Left Atrial Appendage Occlusion in Patients With Anticoagulation Failure vs Anticoagulation Contraindication

Left atrial appendage occlusion (LAAO) provides mechanical cardioembolic protection for atrial fibrillation (AF) patients who cannot use oral anticoagulation therapy (OAT). Patients with a thrombotic event despite OAT are at high risk for recurrence and may also benefit from LAAO. This study sought to investigate the efficacy of LAAO in AF patients with a thrombotic event on OAT compared to: 1) LAAO in AF patients with a contraindication for OAT; and 2) historical data. The international LAAO after stroke despite oral anticoagulation (STR-OAC LAAO) collaboration included patients who underwent LAAO because of thrombotic events on OAT. This cohort underwent propensity score matching and was compared to the EWOLUTION (Evaluating Real-Life Clinical Outcomes in Atrial Fibrillation Patients Receiving the WATCHMAN Left Atrial Appendage Closure Technology) registry, which represents patients who underwent LAAO because of OAT contraindications. The primary outcome was ischemic stroke. Event rates were compared between cohorts and with historical data without OAT, yielding relative risk reductions based on risk scores. Analysis of 438 matched pairs revealed no significant difference in the ischemic stroke rate between the STR-OAC LAAO and EWOLUTION cohorts (2.5% vs 1.9%; HR: 1.37; 95% CI: 0.72-2.61). STR-OAC LAAO patients exhibited a higher thromboembolic risk (HR: 1.71; 95% CI: 1.04-2.83) but lower bleeding risk (HR: 0.39; 95% CI: 0.18-0.88) compared to EWOLUTION patients. The mortality rate was slightly higher in EWOLUTION (4.3% vs 6.9%; log-rank P = 0.028). Relative risk reductions for ischemic stroke were 70% and 78% in STR-OAC LAAO and EWOLUTION, respectively, compared to historical data without OAT. LAAO in patients with a thrombotic event on OAT demonstrated comparable stroke rates to the OAT contraindicated population in EWOLUTION. The thromboembolic event rate was higher and the bleeding rate lower, reflecting the intrinsically different risk profile of both populations. Until randomized trials are available, LAAO may be considered in patients with an ischemic event on OAT.