Long term results were compared in 81 operative survivors with MVR using Hancock valve (Hancock group) and 83 using St. Jude Medical valve (SJM group). The cumulative follow-up was 709.0 patients-year (p-y) and 175.2 p-y in the Hancock and SJM groups. Twenty-four percent of the patients in the Hancock group was permanently anticoagulated, while 100% of patients in the SJM group anticoagulated. The 7-year actuarial survival rate including early mortality was 79.3 +/- 4.6% for the Hancock and 93.9 +/- 3.0% for the SJM group (p less than 0.01). The survival rate was significantly higher in the SJM group than that in the Hancock group. The 7-year actuarial event free rate of the valve-related complications in the Hancock and SJM groups were as follows; thromboembolism 88.6 +/- 3.8% vs 95.0 +/- 2.8% (NS), hemorrhage, 94.0 +/- 3.0% vs 98.7 +/- 1.8% (NS), paravalvular leak 92.6 +/- 3.2% vs 97.4 +/- 1.8% (NS), infection 93.9 +/- 3.0% vs 100% (p less than 0.05), valve malfunction 78.0 +/- 5.2% vs 100% (p less than 0.001), overall valve related complications 58.9 +/- 5.9% vs 91.1 +/- 3.5% (p less than 0.001), reoperation for valve-related complication 83.1 +/- 4.6% vs 100% (p less than 0.01). The event free rate of reoperation for valve related complication in the Hancock group was significantly lower than that in the SJM group. The Hancock valve had the acceptable antithrombogenicity, but had the limited long-term durability.(ABSTRACT TRUNCATED AT 250 WORDS)