibene, MC, USARV Medical Consultant, 1969: Personal communication.
360
More recently, the use of drugs such as opiates and diphenoxylate hydrochloride with atropine (Lomotil), which retard intestinal motility, has been shown to effectively reduce the duration of the diarrhea but at the expense of prolonging the febrile response (DuPont and Hornick 1973a).
Prevention
Man-to-man spread is the primary source of transmission of shigellosis. Sanitation measures involving "food, feces, flies, and fingers" are most important. Inapparent carriers are common and complicate epidemiological control measures. Recently, the use of attenuated live oral vaccines has shown promise in protecting human volunteers (DuPont et al. 1972b). There is need for a vaccine which is multivalent against the activity of S. sonnei and several subtypes of S. flexneri in the United States.
SUMMARY
Shigellosis was recognized as the most common specific bacterial cause of acute diarrhea in Vietnam. While laboratory evidence of tetracycline resistance emerged, no major clinical problems arose. Ampicillin was the drug of choice when antibiotic therapy was indicated. Studies of the R factor were not accomplished in Vietnam, although it was postulated as a mechanism for the emergence of drug resistance.?
Section II. Typhoid Fever and Other Salmonellosis
Colonel Kenneth I. Hedlund, MC, USA, and Brigadier General Andre J. Ognibene, MC, USA
TYPHOID FEVER
History
The genus Salmonella contains more than a thousand distinct serotypes, a large number of which are pathogenic for both animal and man. Frequently, these etiologic agents were not isolated and characterized until the clinical syndromes and associated pathologic lesions they produced were well delineated. The clinical manifestations of salmonella infection range from enteric fevers to septicemias and acute gastroenteritis.
Typhoid fever, the classic form of enteric fever, has a recorded history from the time of Hippocrates. The first complete description of the clinically recognizable syndrome is attributed to Thomas Willis, who, in 1659, described a fever which had an insidious onset and was marked by stepwise increases during the first week. The fever, he said, was maintained during the second and third
361
weeks and then resolved by gradual lysis. Willis also described the relapse in some patients who had recovered (Huckstep 1962, p. 4).
In 1826, Bretonneau delineated the pathological picture of typhoid fever by demonstrating the characteristic inflammatory involvement of both Peyer`s patches and Brunner`s glands. In 1829, Pierre Louis coined the term "Fievre Typhoide." In 1837, William Gerhard, a student of Louis, set down the clinical criteria to differentiate typhoid fever from typhus. At the time, this was a radical and generally unaccepted concept; both diseases were considered to be merely variants of the same disorder. Not until 1850, when Sir William Jenner confirmed and expanded upon Gerhard`s work, did the distinction gain general acceptance (Huckstep 1962, p.5).
Between 1856 and 1860, William Budd demonstrated that typhoid fever was transmitted from man to man through excreta, and that contaminated water and milk could cause typhoid. Eberth, in 1880, described "B. typhosus" in the tissue of patients. Four years later, Gaffky successfully isolated and cultured Salmonella typhosa, the typhoid bacillus, for the first time. After these breakthroughs, the subgroups of the Salmonella grew progressively larger as systematic investigations were made of patients with typhoid like illnesses. The term "paratyphoid fever" was introduced in 1896 by Achard and Bensaude following their isolation of S. paratyphi B (Huckstep 1962, p.6). In that same year, Gruber and Durham (1896) and Widal (1896) detected agglutinating antibodies in the serum of animals and patients previously infected with the typhoid bacillus.???
The first immunizations of humans against typhoid fever were performed by Wright and Semple (1897) in 1896; they injected heat-killed typhoid bacilli subcutaneously in two volunteers. The first attempt at the immunization of U.S. Army troops against typhoid was by means of an oral vaccine in 1904; it proved unsuccessful because of the probable presence of viable bacilli in what was erroneously believed to be a heat-killed vaccine (Tigertt 1959). In 1909, subcutaneous injection of heat-killed typhoid bacilli was given to U.S. troops on a voluntary basis, and by 1911 it became compulsory. In 1915, the vaccine was fortified with S. paratyphi A and B (MD-WW9, pp. 21, 41).
In 1947, chloramphenicol was introduced (Erlich et al. 1947). One year later, Woodward and associates (1948) reported the beneficial effects of this drug in the treatment of typhoid fever. With the advent of an effective antibiotic, the mortality rate of typhoid fever in young men dropped from 10 percent to 1 percent (MD-WW9, p. 17).
Strains of S. typhosa which are resistant to chloramphenicol have been encountered since 1950 (Colquhoun and Weetch 1950). However, it was not until 1972 that reports of the first epidemic caused by a resistant S. typhosa were received (CDC-M&M; Vazquez, Calderon, and Rodriguez 1972; Gangarosa 1972). This epidemic, which began in Mexico in that year, persisted for months and affected thousands of people. In a report given at the Interscience Conference on Antimicrobial Agents and Chemotherapy, Gangarosa noted that the multiresistant organisms caused a clinically severe disease with a high mortality rate. In addition, epidemiologically, there appeared to be greater infectivity and
362
transmissibility. Although water was implicated as a vehicle in the spread of the disease, person-to-person contact has also been implicated. The latter method of transmission is not usually seen in widespread typhoid epidemics.
Anderson and Smith (1972) pointed out that the causative organism in the Mexican epidemic was resistant to chloramphenicol, tetracycline, streptomycin, and the sulfonamides and that the resistance to these drugs is attributable to an R factor. This extrachromosomal element can be transferred to a sensitive S. typhosa by an intestinal commensal such as Escherichia coli or a resistant strain of shigella organisms. These authors suggest that two conditions optimize the possibility for such a transfer of en bloc drug resistance. The first is the presence of endemic typhoid so that the organism is frequently present in the intestine. The second is the continuous, indiscriminate use of chloramphenicol so that its widespread selective pressure promotes the emergence of stable R factors coding for specific resistance. These conditions existed in Mexico and also in Southeast Asia.
Davis and Anandan (1970) have pointed out that in isolated "antibiotic virgin" areas R factors had evolved in nature before the manmade antibiotic era. The members of the enteropathogens that by chance have received R factors by conjugation possess a selective advantage only after the introduction of manmade antibiotics. These enteropathogens have emerged in some instances as the predominant members of their population.
Vivona and coworkers (1966) showed the presence of en bloc antibiotic resistance to sulfonamide drugs, streptomycin, tetracycline, and chloramphenicol in various strains of Shigella flexneri Sh. sonnei, and E. coli cultured from Vietnamese patients with acute diarrhea. Gaines and Nhu-Tuan (1968), although not reporting directly on R factor resistance, demonstrated that, among a native Vietnamese population of 645 strains of Shigella sampled, 437 (67.7 percent) were resistant to chloramphenicol. Of 287 strains of E. coli tested, 196 (68.2 percent) were resistant to the drug, as were 60 (24.3 percent) of 246 strains of Salmonella.
Military Significance
Insight into the past military significance of typhoid fever can be obtained by reviewing both the American and British experiences in an era before controlled immunizations and antibiotics. Walter Reed (Reed, Vaughan, and Shakespeare 1900, pp. 190-92) noted that one-fifth of the soldiers in the national encampment in the United States in 1898 developed typhoid fever. Among the 107,973 officers and men in 92 regiments, the number of cases was 20,738 (an incidence rate of 19.21 percent). The death rate was 7.61 percent (1,580 cases). Deaths from typhoid fever accounted for 86.24 percent of the total deaths from all causes. The morbidity from the disease was 192.65 per 1,000 average strength, or a little less than one-fifth. Mortality per 1,000 average strength was 14.63.
Although a partial attempt at immunization was made by the British, the American experience correlates fairly well with that of the British Army during
363
the Boer War (1899-1901) (MD-WW9, p.17). There 209,404 men developed 59,750 cases of typhoid fever, with 8,227 deaths, or a morbidity of 114.13 per 1,000 and a mortality of 14 per 1,000.
With improved field sanitation and compulsory antityphoid-paratyphoid immunizations, the incidence rates among the U.S. military fell remarkably during the First World War (table 65).
Vaughan`s study (1920) suggests that the high mortality rate given for World War I (14.85 percent) is really somewhat inflated. If one accepts his conclusions, untreated typhoid fever carried then, as it does now, a mortality of approximately 10 percent.
The incidence rates for varying types of Salmonella infections during both World War II and the Korean war are shown in table 66. These reflect the fact that when good field sanitation and compulsory immunizations are maintained, Salmonella infections cease to be a significant threat to the Army`s operations or activities.
In Vietnam, the environment beyond the confines of the base camp could not be controlled with any reliability. Protection against salmonella infections took the form of immunization for all incoming personnel and maintenance of "safe in-house" water and food supplies at the base camps. In the field, canned rations were supplied along with iodine or chlorine tablets to disinfect available "field water."
No one actually knows, with any certainty, the real incidence of typhoid fever which occurred among U.S. troops in Vietnam. For example, in 1968, during a brief 6-week inspection tour of several installations in Vietnam, Dr. Samuel B. Formal,* a WRAIR (Walter Reed Army Institute of Research) bacteriology consultant, was able to gain indirect evidence of at least three deaths following S. typhosa infections. However, only eight cases of typhoid fever among American personnel were reported that year for all of Vietnam (PAD 1968), which would imply a 37.5-percent mortality rate. No matter what the total number of cases or the true mortality rates were, the only reasonable statement that can be made about the incidence of typhoid fever among American troops in Vietnam is that it was underestimated.
TABLE 65.-Admissions and deaths from typhoid fever during the Spanish-American War and World War I
_____
1 Dr. Samuel B. Formal: Personal communication.
364
TABLE 66.-Salmonella infections, active-duty Army, 1942-45 and 1950-53
Incidence
Typhoid fever was endemic in Vietnam. In mid-1965, the indigenous population was 16.1 million with a projected growth rate of 2 percent per year (Smith et al. 1967). The number of cases of typhoid and paratyphoid fevers among the South Vietnamese for each year between 1965 and 1970 was as follows (NISVSY, p. 268):
1966 - 3,321 1968 - 3,420
1966 - 3,934 1969 - 3,805
1967 - 3,531 1970 - 3,305
These numbers represent only the number of reported cases and are best considered a reflection of the problem. During the same period of time, the average U.S. Army troop strength was 250,000 per year and the total number of cases of typhoid fever reported in American Army troops was 29 (table 67). There were also 425 cases of undefined salmonellosis.
365
The incidence of typhoid fever was 141.59 per 1,000 during the Spanish American War, with a mortality rate of 10 percent. Following compulsory immunization procedures, the incidence rate dropped to 0.37 per 1,000; however, the mortality rate was no lower. Statistics from World War II, the Korean war, and the Vietnam war all show very low overall incidence rates. This is a testament not only to the efficacy of immunization but also to the advances of preventive medicine and public health in establishing the techniques of typhoid control in food handling, water supplies, and typhoid carrier surveillance.
Retrospectively, it appears that while control measures reduced the incidence, the introduction of chloramphenicol reduced the mortality rate. The potential threat of en bloc antibiotic-resistant strains of S. typhosa, such as those found in the Mexican typhoid epidemic, is one that should not go unnoticed, especially since person-to-person contact has been implicated in the establishment of the epidemic. New antibiotic combinations may not provide a long lasting solution.
