Carotid bifurcation CT angiography: assessment of interactive volume rendering

The purpose of this study was to compare the accuracy of CT angiography (CTA) for the assessment of carotid bifurcation stenosis, using interactive volume rendering (VR), maximum intensity projection (MIP), and 2D transverse CT technique (t-CT).

Nineteen consecutive patients were prospectively studied with CTA and selective digital subtraction angiography (DSA). There were 13 men and 6 women from 51 to 84 years old (mean 70 years). Results of DSA were compared with those of interactive VR, MIP, and conventional t-CT results, using North American Symptomatic Carotid Endarterectomy Trial criteria for stenosis grading.

There were a total of 38 carotid bifurcations studied, with 9 mild, 10 moderate, and 15 severe stenoses and 4 occlusions. Overall agreement with DSA for VR was achieved in 76%. Eighty percent of the severe stenoses were correctly predicted by VR. The overall agreement between t-CT and DSA was 89%. MIP images, when analyzed independently, showed an overall agreement with angiography of only 71%. VR was not significantly different from MIP (p = 0.60). The difference between VR and t-CT had borderline significance (p = 0.09). MIP had significantly poorer agreement with angiography than t-CT (p = 0.02).

CTA has a high degree of accuracy for the assessment of carotid artery disease compared with catheter angiography. Interactive VR increases the accuracy of diagnosing carotid stenosis and decreases the number of unsatisfactory studies as compared with MIP. Further advances in computation speeds and improvements in software may dramatically alter the future use of VR for the communication of results to clinicians; however, careful analysis of transverse sections is essential to accurate CT interpretation.