Abstract
BACKGROUND: Although older patients with atrial fibrillation are at heightened risk of thromboembolic and bleeding events, their optimal treatment choice remains uncertain. METHODS AND RESULTS: This meta-analysis was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. We searched the PubMed, EMBASE, and Cochrane databases for randomized controlled trials that compared thromboembolic or bleeding outcomes between a direct oral anticoagulant (DOAC) and a vitamin K antagonist (VKA) and reported outcomes for patients aged ≥75 years with atrial fibrillation. The efficacy outcome was the composite of stroke and systemic embolism. Safety outcomes included major bleeding, any clinically relevant bleeding, and intracranial hemorrhage. Each DOAC and VKA was compared pairwise in a network meta-analysis. High- and low-dose regimens and factor IIa and Xa inhibitors were also compared. Seven randomized controlled trials were included in the analysis. Stroke and systemic embolism risks did not differ significantly among DOACs. There were no significant differences in major bleeding between each DOAC and VKA. Intracranial hemorrhage risk was significantly lower with dabigatran, apixaban, and edoxaban than with VKA and rivaroxaban, which had similar risks. High-dose regimens led to lower risks of stroke or systemic embolism compared with VKA and low-dose regimens, with both doses having similar bleeding risks. CONCLUSIONS: In patients aged ≥75 years with atrial fibrillation, DOACs were associated with fewer thromboembolic events compared with VKA, whereas dabigatran, apixaban, and edoxaban were associated with lower risks of intracranial hemorrhage compared with VKA and rivaroxaban. REGISTRATION: URL: www.crd.york.ac.uk/prospero/. Unique identifier: CRD42022329557.