OBJECTIVE The study aim was to examine the long-term durability of the aortic Carpentier-Edwards Perimount pericardial bioprosthesis using actuarial and actual analyses. METHODS A total of 267 patients were implanted at four centers between September 1981 and December 1983. Of these patients, 171 (64%) were males and 96 (36%) females; mean age at implant was 64.9+/-11.8 years (range: 21 to 86 years). Patients have been followed for 9.1+/-4.2 years (total 2335.7 patient-years). Long-term echocardiography data are presented. RESULTS The total operative (<30 days postoperative) mortality rate was 4.9%; of this, 0.4% was valve-related. The total late (> or = 30 days postoperative) mortality rate was 6.2%/pt-yr and included a valve-related mortality rate of 1.6%/pt-yr. Complications of thromboembolism, thrombosis and bleeding showed linearized rates of 1.6%/pt-yr and 0.4%/pt-yr, respectively. Valve dysfunction resulted in an explant rate of 0.9%/pt-yr and an associated mortality rate of 0.1%/pt-yr. At 14 years post implant, actuarial freedom from overall and valve-related death was 39.3% and 78.8%, respectively. Actuarial and actual freedom from valve dysfunction was 70.4% and 81.7%. Actuarial freedom from valve explant as a result of dysfunction was 85.1% in all patients; explant in patients aged < or = 65 years at implant was less (76.1%) than in patients aged >65 years (96.3%). CONCLUSIONS The high actuarial and actual freedom from explant due to structural valve dysfunction supports the long-term durability of this pericardial bioprosthesis and justifies its clinical use in patients older than 65 years at implant.