Noninvasive vascular laboratory diagnosis of deep venous thrombosis

One hundred and seventeen consecutive patients with suspected deep venous thrombosis (DVT) were studied for a total of 233 limbs in a period of two years. Doppler signals were assessed at the femoral, popliteal, and posterior tibial veins. The pulse volume recorder was used to determine segmental venous capacitance and maximum venous outflow. The mean age was 55.2 years and the male to female ratio was 1:1. Of the 113 legs without suggestion of DVT studies as controls, three were found to have DVT by laboratory criteria and resulted in three cases of bilateral DVT. One hundred and ten limbs that were clinically unsuspected and 79 (66%) that were suspected were negative by the laboratory method. Of these, 106 unsuspected limbs and 42 suspected (35%) did not require venography and were not treated. Thirty-seven suspected legs (31%) and four unsuspected were proven to be negative not only by the laboratory criteria, but also by venography. Eighteen (15%) clinically suspected limbs and two (2 per cent) unsuspected limbs were positive both by venography and laboratory criteria. Eighteen (15%) suspected limbs and one unsuspected were positive by laboratory criteria and were treated without venographic confirmation. There was one false-positive and four false-negatives for an overall error of 4.2 per cent. From the true accuracy rate of 92 per cent, sensitivity of 83 per cent, and specificity of 98 per cent, the author concludes that this is a useful method in the diagnosis of DVT. Most of the false-negatives occurred in patients with recurrent DVT.