Background: l fibrillation (AF) is a leading cause of ischemic stroke. The Left Atrial Appendage Occlusion Study (LAAOS) III showed that surgical left atrial appendage occlusion (LAAO) during cardiac surgery reduces stroke risk by one-third in patients with AF. Given the higher baseline stroke risk among women with AF, the benefit of surgical LAAO may differ based on sex. This secondary analysis of LAAOS III aimed to assess whether sex modifies the effect of LAAO on patient outcomes.
METHODS AND RESULTS: In LAAOS III, adults scheduled for on-pump cardiac surgery with a history of AF and a CHA2DS2 -VASc score ≥2 were randomized to concomitant LAAO during cardiac surgery versus no LAAO. Patients were excluded if they underwent mechanical-valve implantation, heart transplantation, surgery for congenital heart disease or implantation of a left ventricular assist device, or had had previous surgical opening of the pericardium or percutaneous LAAO. We built a multivariable logistic regression model with the following predictor variables: LAAO, female sex, age, heart failure, hypertension, diabetes, cerebrovascular event or systemic thromboembolism, and vascular disease. An interaction term between female sex and LAAO treatment was included to compare the full model with a reduced model (without the interaction) using a likelihood ratio test. Our outcomes of interest included ischemic stroke or systemic thromboembolism (primary), myocardial infarction, death from any cause, and major adverse cardiovascular events. We performed a landmark analysis for the primary outcome at ≤30 days after surgery and >30 days. Baseline characteristics were not balanced between men and women; more women had a history of heart failure, but more men had diagnoses of hypertension, diabetes, and vascular disease (Table 1). When adjusting for CHA2DS2 -VASc, female sex did not modify the effect of LAAO on ischemic stroke and systemic arterial embolism (p for interaction = 0.51), myocardial infarction (p for interaction = 0.31), death (p for interaction = 0.77) or major adverse cardiac events (p for interaction = 0.31) (Table 2). The landmark analysis for the primary outcome yielded consistent results.
Conclusion: Sex does not significantly alter the effects of surgical LAAO on stroke reduction or other secondary outcomes.
