Scribd Logo
Upload

Welcome to Scribd!

  • Bhat Bacterial Flora GI Tract in Patients With Tropical Sprue and Norms 562 1972

    Uploaded by

    IG
    15 views11 pages

    Original Title

    Bhat Bacterial Flora GI Tract in Patients with Tropical Sprue and Norms 562 1972

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Report this Document

    Copyright:

    © All Rights Reserved
    Download as PDF or read online from Scribd
    Flag for inappropriate content
    15 views11 pages

    Bhat Bacterial Flora GI Tract in Patients With Tropical Sprue and Norms 562 1972

    Uploaded by

    IG

    Copyright:

    © All Rights Reserved
    Download as PDF or read online from Scribd
    Flag for inappropriate content
    of 11
    * Geprln © 197 The Wim & Wine Ca BACT SOUTH 'ERIAL FLORA OF THE GASTROINTESTINAL TRACT IN ERN INDIAN CONTROL SUBJECTS AND PATIED WITH TROPICAL SPRUE P. Baar, MBS. PUD. 8. Siasraxtreste Pad, CR Karaois, MID, €. Swansaeee S.J. Baw, MD, PRACP. Wellcome Rescarch Unit, Christien Medics Tamitnade, India "ese Hospital. Vellore 4 ‘The viable bacterial flora of the gastrointestinal tract has been studied by peroral intubation in 12 eontzol southern Indian subjects and in 33 patients with tropical spruc. Eight of the control subjects had viable bacteria in the upper jeieri:m in concentrations of 10* to 10" per ml, and all had organism: taroughout the ileum (mean count 10* per mb), Patients with tropics! sorue hed similar flora in the small intestine, there being no statistically” significant differences between the patients and the concrs. In the patient group sig- nificantly more of those with a convaized jejunal mucosa had entero- hacteria and streptococci in the upper jejunum than those with leaf- shaped villi, and more of those with en abnormal barium meal had enterobacteria than those with normet N-ray findings, The individuals with sprue had significantly more enterobacteria and veillonellae in the stools than did the controls, In patients with a history of more than I-year duration, the total number of acrohes in the stoal was greater than the total number of anccrobes. There was no relution- ship between the bacteriological findings and the course of the disease or response to therapy. The presence of bacteria in the small in- testinal lumen of some patients with sprue appears to be a nonspe- fic finding and the relationship of these bacteria to the malabsorption of various substances is obscure and needs further investigation. For many years it has been postulated that Uopical sprue may be an infective disease' and the response of some eases to antimicrobial therapy®* suggests that intestinal bacteria may play a role in its pathogenesis. Search for known bacterial Received May 7, 1971, Accepted September 1 wit Address requests for teprints Wo: De, PL Bhat Welleome Resesreh Unit, Christian Medical College Hospital, Vellure 4, Tamilaada, tndia ‘This work wae supported in pact by PLSD Grant No. 01-329.01. The Wellcome Research Ut is supported by the Welleome Trust in association with the World Heatth Organization, ‘The authors wish to thank the Welleome ‘rust, Professor RE, O. Williams, and Dr, B, Drasit for 1s culture of feces, even in epi- rue, has been unrewarding.® testine have been carried out in relatively few subjects, in some cases without anaero- ie techniques, and with varying results.7* ‘This study was therefore undertaken to quantitate she different types of aerobic ‘eusbling De. Bt to gan experience in anaerobic hac. terilogy ene ‘ce their help and advice. They also wish to Dark De. 8. Gorbsch for his help und en- Mise H. ase, Me, B. T. Somue, 12 snd Mrs. N. Doraipatsdian for vale "2! help; Mr, G. Jesulian and Me. 8, ‘stical help: Me. J. Prabbakaran for ‘20d siotographie asistanee; and Me. 8 ial assistance a y 2 BHAT EP AL Vol. 62, No. and anaerobic bacterial flora of the gas- trointestinal tract in patients with tropical sprue and in control subjects. Material and Methods “Thiety.theee potions with topical sproe {s2 southern Indians and ore America) nd 12 apparently healthy. assmptometie control subjects were stodieg. [The Indian patents and the controls all belonged the poorer s0- economic group and hed a lrgele vegetarian Gietwhich as similar to that deseibiee by Rao tnd Rao."] The Ameren, a tale Pence Corps Solonteeite a good diet containing 2 large mount of animalprtein. All patients and con: trols were admitted in a metabolic ward tod tree given standard ie containing 0 060g af at per day Fall gastcenteroloie! and he mravclogial studs were carried out #0 de: Served previnsly. "8s soon as possible 3. ter adimiscion, the eubjecta inthe festing ater were intubated witha sterile radon Spmque double lamen polyvinyl tube with Ratton smal mectury hag attacked. AS the tube progressed down