OBJECTIVES: There is growing interest in using residual vein obstruction (RVO) to guide the duration of oral anticoagulant therapy (OAT) for unprovoked deep vein thrombosis (DVT). We sought to determine if RVO as determined by compression ultrasonography (CUS) after completion of 5-7 months of anticoagulation for unprovoked DVT is associated with an increased risk of recurrent venous thromboembolism (VTE). MATERIALS AND METHODS: This was a multicentre multinational prospective cohort study undertaken in tertiary care centers. Patients with a first 'unprovoked' major VTE were enrolled over a 4-year period and completed a mean 18-month follow-up in September 2006. All 452 patients with DVT had baseline CUS at inclusion to assess any RVO before stopping OAT at 5-7 months. During follow-up off OAT, all episodes of suspected recurrent VTE were independently adjudicated with reference to baseline imaging. RESULTS: Forty-five out of 231 patients with abnormal CUS (19.5%) had recurrent VTE during follow-up, as compared with 32 out of 220 patients with normal CUS (14.6%), and one patient had inadequate CUS. There was no significant association between an abnormal CUS at inclusion and the risk of recurrent VTE: hazard ratio 1.4 (95% confidence interval, 0.9-2.1), P=0.19. None of the different degrees of clot resolution on baseline CUS was statistically significantly associated with the risk of recurrent VTE. CONCLUSION: In our study, the presence of RVO at the time of OAT withdrawal was not associated with a statistically significant higher risk of recurrent VTE. RVO assessment may not be useful to guide duration of anticoagulation.