Apixaban Efficacy in Atrial Fibrillation

Clinical evidence for apixaban in nonvalvular atrial fibrillation and reduction of stroke/systemic embolism

The evidence base for apixaban in nonvalvular AF rests on two pivotal phase III RCTs — ARISTOTLE and AVERROES — both of which produced strong, consistent results, supported by multiple meta-analyses and large real-world cohorts.

ARISTOTLE (vs. warfarin)

The foundational trial, published in the New England Journal of Medicine in 2011 by Granger et al., enrolled 18,201 patients with AF and at least one additional stroke risk factor. Over a median follow-up of 1.8 years, apixaban 5 mg twice daily demonstrated superiority over warfarin on three simultaneous endpoints — something no prior DOAC trial had achieved. The primary outcome of stroke or systemic embolism occurred at 1.27% per year with apixaban versus 1.60% per year with warfarin — a 21% relative risk reduction. Major bleeding was 31% lower with apixaban (2.13% vs. 3.09% per year), and all-cause mortality was modestly but significantly reduced (3.52% vs. 3.94%). The hemorrhagic stroke reduction was particularly striking: apixaban cut the rate by half (0.24% vs. 0.47% per year). Ischemic stroke rates, however, were similar between arms (0.97% vs. 1.05%), meaning the overall stroke benefit was driven primarily by the hemorrhagic stroke reduction rather than superior thromboembolic prevention over warfarin per se.

A subsequent JACC analysis of ARISTOTLE's bleeding data (Hylek et al., 2014) found that major bleeds followed by death within 30 days occurred half as often with apixaban compared to warfarin, underlining that apixaban's bleeding events were not only fewer but less fatal. The mean time in therapeutic range (TTR) with warfarin was approximately 62–65%, which is on the higher end for real-world practice — meaning the apixaban advantage over warfarin could be even larger in settings where INR control is worse.

AVERROES (vs. aspirin)

For patients unsuitable for vitamin K antagonist therapy, AVERROES randomized 5,599 patients to apixaban or aspirin. The trial was stopped early given clear superiority: apixaban reduced stroke or systemic embolism by 55% compared to aspirin. Crucially, this benefit came without a significant increase in major bleeding (4.5% vs. 3.8%/year, p=0.19), establishing apixaban as the clear anticoagulant of choice in this population — not merely as non-inferior to aspirin but substantially superior, at a bleeding cost that was clinically non-significant. Among the subgroup with prior TIA or ischemic stroke, the benefit was even larger (71% relative risk reduction in stroke/systemic embolism vs. aspirin).

How apixaban compares to other DOACs

No head-to-head RCTs compare apixaban against rivaroxaban or dabigatran directly, so indirect evidence comes from network meta-analyses. A systematic review of 22 NMAs (Cohen et al., International Journal of Cardiology, 2018) found no statistically significant difference between apixaban and any other DOAC for stroke/systemic embolism prevention, but apixaban consistently showed lower major bleeding risk than rivaroxaban in 16 of 20 NMAs and lower bleeding than dabigatran 150 mg in 13 of 16 NMAs. A meta-analysis of 10 real-world studies (Mamas et al., American Journal of Cardiology, 2021) similarly found apixaban associated with fewer strokes/systemic embolism and substantially lower major and gastrointestinal bleeding compared to rivaroxaban.

Elderly patients

A meta-analysis focused on patients over 75 (Malik et al., 2019) found that apixaban was the only DOAC that simultaneously reduced stroke/systemic embolism, major bleeding, and intracranial hemorrhage compared to warfarin — by 29%, 36%, and 66%, respectively — a distinction not shared by rivaroxaban or dabigatran in that age group.

Real-world confirmation

A large UK primary care cohort (QResearch/CPRD, ~18,000 apixaban users) and a US commercial claims study (Adeboyeje et al., 2017, ~44,000 patients) both confirmed the ARISTOTLE safety signal: apixaban was associated with substantially lower major bleeding than warfarin or rivaroxaban in routine clinical practice.

The overall picture is unusually consistent: ARISTOTLE's triple superiority result has been reproduced directionally across real-world observational data and indirect DOAC comparisons. The main residual uncertainty is whether apixaban's ischemic stroke prevention is meaningfully better than other DOACs — NMAs suggest it is not significantly different from dabigatran or rivaroxaban on that outcome. This is from an initial search; a comprehensive deep-dive could surface subgroup analyses (e.g., by CHA₂DS₂-VASc score, renal function, prior stroke) and more recent real-world evidence.