Abstract

• Warfarin, at a dose adjusted according to the INR, is the standard prophylactic anticoagulant for patients with atrial fibrillation and a major risk of thrombosis. Dabigatran, a thrombin inhibitor, is an alternative when warfarin fails to maintain the INR within the therapeutic range most of the time. Warfarin and aspirin are reasonable choices for patients at moderate risk of thrombosis.
   
• Rivaroxaban (Xarelto°) a factor Xa inhibitor, has been approved for the treatment of patients with atrial fibrillation and a moderate or major risk of thrombosis, but with no associated valve abnormalities.
   
• Clinical evaluation of rivaroxaban is mainly based on a double-blind, randomised, non-inferiority trial comparing rivaroxaban (20 or 15 mg taken once daily, according to renal function) versus adjusted-dose warfarin in 14 264 patients at high risk of thrombosis. Most patients were treated for at least 18 months.
   
• Overall mortality was not significantly different between the 2 groups (about 5% annually), nor was the incidence of stroke or systemic embolism (2% annually). Note that the dose of warfarin was not optimised in this trial.
   
• Indirect comparison with dabigatran is too fraught with methodological flaws to provide meaningful results. Overall, data on rivaroxaban are less convincing than those on dabigatran.
   
• About 35% of patients in the 2 groups stopped treatment prematurely, mainly because of adverse effects or withdrawal of consent. The overall incidence of bleeding was similar with rivaroxaban and warfarin (about 15%), including the incidence of serious bleeding (3.5%). Rivaroxaban was associated with fewer bleeding-related deaths (0.24% versus 0.48%), more cases of serious gastrointestinal bleeding (3.2% versus 2.02%) and fewer cases of intracranial haemorrhage (0.8% versus 1.2%).
   
• Combination with cytochrome P450 or P-glycoprotein inhibitors, or with drugs affecting renal function, boosts the effects of rivaroxaban. Combination with other antithrombotic drugs should be avoided.
   
• In practice, warfarin remains the standard prophylactic drug for patients with atrial fibrillation and a major risk of thrombosis, while dabigatran is an alternative in cases that are difficult to manage. In mid-2012, the data on rivaroxaban are not sufficiently convincing to challenge this standard.

©Prescrire 1 November 2012

"Rivaroxaban and atrial fibrillation" Prescrire Int 2012; 21 (132): 257-260 (Pdf, subscribers only)