Late results with Carpentier-Edwards porcine bioprosthesis.

From 1977 to 1984, 429 patients underwent aortic valve replacement (AVR), and 339 underwent mitral valve replacement (MVR) with a Carpentier-Edwards bioprosthesis. Early mortality for AVR was 4.6% (isolated AVR, 1.9%) and for MVR was 5.3% (isolated MVR, 4.1%). Follow-up was 99.3% complete at a mean of 5.9 years. Actuarial event-free rates at 10 years for AVR and MVR were, respectively, 1) for structural valve deterioration, 91.4 +/- 3.2% versus 75.1 +/- 4.0% (p less than 0.01); 2) for nonstructural dysfunction, 100% versus 97.8 +/- 1.6% (p = NS); 3) for thromboembolism, 90.6 +/- 2.3% versus 87.3 +/- 2.6% (p = NS); 4) for anticoagulant-related bleeding, 95.3 +/- 1.1% versus 88.6 +/- 2.4% (p = 0.05); 5) for endocarditis, 87.8 +/- 5.7% versus 90.6 +/- 2.4% (p = NS); 6) for reoperation, 91.0 +/- 2.5% versus 74.4 +/- 3.7% (p less than 0.01); 7) for valve-related mortality, 76.1 +/- 6.9% versus 71.4 +/- 5.2% (p = 0.01); 8) for permanent physical impairment, 85.0 +/- 3.0% versus 71.5 +/- 3.6% (p less than 0.01); and 9) for combined operative mortality, valve-related mortality, and reoperation, 68.7 +/- 6.4% versus 51.5 +/- 4.9% (p = 0.01). No structural valve dysfunction was observed in any AVR patient whose valve was inserted after age 70. Age at operation was the only factor that predicted structural valve deterioration (p less than 0.01).