Background
Residual peri-device leaks (PDL) remain an important issue after left atrial appendage occlusion (LAAO). Complete isolation of the left atrial appendage appears to be associated with improved sealing performance and lower residual leak rates. However, comparative data on the performance of different device systems in relation to LAA morphology and residual leak rates in this specific population remain limited.
Objective
This single-center study aimed to compare the performance of three different LAAO device platforms with regard to baseline characteristics, LAA morphology, procedural findings, follow-up outcomes, and residual leaks in patients undergoing isolated left atrial appendage occlusion.
Methods
A total of 213 consecutive patients who underwent LAAO between March 2019 and March 2026 were retrospectively analyzed. In all patients, the LAA had previously been electrically isolated by catheter ablation, and the absence of electrical conduction was confirmed before the occlusion procedure. Devices were categorized into three groups: Watchman FLX/2.5 (n = 66), AMULET (n = 32), and LAmbre (n = 115). Baseline characteristics, LAA morphology assessed by transesophageal echocardiography (TEE) and fluoroscopy, procedural findings, follow-up outcomes, and residual leak rates were compared between groups. Clinically relevant leaks were defined as residual leaks ≥3 mm on TEE performed 6–8 weeks after implantation. The composite endpoint consisted of all-cause mortality, device-related thrombus, and stroke/transient ischemic attack (TIA) during follow-up.
Results
Baseline characteristics were largely comparable between groups, although HAS-BLED scores differed slightly (p = 0.045). The distribution of LAA morphologies differed significantly between device groups (p = 0.041). Windsock morphology was more common in the Watchman group, whereas Broccoli and bi-lobed morphologies were more frequent in the LAmbre group.After at least 6 weeks of echocardiographic follow-up, clinically relevant leaks (≥3 mm) were observed in 9.1% (n=6/66) of the Watchman group, 0% (n=0/32) of the AMULET group, and 5.2% (n=6/115) of the LAmbre group, without significant differences between groups (p = 0.18). The LAmbre group demonstrated a higher rate of small residual leaks (<3 mm), whereas rates of clinically relevant leaks were comparable across devices. Follow-up duration was similar between groups. Composite endpoints occurred in 3/66 patients (4.5%) in the Watchman group, 2/32 patients (6.3%) in the AMULET group, and 6/115 patients (5.2%) in the LAmbre group, without significant intergroup differences.
Conclusions
Although LAA morphology differed significantly between device groups, rates of clinically relevant residual leaks and adverse clinical outcomes were comparable across devices. While small residual leaks were more frequent in the LAmbre group, this did not translate into a higher incidence of clinically relevant leaks or composite clinical endpoints. Overall, all three devices demonstrated similar mid-term procedural and clinical performance.
