Background: A prospective study was performed comparing the accuracy of digital subtraction angiography (DSA) and Doppler ultrasonography (DUS) stenosis findings with measurements on histological specimens.
Method: DSA and DUS were used to evaluate carotid stenosis and were compared with measurements on histological specimens. Intact carotid plaques from 123 cases were removed in one piece during surgery. The specimens were histologically processed and examined in transverse sections. The smallest inner and correlating outer diameters were measured and the extent of stenosis was calculated. Carotid artery stenoses were compared and statistics done. Specimens in symptomatic cases were divided into 3 groups: stenosis 30-49% (Group 1), stenosis 50-69% (Group 2) and stenosis 70-99% (Group 3). Specimens in asymptomatic cases were divided into two groups: stenosis<or=59% (Group A) and stenosis>or=60% (Group B).
Findings: Wilcoxon paired tests revealed significant differences between DSA, DUS and measurements on histological specimens. In severe stenoses only, no significant difference was observed between stenosis measurement according to the European Carotid Surgery Trial (ECST) angiography methodology and measurements on histological specimens. The most pronounced differences were found between angiography methodology of the North American Symptomatic Carotid Endarterectomy Trial (NASCET) and measurements on histological specimens. When investigating how often preoperative measurement classified stenosis into the same Group of stenoses as postoperative measurement, DUS was the most accurate diagnostic tool.
Conclusions: This study confirmed our previous results, i.e., angiography underestimates the degree of carotid artery stenosis. DUS seems to be more accurate in classifying stenoses into different groups to the extent of narrowing of the carotid arteries. These results make the position of angiography in diagnostic algorithm of carotid stenoses investigations even more questionable.