Accuracy and criteria for localizing arterial occlusion with transcranial Doppler.

Abstract

The authors determined transcranial Doppler (TCD) accuracy for the proximal internal carotid artery (ICA), distal ICA, proximal middle cerebral artery (MCA), distal MCA, anterior cerebral artery (ACA), posterior cerebral artery (PCA), terminal vertebral artery (tVA), and basilar artery (BA) occlusion in cerebral ischemia patients. Detailed diagnostic criteria were prospectively applied for TCD interpretation independent of angiographic findings. Of 320 consecutive patients referred to the neurosonology service with symptoms of cerebral ischemia, 190 (59%) patients also underwent angiography (MRA or DSA). 48 of those 190 patients had angiographic occlusion and 12 of those 48 patients had involvement of multiple vessels. Median time from TCD until angiography was performed was 1 hour (41 patients had angiography before TCD). TCD showed 40 true positive, 8 false negative, 8 false positive, and 134 true negative studies with sensitivity 83.0%, specificity 94.4%, positive predictive value 83.0%, negative predictive value 94.4%, and accuracy 91.6% to determine all sites of occlusion. Sensitivity for each individual occlusion site was: proximal ICA 94%, distal ICA 81%, MCA 93% tVA 56%, BA 60%. Specificity ranged from 96% to 98%. TCD is sensitive and specific in determining the site of the arterial occlusion using detailed diagnostic criteria, including proximal ICA and distal MCA lesions. TCD has the highest accuracy for ICA and MCA occlusions. If the results of TCD are normal, there is at least a 94% chance that angiographic studies will be negative.