Long-term comparative analysis of the Björk-Shiley and Hancock valves implanted in 1975.

The long-term results in all patients undergoing isolated mitral, aortic, or double mitral-aortic heart valve replacement operated upon in 1975 has been retrospectively analyzed. A total of 153 patients received the standard Björk-Shiley (flat pyrolytic disc) mechanical prostheses and 150 patients received the noncomposite Hancock porcine xenograft. Overall operative mortality was not significantly different between groups. All patients receiving a Björk-Shiley prosthesis, but none in the Hancock group, received long-term anticoagulant therapy. Medium and long-term actuarial survival rates (5 and 10 years postoperatively) were comparable for the two groups (88% for Björk-Shiley and 84% for Hancock [NS] at 5 years; 86% for Björk-Shiley and 80% for Hancock at 10 years [NS]). The incidence of systemic embolism was similar in the two groups (1.6% +/- 0.4% per patient-year for the Björk-Shiley group and 1.3% +/- 0.3% per patient-year for the Hancock group [NS]). Also the incidence of endocarditis was similar (0.6% +/- 0.2% per patient-year for the Björk-Shiley group and 0.8% +/- 0.3% per patient-year for the Hancock group [NS]). In the Hancock group the overall incidence of reoperations was significantly higher than in the Björk-Shiley group (4.2% +/- 0.6% per patient-year versus 0.9% +/- 0.3% per patient-year (p = 0.001). The major cause for reoperation in the Hancock group was primary tissue failure (3% +/- 0.5% per patient-year). In the Björk-Shiley group the major cause of reoperation was valve thrombosis (0.5% +/- 0.2% per patient-year). Therefore, accepting the fact that other bioprostheses may behave differently from the Hancock noncomposite xenograft, we currently restrict our indications for valve replacement with bioprostheses.