Deep venous thrombosis of the lower limb: evaluation of duplex ultrasound imaging versus venography prospective blinded comparative study done at the Kenyatta National Hospital (Nairobi)

Abstract

Introduction: Colour duplex ultrasound examination was carried out on 55 limbs of patients referred to the Radiology department of the Kenyatta National Hospital for contrast venography. The indication for the contrast study was clinical suspicion of deep venous thrombosis of the lower limb/s. Objectives: To compare the findings of duplex ultrasound and contrast venography in the detection of DVT of the lower limb and in the process, establish the ability of colour duplex sonography compared to venography, the gold standard. Design: Prospective blinded comparative study Subject: 44 patients (55 limbs) were seen during the study period. Setting: The departments of diagnostic radiology at KNH and University of Nairobi. Methods: An extra colour duplex ultrasound examination was given to patients, free of charge. This was done within 24 hours of a successful venography examination. Main outcome measure: Sensitivity, specificity, predictive values and accuracy of colour duplex ultrasound in detection of DVT of the lower limbs, as it is done locally. Results: A total of 55 limbs in 44 patients were examined. 30 males and 14 females were seen. Most patients seen (66%) were between 26 and 45 years. A comparable number of left (25) and right (30) limbs were studied. 19 limbs were positive for DVT by venography and 18 by ultrasound. There were 2 false positives and 3 false negatives. In 63% of the positive limbs, thrombosis was in the popliteal vein, only 3 cases of isolated calf DVT were seen. The overall sensitivity was 88.9%, specificity 91.8%, positive and negative predictive value 84% and 94.3% respectively. Accuracy was 90.9%. Considering only DVT above the calf, the validity properties improve to 100%. An alternative diagnosis was suggested by duplex ultrasound in 48% of the negative cases. Conclusion: The sensitivity specificity and accuracy of colour duplex sonography in the detection of DVT of-the lower limbs is high. These findings compare favourably to those done in western countries. We feel that all patients with clinical suspicion of DVT should have ultrasound done as the first investigation and venography be reserved for those with equivocal or inadequate sonography results