Comparative evaluation of aortic valve replacement with Starr, Björk, and porcine valve prostheses.

Three groups of 100 consecutive patients with aortic valve disease who were operated on between 1974 and 1978 underwent long-term evaluation. There were 100 aortic valve replacements with porcine bioprosthetic valves (group I), 100 with Starr valves (group II), and 100 with Björk valves (group III). There were no significant differences in the preoperative clinical conditions of the patients in the three groups. Cumulative follow-up was 1688 patient-years. Incidence of valve-related death at 8 years was 4 +/- 2.3% in group I, 13 +/- 3.6% in group II, and 13 +/- 3.8% in group III (p less than .05). At 8 years 95 +/- 2.8% of the patients in group I were free of thromboembolism, compared with 81 +/- 4.8% of those in group II and 84 +/- 4.2% of those in group III (p less than .002). The actuarial risk of a reoperation at 8 years was 16 +/- 6% in group I, 5 +/- 2% in group II, and 2 +/- 1.6% in group III (p less than .025 group I vs group III). At 8 years 98 +/- 1.2% of the patients in group I were free of anticoagulant-related complications, compared with 88 +/- 3.8% of those in group II and 86 +/- 3.9% of those in group III (p less than .005). We conclude that at 8 years porcine bioprosthetic valves performed better than mechanical valves, taking into consideration thromboembolism, anticoagulant-related hemorrhage, and valve-related death.