Discontinuation of Oral Anticoagulation After Successful Atrial Fibrillation Ablation

Importance: There is no clear consensus regarding the discontinuation of oral anticoagulants (OACs) after catheter ablation (CA) for atrial fibrillation (AF). Objective: To evaluate thromboembolic and major bleeding events and all-cause death following OAC discontinuation and characteristics associated with patient prognoses after successful CA. Design, Setting, and Participants: This retrospective cohort study included patients without AF recurrence or adverse events 12 months after CA among those undergoing their first CA between January 1, 2006, and December 31, 2021. The study population was divided into groups according to the continuation and discontinuation of OACs at the landmark period of 12 months after CA. Follow-up data were acquired until December 31, 2023, and the study analysis was conducted from January to April 2024. Exposures: OAC discontinuation. Main Outcomes and Measures: Primary outcomes were thromboembolic and major bleeding events and all-cause death after 12 months. Inverse probability of treatment weighting (IPTW) and propensity score-matched analyses were used to adjust baseline characteristics. Results: This study included 1821 patients (mean SD age, 63.6 11.7 years; 1339 men 73.5%). Overall, 922 patients (50.6%) continued OAC for 12 months, whereas 899 (49.4%) discontinued OAC. During a mean (SD) follow-up of 4.8 (4.0) years, thromboembolic events, major bleeding events, and death occurred in 43 (2.4%), 41 (2.3%), and 71 (3.9%) patients, respectively. After IPTW adjustment, the OAC discontinuation group demonstrated a significantly higher incidence of thromboembolism (incidence rate, 0.86 95% CI, 0.45-1.35 vs 0.37 95% CI, 0.22-0.54 per 100 person-years; log-rank P = .04) and a lower incidence of major bleeding (incidence rate, 0.10 95% CI, 0.02-0.19 vs 0.65 95% CI, 0.43-0.90 per 100 person-years; log-rank P < .001) than in the continuation group. In a subgroup analysis, OAC discontinuation was associated with a higher risk of thromboembolism in patients with asymptomatic AF, left ventricular ejection fraction of less than 60%, and left atrial diameter of 45 mm or greater. In contrast, OAC discontinuation was beneficial for reducing major bleeding risks in patients with a HAS-BLED score of 2 or greater. These outcomes were similar in the propensity score-matched analysis using 1100 paired matched patients, except for insignificant differences in thromboembolic events. Differences in mortality between the 2 groups were not statistically significant. Conclusions and Relevance: In this retrospective cohort study, discontinuation of OACs after successful CA was associated with increased thromboembolic events and decreased bleeding events. The benefits of discontinuing OACs were stratified according to specific characteristics, pending a future prospective randomized study.