November 2, 2021
Does use of proton pump inhibitors affect infliximab treatment response?
Use of proton pump inhibitors was associated with reduced response to infliximab among patients with Crohn’s disease. It remains unclear whether this relationship is causal, although intermediate effects on the microbiome are plausible.
Lu TX, Dapas M, Lin E, et al. The influence of proton pump inhibitor therapy on the outcome of infliximab therapy in inflammatory bowel disease: a patient-level meta-analysis of randomised controlled studies. Gut. 2021;70:2076–84; https://gut.bmj.com/content/70/11/2076
This meta-analysis extracted data from five randomized controlled studies of infliximab therapy in adults with moderate-to-severe IBD. Of those in the study, 147 received infliximab with a proton pump inhibitor (PPI) and 889 received infliximab without a concomitant PPI. Those treated with a PPI were older (43.9 years vs. 37.3 years; P<0.001), more likely to be white and less likely to receive an immunomodulator (18% vs. 36%; P<0.001). Baseline Crohn’s Disease Activity Index (CDAI) scores were similar between groups.
Pooled data revealed that 30% of those receiving infliximab with a PPI achieved remission after 30 weeks of infliximab therapy, compared to 54% of those not receiving a PPI (P<0.001). Multivariate analyses confirmed a 55% reduced likelihood of 30-week remission with PPI use (Odds Ratio [OR], 0.45, P<0.001). Analyzed according to IBD subtype, PPI use was significantly associated with a reduced risk of 30-week remission among Crohn’s disease (CD) patients (OR, 0.39; P<0.001) but not among ulcerative colitis (UC) patients.
Remission rates at 54 weeks among CD patients were 32% and 52% for PPI users and non-users, respectively (P<0.001), but were similar for UC patients.
Thirteen and 8% of CD patients using or not using PPIs, respectively, required hospitalization during the study periods, while 19% and 8% of UC patients using or not using a PPI, respectively, were hospitalized.
There were no differences in rates of adverse events among PPI users and non-users.
Study Design: Meta-analysis
Allocation: Not applicable
Level of Evidence: 1a