SELECTIVE STRATEGY USING BALLOON-EXPANDABLE VERSUS SELF-EXPANDING TRANSCATHETER AORTIC VALVE PROSTHESIS TO MINIMIZE PARAVALVULAR AORTIC VALVE REGURGITATION

Abstract

More than mild paravalvular aortic regurgitation (PAR) after transcatheter aortic valve implantation (TAVI) is associated with increased morbidity and mortality. Aortic valve prosthesis sizing by multidetector computed tomography (MDCT) 3-dimensional aortic annular dimensions reduces the incidence of PAR. We hypothesize that MDCT based aortic annular dimensions based strategy selecting balloon expandable (BE) vs. self expanding (SE) aortic valve prosthesis may further reduce the incidence of PAR. We started our TAVI program using BE valves only (Group 1, n=78) with MDCT based aortic annular sizing. We identified that aortic annulus ellipticity ratio (the ratio of the shortest and longest diameter of the aortic annulus) and the degree of aortic prosthesis oversizing [100 × (Prosthetic valve area - aortic annular area) / prosthetic valve area] independently predicted more than mild PAR. In second phase of our TAVI program we used both BE and SE prosthesis. We postulated that a SE prosthesis may conform to the contours of aortic annulus with higher degree of oversizing (DOS), though with less radial force, thereby minimizing risk of annular rupture. We adopted a preferred strategy of using SE prosthesis with aortic annulus ellipticity ratio (AAER) of < 0.7. Group 2 (n=82) comprised of 51 (62.2%) BE valves and 31 (37.8%) SE valves. PAR was assessed by transthoracic echocardiogram done within 30 days of TAVI procedure. In Group 1 with BE valves, AAER of < 0.7 was present in 12.8% of patients with multivariate analysis showing AAER (p=0.02) and DOS (p=0.005) independently predicting PAR. In Group 2 with selective use of BE and SE valves (n=82), AAER was present in 9.8% of patients. None of the patients with BE valves (n=51) had AAER < 0.7 vs. those with SE valves (n=31) AAER < 0.7 was present in 25.8% of patients (p< 0.001). DOS was 11.6% with BE valves and 19.5% with SE valves (p < 0.001). The incidence of more than mild PAR was 12.8% in Group 1. With selective strategy in Group 2 there was no patient with more than mild PAR at 30 days. Subgroup analysis showed similar results for second and third generation BE valves. Selective strategy of using SE valves based on AAER of < 0.7 and careful oversizing of prosthesis was associated with marked reduction in PAR at 30 days. The reduction in PAR is expected to improve clinical outcomes.