From January 1976 to September 1994, 26 patients underwent a tricuspid valve replacement (TVR) with porcine bioprostheses (Hancock II in 19 patients, Carpentier-Edwards in 3, Hancock standard in 2, Medtronic-Intact in 1, and Liotta in 1). The mean age of these patients was 38.4 years (range, 9 to 70 years). The mean follow-up period was 6.7 years (range, 1.8 to 16.5 years). There was a complete follow-up in 100%. There were two hospital deaths and three late deaths. The actuarial survival rate at 16 years was 78.7%+/-8.6%. Five patients developed structural deterioration. Four of these 5 patients underwent re-TVR. The actuarial freedom rate from structural deterioration at 10 years was 47.1%+/-19.1%. There were no instances of thromboembolism and prosthetic valve endocarditis. Our experience indicated the major problem of the porcine bioprostheses in the tricuspid position thus to be valve durability: however, the long-term actuarial survival rate was satisfactory, including no operative morality with re-TVR due to tricuspid prosthetic structural deterioration. We therefore conclude that the choice of a porcine bioprosthesis in the tricuspid position should be considered not only for patients with isolated tricuspid valve disease, but also for the patients who have not previously received mechanical valves in the mitral or the aortic valve position.