AIMS: To examine the risk of gastrointestinal adverse effects associated with dabigatran use compared with warfarin among patients with atrial fibrillation (AF).

METHODS AND RESULTS: Patients with AF and no history of gastrointestinal diseases initiating dabigatran or warfarin were identified from Danish nationwide registries from 22 August 2011 until 31 December 2012. Patients were classified as naive or experienced users, according to prior use of oral anticoagulant (OAC) therapy. The risk of subsequent proton pump inhibitor (PPI) use, upper dyspepsia-like diagnoses (gastroesophageal reflux, gastritis, gastric, and duodenal ulcer) and gastrointestinal bleeding requiring hospitalization, gastroscopy, and discontinuation of dabigatran and warfarin was examined by cumulative incidence rates and multivariable adjusted Cox regression models. We identified five groups: OAC-naive warfarin (n = 4534); OAC-naive dabigatran 110 mg b.i.d. (dabigatran 110) (n = 1168); OAC-naive dabigatran 150 mg b.i.d. (dabigatran 150) (n = 1844); OAC-experienced dabigatran 110 (n = 1143); and OAC-experienced dabigatran 150 (n = 1748). Compared with OAC-naive warfarin, the rate of initiating PPIs was significantly increased for OAC-naive dabigatran 110 [hazard ratio (HR), 1.24; 95% confidence interval (CI), 1.02-1.50]. Other dabigatran regimes were not associated with a higher risk of initiating PPIs, upper dyspepsia-like diagnoses, gastrointestinal bleeding, or gastroscopy. The risk of discontinuation was increased for OAC-experienced dabigatran 150 (HR, 1.13; 95% CI, 1.02-1.25), but not for the other dabigatran-treated groups, relative to OAC-naive warfarin.

CONCLUSION: Dabigatran was not associated with upper dyspepsia-like diagnoses or gastrointestinal bleeding requiring hospitalization, and gastroscopy. The risk of subsequent PPI use was increased for OAC-naive dabigatran 110 mg users, and the risk of discontinuation was increased for OAC-experienced dabigatran 150 mg users.