Safety and efficacy of direct oral anticoagulation in patients with and without radiofrequency ablation of non-valvular atrial fibrillation: a multicenter retrospective cohort study

Background Based on the few available studies on the prognostic benefit of using direct oral anticoagulants (DOACs) after atrial fibrillation (AF) ablation. Therefore, this study aimed to evaluate the prognostic differences between patients who underwent radiofrequency ablation (RFA) and those without RFA taking DOACs. Methods This is a multicenter retrospective cohort study enrolling 6137 patients with non-valvular AF (NVAF) at 15 hospitals in China. Patient information was collected through a mean follow-up of 10 months and medical record queries. Clinical outcomes included major bleeding, total bleeding, thrombosis, all-cause death, and a composite endpoint of bleeding, thrombosis, and all-cause death. Results After adjusting for confounders and propensity score matching (PSM), patients with RFA of NVAF had a significantly lower risk of major bleeding [OR 0.278 (95% CI, 0.150-0.515), P< 0.001], thrombosis [OR 0.535 (95% CI, 0.316-0.908), P= 0.020] and the composite endpoint [ OR 0.835 (95% CI, 0.710-0.982), P= 0.029]. In the RFA PSM cohort, dabigatran was associated with reduced all-cause death in patients with RFA of NVAF [OR 0.420 (95% CI, 0.212-0.831), P= 0.010]. In the no RFA PSM cohort, rivaroxaban was associated with a reduction in major bleeding [OR 0.521 (95% CI, 0.403-0.673), P< 0.001], total bleeding [OR 0.114 (95% CI, 0.049-0.266), P< 0.001], and the composite endpoint [OR 0.659 ( 95% CI, 0.535-0.811), P< 0.001]. Conclusion Among patients with NVAF treated with DOACs, RFA was a negative correlate of major bleeding, thrombosis, and composite endpoints but was not associated with total bleeding or all-cause mortality.