AbstractThe safest duration of anticoagulation after idiopathic deep vein thrombosis (DVT) is unknown. We conducted a prospective study to assess the optimal duration of vitamin K antagonist (VKA) therapy considering the risk of recurrence of thrombosis according to residual vein thrombosis (RVT). Patients with a first unprovoked DVT were evaluated for the presence of RVT after 3 months of VKA administration; those without RVT suspended VKA, while those with RVT continued oral anticoagulation for up to 2 years. Recurrent thrombosis and/or bleeding events were recorded during treatment (RVT group) and 1 year after VKA withdrawal (both groups). Among 409 patients evaluated for unprovoked DVT, 33.2% (136 of 409 patients) did not have RVT and VKA was stopped. The remaining 273 (66.8%) patients with RVT received anticoagulants for an additional 21 months; during this period of treatment, recurrent venous thromboembolism and major bleeding occurred in 4.7% and 1.1% of patients, respectively. After VKA suspension, the rates of recurrent thrombotic events were 1.4% and 10.4% in the no-RVT and RVT groups, respectively (relative risk = 7.4; 95% confidence interval = 4.9-9.9). These results indicate that in patients without RVT, a short period of treatment with a VKA is sufficient; in those with persistent RVT, treatment extended to 2 years substantially reduces, but does not eliminate, the risk of recurrent thrombosis.