Objective In treating patients with inflammatory bowel disease (IBD), how concomitant medications influence the response to infliximab is largely unexplored. We aim to evaluate whether proton pump inhibitors (PPIs) affect the response to infliximab therapy in patients with IBD.

Design Patient-level data of adult patients with moderate-to-severe IBD treated with infliximab were obtained from the Yale Open Data Access Framework. Multivariable analysis and propensity score-matched analysis were performed to assess week 30 remission rates, week 54 remission rates and hospitalisation rates in patients on infliximab therapy with and without PPI exposure.

Results Among the five randomised controlled studies, there were 147 and 889 patients on infliximab with and without PPI therapy, respectively. Patients on PPI were older, more likely to be Caucasian and were less likely to be on immunomodulator therapy. Patients on PPI were significantly less likely to achieve week 30 remission on multivariable analysis (OR 0.45, p<0.001). Following propensity score matching adjusting for baseline difference in patient characteristics, the week 30 remission rates were 30% and 49% in patients with and without PPI therapy, respectively (p<0.001). Analysing separately for disease, the findings remained statistically significant in Crohn’s disease but did not reach significance in UC. Similar results were seen with week 54 remission rates. Patients on PPI were also more likely to be hospitalised (15% vs 8%, p=0.007). Rates of adverse events such as gastroenteritis were not different between the two groups.

Conclusion In this patient-level meta-analysis of randomised controlled studies, we found that patients with IBD taking PPI were less likely to achieve remission while on infliximab therapy. The results of our study warrant further investigation into the effect of PPI on IBD outcomes and therapies.