Blackwell Science, LtdOxford, UK
JGHJournal of Gastroenterology and Hepatology0815-93192003 Blackwell Publishing Asia Pty Ltd
18
3006
Tropical sprue
UC Ghoshal et al.
10.1046/j.0815-9319.2003.03006.x
Original Article540547BEES SGML
Journal of Gastroenterology and Hepatology (2003) 18, 540–547
INTESTINAL INFECTIONS AND FUNCTIONS
Tropical sprue is associated with contamination of small bowel with
aerobic bacteria and reversible prolongation of orocecal transit time
UDAY C GHOSHAL,* UJJALA GHOSHAL, † ARCHANA AYYAGARI, † PIYUSH RANJAN,*
NARENDRA KRISHNANI, ‡ ASHA MISRA,* RAKESH AGGARWAL,* SITA NAIK § AND
SUBHASH R NAIK*
Departments of *Gastroenterology, †Microbiology, ‡Pathology and §Immunology, Sanjay Gandhi Postgraduate
Institute of Medical Sciences, Lucknow, India
Abstract
Background: In tropical sprue (TS), response to antibiotics may suggest a role for bacterial contamination of the small bowel, which is known in diseases with prolonged orocecal transit time (OCTT).
Methods: We studied 13 patients with TS (diagnosed by standard criteria) for frequency, nature and
degree of bacterial contamination of the small bowel by quantitative culture of jejunal aspirate, glucose
hydrogen breath test (GHBT), and OCTT by lactulose hydrogen breath test before and after treatment.
Twelve patients with constipation-predominant irritable bowel syndrome (IBS) and 12 healthy subjects
served as controls.
Results: Ten of 13 patients with TS had bacterial contamination compared with 3/12 with IBS (all aerobic, P < 0.05). Median colony count in TS (36 000 CFU/mL, 400 to > 100 000) was higher than IBS
(700 CFU/mL, 100–1000, P < 0.05). Gram-negative aerobic bacilli were commonly isolated in TS but
not in IBS. Median OCTT was longer in TS (180 m, 40 - 240) than IBS (110 m, 70 - 150, P = 0.008)
and healthy subjects (65 m, 40 - 110, P = 0.0007, Wilcoxon rank sum test). Orocecal transit time in TS
correlated with fecal fat (Spearman’s rank correlation coefficient 0.69, P < 0.05). Orocecal transit time
and fecal fat, repeated in 8/13 patients, decreased with treatment for TS (195 m, 130–240 vs 125 m, 90–
200, P = 0.02; 8 g/24 h, 6.8–19.6 vs 7 g/24 h, 4.2–9, P = 0.04, respectively).
Conclusion: Aerobic bacterial contamination of the small bowel is common in patients with TS. Prolonged OCTT in TS correlated with fecal fat and normalized in a subset of patients after treatment.
© 2003 Blackwell Publishing Asia Pty Ltd
Key words: breath test, gut transit time, malabsorption syndrome, small bowel aspirate, tropical
enteropathy.
INTRODUCTION
Tropical sprue (TS) is a common cause of chronic diarrhea and malabsorption syndrome in Indian adults; its
etiology and pathogenesis remain unclear. An infectious
etiology is strongly suspected because of the occurrence
of TS in epidemics, particularly in rural areas with poor
sanitation,1 susceptibility of visitors from developed
countries to endemic regions2 and a favorable response
to treatment with antibiotics.3,4 In addition, TS patients
have also been shown to have bacterial contamination of
the small bowel.5–10 In some of these reports, control
subjects had evidence of bacterial contamination of the
small bowel, raising the possibility of: (i) subclinical
tropical enteropathy, or (ii) a non-bacterial etiology for
TS, namely viral and protozoal agents.11,12 The latter,
however, cannot explain the clinical response to antibacterial agents in patients with TS.
In health, fasting small intestinal motility, the socalled housekeeper activity, is an important factor that
Correspondence: Assistant Professor UC Ghoshal, Department of Gastroenterology, Sanjay Gandhi Postgraduate Institute of
Medical Sciences, Raebareli Road, Lucknow 226014, India. E-mail: [email protected]
This paper was presented in part at the 26th International Conference of Internal Medicine at Kyoto, Japan in May 2002.
Accepted for publication 26 October 2002.
