Abstract: Objective: The purpose of this work was to evaluate the differences of plaque composition and volume between carotid artery bifurcation and carotid siphon and its predictive value for cerebrovascular events. Materials and Methods: A total of 100 patients(65 men, 35 women; mean age,(55 ± 9) years) who underwent CT angiography experiment were enrolled. Plaque component types of the carotid artery bifurcation and the carotid siphon were defined depending on its attenuation values(lipid, < 60 HU; mixed tissue, 60~130 HU; and calcification, > 130 HU), and the components volumes were calculated. Statistical analyses were performed using Student's t test and Chi-square test to evaluate the plaque volumes and constituent ratio difference among plaque components between carotid artery bifurcation and carotid siphon. The ROC curve and Az were calculated to evaluate the predict accuracy of the total and each component plaque volumes for cerebrovascular events. Result: There was a statistically significant difference(P< 0.05) between the total volume of the carotid artery bifurcation(mean value,(752.3 ± 3.4) mm³) and the carotid siphon(mean value,(56.2 ± 5.541) mm³)) plaque. The constituent ratio of lipid, calcified, and mixed components were 16.8%, 49.1%, 34.1%, respectively, in the carotid artery bifurcation and 33.3%, 32.0%, 34.6%, respectively, in the carotid siphon, showing a statistically significant difference(χ²=26.077; P< 0.05). Statistical analysis approved the high predict accuracy of the carotid artery bifurcation total, lipid and mixed plaque volumes for cerebrovascular events with Az values of 0.682, 0.773, and 0.653, respectively. The best cut-point for the predict accuracy for cerebrovascular events in the plaque was 203.383 mm³ of fatty components, with sensitivity and specificity of 75% and 82%. Conclusion: Resultsof this study suggest there were significant differences in plaque volumes and constituent ratio among plaque components between carotid artery bifurcation and carotid siphon, and indicate the high predict accuracy of lipid plaque for cerebrovascular events.