RESUMEN
Radionuclide venography (RNV) and contrast venography (CV) were compared in 72 limbs of 59 patients being clinically suspected of deep vein thrombosis (DVT) of the lower extremities. The criteria of positive RNV for DVT regarding flow pattern abnormality included (1) nonfilling or nonvisualization of a deep vein, (2) interruption of the flow, (3) irregular or asymmetric filling of a deep vein, and (4) presence of abnormal collateral vessels. The overall accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and positive likelihood ratio (LR+) were 90 per cent, 88 per cent, 96 per cent, 98 per cent, 79 per cent, and 20.8 respectively. These figures are even higher when we focused on the major veins of the thigh and pelvis. They were 97 per cent, 95 per cent, 98 per cent, 98 per cent, 95 per cent, and 61 respectively. The results indicate that radionuclide venography, while technically simple, is a reliable test for detection of DVT particularly of the major veins of the lower extremities. Combined radionuclide venography and perfusion lung scan can also be performed in the same setting if Tc99m-MAA is used. Contrast venography which is an invasive procedure, should be reserved for questionable cases of those with suspected isolated calf vein thrombosis.