Proton pump inhibitor (PPI) exposure and risk of pancreatic cancer (PC): Meta-analysis.

Abstract

e15755 Background: PPIs are widely used in the treatment of various acid related disorders. Observational studies have raised concern about an association of PPI use and PC but findings have been inconsistent. This meta-analysis aimed to estimate from published studies the pooled PC risk among subjects exposed and not exposed to PPIs. Methods: We searched PubMed, EMBASE, Scopus, Cochrane Library and clinicaltrials.gov to identify relevant studies on the association of PPI exposure and PC risk. Three reviewers independently screened search results for eligibility and extracted data from retained studies. Our primary analysis quantified PC risk among subjects exposed vs not exposed to PPI, expressed as the pooled (adjusted) odds ratio (OR/aOR) and 95% confidence interval (95%CI) as estimated in a random effect model. We performed similar secondary analyses in selected subgroups. Results: Of the 2683 reports yielded by the search 1 randomized trial, 2 cohort, and 9 case-control studies with 1,093,766 subjects (123,214 cases; 970,552 controls) were retained. Our main analysis (exposed vs non exposed) revealed a significant overall association between PPI exposure and PC risk (OR 1.77, 95%CI 1.14-2.74). Secondary analyses showed higher risk estimates from high quality (NOS > 7) studies (OR 2.04, 95%CI 1.03-4.02), observational studies (aOR, 1.70 95%CI 1.08-2.68), and studies with PPI use and PC risk as main outcome (aOR 2.01, 95%CI 1.06-3.83). Subgroup analyses by comorbidities showed an association with diabetes (aOR 1.98, 95%CI 1.02-3.82) but not pancreatitis (aOR 1.77, 95%CI 0.93-3.35). PC risk was not associated with duration of PPI use longer 1 (aOR 1.36 95%CI 0.81-2.28 and > 3 yrs (aOR 1.21 95%CI 0.73-2.02). In addition to the overall PPI class effect in the primary analysis, rabeprazole was singularly associated with PC risk (aOR 5.40 95%CI 1.98-14.70). Conclusions: Pooled risk estimates suggest that the class of PPIs is associated with a general 1.77 fold increase in PC risk, independent of duration of PPI exposure. Diabetic patients are at a significant incremental risk over the base risk.