Echocardiographic and clinical factors related to paravalvular leak incidence in low-gradient severe aortic stenosis patients post-transcatheter aortic valve implantation

Background: Paravalvular leak (PVL) following transcatheter aortic valve implantation (TAVI) is associated with poor outcomes. The association of clinical and echocardiographic parameters in the occurrence of even mild PVL in patients with low-gradient severe aortic stenosis (LGSAS) after TAVI and predictors of all-cause mortality and re-hospitalization during 2-year follow-up was studied, as data are sparse in the literature. Methods: Seventy-five consecutive LGSAS patients (age 80±8 years, 68% male, 89% Caucasians) were enrolled and followed-up for PVL development at hospital discharge, as well as mortality or re-hospitalization for cardiovascular disease, during a 2-year follow-up period. Patients' characteristics were retrospectively extracted from hospital's database. Mortality was determined either by chart review or the Social Security Death Index. Results: PVL was evident in 29 out of 75 patients. Patients with PVL had higher death rates at 6-month, 1-year of follow-up (27 vs. 7%, P = 0.05, 32 vs. 8%, P = 0.04, respectively) and lower stroke rates (0 vs. 7%, P = 0.009), compared with those who had no PVL. Patients with PVL were older, had lower BMI, higher prevalence of atrial fibrillation, higher brain natriuretic peptide (BNP), lower rest left ventricular ejection fraction, higher aortic cross-sectional area (Aa), lower stroke volume index, higher aortic resistance, stroke work, and higher energy loss index at rest and during dobutamine stress. Multivariable predictors of PVL post-TAVI were pre-procedural BNP levels and Aa. Conclusions: PVL remains a common complication after TAVI, related to adverse events in LGSAS patients. Aa and BNP levels remain the most important predictors of PVL.