Chapter 19
Tropical Sprue
Brigadier General Andre J. Ognibene, MC, USA, Donald Catino, M.D., Robert F. Proctor, M.D., Colonel Llewellyn J. Legters, MC, USA (Ret.), Edward J. Colwell, M.D., and Joe A. Dean, M.D.
HISTORY AND BACKGROUND
Tropical sprue has been defined as a syndrome consisting of chronic diarrhea, malabsorption of fat, xylose, and vitamin B12, megaloblastic anemia, and nonspecific intestinal morphological abnormalities, other conditions having been excluded from the diagnosis, which responds to folic acid and/or antibiotic therapy (Lindenbaum 1973). It may have been described in an Indian medical textbook written between 1300 and 600 B.C., which mentions a wasting disease with chronic diarrhea, "a weakness of the digestive fire" leading to "impaired assimilation of ingested food" (Mathan 1973). The term "sprue" was first used by Sir Patrick Manson in 1830 (Klipstein and Baker 1970); an anglicized word, it is derived from the Dutch "sprouw," and is perhaps related to the Flemish word "spruwen" meaning "to sprinkle" (Oxford English Dictionary). Manson`s description of the syndrome is remarkably similar to the now classic clinical description offered by Bahr (1915, p. 8), who characterized it as follows:
* * * a disease, essentially of the tropics. It is characterized by symptoms suggestive of a chronic affection of the alimentary tract and of the glandular organs sub serving digestion.
The disease rarely runs an acute course, usually its progress is subject to periods of quiescence and exacerbation; it may remain latent even for a number of years.
In typical cases the tongue presents at first a peculiar raw appearance, most marked at the tip and edges, due to inflammation of the fungiform papillae. This process eventually leads to atrophy of all the papillae and to an eroded condition of the entire mucous surface of the mouth. Associated with these changes small yellow aphthous ulcers often appear periodically on the tongue and the buccal mucosa, especially on the inner surface of the lower lip, the cheeks, the frenulum or the soft palate.
Flatulent dyspepsia is generally a marked feature and is accompanied by distension of the entire intestinal canal, particularly of the small intestine. This distension is only relieved by the frequent passage, especially in the early morning, of large, pale, frothy acid stools and of much flatus.
In many instances the inflammatory disturbance spreads down the oesophagus, causing great pain and difficulty swallowing; there may be, and usually is, extreme hyperaesthesia of the mouth parts, and the sense of taste is often in abeyance. In the later and profoundly anaemic stage there is a tendency to patchy pigmentation between the scapulae and on the interior aspect of the thighs.
Symptoms persisting, atrophy of all the organs of the body, particularly of the liver, and profound anaemia ensue. The disease unless promptly and properly treated ultimately proves fatal.
_____
A significant portion of this chapter is from the unpublished paper by Catino et al. (undated): Tropical sprue. Prospective studies on incidence, early manifestations, and association with abnormal bacterial flora and intestinal parasitemia, January 1967-March 1968.
444
Tropical sprue was first discussed in the English language by Hillary in 1759, when small bowel disease was generally ignored (Mathan 1973, p. 978). This attitude persisted for a century and a half; one physician, Dr. Evan Evans, is quoted by Crohn, a student of his, as stating, around 1906, "There are no diseases of the small intestine, and what is more, we know nothing about them" (Lepore et al. 1957).
Since Evans`s comment, tropical sprue has accounted for 30,000 deaths among 100,000 cases in one Indian epidemic (1960-62) alone Nathan 1973, p. 979). Although it was ignored in the United States Army volumes on internal medicine in World War II, it accounted for significant morbidity and mortality during that war, with more than 1,500 cases described in Burma (Wellcome Trust 1971, pp. 13-23) and 1,069 cases reported among Italian prisoners of war in India (Stefanini 1948).
As is true of many diseases, variations are now reported in the severity and geographic distribution of tropical sprue-like illness. These reports have led to division of the syndrome into "tropical enteropathy," "subclinical enteropathy," and "idiopathic enteropathy." However, idiopathic enteropathy is in fact tropical enteropathy occurring in nonendemic geographical regions. Further discussion, therefore, will follow the nomenclature and definition of Klipstein and Baker (1970). They proposed that, until its etiology or etiologies could be determined, "tropical sprue should be regarded as a syndrome which occurs among the indigenous population and visitors to certain tropical regions that has variable clinical and laboratory manifestations which are related, in part at least, to the duration of the disorder and to the nutritional reserves of the individual." Malabsorption of two or more unrelated substances for which no etiology can be ascertained would be included in this definition.
Behind this statement are three basic tenets which summarize the current understanding of the illness:
1. The etiology and pathogenesis of the disease are unknown. It is now generally accepted that mucosal damage, probably of an infectious nature, must occur initially, followed by secondary events such as B12 or folate deficiency which promote the illness.
2. Tropical sprue is confined to endemic tropical regions and affects both visitors and the indigenous population.
3. The phenotype of the illness varies from subclinical indolent disease to severe and fatal disease.
The subclinical enteropathy occurs in asymptomatic individuals (usually selected as control subjects) who reside in endemic areas and have a very high incidence of abnormal small bowel function and morphology. Nutritional status affects the expression of the disease (Mats et al.1972).
The diagnosis is one of exclusion. Therefore, other causes of malabsorption syndromes, such as Giardia lamblia, Strongyloides stercoralis, Capillariaphilippinensis, Coccidia, gluten-sensitive celiac disease, protein-calorie deficiency, in testinal tuberculosis, regional enteritis, Whipple`s disease, and pancreatic exocrine insufficiency, must be eliminated as possibilities.
