Patients with atrial fibrillation (AF) experienced more major and non-major clinically relevant GI bleeding when taking rivaroxaban than patients taking warfarin.

As part of the ROCKET AF trial, researchers from multiple institutions, including Harvard Medical School; Mount Sinai Medical Center; Janssen Research & Development; and Bayer HealthCare Pharmaceuticals, randomized 14,264 patients with nonvalvular AF to receive either rivaroxaban or dose-adjusted warfarin. Post-hoc analysis was performed on all patients while receiving study medication until 2 days after the last dose.

Results showed that GI bleeding events (upper, lower, rectal) occurred more frequently in patients receiving rivaroxaban (n=394) than warfarin (n=290). In addition, major (2.00 vs. 1.24%/year) and nonmajor (1.75 vs. 1.39%/year), clinically relevant bleeding occurred more frequently with rivaroxaban use than warfarin use. Patients with major GI bleeding (n=684) were more likely to be older, current/prior smokers, have prior GI bleeding, mild anemia at baseline and a lower creatinine clearance, and less likely to be female or have prior stroke/TIA.  However, there were fewer fatal GI bleeds with rivaroxaban and the absolute fatality rate was very low.  Only one patient on rivaroxaban and five on warfarin developed fatal GI bleeding.

The authors conclude that both major and non-major, clinically relevant GI bleeding were more common in patients with AF taking rivaroxaban than warfarin. The most severe of these events were balanced between treatment groups. There were fewer fatal GI bleeds on rivaroxaban and the absolute fatality rate was very low.  A careful benefit-risk assessment is needed for high-risk patients.

This study was presented during CHEST 2012, the annual meeting of the American College of Chest Physicians, held October 20 – 25, in Atlanta, Georgia.  

The authors are Christopher Nessel, M.D.; Kenneth Mahaffey, M.D.; Jonathan Piccini, M.D.; Guohua Pan, PhD; Manesh Patel, M.D.; Richard Becker, M.D.; Daniel Singer, M.D.; Jonathan Halperin, M.D.; Graeme Hankey, M.D.; Scott Berkowitz, M.D.; Keith Fox, MBChB; Robert Califf, M.D..  Disclosures are in the abstract.