the intestine, speck mers wete aspirated from the stomach, the fist loop of jefnum just beyond the ligament ef Teety (7010110 cin fom the incor tet, From the lover jeu (140 £0 200 em), from the upper ilu (220 to 250 crn), and from the lower Hearn (260 to 315 em). AUeach level the frst espitate wae scared to minimize con tamination from any recital ice ensign the tube fom the. previogs level. Specimens tee collected on ice and transfered imme ately to the laboratory for processing. When bation was prlonged. fas were given by mouth afver tne tower Jejnal sample had been Stained. Sainpen of anive and a tresh spect tren of ston! were also obtained from each pa. Tent and were proceesed immediately Specimens were. tileted. in AO‘. steps from 08 to 10° pth nora sane Bach dition (D1 mi) at apread-on plates ontaining the various media Hated in table 1 Specimens were inoculated 6 rapidly 35 pos Sbisland placed immediatly ian anacete Jar fitted with a cold catalyst and continously shed with nitrogen, Stel woo! seaked with eiified copper sullate solution” was placed fn the ja, which war then closed, evacuated, fited with hydrogen, reevacuated, and finally filled with 90% hydrogen and 10% earbun di trite. The anaerobic plates were incubated at GFC for 5 days, Mlcroerophiic incubation Sos performed sing. anaerbie jars witout addition of steel wool or cold catalyst, The jars were partially evacuated and refitled with 10% CO, andl incubated at 87 C for 3 days. "The acrobie cultures were incubated at 37 C lor 1 to 2 days. "The following groups of organisms were iden: tified by standard bacteriological techniques: enterobacteria," aerobic streptocccci other than enterococei (henceforth termed strep- tococei), enterucoce!, neisseriae, staphylo cocci?" lactobacilli”? yeasts and yeastlike organisms,” fusobacteria including sphaeroph: fous, haeteroides including Bacteroides ‘meta. ninogenicus, bifidobacteria,"* _veillonellae,?* and lecithinase-pevstive clostridia, "mainly Clostridium welehii. Surface viable counts were done by the spreading method.” Re- sults were expressed as the mumber of viable organisms per milliliter of intestinal juice or per gram of feces, ‘The results of the various investigations and the bacteriological findings at all levels were entered on International Computers and Tele- printers punch cards to facilitate data analy- sis and the detection of possible correlations. Results ‘The control subjects all excreted less than 6 g of fat per day in the stool, had normal vitamin By, absorption, normal Dariuim meal findings, and 7 out of 12 had a normal xylose excretion. ‘The jejunal biopsy was normal by Western standards in 2 and in the remainder showed minor changes, such 8 increase in the depth of the glandular layer and increased cellular infiltration of the lamina propria, which occur commonly in normal southern In- dian subjects. °° ‘The patients with sprue all had steator- thea (mean, 18.0 4; range, 7 to 23 g of fat per day) and xylose malabsorption (mean, BG; range, 1 to 23% exeretion in 5 hr) and 22 had malabsorption of vitamin By. The jejunal biopsy was abnormal in all; 13 had a leaflike villous pattern, 20 a convoluted pattern, and histologically 7 had mild Changes, 12 had partial villus. atrophy, land 13 severe partial villous atrophy. Tn the saliva the pattern of bacterial flora was similar in both the control and patient groups, the predominant ot- ganisms being neisseriae (mean count 10" per ml), fusobacteria (10% per ml, streptococci (10" perm), veillonellae (10® per ml) and bacteroides (30° per ml). sfonuary 1972 Sieieay ciaerohieonainisn Sheep blind agar sith enn (25 aad) Fh! vulet agar’ ogosa SI. agar’ Reinforced clostridial agar with seomycin (25 al) egy aye with neomycin (400 gland * Veillonella ajar* with vancomycin Mieeaaerophiti o Rogosa SL agae Aerobie organisms ‘MacConkey agar** Shoop blood agree Sheep blood agar with neomycin (25 sgl) Mannitol salt "Purchase as deydrated w.eda from the DifeoLhoratores, *Parchased as dehydrated media trom Oxoid, Ltd. London. In the fasting gastric juice organisms were recovered in half of the controls and from 28 (68%) of the patients with sprue. These organisms were generally similar to those in the saliva, although two controls and 14 patients also had enterobacteria (fig. 1) Statistically, there was no significant dif: ference between the two groups when ei ther total flora or individual organisms were considered. There was also no cor relation hetween the results of the aug- mented histamine test and the number or type of organisms, In the upper jejunum (lig. 1) there were no viable organisms in’4 of the controls and 4 of the patients. The oral type flora, such as fusobacteria and nonen terococcal streptococei, were present in somewhat smaller numbers than in. the gastric