No abstract

Background:The left atrial volume index (LAVI) is important for predicting thromboembolism in patients with non-valvular atrial fibrillation (AF), but the utility of LAVI for predicting thromboembolism in patients with both bioprosthetic valve replacement and AF remains unclear. Methods and Results:Of 894 patients from a previous multicenter prospective observational registry (BPV-AF Registry), 533 whose LAVI data had been obtained by transthoracic echocardiography were included in this subanalysis. Patients were divided into tertiles (T1-T3) according to LAVI as follows: T1 (n=177), LAVI=21.5-55.3 mL/m 2 ; T2 (n=178), LAVI=55.6-82.1 mL/m 2 ; T3 (n=178), LAVI=82.5-408.0 mL/m 2 . The primary outcome was defined as either stroke or systemic embolism for a mean (±SD) follow-up period of 15.3±4.2 months. Kaplan-Meier curves indicated that the primary outcome tended to occur more frequently in the group with the larger LAVI (log-rank P=0.098). Comparison of T1 with T2 plus T3 using Kaplan-Meier curves indicated that patients in T1 experienced significantly fewer primary outcomes (log-rank P=0.028). Furthermore, univariate Cox proportional hazard regression showed that 1.3-and 3.3-fold more primary outcomes occurred in T2 and T3, respectively, than in T1. Conclusions:Larger LAVI was associated with stroke or systemic embolism in patients who had undergone bioprosthetic valve replacement and with a definitive diagnosis of AF.