Objectives: Small intestinal bacterial overgrowth (SIBO) is considered a frequent cause of abdominal symptoms in patients with surgically altered intestinal anatomy or dysmotility conditions and is recognized as a contributing factor in the exacerbation of irritable bowel syndrome symptoms. Diagnostic testing can be used to detect the condition.
Methods: The study group comprised patients who had breath hydrogen and methane lactulose challenge testing. All of the patients were treated with antibiotic regimens that have shown benefit for SIBO. The lactulose challenge was administered orally at 15 g. Hydrogen and methane in expired air were measured and hydrogen values were recorded as the hydrogen plus twice the methane result. Breath tests were analyzed for positivity based on single and multiple criteria of fasting baseline elevation, early rise, and second peak hydrogen rise. Global improvement of gastrointestinal symptoms was assessed by telephone contact or record review.
Results: One hundred participants (78 women) were included in the analysis. The mean age was 51 years. A total of 15% of participants did not meet any criteria on breath testing for SIBO; 73% had a fasting baseline elevation >10 ppm; and 19% had an increase of >20 ppm above baseline in the first 20 minutes, 48% had a 20-ppm increase in the first 60 minutes, and 32% had a second increase, reflecting a colon peak. Overall, 74% responded to a course of antibiotics, determined by reported improvement in more than half of the symptoms within 3 months. In total, 67% (10/15) of the subjects who tested negative for SIBO, by all criteria, had a favorable response to antibiotics. Among those with positive hydrogen increases, 76.3% with 1 criterion responded, 66.7% with 2 criteria responded, 84.6% with 3 responded, and 76.9% with 4 responded. After antibiotic treatment, 88.7% (47/53) of those who had diarrhea reported improvement, 63.3% (19/30) with constipation reported improvement, and 74.3% (52/70) with baseline bloating experienced improvement. Sixty-five of the 82 patients with pain or discomfort reported improvement (79.3%). Of those treated with a rifaximin regimen, 74.2% (49/66) reported a response to treatment. Twenty of 28 (71.4%) treated with amoxicillin/clavulanate experienced a clinical response.
Conclusions: These data suggest that the response to antibiotic therapy in patients with suspected SIBO is not predicted by carbohydrate SIBO testing results and bring into question the value of such testing.