摘要
目的:探讨国产首款基于计算流体力学仿真技术的计算软件测量的冠状动脉CT血流储备分数(CT-FFR)对心肌缺血的诊断价值。方法:回顾性将临床疑似或已知为冠心病而在本院行冠状动脉CTA和有创性冠脉动脉造影(ICA)检查且有完整FFR测量(ICA-FFR)数据的44例患者(共55支血管)纳入本研究。由独立核心实验室基于CTA图像重建三维血管模型,通过仿真计算得到血管狭窄处的CT-FFR值(≤0.8为心肌缺血风险高),同时计算狭窄位置近端和远端2cm处CT-FFR值的差值(△CT-FFR)。基于CTA和ICA,分别获得冠脉狭窄处的血管狭窄率(SR)。以ICA-FFR≤0.8作为心肌缺血的诊断标准,采用ROC曲线分析评估CT-FFR、△CT-FFR、ICA-SR和CTA-SR对心肌缺血的诊断效能。分别采用Pearson相关分析和Bland-Altman分析评估CT-FFR和ICA-FFR之间的相关性和一致性。结果:基于血管水平,CT-FFR、△CT-FFR、ICA-SR和CTA-SR诊断心肌缺血的AUC分别为0.914(95%CI:0.807~0.973,P<0.001)、0.902(95%CI:0.791~0.966,P<0.001)、0.840(95%CI:0.716~0.925,P<0.001)和0.614(95%CI:0.473~0.742,P>0.05)。CT-FFR和△CT-FFR诊断心肌缺血的符合率、敏感度和特异度分别为90.9%、92.3%、90.5%及89.1%、92.3%、88.1%。相关性分析结果显示,CT-FFR与ICA-FFR具有较好的相关性(r=0.633,P<0.001)。Bland-Altman分析结果显示,CT-FFR与ICA-FFR测量值的差值绝大部分位于95%一致性界限内,表明两种方法的FFR测量值具有较高的一致性。结论:基于计算流体力学技术的CT-FFR与ICA-FFR具有较高的一致性,CT-FFR对心肌缺血的诊断具有一定的临床价值。
Objective:To investigate the diagnostic performance of fractional flow reserve(CT-FFR)obtained by coronary CT angiography with a newly developed calculating software based on computational fluid dynamics(CFD)technique in myocardial ischemia.Methods:A total of 55 coronary vessels in 44 patients with confirmed or clinically suspected coronary artery disease(CAD)were retrospectively enrolled.All of these patients were underwent coronary CT angiography(CTA),invasive coronary angiography(ICA)and FFR measurement(ICA-FFR).The CT-FFR and△CT-FFR(the difference of CT-FFR between the proximal and 2cm distal to the coronary narrow site)were calcula-ted by the software,which provide simulation results from 3D model based on coronary CTA images in independent core laboratory.Stenosis rates(SR)of coronary vessel were also obtained from CTA and ICA,respectively.Meanwhile,using invasive ICA-FFR≤0.8 as the reference standard for myocardial ischemia diagnosis.The diagnostic performance of CT-FFR,△CT-FFR,ICA-SR,CTA-SR for myocardial ischemia were evaluated by receiver operating characteristic(ROC)curve.Pearson correlation analysis and Bland-Altman analysis were used to evaluate the correlation and consistency between CT-FFR and ICA-FFR,respectively.Results:The AUC for CT-FFR,△CT-FFR,ICA-SR,and CTA-SR in the diagnosis of myocardial ischemia on a per-vessel basis were 0.914(95%CI:0.807~0.973,P<0.001),0.902(95%CI:0.791~0.966,P<0.001)、0.840(95%CI:0.716~0.925,P<0.001)and 0.614(95%CI:0.473~0.742,P>0.05),respectively.The diagnostic accuracy,sensitivity,and specificity were 90.9%,92.3%and 90.5%,respectively for CT-FFR,and 89.1%,92.3%and 88.1%,respectively,for△CT-FFR.The correlation analysis results showed that CT-FFR had a good correlation with ICA-FFR(r=0.633,P<0.001).The Bland-Altman analysis showed that the majority of the differences between CT-FFR and ICA-FFR were within the 95%consistency limit,which mean the FFR values obtained by the two methods had a high consistency.Conclusion:CT-FFR based on fluid dynamics simulation
作者
李涯
蔡震宇
李亿华
赖树填
袁旭春
LI Ya;CAI Zhen-yu;LI Yi-hua(Department of Radiology,Fuwai Hospital Chinese Academy of Medical Sciences Shenzhen,Guangdong 518000,China)
出处
《放射学实践》
CSCD
北大核心
2023年第4期426-433,共8页
Radiologic Practice
基金
2018年深圳市科技计划项目(JCYJ20180507182506416)。
关键词
冠心病
血流储备分数
流体力学仿真
冠状动脉CT血管成像
冠状动脉造影
Coronary artery disease
Fractional flow reserve
Computational fluid dynamics
Computed tomography angiography
Invasive coronary angiography