juice. Half the control subjects and two-thirds of the sprue patients had enterobacteria at this level. The mean PACTERIAL FLORA OF THE GASTROINTESTINAL 1 Tam 1 Culture Media FFosobaetera, including sphaerophoras Bacteroides Bacteroides Forobacteria Bifidobaereia Bifidabacteia claninogeaieus ‘including sphserophorus Lecithinase-postive clostridia Veillonellee Lactobacilli Enterobocteria Other aerchie Gram-negative rods, ex Pseudo. Emterocove! rreptecoce fineluding enterococeld Streptococci tncudling enterococci) Staphlees-cus aureus Yeasts Michigan. nd, Other media were prepared in the log count of enterobacteria in the patients (4.3) was slightly higher than in the con- trols (3.7). However, none of these dif- ferences attain statistical significance. In the lower jejunum and upper and lower ileum (fig. 1), the number of individu- als from whom oral type flora were iso- lated decreased and the number of in- dividuals from whom enterobacteria and enterococci were isolated increased. The mean counts of these latter two groups of organisms also increased in the more distal regions. In no case was there any significant difference between the control and the patient group. In the stools (tig. 1) the total number of viable organisms recovered was similar in both roups. However, when individual organisms sere considered, the mean counts for enterobacteria and veillonellae were significantly higher in the patients than in the control subjects (P< 0.025 and Vol. 62, No. BHAT ET AL. enuary 1972 P< 0.005 respectively). In control sub- jects, anaerobic organisms always pre- dominated so that the ratio of aerobic to anaerobic organisms was always less than 1, whereas of the subjects with sprue, 12 had a ratio of 1 or more (P < 0,025). OF those sprue patients with a history of less than L-year duration only 3 out of 20 had a ratio of Lor more, whereas of those with ahistory of a year or more, 9 out of Hf had. a ratio of Lor more, a’ difference sig. | nificant at the 1% level. The reason for | this inerensed ratio is chiefly due to an in- crease in the total number of aerobic or: ganisms, particularly enterobacteria, tuther than to any decrease in the number of anaerobes, Attempts were made in the patient xroup to correlate various parameters with individual and total flora present at the different levels of the small intestine There was no correlation between the Inacteriological findings and the duration of history, the degree of malabsorption of fat, xylose or vitamin By», the urinary indiean excretion, the serum and red cell folate concentration, or the serum vi tamin By, concentration. ‘There was also no correlation between the histological ‘uppearanee of the biopsy and the bae- teriology. However, when villous archi- tecture was considered, _ significantly fewer subjects with leaflike villi had en- terohacteria (P_<_ 0,005) and_ strepto- cveci (P< 0,025) in the upper jejunum than did those with # convoluted pattera There was no similar correlation when other organisms or other levels of the in testine were considered. ‘There was also a correlation between the X-ray grading and the presence of enterobacteria in the upper jejunum—significantly more _pa- tients with abnormal barium meal findings having these organisms in the upper je jonathan those with normal X-ray Endings (P< 0.005). However, no such correlation was present when ‘other or- sanisms or other levels were considered. All the 33 patients showed sympto- ‘atic improvement during their stay in the hospital, although the changes in absorp- Fion tests, cadiologieal, and biopsy findings varied, ‘Ten subjects left the hospital BACTERIAL FLORA OF THE GASTROINTESTINAL TRACT. 6 3 to 6 weeks after the start of the study and could not be folloxed further. The re maining 23 have been followed for periods ranging ftom 2 (0 15 months and the re- sults of treatment are summarized in table Xo relationship could be detected be- tween the course of the illness and the bacteriological findings. Patient W. G., the American with tropieal sprue who hhad quite a severe clinical illness, had very few organisms in the jejunum and upper ileum (fig. 9} yet he apparently re- sponded to tetracycline (fig. 3). Within 2 days alter starting tetracyeline therapy he experienced ‘marked subjective im provement. His vitamin By, absorption returned t0 normal within 5 days, but re- lapsed slightly later. After 2 weeks his steatorrhea was greatly reduced but his xylose absorption remained abnormal. He subsequently suffered a mild relapse even while continuing tetracycline therapy but 11 months later was apparently cured, Another patient with severe illness (RM. had. significant numbers of or- gonisins throughout the small intestine (fig. 0. With no specific therapy. other than hospital diet and medication to con- ‘Toate 2. Results of treatment in 24 patients “it spre allowed for 2 to 15 months Placebo one Placebo sr flew Plecebe. ‘lie acid, and | antibiotics sfsfalti | Fic, 2. Patient W. G. The numbers of anost prevalent bocteria per milter of itestina ice st irene levels ofthe gostrointestinal tract on the Sth Alay of adrsistion. S. stomach; J. upper and lower jejunum: F exper and lower eum: F, feces, weet ke anose % sentune & £ 2». i i ae 3 am Fic. .