ABSTRACT
Renal impairment confers worse prognosis in patients with atrial fibrillation (AF) but there is scarce evidence about the influence of direct-acting oral anticoagulants in routine clinical practice. Herein, we compared clinical outcomes between patients with AF with and without renal impairment on rivaroxaban and investigated predictors for clinical outcomes in patients with AF with renal impairment. This was a multicenter study including patients with AF on rivaroxaban for at least 6 months. During 2.5 years follow-up, ischemic strokes (IS)/transient ischemic attacks (TIA)/systemic embolisms (SE)/myocardial infarctions (MI), major bleeding, and major adverse cardiovascular events (MACE) were recorded. Creatinine clearance (CrCl) was estimated using the Cockroft-Gault equation, renal impairment was defined as a CrCl <60 ml/min, and 1,433 patients (34.8% with CrCl <60 ml/min) were included. Patients with CrCl <60 ml/min showed higher event rates for major bleeding (1.87%/year vs 0.62%/year; p = 0.003) and MACE (1.97%/year vs 0.62%/year; p = 0.002) but similar event rates for IS/TIA/SE/MI (0.66%/year vs 0.67%/year; p = 0.955). In patients with renal impairment, CHA2DS2-VASc was associated with higher risk of IS/TIA/SE/MI; HAS-BLED and any dependency level were associated with higher risk of major bleeding; and male gender and heart failure were associated with higher risk of MACE. Antiplatelets were independently associated with increased risk of IS/TIA/SE/MI and MACE. In conclusion, in patients with AF on rivaroxaban, the incidence of IS/TIA/SE/MI did not increase in those with renal impairment, suggesting that rivaroxaban may be an effective option in this subgroup. In patients with AF, male gender, heart failure, dependency, antiplatelets, CHA2DS2-VASc, and HAS-BLED were associated with increased risk of adverse outcomes.