445
PATHOLOGY
Morphological changes of the small intestinal mucosa in tropical sprue as well as in subclinical enteropathy are well described by Haghighi and Nasr (1975) in a recent review. These changes are not specific. They may include fragmented or indistinct brush border, decrease in cell height so that cuboidal epithelium replaces the normal columnar epithelium, pseudostratification and loss of polarity and regularity of epithelial cell nuclei, various degrees of transepithelial migration of lymphocytes, pallor and enlargement of the crypt epithelial cell nuclei (megalocytosis), nuclear enlargement and rounding of villous epithelial cells with alteration in position of cell nuclei, subnuclear vacuolization, subepithelial edema sometimes involving vesicle formation (Gruenhagen spaces), basophilia of the villous epithelial cytoplasm, villous core edema, villous core lymphangiectasis, loss of the delicate ruffling of the villi, and thickening of the basement membrane with a change from the tinctorial characteristics of reticulin to those of collagen fibers (figs. 82 and 83).
The total absorptive surface is decreased; however, the mucosal thickness may be normal because of shortening of the villi associated with lengthening or tortuosity of the crypts. The villous shortening has been called partial villous atrophy; however, flat mucosa is rarely a feature of tropical sprue. All of the preceding changes are more severe in the proximal small intestine; the ileum may sometimes be spared. The proximal changes probably antedate the distal abnormalities.
ETIOLOGY
The unique epidemiological features of tropical sprue have suggested a possible infectious etiology, as indicated by the following studies. During an outbreak of dengue fever among Puerto Ricans, Sheehy, Artenstein, and Green (1964) obtained two jejunal biopsies. In one of these, villi were shortened and a striking inflammatory exudate was present in the lamina propria; the other specimen had a normal appearance. No d-xylose absorption tests were performed in these patients, and no patients without dengue were included as controls. Tropical sprue is known to be endemic in Puerto Rico.?
In this study, adenovirus-4 was isolated from three asymptomatic patients with mild jejunitis and villous blunting, but polio virus and herpes simplex virus were also recovered from two of these patients. Adenovirus 2 infection has been associated with partial villous atrophy of the jejunum.* Reovirus was isolated from several members of a family with "steatorrheic enteritis" in a study by Sabin (1956). Stanley (1961) reviewed the subject of reoviruses and noted that three strains of this virus have been isolated in seven cases of steatorrhea in children. He also noted that steatorrhea is a regular feature of reovirus infection in mice. Thomas (1952) reported an epidemic of gastroenteritis with steatorrhea in children. An exhaustive search failed to reveal a bacterial or parasitic pathogen; a viral agent was suggested.
______
1 Joe A. Dean, M.D.: Personal experience.
446
Astaldi and associates (1964) were among the first to describe jejunitis and villous atrophy in "epidemic hepatitis." Partial villous atrophy and steatorrhea with normal d-xylose excretion were demonstrated by Sheehy, Artenstein, and Green (1964, p.1025) in four American soldiers with hepatitis, but no control subjects were included for comparison. Conrad, Schwartz, and Young (1964) studied 25 American soldiers who contracted hepatitis in Korea after 6 weeks to 13 months of service there. They found diffuse granulomatous inflammation of the jejunum in all subjects and partial villous atrophy in those with a prolonged clinical course. During convalescence, the crypt areas were prominent and contained many Paneth`s cells; the villi varied markedly in length and shape, and fusion of adjacent villi was apparent. The results of biopsies performed after com-
447
plete recovery were not given. The illustrations provided suggest that the villous atrophy seen could have been caused by tangential sectioning or Brunner`s gland artifact. No control subjects were included and no attempt was made to rule out tropical sprue or adult celiac disease. LIVIM (lethal intestinal virus infection of mice) in newborn mice is also associated with subtotal villous atrophy of the intestinal tract (Biggers, Kraft, and Sprinz 1964).
Bacterial infections have, in several studies, been associated with acute and chronic malabsorptive states. Among the causative agents were Salmonella, Staphylococcus, Vibrio cholerae, nonvibrio cholera organisms, and Shigella (King and Joske 1960; Alexander 1958; Lindenbaum 1965). In some cases, malabsorption persisted for a year or more after the acute infection; these studies, which
448
were done on Pakistanis, who often have asymptomatic jejunitis and malabsorption (Lindenbaum, Jamiul Alam, and Kent 1966), did not include controls.
Kent (1966) demonstrated jejunitis and villous atrophy in the jejunal mucosa of rhesus monkeys given staphylococcus enterotoxin B but attributed the change to a nonspecific reaction of the jejunum to injury. Sprinz and coworkers (1966) demonstrated nonspecific, chronic, nonsuppurative inflammation or an acute granulomatous jejunitis in typhoid fever patients; these changes disappeared during the convalescent stage of the illness. In another study, Sprinz et al. (1962) showed the jejunal histopathology and results of absorption studies in asymptomatic natives of Thailand to be identical with those in Thais with a history of cholera or other gastroenteritis. Both groups show the nonspecific changes typical of socioeconomically depressed areas of the Middle and Far East. Rubin and Dobbins (1965) apparently agree that this picture is identical to that seen in tropical sprue.
