Mechanisms, predictors, and evolution of severe peri-device leaks with two different left atrial appendage occluders.

Aims Incomplete left atrial appendage occlusion (LAAO) due to peri-device leak (PDL) is a limitation of the therapy. The Amulet IDE trial is the largest randomized head-to-head trial comparing the Amulet and Watchman 2.5 LAAO devices with fundamentally different designs. The predictors and mechanistic factors impacting differences in PDLs within the Amulet IDE trial are assessed in the current analysis. Methods and results An independent core lab analysed all images for the presence or absence of severe PDL (>5 mm). The incidence, mechanistic factors, predictors using propensity score-matched controls, and evolution of severe PDLs through 18 months were assessed. Of the 1878 patients randomized in the trial, the Amulet occluder had significantly fewer severe PDLs than the Watchman device at 45 days (1.1 vs. 3.2%, P < 0.001) and 12 months (0.1 vs. 1.1%, P < 0.001). Off-axis deployment or missed lobes were leading mechanistic PDL factors in each device group. Larger left atrial appendage (LAA) dimensions including orifice diameter, landing zone diameter, and depth predicted severe PDL with the Watchman device, with no significant anatomical limitations noted with the Amulet occluder. Procedural and device implant predictors were found with the Amulet occluder attributed to the learning curve with the device. A majority of Watchman device severe PDLs did not resolve over time through 18 months. Conclusion The dual-occlusive Amplatzer Amulet LAA occluder provided improved LAA closure compared with the Watchman 2.5 device. Predictors and temporal observations of severe PDLs were identified in the Amulet IDE trial. Clinical trial registration https://clinicaltrials.gov Unique identifier NCT02879448. [ABSTRACT FROM AUTHOR]