50 patients (age 44 ± 8.5 years, 31 males) with chronic diarrhea, weight loss and anemia diagnosed as having MAS attending Luminal Gastroenterology Clinic of our department were studied. Diagnosis of MAS was established by abnormal D-xylose test (< 1 g/5 g/5 h) with or without abnormal 24-h fecal fat estimation (≥ 7 g/d) while on fat loading by Van de Kamer's technique 12. The clinical details of these patients have been reported elsewhere 61314. No patient received antibiotics within 8 weeks preceding the study. The criteria used for diagnosis of various causes of MAS were as follows: Celiac disease, a) presence of anti-endomysial antibody, b) suggestive histology and c) response to gluten free diet; giardiasis, demonstration of trophozoite form of the organism in stool and/or duodenal biopsy; intestinal tuberculosis, a) demonstration of acid fast bacilli in intestine or extra-intestinal site and b) response to therapy; strongyloidiasis, demonstration of the parasite in intestinal biopsy and/or stool microscopy; intestinal lymphangiectasia, dilated lymphatic channels in subepithelial location in intestinal biopsy; acquired immunodeficiency syndrome by serology; hypogammaglobulinemia by serum immunoglobulin estimation; TS, a) no specific cause for MAS and, b) persistent response to tetracycline and folic acid.

The diagnoses of patients included in this study were as follows, tropical sprue (16), celiac disease (5), intestinal tuberculosis (3), panhypogammaglobulinemia (2) one of whom had strogyloidiasis, selective IgA deficiency (1), acquired immunodeficiency syndrome (1), giardiasis (2) and intestinal lymphangiectasia (1). Seven patients had SIBO secondary to either structural lesions or motility abnormality of gut (ileocolic anastomosis in 4, diverticulosis in 1, intestinal hypomotility due to scleroderma in 1 and diabetic autonomic neuropathy and hypothyroidism in 1). In 12/50 patients cause of MAS could not be ascertained. The protocol of the study was approved by the ethics committee of the institute and patients gave consent for inclusion into the study.

12 patients with constipation predominant irritable bowel syndrome (IBS) diagnosed by Rome criteria 15 were used as controls. Clinical details of these patients have been reported elsewhere 6. No patient with IBS had biochemical evidence of MAS [normal D-xylose test (≥ 1/g/5 g/5 h) and fecal fat estimation by Sudan III stain of spot stool specimen (< 10 droplets/high power field)]. Endoscopic jejunal biopsy was normal in them. None of them had received antibiotics, pro or anti-motility and anti-secretory drugs within 8 weeks preceding the study.

Jejunal aspirate was collected during jejunoscopy with a pediatric colonoscope and a catheter described earlier 13. Briefly, the catheter assembly had an inner tube that was 3 cm longer than the outer tube. A rubber obturator blocked the mouth of the outer tube (Fig. 1). The assembly was sterilized by autoclaving. On reaching the jejunum (confirmed by fluoroscopic examination in initial few patients and by length of endoscope inserted), the catheter assembly was introduced through the biopsy channel of a sterilized pediatric colonoscope. The inner tube was then pushed beyond the tip of the outer tube once the outer tube was 4–5 cm ahead of the tip of the endoscope. This led to dislodgement of the rubber obturator from the tip of the outer tube. Jejunal aspirate was collected through the inner tube with a sterile syringe, which was used for aerobic and anaerobic bacterial culture (transported in Robertson's cooked meat medium for the latter), colony count and drug sensitivity pattern.

Smears were prepared from jejunal aspirates, fixed, Gram stained and examined for presence of Gram positive and negative organisms. Bacterial species were cultured and isolated using standard techniques 16. Briefly, for aerobic culture, samples were homogenized by vortexing and then diluted serially with sterile distilled water. Dilutions from 5 × 10^-1 to 5 × 10^-4 were prepared. Aliquots of non-diluted sample and each dilution (100 μl each) were plated on blood agar and MacConkey agar. After 24 to 48-h incubation at 37°C, colonies were counted and bacterial species identified using standard techniques 1617. Dilutions were made in Robertson's cooked meat broth for anaerobic culture. Undiluted sample and each dilution (100 μl each) were subcultured on Wilkins-Chalgren agar. For anaerobic culture anaerobic jars (McIntosh) were used. For gassing the culture evacuation replacement system (Anoxomate, Mart-Netherland) was used. The standard anaerobic recipe has 3 evacuation replacement cycles. In the first evacuation phase of the standard anaerobic culture 80% of the jar volume was evacuated. The anoxomate achieves this with high accuracy due to high precision pressure sensor. In the successive replacement, the jar was filled with oxygen free gas mixture (Nitrogen). The jar was finally filled with a gas mixture containing 80%-90% nitrogen, 5%-10% hydrogen and carbon dioxide. Strains of Clostridium difficile, Clostridium perfringens, Bacteroides fragilis and gram negative cocci were used as positive control with each batch. Pseudomonas aeruginosa was used as negative control. Cultures were examined for bacterial growth after incubation for 48-h and if negative, after 5 days at 37°C in an anaerobic chamber. In case of bacterial growth, colonies were counted and bacterial species identified using standard techniques. Rogosa agar was used 16 for lactobacilli and was incubated in anaerobic conditions. Bacterial counts were expressed as logarithm of colony forming units per ml of jejunal fluid. Total bacterial colonies and count of each species were obtained.

