Twenty years' experience with the Model 6120 Starr-Edwards valve in the mitral position.

A total of 549 nonconsecutive patients underwent isolated mitral valve replacement with a Starr-Edwards valve prosthesis (Model 6120) at the University of Louvain (Belgium) from 1965 to 1985. Ninety-seven percent of the patients could be traced and only 17 patients were lost to follow-up. Cumulative follow-up totalled 3,130 patient-years. Actuarial and linearized statistical techniques were used to describe the survival and the incidence of valve-related complications (according to stringent criteria). Long-term overall survival rate including early deaths was 79% +/- 0.02% at 5 years, 65% +/- 0.03% at 10 years, and 54% +/- 0.04% at 19 years, with a linearized incidence of late deaths of 3.9% +/- 0.5% per patient-year. The incidence of late valve-related deaths was 1.25% per patient-year. Valve-related complications occurred at the following rates: thromboembolism 3.1% per patient-year, anticoagulant-related hemorrhage 1.08% per patient-year, endocarditis 0.26% per patient-year, reoperation and periprosthetic leak 0.45% per patient-year, and structural failure 0% per patient-year. All valve-related mortality and morbidity were calculated at 4.9% per patient-year and the rate of valve failure (deaths and reoperations) at 1.4% per patient-year. Among 376 survivors, 352 clinical functions could be obtained: 95% of patients belong to Class I or II of the New York Heart Association after operation versus 24% before operation. The study shows the structural durability of the Starr-Edwards mitral valve with a follow-up over 20 years. The Model 6120 valve may be considered a faithful standard.