Case-control studies have indicated increased risk of venous thrombosis associated with the prothrombin gene G20210A polymorphism and with elevated plasma prothrombin levels. We sought to confirm these results in a prospective population-based study of 21,690 persons. We measured G20210A and prothrombin antigen on pre-event blood samples of 302 participants who developed venous thromboembolism (VTE) and 626 participants who remained free of VTE. Approximately 4.0% of cases and 2.4% of controls carried the G20210A polymorphism, but only one of 137 African Americans did. The odds ratio in whites was 1.87 (95% CI = 0.85, 4.11)--higher for those who reported a prior history of VTE (OR = 5.44) than those reporting no VTE history (OR = 1.41) and in those with idiopathic VTE (OR = 2.51) than those with secondary VTE (OR = 1.38). There was no association between venous thromboembolism and plasma prothrombin antigen level. We estimated that the G20210A polymorphism may account for approximately 2.5% of venous thromboembolism events in United States whites.
Copyright 2002 Wiley-Liss, Inc.