Residual Thromboembolic Risk and Outcomes of Oral Anticoagulation Discontinuation After Atrial Fibrillation Ablation: A Target Trial Emulation
Xiaodong Peng, Liu He, Jue Wang, Sitong Li, Qifan Li, Zixu Zhao, Zejun Yang, Mingxiao Li, Fadong Li, Xiaoxia Liu, Song Zuo, Xueyuan Guo, Junmeng Zhang, Ning Zhou, Yunlong Wang, Deyong Long, Caihua Sang, Xin Du, Jianzeng Dong, Changsheng MaAbstract
Background and Aims
Optimal management of oral anticoagulation (OAC) after atrial fibrillation (AF) ablation remains uncertain. We evaluated thromboembolic and bleeding outcomes associated with OAC discontinuation versus continuation at a clinically relevant 6-month post-ablation landmark.
Methods
This target trial emulation used data from a multicentre prospective registry in China. Patients with CHA2DS2-VA scores ≥2, no prior thromboembolism, and no atrial arrhythmia recurrence within 6 months after ablation were classified according to OAC discontinuation or continuation at the 6-month landmark. The primary outcome was the composite of stroke, systemic embolism, and major bleeding. Inverse probability weighting was applied, with intention-to-treat as the primary analysis.
Results
Among 8,339 patients (mean age 68 years; 40.5% women), 4,406 discontinued and 3,933 continued OAC. The risk of the primary outcome did not differ significantly between groups (weighted HR 0.93; 95% CI 0.67–1.29). Thromboembolic risk was similarly comparable (HR 0.96; 95% CI 0.69–1.35). Clinically relevant non-major bleeding occurred less frequently after OAC discontinuation (HR 0.68; 95% CI 0.47–0.97). Findings were consistent in sensitivity analyses. Annualized thromboembolic rates after discontinuation were <1% in patients with CHA2DS2-VA scores 2–3 but 1.52% in those with scores ≥4.
Conclusions
Among patients without prior thromboembolism who remained arrhythmia-free at 6 months after AF ablation, OAC discontinuation was not associated with a difference in the composite outcome of stroke, systemic embolism and major bleeding, compared with OAC continuation. Residual risk remained low in patients with CHA2DS2-VA scores 2–3 but exceeded conventional thresholds in those with scores ≥4.