The porcine bioprosthetic valve was used in 440 patients having isolated mitral valve replacement (MVR), 522 patients having isolated aortic valve replacement (AVR), and 88 patients having MVR + AVR between 1974 and 1981. Patients with associated surgical procedures were excluded. Mean follow-up was 8.3 years. At 10 years, there was no difference in patient survival between those having AVR and those having MVR. Reoperations were performed on 192 patients. Endocarditis was the reason for reoperation in 3.7% of patients who had MVR and 10.6% of those who had AVR. Structural valve degeneration was the reason for reoperation in 89.7% of MVR patients and 78.8% of AVR patients (p = 0.04). Hospital mortality among patients having valve reoperations was 4.7%. At 10 years, the freedom from valve reoperation for all causes and from structural valve degeneration was significantly better for the AVR group than the MVR group (74% +/- 3% versus 61% +/- 4%, p = 0.004; and 79% +/- 3% versus 63% +/- 4%, p = 0.0006, respectively). For patients in their 60s, the 10-year freedom from reoperation was 92% +/- 2% for AVR and 80% +/- 6% for MVR (p = not significant). At 10 years, freedom from cardiac-related death and valve reoperation was best for both MVR and AVR patients in their 60s. Patients 70 years old or older rarely had reoperation but died before valve failure occurred. The 10-year freedom from all major valve-related events (cardiac-related death, reoperation, thromboembolism, endocarditis, and anticoagulant-related bleeding) was practically the same for both MVR and AVR patients (48% +/- 3% versus 49% +/- 3%, respectively). The porcine bioprosthetic valve is the valve of choice only for patients 60 years old or older. Patients in their 70s have an extremely low rate of reoperation but a high rate of cardiac-related death and do not outlive the prostheses.