Grading of internal carotid artery stenosis: can CTA overcome the confusion?

To compare the measurements of internal carotid artery (ICA) stenosis obtained from multidetector computed tomographic angiography (CTA) and digital subtraction angiography (DSA) based on the NASCET and ECST grading methods.

In a retrospective cohort analysis from data at a tertiary care university clinic, the CTA and DSA images from 69 consecutive patients (52 men; mean age 70.3+/-8.0 years, range 51-85) who underwent both imaging studies within a maximum period of 28 days were interpreted by 2 radiologists blinded to the results of the other modality. The exact degree of ICA stenosis was calculated for both modalities according to NASCET and ECST guidelines.

The agreement between both stenosis grading methods was comparable for CTA (R2=0.87) and DSA (R2=0.84); mean differences in stenoses grades between ECST and NASCET were 13.9% (CTA) and 12.9% (DSA, p>0.05). Corresponding results for the intermodality correlation were almost equal for NASCET (R2=0.59) and ECST (R2=0.55), with mean differences of 13.4% and 13.5%, respectively (p>0.05). Sensitivity and specificity of CTA for detecting occlusions was 100% for both modalities and grading systems. For detecting stenoses >70%, the sensitivity and specificity were 90.9% and 54.9%, respectively, for NASCET and 94.7% and 46.3%, respectively, for ECST. For stenoses >50%, the values were 95.8% and 59.6%, respectively, for NASCET and 96.4% and 42.5%, respectively, for ECST.

The introduction of multidetector CTA cannot overcome the confusion in the exact grading of ICA stenosis because the application of both tested modalities as well as both grading methods results in clinically important differences.