ABSTRACT NUMBER: ESOC2026A876 INCIDENCE OF CEREBROVASCULAR EVENTS BY CHA2DS2-VASC SCORE STRATIFICATION AFTER INTERRUPTED ORAL ANTICOAGULATION IN SUCCESSFUL ATRIAL FIBRILLATION CATHETER ABLATION: A META-ANALYSIS OF 128,144 PATIENTS

Abstract Background and aims The safety of suspending oral anticoagulation (OAC) after successful atrial fibrillation (AF) catheter ablation remains unclear, and prior studies have not stratified outcomes by CHA₂DS₂-VASc score. This meta-analysis evaluates thromboembolic and hemorrhagic cerebrovascular events in patients with interrupted versus uninterrupted OAC following ablation. Methods A comprehensive search of electronic databases through December 2025 identified studies comparing interrupted versus uninterrupted OAC after successful AF ablation, including vitamin K antagonists and direct oral anticoagulants. One non-randomized controlled trial, eight retrospective cohort studies (two with propensity score matching), and two prospective cohort studies were included. Outcomes included thromboembolic and hemorrhagic events stratified by CHA₂DS₂-VASc scores ≤1 and ≥2. Results Among 128,144 patients with AF after ablation (62,224 uninterrupted; 65,920 interrupted; 11 studies), no significant difference in thromboembolic events was observed between groups: CHA₂DS₂-VASc ≤1 (RR 0.87, 95%CI 0.51–1.48, p=0.61) and CHA₂DS₂-VASc ≥2 (RR 1.12, 95%CI 0.82–1.52, p=0.49). Interrupted OAC was associated with a significant reduction in hemorrhagic events in both risk strata: CHA₂DS₂-VASc ≤1 (RR 0.18, 95%CI 0.06–0.61, p=0.006) and CHA₂DS₂-VASc ≥2 (RR 0.27, 95%CI 0.09–0.78, p=0.02). Conclusions Interruption of OAC after successful AF ablation appears safe across CHA₂DS₂-VASc strata, with reduced hemorrhagic risk and no increase in thromboembolic events. These findings question routine post-ablation OAC continuation and warrant confirmation in randomized controlled trials. Conflict of interest Amiel Armando Aragon Cortes. Nothing to disclose Figure 1 - belongs to Results Figure 2 - belongs to Conclusions Figure 3 - belongs to Conclusions