22 carotid specimens following eversion-endarterectomy were compared with preoperative assessment of carotid stenosis obtained angiographically and by c-w-Doppler-sonography. The intact, unsplit specimens were perfused with a liquid plastic material (Palavit M). After hardening of the plastic material the specimens were removed. The local degree of carotid stenosis with respect to diameter reduction was assessed by direct measurement of the plastic specimens at the narrowest site compared with the diameter of the carotid eversion specimens at the place of the maximum stenosis. The distal degree of carotid stenosis was assessed by comparison of the diameter of the distal internal carotid artery obtained intraoperatively with the measurements of the plastic specimens (1 mm vascular wall thickness of distal internal carotid artery was taken into account). RESULTS Both the local and the distal degree of carotid stenosis diameter (mean 84.7% +/- 8.4% and 82.1% +/- 9.1% respectively) were underestimated in the preoperative angiogram (79.8 +/- 9% by ECST-criteria and 69 +/- 10.3% by NASCET-criteria) in most of the cases. The difference of the diameter reduction was statistically significant (p < 0.05 and p < 0.01 respectively, Wilcoxon signed rank test). The c-w-Doppler assessments were 82.6 +/- 8.2% (n.s.). CONCLUSIONS Our results suggest that the preoperative assessment of internal carotid stenosis obtained angiographically or by c-w-Doppler-sonography easily underestimate the true degree of carotid stenosis.