Transcatheter Replacement of Failed Bioprosthetic Valves

Abstract

© 2016 American Heart Association, Inc.Background - Transcatheter valve implantation inside failed bioprosthetic surgical valves (valve-in-valve [ViV]) may offer an advantage over reoperation. Supra-annular transcatheter valve position may be advantageous in achieving better hemodynamics after ViV. Our objective was to define targets for implantation that would improve hemodynamics after ViV. Methods and Results - Cases from the Valve-in-Valve International Data (VIVID) registry were analyzed using centralized core laboratory assessment blinded to clinical events. Multivariate analysis was performed to identify independent predictors of elevated postprocedural gradients (mean ≥20 mm Hg). Optimal implantation depths were defined by receiver operating characteristic curve. A total of 292 consecutive patients (age, 78.9±8.7 years; 60.3% male; 157 CoreValve Evolut and 135 Sapien XT) were evaluated. High implantation was associated with significantly lower rates of elevated gradients in comparison with low implantation (CoreValve Evolut, 15% versus 34.2%; P=0.03 and Sapien XT, 18.5% versus 43.5%; P=0.03, respectively). Optimal implantation depths were defined: CoreValve Evolut, 0 to 5 mm; Sapien XT, 0 to 2 mm (0-10% frame height); sensitivities, 91.3% and 88.5%, respectively. The strongest independent correlate for elevated gradients after ViV was device position (high: odds ratio, 0.22; confidence interval, 0.1-0.52; P=0.001), in addition to type of device used (CoreValve Evolut: odds ratio, 0.5; confidence interval, 0.28-0.88; P=0.02) and surgical valve mechanism of failure (stenosis/mixed baseline failure: odds ratio, 3.12; confidence interval, 1.51-6.45; P=0.002). Conclusions - High implantation inside failed bioprosthetic valves is a strong independent correlate of lower postprocedural gradients in both self- and balloon-expandable transcatheter valves. These clinical evaluations support specific implantation targets to optimize hemodynamics after ViV.