Tropical sprue
prevents bacterial colonization of the small bowel.13
Thus, conditions with poor intestinal motility, as evidenced by a prolonged orocecal transit time (OCTT),
are associated with bacterial colonization of the small
bowel.14 However, data on gut transit time in patients
with TS are scant.15,16 Of the two studies available on
this issue, one reported prolonged OCTT15 whereas the
other reported normal whole gut transit time.16 Factors
associated with prolonged OCTT, if any, in patients
with TS, are not known. Effect of treatment for TS on
OCTT has not been studied. We therefore hypothesized
that bacterial contamination of the small bowel is an
important factor in the pathogenesis of TS, and it
results from prolongation of OCTT that might be a
primary cause of this disease or could be a secondary
phenomenon. Accordingly, we decided to study the
alteration in OCTT and its relationship with jejunal
bacterial flora in patients with TS, and the effect of
treatment for TS on OCTT.
541
Controls
Twelve healthy volunteers (authors and staff members
of their institution) were studied as healthy controls and
subjected to measurement of OCTT and GHBT. They
did not undergo jejunal aspiration. Twelve patients with
irritable bowel syndrome (IBS, all constipationpredominant) diagnosed using Rome criteria,17 with
normal D-xylose test, fecal fat estimation by Sudan III
stain of a spot-stool specimen (normal £ 10 droplets/
high power field) and endoscopic jejunal biopsy served
as disease controls. All the patients with IBS included as
controls had socioeconomic status similar to that of
patients with TS, and belonged to rural (10/12) and
urban (2/12) areas of northern India. None had
received antibiotics, prokinetics, antimotility or antisecretory drugs within 8 weeks preceding the study.
They underwent GHBT and lactulose hydrogen breath
tests, and jejunal aspiration for aerobic, anaerobic culture and colony count.
METHODS
Estimation of fecal fat and daily fecal weight
Study protocol and study population
Patients presenting with chronic large volume diarrhea
to the Luminal Gastroenterology Clinic of our department were evaluated for biochemical evidence of malabsorption by the D-xylose absorption test (normal
≥ 1 g/5 g/5 h) and 72 h fecal fat (normal < 7 g/24 h) estimation. Those with abnormal results on one or both
tests underwent endoscopic jejunal biopsy using a pediatric colonoscope (Pentax, Japan). Other causes of malabsorption were excluded using various tests in each
case, for example, celiac disease by anti-endomysial
antibody, anatomic abnormalities in the small bowel by
barium study, associated acquired immunodeficiency
syndrome by serology, giardiasis by stool microscopy
and small intestinal biopsy, associated hypogammaglobulinemia by serum immunoglobulin estimation and
immunoproliferative small intestinal disease by a test for
abnormal alpha-heavy chain in the serum and by small
intestinal histology. At entry into the study, evidence of
bacterial contamination of the small bowel was sought
using bacterial culture and colony count of jejunal
aspirate, and glucose hydrogen breath test (GHBT).
Orocecal transit time was measured by a lactulose
hydrogen breath test. Our institution’s Ethics Committee approved the study protocol.
Fecal fat was estimated by Van de Kamer’s technique
after 3 days fat loading.18 Daily stool collection was performed during the next 3 days during which fat loading
(75 g/day) was continued. Average of 3 days’ measurement of stool weight and fat content were taken as daily
values.
Jejunal aspiration
We designed a catheter for jejunal aspiration based on
principles described previously.19,20 Briefly, the catheter
assembly consisted of an outer and an inner tube, and
an obturator. The inner tube was 4–5 cm longer than
the outer tube. The inner tube was housed within the
outer tube, whose tip was blocked with a rubber obturator. Pre-autoclaved catheter assembly was introduced
through the biopsy channel of a sterilized pediatric
colonoscope after the latter had been negotiated into
the jejunum, as confirmed by length of the endoscope
inserted. The tip of the assembly was advanced for 3–
4 cm beyond the tip of the endoscope. The inner tube
was then pushed, resulting in dislodgement of the obturator. Jejunal aspirate was obtained through the inner
tube and transported to the laboratory in a sterile tube
and in Robertson’s cooked meat medium for aerobic
and anaerobic cultures, respectively.
Treatment and follow-up
All patients were treated with tetracycline (500 mg
thrice daily for 1 month followed by 500 mg twice daily
for another month) and folic acid (10 mg/day) for at
least 6 months. At least 2 weeks after stopping treatment
with tetracycline, 72 h fecal fat, D-xylose, histologic
examination of endoscopic jejunal biopsy, and lactulose
and glucose hydrogen breath tests were repeated.
