Randomized trials have shown that direct oral anticoagulants (DOACs) are superior to warfarin in patients with nonvalvular atrial fibrillation. However, long-term use of anticoagulation carries an inherent risk of bleeding and nonadherence. Although the use of percutaneous left atrial appendage occlusion (LAAO) has become readily available, its effectiveness relative to oral anticoagulants is still unclear. The present study aimed to compare the outcomes of warfarin, DOACs, and LAAO in patients with atrial fibrillation. Medline, Embase, CENTRAL, and Web of Science were systematically searched through December 2021 for randomized controlled trials comparing warfarin, DOACs, or LAAO, reporting on all-cause mortality, stroke, and clinically relevant bleeding. A random-effects model was used to assess the safety and efficacy outcomes of these 3 treatments relative to each other in a Bayesian network meta-analysis. A total of 40 trials with 95,469 patients (LAAO: 5 trials, 3,032 patients; DOAC: 36 trials, 54,327 patients; warfarin: 37 trials, 38,110 patients) were included. LAAO was associated with significantly lower mortality than warfarin (odds ratio [OR] 0.68; 95% credible interval [CrI] 0.50 to 0.90) and DOACs (OR 0.75, 95% CrI 0.55 to 0.99). LAAO was the best-ranked treatment with respect to mortality reduction (surface under the cumulative ranking curve [SUCRA] 98.77%) and bleeding avoidance (SUCRA 72.26%). Compared with warfarin, DOACs significantly reduced mortality (OR 0.91, 95% CrI 0.85 to 0.97), stroke (OR 0.80, 95% CrI 0.63 to 0.93), and bleeding (OR 0.78, 95% CrI 0.63 to 0.95) and were ranked as the best option at preventing stroke (SUCRA 82.63%). In conclusion, LAAO was associated with lower mortality compared with DOACs, and both LAAO and DOACs significantly reduce mortality compared with warfarin. Future trials are needed to rule out a significant inferiority of LAAO compared with DOACs in terms of stroke and bleeding risks.
