The mean CHADS2 score was 3.25, and the most frequent scores were 3 and 4. No statistically significant interactions were observed between principal safety outcome event rates and CHADS2 scores with respect to treatment groups (P value for interaction聽=聽.700). Irrespective of stratification into moderate- and high-risk groups based on CHADS2 scores of 2 and 3 or more, respectively, no differences in principal safety outcome event rates were observed between rivaroxaban- and warfarin-treated patients (moderate-risk group: hazard ratio [HR], 1.06; 95% confidence interval [CI], .58-1.95; high-risk group: HR, 1.11; 95% CI, .86-1.45; P value for interaction聽=聽.488). The primary efficacy end point rate in the rivaroxaban-treated group was numerically lower than in the warfarin-treated group, regardless of risk group stratification (moderate-risk group: HR, .46; 95% CI, .09-2.37; high-risk group: HR, .49; 95% CI, .22-1.11; P value for interaction聽=聽.935).
This subanalysis indicated that the safety and efficacy of rivaroxaban compared with warfarin were similar, regardless of CHADS2 score.