Early experience using duplex ultrasonography in the diagnosis of deep venous thrombosis; a prospective evaluation

Duplex ultrasound is used in many radiology departments as the first line o f investigation for symptomatic deep venous thrombosis. Before changing the practice of our department from venography to duplex ultrasonography, we w anted to assess our ability to identify deep venous thrombosis on ultrasoun d. Thirty-eight patients were investigated for suspected deep venous thromb osis by venography and duplex ultrasound. The results were compared using v enography as the 'gold standard'. Duplex ultrasound correctly identified 13 out of 16 limbs with deep venous thrombosis. Four of the 38 duplex ultraso und examinations (11%) were described as inadequate at the time of examinat ion, and when these are excluded from the analysis a sensitivity of 93%, an d specificity of 80% are achieved. We conclude that there is a significant learning curve when performing duplex ultrasound of the lower limb, and tha t change-over from venography to ultrasound should include a period during which both examinations are routinely performed.