Aortic valve replacement: a ten-year follow-up of non-cloth-covered vs cloth-covered caged-ball prostheses.

From 1965 to 1976, 721 isolated aortic valve replacements were performed at the University of Oregon hospitals, utilizing Starr-Edwards caged-ball prostheses. Three models of aortic prostheses were introduced during this period: a non-cloth-covered model has been in continuous use since 1965; a cloth-covered model was begun in 1968 and has been supplanted by the modified composite-strut or "track" model since 1972. The 5-year actuarial survival rate for operative survivors is about 80% for both non-cloth-covered and cloth-covered valves, while the 10-year survival is 61%, based on the older model only. The actuarially-determined percentages of patients experiencing significant thromboembolic episodes (i.e., all except transient ischemic episodes) at 5 years are 7% for the cloth-covered and 9% for the non-cloth-covered model. If transient ischemic attacks are included, the cloth-covered model has only an 8% incidence at 5 years compared to a 22% incidence for the older model. However, the cloth-covered valves are subject to a higher risk of reoperation because of the possibility of cloth injury. The "track" valve, therefore, was designed with exposed metal on the inner surface of each strut to prevent ball-cloth contact. In 107 patients (mean follow-up period 1 year) receiving anticoagulation, this prosthesis has maintained the same low incidence of thromboembolism as the previous cloth-covered model, with no reoperations for valve failure.