Proton pump inhibitors alter the composition of the gut microbiota

Abstract

Objective: Proton pump inhibitors (PPIs) are drugs used to suppress gastric acid production and treat GI disorders such as peptic ulcers and gastro-oesophageal reflux. They have been considered low risk, have been widely adopted, and are often over-prescribed. Recent studies have identified an increased risk of enteric and other infections with their use. Small studies have identified possible associations between PPI use and GI microbiota, but this has yet to be carried out on a large population-based cohort.

Design: We investigated the association between PPI usage and the gut microbiome using 16S ribosomal RNA amplification from faecal samples of 1827 healthy twins, replicating results within unpublished data from an interventional study.

Results: We identified a significantly lower abundance in gut commensals and lower microbial diversity in PPI users, with an associated significant increase in the abundance of oral and upper GI tract commensals. In particular, significant increases were observed in Streptococcaceae. These associations were replicated in an independent interventional study and in a paired analysis between 70 monozygotic twin pairs who were discordant for PPI use. We propose that the observed changes result from the removal of the low pH barrier between upper GI tract bacteria and the lower gut.

Conclusions: Our findings describe a significant impact of PPIs on the gut microbiome and should caution over-use of PPIs, and warrant further investigation into the mechanisms and their clinical consequences.

Keywords: GASTRIC ACID; INTESTINAL MICROBIOLOGY; LACTIC ACID BACTERIA; LACTOBACILLUS; PROTON PUMP INHIBITION.

Figure 1

Distributions of covariates included for analysis, compared between proton pump inhibitor (PPI) users and non-users. Wilcoxon rank sum tests were carried out to compare the distribution of covariates in the modelling analysis. All variables were on a different scale so were centred and scaled before plotting. PPI users were older, frailer, had higher body mass index (BMI) and lower scores on the high alcohol food frequency questionnaire (FFQ) principle component (PC). Significant differences are indicated where ***p<0.001 and **p<0.01.

Figure 2

Comparison of α diversity in proton pump inhibitor (PPI) users and non-users, and in individuals with and without GI indications. Four metrics of α diversity were calculated on rarefied samples and one-tailed Wilcoxon rank sum tests carried out to test for significantly lower diversity with PPI use, or with GI indication. Significantly lower diversity was observed for all metrics in PPI users versus non-users (*); no difference was found splitting by GI indication (NS).

Figure 3

Summary of taxonomic associations with proton pump inhibitor (PPI) use. Shown are all collapsed groups used in taxonomic association analyses that had complete taxonomic assignment (not including collapsed species). Connecting lines highlight the taxonomic relationships between groups (not considering genetic relatedness). Taxa significantly associated with PPI use are highlighted with circles, larger circles representing a larger absolute coefficient of association. Association analyses were carried out at each taxonomic level independently. Taxa at higher abundance with PPI use are shown in blue and at lower abundance in red. Lines connecting taxa of similar association are also coloured and weighted by the average coefficient between the taxa. Names are shown for significant results only. Those in bold retained significance between 70 discordant monozygotic (MZ) twins, and underlined taxa replicated in analysis of interventional study data.

Figure 4

Paired plots of the relative abundances of Streptococcaceae and Lactobacillales within monozygotic (MZ) twins discordant for proton pump inhibitor (PPI) use. These were the only collapsed family and order traits found to be significantly different in discordant MZ twin pairs, both higher in abundance in PPI users.