Deformation of Transcatheter Aortic Valve Prostheses: Implications for Hypoattenuating Leaflet Thickening and Clinical Outcomes

Abstract

Background: Although transcatheter aortic valve replacement (TAVR) therapy continues to grow, there have been concerns about the occurrence of hypoattenuating leaflet thickening (HALT), which may affect prosthesis function or durability. This study aimed to examine prosthesis frame factors and correlate their extent to the frequency of HALT and clinical outcomes. Methods: We prospectively examined 565 patients with cardiac computed tomography screening for HALT at 30 days after balloon-expandable SAPIEN3 and self-expanding EVOLUT TAVR. Deformation of the TAVR prostheses, asymmetric prosthesis leaflet expansion, prosthesis sinus volumes, and commissural alignment were analyzed on the postprocedural computed tomography. For descriptive purposes, an index of prosthesis deformation was calculated, with values >1.00 representing relative midsegment underexpansion. A time-to-event model was performed to evaluate the association of HALT with the clinical outcome. Results: Overall, HALT was present in 21% of SAPIEN3 patients and in 16% of EVOLUT patients at 30 days after TAVR. The occurrence of HALT was directly associated with greater prosthesis frame deformation ( P <0.001), worse asymmetry of the leaflets ( P <0.001), and smaller TAVR neosinus volumes ( P <0.001). These relations were present in both prosthetic types and in all of their size ranges (all P <0.05). In multivariable analyses that include clinical variables previously associated with HALT (eg, anticoagulant therapy), variables of TAVR prosthesis deformation remained predictive of HALT. Although HALT was not associated with changes in prosthetic hemodynamics, its presence was associated with the risk of mortality at 1 year, with respect to greater incidences of all-cause mortality (hazard ratio, 2.98 [95% CI, 1.57–5.63]; P =0.001), cardiac death (hazard ratio, 4.58 [95% CI, 1.81–11.6]; P =0.001), and a composite outcome of all-cause mortality and heart failure hospitalization (hazard ratio, 1.94 [95% CI, 1.14–3.30]; P =0.02) with adjustment for age, sex, and comorbidities. Conclusions: Nonuniform expansion of TAVR prostheses resulting in frame deformation, asymmetric leaflet, and smaller neosinus volume is related to occurrence of HALT in patients who undergo TAVR. These data may have implications for both prosthesis valve design and deployment techniques to improve clinical outcomes for these patients.