Patients were then followed up regularly for a period of
18.5 ± 2 months.
Microbiological analysis of jejunal aspirate
Smears prepared from jejunal aspirates were fixed,
Gram-stained and examined for presence of organisms.
Bacterial species were cultured and isolated using standard techniques.21 Briefly, for aerobic culture, samples
were homogenized by vortexing and serial dilutions
(5 ¥ 10-1 to 5 ¥ 10-4) were prepared with sterile distilled
water. Aliquots of undiluted sample and each dilution
542
UC Ghoshal et al.
(100 mL each) were plated on blood agar, MacConkey
agar and incubated at 37∞C for 48 h. For anaerobic
culture, undiluted sample and serial dilutions in
Robertson’s cooked meat broth (100 mL each) were
cultured on Wilkins-Chalgren agar and examined after
incubation for 48 h and 5 days at 37∞C in an anaerobic
chamber. In case of bacterial growth, colonies were
counted and bacterial species identified using standard
techniques.21,22 For lactobacilli, Rogosa agar was used
and incubated in anaerobic conditions. Bacterial
counts were expressed as a logarithm of colony forming
units (CFU) per mL of jejunal fluid; total bacterial
colony count and counts of each individual species were
obtained.
following glucose or lactulose administration was calculated by subtracting the fasting value from the highest
value of hydrogen excretion obtained.26 Rise of breath
hydrogen by 14 ppm above basal level following glucose
administration was taken as evidence of small intestinal
bacterial overgrowth.27 Time interval between lactulose
administration and sustained (for at least three consecutive 10 min recordings) rise of breath hydrogen by
20 ppm above basal level was considered as OCTT.28 If
two peaks were observed after lactulose administration,
the first peak was taken as evidence of small bowel
bacterial overgrowth and the second ‘colonic’ peak
(> 20 ppm) was used for measurement of OCTT.
Statistical analysis
Hydrogen breath tests
Glucose hydrogen breath test (GHBT) and lactulose
hydrogen breath test were performed on two separate
consecutive days using a breath gas analyzer (Lactoscreen H2 breath tester, Hoek Loos, Amsterdam,
Netherlands). Basal breath specimens were obtained
after an overnight fast; the subjects avoided slowly
absorbed carbohydrates (bread, potato, corn) and fiber
the previous evening to avoid delayed excretion of
hydrogen in the breath.23 Cigarette smoking and physical exercise were not permitted for 2 h before and during the test, to prevent hyperventilation and consequent
changes in breath hydrogen content.24 The subjects then
brushed their teeth, rinsed their mouths with an antiseptic wash followed by tap water, to eliminate an early
hydrogen peak due to action of oral bacteria on test sugars.25 An average of four values was taken as the basal
breath hydrogen level. Subjects then ingested 100 g glucose dissolved in 200 mL water or 15 mL lactulose solution. Thereafter, breath hydrogen was estimated every
10 min for 3 h. Lactulose hydrogen breath test was continued for up to 4–5 h if no peak was obtained. An
increase in hydrogen excretion, parts per million (ppm)
Table 1
Qualitative parameters such as presence of bacterial
contamination of the small bowel in patients with TS
and control subjects were compared by c2 test. Differences in quantitative parameters were evaluated by
Wilcoxon’s rank sum test. Spearman’s rank correlation
coefficient was used to evaluate factors affecting OCTT.
RESULTS
Of the 50 patients with malabsorption syndrome evaluated during a 14 month period (November 2000 December 2001), 13 were diagnosed as having TS and
were included in this study. Demographic, clinical
and biochemical parameters of patients with TS before
and after treatment are summarized in Table 1. One
patient presented with hypokalemic paralysis and
hypocalcemic tetany in addition to chronic diarrhea.
Healthy (n = 12, median age 30 years, range 23 - 44;
8 male) and IBS (n = 12, median age 39.5 years, range
26 - 44; 9 male) controls were similar to patients with
TS (n = 13, median age 28 years, range 19 - 46; 6 male)
in age and sex distribution.
