[Diagnosis of post-operative venous thrombosis using determination of plasma D-dimer]

The D-Dimer test has been shown to be highly sensitive for the diagnosis of deep vein thrombosis (DVT) and pulmonary embolism. Two automatic quantitative tests giving a rapid response within 10 and 30 minutes have been recently marketed. In the postsurgery situation however, the role of the D-Dimer test remains controversial and the optimal cutoff value remains open. The aim of this study was to determine the cutoff value during the postoperative period.

One hundred three consecutive patients admitted to surgery were included. In all patients, D-Dimer test was performed every 2 or 4 days from admission to hospital discharge. The Vidas D-Dimer (Biomerieux, Marcy l'Etoile, France) and the STA Liatest D-DI (Diagnostica Stago, Asnières, France) were performed in parallel in all cases.

Thirty-five patients were excluded because the follow-up period was too short. Results suggest that a D-Dimer value below 2 micrograms/ml has a negative predictive value of 100%. A D-Dimer value over 4 micrograms/ml would indicate suspected deep vein thrombosis in half of the cases, even without clinical signs. Dividing the patients into three groups according to the D-Dimer value, the two tests correlated poorly (r = 0.36 and 0.57) in the middle group (between 2 and 3 and between 3 and 4) and correctly for values below 2 or over 4 micrograms/ml (r = 0.83 and 0.78 respectively).

These two optimum cutoff values (< 2 micrograms/ml and > 4 micrograms/ml) are useful for determining the need for further explorations for DVT. By limiting need for ultrasonography and contrast venography, the cost-efficacy ratio for the detection of DVT during the postoperative period is greatly improved with the D-Dimer screening strategy.