We propose to analyse the long-term follow-up in patients older than 65 years of age who received a mechanical valve in the aortic position, using death and prosthetic-related complications as endpoints. From April 1988 to December 1995, 144 consecutive patients 65-75 years of age (mean 67.7+/-2.5) were enrolled. Total duration of follow-up was 1663 patient-years (median 13.0 years) and was complete for 99% of the patients. Thirty-day mortality was 1.4% (n=2). At the end of the study, 77 patients (53.8%) were alive, with ages ranging from 77 to 91 years (mean 82.1+/-3.2 years). The overall 5-, 10- and 15-year actuarial survival was 87.4%+/-3.0, 67.7%+/-4.3 and 58.5%+/-4.5, respectively. Freedom from stroke was 93.3+/-3.1%, 84.6+/-3.3% and 71.7+/-4.5%, respectively, after identical periods. Freedom from major bleeding was 97.2+/-1.1%, 90.4+/-3.5% and 86.4+/-4.0%, respectively. Freedom from endocarditis was 95.7+/-2.3%, 95.0+/-2.1% and 94.4+/-2.5%, respectively, and freedom from reoperation was 98.0+/-1.2%, 97.6+/-1.3%, 96.9+/-2.4% and 96.4+/-2.6%, respectively. Freedom from major valve-related events was 87.7+/-2.6%, 73.9+/-3.4% and 61.5+/-4.6%, respectively. Nearly two-thirds of the patients were alive and free from major adverse valve-related events. Hence, we consider implantation of a mechanical prosthesis in elderly patients safe and appropriate, but the choice must be tailored for each specific patient.