In the SCOPE I trial (Safety and Efficacy of the Symetis ACURATE Neo/TF Compared to the Edwards SAPIEN 3 Bioprosthesis), transcatheter aortic valve implantation with the self-expanding ACURATE neo (NEO) did not meet noninferiority compared with the balloon-expandable SAPIEN 3 (S3) device regarding a composite end point at 30 days due to higher rates of prosthetic valve regurgitation and acute kidney injury. Data on long-term durability of NEO are scarce. Here, we report whether early differences between NEO and S3 translate into differences in clinical outcomes or bioprosthetic valve failure 3 years after transcatheter aortic valve implantation. Patients with severe aortic stenosis were randomized to transfemoral transcatheter aortic valve implantation with NEO or S3 at 20 European centers. Clinical outcomes at 3 years are compared using Cox proportional or Fine-Gray subdistribution hazard models by intention-to-treat. Bioprosthetic valve failure is reported for the valve-implant cohort. Among 739 patients, 84 of 372 patients (24.3%) had died in the NEO and 85 of 367 (25%) in the S3 group at 3 years. Comparing NEO with S3, the 3-year rates of all-cause death (hazard ratio, 0.98 [95% CI, 0.73-1.33]), stroke (subhazard ratio, 1.04 [95% CI, 0.56-1.92]), and hospitalization for congestive heart failure (subhazard ratio, 0.74 [95% CI, 0.51-1.07]) were similar between the groups. Aortic valve reinterventions were required in 4 NEO and 3 S3 patients (subhazard ratio, 1.32 [95% CI, 0.30-5.85]). New York Heart Association functional class ≤II was observed in 84% (NEO) and 85% (S3), respectively. Mean gradients remained lower after NEO at 3 years (8 versus 12 mm Hg; P<0.001). Early differences between NEO and S3 did not translate into significant differences in clinical outcomes or bioprosthetic valve failure throughout 3 years. URL: https://clinicaltrials.gov, Unique identifier: NCT03011346.