Demographic, clinical and biochemical findings of patients with tropical sprue before and after treatment
Parameter
Age (years)
Sex (M : F)
Duration of diarrhea (years)
Stool frequency (number/day)
Hemoglobin (g/L)
Fecal weight (g/day)
D-xylose (g excreted/5 g/5 h)
Fecal fat (g/day)
OCTT (min)
Before treatment (n = 13)
After treatment (n = 13)
28 (19 - 46)
6:7
2 (0.6–4.0)
9 (4 - 15)
94 (68 - 150)
716 (350 - 1500)
0.36 (0.16 - 0.8)
8.1 (6.6 - 19.6)†
180 (40 - 240)§
–
–
–
2 (1 - 3)**
120 (70 - 140)*
550 (275 - 850)
0.8 (0.63 - 01.8)
7 (4.2 - 9)*‡
125 (90 - 200)*‡
All values are shown as median (range). § Data are for 11 patients because two patients were hydrogen non-producers.
†
Two of 13 patients had fecal fat excretion < 7 g/24 h but had abnormal D-xylose. One of them presented with hypokalemic
paralysis and hypocalcemic tetany in addition to chronic diarrhea. ‡Repeat study was performed in only eight patients.
* P < 0.05; **P < 0.001.
Tropical sprue
543
Small bowel bacterial contamination
Effect of treatment
Ten out of 13 patients with TS had contamination of the
jejunum with aerobic bacteria (Gram-negative in 8/10)
but none had contamination with anaerobic bacteria; in
contrast, only three of 12 IBS controls had evidence of
such bacterial contamination (P = 0.03, c2 test). Colony
count in patients with TS (median 36 000; range 400 to
> 100 000 CFU/mL) was higher than that in patients
with IBS (700; 100 - 1000; P < 0.05); in four patients,
the colony count was > 105 CFU/mL and three patients
had colony counts between 103 CFU/mL and 105 CFU/
mL. The other three patients grew fewer than 103 CFU/
mL of bacteria. The bacteria isolated in patients with TS
were Pseudomonas aeruginosa (2/10), Escherichia coli (2/
10), Enterococcus faecalis (2/10), Klebsiella pneumoniae
(2/10), a mixture of Staphylococcus aureus and Acinetobacter baumanii (1/10), and Streptococcus species (1/10).
Among patients with IBS, jejunal aspirate of two grew
Acinetobacter baumanii and another grew Streptococcus
species. No patient or control had a positive GHBT.
One of four patients with colony counts higher than
105 CFU/mL was a hydrogen non-producer.
All patients with TS showed symptomatic and histological responses to treatment (Table 1). Reduction of steatorrhea was observed in 8/13 patients, in whom the test
was repeated (8.0 g/24 h, 6.8 - 19.6 vs 7.0 g/24 h, 4.2 9.0; before vs after treatment, P = 0.04, Fig. 2). Treatment resulted in shortening of OCTT in 8/13 patients
in whom it was repeated (195 min, 130 - 240 vs
125 min, 90 - 200; before vs after treatment, P = 0.02,
Fig. 3). The patient with hypokalemic paralysis and
hypocalcemic tetany had normal serum potassium and
calcium levels after treatment for TS in the absence of
any supplementation during follow-up.
Two of 13 patients with TS were hydrogen non-producers; therefore, OCTT could be assessed in only 11
patients. Orocecal transit time in patients with TS
(median 180 min; range 40 - 240) was higher than that
in IBS controls (110; 70 - 150; P = 0.08) and healthy
volunteers (65; 40 –110; P = 0.0007; Wilcoxon’s rank
sum test); OCTT in patients with IBS was longer than
that in healthy subjects (P = 0.001, Wilcoxon’s rank sum
test). Orocecal transit time in patients with TS had a
positive correlation with amount of daily fecal fat excretion (r = 0.69, P < 0.05, Fig. 1) but not with fecal weight
or D-xylose excretion.
20
Fecal fat (g/24 h)
15
Among patients with TS, endoscopic jejunal biopsy
revealed subtotal villous atrophy in one, partial villous
atrophy in three, and blunting of villi with excessive
mononuclear infiltrate in mucosa and lamina propria in
nine patients before treatment. After treatment, villous
morphology improved to near normal although mononuclear infiltrate persisted. Median number of intra-
25
20
Fecal fat (g/24 h)
Orocecal transit time in patients with
tropical sprue and controls
Histological findings of endoscopic jejunal
biopsies before and after treatment
P = 0.04
15
10
10
5
5
0
0
50
100
150
Orocecal transit time (min)
200
250
Figure 1 Significant correlation between fecal fat and orocecal transit time in patients with tropical sprue (Spearman’s
r = 0.69, P < 0.05).
