Early regression of severe left ventricular hypertrophy after transcatheter aortic valve replacement is associated with decreased hospitalizations.

Abstract

Objectives: This study sought to examine the relationship between left ventricular mass (LVM) regression and clinical outcomes after transcatheter aortic valve replacement (TAVR). Background: LVM regression after valve replacement for aortic stenosis is assumed to be a favorable effect of LV unloading, but its relationship to improved clinical outcomes is unclear. Methods: Of 2,115 patients with symptomatic aortic stenosis at high surgical risk receiving TAVR in the PARTNER (Placement of Aortic Transcatheter Valves) randomized trial or continued access registry, 690 had both severe LV hypertrophy (left ventricular mass index [LVMi] 149 g/m2 men, 122 g/m2 women) at baseline and an LVMi measurement at 30-day post-TAVR follow-up. Clinical outcomes were compared for patients with greater than versus lesser than median percentage change in LVMi between baseline and 30 days using Cox proportional hazard models to evaluate event rates from 30 to 365 days. Results: Compared with patients with lesser regression, patients with greater LVMi regression had a similar rate of all-cause mortality (14.1% vs. 14.3%, p ¼ 0.99), but a lower rate of rehospitalization (9.5% vs. 18.5%, hazard ratio [HR]: 0.50, 95% confidence interval [CI]: 0.32 to 0.78; p ¼ 0.002) and a lower rate of rehospitalization specifically for heart failure (7.3% vs. 13.6%, p ¼ 0.01). The association with a lower rate of rehospitalization was consistent across subgroups and remained significant after multivariable adjustment (HR: 0.53, 95% CI: 0.34 to 0.84; p ¼ 0.007). Patients with greater LVMi regression had lower B-type natriuretic peptide (p ¼ 0.002) and a trend toward better quality of life (p ¼ 0.06) at 1-year follow-up than did those with lesser regression. Conclusions: In high-risk patients with severe aortic stenosis and severe LV hypertrophy undergoing TAVR, those with greater early LVM regression had one-half the rate of rehospitalization over the subsequent year compared to those with lesser regression.