-Cliniea) course of patient, G. Tetre ceseline was given (1g per das) during she period in Gicated at the top of the chart. Biopsy warrausl: 1, rmild changes: 2 partial villous atropine: 3. severe partial villous atrophy. ** trol the diarrhea, his general condition, his absorptive funetion, X-ray, and biop: findings improved ‘remarkably (fx. yet the bacteriological findings remained essentially unchanged. At the time of the third intubation his xylose absorption had decreased, he had gained some more weight, and the organisms were reduced jin numbers. When seen again 7 months Inter he had suffered a mild relapse, yet no organisms were recovered trom the whole of the small intestine (fiz. 4 Discussion Control subjects. The pattern of bacteria isolated from the saliva of the control sub: jects resembles that found by Richardson ‘and Jones,?* except that ther did not BHAT ET At. Vol. 62, No. 1 isolate bacteroides. These organisms have, however, been found by others in saliva’® and in the gingival crevice.” Half the controls in this study had bacteria in numbers up to 10" per ml in the fasting gastric juice. Similar findings were re- ported in normal subjects by Franklin and Skoryna.** Tn subjects living in temperate regions the upper small intestine is usually sterile or has less than 10? organisms per ml," 38-38 although occasional subjects may have 10® to. 10° organisms per ml. {ln the tower ileum, however, mean counts i } ( h i a NAS Fic, 4, Patient R, M, Bacteria of the gastroin, testinal tract depicted as in figure 2. @———®, First intubation on day 8; O———O, seeond in tubation on day 106, 2———x, third intubation fo dey 223 fourth intubation on day 405. Fic. 6 Clinical course of patient RM, Numbers indicating bier Finding #8 in figure 8.‘The arrows long the bottom indicate the days when intubation was performed Jonuary 1972 ‘of 10° or 10* organisms per ml are fre quently encountered." 2%9% Studies of the microbial flora of the small intestine of normal individuals living in the tropics are very limited. Nadel and Gardner’ sampled the upper jejunum. in 13 control subjects and found bacteria in 5 (streptococci and/or micrococei and Gram-positive bacilli). Gorbach et al." studied 13 Bengali control subjects and found no coliforms in the jejunum but in 11 out of 13 these were present in. the ileum with counts ranging from 10° to 10° per ml. In an earlier study by the same workers," out of eight controls, bac- teroides, clostridia, and anaerobic Inc- tobacilli were present in the mid or lower ileum in 2, 6, and I subjects respectively. While meaningful comparisons between results in different laboratories are diffi cult because of differences in. sampling and culturing techniques and differences in the populations studied, it is apparent. that [the control subjects’ in this. study haveltfore bacteria, and more fecal type organisms in the upper gastrointestinal tract than has been found in previous studies. The explanation for the presence of this increased flora in apparently healthy individuals with normal gastrointestinal motility is obscure. Tt has been shown that| bacteria in the environment and in foo can colonize the gastrointestinal tra Perhaps the lack of sanitation and resultant} genetal contamination of the environment (@s exemnnlified by the frequently observed| ross bacterial contamination of drinkin ‘water sources) may play a part in | this observed increase in gastrointestinal Gorbach et al." found that subjects} with malnutrition had inereased numbers of organisms, particularly coliforms, in the upper intestine and suggest that “some control patients reported by other in- vestigators were similar to our malnour- ished group, and this may explain their abnormal sinall bowel flora.” Unfortu. nately these authors gave no specifie de- tails reyarding dietary intake, but, in that their controls were stated to have been eating a “good diet,” it seems possible that they may have been eating a diet rather different from that of the ‘controls BACTERIAL FLORA OF THE GASTROINTESTINAL TRACT. in this study, who, although they had no clinical evidence of malnutrition, could certainly not be regarded as eating a good diet. It is possible that the presence of these bacteria in the upper small intestine may in some way be related to the as yet un. known factor(s) responsible for the minor morphological abnormalities of the jejunal ‘mucosa seen in these subjects and which are prevalent in many tropical regions.