Exclusion and diagnosis of deep vein thrombosis by a rapid ELISA D-dimer test, compression ultrasonography, and a simple clinical model

Abstract

The classical clinical signs of deep vein thrombosis (DVT) are unspecific and may be found in several other conditions besides DVT. Therefore, patients suspicious of DVT are subjected to elaborate invasive or noninvasive evidence-based procedures that actually confirm DVT in only 20% to 30% of patients in this setting. However, simple laboratory tests and noninvasive strategies to exclude and diagnose DVT are becoming available in the clinical emergency setting of outpatients. In the presented literature, a sound basis is provided for quantifying clinical judgment for the diagnosis of acute proximal DVT. The number of positive clinical findings at time of first suspicion of DVT appears to correlate directly with the probability of acute proximal DVT. The modified clinical model of Landefeld and Wells for DVT allows reasonable accurate classification of patients into low, moderate, and high probability for suffering DVT. The rapid automated enzyme-linked immunoabsorbant assay (ELISA) VIDAS D-dimer presently available can be rapidly performed in daily practice and emergency situations and is accurate to a high degree, especially in ruling out ongoing venous thromboembolic processes. The sequential use of the rapid ELISA VIDAS D-dimer test and compression ultrasonography in a well-designed clinical setting using a simple clinical model predicts a significant improvement due to a high sensitivity near 100% for the exclusion and diagnosis of DVT in the majority of outpatients with suspect DVT. A prospective decision analysis management study is proposed to exclude and diagnose DVT based on the rapid ELISA VIDAS D-dimer test and compression ultrasonography within the context of a ready-to-use simple clinical model. The proposed simple model of a rational diagnosis of deep vein thrombosis (RADIA DVT) has to be tested in a large multicenter study of more than 1,000 outpatients with suspected DVT. This model would be less expensive, easy to perform, and likely yield a significant simplification and improvement of highly accurate evidence-based exclusion or diagnosis of DVT on the basis of which clear-cut indications of anticoagulation could be appropriately initiated or safely withheld.