Parasitic infections have also been implicated in certain malabsorptive states. Olsson and Johnston (1969) demonstrated jejunitis and duodenitis without villous atrophy in 20 U.S. Army personnel in Vietnam with Plasmodium falciparum malaria; 8 of 10 men tested failed to absorb d-xylose normally. No follow up biopsies were done.
Necator americanus (hookworm) and Strongyloides stercoralis (stronglyoidiasis) infections have been associated with malabsorption and jejunitis with partial villous atrophy in some patients (O`Brien and England 1966; Sheehy et al. 1962; Chaudhuri and Saha 1964; Milner et al. 1965). None of these studies was well enough controlled to rule out tropical sprue or the nonspecific jejunal changes seen in tropical areas. Other investigators have noted the absence of jejunitis in ancylostomiasis (Layrisse et al. 1964), and the histological response to adequate treatment is not well documented.
Giardiasis has been found in association with nonspecific jejunitis, villous atrophy, and steatorrhea, usually in pediatric patients (Hoskins et al. 1963; Yardley, Takano, and Hendrix 1964; Amini 1963). Da Silva et al. (1964) reported five adult Brazilian patients with giardiasis and jejunitis with villous atrophy, but all had other parasitic infections as well. Studies in adult populations in underdeveloped tropical areas have not been well enough controlled to exclude commonly occurring nonspecific jejunitis, tropical sprue, and adult celiac disease.
Subtotal villous atrophy of the jejunum and steatorrhea have been reported with Isospora belliinfection. Nitrofurantoin therapy relieved the steatorrhea but produced no change in the biopsy picture (French, Whitby, and Whitfield 1964). A gluten-free diet to rule out adult celiac disease was not tried.
An unidentified species of the nematode Capillaria caused 500 cases of chronic malabsorption with 80 deaths in the Philippines. Autopsies were performed, but the nature of the jejunal lesion was not mentioned (NamruGram).
The past decades have been marked by multiple studies of intestinal microflora. Improvements in such studies have recently been made through new and more accurate anaerobic culture systems. Bacterial overgrowth in the small
449
intestine has been reported (Gorbach et al. 1970; Bhat et al. 1972). There is debate about the normal flora in the control group patients; in the series by Bhat and associates, there was no difference between the flora of healthy subjects and that of those with sprue.
The observations which follow, along with the studies just discussed, support the hypothesis that tropical sprue is an infectious disease. The beneficial effect of long-term antibiotic therapy provides the most impressive indirect evidence that antibiotic-sensitive infectious agents play an important part in the pathogenesis of the syndrome (French, Gaddie, and Smith 1956; Klipstein 1964; Guerra, Wheby, and Bayless 1965).
Epidemiologically, tropical sprue can behave like an infectious disease, as demonstrated by an outbreak reported from India (Keele and Bound 1946). It presented as an acute, explosive enteritis among British soldiers on a well balanced diet during the Burma campaign (Keele 1946), and it was observed to spread through an isolated family in a manner suggestive of an infectious disease (Mathan, Ignatius, and Baker 1966).
The tropical distribution of the illness, affecting both permanent and transient inhabitants, suggests an endemic infectious agent. If patients with mild tropical sprue are evacuated to a temperate climate, they often have remission of symptoms without antibiotic or folate treatment, suggesting that they are thus freed from continued exposure to a tropical infectious agent (Gardner 1958).
The occasional clinical presentation of tropical sprue as an explosive enteritis is compatible with enterovirus infection. Coxsackie B3 virus has been isolated in tissue cultures of rectal swab material from 1 of 50 patients with untreated, symptomatic tropical sprue. However, the same virus was isolated from 1 of 10 asymptomatic patients with treated sprue. ECHO 8 virus was isolated from 1 of 48 controls (Bayless, Guardiola-Rotger, and Wheby 1966).
Many different parasites have been recovered from the stools of patients with tropical sprue, but with no greater frequency than in control groups of the general population. Milanes and associates (1946) found no parasites in the stools of 44 percent of the sprue patients they studied.
Ashford (1930-31) isolated Candida albicans in 80 percent of sprue patients. On the other hand, Swanson, Haley, and Wheby (1965) found no evidence of fungal elements in jejunal biopsy samples from 15 patients with tropical sprue cultured on Sabouraud`s agar, nor in tissue sections stained with periodic acid Schiff, Gomori methenamine silver, or hematoxylin-eosin which were examined specifically for hyphal elements. The data to date still do not conclusively establish a specific infectious etiology for tropical sprue.
RADIOLOGY
One of the most constant radiographic features of tropical sprue is dilatation of the lumen of the small intestine, especially of the distal jejunum. Generally, the dilated loops are long and tortuous, and the valvulae conniventes are
450
451
prominent. Dilatation may also be seen in the large intestine, where it may be pronounced. Segmentation is the second most common finding; this is usually most prominent in the ileum and is best seen in more advanced cases. Another constant phenomenon inmost cases is excessive secretion in the intestinal tract (Lepore et al. 1957, pp.119-20).?
The term "moulage sign" describes the X-ray appearance of the jejunum, in which the folds appear to be entirely obliterated and the barium column resembles a tube filled with hardened wax. This picture is most frequently associated with hypersecretion and segmentation (Lepore et al. 1957, pp. 121-22). The X-ray patterns are shown in figure 84.