All bacterial strains were subjected to in vitro antibiotic sensitivity test. Antibiotic discs (Himedia, Mumbai, India), namely ampicillin (10 μg), ampicillin and salbactum (10 μg each), tetracycline (30 μg), co-trimoxazole (trimethoprim / sulfamethoxazole 1.25 / 23.75 μg), erythromycin (15 μg), norfloxacin (10 μg), ciprofloxacin (5 μg) and cephalexin (30 μg) were used. We chose these antibiotics as these are orally administered and therefore, are likely to be preferred by the clinicians in the treatment of SIBO. The antibiotic discs were stored at 4°C before use. Antibiotic sensitivity test was performed by modified Stoke's method 17. Briefly, Muller Hinton's agar was prepared and its pH was adjusted to 7.3. It was poured to a depth of 4 mm on flat-bottomed 9-cm petri dishes. These were dried before use. The area of petri dish was arbitrarily divided into three parts, central area for control organisms and two areas on either side of it, for test organisms. The control organisms consisted of Escherichia coli [National Collection of Type Culture (NCTC) 10418], Staphylococcus aureus (NCTC 6571) and Pseudomonas aeruginosa (NCTC 10662) for coliforms, Staphylococcus and pseudomonads respectively. An inoculum of the test organism was prepared by emulsifying part of the growth from each of 5 similar colonies in saline. Turbidity of the suspension was adjusted to 0.5 McFarland standard spectrophotometrically at 530 nm wavelength. Overnight incubation of this inoculum produced semi-confluent growth. Sterile cotton swabs were impregnated with the test and control organisms separately. These swabs were used to inoculate the specified areas of the petri dishes with test and control organisms. A gap of 2–3 mm was left between test and control strains. Antibiotic discs were applied with light pressure on the agar surface using flamed forceps after the inoculum had dried. The petri dishes were incubated at 37°C for 16 to 18-h.

The radial width of the zones outside the antibiotic discs was measured in mm. The zones of inhibition of the test organisms were compared with those of control strains. The results were interpreted based on measurement of zone of inhibition (mm) in test organisms as compared with the controls as follows: (a) sensitive: equal to, greater than or not less than 3-mm as compared with controls, (b) intermediate: ≥ 2-mm but smaller than controls by >3-mm, (c) resistant: ≤ 2-mm.

Nominal variables were analyzed by Chi-square test with Yates' correction as applicable. Numerical variables were analyzed by Mann-Whitney U test as the data were not expected to have normal distribution. A p value of < 0.05 was considered as significant.

Gram positive cocci were found in 8/50, Gram negative bacilli in 13/50 and both Gram positive cocci and negative bacilli in 7/50 patients with MAS. In other 22 patients Gram staining did not show any bacteria.

Culture revealed growth of bacteria in 34/50 (68%) patients with MAS and 3/12 patients with IBS (p < 0.05). Total colony count ranged from 3 × 10² to 1 × 10¹⁵ colony forming unit (CFU)/ml (median 10⁵) in MAS and 100–1000 (median 700) CFU/ml in patients with IBS (p 0.003); 21/50 (42%) patients had with MAS and none of 12 with IBS had colony count ≥10⁵ CFU/ml (p < 0.05). In patients with MAS, aerobes were isolated in 34/34 and anaerobe in 1/34 with positive culture. Two bacterial species were grown in jejunal aspirate of 13 patients and three bacterial species in one patient. In the remaining 20 patients, growth of a single bacterial species was obtained. Frequency of isolation of different bacteria and colony count of each species of bacteria in patients with MAS are shown in table 1. ≥10⁵ CFU/mL bacteria grew in 6/16 patients with TS, 2/5 with celiac disease, 1/3 with intestinal tuberculosis, 1/1 with panhypogammaglobulinemia and Strongyloidiasis, 1/1 with selective IgA deficiency, all seven with SIBO secondary to structural and motility abnormality and 3/12 with MAS without a known cause.

In vitro sensitivity to commonly used orally absorbed antibiotics showed (table 2) that bacteria isolated from small bowel were more often sensitive to quinolones (ciprofloxacin and norfloxacin) than to tetracycline (39/47 and 34/47 vs 19/47, p < 0.0001 and <0.01, respectively); it is important to note that tetracycline is commonly used to treat chronic diarrhoea and malabsorption associated with bacterial contamination of small bowel 1011. There was no difference in sensitivity to ampicillin, erythromycin, co-trimoxazole with that of tetracycline (21/44, 14/22 and 24/47 vs. 19/47, p = ns). 30/44 and 25/40 bacterial species were sensitive to ampicillin and sulbactam combination and cephalexin respectively.