Sex-specific effects of apixaban versus aspirin in subclinical atrial fibrillation: insights from the ARTESiA trial

Background The ARTESiA trial showed that apixaban reduces stroke and increases major bleeding compared to aspirin in patients with subclinical atrial fibrillation (SCAF). Purpose To assess the sex-specific efficacy and safety of apixaban versus aspirin in patients with SCAF. Methods We performed a pre-specified subgroup analysis of the ARTESiA trial according to sex. The primary endpoints were stroke/systemic embolism (SE) (efficacy) and major bleeding (safety). We adjusted for the individual components of the CHA2DS2-VA score. Results Female patients had a lower mean CHA2DS2-VA score than male patients (3.3±1.1 vs. 3.8±1.1; p<0.001). All of congestive heart failure, diabetes, prior stroke and coronary artery disease were less prevalent in female compared to male patients (all p<0.001). The proportion of patients with a CHA2DS2-VA score ≥4 was 35% in females and 52% in males (p<0.001). Among females, the crude rate of stroke/SE was 0.6%/year with apixaban and 0.9%/year with aspirin (adjusted hazard ratio [HR] 0.68, 95% confidence interval [CI] 0.36-1.28); including 0.5%/year with apixaban and 0.7%/year with aspirin (HR=0.68; 95% CI 0.28-1.64) in those with a CHA2DS2-VA score <4 and 0.9%/year with apixaban and 1.4%/year with aspirin (HR=0.65, 95% CI 0.26-1.62) in those with a CHA2DS2-VA-score ≥4. Among males, the crude rate of stroke/SE was 0.9%/year on apixaban and 1.5%/year on aspirin (adjusted HR 0.61, 95% CI 0.41-0.91), including 0.9%/year with apixaban and 1.1%/year with aspirin (HR=0.82, 95% C I 0.45-1.49) in those with a CHA2DS2-VA score <4 and 0.9%/year with apixaban and 1.9%/year with aspirin (HR=0.48, 95% CI 0.28-0.83) in those with a CHA2DS2-VA score of ≥4. We did not find evidence for interaction of the reduction in stroke with apixaban vs. aspirin with sex (p for interaction=0.75). In female patients, the crude rate of major bleeding was 1.5%/year on apixaban and 1.1%/year on aspirin (adjusted HR 1.39, 95% CI 0.86-2.25), including 1.4%/year with apixaban and 0.8%/year with aspirin (HR=1.72, 95% CI 0.91-3.24) in those with a CHA2DS2-VA score <4 and 1.7%/year on apixaban and 1.6%/year on aspirin (HR=1.04, 95% CI 0.50-2.16) in those with CHA2DS2-VA score of ≥ 4. In male patients, the crude rate of major bleeding was 1.5%/year on apixaban and 1.2%/year on aspirin (adjusted HR 1.29, 95% CI 0.89-1.87), including 1.2%/year on apixaban and 1.3%/year on aspirin (HR=0.83, 95% CI 0.49-1.42) in those with a CHA2DS2-VA <4 and 1.9%/year on apixaban and 1.0%/year on aspirin (HR=1.92, 95% CI 1.13-3.26) in those with CHA2DS2-VA score ≥4. We did not find evidence for interaction of the increase in bleeding with apixaban vs. aspirin with sex (p for interaction=0.76). Conclusions The reduction in stroke/SE and the increase in major bleeding with apixaban compared to aspirin were consistent in female and male patients, including in high-risk patients with a CHA2DS2-VA score ≥4.