DIAGNOSIS AND TREATMENT
The diagnosis of tropical sprue is suggested by red cell macrocytosis, hypersegmentation of the polymorphonuclear leukocytes, and malabsorption as demonstrated by oral glucose tolerance, xylose excretion, and 72-hour fecal fat, in patients residing in tropical environments. Hypokalemia, hyponatremia, and hypochloremia are common. Hypocalcemia may be present, and hypoproteinemia is a frequent finding. IgG synthesis may be depressed. Bone marrow aspiration or biopsy may reveal a megaloblastic pattern. Anemia is generally present. Serum iron is normal to low, and serum vitamin B12 levels are usually depressed. Serum folate levels may be low but are not necessarily depressed. Macrocytosis may be demonstrated in the gastric mucosa. Small bowel biopsy will disclose the morphologic changes previously discussed. Lactase deficiency is expected, but sucrase and maltase activity may be normal. Chronic sucrosuria may occur and reflects deficiency of disaccharidase activity. The presence of steatorrhea depends upon the severity of the mucosal lesion (Spiro 1970, pp. 465-71).
Tropical sprue usually resolves when the patient leaves the tropical endemic area and returns to a temperate climate. Therapy for patients with ongoing disease should aim at managing diarrhea, correcting nutritional deficiencies, and reversing the intestinal mucosal lesion. Diarrhea can usually be controlled with Lomotil, belladonna, opium, paregoric, or bismuth subcarbonate. Attention should be paid to fluid and electrolyte balance. Specific therapy for dehydration, lactic acidosis, hypokalemia, and hyponatremia should be instituted. Appropriate therapy for megaloblastic anemia and iron deficiency is also indicated. Orally administered iron and folic acid appear to be absorbed satisfactorily; however, vitamin B12 should be given parenterally. A high-calorie and high-protein diet should be given, but unfortunately there are no clear guidelines for diet therapy in tropical sprue patients (Mathan 1973, pp. 986-87).
452
In addition to therapy with folic acid and vitamin B12, broad spectrum antimicrobial agents have been reported to be effective. Tetracycline has been used most widely and has had the most success. Erythromycin is ineffective. Lincomycin may be effective, as may some sulfonamides (Lindenbaum 1973, p. 644). A short course of antibiotics (2 weeks) is usually effective; however, therapy may be continued up to 6 months, when necessary, to achieve a higher percentage of remission (Mathan 1973, p. 987). The overall prognosis is generally good, especially in those patients repatriated to temperate zones.
VIETNAM STUDIES
Data collected by the Surgeon`s Office, U.S. Army John F. Kennedy Center for Special Warfare (Airborne), in collaboration with the Division of Medicine, WRAIR (Walter Reed Army Institute of Research), attest to the military significance of diarrheal disease and tropical sprue in operations by U.S. Army Special Forces in Southeast Asia (OS-CSW). Of the SF (Special Forces) troops studied, 56 percent experienced one or more attacks of acute diarrhea during a 6-month tour of duty; 5 percent experienced a diarrheal illness lasting longer than 2 weeks. Of randomly selected SF troops studied after return from a 6-month tour in Vietnam, 6.5 percent were found to have jejunitis associated with malabsorption. The continued deployment of troops to an area in which preliminary data showed a high incidence of acute diarrheal illness and jejunitis with malabsorption provided a unique opportunity for prospective studies of the natural history of early tropical sprue.
The U.S. Army Medical Research Team (WRAIR) Vietnam was deployed to the Special Forces operations detachment C in Can Tho, IV CTZ (Corps Tactical Zone) to study the tropical sprue syndrome (Catino et al.). Tropical sprue was known to be endemic in this area of the Mekong Delta (Colwell et al. 1968; 1971). Sixty-nine soldiers volunteered for a continuing study of intestinal absorption and small bowel mucosal abnormalities during their tour of duty. They were initially studied 1 to 2 weeks after arrival in Vietnam; all but two had normal small bowel biopsies. The importance of the further findings of the WRAIR team in these soldiers requires that they be discussed in some detail.
The relative proportion of fingerlike villi found, plotted according to the length of swallowed tubing, was progressively greater in biopsies taken from more distal sites, although there were wide individual variations. The association between the proportion of fingerlike villi and tube length was statistically significant (p<.001). Mean villous width in the 67 normal jejunal biopsies plotted against tube length showed a decrease with increase in tube length, but the association was not statistically significant (p>0.1).
Histologic sections indicated that duodenal or very early jejunal tissue with Brunner`s glands was found at tube lengths o£ 80 cm or less. The average tube length of initial jejunal biopsies was 107 cm, similar to the tube lengths used by
453
other investigators (Hirsch, Ahrens, and Blankenhorn 1956; Sprinz et al. 1962; Blankenhorn, Hirsch, and Ahrens 1955; Madanagopalan, Shiner, and Rowe 1965).
Mucosal measurements of 48 perpendicularly sectioned initial jejunal biopsies are summarized in table 84. There was close agreement between these results and those of other investigators (Madanagopalan, Shiner, and Rowe 1965; Doniach and Shiner 1957). The ratio of VH:CD (villus height to crypt depth) has been used as a measure of villous atrophy; a ratio of 4:1 is usually considered normal (Kent and Lindenbaum 1967; Swanson and Thomassen 1965). While the average VH:CD ratio was 3.59 in these subjects, the normal range was wide (2.39-5.09).
