Triplex US in the diagnosis of asymptomatic deep venous thrombosis

Abstract

Purpose: To evaluate the accuracy of triplex ultrasound (TUS) compared with venography as a screening test for deep venous thrombosis (DVT), and to evaluate interobserver variation in the interpretation of the venographic studies.

Material and methods: A total of 133 postoperative hip fracture patients, asymptomatic of DVT, were prospectively examined with TUS and venography. All venograms were reviewed blindly and in case of disagreement a consensus was arrived at.

Results: The incidence of DVT was 20%, with isolated calf vein thrombi in 63% of the cases. There were 7 false-negative and one false-positive result/s at TUS, with a sensitivity of 74%, specificity of 99% and accuracy of 97%. The kappa values ranged from 0.58 to 0.82. The false-negative results were all caused by missed calf vein thrombi in technically inadequate examinations. At sonography 2% of vein segments were noninterpretable, compared to 29% at venography.

Conclusion: Venous US is less sensitive as a test for DVT in this study of asymptomatic patients than in earlier studies on symptomatic patients. Still, sonographic screening of high-risk patients would be both effective and cost effective. Fresh thrombi may cause a false-negative compression test.