Background: CT angiography (CTA) was routinely performed for carotid artery disease, but the plaque characterization based on measurement of density may be inaccurate, influenced by adjacent calcification and contrast agent. Remodeling ratio was one of the additional new objective indexes for plaque characterization. On coronary CTA, the utility of histogram analysis was reported in a few studies. Objectives: The purpose of this study was to evaluate the diagnostic value of histogram analysis of carotid artery plaque density and positive remodeling on CTA for the diagnosis of pathologically proven complicated carotid artery plaques. Methods: We retrospectively evaluated 29 patients with atherosclerotic carotid artery stenosis treated by carotid endarterectomy. On CTA, the boundaries of non-calcified plaques were manually traced at the site of the greatest stenosis. In addition to the mean plaque density, the percentage of pixels (PP) with a density of less than 30 Hounsfield units (HU) (PP < 30) was calculated using the histogram analysis. We calculated the remodeling ratio by dividing the cross-sectional artery area at the greatest stenosis by normal reference-segment artery area. These results of CTA analysis were compared with the results of histological analysis based on the American heart association criteria. Results: Histologically, 13 of the 29 patients had complicated plaques. PP < 30 and the remodeling ratio were higher in patients with complicated plaques than in those with uncomplicated plaques (P = 0.023 and 0.002, respectively). Whereas, the mean plaque density did not differ between the two groups (P = 0.062). Correlation between PP < 30 and the remodeling ratio was stronger than the correlation between PP < 30 and the mean density (Spearman's rs = 0.76 vs. 0.65, respectively). Conclusions: Histogram analysis of the carotid artery plaque density may be more useful for the evaluation of atherosclerosis on CTA compared with the conventional analysis of plaque densities, and the combined analysis of histogram analysis and the remodeling ratio may help to predict future stroke events.