Percutaneous left atrial appendage occlusion versus oral anticoagulation in atrial fibrillation: A pragmatic meta-analysis of randomized trials within the DOAC era

Background Percutaneous left atrial appendage occlusion (LAAO) is an established alternative to oral anticoagulation (OAC). However, its safety and efficacy boundaries remain incompletely defined in dedicated comparisons against direct oral anticoagulants (DOACs). Objective This study applied a pragmatic framework to evaluate the comparative efficacy and safety of LAAO versus OAC in a DOAC-predominant setting. Methods PubMed, Embase, and Cochrane Library were searched for randomized controlled trials comparing LAAO with OAC, restricted to a DOAC-predominant comparator. Risk ratios (RR) with 95% confidence intervals (CI) were calculated via a Mantel-Haenszel random-effects model. Results Four trials comprising 5890 patients were included (96.4% of controls assigned to DOACs). Overall stroke (RR 1.18, 95% CI 0.87–1.60; p = 0.28), systemic embolism (RR 1.09, 95% CI 0.29–4.08; p = 0.89), major bleeding (RR 0.98, 95% CI 0.81–1.19; p = 0.84), hemorrhagic stroke (RR 0.82, 95% CI 0.45–1.51; p = 0.53), and all-cause mortality (RR 1.00, 95% CI 0.85–1.17; p = 0.98) did not differ significantly. Ischemic stroke presented numerical asymmetry favoring OAC that approached statistical significance (RR 1.36, 95% CI 0.99–1.89; p = 0.06). Conversely, within available protocol-defined datasets, LAAO significantly reduced nonprocedural clinically relevant bleeding (RR 0.50, 95% CI 0.43–0.60; p < 0.01). Conclusions In a pragmatic analysis of the DOAC era, randomized data show no statistically significant differences for hard individual endpoints, though broad confidence intervals reflect precision limits. A numerical asymmetry for ischemic stroke favored OAC, while LAAO reduced nonprocedural clinically relevant bleeding. Current evidence remains non-definitive, warranting shared clinical decision-making to balance individual thromboembolic risks against long-term bleeding susceptibility.