Objective:To perform a systematic review and meta-analysis of studies reporting recurrent intracranial hemorrhage (ICH) and ischemic stroke (IS) in ICH survivors with atrial fibrillation (AF) during long-term follow-up.Methods:A comprehensive literature search including MEDLINE, EMBASE, Cochrane library, clinical trials registry was performed following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement. We considered studies capturing outcome events (ICH recurrence and IS) for ≥3 months and treatment exposure to vitamin K antagonists (VKAs), antiplatelet agents (APAs), or no antithrombotic medication (no-ATM). Corresponding authors provided aggregate data for IS and ICH recurrence rate between 6 weeks after the event and 1 year of follow-up for each treatment exposure. Meta-analyses of pooled rate ratios (RRs) were conducted with the inverse variance method.Results:Seventeen articles met inclusion criteria. Seven observational studies enrolling 2,452 patients were included in the meta-analysis. Pooled RR estimates for IS were lower for VKAs compared to APAs (RR = 0.45, 95% confidence interval [CI] 0.27–0.74, p = 0.002) and no-ATM (RR = 0.47, 95% CI 0.29–0.77, p = 0.002). Pooled RR estimates for ICH recurrence were not significantly increased across treatment groups (VKA vs APA: RR = 1.34, 95% CI 0.79–2.30, p = 0.28; VKA vs no-ATM: RR = 0.93, 95% CI 0.45–1.90, p = 0.84).Conclusions:In observational studies, anticoagulation with VKA is associated with a lower rate of IS than APA or no-ATM without increasing ICH recurrence significantly. A randomized controlled trial is needed to determine the net clinical benefit of anticoagulation in ICH survivors with AF.