基于CT血管成像的斑块定量分析在冠状动脉血流动力学异常诊断中的应用价值  

Application value of quantitative plaque analysis based on CT vascular imaging in the diagnosis of coronary artery hemodynamic abnormalities

作　　者：南丽虹 李睿君 冯进堂 张天助 门玥琪 李东[1] 李锋坦[1] 张璋[1] Nan Lihong;Li Ruijun;Feng Jintang;Zhang Tianzhu;Men Yueqi;Li Dong;Li Fengtan;Zhang Zhang(Department of Radiology,Tianjin Medical University General Hospital,Tianjin 300052,China)

机构地区：[1]天津医科大学总医院医学影像科,天津300052

出　　处：《中华解剖与临床杂志》2021年第5期504-510,共7页Chinese Journal of Anatomy and Clinics

基　　金：国家自然科学基金(81301217);天津市自然科学基金(18JCYBJC25100);科技部“十三五”国家重点研发计划项目(2016YFC1300402)。

摘　　要：目的探讨基于冠状动脉CT血管成像(CCTA)分析斑块定量指标在冠状动脉血流动力学异常诊断中的应用价值。方法回顾性病例对照研究。纳入2020年7—10月天津医科大学总医院行CCTA检查的冠状动脉单支或多支局限性狭窄患者104例(共检出病变血管200支)的临床及影像学资料,其中男63例、女41例,年龄45~83岁。依据靶血管病变部位的CT血流储备分数(CT-FFR)值将200支血管分为两组:CT-FFR≤0.80为血流动力学异常组、CT-FFR>0.80为血流动力学正常组。用半自动斑块分析软件测量靶血管病变长度、管腔体积、斑块总体积、钙化体积、脂质体积、纤维体积、最小管腔面积(MLA)、狭窄程度(SG)、重构指数(RI)、偏心指数(EI)、周围脂肪密度(FAI)等斑块定量指标。采用单因素分析及多因素logistic回归分析冠状动脉血流动力学异常的影响因素,采用Pearson相关法分析法分析CT-FFR与斑块定量分析指标之间的相关性。基于以上统计分析,设计两种模型:模型一(使用SG判断心肌血流动力学异常)、模型二(SG+有统计学意义的斑块定量指标);采用受试者操作特征曲线(ROC曲线)比较两种模型预测血流动力学异常的效能。结果单因素分析显示,两组的病变长度、斑块总体积、钙化体积、MLA、SG差异均有统计学意义(Z=-2.613、-2.356、-2.435、-7.499、-5.516,P值均<0.05);多因素logistic回归分析显示,MLA是冠状动脉血流动力学异常的独立预测因子(OR=0.653,95%CI 0.509~0.838,P<0.01);Pearson相关分析,病变长度、MLA、SG与CT-FFR呈相关关系(r=-0.41、0.49、-0.49,P值均<0.05)。模型一和模型二在判断血流动力学异常的ROC曲线下面积分别为0.730(95%CI 0.659~0.800)和0.804(95%CI 0.757~0.875),模型二显示出了更高的诊断效能(重分类改善指标为0.181,P<0.05)。结论基于CCTA的冠状动脉斑块定量分析表明,病变长度、MLA、SG是血流动力学异常的影响因素,MLA是血流�Objective To explore the value of quantitative plaque analysis using coronary computed tomography angiography(CCTA)in the diagnosis of coronary artery hemodynamic abnormalities.Methods Retrospective(case-control)study.A retrospective analysis was performed on 104 patients(with a total of 200 vessels)who underwent CCTA examination in Tianjin Medical University General Hospital from July to October 2020.The patients included 63 males and 41 females aged 45-83 years old.The 200 blood vessels were divided into two groups according to the computed tomography-derived fractional flow reserve(CT-FFR)value(abnormal hemodynamics group:CT-FFR≤0.80,normal hemodynamics group:CT-FFR>0.80).A semi-automatic plaque analysis software was used to measure quantitative plaque indexes,such as target vascular lesion length,lumen volume,total plaque volume,calcification volume,lipid volume,fiber volume,minimal lumen area(MLA),stenosis grade(SG),remodeling index,eccentricity index,and fat attenuation index.Univariate analysis,multivariate logistic regression analysis,and Pearson correlation analysis were used to analyze the data.Two models were designed based on the statistical analyses:Model 1(use SG to determine cardiac hemodynamic abnormalities)and Model 2(use SG+statistically significant quantitative plaque indicators to determine cardiac hemodynamic abnormalities).Receiver operating characteristic(ROC)curve was used to compare the efficacy of the two models in the diagnosis of hemodynamic abnormalities.Results Univariate analysis showed statistical differences in lesion length,total plaque volume,calcification volume,MLA,and SG(Z=-2.613,-2.356,-2.435,-7.499,-5.516;all P values<0.05).Multivariate logistic regression analysis showed that MLA was an independent predictor of abnormal coronary artery hemodynamics(odds ratio=0.653,95%CI interval=0.509-0.838,P<0.01).Pearson correlation analysis showed that lesion length,MLA,and SG were linearly correlated with CT-FFR(r=-0.41,0.49,-0.49;all P values<0.05).The areas under the ROC curve wer

关 键 词：血流动力学 血流储备分数 冠状动脉疾病 计算机体层摄影血管造影术 

分 类 号：R816.2[医药卫生—放射医学] R543.3[医药卫生—临床医学]