BACKGROUND: There are scant data on long-term clinical outcomes and bioprosthetic-valve function after transcatheter aortic-valve replacement (TAVR) as compared with surgical aortic-valve replacement in patients with severe aortic stenosis and intermediate surgical risk. METHODS: We enrolled 2032 intermediate-risk patients with severe, symptomatic aortic stenosis at 57 centers. Patients were stratified according to intended transfemoral or transthoracic access (76.3% and 23.7%, respectively) and were randomly assigned to undergo either TAVR or surgical replacement. Clinical, echocardiographic, and health-status outcomes were followed for 5 years. The primary end point was death from any cause or disabling stroke. RESULTS: At 5 years, there was no significant difference in the incidence of death from any cause or disabling stroke between the TAVR group and the surgery group (47.9% and 43.4%, respectively; hazard ratio, 1.09; 95% confidence interval [CI], 0.95 to 1.25; P = 0.21). Results were similar for the transfemoral-access cohort (44.5% and 42.0%, respectively; hazard ratio, 1.02; 95% CI, 0.87 to 1.20), but the incidence of death or disabling stroke was higher after TAVR than after surgery in the transthoracic-access cohort (59.3% vs. 48.3%; hazard ratio, 1.32; 95% CI, 1.02 to 1.71). At 5 years, more patients in the TAVR group than in the surgery group had at least mild paravalvular aortic regurgitation (33.3% vs. 6.3%). Repeat hospitalizations were more frequent after TAVR than after surgery (33.3% vs. 25.2%), as were aortic-valve reinterventions (3.2% vs. 0.8%). Improvement in health status at 5 years was similar for TAVR and surgery. CONCLUSIONS: Among patients with aortic stenosis who were at intermediate surgical risk, there was no significant difference in the incidence of death or disabling stroke at 5 years after TAVR as compared with surgical aortic-valve replacement. (Funded by Edwards Lifesciences; PARTNER 2 ClinicalTrials.gov number, NCT01314313.).