The limitations of impedance plethysmography in the diagnosis of acute deep venous thrombosis.

We reviewed our experience with impedance plethysmography (IPG) and duplex scanning in the diagnosis of acute deep venous thrombosis (DVT) to determine their respective accuracy and current role in our noninvasive vascular laboratory. During a recent 22-month period 1776 patients were evaluated in our laboratory for DVT. Sixty patients (64 limbs) underwent ascending venography within 48 hours of testing (49 limbs were evaluated by all three modalities). With the venograms used as the reference standard, B-mode scanning correctly identified the presence of acute thrombus in 24 of 27 limbs (88.8%) and the absence of thrombus in 31 of 34 limbs (91.2%), for an overall accuracy of 90.6%. IPG alone was less sensitive (75%) and less specific (44.8%), with an overall accuracy of only 57.1%. Twenty-eight IPGs were performed on patients with negative venous scans. Two positive IPGs were the result of chronic venous occlusion and two others detected clinically significant isolated iliac vein thrombi, but 13 patients had false positive IPGs. One false negative IPG occurred. The difference in the sensitivity of scan alone vs scan plus IPG was not significant (chi 2 = 0.045; difference not significant), but the decrease in specificity was chi 2 = 17.3; p less than 0.001). The rarity of isolated iliac vein thrombosis and the high false positive rate for IPG do not justify its continued use if B-mode venous scanning is available. Although positive scan results may be used confidently to institute therapy without the need for venography, in high-risk patients with a strong clinical suspicion of proximal DVT despite a negative scan venography should be obtained before withholding anticoagulation.