Breath Test Gas Patterns in Inflammatory Bowel Disease with Concomitant Irritable Bowel Syndrome-Like Symptoms: A Controlled Large-Scale Database Linkage Analysis.
Dig Dis Sci. 2019 Nov 21;:
Authors: Gu P, Patel D, Lakhoo K, Ko J, Liu X, Chang B, Pan D, Lentz G, Sonesen M, Estiandan R, Lin E, Pimentel M, Rezaie A
INTRODUCTION: Breath testing (BT) has gained interest for diagnosing small intestinal bacterial overgrowth (SIBO) in IBD patients with irritable bowel syndrome (IBS) overlap. We aim to characterize the rate of SIBO and BT gas patterns in IBD patients with IBS-like symptoms compared to non-IBD patients.
METHODS: A database of 14,847 consecutive lactulose BTs was developed from patients with IBS-like symptoms between November 2005 and October 2013. BTs were classified as normal, H2 predominant, CH4 predominant, and flatline based on criteria established from the literature. BT data linkage with electronic health records and chart review identified IBD patients along with disease phenotype, location, severity, and antibiotic response. Poisson loglinear model evaluated differences in gas patterns between the two groups.
RESULTS: After excluding patients with repeat breath tests, we identified 486 IBD and 10,505 non-IBD patients with at least one BT. Positive BT was present in 57% (n = 264) of IBD patients. Crohn’s disease (odds ratio (OR) 0.21, [95% confidence interval (CI) 0.11-0.38]) and ulcerative colitis (OR 0.39, [95% CI 0.22-0.70]) patients were less likely to produce excess CH4. IBD patients were more likely to have flatline BT (OR 1.82, [95% CI 1.20-2.77]). In IBD patients with SIBO, 57% improved symptomatically with antibiotics.
CONCLUSION: In a cohort of IBD patients with IBS-like symptoms, a high rate of patients had positive BT and symptomatic improvement with antibiotics. In IBD, methanogenesis is suppressed and flatline BT is more frequent, suggesting excess hydrogenotrophic bacteria. These findings suggest methanogenic and hydrogenotrophic microorganisms as potential targets for microbiome-driven biomarkers and therapies.
PMID: 31754993 [PubMed – as supplied by publisher]