IMPACT OF MARKED CAROTID SIPHON WALL CALCIFICATION ON THE EVALUATION OF DEGREE OF VASCULAR STENOSIS BY COMPUTED TOMOGRAPHY ANGIOGRAPHY
Objective To investigate the impact of marked carotid siphon wall calcification on the evaluation of the degree of vascular stenosis by computed tomography angiography (CTA),and to find out a relatively reliable type of CTA image for evaluating the degree of vascular stenosis.Methods A retrospective analysis was performed for the imaging data of 66 patients who underwent digital subtraction angiography (DSA) due to intracranial aneurysm and had marked carotid siphon wall calcification on CTA (major axis of calcified lesion ≥1.5 mm on axial computed tomography).The DSA image was used as the gold standard to analyze the sensitivities,specificities,and correct rates of original axial CTA image,subtracted axial image,subtracted volume rendering (VR) image,curve planar reconstruction (CPR) image,and subtracted maximal intensity projection (MIP) image in the diagnosis of the degree of vascular stenosis in calcified area.Results In the diagnosis of the degree of vascular stenosis in calcified area,the original axial CTA image had a sensitivity of 98.33％,a specificity of 24.49％,and a correct rate of 65.14％;the subtracted axial image had a sensitivity of 96.67％,a specificity of 79.59％,and a correct rate of 88.99％;the subtracted VR image had a sensitivity of 93.33％,a specificity of 71.43％,and a correct rate of 83.49％;the CRP image had a sensitivity of 98.33％,a specificity of 44.90％,and a correct rate of 74.31％;the subtracted MIP image had a sensitivity of 98.33％,a specificity of 55.10％,and a correct rate of 78.90％.The subtracted axial image was the most reliable image for determining the degree of vascular stenosis in calcified area (x2=3.84,P＜0.05).Conclusion Marked carotid siphon wall calcification significantly reduces the reliability of original axial CTA image in the diagnosis of stenosis,and subtracted axial image should be used for determining the degree of vascular stenosis in calcified area.
NETL
NSTL
万方数据
维普
