Can laboratory testing improve screening strategies for deep vein thrombosis at an emergency unit?

To study various markers of blood coagulation and fibrinolysis in relation to the extension of deep vein thrombosis (DVT), and to compare the diagnostic usefulness of these markers as screening tests for excluding DVT.

A clinical study of patients admitted to an emergency unit.

Ostra Hospital, Göteborg, Sweden.

One hundred and five patients with a clinical suspicion of DVT.

Phlebography was used as the reference method for a diagnosis of DVT. Small distal thromboses as well as large proximal thromboses were included. Plasma D-dimer as well as other markers of coagulation and fibrinolysis were analysed.

Twenty-eight proximal and 20 distal DVTs were found. Plasma D-dimers (one ELISA and two latex assays), fibrin monomer, prothrombin fragment 1 + 2 (F1+2), thrombin-antithrombin III complex (TAT) and the t-PA-PAI-1 complex were all significantly correlated to the extension of DVT, whilst fibronectin, tissue-type plasminogen activator (t-PA), single-chain urokinase-type plasminogen activator (scru-PA) and plasminogen activator inhibitor 1 (PAI-1) were not. The sensitivity was 94% for the D-dimer ELISA and one of the latex methods (latex-B), at a specificity of 60% and 68%, respectively. The negative predictive value was 92% for ELISA and 93% for latex-B, and both assays showed a negative predictive value of 100% for proximal DVTs. Fibrin monomer, F1+2, TAT, D-dimer (latex-S) and the t-PA-PAI-1 complex all showed lower negative predictive values (88, 84, 79, 78 and 65% respectively).

Sensitivity and negative predictive values for a latex assay (D-dimer latex-B) was similar to that of a D-dimer ELISA: With a sensitivity of 94% (100% for proximal DVTs) such a latex assay may be included in a screening strategy for DVT at an emergency unit. However, the safety of such an approach has to be tested in other prospective studies.