Identification of the Optimal Approach to the Ultrasound Diagnosis of Deep Vein Thrombosis of the Lower Extremities in Symptomatic Patients

All eligible patients will undergo D-dimer measurement and a formal assessment of pre-test clinical probability (PTP) by means of the Wells score. Patients with unlikely PTP and negative D-dimer will have DVT ruled out and will undergo a 3-month clinical follow-up. If at least one of the two tests is positive, patients will undergo CUS of the proximal vein system by investigating at least the common femoral vein at the groin, the superficial femoral vein at the mid thigh and the popliteal vein(s) in the popliteal fossa. A positive CUS will adjudicate proximal DVT. In the case of negative CUS of the proximal vein system, patients with either negative D-dimer or unlikely PTP will have DVT ruled out, and will undergo a 3-month follow up. Only patients with concomitant likely PTP and positive D-dimer will undergo whole-leg ultrasonography with interrogation of the calf veins. A positive test will adjudicate distal DVT. Patients with negative test will undergo a 3-month clinical follow up.

Follow-up. All patients in whom DVT is ruled out will be monitored for 3 months (either by clinical visit or by telephone contact) after the enrolment in the study. Patients with clinical symptoms of DVT or PE during follow-up will undergo objective diagnostic testing (venous ultrasound for suspected DVT and spiral CT-scan or VQ scan for suspected PE). All source documents will be sent out for an independent adjudication. In case of death, PE diagnosis will be adjudicated by means of autopsy, if available. If autopsy is not available, the adjudication of PE-related death will be done based on clinical documentation.

D-dimer. All quantitative D-dimer tests available at each center will be acceptable for the purpose of this study. D-dimer will be considered negative according to the cut-offs provided by manufacturers.

Sample size. The investigators expect that in every 100 symptomatic outpatients DVT will be excluded without the need for objective testing in approximately 30%, while approximately 20% will have a CUS-detected proximal DVT, approximately 15% will have negative CUS and unlikely PTP in spite of positive D-Dimer (no need for further investigation), approximately 15% will have negative CUS and negative D-Dimer in spite of likely PTP (no need for further investigation, and approximately 20% will have negative CUS and both likely PTP and positive D-Dimer (thus requiring interrogation of the calf vein system).

The investigators hypothesize that the incidence of venous thromboembolic events during follow up in patients labeled as not having DVT will not exceed 1% with the application of the proposed algorithm, and that the upper limit of the 95% confidence intervals around this proportion will not exceed 2%. Accordingly, ap proximately 1100 patients should be enroled.