Shinichi Ohtani

Long-term outcome of patients who received Starr-Edwards valves between 1965 and 1977

To determine the risk factors relating to long-term outcomes of the Starr-Edwards ball valve, data from 150 patients (75 males, 75 females; age range 9-67 years); who underwent valve replacement (44 aortic, 106 mitral) between 1965 and 1977 at Niigata University were reviewed retrospectively. Valve models 6120/6320 and 1260/2320 were mainly used in mitral and aortic positions, respectively. The cumulative follow-up period was 2227.4 patient-years. Using the Kaplan-Meier life-table method, actuarial probability of freedom from valve-related mortality, morbidity and thromboembolism were analysed. Thirteen patients died from various causes during hospitalization. Some 80 patients suffered 97 instances of valve-related morbidity and 58 died during follow-up (up to 28 years). Patient age > 50 years affected the actuarial probability of freedom from valve-related mortality in the entire group (P < 0.05). Valve type and the year of operation were discriminative predictors of thromboembolic complications in the entire group (both P < 0.05) and in the mitral valve replacement group (P < 0.01 and P < 0.001). However, only the year of operation affected the actuarial probability of freedom from all valve-related morbidity in the entire group and in the mitral valve replacement group (both P < 0.05). Univariate analysis revealed that age at surgery, valve type, and the year of operation were related to late postoperative outcome in patients who received a Starr-Edwards ball valve between 1965 and 1977.