Accuracy of magnetic resonance imaging in the diagnosis of deep vein thrombosis in asymptomatic patients with injuries of the lower extremity. A pilot study.

This is a prospective comparative study of magnetic resonance imaging (MRI) of the deep veins versus contrast venography in consecutive patients treated for various injuries to their lower extremities, showing no clinical symptoms of deep vein thrombosis. The majority of examinations referred to in this study were performed according to the following methodology: First, the patient was subjected to MRI. Subsequently, within a 24-h interval, he/she was subjected to contrast venography. The acquired results were compared in a blinded manner. The diagnostic indices for MRI were calculated on the assumption that the results of contrast venography were sure to give an accurate indication of either presence or absence of thrombosis. Thirty-six patients were included in the study, of which 27 (15 males) completed it. The overall incidence of distal deep venous thrombosis (DVT) was 22% (6/27). One patient showed extension of a crural thrombus into the popliteal vein. MRI did not detect any of the thrombi. This lack of result was ascribed to failure to fully demonstrate all segments of the crural veins. However, MRI did show three proximal thrombi in the superficial femoral vein, which were not shown by the venograms. Thus, both the sensitivity and specificity of MRI were 0%, so MRI proved to be of no value in the diagnosis of asymptomatic deep venous thrombosis in this study.