Carotid artery plaque characterization with a wide-detector computed tomography using a dedicated post-processing 3D analysis: comparison with histology

Purpose The characterization of atherosclerotic carotid plaque plays a key role in the identification of patients at risk. The aim of our work was to evaluate the potentialities of carotid computed tomography angiography (CCTA) in assessing composition of atherosclerotic plaque. Materials and methods We retrospectively evaluated 29 patients (7 women and 22 men, age range 54-81; mean age 69) who underwent carotid endarterectomy. All patients underwent pre-surgical CCTA using a 320-slice scanner. Post-processing reconstructions and analysis were performed using a specific software. Percentage of three different components of the atherosclerotic plaque (adipose, fibrotic and calcific) were classified based on Hounsfield unit values. Post-processing results were compared with histological analysis. Vessel and plaque parameters were compared using the Pearson correlation coefficient (r). Bland-Altman plots with 95% confidence intervals were calculated for correlation. McNemar's test was used for comparison of dichotomous variables. Results A significant correlation between histology and CCTA was found with respect to the areas corresponding to adipose, fibrotic and calcified plaques. The existence of proportional bias was observed between the two quantifying methods with lower discrepancies found for the adipose and fibrotic plaque areas. The Bland-Altman analyses showed a mean bias of 3.2%, 2.5% and 0.6% between histology and CCTA, for adipose, fibrotic and calcified plaque areas, respectively. Conclusions Multi-detector CT angiography represents a valuable technique to assess quantitatively the composition of atherosclerotic plaques, with particular reference to the prevalence of fibrotic tissue, and is a useful diagnostic tool to improve risk stratification of patients for cerebral stroke.