0
Before
After treatment
Figure 2 Quantitative fecal fat in patients with tropical
sprue (n = 8) before and after treatment with tetracycline and
folic acid for 2 months.
544
UC Ghoshal et al.
300
P = 0.02
Orocecal transit time (min)
250
200
150
100
50
0
Before treatment
After treatment
Figure 3 Orocecal transit time in patients with tropical
sprue (n = 8) before and after treatment with tetracycline and
folic acid for 2 months.
epithelial lymphocytes (IEL) was 27/100 enterocytes
(range 6–72) and 12/100 enterocytes (range 9–21)
before and after treatment, respectively (P = 0.02, Wilcoxon’s rank sum test). Villous atrophy or blunting was
not found in any patient with IBS. Median IEL count in
jejunal biopsy in patients with IBS was 8/100 enterocytes (range 3 - 17, normal £ 25/100 enterocytes).29
DISCUSSION
The present study shows that patients with TS had colonization of the small bowel with aerobic bacteria more
often than did patients with IBS, and had longer OCTT
than healthy subjects and patients with IBS. Orocecal
transit time correlated with amount of fecal fat excretion. Clinical and biochemical improvement in malabsorption with antibiotic treatment was associated with
normalization of OCTT in a subset of patients.
A previous study found OCTT to be prolonged in
patients with TS.15 Another study however, found whole
gut transit time to be similar in patients with TS and in
controls.16 Small bowel transit constitutes only 15% of
whole gut transit time.30 Therefore, whole gut transit
time may remain unchanged despite prolongation of
small bowel transit time. Our data showed that, in
patients with TS, OCTT was prolonged and that this
prolongation correlated with level of fecal fat excretion,
and reversed in some patients with decreases in fecal fat
excretion after treatment. Infusion of fat into the ileum
has been shown to inhibit jejunal motility mediated by
peptide YY31 and enteroglucagon32 in healthy subjects.
It is therefore possible that prolongation of OCTT in
some patients with TS is a secondary phenomenon
resulting from stimulation of ‘ileal brake’ mechanism by
unabsorbed fat in the intestinal lumen. Increased concentration of peptide YY in plasma of patients with TS
and chronic pancreatitis with steatorrhea as compared
to healthy subjects supports this possibility.33 Shortening of OCTT after treatment and lack of any correlation
between prolonged OCTT and other markers of severity of disease (e.g. daily stool weight and degree of
abnormality in D-xylose test) support our contention. A
previous report documenting prolongation of OCTT in
celiac disease, which is also associated with mucosal
malabsorption, and its normalization with treatment
also support our hypothesis.34 However, successful
treatment did not lead to normalization of OCTT in all
patients. We therefore believe that in some patients with
TS, prolongation of OCTT is a primary cause of this
disease; small bowel stasis resulting from prolonged
OCTT could lead to colonization of the small bowel
with ingested bacteria that commonly contaminate
foods and drinks in developing countries with poor
hygiene. This might initiate malabsorption in them
(Fig. 4). In the present study, mean OCTT in healthy
subjects was similar to that in previous reports, which
showed it to be shorter in tropical countries including
India than in residents of developed countries.35,36 A
falsely short OCTT due to small bowel bacterial overgrowth in the control group is unlikely as we excluded
small bowel bacterial overgrowth by glucose hydrogen
breath test in all cases and by jejunal aspirate culture
and colony count in the control group with IBS. Prolonged OCTT in patients with IBS as compared to
healthy subjects in the present study is in accordance
with an earlier report that showed OCTT to be prolonged in constipation-predominant IBS as compared
with healthy subjects.37
Our observation of frequent bacterial contamination
of the small bowel in TS is in accordance with previous
reports.5–9 However, in some studies, bacterial contamination of the small bowel was equally frequent in
healthy subjects and in patients with TS.5 Interestingly,
in none of these studies were adequate precautions
taken to prevent contamination of jejunal aspirate by
oropharyngeal flora during the passage of the catheter.
This might explain the frequent bacterial contamination
of the small bowel in healthy adults observed in these
studies. Gastric acid and normal small bowel motility
are known to keep the small bowel sterile in healthy subjects. Jejunal aspirate of some healthy subjects may grow
bacteria, but colony count is usually below 104 CFU/
mL38 Absence of significant bacterial contamination in
our control group may be explained by the use of a special catheter that we designed to avoid contamination of
jejunal aspirate with oropharyngeal flora. Nine of 10
healthy subjects studied as controls in another study
using a similar catheter grew no bacteria.20 The other
reason could be related to poor hygiene in the control
Tropical sprue
545
Pre-existing
prolonged OCTT in
some patients
Other primary etiological
factors ?
