Purpose: The etiology of small intestinal bacterial overgrowth (SIBO) is diverse and frequently multi-factorial. SIBO is thought to result from a disruption of the normal peristalsis of the stomach and small intestine as well as hypochlorhydria due to anti-secretory medications. Alcoholics have previously been shown to have higher rates of SIBO as diagnosed by jejunal aspirate; however, little is known regarding the relationship between moderate alcohol consumption and SIBO.
Methods: A retrospective chart review was completed for 198 consecutive patients who underwent lactulose hydrogen breath testing (LHBT) over a 24 month time period. We defined a positive LHBT by the following criteria: (a) baseline hydrogen >40ppm; (b) hydrogen increase of >20ppm within 60min; (c) baseline methane >60ppm; (d) methane increase of >10ppm within 60min; (e) fulfillment of both hydrogen and methane criteria. We reviewed demographic data including age, race, BMI, alcohol and tobacco history, medication use, comorbid medical conditions, and history of abdominal surgery. All data were analyzed using SPSS v18.0.
Results: The study included 198 patients (68% female; mean age=55), 97 of whom had a positive LHBT (49%). Any current alcohol consumption was a strong predictor of a positive LHBT (OR 2.28, 95% CI 1.28-4.08). When stratified by rate of consumption, there was a trend toward a dose response (p=0.08); however, rates of positive LHBT were higher for all levels of consumption when compared to non-drinkers. Tobacco use was not associated with an abnormal breath test, although there was a non-significant trend toward higher rates of positive LHBT for non-smokers compared to smokers (50.3% vs. 41.4%, p=0.38). Current PPI use was not associated with positive LHBT when compared to patients without a history of PPI use (47.8% vs. 50%, p=0.76); this was also true for history of any PPI use (47.2% vs. 52.1%, p=0.51). There was no association between positive LHBT and age (54.7 vs. 55.3; p=0.76), sex (females=47.8%, males=51.6%; p=0.23), BMI (27.2 vs. 27.1; p=0.87) or H2RA use (40% vs. 49.5%; p=0.56).
Conclusion: In this retrospective review, any current alcohol consumption was associated with the presence of SIBO. Neither smoking nor PPI use was associated with increased risk of SIBO. Further studies are needed to confirm the association between moderate alcohol consumption and SIBO as well as determine if alcohol cessation can be utilized in the treatment of SIBO.