Haemodynamic and clinical outcomes at 5 years according to predicted prosthesis-patient mismatch after transcatheter aortic valve replacement.

ABSTRACT

BACKGROUND/PURPOSE: Although the impact of predicted prosthesis-patient mismatch (PPMP) on outcomes after surgical aortic valve replacement is well established, studies on PPMP in transcatheter aortic valve replacement (TAVR) are limited. This study investigated the effects of PPMp on haemodynamic and 5-year clinical outcomes after TAVR. METHODS/MATERIALS We analysed 1733 patients who underwent TAVR. PPMp was defined using two different methods: 1) normal reference values of the effective orifice area for each valve type and size indexed to body surface area (PPMp1; n = 1733) and 2) reference values for aortic annulus area or perimeter assessed with pre-procedural computed tomography indexed to body surface area (PPMp2; n = 1227). The primary endpoint was the composite of all-cause death and/or rehospitalisation for heart failure at 5 years. RESULTS: The incidence of PPMp1 was 11.7 % and 0.8 % in moderate and severe cases, respectively. PPMp2 was classified as either moderate (3.8 %) or severe (0 %). Rates of residual mean aortic gradient ≥20 mmHg significantly increased depending on PPMp1 severity (no PPMp1 3.1 % vs. moderate PPMp1 26.8 % vs. severe PPMp1 53.9 %, p < 0.0001) and PPMp2 (no PPMp2 4.1 % vs. moderate PPMp2 12.8 %, p = 0.0049). Neither of PPMP methods were associated with the composite outcome in total cohort; however, PPMP1 was significantly related to worse clinical outcomes at 5 years among patients with reduced left ventricular ejection fraction (LVEF) in multivariate analysis (HR 1.87; 95 % CI 1.02-3.43). CONCLUSIONS: The impact of PPMP on TAVR clinical outcomes may not be negligible in patients with low LVEF.