Ingested
bacteria
Malabsorption
SIBC
Figure 4 Proposed model of
pathogenesis of tropical sprue
based on the present study.
OCTT, orocecal transit time;
SIBC, small bowel bacterial
colonization.
Increased intraluminal fat
population in the past that might have improved over
years. The differences may also be due to geographic
reasons because most of the previous studies were from
southern parts of India. Failure to detect bacterial contamination of the small bowel with GHBT might be
related to low colony counts because GHBT has sensitivity of 75% in patients with bacterial overgrowth
syndrome in whom colony count is > 105 CFU/mL
of jejunal aspirate.27 Only 4/13 patients in this study
had colony count > 105 CFU/mL, one of whom was a
non-producer of hydrogen. In most of the previous
studies,6–10 colony counts of bacteria in jejunal aspirate
in patients with TS were not higher than 105 CFU/mL,
a level which is considered diagnostic of small bowel
bacterial overgrowth syndrome. However, bacterial
contamination of the small bowel in these previous
studies and in the present study cannot be ignored as
such bacterial contamination is clearly not the result
merely of constant exposure to a contaminated environment, as most healthy native residents of tropics do not
harbor coliform bacteria in the proximal small bowel,6,10
although some in southern India have been reported to
do so.5 Growth of > 103 CFU/mL of colonic type flora
has been regarded as significant bacterial contamination
of the small bowel.39 Other reasons for considering
such small bowel bacterial contamination as significant
include the presence of Gram-negative bacteria in 8/10
patients with TS in our study and signficant difference
in colony ocunts in them as compared with controls
residing in a similar environment.
It is possible that slowed gut transit, as evidenced by
prolonged OCTT, leads to colonization of the small
bowel with bacteria in the later stage of the disease
resulting in further exacerbation of malabsorption. It is
important to note that no particular bacterial species
was consistently isolated in patients with TS, suggesting
bacteria may be a secondary invader exacerbating malabsorption rather than a primary cause of this disease.
SIBC
Further prolongation of OCTT
Ileal brake
Bacterial contamination of the small bowel may worsen
steatorrhea and leads to further prolongation of small
bowel transit due to activation of the ‘ileal brake’, resulting in a vicious cycle. This hypothesis of pathogenesis of
TS has been proposed previously, but evidence to support it was lacking.34 We believe that our study provides
some of this evidence. However, primary factors that
initiate the disease in most patients need elucidation.
Clinical, biochemical and histological improvement
observed with antibacterial therapy in advanced stages
of the disease may be related to reduction in bacterial
colonization in addition to improvement of mucosal
architecture and absorptive capacity, and shortening of
OCTT due to amelioration of fat-induced ‘ileal brake’.
The latter in turn, may reduce the degree of small bowel
bacterial contamination by abolishing the proposed
vicious cycle (Fig. 4). This hypothesis needs to be
studied further.
Diagnosis of TS in our patients was based on standard criteria.40 Although median fecal fat excretion in
our patients was not very high, D-xylose was markedly
abnormal in all of them. Changes in mucosal morphology in our patients cannot be explained merely as
tropical enteropathy as our patients had persistent and
progressive symptoms, biochemical evidence of malabsorption, weight loss and nutritional deficiencies, and
improved with specific treatment.40–42 Small intestinal
bacterial contamination and prolonged OCTT even in
the absence of very high fecal fat excretion in our
patients led us to believe that in more severe disease
with higher fat excretion, higher OCTT and consequent increase in small bowel bacterial contamination
are expected.
In conclusion, we have shown that TS is associated
with contamination of the small bowel with aerobic bacteria and prolongation of OCTT. Clinical improvement
following antibacterial treatment normalizes OCTT in
a subset of patients.
546
UC Ghoshal et al.
ACKNOWLEDGMENTS
This work was funded by an intramural research grant
to UCG from the authors’ institution. The authors
thank the endoscopy and laboratory staff of the Department of Gastroenterology at their institution for technical help.
19
20
REFERENCES
21
1 Ayrey F. Outbreaks of sprue during the Burma campaign.
Trans. R. Soc. Trop. Med. Hyg. 1948; 41: 377–406.
2 Peetermans WE, Vonck A. Tropical sprue after travel to
Tanzania. J. Travel Med. 2000; 7: 33–4.
