Are Factor Xa Inhibitors Efficacious for Ischemic Stroke Prevention in Patients Without Atrial Fibrillation? Evidence From Randomized Clinical Trials
Journal Articles
Overview
Research
Identity
Additional Document Info
View All
Overview
abstract
BACKGROUND: Clinical trials provide conflicting evidence regarding oral factor Xa inhibitors for prevention of ischemic stroke in patients without a history of atrial fibrillation (AF). METHODS: We performed a critical appraisal of randomized clinical trials that tested oral factor Xa inhibitors in patients without AF that reported ischemic stroke. RESULTS: Considering the 11 trials that reported > 10 ischemic strokes during follow-up (97,578 participants, 1195 ischemic strokes), 1 tested apixaban (57 strokes), 1 betrixaban (52 strokes), and 9 rivaroxaban (1086 strokes). In 7 trials with placebo comparisons, numerically fewer ischemic strokes occurred among those assigned factor Xa inhibitors in 7 of 8 randomized comparisons (range of hazard ratios [HRs], 0.89-0.51), including statistically significant reductions in 2 trials that compared rivaroxaban 2.5 mg twice daily vs placebo on a background of aspirin in patients with cardiovascular disease, COMPASS (HR, 0.51; 95% confidence interval [CI], 0.38-0.68) and COMMANDER-HF (HR, 0.64; 95% CI, 0.43-0.95). Compared with aspirin in 4 trials, oral factor Xa inhibitors were associated with fewer ischemic strokes in 2, with statistically significant reduction in 1 (rivaroxaban 5 mg twice daily in COMPASS; HR, 0.69; 95% CI, 0.53-0.90). Major bleeding was increased by oral factor Xa inhibitors in all 7 placebo-controlled trials (HR range, 1.42-4.08), with statistically significant increases reported in 5 trials, and in all 4 aspirin-controlled trials (all statistically significant increases; HR range, 1.52-2.72). CONCLUSIONS: Aggregate evidence on the basis of placebo comparisons from randomized trials supports the potential for oral factor Xa inhibitors to reduce ischemic stroke in patients without AF, but major bleeding is increased.
