Abstract
In our sequential studies of 67 and 21 patients, testosterone therapy (TT) interacted with thrombophilia⁻hypofibrinolysis, leading to venous thromboembolism (VTE). Compared to 111 VTE controls not taking TT (VTE-no TT), the 67 and 21 cases were more likely (p < 0.05 for all) to have Factor V Leiden (FVL) heterogeneity (24% and 33% vs. 12%), the lupus anticoagulant (14% and 33% vs. 4%), and high lipoprotein(a) (33% vs. 13%, n = 21). After a first VTE and continuing TT, 11 thrombophilic cases had a second VTE despite adequate anticoagulation, 6 of whom, still anticoagulated, had a third VTE. The greatest density of thrombotic events was at three months after starting TT, with a rapid decline by 10 months. From <1 to 8 months after starting TT, 65% of VTE occurred, which may reflect TT-induced depletion of susceptible thrombophilic patients, leaving a winnowed residual group with fewer VTE events despite the continuation of TT. Before starting TT, we suggest screening for FVL, lipoprotein(a), and the lupus anticoagulant to identify patients at increased VTE risk, with an adverse risk-to-benefit ratio for TT. We suggest that TT should not be started in patients with known thrombophilia⁻hypofibrinolysis, and should not be continued after a first VTE. When TT is given to patients with thrombophilia⁻hypofibrinolysis, VTE may occur and then recur despite adequate anticoagulation.