*. “4 Alternatively it is possible that the organisms theinselvés are responsible for histological abnormalities such as have been shown to occur in gnotobiotie animals when bacteria are introduced into the in. testinal lumen,**" However, in man the bacteria are more numerous lower down the intestine where morphological abnor- malities are ususlly less marked,** ft there. fore seems improbable that the two are directly connected. Mereover, in this small group of control subjects there was no relation between the jejunal histology and the bacteriological findings. Five of these control subjects had an. abnormal xylose absorption, but this does not seem to have been due to the bacteria in the Jejunum since there was no correlation be- tween the results of this test and the bae- teriotogical findings ‘The results obtained from the stools of the control subjects agree in general with the findings of other workers, anaerobes being predominant #53 Patients uith sprue. The studies on the patients with sprue show that many have more than 10" orgenisms per ml at all levels of the small intestine, However, ‘when compared with the control subjects there are no significant differences. be- tween the two groups, either in terms of the number of individuals with or without 8 particular organism at a given level, or in terms of the mumber of organisms found. ‘The only situation where a significant dit ference existed was in the stools—the pa- tients with sprue having significantly more enterobacteria and veillonellae than the controls. The reason for this is unknown, Dut it presumably reflects some alteration in the lumninal environment favoring the growth of these particular organisms. A 18 BHAT ET AL. similar increase in coliforms has been noted in a study of patients with regional enteritis and severe ulcerative colitis." ‘The explanation for the observed corre- lation between the presence of enterobac: teria and total streptococci in the upper jejunum and the villous architecture is un- known. The organisms may in some way be responsible for the morphological changes; the presence of a convoluted mu. cosa may encourage the proliferation of the organisms, or both may be related to some ther factor. The finding of an association between the presence of enterobacteria and radiological abnormalities is also of in- terest. Perhaps the altered peristalsis favors the proliferation of these organisms. However, it should be noted that entero. bacteria were also present in some con- trot subjects with no radiological abnor- malities. The significance of this finding is therefore difficult to interpret ‘There was no over-all correlation be- tween the bacteriological findings, the ab sorption tests, and the results of therapy Patient R, M. clearly demonstrates that, in an individual, it was not possible to re- late the absorption tests to the bacterial flora or vice versa. The apparent response of W. G. to tetracycline, in spite of the presence of only minimal intraluminal flora, raises the question of the relation- ship of tetracycline administration to the response, Since many patients undergo spontaneous remission®* the condition might have pursued a similar course even ‘without antimicrobial therapy. It may also be that the tetracycline had some function other than that mediated by its antimicro- bial effect. It is also possible that sampling the luminal contents may not necessarily sample the bacteria adherent to the intes tinal mucosa.‘ * The only available study in man showed no evidence of a microbial population adherent to the mucosa as dis. tinct from the luminal one.*" Nevertheless it is possible that in W. G. there were such organisms which were tetracycline sensi tive. Finally the tetracycline may have acted on an as yet unrecognized infectious agent. which may either be the cause of tropical sprue oF a secondary perpetuat- ing factor.”* “Most of the previous studies of intesti- Vo. 62, No.1 nal bacteria in tropieal sprue have been carried out only on jejunal aspirates and with limited bacteriological techniques. Milanes et al.** cultured jejunal aspirates from 19 patients and found no organisms in 7, coliforms in 7, and other organisms in 5. No controls were studied. Nadel and Gardner? sampled the jejunum in normal controls and in patients with sprue and found no difference between the two groups, about 60% in each group being sterile, Klipstein et al.” obtained coliforms in the jejunal aspirate in 3 out of 6 patients and Desai et al.