BACKGROUND A retrospective study was conducted to analyze the results of St. Jude Medical mitral valve replacement. METHODS From January 1979 to December 1989, 870 patients (54% women, 46% men; mean age, 55.8 +/- 6.2 years) underwent mitral valve replacement with the St. Jude Medical prosthesis. Of these operations 616 were isolated mitral valve replacements and 254 were double valve replacements. Coronary artery bypass grafting was performed concomitantly in 55 patients (6.3%). RESULTS Overall, early mortality was 5.05%, with 4.2% for the isolated mitral valve procedure and 7.08% for the double valve replacement. Follow-up at 15 years was complete in 859 patients (98.74%). Mean follow-up time was 93.5 months, for a total of 6,436 years. Actuarial survival at 15 years was 59.5% +/- 5%, 60.5% +/- 6%, and 56.9% +/- 9%, for the entire group, the isolated mitral valve and double valve procedures, respectively. Multivariate analysis identified age, sex, hospital stay, and preoperative mitral regurgitation as independent prognosis factors for overall mortality. Of 606 patients alive at the latest follow-up, the New York Heart Association class improved significantly (from 67% class III/IV before the operation to 88% class I/II after the operation). All patients received warfarin to maintain an international normalized ratio between 3.5 and 4. The linearized rates (% per patient-year) of thrombosis, thromboembolism, and major hemorrhage were, respectively, 0.21, 0.75, and 0.94 for the entire group; 0.18, 0.67, and 0.88 for the isolated mitral valve operation; and 0.15, 0.92, and 1.08 for the double valve replacement. For the entire group the freedom from thrombosis and thromboembolism at 15 years was 98.1% +/- 1% and 88% +/- 4%, respectively. No case of structural dysfunction occurred. The freedom from paravalvular leak and endocarditis at 15 years was 95.3% +/- 2% and 97.3% +/- 2.4%, respectively. The probability of remaining free from reoperation at 15 years was therefore 95.6% +/- 2.5%. CONCLUSIONS These results confirm that the St. Jude Medical valve is a reliable prosthesis with very low thrombosis and thromboembolism rates, allowing the use of a low dose of anticoagulation with an international normalized ratio of about 3.