Effects of Proton Pump Inhibitors on the Small Bowel and Stool Microbiomes

Proton pump inhibitor (PPI) use is extremely common. PPIs have been suggested to affect the gut microbiome, and increase risks of Clostridium difficile infection and small intestinal bacterial overgrowth (SIBO). However, existing data are based on stool analyses and PPIs act on the foregut.

To compare the duodenal and stool microbiomes in PPI and non-PPI users.

Consecutive subjects presenting for upper endoscopy without colonoscopy were recruited. Current antibiotic users were excluded. Subjects taking PPI were age- and gender-matched 1:2 to non-PPI controls. Subjects completed medical history questionnaires, and duodenal aspirates were collected using a validated protected catheter. A subset also provided stool samples. Duodenal and stool microbiomes were analyzed by 16S rRNA sequencing.

The duodenal microbiome exhibited no phylum-level differences between PPI (N = 59) and non-PPI subjects (N = 118), but demonstrated significantly higher relative abundances of families Campylobacteraceae (3.13-fold, FDR P value < 0.01) and Bifidobacteriaceae (2.9-fold, FDR P value < 0.01), and lower relative abundance of Clostridiaceae (88.24-fold, FDR P value < 0.0001), in PPI subjects. SIBO rates were not significantly different between groups, whether defined by culture (> 10³ CFU/ml) or 16S sequencing, nor between subjects taking different PPIs. The stool microbiome exhibited significantly higher abundance of family Streptococcaceae (2.14-fold, P = 0.003), and lower Clostridiaceae (2.60-fold, FDR P value = 8.61E-13), in PPI (N = 22) versus non-PPI (N = 47) subjects.

These findings suggest that PPI use is not associated with higher rates of SIBO. Relative abundance of Clostridiaceae was reduced in both the duodenal and stool microbiomes, and Streptococcaceae was increased in stool. The clinical implications of these findings are unknown.