The measurement of stenosis of the internal carotid artery: comparison of doppler ultrasound, digital subtraction angiography and the 3D CT volume rendering technique.

OBJECTIVES

To determine the variability of measurement of stenosis of the internal carotid artery (ICA) using Doppler ultrasound, digital subtraction angiography (DSA) and the three-dimensional volume rendering technique (3D CT) in the high-grade stenosis band (70-100%).

DESIGN

A prospective study of preoperative findings in 64 patients.

MATERIAL AND

METHODS

Doppler ICA stenosis was assessed according to combined Doppler acoustic standard criteria (CDASC) and peak systolic velocity (PSV). DSA and 3D CT stenosis were measured using the NASCET method. The results were compared using kappa statistics and Pearson's correlation coefficient.

RESULTS

Agreement on the degree of ICA stenosis was statistically very good for DSA and 3D CT (kappa = 0.81, CI 0.69-0.93): Pearson's correlation coefficient was 0.88 (CI 0.76-0.90). The comparison of Doppler ultrasound with DSA showed good agreement (kappa 0.70, CI 0.58-0.82): Pearson's correlation coefficient was 0.61 (CI 0.43-0.74). The disagreement rate on occlusion was 3% between DSA and 3D CT and was rather high (10.9%) between Doppler ultrasound and DSA.

CONCLUSIONS

The clinical decision to operate on an ICA stenosis will be strongly influenced by the diagnostic method used. DSA and 3D CT correlate well, whereas Doppler ultrasound tends to differ at the high end of the high-grade stenosis band (90% occlusion), underestimating (compared with DSA) the degree of stenosis.