Proton Pump Inhibitor Use Predisposes to Small Bowel Bacterial Overgrowth in Patients with Post-abdominal Surgery

Abstract

Purpose: To study the association between proton pump inhibitor (PPI) use and SIBO in patients with prior abdominal surgery involving the loss of the ileocecal valve, or gastric by-pass with creation of a blind loop when compared to patients with an intact gastrointestinal tract. Methods: Six hundred and seventy eight (509 female) patients underwent a glucose breath test (GBT) for evaluation of SIBO. Demographic data, list of medical diagnosis, etiology of GP if applicable, and information regarding PPI use was recorded. During the GBT, two breath samples were obtained at baseline and then at 30, 45, 60, 75 and 90 minutes following ingestion of 50 g of glucose in 150 cc water (isosmotic solution). The criteria for a positive GBT were: 1) A baseline hydrogen or methane concentration of > 15parts per million (ppm), 2) When baseline hydgrogen or methane was <10ppm, a rise during the test reached a peak that exceeded 20 ppm, or 3) A doubling of the baseline methane concentration at any time during the test (measured by Quintron MicroLyzer DP; Milwaukee, WI). Patients with co-existing DM, celiac sprue, cirrhosis, chronic pancreatitis, narcotic analgesic use, and scleroderma were excluded from the analysis. Results: Sixty two patients (48 female) with prior abdominal surgery and 350 (265 female) with an intact gastrointestinal tract met the inclusion criteria. Twenty six (42%) patients in the abdominal surgery group and 149 (43%) in the intact gastrointestinal tract group were on PPI therapy preceding the GBT. Thirty four (55%) patients in the abdominal surgery group, and 103 (29%) in the intact gastrointestinal tract group had a positive GBT. PPI users were more likely to have a positive GBT in the abdominal surgery group (OR: 3.2; 95% CI: 1.1-9.4). In contrast, PPI therapy did not predispose patients with an intact gastrointestinal tract (OR: 0.7; 95% CI: 0.4-1.1) to SIBO. Conclusion: 1) PPI use predisposes to SIBO in patients with loss of the ileocecal valve, or gastric by-pass with creation of a blind loop; 2) The added antisecretory property of a PPI combined with host defense impairments related to surgical alterations would explain this observation; 3) The clinical message is to alert physicians to the propensity for SIBO when PPI's are used in these surgical settings.