Safety and Efficacy of Minimally Interrupted Dabigatran vs Uninterrupted Warfarin Therapy in Adults Undergoing Atrial Fibrillation Catheter Ablation: A Randomized Clinical Trial.

IMPORTANCE Uninterrupted dabigatran therapy reduces stroke risk in patients with nonvalvular atrial fibrillation (NVAF) undergoing ablation and is associated with a lower bleeding risk than uninterrupted warfarin therapy. Minimally interrupted direct oral anticoagulant therapy is widely used, but data from controlled studies are insufficient. OBJECTIVE To compare the safety and efficacy of minimally interrupted dabigatran vs uninterrupted warfarin therapy in patients undergoing catheter ablation for NVAF. DESIGN, SETTING, AND PARTICIPANTS The ABRIDGE-J (ABlation peRIoperative DabiGatran in use Envisioning in Japan) trial is a open-label, randomized clinical trial performed in 28 Japanese treatment centers. A total of 504 patients scheduled for NVAF ablation were enrolled; 500 were randomized to the study treatments; 499 received at least 1 dose of dabigatran etexilate (n = 248) or warfarin potassium (n = 251); and 442 underwent ablation (220 in the dabigatran group and 222 in the warfarin group). Data were collected from May 1, 2014, through September 14, 2015, and analyzed from March 7, 2017, through January 28, 2019. INTERVENTIONS Appropriate dose anticoagulation was administered 4 weeks before and at least 3 months after ablation in all patients. Dabigatran therapy was interrupted before catheter ablation (holding of 1-2 doses) and resumed after ablation. MAIN OUTCOMES AND MEASURES Primary end points were the incidence of embolism during the perioperative period and atrial thrombus just before the ablation. The main secondary end point was the incidence of major bleeding events until 3 months after ablation. RESULTS Of the 442 patients who underwent ablation, 74.9% were men and the median age was 66 years (interquartile range, 59-71 years). Before ablation, 1 cerebral infarction and 1 thrombus in the left atrium occurred in the warfarin group, but no events occurred in the interrupted dabigatran group. After ablation, the mean (SD) incidence of major bleeding events was significantly lower with dabigatran (3 patients [1.4% {0.8%}; 95% CI, 0.4%-4.2%]) vs warfarin (11 patients [5.0% {1.5%}; 95% CI, 2.8%-8.8%]; P = .03). No thromboembolic events occurred after ablation in the dabigatran group; 1 (0.5%) occurred in the warfarin group. CONCLUSIONS AND RELEVANCE In patients undergoing ablation for NVAF, anticoagulation with minimally interrupted dabigatran therapy did not increase thromboembolic events and was associated with fewer bleeding complications than uninterrupted warfarin therapy. TRIAL REGISTRATION umin.ac.jp Identifier: UMIN000013129.