TABLE 67.-Incidence of typhoid fever and salmonellosis in U.S. Army troops in Vietnam, 1965-70
Etiology
The Salmonella are gram-negative, nonspore-forming, motile bacilli. In classic form they do not ferment lactose or sucrose; however, variants have been found that do ferment lactose. Although the genus can be separated from many other varieties of enteric bacilli by differing metabolic reactions, the final identification of individual Salmonella species is based on differences in specific H, 0, and Vi (virulence) surface antigens. More than a thousand subtypes have been distinguished on the basis of these specific antigens and identified by exhaustive cross-absorption and cross-reaction serological tests.
Previously, attempts were made to categorize Salmonella on the basis of infectivity. For practical purposes, one can say that S. typhosa produces a disease-typhoid fever-which is found only in man, while the bulk of other
366
Salmonella organisms can produce disease in both man and animals. The propensity spectrum ranges from those organisms more frequently found to cause human disease, such as S. paratyphi A and B, which produce typhoidlike disease, to those organisms which seem to be uniquely animal pathogens, such as S. gallinarum. The middle ground includes S. typhimurium, S. choleraesuis, and S. enteritidis, which produce disease primarily in animals but can also infect man.
Epidemiology
Salmonella typhosa organisms are spread primarily by human contamination of either water or food. In the more advanced countries, periodic modern outbreaks of typhoid have occurred when public water supply surveillance has lapsed or when unprotected and untreated water and food stores are used (Bernard 1965). In Vietnam, U.S. troops who were fed in carefully controlled mess halls could and would drink from and swim in polluted waterholes. Among the 17 patients treated by the author for typhoid fever at the 85th Evacuation Hospital during 1970, there were at least three distinct sources of infection: a polluted waterhole along the highway where troops swam, a food handler in Phu Bai, and a popsicle vendor in the city of Hue.?
The nontyphoid salmonellae capable of causing human disease have a wide host distribution in nature. Domestic animals, poultry, and hen and duck eggs are potential sources of infection, as are human beings with clinical or inapparent disease.
Clinical Features, Course, and Therapy
Typhoid fever is one of the historically classic diseases that very few American physicians have seen or treated. The signs and symptoms which are present at the time of admission depend to a large degree on how long the pa tient has been ill; the classic onset is insidious. Early symptoms -frontal headache, dizziness, anorexia, and general malaise - are similar to those of many other febrile illnesses. The patient in Vietnam usually ignored these and a large variety of other nonspecific symptoms until the onset of fever, which was seen as a valid reason for seeking medical aid.
The classic "stepladder" increases in fever were rarely, if ever, seen. One reason was that symptoms usually began between 5 and 8 days before the patient came to a hospital. On admission, fevers ranged from 101? to 104?, and a relative bradycardia was common. Signs of meningismus were seen in 17 percent of the author`s patients, suggesting meningitis, but examinations of the spinal fluid were negative.
One-third of the patients had an intermittent nonproductive cough and auscultatory signs of bronchitis. The abdomen was characteristically diffusely tender, with distention. While the spleen was not palpable, left upper quadrant tenderness was noted frequently. "Rose spots" were identified in only 1 of the author`s 17 patients.
367
In the interval between hospitalization and laboratory confirmation of the diagnosis of typhoid fever, there was usually a progression of toxicity. In some of the more seriously ill patients, mental confusion replaced apathy, and the initial tenderness and cramping pain in the abdomen progressed to the ileus with overt distention. In some, the skin became sallow and doughy feeling.
Chloramphenicol therapy brought the temperature to normal within 3 to 4 days. Bronchitic signs and toxemia diminished with the fever, and abdominal distention resolved within 72 to 96 hours. Chart 23 shows the course of a previously immunized American patient who contracted typhoid fever in Vietnam. It illustrates the characteristically low white count, which is unfortunately shared with several other febrile illnesses endemic to Southeast Asia, such as dengue and malaria. The response to chloramphenicol therapy is reflected in the temperature chart after 72 hours. Also shown is the unpredictability of the Widal reaction, perhaps influenced in this case by prompt treatment with chloramphenicol.
Chloramphenicol is the drug of choice against sensitive typhoid strains. It is usually given 50 mg per kg per day, either orally or intravenously, until the patient is afebrile. At that point, the daily dosage is reduced by half until a total of 2 weeks of chloramphenicol therapy has been given. An effective alternative drug, when chloramphenicol is contraindicated, is amoxicillin, 1 g every 6 hours for 14 days (Afifi, Adnan, and El Garf 1976).
Stamps and Wicks (1972) treated 92 South African typhoid fever patients with a combination of trimethoprim and sulfamethoxazole. There were only three treatment failures and the side effects were minimal. The average time to bring the temperature to normal was 4.1 days, which compares favorably with that required by chloramphenicol. This drug combination and others like it will doubtlessly be useful in areas where antibiotic-resistant strains of S. typhosa are present.
The general supportive treatment of patients with typhoid fever, paratyphoid fever, or severe acute Salmonella gastroenteritis with suspected bacteremia is best undertaken in a hospital and in an area separated from the general patient population. Cooling blankets and antipyretics were used in Vietnam for temperatures above 103 0 F. Antiemetics were used to control nausea and vomiting. Diarrhea was not a major problem and, when present, was easily controlled.
Laboratory Diagnosis
The cornerstone of the diagnosis of typhoid fever is isolation of S. typhosa from the blood, stool, or urine of the patient. As seen in chart 24, positive blood cultures are obtained in 80 to 90 percent of the cases during the first week. As many as 25 percent have positive urine cultures in the later stages of the disease. In acute Salmonella gastroenteritis while positive stool cultures are found in the early stages of the disease, blood cultures are usually negative. Specific agglutinins formed in response to the surface antigens of the Salmonella organism can usually be demonstrated in the serum of patients 1 to 2 weeks
368
CHART 23.-Course of typhoid fever of a previously immunized American patient in Vietnam
369
after the onset of typhoid fever, septicemia, or the more severe acute gastroenteritis cases.
Variations occur between the H and 0 titers of unimmunized persons and those of immunized persons. In the unimmunized individual, the 0 agglutinins appear before the H and are usually higher at first. (Their positions are reversed in the convalescent phase.) The presence of high 0 titers (1:160) with low H titers thus suggests typhoid fever in an unimmunized individual. Active immunization with typhoid-paratyphoid vaccine produces both H and 0 agglutinins; therefore,
370
high titers of H agglutinins (1:1280) can be invoked as an anamnestic response to typhoid bacilli in a previously immunized individual. However, it should be stressed that a generalized increase in H titers occurs in patients with certain non-Salmonella infections. Therefore, the only realistic way to follow possible typhoid infections in previously immunized individuals is by serial determinations of H and 0 titers at 3-day intervals. If the titers progressively increase, and especially if the 0 titers rise, then typhoid fever is a strong possibility.
Early treatment with chloramphenicol or any other effective antibiotic will cause the agglutinin titers to remain low or within normal limits (El-Rooby and Gohar 1956). In addition, nontyphoid Salmonella organisms may cause an in crease in agglutinin levels but usually to a lesser degree than typhoid salmonellae. Serological determinations are therefore at best an indication of possible Salmonella infection.
Complications
Typhoid fever has a mortality rate of 10 to 14 percent in untreated cases, as mentioned earlier. If one substitutes the word "undiagnosed" for "untreated," the true dimensions of the problem become apparent. The following case illustrates the point:*
The patient, a 22-year-old male, was admitted to the 9th Surgical Hospital on 31 August 1968, with a 2-week history of fever. On 9 September he was transferred to the 2d Surgical Hospital with fever, tachycardia, and right upper quadrant tenderness. On 13 September he was transferred to the 67th Evacuation Hospital with a persistent temperature of 103 0 F. Four days later, blood cultures were reported as positive for S. typhi.
The patient was apparently then started on colistin and later on chloramphenicol. By 21 September, he had developed overt gastrointestinal hemorrhage and perforation. The hematocrit was 23%. After transfusion, an exploratory laparotomy was performed on 24 September and a portion of the ileum was resected. One day later, a subtotal colectomy was done. The patient died on 28 September 1968. At postmortem, the stomach showed no areas of ulceration. What remained of the small and large bowel showed focal areas of hyperemia over the serosal surfaces. The mucosal surface of the small intestine was described as discolored and hemorrhagic.?
This case demonstrates that the longer diagnosis and treatment are delayed, the greater the possibility of perforation and hemorrhage. Once formed, these lesions are generally unaffected by antibiotics. Gastrointestinal hemorrhage and perforation account for 75 percent of all the deaths caused by typhoid fever (Boyd 1965, p. 416).
Hemorrhage, like perforation, is rare before the 10th day of illness and, more importantly, hemorrhage and perforation in the antibiotic era may occur in afebrile patients who clinically are doing well. Osler (1935, pp.1-33) noted a 7-percent incidence of overt intestinal hemorrhage in the 23,721 cases he reported during the preantibiotic era. The incidence of overt hemorrhage has been
______
1 Dr. Samuel B. Formal: Personal records.
371
relatively uninfluenced in the antibiotic era(Woodward and Smadel 1964; Rowland 1961).Stuart and Pullen (1946) reported chemically detectable blood in the stool in 52 percent of their 360 patients. Obviously, small hemorrhages may go unnoticed. The larger ones may be preceded by abdominal pain, vomiting, and dizziness, progressing to overt shock. Prompt diagnosis and blood replacement therapy are important in lowering mortality.
The incidence of perforation reported is about 1.3 to 5.1 percent (Woodward and Smadel 1964, p. 150; Rowland 1961, p. 107).Many investigators believe that the incidence is not appreciably affected by treatment with chloramphenicol (Woodward and Smadel 1964, p.150; El Ramli 1950). The immediate cause of perforation, as of hemorrhage, is ulceration followed by necrosis. The onset of perforation may be either insidious or sudden. In sudden perforation, prodromal signs of increasing pain and hemorrhage are not invariably present. Abdominal pain, rigidity, vomiting, and rapidly progressive collapse occur. In the insidious type of perforation, there may be little pain and the signs and symptoms of generalized peritonitis may be absent or slow in appearing. Typhoid fever patients are frequently apathetic and even obtunded. Abdomens are often mildly distended, white cell counts low, and temperatures elevated. A sudden drop in temperature may be seen following perforation, in some after a modest initial 1 to 2 rise. The white cell count may rise rapidly following perforation, going to 10,000/mm3 or 15,000/mm3 within a few hours. In patients who are being given high levels of antibiotics and steroids, the signs and symptoms may be markedly atypical or masked altogether.
The treatment of intestinal perforation in patients with typhoid fever has been debated for some time. There is classic disagreement between proponents of conservative medical management and aggressive surgical intervention. In the preantibiotic era, when perforation occurred, death was almost inevitable. However, Osler (1935, pp.18, 32) suggested that early diagnosis and surgery could save about one-third of patients who developed this otherwise fatal complication.???
