Percutaneous Left Atrial Appendage Closure with Concomitant Dual-Device Implantation: A Single-Center Observational Study

Antanas Gasys¹Roberto Galea²Tommaso Bini^3,4Juan Perich-Krsnik³Marco Gamardella³Roten Laurent³Georgios Siontis³Bernhard Meier³Lorenz Räber^3*

• ¹University Hospital of Bern, Bern, Switzerland
• ²Spitalzentrum Biel AG, Biel/Bienne, Switzerland
• ³Inselspital Universitatsspital Bern, Bern, Switzerland
• ⁴Universitat Bern, Bern, Switzerland

Background Percutaneous left atrial appendage (LAA) closure (LAAC) is a proven stroke prevention strategy for atrial fibrillation (AF) patients. Incomplete sealing in complex LAA anatomies may compromise efficacy. Objectives This study evaluates the safety, feasibility, and efficacy of concomitant dual-device LAAC in multilobed anatomies, representing the largest cohort examined to date Methods We reviewed all LAAC procedures at University Hospital of Bern (2009–2025). Baseline characteristics, procedural details, and follow-up data were analyzed for patients receiving dual-device LAAC. Endpoints included technical success, complications, thromboembolic events, and device-related issues. Continuous data were expressed as mean ± standard deviation or median, categorical data as percentages. Groups were compared using t-test, Mann–Whitney U test, or chi-square test. Differences were expressed as 95 % confidence intervals; P < 0.05 was considered significant. Results Of 1307 LAAC procedures, 10 included dual-device implantation. Mean age was 71 years, and all patients were men, CHA₂DS₂-VASc and HAS-BLED scores were 3.5±1.8 and 2.9±1.4, respectively. Most (70%) had paroxysmal AF. Pre-and intraprocedural transesophageal echocardiography (TEE) confirmed multilobed LAA anatomy in all cases. Half of the procedures were fluoroscopy-guided with one delivery sheath and transseptal puncture, while the other half were TEE-guided with a double-sheath and two transseptal punctures. Only Amplatzer devices were used: 8 procedures used two of the same type (5 Amulet, 3 Amplatzer Cardiac Plug (ACP)), and 2 combined different types (Amulet+ACP, ACP+ Amplatzer Vascular Plug). Technical success was achieved in all cases. Within the first week, one (10%) patient experienced a clinically non-clinically relevant pericardial effusion. At one-year follow-up (completed in 9 patients), 3 (30%) patients developed pericarditis. No thromboembolic events, device-related thrombosis, or device embolizations occurred. Conclusion In a small cohort of patients with complex multilobed LAA anatomy, concomitant implantation of two Amplatzer devices proved to be a feasible strategy with acceptable short-term safety, although potentially associated with an increased risk of pericarditis.