Comparing Efficacy And Safety Of Left Atrial Appendage Closure Devices: A Network Meta-Analysis Of Randomized Controlled Trials

IntroductionAtrial fibrillation-related stroke is a leading cause of morbidity and mortality. Options for prevention include left atrial appendage closure devices or oral anticoagulation. However, it remains unclear which option may be superior overall. MethodsWe conducted a systematic review and network meta-analysis of all clinical trials comparing the WATCHMAN, Amplatzer Amulet (Amulet), and/or OACs. The primary outcomes of interest were any stroke and all-cause death. Safety outcomes included any thromboembolism, device embolization, and pericardial effusion. We calculated risk ratios and heterogeneity statistics for each comparison, and calculated the probability of intervention superiority where at least one comparison was significant. ResultsThere were 441 articles identified from the search, from which 5 eligible RCTs were identified (n=1,811). Compared to OACs (all warfarin), risk of stroke was non-significantly decreased with WATCHMAN (RR=0.90, 95% CI: 0.48, 1.72, I2=13.4%), but risk with Amulet was non-significantly lower than WATCHMAN (RR=0.95, 95% CI: 0.50, 1.81). However, observed risk of all-cause death was significantly lower with Amulet than OAC (RR=0.53, 95% CI: 0.33, 0.85, I2=0%) and trended towards significance versus WATCHMAN (RR=0.74, 95% CI: 0.55, 1.02, p=0.06). The P-score was 0.982, signifying a >98% probability Amulet was superior to all alternatives. Risk of thromboembolism was non-significantly increased with WATCHMAN (RR=2.04, 95% CI: 0.23, 18.4) and Amulet (RR=1.54, 95% CI: 0.11, 22.1), with head-to-head comparison favoring Amulet (RR=0.76, 95% CI: 0.17, 3.38, I2=0%). Versus WATCHMAN, device embolization risk was non-significantly elevated with Amulet (RR= 2.38, 95% CI: 0.67, 8.43, I2=0%). Finally, risk of pericardial effusion was significantly elevated with Amulet versus OACs (RR=27.0, 95% CI: 3.48, 210) and versus WATCHMAN (RR=2.11, 95% CI: 1.41, 3.17, I2=0%). The inverse P-score for Amulet (0.9995) indicated a very high probability Amulet was inferior to alternatives. ConclusionWhile risk of some adverse events was greater with Amulet, we estimated >98% probability Amulet is superior to alternatives in risk of death. Pooled patient-level analyses are warranted.