From May 1982 to August 1991, 1200 patients underwent valve replacement with the St. Jude Medical (St. Jude Medical, Inc., St. Paul, Minn.) valve: 615 men (51%) and 585 women, mean age 58 years. Preoperatively, 830 patients (69%) were in functional class III or IV. A total of 611 patients (51%) had the aortic valve replaced, 490 (41%) the mitral valve, 2 (0.2%) the tricuspid valve, and 97 (8%) multiple valves. There were 81 hospital deaths (6.8%). Risk factors included older age (p = 0.0001), female gender (p = 0.02), higher preoperative left ventricular end-diastolic pressure (p = 0.05), previous cardiac operation (p = 0.003), longer aortic crossclamp time (p = 0.0001), and longer cardiopulmonary bypass time (p = 0.0001). Follow-up was 98% complete (3153 patient-years). There were 152 late deaths; 32 (21%) were considered valve-related: six thromboembolism, four valve thrombosis, five anticoagulant-related hemorrhage, eight prosthetic valve endocarditis, one paravalvular leak, and seven sudden death. The 5-year actuarial survival was 75%. Risk factors for late death included older age (p = 0.03), lower preoperative ejection fraction (p = 0.005), longer aortic crossclamp time (p = 0.001), longer cardiopulmonary bypass time (p = 0.0001), previous cardiac operation (p = 0.02), and higher preoperative functional class (p = 0.0001). Actuarial freedom at 5 years from major thromboembolic events and anticoagulant-related hemorrhage was 97% and 95%, respectively. This value for valve thrombosis was 99%, for reoperation 96%, for prosthetic valve endocarditis 98%, and for paravalvular leak 96%. Actuarial freedom from all valve-related events and valve-related death at 5 years was 74% and 94%, respectively. We conclude that the low incidence of valve-related events and low mortality supports the continued use of the St. Jude Medical valve.