Résumé
About 20% of patients who present with symptoms compatible with deep vein thrombosis (DVT) have the diagnosis confirmed on objective testing. Diagnostic performances of compression ultrasonography (CUS) for proximal DVT are excellent and the necessity of treating these thromboses is well recognized. Two strategies are used for the diagnosis of DVT. The first consists of CUS limited to the proximal veins performed on two occasions with a one-week interval, a widely validated strategy, called "serial" ultrasonography. This approach does not search for distal DVT but allows detection of possible extension of distal DVT to the proximal veins. Another diagnostic strategy relies on a single complete (proximal and distal) CUS. This obviously enables the diagnosis of calf vein thrombosis, which is usually treated by anticoagulants. However, some robust data suggest that this strategy may increase the number of patients given anticoagulant therapy without improving their 3-month thromboembolic risk. Indeed, a recent randomized non-inferiority trial showed similar safety of these two strategies. Despite this reassuring study, many clinicians still search for and treat distal DVT with anticoagulant therapy. The ongoing randomized double blind CACTUS trial assesses the usefulness of anticoagulation treatment in symptomatic distal DVT. This study should answer questions about the usefulness of anticoagulation for distal DVT, a widely debated issue.