Abstract 4369839: Thromboembolic Events and Bleeding Risk with Antithrombotic Strategies for Peri-Device Leak After Percutaneous Left Atrial Appendage Occlusion

Background: The 2023 American Heart Association guidelines recommend left atrial appendage occlusion (LAAO) for nonvalvular atrial fibrillation patients with oral anticoagulants (OAC) contraindications. Incomplete closure, or peri-device leak (PDL), has been associated with increased thromboembolic (TE) risk, but optimal antithrombotic management of moderate PDL (3–5 mm) remains unclear, leading to variability in clinical practice. Objective: This study compares TE and bleeding outcomes in patients with 3–5 mm PDL managed with direct OAC (DOAC) or dual antiplatelet (DAPT) therapy versus aspirin (ASA) monotherapy. Methods: This single-center study at The Ohio State University included patients with PDL (3–5 mm) after LAAO. Data on age, sex, CHA2DS2-VASc and HAS-BLED scores, and antithrombotic strategies were collected retrospectively. TE and bleeding events within one year were assessed. Chi-square and Mann–Whitney U tests were used for group comparisons. Logistic regression identified independent predictors of bleeding. Results: Between 2017 and 2024, 942 patients underwent LAAO. Transesophageal echocardiography at post-procedure day 45 identified a PDL measuring 3–5 mm (mean 3.6±0.7 mm) in 162 patients (17%). The mean age was 74±8 years, 44% were female. Mean CHA2DS2-VASc and HAS-BLED scores were 4.3±1.3 and 3.1±1.1, respectively. Within one year, 62 patients (38%) were treated with either therapeutic-dose DOAC (n=28) or DAPT (n=34), while 100 (62%) were on single antiplatelet therapy (SAPT); primarily aspirin (ASA, n=94) or clopidogrel (n=6). There were no significant differences between the DOAC/DAPT and ASA groups in CHA2DS2-VASc, HAS-BLED, or leak size. Overall, TE events occurred in 22 patients (14%) and bleeding incidents in 41 (25%). TE occurred in 16% of DOAC/DAPT patients and 10% of ASA patients with no significant difference between groups. Bleeding events were more frequent in the DOAC/DAPT group than in the ASA group (37% vs 16%, p<0.05). On logistic regression adjusting for HAS-BLED score, DOAC/DAPT use remained independently associated with higher bleeding risk (OR 2.81, p=0.007). Conclusion: Among patients with 3-5 mm PDL, ASA monotherapy was associated with significantly fewer bleeding events compared to DOAC or DAPT, with no difference in thromboembolic risk. Further prospective studies are warranted to guide antithrombotic therapy in this population.