* grew organisms in small numbers in 12 of 28 Indian patients, but neither studied control subjects. Lahti et al." studied jejunal aspirates from 29 Indian control subjects and 18 patients by limited bacteriological methods and ob- served no significant difference between the two groups, O'Brien and England, in Caucasian subjects with sprue, found that jejunal aspirates from 7 were either sterile for showed insignificant growth. Tleal aspi- rates from 5 showed counts within normal limits. These findings are similar to that in the American subject in this study, but unlike the majority of the Indian subjects. Gorbach et al.!*" carried out careful bacteriological studies of Indian patients with tropical sprue, Full details of their bacteriological findings are not given but it appears that their results in sprue pa- tients are similar to those reported here. However, their control subjects are differ- cent in that none of them had coliforms in the prosimal small intestine. ‘These au- thors also studied a group of 7 patients, designated as “protein calorie malnutri- tion" with normal fat_and vitamin By, ab- sorption, 4 of whom had coliforms in the stomach or jejunum. They conclude that there is “a spectrum of bacteriological findings” ranging from the healthy controls, with no coliforms in the upper bowel, through the malnourished group, 60% of whom had coliforms in the upper bowel, to the patients with sprue, 93% of whom had coliforms in the stomach or jejunum. Unfortunately, there are no specific de tails given regarding the dietary intake of any of the groups, but. in that the controls ‘were stated to have been eating a good diet it seems possible that they may have sorry 1972 heen eating a diet rather different from that of the patients with tropical sprue and may therefore not have been strictly comparable, From the studies reported here it is con- eluded that, although bacteria are fre- quently present in numbers of 10° per ml of more throughout the small intestine of patients with sprue, this cannot be can- sidered a specific abnormality| Tt is pos. sible that in some cases these fimminal hae. teria may’ contribute to the pathogenesis the malabsorption of tropical sprue but. itso, the way in which they do is not clear In the stagnant loop syndrome, the intes. tinal luminal bacteria may produce bile salt deconjugation which reduces the con. centration of bile salts below the critical micellar concentration and thus produces steatorrhea, However, in 20 of the patients reported in this study, qualitative bile salt studies showed evidence of deconju- xation in only 1 subject and that only in the lower ileum. The bacteria therefore cleuriy do not produce steatorrhea by this mechanism, Whether intraluminal bacteria can produce steatorrhea by some other tmechanism such as the elaboration of toxie metabolites is not known, but if so, then the findings reported here could be ex- plained either by assuming that the mu- cesal cells in the patients with tropical sprue are peculiarly sensitive to some ic metabolite(s),"" or that the patients ‘sith sprue have as yet unrecognized or anisms which are not preseut in the con- trols. In the stagnant loop syndrome bac. teria also play a role in producing vitamin Bre malabsorption, although the precise mechanism by which this occurs is not known. ‘The rapid response of the vitamin Bi. absorptive defect to antibiotie admin- Stration in some patients with tropical smue,* suzgests that bacteria are also, insome way, responsible, However, in this study there was no correlation between Ntamin By, absorption and the bacterio- Josical findings at any level of the intestine, sad the precise mechanism by which vita, ‘in Bya absorption is impaired and by “hich ie is improved following antibiotie dministeation remains conjectural Finally, it should be emphasized that ince tropical sprue may well he a syn- BACTERIAL FLORA OF THE GASTRI TESTINAL TRACT. 19 drome’ and since it shows certain varia. tions in different parts of the world,*® © these findings can only he taken as repre- sentative of tropical sprue as seen in southern India, REFERENCES nex DU: A dis 1. Gal fom on sprue and bil dar. 105 Freach JM, Gaddie R, Smith NM: ‘Tropical study of seven cases and their response ned chemotherapy, QJ Med 25:2 a TW, Peréz-Santiago Bs Antibiotic ther 2 tropieal sprue. Gastroenterology 41:208. 1st 4. Guerre R, Wheby MS, Hayless TM: Long-term savibiewe therapy in tmpieal sprue. Ann intern Med 6519-694, 1985 4. Klipstein Fa, SamloffIM, Smarth G, etal: Treat- rent /f overt and subelinical malabsorption in biel Gur 10:215-022, 1969, 5, Mether VI. Bakee SJ: The epidemiology of trop wal speee. Tropical Speue and Megaloblostic Anaem's, Welleome Trust Suidy, London, J end A Churchill, 1971 7 Nadel H. Gardner FH: Bacterotogial assay of sel secretion in tropieal speue, Am ‘Toop Med Hs 5080-659, 1956, 8, Desai iG. Parekh DV, Jecjeebhoy KN: Boe teriolosical study of smell intestinal fluid. tn tropical

    You might also like