(BSPOT25) Proton Pump Inhibitors Use and Risk of Inflammatory Bowel Disease in Children: A Population-based Cohort Study

Background: There is growing evidence about the potential adverse effects of proton pump inhibitors (PPIs), along with various mechanisms through which they affect gut and immune functions. However, despite the dramatic rise in PPI use in children over the past decade, their potential adverse effects are largely unknown in the pediatric population.

Objectives: To assess whether PPIs are associated with a higher risk of inflammatory bowel disease (IBD) compared with H2 receptor antagonists (H2RAs) in children.

Methods: Using Korea’s nationwide healthcare database (2002-2020), we conducted a new-user, active comparator cohort study by identifying children aged 6-17 years who initiated a PPI or H2RA during 2003-2014. Based on an intention-to-treat approach, patients were followed from cohort entry (the first date of prescription of study drugs) until the earliest of outcome occurrence, death, or end of the study period, applying a 2-year lag period to account for protopathic bias and diagnostic delays. The primary outcome was an IBD diagnosis, and secondary outcomes were the type of IBD: Crohn’s disease (CD) and ulcerative colitis (UC). To control potential confounders, we used standardized morbidity ratio weighting using calendar time-specific propensity score and estimated hazard ratios (HRs) with 95% CIs using a weighted Cox proportional hazards model.

Results: Of the 8.7 million children, we included 33,710 PPI users and 2,751,592 H2RA users, with a mean age of 11.8 years and 49.8% male. During a mean follow-up of 7.9 years, there were 77 and 6,708 IBD events in PPI and H2RA users, respectively (weighted incidence rate: 2.91 vs 2.16 per 10,000 person-years). PPI was associated with a higher risk of IBD (weighted HR 1.37, 95% CI 1.09-1.72), including CD (1.16, 0.79-1.70) and UC (1.54, 1.15-2.05), compared with H2RA. When stratifying by indication of use at baseline, there were elevated risks of IBD in children with off-label use (weighted HR, 1.27, 95% CI 0.70-2.30) and no record of indications (1.42, 1.09-1.85), compared with children with a labeled indication (1.19, 0.61-2.31). The main finding was consistent when applying a 3, 4, and 5-lag period and considering switching between the study drugs.

Conclusions: In this large cohort study, PPIs compared with H2RAs were associated with a higher risk of IBD in children. The incidence rate was very low, but the observed risk suggests a need for PPI prescriptions only when clearly indicated, while the therapeutic benefits of PPI and other known risks should be properly considered.