TABLE 84.-Jejunal mucosal measurements of 48 Americans on arrival in Vietnam
The results of initial d-xylose excretion tests are shown in chart 28 with d-xylose excretion plotted against 5-hour urine volume. Subjects who excreted less than 200 ml of urine in 5 hours tended to have lower levels of d-xylose excretion. Repeat testing, with the ingestion of at least 700 cc of water over the 5-hour collection period, produced larger urine volumes and normal d-xylose excretion. However, after excluding "low volume" tests, the lower limit of d-xylose excretion in initial studies was still 3.01 g (normal 6.99 + 1.99 g). The prospective approach allowed judging of each patient`s variation in d-xylose excretion against his own initial value. The 13 patients who eventually developed tropical sprue had an initial d-xylose excretion of 8.2 g (7.10-9.51 g) in 5 hours.
Qualitative and quantitative bacteriology was done on 52 patients in the initial group. The results are shown in tables 85 and 86. These are similar to other published data on normal patients (Kalser et al. 1966; Dellipiani and Girdwood 1964; Cohen et al. 1967).
There was a statistically significant association (p< 0.025) between mild or moderate jejunitis and recent ADI (acute diarrheal illness), using Fisher`s exact probability test. The two patients who had jejunitis with villous atrophy on ini tial examination had acute idiopathic gastroenteritis at the time of study. They were not considered to have had tropical sprue. Thirteen patients went on to develop tropical sprue, representing 19 percent of the total group (table 87).
454
CHART 28.-d-Xylose excretion and 5-hoururinevolume at initial examination of Americans in Vietnam
Earlier studies of Special Forces returnees from Vietnam (OS-CSW) revealed that almost all those who developed tropical sprue had been deployed in IV CTZ. When the detachment locations of the patients who developed tropical sprue during the course of this study were plotted on a map of the Mekong River Delta, it appeared that most cases occurred in the Plain of Reeds area and the U
455
TABLE 87.-Incidence of tropical sprue in Vietnam
Minh Forest. Both are low, inundated areas of sparsely inhabited swampland where conditions favor the spread of waterborne or mosquito borne diseases.
Twelve of the 13 patients with tropical sprue in the study of Catino et al. served as members of the remote Special Forces A detachments. Only one patient with sprue served at an urban B detachment headquarters; his illness was mild and abated spontaneously. The relationship of tropical sprue and service in A detachments is statistically significant (p< 0.025). Other data collected by the WRAIR team at the Dong Tam base of the 9th Infantry Division indicated that personnel at this location did not develop tropical sprue. They remained on this large post most of the time and went on 2- to 3-day combat operations, returning to the "protected" base environment.
456
There were no significant differences between the normal and sprue groups with regard to age, race, or rank. Prior duty in tropical areas did not seem to protect from or predispose to the development of tropical sprue. There was no relationship between the frequent ingestion of Vietnamese food, protein-deficient diet, scarcity of fresh vegetables in the diet, or consumption of contaminated food or water or of rancid fats, and the development of tropical sprue. Those who took army vitamins ("Non-A-Vits") were not protected; however, this preparation does not contain folic acid.
Two of the tropical sprue patients had FUO (fever of undetermined origin) during the 3-month interval before diagnosis, while none of the normal subjects had such an illness. Those with sprue had a total weight loss almost double that of the control group. In general, these 13 patients were not severely ill and would fall into Gardner`s (1958) categories of stage 1 or 2 tropical sprue. Complaints of weakness, malaise, intermittent noneffectiveness, anorexia, nausea, vomiting, dyspepsia, and flatulence were all more common in patients with sprue than in controls, but these complaints did not permit clinical diagnosis of tropical sprue. Both groups had complaints of diarrhea, but the tropical sprue group had more days of diarrhea per month and more bowel movements per day during these episodes. Those with sprue more often had a constant diarrhea with loose, fatty-looking stools. However, these characteristics were not unique, and the symptoms of individual patients varied considerably. Three of the tropical sprue patients had subclinical disease, while those with normal intestinal biopsies occasionally had a classical clinical picture of sprue (table 88).
457
The jejunal biopsies in the sprue patients were obtained at an average tube length of 99.1 cm (range 75 to 110 cm). None contained Brunner`s glands, so they must have been taken from the first portion of the jejunum, just past the ligament of Treitz. Although most jejunal mucosal specimens from sprue patients were not grossly abnormal upon examination through the dissecting microscope, fingerlike villi always appeared decreased. Ridgelike villi and/or thick leaflike villi were present in all; two specimens contained convoluted villus ridges. Only two specimens showed white, dull, poorly defined, shortened villi with increased mucus adherent to them. Therefore, although the mucosal appearance with stereoscopic microscopy was often suggestive, the diagnosis of tropical sprue was of necessity made by light microscopy.
Inflammation of the lamina propria of the villi and subvillous thickness of the mucosa was moderate to severe in most cases, though one patient had minimal changes (table 89). Branching and fusion of villi were often seen. Mucosal measurements were abnormal; they showed the characteristic decrease in total mucosal thickness, villous shortening, and some broadening, with distinct deepening of the glandular crypts. Thus, the villus: crypt ratios were low (0.9 to 2.4) and always less than the initial values.
TABLE 89.-Clinical picture and laboratory data of 13 tropical sprue patients, Vietnam
The average d-xylose excretion for the 13 tropical sprue patients was 4.3 g/5 hours. While five patients had d-xylose excretion above the "lower limit of normal" (5 g/5 hours), all of them excreted 1.2 to 4.7 g less than they had in their baseline studies; and thus these later findings may be considered abnormal. These five patients tended to have milder inflammatory (minimal to mild) and atrophic (1.4 to 2.4) changes than the patients with grossly abnormal d-xylose excretion. Six patients with normal intestinal biopsies demonstrated abnormal xylose absorption (0.93 to 3.45 g/5 hr). All of these patients had acute diarrheal illness; five of the six had 10 4 to 10 6 intestinal bacteria/cc of aspirate, but none had parasitism or enteric bacterial pathogens.