Huckstep (1962, pp.187-95) in a 23-patient series, points out that 27 percent of 15 patients with intestinal perforation alone died following aggressive surgical management, while 13 percent of eight patients with peritonitis and probable perforation died with medical management alone. Medical management in the series consisted of gastric aspiration, large doses of chloramphenicol, careful monitoring and maintenance of electrolyte balance, and mild sedation as well as a high level of nursing care. Huckstep notes two important exceptions to conservative medical management. The first is when sudden perforation is seen in the convalescent patient within 6 hours of onset. The second is when perforation has led to complications, such as adhesions causing obstruction. In both of these exceptions, he advocates surgical intervention.
Li (1963), in a 20-case series of proven typhoid perforation, reported only a 10-percent mortality following surgical intervention and chloramphenicol therapy. Archampong (1969), in a 121-case series, advocated early surgical intervention. His overall mortality rate following surgical intervention was 29.8 percent. However, 90 percent of the patients he operated on had perforation before their admission to the hospital and therefore had no prior antibiotic
372
therapy. More importantly, he points out that there is only a 13.3-percent mortality if surgery is performed in the first 24 hours. If a delay of 2 to 5 days occurs, the mortality rate rises to 26.2 percent; after the fifth day, it jumps to 76 percent.???
Woodward and Smadel (1964, p.152) emphasize that when it becomes obvious to the clinician and the attending surgeon that the infectious process is failing to localize under antibiotic therapy, as evidenced by the findings on examination and by the persistence of shock and leukocytosis, it may be surmised that the ulcer is not healing and that surgical intervention is indicated.
Early surgical intervention became the standard of practice in Vietnam for perforation.
The other complications of typhoid fever include cholecystitis, pneumonia, hemolytic anemia, nephritis, arthritis, psychosis, diffuse intravascular coagulation syndromes, hepatitis, thrombophlebitis, toxic myocardiopathy, and metastatic abscesses throughout the body.
Pathogenesis
Whether or not one becomes ill following infection depends upon the interaction of a large number of variables. Among these are the virulence of the pathogen and the size of the inoculum. The "natural" and "acquired" resistance of the host also plays an important role. Using human volunteers immunized against typhoid fever, Hornick and coworkers (1970a) were able to establish an I.D.50 (median infective dose) of 107 S. typhosa (Quailes strain) organisms. This strain contains the Vi antigen. In contrast, when a numerically equivalent inoculum of S. typhosa organisms lacking the Vi antigen was used, the disease rate was only 26 percent. Once induced, the disease in immunized volunteers is identical to the naturally acquired disease regardless of the size of the initial inoculum. Following ingestion, the bacilli multiply rapidly in the intestinal tract and stools are frequently, but not always, positive for S. typhosa within 24 hours. S typhosa multiplication within the intestinal tract, however, is not an absolute indication that the volunteer will contract typhoid fever or even become ill. In addition to multiplication, there is rapid penetration. Sprinz (1969) points out that, in contrast to shigellae, salmonellae are readily transported through the intestinal epithelial lining without marked local inflammatory response.
Once through the intestinal wall, the bacilli migrate to the mesenteric lymph nodes where multiplication continues. From there, they gain access to the blood via the thoracic duct and are filtered by the reticuloendothelial system, primarily the spleen and liver. They subsequently reinvade the bloodstream.
Microscopically, the activated reticuloendothelial system responds to typhoid bacilli challenge by an active proliferation and accumulation of macrophages (mononuclear phagocytes) within the lymphoid tissue of the intestine. The packing of these lymphoid pockets by enormous numbers of macrophages disrupts the normal architecture. During the first week of illness, the pockets of lymphoid tissue become grossly swollen and project above the mucosal surfaces.
373
The macroscopic lymphoid swelling is caused by the large number of phagocytic cells and by the edema which results from local circulatory compromise. Vascular occlusion leads to ischemic necrosis and then the formation of a superficial slough. The necrosis may extend laterally into the mucosa or deeply into the tunica muscularis or the peritoneum. The areas most frequently involved are the lower portion of the ileum, the cecum, and the proximal portions of the colon. If the local processes leading to necrosis are slow, the vessels involved are usually obliterated; if the process is rapid, the blood vessel walls are frequently eroded.???
As Boyd (1965, pp. 414-21) points out, the gravity of hemorrhage is appreciated when one considers the blood supply of the Peyer`s patches. A large number of parallel vessels run into the patch from the plexus in the submucosa, so that even a superficial ulceration may cause a very severe hemorrhage.
The role of bacterial endotoxins is still being elucidated. These lipopolysaccharides form an integral part of the gram-negative cell wall and are released into the host circulation upon destruction of the bacterium. Hornick et al. (1970b) found that as little as 0.25 u, of purified endotoxin from S. typhosa given intravenously can produce chills, fever, and headaches as well as a depression of platelets and leukocytes. Neither the tolerance developed to increasing amounts of pure endotoxin nor the resultant high anti-0 antibody titer affords protection in volunteers to the subsequent development of classic typhoid fever. Endotoxin, a potent inflammatory agent, also provides the endogenous release of pyrogens from polymorphonuclear cells and monocytes, and this explains in part the sustained temperature elevations. In addition, chemotactic factors are generated for both polymorphonuclear leukocytes (Snyderman, Gewurz, and Mergenhagen 1968) and monocytes (Ward 1968) by the interaction of complement components with endotoxin or other bacterial products. The characteristic mononuclear cell infiltrate in typhoid fever, in contrast to the polymorphonuclear leukocyte inflammatory response in other Salmonella infections remains incompletely explained.
Summary
In areas where typhoid fever is endemic, it is common for local physicians to miss the diagnosis on admission (Chalmers 1971). This was as true for American physicians in Vietnam as for physicians elsewhere. There were a number of reasons for this, but perhaps the most important was that the disease was simply not included in the differential diagnosis because of a misunderstanding of the effectiveness of typhoid immunizations. In addition, the classic textbook presentation of epidemic typhoid fever is the exception rather than the general rule in the endemic disease (Wicks, Holmes, and Davidson 1971; Paul 1952, pp.291-315).???
Typhoid fever can be controlled in large measure by immunizations and by the enforcement of common public health measures; however, immunizations per se do not give complete protection to a person if he receives a large enough inoculum (Hornick et al. 1970a, pp.687-88). Typhoid fever now (Chalmers 1971), as in the past, remains a great imitator of other febrile illnesses. It always must
374
be considered in the differential diagnosis of febrile illnesses in areas where the disease is endemic. Early antibiotic treatment and fluid replacement are the basis of therapy.
OTHER SALMONELLOSES
The approximately 1,500 serotypes which comprise the genus Salmonella make taxonomy unwieldy. They may more easily be classified into three ecological groups according to host preference-those specifically adapted to man, those adapted to animals, and those with no specific host preference. Table 68 clarifies the relationship of Salmonella species and representative serotypes to human disease.
Although Salmonella infections other than typhoid cause relatively mild and self-limited disease, they are of military significance because of their abrupt onset and the frequent epidemic clustering of cases.
Kuhns and Learnard reviewed the subject of typhoid and paratyphoid fevers from the beginning of the Army`s reporting system in 1818 through World War II (MD-PM4, pp. 463-82). (Kuhns reviewed Salmonella food poisoning separately [MD-PM4, pp. 417-31].) Table 69 reflects the experience of the U.S. Army in that conflict with "paratyphoid fevers."
Precise data on the occurrence of salmonella infection in Vietnam are not available, although USARV Command Health Reports (USARV-CHR) from 1969 to 1970 indicate a monthly incidence of one or two confirmed cases.
In the United States, the number of salmonella infections other than typhoid increased from 504 in 1942 to 19,723 bacteriologically confirmed cases in 1967 (NCDC-SS). This number is believed to reflect only a fraction of actual cases. This increase represents increased surveillance and changing patterns of mass food processing and distribution. The major reservoirs of human salmonellosis are livestock and poultry. Outbreaks have occurred as a result of contamination of dried milk, poultry, pork, shellfish, eggs, water, and carmine dye (DuPont and Hornick 1969; Bauer 1973).
The disease varies with the infecting species of Salmonella nd the number of organisms ingested. While salmonella infection does involve intestinal epithelial invasion, it does not result in extensive destruction of the intestinal mucosa. The epithelial lining is left intact, and an inflammatory response is elicited when the organisms reach the lamina propria. The nature of the inflammatory response seems to be important in determining the pathogenesis of the disease and the resultant symptomatology. S. typhosa organisms, for reasons that are not understood, produce a predominantly mononuclear inflammatory response in which the organisms are carried into the circulation, resulting in enteric fever with systemic symptoms and bacteremia. In contrast, nontyphoid salmonellae produce a predominantly polymorphonuclear reaction in which the organisms are phagocytized and contained in the lamina propria, the clinical expression being a gastroenteritis. Although suitable pathologic studies have not been conducted on human salmonella gastroenteritis, it is probable that in most cases invasion of the distal small bowel as well as the colonic mucosa occurs (DuPont and Hornick 1973).
375
TABLE 68.-Relation of Salmonella species and representative serotypes to human disease
TABLE 69.-Incidence of paratyphoid fever in the U.S. Army, by area and year, 1942-451
376
While the O or somatic antigen relates to production of an exotoxin which is toxic to man when injected under experimental conditions, it is likely that endotoxins from senescent bacteria are continuously present in the gut and that few or none of them are absorbed under normal circumstances. Other factors that may influence the pathogenicity of salmonella are intestinal immune factors, bowel motility, gastric acidity, and the bacterial flora of the gut. Short chain fatty acids, such as acetic and butyric acid, are produced by many intestinal bacteria in sufficient concentrations to inhibit the growth of virulent enteric pathogens (DuPont and Hornick 1973b).
Salmonella gastroenteritis is the most common clinical manifestation of salmonella infections in man. Symptoms begin 12 to 24 hours after ingestion of the organism. There is sudden onset of abdominal pain, nausea, diarrhea, and vomiting. Fever nearly always occurs. Dehydration in infants may be severe. Anorexia and loose stools may continue for several days. The process is self-limited, and recovery occurs in 2 to 5 days (Bauer 1973, p.324).???
The enteric fever pattern of salmonellosis is less common than salmonella gastroenteritis but may occur with any of the species of Salmonella infecting man. Clinically, the illness is less severe than typhoid, but on occasion it may be serious and, rarely, fatal. The onset is characterized by malaise, anorexia, and diarrhea. Shortly, the temperature begins to rise and will persist for 1 to 3 weeks if untreated; initially, it may reach 102? to 104. During the first week, blood cultures will be positive and stool cultures often negative. S. enteritidis serotypes, S. paratyphi A, and S. schottmlleri tend to be more common causes of enteric fever but also may cause gastroenteritis.
Septicemia is the most serious form of salmonella infection. When septicemia occurs, the patient manifests high fever and prostration while gastrointestinal symptoms are often minimal. Focal abscess formation may oc cur with endocarditis, septic arthritis, osteomyelitis, meningitis, cholecystitis, or abscess of the lungs or kidneys. While any of the species of Salmonella can cause widespread sepsis, it is most common with S. choleraesuis.
