基于非门控胸部CT的冠状动脉钙化积分与血管内超声评估钙化病变的对比观察研究

Comparative observation between coronary artery calcium score derived from non-gated chest tomography and calcified lesions assessed by intravascular ultrasound

目的探究基于不同层厚非门控胸部CT(nongated chest computed tomography,NCCT)及冠状动脉CT血管造影(coronary computed tomography angiography,CCTA)的冠状动脉钙化积分(coronary artery calcium score,CACS)与血管内超声(intravascular ultrasound,IVUS)评估钙化病变的关系。研究对象回顾性分析2018年1月至2021年6月因冠心病在上海市胸科医院接受冠状动脉造影及IVUS检查,且在术前1个月内行NCCT或CCTA检查的患者。干预措施根据CT类型及层厚将病变分为4组:NCCT/5 mm组、NCCT/1 mm组、NCCT/1.25 mm组和CCTA/0.625 mm组。观测指标及测量方法在CT图像上使用Agatston评分计算靶病变CACS,利用IVUS测量靶病变钙化长度、最大钙化弧度和钙化斑块表面积,通过计算Spearman相关系数分析CACS与钙化斑块表面积的相关性,并使用受试者工作特征曲线(receiver operating characteristic curve,ROC曲线)评估CACS识别弧度>180°钙化斑块的能力及阈值。结果共纳入96例患者的104个病变,年龄为(66.7±8.7)岁,男性67例(69.8%),合并糖尿病或肾功能不全的患者分别为32例(33.3%)与8例(8.3%)。104个病变中有58个病变(55.8%)位于左前降支(leftanteriordescending,LAD),钙化长度中位数为11.8mm(四分位数9.6 mm,15.5 mm),最大钙化弧度中位数为154°(四分位数96°,215°),钙化斑块表面积为中位数20.0 mm^(2)(四分位数13.8 mm^(2),26.4 mm^(2))。各组NCCT及CCTA的CACS均与钙化斑块表面积显著相关:NCCT/5mm组、NCCT/1mm组、NCCT/1.25 mm组和CCTA/0.625 mm组相关系数r分别为0.6379、0.7831、0.7968和0.7909(P均<0.0001)。ROC曲线分析显示:NCCT/5 mm组识别弧度>180°钙化斑块的CACS阈值是184.5[曲线下面积(area under the cure,AUC)=0.7011;P=0.0013];NCCT/1 mm组的阈值是202.5(AUC=0.7943;P=0.0008);NCCT/1.25 mm组的阈值是194.0(AUC=0.7509;P=0.0131);CCTA/0.625 mm组的阈值是233.0(AUC=0.7527;P=0.0256);NCCT/5 mm组与其他各组AUC的差异均无统计学意义。结论基于多种层厚NCCT及CCTA的CACS都与钙化斑块表面积具有良好的相关性,并可有效识别弧度>180°的钙化斑块。

Objective To assess the association between the coronary artery calcium score(CACS)derived from ECG-gated coronary computed tomography angiography(CCTA)or non-gated chest computed tomography(NCCT)and coronary calcium measured by intravascular ultrasound(IVUS).Subjects We retrospectively enrolled 96 patients(104 vessels)who underwent coronary angiography and IVUS examination due to coronary artery disease and underwent CT(NCCT or CCTA)scan within 1 month before their procedure,which took place between January 2018 and June 2021 in Shanghai Chest Hospital.Interventions According to the type and slice thickness of CT,lesions were divided into the following four groups:NCCT/5 mm group,NCCT/1 mm group,NCCT/1.25 mm group,and CCTA/0.625 mm group.Main Outcomes and Measurements The CACS of target lesions were obtained using the Agatston method.Calcium length,the maximum calcium angle,and calcified plaque surface area were measured by IVUS.A correlation analysis was performed between CACS and the calcified plaque surface area.The receiver operating characteristic(ROC)curve was used to evaluate the ability and threshold of CACS to identify a calcific plaque with a total calcium arc>180°.Results Hundred and four lesions were observed in this cohort of 96 patients.The mean age of patients was(66.9±8.7)years,and 67(69.8%)of them were male.Thirty-two patients(33.3%)had diabetes mellitus and eight(8.3%)had renal insufficiency.Fiftyeight(55.8%)of 104 lesions were located in the left anterior descending artery(LAD).The median calcium length was 11.8 mm(Q1-Q3,9.6 mm-15.5 mm),the median maximum calcium arc was 154°(Q1-Q3,96°-215°),and the median calcified plaque surface area was 20.0 mm^(2)(Q1-Q3,13.8 mm^(2)-26.4 mm^(2)).A significant correlation was observed between CACS and calcified plaque surface area in each group(NCCT/5 mm[Spearman r=0.6379;P<0.001],NCCT/1 mm[Spearman r=0.7831;P<0.001],NCCT/1.25 mm[Spearman r=0.7968;P<0.001],and CCTA/0.625 mm[Spearman r=0.7906;P<0.001]).In the ROC curve analysis,the optimal cut off value to identify a calcific plaque with a total calcium arc>180°was 184.5 in the NCCT/5 mm group(AUC=0.701;P=0.0013),202.5 in the NCCT/1 mm group(AUC=0.7943;P=0.0008),194.0 in the NCCT/1.25 mm group(AUC=0.7509;P=0.0131),and 233.0 in the CCTA/0.625 mm group(AUC=0.7527;P=0.0256).There was no significant difference in the AUC between each group.Conclusions Regardless of slice thickness,the CACS derived from NCCT or CCTA had a good correlation with the calcified plaque surface area and proved to be a useful tool to identify calcific plaques with a total calcium arc>180°.