Contribution of D-Dimer determination in the exclusion of deep venous thrombosis in spinal cord injury patients

Deep vein thrombosis (DVT) is a common complication of paraplegia despite p rophylactic anticoagulant therapy. The diagnosis relies primarily on ultras onography or phlebography; these investigations are difficult, expensive an d can be time-consuming in paraplegic patients. Study Design: To evaluate the usefulness of coagulation activation markers in excluding a diagnosis of DVT, D-Dimers, thrombin-antithrombin complexes, prothrombin fragments (F1+2) and activated factor VIIa. Objectives: To improve the diagnosis of deep venous thrombosis in paraplegi c patients. Setting: This collaborative work was done at Raymond Poincare Hospital, Gar ches, France. Methods: To evaluate the usefulness of coagulation activation markers in ex cluding a diagnosis of DVT, D-Dimers (D-Di), thrombin-antithrombin (TAT) co mplexes, prothrombin fragments (F1+2) and activated factor VIIa (FVIIa), we re determined in a prospective study of 67 consecutive patients with parapl egia or tetraplegia. Doppler ultrasonography and/or phlebography of the low er limbs and D-Di, TAT, F1+2 level determination were systematically done i n each patient at admission to our rehabilitation unit. Results: Despite prophylactic low molecular weight heparin therapy, six of the 67 patients developed DVT diagnosed by radiologic explorations. D-Di le vels measured by a reference ELISA (Asserachrom D-Di, Diagnostica Stage) or a new rapid automated turbidimetric test (STA-Liatest D-Di) were greater t han 500 ng/ml in all DVT patients and in 40 non-DVT patients, of whom most had urinary tract infections, osteomas, or pressure sores. D-Di values were normal in only 21/67 patients (31%). The negative predictive value of D-Di in our study was 100% since all DVT patients had D-Di values greater than 500 ng/ml. TBT and F1+2 levels were not correlated with D-Di levels but als o had a negative predictive value of 100%. Comparison of D-Di levels obtain ed using the two tests showed that results of the reference ELISA were clos ely correlated to those of the new rapid automated turbidimetric. TAT, F1+2 , and factor VIIa are not useful for measuring hypercoagulability in parapl egic or tetraplegic patients since no rapid tests for determining these par ameters are available. Conclusion: D-Di levels determined using an ELISA or a new rapid automated turbidimetric test have a good negative predictive value for DVT in paraple gic or tetraplegic patients and may reduce the need for Doppler ultrasonogr aphy and/or phlebography by 31%.