BACKGROUND AND AIM OF THE STUDY: The study aim was to evaluate our clinical experience with the Xenomedica heart valve prosthesis, a low-pressure glutaraldehyde-preserved porcine aortic valve with low-profile mounting. METHODS: Between January 1983 and July 1990, 242 consecutive patients (75 men, 167 women; mean age 59.8+/-8.0 years; range: 32-77 years) underwent mitral valve replacement with the Xenomedica prosthesis. Preoperatively, patients were in NYHA classes III (66%) and IV (26%); 94 patients (39%) had undergone previous cardiac surgery and 201 (83%) had chronic atrial fibrillation. Etiology was rheumatic in 51%, myxomatous in 7%, ischemic in 1%, endocarditis in 2%, and due to dysfunction of a previously implanted device in 39%. In total, 115 (47%) patients underwent an associated procedure. Mean follow up was 142+/-24 months (range: 2-181 months); total follow up was 2,627 patient-years. RESULTS: Early mortality was 8.2% (14 patients with low-output syndrome, three with multi-organ failure, one with stroke and two with major bleeding, 2). Late mortality was 45% (3.8%/pt-year) (84 cardiac deaths, 38 being valve-related). Actuarial survival at 5, 10 and 15 years was 69+/-3%, 52+/-3% and 38+/-4%, respectively. Actuarial estimates of freedom from structural valve deterioration (SVD) at 5, 10 and 15 years were 93+/-2%, 64+/-4%, and 25+/-9%; in almost all cases SVD occurred in progressive fashion. At 5, 10 and 15 years, estimates of freedom from thromboembolism were respectively 90+/-2%, 83+/-3% and 83+/-3%, for anticoagulant-related hemorrhage 96+/-1%, 88+/-3% and 88+/-3%, for endocarditis 96+/-1%, 94+/-2% and 94+/-2%, and for reoperation 93+/-2%, 67+/-4% and 54+/-6%. Estimates of freedom from all valve-related mortality at 5, 10 and 15 years were 87+/-2%, 80+/-3% and 75+/-4%. Multivariate analysis (Cox model) showed younger age to be a significant risk factor for reoperation. CONCLUSION: Long-term results with the Xenomedica device implanted in the mitral position appear in line with those achieved for other first-generation bioprostheses. However, incidence of primary tissue failure within 10 years was unsatisfactory. Although sudden dysfunction of the device never occurred, close monitoring of survivors is warranted.