Prospective cohort study aimed at optimizing the duration of anticoagulant treatment in patients at their first episode of proximal deep venous thrombosis (DVT) of the lower extremities, whose pathogenesis is either unknown (idiopathic DVT) or associated with minimal risk factors for thrombosis, with the help of an algorithm which incorporates both ultrasonography and D-dimer information. All patients will be followed-up until 1) the achievement of a major end-point; 2) the date of lost to to followup; 3) the date of death; 4) the date of study stop. The purpose of this study is to demonstrate the safety of withholding anticoagulation from a subgroup of patients with proximal DVT whose veins have recanalized and present with a repeatedly negative D-dimer (at baseline, after 1 and 3 months). The approach will be deemed to be safe if the annual rate of recurrent VTE in patients who will have their anticoagulation discontinued is lower than 5%.

After giving informed consent, patients will receive an ultrasound investigation of the proximal-vein system (common femoral vein at groin, popliteal vein up to its trifurcation):

• Patients with residual thrombosis (defined as a diameter of at least 4 mm in at least one spot) will have their anticoagulation continued. A repeat ultrasound is scheduled after 6, 12, 18, 24 and 36 months. In patients with persistently residual thrombosis anticoagulation will not be discontinued, while those whose veins have recanalized will have a decision making process based on the behaviour of D-dimer (see below).
• Patients whose veins have recanalized (either at the recruitment or later on) will receive the D-dimer determination before discontinuing anticoagulation. In those with negative D-dimer anticoagulation will be discontinued. These patients will have two further determinations of D-dimer (after 1 and 3 months, respectively). While patients with persistently negative D-dimer will no longer receive anticoagulation, those in whom D-dimer becomes positive will have their anticoagulation resumed and no longer discontinued.
• All patients will be followed up to completion of 4 years since recruitment. For the purpose of this study each quantitative D-dimer is allowed. The criterion for D-dimer interpretation will rely on the cut-off that is recommended by manufacturers for diagnostic purposes.

D-dimer.For the purpose of this study each quantitative D-dimer is allowed. The criterion for D-dimer interpretation will rely on the cut-off that is recommended by manufacturers for diagnostic purposes.

Sample size 1000 years of observation without anticoagulation are required to demonstrate (power 90%, type I error 0.05, two sided) that with this approach the annual rate of recurrent VTE is lower than 5%. Approximately 600 patients with proximal DVT satisfying the eligibility criteria are required to obtain 1000 years of observation without anticoagulation.

Inclusion Criteria:

• Patients with proximal DVT that is idiopathic or secondary to minor factors for thrombosis, with or without contemporary manifestations of PE, who have completed an uneventful 3 to 24-month period of anticoagulation and are available for an overall 48-month follow-up at the study centre.

Exclusion Criteria:

• previous thromboembolism
• recent (less than 3 months) major trauma or surgery
• active cancer
• immobilization resulting from chronic irreversible medical diseases
• need for indefinite anticoagulation for medical reasons other than VTE
• impossibility to attend the follow-up visits or to have D-dimer determinations
• already known major thrombophilia: carriage of deficiencies of natural anticoagulants, lupus-like anticoagulants, homozygosis for factor V Leiden or prothrombin mutation, heterozygosis for both abnormalities
• short (less than 1 year) life expectancy
• pregnancy
• age younger than 18
• refusal of informed consent