ABSTRACT NUMBER: ESOC2026A1020 LEFT ATRIAL APPENDAGE OCCLUSION FOR STROKE PREVENTION IN PATIENTS WITH ATRIAL FIBRILLATION AND MALIGNANCY: A SYSTEMATIC REVIEW AND META-ANALYSIS OF 177,896 PATIENTS

Abstract Background and aims Atrial fibrillation (AF) is highly prevalent in cancer patients, creating a "competing risk" scenario where prothrombotic states increase stroke risk while malignancy-related factors (thrombocytopenia, mucosal fragility) amplify bleeding risks with oral anticoagulation. Percutaneous left atrial appendage occlusion (LAAO) is a non-pharmacological alternative, yet its safety and efficacy in this high-risk population are not fully established. We aimed to synthesize global evidence on LAAO outcomes in the cardio-oncology population. Methods We searched PubMed, Scopus, and Web of Science through August 2025 for studies comparing LAAO outcomes in AF patients with and without cancer. Quantitative synthesis was performed using random-effects models. Results Seven studies (n=177,896) were included, with five (n=61,632) contributing to the meta-analysis. LAAO demonstrated comparable stroke prevention in cancer versus non-cancer cohorts (OR 0.74; 95% CI, 0.41–1.34; p=0.32). While overall bleeding risk appeared similar (p=0.08), sensitivity analysis excluding a single large dataset revealed an increased risk in cancer patients (OR 1.62; 95% CI, 1.02–2.56; p=0.04). Cancer patients faced significantly higher risks of pericardial complications (OR 2.22; p<0.00001) and all-cause mortality (HR 1.57; p=0.002). Mortality likely reflects underlying oncological prognosis rather than procedural failure. Conclusions LAAO provides effective stroke prevention for cancer patients with AF, achieving ischemic outcomes comparable to the general population. However, the increased risk of procedural complications and mortality necessitates meticulous patient selection and specialized periprocedural care in this complex cardio-oncology population. Conflict of interest All authors declare that they have nothing to disclose.