The mortality rate is highest among patients over 50 years of age. Saphra (1950) reported 174 fatalities (5.3 percent) in 3,279 human infections, with a fatality rate of 15 percent in patients over age 50, 5.8 percent in infants, and 2 percent in patients between ages 1 and 50. S. choleraesuis had the highest fatality rate, 21.3 percent, while S. typhimurium and S. oranienburg had fatality rates of 5.5 percent.
Following acute infections, 40 to 60 percent of patients continue to shed salmonella when tested at 1 month, with a rapid decrease thereafter (Rubenstein, Feemster, and Smith 1944). The carrier state is usually temporary in un treated patients. Permanent carriers are primarily typhoid and paratyphoid patients. The carrier rate in the general population is estimated at 0.2 percent (Saphra and Winter 1957).
There is often a leukocytosis, although patients with septicemia may have a leukopenia. In salmonella gastroenteritis, the white count may be normal or only slightly elevated. Definitive diagnosis depends on isolation and identification of the offending salmonella from blood, feces, or urine. Blood cultures obtained dur-
377
ing the first week of illness are often the most productive source of material, although a rectal swab or fresh stool specimen plated promptly may also be positive.
While indirect serologic means of diagnosis are available, there has recently been some concern about their validity. If serologic tests are used, the only meaningful value is the titer for 0 antigen. However, even this may be sup pressed by early antibiotic treatment or elevated by immunization (Bauer 1973; Schroeder 1968).
Proper processing, preparation, and handling of food are fundamental to the prevention of salmonellosis. As mentioned earlier, mass production and distribution of food for both humans and animals are conducive to wide dissemination of salmonellosis. This implies appropriate education of food processors and handlers and adequate bacteriological checks on production items. There is no salmonella vaccine for man except that for S. typhosa and S paratyphi A and B (Hornick et al. 1970a, b).
The appropriate therapy for salmonella gastroenteritis is strictly supportive, with fluid and electrolyte replacement. Continued emphasis of this point is required in a combat theater. Antimicrobial therapy is indicated only in patients with positive blood cultures or with septicemia and focal abscesses. The latter cases require adequate local management as well. The antibiotics of choice are chloramphenicol in doses of 1 to 2 g per day, or ampicillin in doses of 2 to 4 g per day, each in four divided doses. There is growing evidence, however, that the price paid for early control of toxicity is prolonged shedding of bacteria in the stool (Aserkoff and Bennett 1969). Therapy of the chronic carrier is frustrating unless an accessible focal lesion, such as cholelithiasis, is present.
There was a continuing awareness of the problem of salmonellosis throughout the Vietnam conflict, although no concerted effort was made to determine the incidence of carriers or to culture all patients with diarrheal illness. In 1969, a complete study of 204 patients yielded only 6 with salmonellosis (Kalas and Bearden 1969). This information paralleled clinical impressions; the reported incidence remained low and was overshadowed by acute nonspecific diarrhea and shigellosis. Antibiotic resistance was not recognized as a major management problem in salmonellosis and was not studied. The impact of widespread unnecessary antibiotic usage in diarrheal disease was not delineated.
Section III. Cholera and Vibrio parahaemolyticus Gastroenteritis
Colonel Ralph F. Wells, MC, USA (Ret)
CHOLERA
History and Military Significance
The U.S. military experience with cholera through World War II has been well summarized by Mosley (MD-PM4, pp. 451-62). The disease has had remark-
378
ably little impact on U.S. forces during any major conflict. In World War II, only 13 cases and two deaths were reported among American military personnel. These cases stemmed from two epidemics of limited extent, both occurring in China during the summer of 1945. One epidemic involved six men and the other, seven; each involved one death. In both instances, breaches o sanitary discipline were directly responsible. The only other case cited by Mosley was that of a Red Cross worker who apparently acquired a mild but bacteriologically confirmed infection while serving in Calcutta. Despite deployment of large numbers of military personnel to the India-Burma theater, the disease was not reported in American troops there, in contrast to British and Indian military experience and experience in the civilian populace. The two factors believed to be responsible for the remarkably low incidence of cholera among U.S. forces were an aggressive immunization program and rigid sanitary policies directed primarily toward the more common diarrheal diseases.
Calcutta was a major endemic focus throughout World War II and, in the early 1960`s, became one of two sites on the Indian subcontinent for major clinical investigation (Carpenter, Mitra, and Sack 1966). Studies in Calcutta were conducted jointly by the Calcutta School of Tropical Medicine, the Calcutta Infectious Disease Hospital, and the Johns Hopkins University Center for Medical Research and Training, Calcutta. The other major investigative program had been launched in the late 1950`s with the establishment of the Pakistan-SEATO (South East Asia Treaty Organization) Cholera Research Laboratory in Dacca, East Pakistan (now Bangladesh) (Gordon et al. 1966). These two programs are mentioned because of their fundamental contribution to the pathophysiology and therapy of cholera and, perhaps more importantly, for their pioneering work in the study of diarrheal disease in general.
Endemic cholera remained a continuing problem in India from the end of World War II throughout the Vietnam conflict.
When the buildup began in Vietnam, the disease, particularly the El Tor type, occurred in epidemic form in the Philippines. In 1964, the Republic of Vietnam had its first epidemic in 20 years; 20,000 cases were reported from areas under governmental control, concentrated in major urban centers such as Saigon, Hue, Da Nang, and Nha Trang. The continuing nature of the problem became apparent when more than 2,800 cases were reported in the first 3 months of 1966 (Sheehy 1968). Precise data on the incidence of cholera in North Vietnam are not available, although the existence of rigid immunization requirements for foreign visitors entering the country suggested the possibility of a significant problem (HD-25, p. 19).
Incidence and Epidemiology
The period beginning in 1961 has been classified as the "Seventh Cholera Pandemic." Although epidemic cholera occurred among the Vietnamese civilian populace during the peak years of the U.S. commitment in Vietnam, no cases were reported among American military personnel under U.S. military control. This generalization apparently also applies to military advisers. The incidence of
379
cholera among American prisoners of war is unknown. One possible explanation for this rather remarkable record is the aggressive policy of immunization before entry into the country with a booster immunization at approximately 6 months. A second consideration is the generally good nutritional state our personnel enjoyed, and a third is that the diagnosis may simply have been missed and the disease treated supportively as nonspecific diarrhea because of a lack of refined laboratory methodology.
Etiology and Pathogenesis
If an apology need be presented for discussing an uncommon disease at length, it is that the knowledge gained from the study of cholera not only promises to be of value in its prevention and treatment but also has led to isolation of an enterotoxin responsible for the massive fluid loss it causes. This, in turn, has resulted in a better understanding of the pathogenesis of other infectious diarrheas and added to our knowledge of normal mechanisms of fluid and electrolyte movement across the gut and other mucosal membranes.??
Vibrio cholerae is a gram-negative, nonspore forming, usually slightly curved (comma shaped) rod that is motile by means of a single polar flagellum. The two classic serotypes are the Ogawa and Inaba. Another biotype is the El Tor. A schema for its distinction from classic cholera will be presented when laboratory diagnosis is discussed. The El Tor vibrios are so named because they were first isolated from pilgrims at the El Tor quarantine station in the Sinai Peninsula in 1905.
Vibrio cholerae organisms neither invade the host nor cause morphologic damage to the intestinal epithelium. However, they proliferate rapidly, reaching tremendous numbers (104 to 108 organisms per ml) in the luminal content of the small bowel and the colon (Elliott et al. 1970; Gorbach et al. 1970). The pathologic effects of the organism are attributed entirely to the action of the exotoxin (enterotoxin or choleragen) produced in the intestinal lumen. This exotoxin has been purified and described as a heat- and acid-labile protein having a molecular weight of about 90,000 (Finkelstein and Lo Spalluto 1969).
The rapid loss of water and electrolytes in the stool, which may reach 20 liters or more in a 24-hour period, explains the clinical signs and metabolic derangements characteristic of cholera. As the colon is relatively normal, the small bowel has been incriminated as the site of this fluid loss. Four mechanisms of fluid loss have been proposed: exudation, increased transudation or filtration, inhibition of absorption, and increased intestinal secretion (Hendrix 1972). Several recent reviews have summed up the evidence, dismissing the first three mechanisms and implicating the fourth (Field 1971; Carpenter 1971b; Sladen 1973; Banwell and Sherr 1973).
John Snow, in 1855, had suggested that "cholera poison" played a key role, a viewpoint further spelled out in 1882 by Cohnheim, who concluded that under the influence of cholera organisms "there takes place an extra-ordinary profuse secretion from the glands of the small intestine" (Hendrix 1972). Further delineation of the mechanism was contingent on the definition of the functional anatomy
380
of the small bowel, the development of appropriate laboratory methods such as the rabbit ileal loop and the Using chamber, and some knowledge of the role of cyclic AMP (adenosine 3`, 5`-monophosphate).
Sutherland and Rall (1958) found that the glycogenolytic effect of epinephrine and glucagon was associated with an increase in cellular cyclic AMP. This observation led to extensive investigations which have clarified the subcellular biochemical changes responsible for mediating the physiological effect of catecholamines as well as many other hormones. Cyclic AMP serves as the intracellular mediator for hormonal action as follows (Field 1971, p.1142). The hormone, or other stimulating substance, binds on the cell membrane with its specific receptor. The hormone receptor binding stimulates adenyl cyclase activity. An increase in adenyl cyclase activity catalyzes the conversion of ATP (adenosine triphosphate) to cyclic AMP, initiating a series of intermediate steps which ultimately result in the physiological response or responses characteristic of a cell or an organ. Cyclic AMP is degraded to 5`-AMP by the enzyme phosphodiesterase. Thus, increased cyclic AMP activity may result either when adenyl cyclase is stimulated or when the breakdown via phosphodiesterase is inhibited. Apparently all nucleated mammalian cells may contain a cyclic AMP mechanism. ACTH (adrenocorticotropic hormone), TSH (thyroid-stimulating hormone), and thyroxine, among other hormones, seem to act through a cyclic AMP intermediate.
Both cyclic AMP and cholera toxin produce active secretion of chloride by ileal mucosa in an Ussing chamber; they also cause similar changes in transmucosal potential difference. These in vitro effects parallel the events in naturally occurring and experimentally induced cholera, which can be produced by cyclic AMP itself, by inhibition of phosphodiesterase and by at least one group of hormones, the prostaglandins, which increase the levels of cyclic AMP in the mucosa by activating intestinal adenyl cyclase (Field 1971).
In summary, the enterotoxin or choleragen released by Vibrio cholerae binds firmly and rapidly to the mucosa, fixing to the tips and crypts of the villi. Either directly or by means of an undefined mediator, the interaction between cholera toxin and the epithelium activates adenyl cyclase in cell walls of the secretory epithelium, resulting in an increase in the concentration of cyclic AMP. Binding appears to occur near the villus tip but does not involve the brush border. The effect of even brief contact of the cholera toxin with the intestinal mucosa persists for some hours. It has been suggested that exposure of new cells in the crypts to a substance such as cycloheximide, which inhibits the secretory process, may apparently take 2 to 3 hours and, not uncommonly, 12 hours to reverse the secretory process (Hendrix 1972; Sladen 1973).
