Creatinine clearance and inappropriate dose of rivaroxaban in Japanese patients with non-valvular atrial fibrillation.

In Japanese surveillance in an early phase after the approval of rivaroxaban, inappropriate underdose was frequently utilized. The aim of this study was to describe the prevalence and predictors of the inappropriate usage of rivaroxaban in a single-center, cardiovascular-specialized hospital. Consecutive 661 non-valvular atrial fibrillation (NVAF) patients treated with rivaroxaban between 2012 and 2017 were recruited. After excluding 30 patients without assessment of creatinine clearance (CCr), the proportion and predictors of inappropriate underdose were analyzed. Additionally, patient outcomes, including thromboembolism (ischemic stroke or systemic embolism) and major bleeding, were determined. In patients with CCr ≥ 50 mL/min (n = 532) and < 50 mL/min (n = 98), inappropriate underdose and overdose were used in 123 (23%) and 8 (8%), respectively. The predictors of inappropriate underdose (in patients with CCr ≥ 50 mL/min) were CCr [50-63 mL/min (the lowest tertile) compared to ≥ 64 mL/min], age ( ≥ 75 years), female gender, prescription of antiplatelet, and coexistence of heart failure. Although PT under rivaroxaban was lower in patients with inappropriate underdose than in those with an appropriate dose, no significant increase in the incidence of thromboembolism or major bleeding was observed within the mean follow-up of 683 days. Inappropriate underdose of rivaroxaban was frequently observed for NVAF patents even in a cardiovascular hospital, particularly in patients with CCr adjacent to the dose reduction criteria. The responses of PT and the incidence of adverse outcomes under an inappropriate dose of rivaroxaban should be further investigated.