Oral anticoagulation after catheter ablation: does continuation or discontinuation affect major cardiovascular outcomes? a systematic review and meta-analysis of 143,855 patients

Background Atrial fibrillation (AF) imposes a substantial health-care burden, and catheter ablation improves rhythm control and quality of life; however, long-term thromboembolic risk after ablation remains uncertain. Guidelines recommend oral anticoagulation (OAC) post-ablation according to thromboembolic risk, yet the net benefit—stroke prevention vs bleeding—remains incompletely defined, particularly without documented arrhythmia recurrence. We performed a systematic review and meta-analysis comparing OAC continuation vs discontinuation after AF ablation and assessed whether effects differ by CHA2DS2-VASc risk. Purpose To compare thromboembolic, bleeding, and major cardiovascular outcomes with OAC continuation versus discontinuation after AF catheter ablation, and to evaluate whether effects differ across CHA2DS2-VASc risk strata. Methods PubMed and Embase were searched through 15 February 2026 for comparative studies (n=143,855). Outcomes included thromboembolic events, stroke, ischaemic stroke, TIA, major bleeding, intracranial haemorrhage (ICH), AF recurrence, and all-cause mortality. Random-effects models estimated RR (95% CI); heterogeneity was assessed using I². Prespecified subgroup analyses were performed by CHA2DS2-VASc. Results Twenty-seven studies, including 143,855 patients (70,996 in OAC continuation; 72,859 in OAC discontinuation), were analysed. In pooled overall analyses, OAC continuation was associated with higher thromboembolic events (RR 1.42 [1.02–1.95]), major bleeding (RR 3.45 [2.16–5.52]), and ICH (RR 5.20 [2.70–10.03]). Stroke was more frequent with OAC continuation (RR 2.08 [1.45–2.99]), including ischaemic stroke (RR 1.58 [1.10–2.27]), and AF recurrence was higher in the continuation group (RR 1.69 [1.26–2.25]). No significant differences were observed for all-cause mortality and TIA. Overall estimates may reflect baseline risk imbalance and heterogeneity across studies; risk-stratified analyses suggested benefit of continuation in CHA2DS2-VASc ≥2, with fewer thromboembolic events (RR 0.55 [0.31–0.98]) and fewer ischaemic strokes (RR 0.37 [0.15–0.90]), at the expense of higher major bleeding (RR 4.04 [1.48–11.06]). Conclusion After AF ablation, OAC continuation is consistently associated with increased major bleeding and ICH, while thromboembolic effects appear risk-dependent. Overall findings should be interpreted in light of heterogeneity and potential baseline risk imbalance; the most clinically actionable signal was a thromboembolic benefit of continuation in CHA2DS2-VASc ≥2, at the expense of substantially more major bleeding. These data support a personalised, dual-risk approach to post-ablation anticoagulation integrating thromboembolic and bleeding-risk assessment and patient preferences, and highlight the need for adequately powered randomised trials with standardised post-ablation strategies and rhythm surveillance.