Cholera toxin completely inhibits the absorption of sodium through the brush border; however, sodium transport via glucose or amino acid carrier is not affected. Since the glucose-mediated mechanism for sodium transport remains intact, patients with cholera can be given a balanced glucose-electrolyte solution for oral repletion. The active chloride pump moving chloride into the cell is likewise inhibited or even reversed by cholera toxin with a resultant flux of chloride into the gut lumen. The net effect is that the enterocyte becomes a
381
secretory cell with a large efflux of sodium and chloride associated with corresponding fluid loss accounting for the 10- to 20-liter fluid loss per day in the cholera patient. The bicarbonate loss remains unexplained (Field 1971; Carpenter 1971b; Sladen 1973; Banwell and Sherr 1973).
The mechanisms described above are applicable not only to Vibrio cholerae but also to Escherichia coli, Clostridium perfringens, Shigella dysenteriae, and staphylococci as well as a number of noninfectious but humorally mediated diarrheas (Field 1971; Carpenter 1971b; Sladen 1973; Banwell and Sherr 1973).
Clinical Features, Course, and Complications
Cholera usually occurs in an epidemic pattern, isolated cases being extremely rare. However, documented cases among American travelers in endemic areas have been infrequent, paralleling the American military experience. This may reflect travel at times when the vibriocidal effect of recent immunization is at its peak (Gangarosa 1971).
The incubation period varies from 1 to 3 days or longer. The prodrome may range from mild malaise and mild diarrhea to depression and prostration. The onset of clinical illness is usually explosive with profuse watery diarrhea. The stools may initially be normal colored or yellow but rapidly become colorless and odorless, the "rice water" stool that is the hallmark of cholera. Vomiting may occur; fever is uncommon. While colic and tenesmus are not often seen, severe muscle cramps of the extremities and abdominal wall are noted, reflecting fluid and electrolyte losses. The character of stool electrolyte losses is shown in table 70. Stool volumes, as noted earlier, may be as high as 20 liters per day. Oliguria and anuria may follow. In endemic areas, the cholera cot is widely used; this is a canvas cot with a "porthole" for the perineum, under which a large receptacle is placed allowing constant monitoring of fecal fluid losses. Volume for volume replacement is critical. Untreated severe cases may result in shock and death in 4 hours or less. Before the advent of current therapeutic measures, epidemic mortality rates varied from 30 to 80 percent (Carpenter, Mitra, and Sack 1966, p.165; Phillips 1966). Milder cases and asymptomatic carriers constitute the bulk of the cases in most epidemics and are a major public health hazard.
TABLE 70.-Composition of intestinal fluid (m Eq/l)
382
Laboratory Diagnosis
As Feeley pointed out (Gordon et al. 1966), laboratory diagnosis is of no significance to immediate treatment in the critically ill patient. To confirm the diagnosis, cultures should be obtained from rectal swabs or fluid stool. Conventional media used for "enteric pathogens," such as MacConkey`s agar, eosinmethylene blue, or Salmonella-Shigella agar, are suboptimal and even inhibitory. The simplest and most effective media are gelatin agar plates, tellurite taurocholate gelatin agar plates, and enrichment-transport media. Typical colonies are recognized by a turbid or cloudy zone caused by gelatinase production. Alkaline peptone water is a suitable enrichment medium and may be substituted for the tellurite medium. In the field, the transport medium is used for both transmission and enrichment of the vibrios during shipment. Suspicious colonies are then confirmed by appropriate agglutination and biochemical tests. Other diagnostic modalities include dark field microscopy, fluorescent antibody methods, and serologic diagnosis. A schema for the distinction of El Tor from classic biotypes is given in table 71.
Prevention and Treatment
Appropriate sanitary measures and immunization are the cornerstone of prevention. The actual efficacy of cholera immunization has been under scrutiny, as has typhoid immunization. There is considerable variation from manufacturer to manufacturer, and the duration of effectiveness is unknown. The U.S. military policy of recommending booster injections at 6-month intervals while in endemic areas undoubtedly contributed to the degree of success in Vietnam. A summary of control measures in the 1970`s as outlined by Benenson (1970, p.1207), follows.
1. Effective treatment of cholera as a diarrheal case.
2. Bacteriological surveillance of diarrheal diseases.
3. Chemoprophylaxis for members of the patient`s hearth-group.
4. Sanitary improvements:
Water supply.
Disposal of excreta.
5. Health education.
6. Immunizations on a voluntary basis.
7. Elimination of quarantine measures.
While intravenous fluids of appropriate volume and composition may be lifesaving, the logistical problems of this approach are insurmountable in underdeveloped countries (Carpenter 1971a; Phillips 1966). Fortunately, tetracycline therapy has been shown to rapidly eliminate the vibrio from the stool and significantly reduce the amount of fecal fluid loss (table 72).
Of equal importance has been the demonstration that cholera enterotoxin does not inhibit glucose transport and that it is thus feasible to administer a balanced glucose and electrolyte solution orally, which further reduces the requirement for intravenous fluids. The composition of this oral glucose electrolyte solution is as follows (Carpenter 1971a, p. 1201):
383
Sodium - 100 m Eq/l Bicarbonate - 40m Eq/l
Potassium - 10 m Eq/l Glucose - 120 m Eq/l
Chloride - 70 m Eq/l Osmolarity - 323 m Os ml
This solution may temporarily increase fecal fluid loss but when it is used in conjunction with tetracycline therapy, this untoward effect is short-lived.
New Advances
It is paradoxical that study of a disease infrequently encountered among U.S. military personnel provided a major breakthrough in the understanding of all enterotoxin-related diarrheal diseases. Because cholera was endemic among the civilian populace in the entire area of operation, there was ample opportunity for clinical study. Both NAMRU (Naval Medical Research Unit)-2 and the SEATO laboratories were intimately involved in fieldwork in conjunction with civilian activities in Bangkok, the Philippines, and Vietnam. Additional work was performed by other agencies, particularly in Calcutta and Dacca. Much of the fundamental laboratory work on the pathogenic mechanism of diarrhea, both in cholera and E. coli, was subsequently accomplished in the laboratory at WRAIR (Walter Reed Army Institute of Research).
TABLE 71.-Laboratory tests used to distinguish classical from El Tor biotypes of Vibrio cholerae
TABLE 72.-Effect of tetracycline on stool volume in cholera patients, 1963 study
384
VIBRIO PARAHEMOLYTICUS GASTROENTERITIS
History and Military Significance
Vibrio parahemolyticus, a halophilic, noncholera vibrio, is a relative newcomer on the military scene. First isolated by Japanese workers in the 1950`s, it was not recognized as a foodborne pathogen until the early 1960`s. It is of some importance in that, in the few years since its recognition as a pathogen, a near global distribution has been identified. Barker (1974) cites reports of V. parahemolyticus from India, Thailand, Malaysia, the Philippines, South Vietnam, Australia, Togo, Mexico, Panama, England, and the United States in a recent review from the Bureau of Epidemiology, Center for Disease Control, in Atlanta, Ga. The organism is discussed here because of the contributions of the WRAIR Medical Research Team, Vietnam, to the study of it (Neumann et al. 1972).
Incidence and Epidemiology
As part of their continuing effort to identify specific etiologic factors in patients with acute "nonspecific" gastroenteritis, the WRAIR team conducted an investigation of V. parahemolyticus infection from 12 October 1970 to 3 April 1971. Rectal swabs were obtained from 965 Vietnamese gastroenteritis patients at Saigon`s Cho Quan Infectious Disease Hospital and Nhi Dong Pediatric Hospital. Hospitalized U.S. servicemen were studied at the 3d Field Hospital, Saigon, and ambulatory patients, at the 299th Medical Dispensary, Tan Son Nhut. During preemployment physicals at the 218th Medical Dispensary, Saigon, 237 swabs were obtained from Vietnamese controls. In addition, cultures of seawater, sand, river water, and seafood purchased at local markets were collected from the Saigon and Cam Ranh Bay areas. V. parahemolyticus was isolated from 59 of 702 Vietnamese adults with diarrhea, 1 of 263 Vietnamese children, and none of the Vietnamese control populations. Vibrios were isolated from 2 of 82 U.S. servicemen with gastroenteritis. Isolates of V. parahemolyticus were obtained from fish, crabs, clams, shrimps, seawater, and sand but not from the fish or fresh water samples from the Saigon area. These results conform with the evidence accumulated in other epidemiological surveys and emphasize the halophilic nature of the organism.
Etiology and Pathogenesis
Crab, shrimp, lobster, and oysters have been incriminated in one or more outbreaks; these foods were eaten raw or cooked. In the former instance, poor refrigeration between preparation and serving allowed significant proliferation of organisms in shellfish naturally contaminated with small numbers of vibrios. When cooked foods have been implicated, either poor refrigeration or cross contamination of cooked seafood by uncooked appears to have been the cause. Crowded kitchens, the use of a common table for initial handling and subsequent
385
processing, and inadequate handwashing facilities for food handlers all played a role (Hooper, Barrow, and McNab 1974).
Observations suggest both bacterial invasion of the intestinal mucosa and enterotoxin production (Barker 1974, p. 553). The role of these and other mechanisms in pathogenesis is not yet resolved. While there may be serious morbidity, the mortality to date is exceedingly low.
Clinical Features, Course, and Complications
The common feature in all V. parahemolyticus infection is the history of recent seafood ingestion. The incubation time is typically 12 to 24 hours, though it may be as short as 2 to 4 hours or as long as 96. Inoculum size and host factors, such as gastric acidity, appear to modify the incubation time. Commonly, a cluster of cases may occur; the preparation of food for a large group predisposes to inadequate refrigeration or cross contamination (Barker 1974; Peffers et al. 1973).
The illness involves abdominal cramps, profuse watery diarrhea, nausea, vomiting, headache, fever, and chills, in decreasing order of frequency. Diarrhea, cramps, and nausea are almost universally present. The illness usually subsides spontaneously in 48 to 72 hours, though isolated cases lasting as long as 10 days have been reported (Barker 1974).
The case fatality ratio is low. In some, however, significant morbidity may occur. In a single epidemic of 12 patients who acquired their infection while aboard a chartered aircraft, 3 developed such severe dehydration and shock that Vibrio cholerae was initially suspected (Peffers et al. 1973).
Laboratory Diagnosis
When the organism is suspected, the most effective means of identification is direct culture on TCBS (thiosulfate, citrate, bile salts, sucrose) media. One may also use BTBST (bromothymol-blue, salt, "Teepol"). An alternate method, the technique used in Vietnam (Neumann et al. 1972), is to place the rectal swabs in Cary-Blair transport media for shipment to the laboratory. The specimen is then streaked on MacConkey`s agar and Salmonella-Shigella agar as well as selenite F broth and alkaline peptone water. To this routine should be added culture on TCBS. Overnight growth produces a characteristic green, domed colony. Other biochemical tests such as fermentation provide positive identification. The Kanagawa hemolysin test has also proved extremely useful. Most human isolates are Kanagawa positive. Thus far, more than 100 serotypes have been identified.