458
Only 3 of the 13 patients with sprue had low serum carotene. Carotene is stored in the liver and may take months to become depleted. The short duration of illness in these 13 patients may explain the normal serum carotene found in most cases.
?
A comparison of normal jejunal flora with the flora of the 13 sprue patients is presented in table 85. There was no significant difference between the two groups. Table 86 shows a comparison of the types of bacteria found in the jejunal aspirate of tropical sprue patients and the control population. Both groups had a predominance of swallowed oral and upper respiratory organisms; there is no evidence of a predominance of fecal flora in sprue patients. In fact, most oral and fecal organisms were more common in the group without tropical sprue.
Only 1 of the 13 patients with tropical sprue harbored an organism which might be construed as a pathogen, Bethesda baller up. The same organism was isolated from one other patient with sprue 6 months after the diagnosis was made, and from another sprue patient 4 months after diagnosis, following a spontaneous remission. Bethesda ballerup was also isolated from a patient with duodenitis and villous atrophy and one with ADI. Shigella species and Providence species were isolated from 12 patients without sprue, some of whom had enteritis at the time.
No parasites were found in the stool or jejunal aspirate of patients with sprue at the time of diagnosis. Ascaris lumbricoides was found in one patient 5 months before the development of sprue. In another patient, hookworm infection developed 5 months after the diagnosis of sprue was made; mild jejunitis and villous atrophy were still present in this patient despite folate and tetracycline therapy. Giardia lamblia was identified in two patients with diarrheal illness. G. lamblia and hookworm species were isolated from another patient who was asymptomatic. The latter three patients did not have tropical sprue.
The WRAIR team studies were undertaken following the report of Sheehy and associates (1965) in which histologic abnormalities and malabsorption had been noted in 12 percent of American servicemen returning from Vietnam. The WRAIR team found a 28-percent peak incidence in selected troops in the Mekong Delta. Despite these dramatic reports, the impact of tropical sprue upon U.S. troops in Vietnam was never fully investigated. McCloy, who served as a gastroenterologist at the 8th Field Hospital, and Hofmann, of the Mayo Clinic, suggested the existence of tropical sprue in the Nha Trang area in their treatise (1971) on tropical diarrhea. Air Force physicians at nearby Cam Ranh Bay reported 12 cases in a 1-year period (Miller et al. 1974). Although biopsies were not performed, clinical findings were extensively summarized (table 90). Laboratory abnormalities found in the 12 sprue patients, plus two patients with clinical evidence of sprue, were as follows (number with finding/number tested) (Miller et al. 1974, p. 20):
Abnormal d-xylose - 14/14 Low carotene - 3/3
Flat glucose tolerance - 9/10 Anemia - 4/0
Neutral fat in stools - 11/12 Abnormal small bowel biopsy - 8/10
Low cholesterol - 6/7
459
The presence of U.S. forces in Vietnam allowed further elucidation of the syndrome of tropical sprue. The spectrum of disease, subclinical to moderately severe, was reemphasized, as was the lack of specificity of clinical symptomatology. The overall impact on the combat effort is unknown, but understanding of diarrheal syndromes was increased and tropical sprue was shown to be a cause of non effectiveness, especially in troops operating with indigenous populations in tropical areas.
REFERENCES
Alexander, J.G. 1958. Congenital agenesis of the spleen with chronic enterocolitis in an adult. J. Clin. Path. 11: 396-98.
Amini, F. 1963. Giardiasis and steatorrhea. J. Trop. Med 66: 190-92.
Ashford, B. K.1930-31. The relation of Moniliapsilosis to tropical sprue and an evaluation of fermentation of sugar as a criterion for specificity. Porto Rico J. Pub. Health 6:310-33.
Astaldi, G.; Grandini, U.; Poggi, C.; and Strosselli, E. 1964. Intestinal biopsy in acute hepatitis. Am. J. Digest. Dis. 9: 237-45.
Bahr, P. H. 1915. A report on researches on spree in Ceylon, 1912-1914. Cambridge: University Press.
Bayless, T. M.; Guardiola-Rotger, A.; and Wheby, M. S. 1966. Tropical sprue: Viral cultures of rectal swabs. Gastroenterology 51: 32-35.
Bhat, P.; Shantakumari, S.; Rajan, D.;Mathan, V. I.; Kapadia, C. R.; Swarnabai, C.; and Baker, S. J. 1972.Bacterialflora of the gastrointestinal tract in Southern Indian control subjects and patients with tropical sprue. Gastroenterology 62: 11-21.
Biggers, D. C.; Kraft, L. M.; and Sprinz,H.1964. Lethal intestinal virus infection of mice (LIVIM). An important new model for study of the response of the intestinal mucosa to injury .Am. J. Path. 45:413-22.
Blankenhorn, D. H.; Hirsch, J.; and Ahrens, E. H.; Jr. 1955. Transintestinal intubation: Technic for measurement of gut length and physiologic sampling at known loci. Proc. Soc. Exper. Biol. & Med 88:356-62.
Catino, Capt. Donald; Proctor, Maj. Robert F.; Colwell, Capt. Edward J.; Legters, Lt. Col. Llewellyn J.; and Webb, Lt. Col. Charles R., Jr. Tropical sprue. Prospective studies on incidence, early manifestations, and association with abnormal bacterial flora and intestinal parasitemia, January 1967-March 1968. Unpublished paper, undated.
