Outcomes and feasibility of redo-TAVR after Sapien 3 Ultra TAVR in extremely-undersized versus nominally-sized annuli

Objectives To compare outcomes in Sapien 3 Ultra (S3U) transcatheter aortic valve replacement (TAVR) with extreme annular undersizing (EAU) versus nominal annular sizing (NAS). Background The Edwards S3U valve has reduced paravalvular leak (PVL) in TAVR but outcomes remain unknown in extremely undersized anatomy. Implanting a smaller S3U valve may facilitate future redo-TAVR but risk compromising hemodynamics. Methods From December 2019 to July 2021, 366 patients with native aortic stenosis underwent S3U TAVR. Patients with EAU (annular areas >430 mm(2) for 23 mm or >546 mm(2) for 26 mm) were compared to NAS (338-430 mm(2) for 23 mm or 430-546 mm(2) for 26 mm). In-hospital and 30-day outcomes, and redo-TAVR feasibility were determined. Results There were 79 (21.6%) EAU patients, with more bicuspid (p = 0.0014) and >= moderate annular/left ventricular outflow tract calcification (p < 0.001). The EAU group had less annular oversizing than NAS group (23 mm: -8.2 +/- 2.6% vs. 4.0 +/- 7.0%, p < 0.001; 26 mm: -8.9 +/- 2.2% vs. 6.7 +/- 6.9%, p < 0.001), more balloon overfilling (71.3% vs. 11.6%, p < 0.001), and postdilatation (15.0% vs. 5.8%, p = 0.016). No differences were found in in-hospital or 30-day mortality and stroke (p > 0.05). Mild PVL (13.4% EAU vs. 11.5% NAS, p = 0.56) and mean gradients (23 mm: 13.0 +/- 4.5 vs. 14.1 +/- 5.4 mmHg, p = 0.40; 26 mm: 11.4 +/- 4.1 vs. 11.5 +/- 3.9 mmHg, p = 1.0) were similar at 30 days. Had the EAU group undergone NAS with the larger Sapien 3/S3U, by computed tomography analysis simulating 80:20 or 90:10 target implant depth, 33.3%-60.9% (vs. 4.3%-23.2%) would not be feasible for redo-TAVR due to high risk of coronary obstruction. Conclusions In this first report of EAU with S3U TAVR, similar excellent short-term outcomes can be achieved compared to NAS, and may preserve future redo-TAVR option.