The organism proliferates rapidly. Generation time is 12 to 15 minutes at 37 C (Peffers et al. 1973). This rate may result in profuse stool passage of the organism, but as the infection is short-lived, stool studies must be done early.
386
Prevention and Treatment
The best prevention is good sanitation in the kitchen or food processing area. Adequate hand washing facilities properly used are mandatory. Specific therapy is supportive, consisting of fluid and electrolytes. Rarely, tetracycline may be necessary in an extremely severe case.
New Advances
This is the first conflict in which V. parahemolyticus has been recognized as a definitive enteric pathogen. Its occurrence in epidemic fashion and in strategic areas of the globe implies a potential for major military importance.
Section IV. Pathogenic Escherichia coli Diarrhea
Brigadier General Andre J. Ognibene, MC,USA, and Colonel Ralph F. Wells, MC, USA (Ret.)
HISTORY AND MILITARY SIGNIFICANCE
Acute nonspecific or "traveler`s" diarrhea is accepted as an inherent risk in foreign travel. The syndrome is characterized by the abrupt onset of diarrhea, usually within a few weeks of arrival in an underdeveloped area, and corresponds with the less severe of the two syndromes described in the report of the AFEB (Armed Forces Epidemiological Board) Commission on Enteric Diseases discussed in chapter 15 (Gezon 1966). It is dismissed by the casual traveler with a euphemistic descriptive term - for example, "Delhi belly" or "Montezuma`s revenge"-and quelled with a bit of paregoric.
The problem is far more significant in the military where the question of personal inconvenience is overshadowed by that of reduced operational effectiveness. A review of military history quickly spells out the magnitude of the problem. During World War II, the Allied experience in the Middle East implicated acute diarrheal disease as an important cause of loss of effectiveness among newly arrived troops (Hone, Keogh, and Andrew 1942; Bulmer 1944; Wirts and Tallant 1944), as did the American experience in the China-Burma India theater. In North Africa the tide may well have been turned by the diarrheal disease which afflicted Rommel`s troops (MD-PM4, pp. 319, 376). Several excellent British surveys done in recent years have defined diarrheal disease as a continuous problem in Arabia and to a lesser extent in British Guiana (now Guyana) and the Far East (Barnes and Moylan-Jones 1966).
Rowe, Taylor, and Bettelheim (1970), in a study of acute diarrhea among 540 British soldiers airlifted to Aden, were able to identify recognized enteric pathogens, such as salmonellae, in only 5.7 percent of 35 cases. In 33 cases,
387
pathogenic Escherichia coli was implicated, with positive cultures for serotype 0148k H28 in 19 of these. This corresponds with the early experience in Vietnam, where approximately 80 percent of reported diarrheal disease was short lived and of undetermined etiology (Kalas and Bearden 1969).
INCIDENCE AND EPIDEMIOLOGY
Diarrheal disease rates for Vietnam, abstracted from Command Health Reports from January 1966 through December 1970, are shown in table 73 and in chart 25. Only rarely did the monthly rate fall below 30 cases per 1,000 strength per annum. As specific pathogens were identified infrequently, it is safe to deduce that these rates basically reflect diarrhea induced by pathogenic strains of E. coli or by viruses. Serotyping of E. coli was not available in Vietnam during these years. Cultures obtained by a WRAIR research team were subsequently studied in depth at WRAIR and the University of Maryland, providing some useful retrospective data (DuPont et al. 1971). The problem of viral enteritis had not been addressed at the time of this writing.
Despite the fact that command responsibility was spelled out in several military publications (DA-FM, pp.5-8; AR 40-5, p.5-2; USARV Reg, par. 4), a large number of cases of diarrheal disease were assumed to stem from breaches of sanitary discipline in base camp or support areas. The possibility of troops consuming food, drink, or ice from indigenous sources remained throughout the conflict a risk which was hard to assess.
TABLE 73.-Monthly diarrheal disease rates, U.S. Army, Vietnam, January 1966-December 1970
388
CHART 25.-Monthly diarrheal disease rates, U.S. Army in Vietnam, January 1966-December 1968
ETIOLOGY AND PATHOGENESIS
Escherichia coli strains B2C and B7A were isolated from diarrheal stools of two American soldiers in Vietnam by the WRAIR team (DuPont et al. 1971). Both men had the sigmoidoscopic appearance of acute colitis at the time of culture. The CDC (Center for Disease Control) determined strain B2C to be serotype 06016. The B7A strain typed by WRAIR was found to be antigenically identical to the 0148:H28 strain reported by Rowe, Taylor, and Bettelheim (1970) as a cause of "traveler`s diarrhea" in British troops in the Middle East.
The two strains from Vietnam produced an enterotoxin in a rabbit ileal loop; in volunteers, they caused a diarrheal syndrome resembling that of cholera. Studies to date suggest two types of enterotoxin, a heat-labile and a heat-stable toxin. An organism may produce both toxins or the stable toxin alone. Thus far, the strains pathogenic for man have been combined toxin producers (Donta et al. 1974; Gorbach 1974).
Two nontoxigenic E. coli strains studied by DuPont et al. (1971) appeared to penetrate epithelial cells in laboratory models and produced a shigella-like illness in man which was characterized by dysentery, tenesmus, urgency, hyper pyrexia, and hypotension, with systemic toxemia. The organisms capable of producing a colitis-like syndrome include OK types 0144:K?B, 0143:K?B, 0136:K78,
389
0124:K72, and 028a,c:K73. It is interesting that all of these types except one (0136) were found to possess somatic antigens related to one shigella serotype or another. Recently, an outbreak of invasive enteropathic E. coli dysentery caused by 0124 serotype was reported in 28 adults (Tulloch et al. 1973).
CLINICAL FEATURES, COURSE, AND COMPLICATIONS
The clinical picture produced by the invasive E. coli varies with the host. In its most severe form, it includes dysentery with blood, mucus, and inflammatory cells in the diarrheal stool, tenesmus, and severe systemic toxicity. The colon ap pears to be the predominant site of multiplication of the organism, and the proctoscopic appearance may be indistinguishable from that of shigellosis. The disease has been studied primarily in volunteers, and in severe cases the natural course has been interrupted by therapy with ampicillin; therefore, few data are available on complications.
The effects of infection with toxigenic strains of E. coli also reflect significant host variability. In the Vietnam era, few documented cases of diarrhea caused by toxigenic strains were described. The spectrum of symptoms may range from a mild diarrhea lasting 2 to 5 days to a fulminant cholera-like illness with 5 to 10 watery stools per day for up to 19 days. The incubation period appears to be somewhat longer (24 to 30 hours) than for acute diarrheal syndromes (10 to 12 hours).
LABORATORY DIAGNOSIS
Because of limited laboratory capability, some selection of patients to be studied was necessary. In general, any patient with a bloody diarrhea, a temperature over 101? F, or diarrhea persisting more than 48 to 72 hours war ranted study by proctoscopy, microscopic examination, and culture. A useful but underused technique was simple examination of a rectal swab stained with methylene blue for the presence of leukocytes. Details of the technique have recently been reviewed by Harris, DuPont, and Hornick (1972). The fecal leukocyte study could be prepared at the same time as rectal scrapings were screened for the trophozoites of E. histolytica.
Thomas (1969), at the 9th Medical Laboratory, reviewed the types of media used for the identification of enteric pathogens. These fell into three categories: differential media (such as eosin-methylene blue), differential-selective media (such as Salmonella-Shigella agar), and enrichment media. Unfortunately, toxigenic E. coli are indistinguishable from "normal flora" by such conventional tests. Demonstration of pathogenicity relies on a highly complex serotyping procedure; on the production of diarrhea or toxicity in a laboratory animal or model such as the guinea pig eye model (Sereny test) or rabbit ileal loop (DuPont et al. 1971); or on the demonstration of toxin effect on adrenal cell culture, as described by Donta et al. (1974). Most of these techniques were too complex for use even by the 9th Medical Laboratory and were clearly beyond the capability of other laboratories in-country, thus rendering the diagnosis one of exclusion.
390
PREVENTION AND TREATMENT
Prevention rests on the exercise of good sanitary discipline, an often unattainable goal under field conditions. Nevertheless, since the most significant outbreaks in Vietnam occurred in base camp situations, a responsible attitude by both command and medical staff is essential. There is an urgent need for the development of a vaccine, but in the interim proper sanitation and adequate nutrition are the key preventive measures.
Rehydration and symptomatic relief remain the basis of rational therapy. The use of balanced electrolyte solutions and glucose orally, as in cholera, may reduce or eliminate the requirement for parenteral fluids. Although antibiotics have been used to terminate apparently severe infection with invasive strains of E. coli in volunteers experimentally infected, there is little evidence for their efficacy and they should not be used routinely. There is no role for mass oral antibiotic prophylaxis or for the use of furazolidone or sulfonamide preparations.
NEW ADVANCES AND LESSONS LEARNED
Perhaps the key lesson learned was the definition and distinction of the two clinical syndromes related to pathogenic E. coli. The studies with the toxigenic strains of E. coli and cholera did much to clarify the mechanism for transport of fluid, glucose, and electrolytes across the small bowel mucosa and have provided a rational basis for symptomatic therapy. The recognition of a dysenteric phase places an even greater responsibility on the physician to correctly identify the patient with massive fluid loss.
In a study of the clinical records of 2,050 admissions to the 3d Platoon, 568th Medical Company (Clearing) between November 1966 and October 1967, Scott, Ardison, and Wells (1967) demonstrated a lack of correlation between acute diarrheal disease and urolithiasis, in contrast to the situation in the diarrhea of chronic inflammatory bowel disease.
McCloy and Hofmann (1970) investigated the possibility that tropical diarrhea might be bile-salt induced. In a clinical trial conducted at the 8th Field Hospital, Nha Trang, 31 patients were studied, who were receiving either 4 g of cholestyramine, a bile sequestering resin, four times a day for 4 consecutive days, or a placebo on a similar schedule. A majority of the patients still had diarrhea at the end of the 4-day study period. There was no significant difference of any sort between the treated and placebo groups.
Much remains to be learned about diarrhea in troops. From what little is known, less than half the cases can be attributed to a specific agent. To fully comprehend the problem, it is crucial to determine the days lost from duty in cases of known and unknown etiology. Attention should be focused on controlling and treating those disorders of most significance. Statistical reporting methods must be modified to accept this new information and methodology established to provide it. The reporting of various types of "diarrheal disease" as a single entity is imprecise; further definition is required for future reporting systems.
391
REFERENCES
Afifi, A. M.; Adnan, M.; and El Garf, A.A.1976. Amoxycillin in treatment of typhoid fever in patients with hematological contraindications to chloramphenicol. Brit. M. J.2: 1033-34.
Anderson, E. S., and Smith, H. R.1972.Chloramphenicol resistance in the typhoid bacillus. Brit. M. J.3: 329-31.
AR 40-5-Department of the Army. 1969.Preventive medicine. Army Regulation No. 40-5, chap. 5, sec. II, 13Mar.69.
Archampong, E. Q. 1969. Operative treatment of typhoid perforation of the bowel. Brit. M. J. 3: 273-76.
