Peri-device sealing and procedural outcomes of LAA closure: a comparative study of device types in patients with and without LAA isolation

Abstract Background Catheter-based electrical isolation of the left atrial appendage (LAA) is increasingly utilized in patients with symptomatic atrial fibrillation (AF) that is refractory to pulmonary vein isolation (PVI). Due to the risk of thromboembolism associated with electrically isolated LAA, endovascular LAA occlusion (LAAO) has become an essential adjunctive procedure. Objective This study aimed to analyze procedural and 12-month clinical outcomes in patients undergoing LAAO with various occluder devices, comparing those with LAA isolation (LAAI) versus those without (No-LAAI). Methods A single-center, retrospective analysis was conducted on patients who underwent catheter ablation for AF between March 2019 and March 2024. All patients had undergone at least one prior AF ablation. The primary indication for LAAO was prior electrical LAA isolation. Procedural data, peri-procedural complications, and peridevice leaks at 45-day follow-up were analyzed, along with clinical outcomes (death, TIA/stroke, device thrombus, bleeding) at 12 months. Results The study included 249 patients (median age 73.2 years; 166 men [66.7%]), with 148 (59.4%) in the LAAI group and 101 (40.6%) in the No-LAAI group. The primary indication for LAAO was prior electrical LAA isolation in 82.4% (122/148) of the LAAI group and 57.4% (58/101) of the No-LAAI group. The distribution of LAAI/No-LAAI was as follows: Watchman FLX [62.8% (49/78) / 29/78], Watchman 2.5 [80% (8/10) / 2/10], AMULET [60% (12/20) / 8/20], and LAMBRE [56% (79/141) / 62/141]. At 45 days, peridevice leaks were observed in 20.3% (30/148) of the LAAI group and 44.5% (45/101) of the No-LAAI group (p < 0.001). Minor leaks (1-5 mm) constituted 86.7% (65/75) of all leaks, while major leaks (>5 mm) were seen in 10 patients. Leak distribution categorized as no leak/<3 mm/3-5 mm/>5 mm differed significantly between LAAI and No-LAAI groups (LAAI: 110/21/5/4 vs. No-LAAI: 55/27/12/6; p = 0.001 for none, p = 0.012 for <3 mm, p = 0.008 for 3-5 mm). Major leak occurrence was not significantly different between the groups (p = 0.195). Procedural complications occurred in 4,7% (7/148) of the LAAI group and 5.9% (6/101) of the No-LAAI group (p = not significant). The primary composite endpoint (death, stroke, severe bleeding, thromboembolism) was reached in 6.2% (9/148) of the LAAI group and 5.0% (5/101) of the No-LAAI group. Conclusion LAAO can be performed safely and effectively in patients with LAAI. Although the primary composite endpoint was similar between the groups, patients with LAAI demonstrated improved peridevice sealing. These findings highlight the potential benefits of LAAI for enhanced leak prevention, which is further complemented by the potential advantage regarding its rhythm-maintaining effect. Larger studies with extended follow-up are warranted to further evaluate these findings.