3 Keele KD, Bound JP. Sprue in India: a clinical survey of
600 cases. Br. Med. J. 1946; 2: 77–81.
4 Guerra R, Wheby MS, Bayless TM. Long term antibiotic
therapy in tropical sprue. Ann. Intern. Med. 1965; 63:
619–34.
5 Bhat P, Shantakumari S, Rajan D et al. Bacterial flora of
the gastrointestinal tract in southern Indian control subjects and patients with tropical sprue. Gastroenterology
1972; 62: 11–21.
6 Klipstein FA, Holdeman LV, Corcino JJ, Moore WE.
Enterotoxigenic intestinal bacteria in tropical sprue. Ann.
Intern. Med. 1973; 79: 632–41.
7 Tomkins AM, Drasar BS, James WPT. Bacterial colonization of jejunal mucosa in acute tropical sprue. Lancet
1975; 1: 59–62.
8 Applebaum PC, Moshal MG, Hift W, Chatterton SA.
Intestinal bacteria in patients with tropical sprue. S. Afr.
Med. J. 1980; 57: 1081–3.
9 Tomkins AM, Wright SG, Drasar BS. Bacterial colonization of upper intestine in mild tropical malabsorption.
Trans. R. Soc. Trop. Med. Hyg. 1980; 74: 752–5.
10 Gorbach SL, Mitra R, Jacobs B, Banwell JG, Chatterjee
BD, Mazumder DN. Bacterial contamination of the
upper small bowel in tropical sprue. Lancet 1969; 1: 74–7.
11 Mathan M, Mathan VI, Swaminathan SP, Yesudoss S.
Pleomorphic virus-like particles in human feces. Lancet
1975; 1: 1068–9.
12 Cook GC. ‘Tropical sprue’: some early investigators
favoured an infective cause, but was a coccidian involved?
Gut 1997; 40: 428–9.
13 Stotzer PO, Bjornsson ES, Abrahamsson H et al. Interdigestive and postprandial motility in small-intestinal
bacterial overgrowth. Scand. J. Gastroenterol. 1996; 31:
875–80.
14 Matsumoto T, Iida M, Hirakawa M et al. Breath hydrogen
test using water-diluted lactulose in patients with gastrointestinal amyloidiasis. Dig. Dis. Sci. 1991; 36: 1756–
60.
15 Cook GC. Delayed small-intestinal transit in tropical malabsorption. Br. Med. J. 1978; 2: 238–40.
16 Jayanthi V, Chako A, Karim G, Mathan VI. Intestinal
transit in healthy southern Indian subjects and in patients
with tropical sprue. Gut 1989; 30: 35–8.
17 Thompson WG. Irritable bowel syndrome: pathogenesis
and management. Lancet 1993; 341: 1569–72.
18 Raffensperger EC, D’Agostino F, Manfredo H, Ramirez
M, Brooks FP, O’Neill F. Fecal fat excretion: an analysis
22
23
24
25
26
27
28
29
30
31
32
33
of four years’ experience. Arch. Intern. Med. 1967; 119:
573–6.
Ghoshal UC, Ghoshal U, Jain M et al. Strongyloides stercoralis infestation associated with septicemia due to intestinal transmural migration of bacteria. J. Gastroenterol.
Hepatol. 2002; 17: 1331–33.
Bardhan PK, Gyr K, Beglinger C, Vogtlin J, Frey R,
Vischer W. Diagnosis of bacterial overgrowth after culturing small bowel aspirate obtained during routine upper
gastrointestinal endoscopy. Scand. J. Gastroenterol. 1992;
27: 253–6.
Collee JG, Miles RS, Walt B. Tests for the identification of
bacteria. In: Mackie and Mccartney Practical Medical
Microbiology. (eds. Collee, JG, Fraser, AG, Marmion, BP,
Simmons, A.) New York: Churchill Livingstone, 1996;
131–49.
Anonymous. Processing clinical specimens for anaerobic
bacteria: Isolation and identification procedures. In:
Bailey and Scott’s Diagnostic Microbiology. (eds. Baron, EJ,
Peterson, LR, Finegold, SM) Philadelphia: Mosby, 1994;
474–503.
Anderson IH, Levine AS, Levitt MD. Incomplete absorption of the carbohydrates in all-purpose wheat flour. N.
Engl. J. Med. 1981; 304: 891–2.