Aserkoff, B., and Bennett, J. V. 1969. Effect of antibiotic therapy in acute salmonellosis on the fecal excretion of salmonellae. New England J. Med. 281: 636-40.
Banwell, J. G., and Sherr, H. 1973. Effect of bacterial enterotoxins on the gastrointestinal tract. Gastroenterology65:467-97.
Barker, W. H., Jr. 1974. Vibrioparahemolyticus outbreaks in the United States. Lancet 1: 551-54.
Barnes, J., and Moylan-Jones, R. J. 1966.Travellers` diarrhea: An epidemiological study. Report, Army Personnel Research Committee, London.
Bauer, H. 1973. Growing problem of salmonellosis in modern society. Medicine 52: 323-30.
Benenson, A. S. 1971.The control of cholera. In Symposium on cholera. Bull. New York Acad. Med. 47:1204-10.
Bernard, R. P. 1965. The Zermatt typhoid outbreak in 1963. J. Hyg. 63: 537-63.
Bloom, H. 1944. Dysentery in British prisoners of war. Lancet 2: 558-60.
Boyd, W. 1965. Pathology for the physician. 7th ed. Philadelphia: Lea & Febiger.
Bulmer, E.1944. A survey of tropical diseases as seen in the Middle East. Tr. Roy. Soc. Trop. Med. & Hyg. 37:225-42.
Carpenter, C. C. J. 1971a. Cholera: Diagnosis and treatment. In Symposium on cholera. Bull. New York Acad. Med.47: 1192-1203.
Carpenter, C. C. J., Jr. 1971b. Choleraentero toxin-recent investigations yield insights into transport processes. Am. J. Med 50: 1-7.
Carpenter, C. C. J.; Mitra, P. P.; and Sack, R. B. 1966. Clinical studies in Asiatic cholera. I. Preliminary observations, November 1962-March 1963. Bull. Johns Hopkins Hosp. 118:165-73.
CDC-M&M-Center for Disease Control, Department of Health, Education and Welfare. 1972. Morbidity and Mortality Weekly Report 21: 177-78,27 May 72.
CDC-Center for Disease Control, Department of Health, Education and Welfare. 1973. Shigella surveillance. Report no. 33, Apr. 73.
Chalmers, I. M. 1971. Typhoid fever in an endemic area: A "great imitator." South African M. J. 45:470-72.
Chow, E. A.; Dimaio, C. M.; Farkas, J. R.; and Franger, J. E. 1971. Changing therapy in bacillary dysentery. USARVM. Bull. (USARV Pam 40-27), May-June, p. 41. Copy in Joint Medical Library, Office of the Surgeons General.
Colquhoun, J., and Weetch, R. S.1950. Resistance to chloramphenicol developing during treatment of typhoid fever. Lancet 2: 621-23.
Commander, U.S. Army Medical Command, Vietnam. 1969. Army Medical Department Activities Report to The Surgeon General. On file at U. S. Army Center of Military History.
Communicable and other diseases, The Medical Department of the U.S. Army in the World War. See MD-WW9.
Communicable diseases transmitted chiefly through respiratory and alimentary tracts, Preventive Medicine in World War II See MD-PM4.
DA-FM- Department of the Army. 1957. Military sanitation. DA Field Manual 21-10, chap. I, sec. II, 6 May 67.
Davis, C. E., and Anandan, J. 1970. The evolution of "R" factor. A study of a "pre antibiotic" community in Borneo. New England J. Med 282: 117-22.
Democratic Republic of Viet-Nam. North Viet-Nam, Health Data Publication. See HD-25.
Donta, S. T.; Sack, D. A.; Wallace, R. B. DuPont, H. L.; and Sack, R. B. 1974. Tissue-culture assay of antibodies to heat-labile Escherichia coli enterotoxins. New England J. Med 291: 117-21.
DuPont, H. L., and Hornick, R. B. 1969. Diarrheal diseases. Disease-a-Month, July 69, pp. 14-16. .
______1973a. Adverse effect of Lomotil therapy in shigellosis. J.A.M.A. 226: 1525-28.
______. 1973b. Clinical approach to infectious diarrheas. Medicine 52: 265-70.
DuPont, H. L.; Formal, S. B.; Hornick, R. B.; Snyder, M. J.; Libonati, J. P.; Sheahan, D. G.; LaBrec, E. H.; and Kalas, J. P. 1971. Pathogenesis of Escherichia coli diarrhea. New England J. Med 285: 1-9.
DuPont, H. L.; Hornick, R. B.; Snyder, M. J.; Libonati, J. P.; Formal, S. B.; and Gangarosa, E. J. 1972a. Immunity in shigellosis. I. Response of man to attenuated strains of Shigella J. Infect. Dis. 125: 5-11.
DuPont, H. L.; Hornick, R. B.; Snyder, M. J.; Libonati, J. P.; Formal, S. B.; and Gangarosa, E. J. 1972b. Immunity in shigellosis. II. Protection induced by oral live vaccine or primary infection. J. Infect. Dis. 125:12-16.
Elliott, H. L.; Carpenter, C. C. J.; Sack, R.B.; and Yardley, J. H. 1970. Small bowel morphology in experimental canine cholera. Lab. Invest. 22: 112-20.
El Ramli, A. H. 1950. Chloramphenicol in typhoid fever. Lancet 1: 618-20.
El-Rooby, A., and Gohar, M. A. 1956. The effect of chloramphenicol on the agglutin in titre in enteric fevers. J. Trop. Med. 59: 47-51.
Erlich, J.; Bartz, Q. R.; Smith, R. M.; Joslyn, D. A.; and Burkholder, P. R. 1947. Chloromycetin, a new antibiotic from a soil actinomycete. Science 106: 417.
Field, M.1971. Intestinal secretion: Effect of cyclic AMP and its role in cholera. New England J. Med 284:1137-44.
Finkelstein, R. A., and Lo Spalluto, J.J. 1969. Pathogenesis of experimental cholera. Preparation and isolation of choleragen and choleragenoid. J. Exper. Med. 130: 185-202.
Gaines, S., and Nhu-Tuan, N. T. 1968. Types and distribution of bacterial entero pathogens associated with diarrhea in Vietnam. Mil Med. 133: 114-27.
Gangarosa, E. J.1971. The epidemiology of cholera: Past and present. In Symposium on cholera. BulL New York Acad Med. 47:1140-51.
Gangarosa, E. J. 1972. Report to Interscience Conference on Antimicrobial Agents and Chemotherapy, Nov. 72, at Atlantic, N.J.
Gangarosa, E. J.; Perera, D. R.; Mata, L. J.;Mendizabal-Morris, C.; Guzman, G.; and Reller, L. B. 1970. Epidemic Shiga bacillus dysentery in Central America. II. Epidemiologic studies in 1969. J. Infect Dis. 122: 181-90.
Garfinkel, B. T.; Martin, G. M.; Watt, J.;Payne, F. J.; Mason, R. P.; and Hardy, A. V. 1953. Antiobiotics in acute bacillary dysentery: Observations in 1,408 cases with positive cultures. JA.M.A. 151:1157-59.
Gemski, P., Jr.; Takeuchi, A.; Washington,0.; and Formal, S. B. 1972. Shigellosis due to Shigelladysenteriae. 1.Relative importance of mucosal invasion versus toxin production in pathogenesis. J. Infect. Dis. 126: 523-30.
Gezon, H. M.1966. Special report on the visit to Cairo and S.E. Asia in August and September, 1965. Report, Armed Forces Epidemiological Board, Commission on Enteric Infections, 22 Apr 66.
Gorbach, S. L. 1974. Scatologic aspects of Escherichiacola: New England J. Med 291: 150-51.
Gorbach, S. L.; Banwell, J. G.; Jacobs, B.; Chatterjee, B. D.; Mitra, R.; Bugham, K.L.; and Neogy, K. N. 1970. Intestinal microflorain Asiatic cholera. II. The small bowel. J. Infect. Dis. 121:38-45.
Gordon, J. E., ed. 1965. Control of communicable diseases in man. 10th ed. New York: The American Public Health Association.
Gordon, R. S., Jr.; Feeley, J. C.; Greenough, W. B., III; Sprinz, H.; and Oseasohn, R. 1966. Cholera. Combined clinical staff conference at the National Institutes of Health. Ann. Int. Med. 64: 1328-51.
Grady, G. F., and Keusch, G. T. 1971.Pathogenesis of bacterial diarrheas. Part I. New England J. Med. 285: 831-41.
Gruber, M., and Durham, H. E. 1896. Eine neuemethode zur raschen erkennung des choleravibrio and destyphusbacillus. Munchen. med. Wchnschr. 43: 285-86.
Harris, J. C.; DuPont, H. L.; and Hornick, R.B. 1972. Fecal leukocytes in diarrheal illness. Ann. Int. Med. 76:697-703.
HD-25-Health Data Publication No. 25(Revised). The Democratic Republic of Viet-Nam. North Viet-Nam. Walter Reed Army Institute of Research, Walter Reed Army Medical Center, Oct. 66.
Heggers, J. P., and Smith, C. D. 1967.Sensitivity of shigellae to various antiobiotics. USARV M. Bull. (USARV Pam40-3), May-June, p. 53. Copy in Joint Medical Library, Office of the Surgeons General.
Hendrix, T. R. 1972. Cholera: New lessons from an old disease. Viewpoints Digest. Dis. 4 (May).
Hone, F. R.; Keogh, E. V.; and Andrew, R.1942. Bacillary dysentery in an Australian hospital in the Middle East.M. J. Australia 1: 631-35.
Hooper, W. L.; Barrow, G. I.; and McNab, D.J. N. 1974. Vibrio parahaemolyticus food-poisoning in Britain.Lancet1: 1100-2.
Hornick, R. B.; Greisman, S. E.; Woodward, T. E.; DuPont, H. L; Dawkins, A. T.; and Snyder, M. J. 1970a. Typhoid fever: Pathogenesis and immunologic control. Part I. New England J. Med. 283: 686-91.
___ 1970b. Typhoid fever: Pathogenesis and immunologic control. Part II. New England J. Med. 283:739-46.
Huckstep, R. L. 1962. Typhoid fever and other salmonella infections. London: E. & S. Livingstone.
Kalas, Lt. Col. John P., MC, and Bearden, Maj. James H., MC. 1969. Trip report to Vietnam, 18 February 1969 to 8June 1969; to evaluate the problem of gastroenteritis in combat troops at the hospital battalion aid station, and platoon (or company) level. Report, 25 Jul. 69.
La Brec, E. H.; Schneider, H.; Magnani, T.J.; and Formal, S. B.1964. Epithelial cell penetration as an essential step in the pathogenesis of bacillary dysentery. J. Bact. 88: 1503-18.
Levine, M. M.; DuPont, H. L.; Formal, S. B.; Hornick, R. B.; Takeuchi, A.; Gangarosa, E. J.; Snyder, M. J.; and Libonati, J. P. 1973. Pathogenesis of Shigella dysenteriae 1 (Shiga) dysentery. J. Infect. Dis. 127: 261-70.