Chaudhuri, R. N., and Saha, T. K. 1964.Jejunal mucosa in hookworm disease. Am. J. Trop. Med. 13:410-11.
Cohen, R.; Kalser, M. H.; Arteaga, I.; Yawn, E.; Frazier, D.; Leite, C. A.; Ahearn, D. G.; and Roth, F. 1967.Microbial intestinal flora in acute diarrheal disease. J.A.M.A. 201:835-40.
Colwell, E. J.; Welsh, J. D.; Boone, S. C.; and Legters, L. J. 1971. Intestinal parasitism in residents of the Mekong Delta of Vietnam. Southeast Asian J. Trop. Med. Pub. Health 2:25-28.
Colwell, E. J.; Welsh, J. D.; Legters, L. J.; and Proctor, R. F. 1968. Jejunal morphological characteristics in South Vietnamese residents. JA.M.A. 206: 2273-76.
Conrad, M. E.; Schwartz, F. D.; and Young, A.A.1964. Infectious hepatitis -a generalized disease. A study of renal, gastrointestinal and hematologic abnormalities. Am. J. Med 37:789-801.
Da Silva, J. R.; Coutinho, S. G.; Dias, L.B.; and De Figueiredo, N. 1964. Histopathologic findings in giardiasis: Abiopsy study. Am. J. Digest. Dis. 9: 355-65.
Dellipiani, A. W., and Girdwood, R. H. 1964.Bacterial changes in the small intestine in malabsorptive states and inpernicious anemia. Clin Sc. 26: 359-74.
Disease and morbidity associated with Special Forces operations in Vietnam, report, Center for Special Warfare. See OS-CSW.
Doniach, I., and Shiner, M. 1957. Duodenal and jejunal biopsies. II. Histology. Gastroenterology 33:71-86.
French, J. M.; Gaddie, R.; and Smith, N. M. 1956. Tropical sprue: A study of seven cases and their responses to combined chemotherapy. Quart J. Med. 25: 333-51.
French, J. M.; Whitby, J. L.; and Whitfield, A. G. W. 1964. Steatorrhea in a man infected with coccidiosis (Isospora bellil. Gastroenterology 47: 642-48.
Gardner, F. H. 1958. Tropical sprue. New England J. Med. 258: 791-96, 835-42.
Gorbach, S. L.; Banwell, J. G.; Jacobs, B.; Chatterjee, B. D.; Mitra, R.; Sen. N. N.; and Guha Mazumder, D. N. 1970. Tropical sprue and malnutrition in West Bengal. Am. J. Clin Nutrition 23:1545-58.
Guerra, R.; Wheby, M. S.; and Bayless, T. M.1965. Long-term antibiotic therapy in tropical sprue. Ann Int. Med 63:619-34.
Haghighi, P., and Nasr, K. 1975. Tropical sprue: Subclinical and idiopathic enteropathy. Pathology Annual 10:177-203.
Hirsch, J.; Ahrens, E. H., Jr.; and Blankenhorn, D. H. 1956. Measurement of the human intestinal length invivo and some causes of variation. Gastroenterology 31: 274-84.
Haskins, L. C.; Winawer, S. J.; Gottlieb, L.S.; Broitman, S. A.; and Zamcheck, N. 1963. Pathogenetic features of malabsorption accompanying intestinal giardiasis. Clin Res. 11: 184.
Kalser, M. H.; Cohen, R.; Arteaga, I.; Yawn,E.; Mayoral, L.; Hoffert, W. R.; and Frazier, D. 1966. Normal viral and bacterial flora of the human small and large intestine. New England J. Med. 274:500-505,558.
Keele, K. D. 1946. A study of the onset and cyclic development of the sprue syndrome. Brit. M.J. 2: 111-14.
Keele, K. D., and Bound, J. P. 1946. Sprue in India: A clinical survey of 600 cases. Brit. M.J. 1: 77-81.Kent, T. H.1966. Staphylococcal enterotoxin gastroenteritis in rhesus monkeys. Am. J. Path. 48: 387-98.
Kent, T. H., and Lindenbaum, J.1967.Correlation of jejunal function and morphology with acute and chronic diarrhea in East Pakistan. Gastroenterology 52: 972-84.
King, M. J., and Joske, R. A. 1960. Acute enteritis with temporary intestinal malabsorption. Brit. M.J. 2:1324-27.
Klipstein, F. A. 1964. Antibiotic therapy in tropical sprue. The role of dietary folic acid in the hematologic remission associated with oral antibiotic therapy. Ann Int. Med. 61:721-28. Klipstein, F. A., and Baker, S. J. 1970. Regarding the definition of tropical sprue. Gastroenterology 58:717-21.
Layrisse, M.; Blumenfeld, N.; Carbonell, L.;Desenne, J.; and Roche, M. 1964. Intestinal absorption tests and biopsy of the jejunum in subjects with heavy hookworm infection. Am. J. Trop.Med.13:297-305.
Lepore, M. J.; Almy, T.; Marshak, R. H.; Lattes, R.; and Porter, M. R. 1957. Panel discussion on diseases of the small intestine. Am. J. Gastroenterology 28: 113-40.
Lindenbaum, J. 1965. Malabsorption during and after recovery from acute intestinal infection. Brit. M.J.2:326-29.
