Left Atrial Appendage Closure in Patients Considered Unsuitable for Oral Anticoagulation

Background Oral anticoagulation (OAC) use in atrial fibrillation is suboptimal due to real or perceived bleeding risks. Left atrial appendage closure (LAAC) is an alternative to OAC, but pivotal trials used OAC comparators. Objectives The study sought to assess the long-term safety and effectiveness of LAAC in patients deemed unsuitable for OAC. Methods The ASAP-TOO trial was a multinational randomized trial of patients with atrial fibrillation, CHA2DS2-VASc (congestive heart failure, hypertension, age ≥75 years, diabetes mellitus, prior stroke or transient ischemic attack or thromboembolism, vascular disease, age 65-74 years, sex category) score ≥2, and OAC unsuitability; target enrollment was 888 patients. Patients were assigned 2:1 to LAAC with a Watchman device or control treatment-single antiplatelet therapy or no therapy. Device patients received dual antiplatelet therapy for 3 months and aspirin through 12 months. The primary effectiveness endpoint was ischemic stroke or systemic embolism; the primary safety endpoint was death, ischemic stroke, systemic embolism, or 7-day device/procedure-related events requiring major intervention. Results Enrollment stopped early because of slow accrual, accentuated by the COVID-19 pandemic: only 481 patients were randomized (320 LAAC, 161 control treatment) and withdrawal was significant (7.8% LAAC, 24.2% control treatment), with disparate follow-up durations (median 876 days and 440.1 total patient-years of follow-up for control patients; median 1,768 days and 1,154.6 total patient-years of follow-up for LAAC patients), reducing statistical power. Patients were 75 ± 8 years of age, the CHA2DS2-VASc score was 4.6 ± 1.7, and 34% had prior intracranial/subdural hemorrhage. Implantation success was 97.8%. Only 31 of the 101 planned primary effectiveness events occurred, fewer than the 34 events required to trigger the first planned interim analysis. The primary safety endpoint occurred in 1.0% of implant-attempted patients, meeting the prespecified performance goal (P < 0.0001); there were 3 pericardial effusions requiring intervention. At up to 5 years, ischemic stroke or systemic embolism occurred in 7.8% with LAAC and 11.4% with control treatment (HR: 0.73; 95% CI: 0.35-1.52; P = 0.39). Major bleeding was 22.1% with LAAC and 15.7% with control treatment (HR: 1.64; 95% CI: 0.96-2.81; P = 0.067), concentrated during the first 3 months. In an exploratory analysis, disabling/fatal stroke was nominally less frequent with LAAC (2.2% vs 5.7%; HR: 0.32; 95% CI: 0.10-1.05). Conclusions In this prematurely terminated and underpowered trial, LAAC implantation was successful, with acceptable safety. Long-term estimates with LAAC were numerically favorable for thromboembolism but imprecise, while early bleeding was increased with the mandated antiplatelet regimen. These findings should be interpreted as exploratory and integrated with contemporary evidence and shared decision making. (Assessment of the WATCHMAN Device in Patients Unsuitable for Oral Anticoagulation [ASAP-TOO]; NCT02928497).