Thompson DG, Binfield P, De Belder A, O’Brien J.
Warren S, Wilson M. Extra intestinal influences on exhaled
breath hydrogen measurements during the investigation of
gastrointestinal disease. Gut 1985; 26: 1349–52.
Thompson DG, O’Brien JD, Hardie MJ. Influence of
the oropharyngeal microflora on the measurement of
exhaled breath hydrogen. Gastroenterology 1986; 91: 853–
60.
Barr RG, Watkins JB, Perman JA. Mucosal function and
breath hydrogen excretion: comparative studies in the
clinical evaluation of children with non-specific abdominal complaints. Pediatrics 1981; 68: 526–33.
Kerlin P, Wong L. Breath hydrogen testing in bacterial
overgrowth of small intestine. Gastroenterology 1988; 95:
982–8.
King CE, Toskes PP. Comparison of 1-gram [14C] xylose,
10-gram lactulose-H2, and 80-gram glucose-H2 breath
tests in patients with small intestinal bacterial overgrowth.
Gastroenterology 1986; 91: 1447–51.
Hayat M, Cairns A, Dixon MF, O’Mahony S. Quantitation of intraepithelial lymphocytes in human duodenum:
what is normal? J. Clin. Pathol. 2002; 55: 393–5.
Read NW, Miles CA, Fisher D et al. Transit of a meal
through the stomach, small intestine, and colon in normal
subjects and its role in the pathogenesis of diarrhoea. Gastroenterology 1980; 79: 1276–82.
Pironi L, Stanghellini V, Miglioli M et al. Fat-induced
ileal brake in humans: a dose-dependent phenomenon
correlated to the plasma levels of peptide YY. Gastroenterology 1993; 105: 733–9.
Farthing MJG. Tropical malabsorption and tropical diarrhea. In: Gastrointestinal and Liver Disease. Pathophysiology,
Diagnosis, Management. (eds. Feldman, M, Scharschmidt,
BF, Sleisenger, MH.) Philadelphia: W.B. Saunders, 1998;
1575–84.
Adrian TE, Savage AP, Bacarese-Hamilton AJ, Wolfe K,
Besterman HS, Bloom SR. Peptide YY abnormalities in
gastrointestinal diseases. Gastroenterology 1986; 90: 379–
84.
Tropical sprue
34 Chiarioni G, Bassotti G, Germani U, et al. Gluten-free
diet normalizes mouth-to-cecum transit of a caloric meal
in adult patients with celiac disease. Dig. Dis. Sci. 1997;
42: 2100–5.
35 Jain NK, Patel VP, Pitchumoni CS. Sorbitol intolerance
in adults. Prevalence and pathogenesis in two continents.
J. Clin. Gastroenterol. 1987; 9: 317–9.
36 Patel VP, Jain NK, Pitchumoni CS. Factors affecting fasting breath hydrogen levels in healthy adults: a study in two
continents. Am. J. Gastroenterol. 1986; 81: 771–3.
37 Lu CL, Chen CY, Chang FY, Lee SD. Characteristics of
small bowel motility in patients with irritable bowel syndrome and normal humans: an Oriental study. Clin. Sci.
(Lond.) 1998; 95: 165–9.
38 Toskes PP, Donaldson M. Enteric bacterial flora and
bacterial overgrowth syndrome. In: Gastrointestinal and
Liver Disease. Pathophysiology, Diagnosis, Management.
547
39
40
41
42
(eds. Sleisenger, MH, Fordtran, JS, Scharschmidt, BF,
Feldman, M.) Philadelphia: W.B. Saunders, 1993; 1106–
18.
Justesen T, Haagen Nielsen O, Jacobsen IE et al. The
normal cultivable microflora in upper jejunal fluid in
healthy adults. Scand. J. Gastroenterol. 1984; 19: 279–
82.
Klipstein FA. Tropical sprue. In: Gastrointestinal and
Liver Disease. Pathophysiology, Diagnosis, Management.
(eds. Sleisenger, MH, Fordtran, JS, Scharschmidt, BF,
Feldman, M.) Philadelphia: W.B. Saunders, 1993; 1097–
105.
Haghighi P, Wolf PL. Tropical sprue and subclinical
enteropathy. a vision for the nineties. Crit. Rev. Clin.
Laboratory Sci. 1997; 34: 313–41.
Lim ML. A perspective on tropical sprue. Curr. Gastroenterol. Rep. 2001; 3: 322–7.