Hemodynamic and Echocardiographic Comparison of the Lotus and CoreValve Transcatheter Aortic Valves in Patients With High and Extreme Surgical Risk

Abstract

Background: Comparative echocardiographic data on transcatheter aortic valve replacement systems from randomized trials are limited. The REPRISE III trial (Repositionable Percutaneous Replacement of Stenotic Aortic Valve through Implantation of Lotus Valve System – Randomized Clinical Evaluation) is a multicenter, randomized comparison of a mechanically expanded (Lotus) versus self-expanding (CoreValve) transcatheter aortic valve replacement device. This analysis rigorously assesses Doppler-derived valve hemodynamics and the impact on outcomes at 1 year in patients with extreme/high surgical risk treated with Lotus and CoreValve from REPRISE III. Methods: REPRISE III includes patients with extreme- and high-risk aortic stenosis. Patients were enrolled at 55 centers. All transthoracic echocardiograms with Doppler were obtained following a standard protocol up to 12 months postimplant and analyzed by a core laboratory. Valve size, mean gradient, aortic valve area, and Doppler velocity index and their impact on clinical outcomes are reported. Additional parameters including paravalvular leak were evaluated using a multiparametric approach. Results: A total of 912 patients were randomly assigned (2:1 ratio; 607 Lotus:305 CoreValve). Median age was 84 years, 51% of the patients were women, and the Society of Thoracic Surgeons score was 6.8±4.1. CoreValve demonstrated lower gradients and larger aortic valve area and Doppler velocity index than Lotus at discharge; the difference decreased in subsequent follow-up up to a year (all P <0.01). Lotus had lower rates of paravalvular leak that persisted over time ( P <0.05). Similar outcomes were seen when comparing each valve type by size group (small, medium, large). The hemodynamic differences between valves did not translate into worse clinical outcomes. All-cause mortality was not different between the 2 groups in any of the 3 valve sizes. When comparing patients with normal valve gradients (<20 mm Hg, n=780) with those with abnormal gradients (>20 mm Hg, n=48) in the entire patient population, all-cause mortality was not different. This was also not significant when evaluating each valve type separately. Similarly, there were no differences for aortic valve area >1.1 cm 2 or <1.1 cm 2 and for Doppler velocity index >0.35 or <0.35 (all P =not significant). Conclusions: Lotus had significantly greater freedom from moderate or severe paravalvular leak and smaller valve area and higher gradients than CoreValve. The hemodynamic differences were not associated with any clinical differences in the composite end point of mortality, disabling stroke, and moderate paravalvular leak or with quality of life at 1 year of follow-up. Clinical Trial Registration: URL: https://www.clinicaltrials.gov . Unique identifier: NCT02202434.