Li, F. W. P. 1963. Surgical treatment of typhoid perforation of the intestine. Brit. J. Surg. 50:976-79.
Marmion, D. E. 1952. The treatment of typhoid fever with chloramphenicol. A clinical study of 330 cases of enteric fever treated in Egypt. Tr. Roy. Soc. Trop. Med. & Hyg. 46:619-38.
Martin, D. G.; Tong, M. J.; Ewald, P. E.; and Kelly, H. V. 1970. Antiobiotic sensitivities of shigella isolates in Vietnam, 1968-1969. Mil. Med135: 560-62.
Mata, L. J.; Gangarosa, E. J.; Caceres, A.;Perera, D. R.; and Mejicanos, M. L. 1970. Epidemic Shiga bacillus dysentery in Central America. I. Etiologic investigations in Guatemala,1969. J. Infect. Dis. 122:170-80.
McCloy, R. M., and Hofmann, A. F. 1970.Tropical diarrhea in Vietnam -a controlled study of cholestyramine therapy. New England J. Med. 284:139-40.
MD-PM4-Medical Department, U.S. Army. 1958. Communicable diseases transmitted chiefly through respiratory and alimentary tracts. Internal Medicine in World War II, vol. IV. Washington: Government Printing Office.
MD-WW9-Medical Department, U.S. Army. 1928. Communicable and other diseases. The Medical Department of the U.S. Army in the World War, vol. IX. Washington: Government Printing Office.
Mess sanitation, USARV Regulation. See USARV Reg.
Military Assistance Command, Vietnam. 1973.U.S. military personnel in South Vietnam, by month, by service. Report, 7 Dec. 73.
Military sanitation, Department of the Army Field Manual. See DA-FM.
Mitsuhashi, S. 1969. Review: The R factors. J. Infect. Dis. 119: 89-100. Morbidity and Mortality, Center for Disease Control. See CDC-M&M.
Morgan, H. R.1965. The enteric bacteria. In Bacterial and mycotic infections of man, eds. R. J. Dubos and J. G. Hirsch, pp. 610-48. Philadelphia: J. B. Lippincott Co.
NCDC-ShS-National Communicable Disease Center, Department of Health, Education and Welfare. 1969. Shigella surveillance. Report no. 18, 3 Mar. 69.
NCDC-SS-National Communicable Disease Center, Department of Health, Education and Welfare. 1967. Salmonella surveillance. Annual summary.
Neumann, D. A.; Benenson, M. W.; Hubster, E.; Tuan, N. T. N.; and Le-Tien-Van. 1972. Vibrio parahemolyticus in the Republic of Vietnam. Am. J. Trop. Med. 21: 464-66.NIS-VSY-National Institute of Statistics, Republic of Vietnam. 1971. Vietnam statistical yearbook,vol.17.
Office of the Surgeon General, Department of the Army. Health of the Army, May 1965, May 1966, May 1967, May 1968, May 1969, May 1970. Copies at Uniformed Services University of the Health Sciences.
Ogawa, H.; Nakamura, A.; Nakaya, R.; Mise, K.; Honjo, S.; Takasaka, M.; Fujiwara, T.; and Imaizumi, K. 1967.Virulence and epithelial cell invasiveness of dysentery bacilli. Japanese J. M. Sc. & Biol. 20: 315-28.
Osler, Sir William. 1935. The principles and practice of medicine, rev. T. McCrae.12th ed. New York: D. Appleton-Century Co.
PAD-Patient Administration Division, Health Services Command, Department of the Army. Individual Medical Records (IMR), 1965-70.
Patient Administration Division, Health Services Command. See PAD.
Paul, H. 1952. The control of communicable diseases. London: Harvey & Blythe.
Peffers, A. S. R.; Bailey, J.; Barrow, G. I.; and Hobbs, B. C. 1973. Vibrio parahemolyticus gastroenteritis and international air travel. Lancet 1: 143-45.
Philbrook, F. R.; Barnes, L. A.; McCann, W.J., Jr.; and Harrison, R. R. 1948. Prolonged laboratory observations on clinical cases and carriers of "Shigella flexneri III" following epidemic. U.S. Naval M. Bull. 48: 405-14.
Phillips, R. A. 1966. Cholera in the perspective of 1966. Ann. Int. Me& 65: 922-30. Preventive medicine, Army Regulation. See AR 40-5.
Reed, W.; Vaughan, V. C.; and Shakespeare, E.0.1900. Abstract of report on the origin and spread of typhoid fever in U.S. military camps during the Spanish war of 1898. Washington: Government Printing Office.
Rowe, B.; Taylor, J.; and Bettelheim, K. A.1970. An investigation of travellers` diarrhoea. Lancet 1: 1-5.
Rowland, H. A. K. 1961. The treatment of typhoid fever. J. Trop. Med. 64: 101-11.
Rubenstein, A. D.; Feemster, R. F.; and Smith, H. M. 1944. Salmonellosis as a public health problem in wartime. Am. J. Pub. Health 34: 841-53.
Salmonella surveillance, National Communicable Disease Center. See NCDC-SS.
Saphra, I. 1950.Fatalities in salmonella infections. Am. J. M. Sc. 220: 74-77.
Saphra, I., and Winter, J. W. 1957. Clinical manifestations of salmonellosis in man; an evaluation of 7,779 human infections identified at the New York Salmonella Center. New England J. Me & 256: 1128-34.
Schroeder, S. A. 1968. Interpretation of serologic tests for typhoid fever. J.A.M.A. 206: 839-40.
Scott, N. R.; Ardison, G.; and Wells, R. F. 1967. Urolithiasis in Vietnam. USARVM. Bull. (USARV Pam 40-6), Nov.-Dec., pp.62-66. Copy in Joint Medical Library, Office of the Surgeons General.
Sheehy, T. W. 1968. Digestive disease as a national problem. VI. Enteric disease among United States troops in Vietnam. Gastroenterology 55: 105-12.
Shigella surveillance. See NCDC-ShS(1969) and CDC (1973).
Sladen, G. E. 1973. The pathogenesis of cholera and some wider implications. Gut 14: 671-80.
Smith, H.H.; Bernier, D. W.; Bunge, F. M.; Rintz, F. C.; Shinn, R. S.; and Teleki, S. 1967. Area handbook for South Vietnam. Department of the Army Pamphlet No. 550-55, Apr. 67. Washington: Government Printing Office.
Snyderman, R.; Gewurz, H.; and Mergenhagen, S. E. 1968. Interactions of the complement system with endotoxiclipo polysaccharide. Generation of a factor chemotactic for polymorphonuclear leukocytes. J. Exper. Med. 128: 259-75.
Sprinz, H. 1969. Pathogenesis of intestinal infections. Arch. Path. 87: 556-62.
Stamps, T. J., and Wicks, A. C. B. 1972. Trimethoprim-sulfamethoxazole (Bactrim) in the treatment of typhoid fever. South African M.J. 46: 652-55.
Stone, G. D. 1968. Shigellosis in Saigon, Vietnam. USARV M. Bull. (USARV Pam 40-7), Jan.-Feb., pp. 57-58. Copy in Joint Medical Library, Office of the Surgeons General.
Stuart, B. M., and Pullen, R. L. 1946.Typhoid. Clinical analysis of three hundred and sixty cases. Arch. Int. Med 78: 629-61.
Sutherland, E. W., and Rall, T. W. 1958.Fractionation and characterization of a cyclic adenine ribonucleotide formed by tissue particles. J. Biol. Chem. 232: 1077-91.
Thomas, E. T. 1969. Laboratory aspects of diarrheal disease. USARV M. Bull (USARV Pam 40-16), Jul.-Aug., pp. 21-23. Copy in Joint Medical Library, Office of the Surgeons General.
Tigertt, W. D.1959. The initial effort to immunize American soldier volunteers with typhoid vaccine. Mil Med.124:342-49.
Tulloch, E. F., Jr.; Ryan, K. J.; Formal, S.B.; and Franklin, F. A. 1973. Invasive enteropathic Escherichiacoli dysentery. An outbreak in 28 adults. Ann. Int. Med 79:13-17.
USARV-CHR-USARV surgeons. Monthly Command Health Reports to USARV commander, 1965-70. On file at U.S. Army Center of Military History.
USARV monthly Command Health Reports. See USARV-CHR.
USARV Reg-Headquarters USARV. 1968. Mess sanitation. USARV Regulation No. 40-31, 24 Mar. 68.
Vandevelde, A. G. 1966. Hookworm epidemic in the 1st Cavalry Division. USARV M. Newsletter, Aug.-Sept., pp.48-49. Copy in Joint Medical Library, Office of the Surgeons General.
Vaughan, V. C., Jr. 1920. Typhoid fever in the American Expeditionary Forces. J.A.M.A. 74: 1074-81.
Vazquez, V.; Calderon, E.; and Rodriguez, R.S.1972. Chloramphenicol-resistant strains of Salmonella typhosa New England J. Med 286: 1220.
Vietnam statistical yearbook. See NIS-VSY.
Vivona, S.; Minh Ha, T. T.; Gibson, F. L.; and Cavanaugh, D. C. 1966. Antibiotic sensitivities of Enterobacteriaceae isolated in Vietnam. Mil Med. 131: 68-71.
Voino-Iasenetskii, M. V., and Khavin, T. N.1964. A study of intraepithelial localization of dysentery causative agents with the aid of fluorescent antibodies. Zhur. Microbiol(Moscow) 41: 98-100.
Ward, P. A. 1968. Chemotaxis of mononuclear cells. J. Exper. Med. 128:1201-21.
Wicks, A. C. B.; Holmes, G. S.; and Davidson, L. 1971. Endemic typhoid fever: A diagnostic pitfall. Quart. J. Med40: 341-54.
Widal, F. 1896. Serodiagnostic de la fivretyphoide. Semaine Med. 16: 259.
Wirts, C. W., Jr., and Tallant, E. J. 1944.Dysentery in American troops in the Middle East. Am. J. Digest. Dis. 11:252-55.
Woodward, T. E., and Smadel, J. E. 1964.Management of typhoid fever and its complications. Ann. Int. Med 60:144-57.
Woodward, T. E.; Smadel, J. E.; Ley, H. L., Jr.; Green, R.; and Mankikar, D. S. 1948. Preliminary report on the beneficial effect of Chloromycetin in the treatment of typhoid fever. Ann. Int. Med. 29:131-34.
Wright, A. E., and Semple, D. 1897. Remarks on vaccination against typhoid fever. Brit. M. J. 1: 256-59.
TB MED-Department of the Army. 1958.Amebiasis. Technical Bulletin (Medical) 159, 21 May 58.
Trent, S. C. 1963. Re-evaluation of World War II veterans with filariasis acquired in the South Pacific. Am. J. Trop. Med. 12:877-87.
Vandevelde, A. G. 1966. Hookworm epidemic in the 1st Cavalry Division USARV M. Newsletter, Aug.-Sept., pp.48-49. Copy in Joint Medical Library, Office of the Surgeons General.
Wartman, W. B., and King, B. G.1944. Early filariasis in American soldiers. Bull U.S. Army M. Dept. 76:45-51.