Lindenbaum, J. 1973. Tropical enteropathy. Gastroenterology64: 637-52.
Lindenbaum, J.; Jamiul Alam, A. K. M.; and Kent, T. H. 1966. Subclinical small-intestinal disease in East Pakistan. Brit. M.J. 2: 1616-19.
Madanagopalan, N.; Shiner, M.; and Rowe, B.1965. Measurements of small intestinal mucosa obtained by peroral biopsy. Am. J. Med. 38: 42-53.
Mata, L. J.; Jim6nez, F.; Cordon, M.; Rosales, R.; Prera, E.; Schneider, R. E.; and Viteri, F. 1972.Gastrointestinalflora of children with protein-calorie malnutrition. Am. J. Clin. Nutrition 25: 1118-26.
Mathan, V. I. 1973. Tropical sprue. In Gastrointestinal disease: Pathophysiology, diagnosis, management, ed. M. H. Sleisenger and J. S. Fordtran, pp. 978-88. Philadelphia: W. B. Saunders Co.
Mathan, V. I.; Ignatius, M.; and Baker, S. J. 1966. A household epidemic of tropical sprue. Gut 7:490-96.
McCloy, R. M., and Hofmann, A. F. 1971.Tropical diarrhea in Vietnam-a controlled study of cholestyramine therapy. New England J. Med. 284: 139-40.
Milanes, F.; Curbelo, A.; Rodriguez, A.;Kouri, P.; and Spies, T. D.1946. A note on bacteriological and parasitic studies of the intestinal contents of patients with sprue. Gastroenterology7: 306-13.
Miller, M. B.; Loftus, P. M.;Lohr, D. C.; Reynolds, R. D.; Bratton, J. L.; Hanson, J. P.; and Keil, P. G. 1964. Tropical sprue in Vietnam. Mil.Med. 139: 17-20.
Milner, P. F.; Irvine, R. A.; Barton, C. J.; Bras, G.; and Richards, R.1965. Intestinal malabsorption in Strongyloidesstercoralis infestation. Gut 6: 574-81.
NamruGram-U.S. Naval Medical Research Unit No. 2. NamruGram 1: 1-7, 1 Sept. 1967.
O`Brien, W., and England, M. W. J. 1966. Military tropical sprue from Southeast Asia. Brit. M.J. 2:1157-62.
Olsson, R. A., and Johnston, E. H. 1969. Histopathologic changes and small-bowel absorption in falciparummalaria. Am. J. Trop. Med. 18: 355-59.
OS-CSW-Office of the Surgeon, U.S. Army John F. Kennedy Center for Special Warfare (Airborne). 1965. Interim report, Disease and morbidity associated with Special Forces operations in Vietnam. Report, 12 Apr. 65. On file at U.S Army Center of Military History.
Rubin, C. E., and Dobbins, W. 0.1965. Peroral biopsy of the small intestine. A review of its diagnostic usefulness. Gastroenterology 49: 676-97.
Sabin, A. B. 1956. The significance of viruses recovered from the intestinal tracts of healthy infants and children. Ann. New York Acad Sc. 66: 226-30.
Sheehy, T. W.; Artenstein, M. S.; and Green, R. W. 1964. Small intestinal mucosa in certain viral diseases. JA.M.A. 190: 1023-28.
Sheehy, T. W.; Cohen, W. C.; Wallace, D. K.; and Legters, L. J. 1965. Tropical sprue in North Americans. JA.M.A. 194: 1069-76.
Sheehy, T. W.; Meroney, W. H.; Cox, R.S., Jr.; and Soler, J. E. 1962. Hookworm disease and malabsorption. Gastroenterology 42: 148-56.
Spiro, H.M. 1970. Clinical gastroenterology. London: Collier-Macmillan.
Sprinz, H.; Gangarosa, E. J.; Williams, M.; Hornick, R. B.; and Woodward, T. E. 1966. Histopathology of the upper small intestines in typhoid fever. Am. J. Digest Dis. 11: 615-24.
Sprinz, H.; Sribhibhadh, R.; Gangarosa, E.J.; Benyajati, C.; Kundel, D.; and Halstead, S.1962. Biopsy of small bowel of Thai people. With special reference to recovery from Asiatic cholera and an intestinal malabsorption syndrome. Am. J. Clin. Path. 38: 43-51.
Stanley, N. F. 1961. Reovirus-a ubiquitous orphan. M. J. Australia 2: 815-18.
Stefanini, M. 1948. Clinical features and pathogenesis of tropical sprue. Observations on a series of cases among Italian prisoners of war in India. Medicine 27: 379-427.
Swanson, V. L., and Thomassen, R. W.1965.Pathology of the jejunal mucosa in tropical sprue. Am. J. Path. 46:511-36.
Swanson, V. L.; Haley, L. D.; and Wheby, M.S. 1965. Mycological study of jejunal biopsy specimens from patients with tropical sprue. Am. J. Trop. Med 14: 1066-68.
Thomas, M. E. M.1952. "Epidemic" abdominal colic associated with steatorrhea. Brit. M. J. 1: 691-92.
Wellcome Trust, The. 1971. Tropical sprue and megaloblastic anaemia Wellcome Trust Collaborative Study, 1961-1969. London: Churchill Livingstone.
Yardley, J. H.; Takano, J.; and Hendrix, T.R. 1964. Epithelial and other mucosal lesions of the jejunum in giardiasis. Jejunal biopsy studies. Bull, Johns Hopkins Hosp. 115: 389-406.
