Prospective controlled randomized clinical trial. Consecutive patients with acute proximal deep vein thrombosis (DVT) of the lower extremities, with or without contemporary manifestations of pulmonary embolism, are randomized to receive either a fixed duration of anticoagulant therapy (three months in patients with DVT secondary to transitory risk factors, six months in patients with idiopathic DVT) or a flexible duration of anticoagulant therapy, according to the persistence of residual thrombi, as shown by leg veins ultrasonography (up to 1 year in patients with secondary DVT, up to 2 years in those with idiopathic DVT). All patients are followed up to three years to assess the development of (objectively documented) recurrent thromboembolism. The rate of recurrent thromboembolism is compared between the two study groups, as well as the rate of major bleeding complications occurring during anticoagulation.

Study patients Consecutive patients with symptomatic proximal-vein thrombosis who have completed three months of anticoagulation will be eligible for the study. Patients with thrombosis occurring in association with recent (less than three months) trauma, surgical intervention or puerperium, prolonged (more than seven days) immobilization from any cause or use of hormonal treatment will be regarded as "secondary thrombosis". All other patients will defined as having idiopathic thrombosis.

Study design and interventions The AESOPUS study is a randomized, multicenter, open trial, with independent and blinded assessment of study outcomes. The study is designed to evaluate the long-term clinical benefit and risk of adjusting the duration of oral anticoagulant therapy based on to the persistence or recanalization of venous thrombosis, as shown by repeated ultrasonography.

Randomization is stratified for secondary versus idiopathic thrombosis and for center. After completing the first uneventful three months of anticoagulation, patients will be randomized to fixed durations of warfarin or to flexible durations. In the fixed duration group, patients with idiopathic deep-vein thrombosis will receive 3 additional months of treatment (for a total treatment duration of 6 months), and patients with secondary deep-vein thrombosis will discontinue treatment (for a total treatment duration of 3 months). In the flexible duration group, patients will have ultrasound at the time of randomization and then at 3, 9, 15 and 21 months if the index event is idiopathic or at 3 and 9 months if the index event is secondary. Anticoagulant therapy will be discontinued if the thrombosis has recanalized. If residual venous thrombosis is detected, treatment will be continued until the following ultrasound testing, and for a maximum of 9 months for secondary and 21 months for idiopathic venous thrombosis. In both groups, warfarin will be adjusted to a target international normalized ratio of 2.0 to 3.0.

Ultrasound assessments of the common femoral and popliteal vein (transverse plane) will be done according to a standardized procedure by independent experts unaware of clinical details and of previous ultrasound findings. Vein diameters will be measured during maximal compression, and considered recanalized in case of a diameter < 2.0 millimeters in a single determination, or a diameter < 3.0 millimeters in two consecutive determinations.

Follow-up, recurrent VTE and bleeding Patients will be followed-up for 33 months to document the incidence of symptomatic recurrent thromboembolism. Follow-up visits will be scheduled in all patients at 3, 9, 15, 21, and 33 months after randomization. Recurrent thromboembolism will be diagnosed by compression ultrasound, ventilation/perfusion scanning, or helical tomography as appropriate. If recurrent thrombosis is suspected in a previously unaffected leg, the sole diagnostic criterion will be incompressibility of a proximal vein. Ultrasound criteria for recurrent ipsilateral thrombosis will be incompressibility of a proximal vein segment initially free from thrombi and/or incompressibility of a proximal vein that has completely recanalized. Nonfatal pulmonary embolism will be defined by a (sub)segmental ventilation-perfusion mismatch on lung scanning or an intraluminal filling defect on spiral computed tomography of the chest. Fatal pulmonary embolism will be diagnosed if it is confirmed at autopsy, if it is anteceded in the immediate period before death by objectively confirmed pulmonary embolism or venous thrombosis, or if it is a sudden death that cannot be explained by a disease or condition other than pulmonary embolism.

Bleeding will defined as major it is clinically overt and associated with a hemoglobin drop of at least 20 g/L or transfusion of al least two units of red cells, is retroperitoneal or intracranial. All outcome events will be reviewed by an independent adjudication committee whose members were unaware of the treatment assigned.

Inclusion Criteria:

• acute proximal DVT associated or not with clinically symptomatic pulmonary embolism

Exclusion Criteria:

• history of previous VTE
• active cancer
• indications for permanent anticoagulation
• contraindications to anticoagulation
• pregnancy
• geographical inaccessibility for long-term follow-up
• life expectancy shorter than 1 year
• refusal of informed consensus