Suppr超能文献

基于冠状动脉 CT 血管造影的形态学指数预测有血流动力学意义的冠状动脉狭窄

Coronary CT Angiography-based Morphologic Index for Predicting Hemodynamically Significant Coronary Stenosis.

机构信息

From the National Heart Research Institute Singapore, National Heart Centre Singapore, 5 Hospital Dr, 169609 Singapore (C.W., S.L., R.S.T., J.M.F., C.Y.C., L.B., Y.J.F.K., A.S.L.W., S.J.T.C., S.Y.T., S.T.L., L.Z.); The Second Affiliated Hospital of Nanchang University, Nanchang, China (C.W., Q.W.); Duke-NUS Medical School, Singapore (S.L., R.S.T., J.M.F., C.Y.C., L.B., Y.J.F.K., A.S.L.W., S.J.T.C., S.Y.T., S.T.L., L.Z.); Department of Cardiology, National University Heart Centre, Singapore (P.C., A.F.H.L., M.Y.Y.C.); Yong Loo Lin School of Medicine (P.C., L.L.S.T., C.C.O., Y.J.F.K., A.F.H.L., M.Y.Y.C.) and Department of Biomedical Engineering (L.Z.), National University of Singapore, Singapore; and Department of Diagnostic Imaging, National University Hospital, Singapore (L.L.S.T., C.C.O.).

出版信息

Radiol Cardiothorac Imaging. 2023 Dec;5(6):e230064. doi: 10.1148/ryct.230064.

Abstract

Purpose To develop a new coronary CT angiography (CCTA)-based index, α×LL/MLD, that considers lesion entrance angle (α) in addition to lesion length (LL) and minimal lumen diameter (MLD) and to evaluate its efficacy in predicting hemodynamically significant coronary stenosis compared with invasive coronary angiography (ICA)-derived fractional flow reserve (FFR). Materials and Methods This prospective study enrolled participants (September 2016-March 2020) from two centers who underwent CCTA followed by ICA (ClinicalTrials.gov identifier: NCT03054324). CCTA images were processed semiautomatically to measure LL, MLD, and α for calculating α×LL/MLD. Diagnostic performance and accuracy of α×LL/MLD and LL/MLD in detecting hemodynamically significant coronary stenosis were compared against the reference standard (invasive FFR ≤ 0.80). Results In total, 133 participants (mean age, 63 years ± 9 [SD]; 99 [74%] men) with 210 stenosed coronary arteries were analyzed. Median α×LL/MLD was 54.0 degree/mm (IQR, 25.3-128.7) in participants with invasive FFR of 0.80 or less and 6.7 degree/mm (IQR, 3.3-12.8) in participants with invasive FFR of more than 0.80 ( < .001). The per-vessel accuracy, sensitivity, specificity, positive predictive value, and negative predictive value for discriminating ischemic lesions were 86.2%, 83.1%, 88.4%, 84.1%, and 87.7% for α×LL/MLD and 80.5%, 66.3%, 90.9%, 84.3%, and 78.6% for LL/MLD, respectively. Area under the receiver operating characteristic curve for discriminating hemodynamically significant stenosis was 0.93 for α×LL/MLD, which was significantly greater than the values of 0.84 for LL/MLD and 0.63 for diameter stenosis (both < .001). Conclusion The new morphologic index, α×LL/MLD, incorporating lesion entrance angle achieved higher diagnostic performance in detecting hemodynamically significant lesions compared with diameter stenosis and LL/MLD. CT Angiography, Cardiac, Coronary Arteries, Ischemia, Infarction, Technology Assessment Clinical trial registration no. NCT03054324 © RSNA, 2023 See also the commentary by Fairbairn and Nørgaard in this issue.

摘要

目的 开发一种新的基于冠状动脉 CT 血管造影(CCTA)的指数,即α×LL/MLD,该指数除了病变长度(LL)和最小管腔直径(MLD)外,还考虑了病变入口角度(α),并评估其预测血流动力学意义上的冠状动脉狭窄的效果与有创冠状动脉造影(ICA)衍生的血流储备分数(FFR)相比。

材料与方法 本前瞻性研究于 2016 年 9 月至 2020 年 3 月从两个中心招募了接受 CCTA 后行 ICA 的参与者(ClinicalTrials.gov 标识符:NCT03054324)。使用半自动方法处理 CCTA 图像以测量 LL、MLD 和 α,用于计算α×LL/MLD。与参考标准(有创 FFR≤0.80)相比,比较了α×LL/MLD 和 LL/MLD 在检测血流动力学意义上的冠状动脉狭窄方面的诊断性能和准确性。

结果 共分析了 133 名(平均年龄 63 岁±9[标准差];99[74%]名男性)有 210 处狭窄冠状动脉的参与者。在有创 FFR 为 0.80 或更低的参与者中,α×LL/MLD 的中位数为 54.0°/mm(IQR,25.3°-128.7°),在有创 FFR 大于 0.80 的参与者中为 6.7°/mm(IQR,3.3°-12.8°)(<0.001)。对于缺血性病变,α×LL/MLD 的血管准确性、敏感度、特异度、阳性预测值和阴性预测值分别为 86.2%、83.1%、88.4%、84.1%和 87.7%,而 LL/MLD 分别为 80.5%、66.3%、90.9%、84.3%和 78.6%。用于区分血流动力学意义上的狭窄的受试者工作特征曲线下面积,α×LL/MLD 为 0.93,显著大于 LL/MLD 的 0.84 和直径狭窄的 0.63(均<0.001)。

结论 与直径狭窄和 LL/MLD 相比,新的形态学指数,即纳入病变入口角度的α×LL/MLD,在检测血流动力学意义上的狭窄方面具有更高的诊断性能。

https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8ebd/11163246/500b0e20d819/ryct.230064.VA.jpg

文献检索

告别复杂PubMed语法,用中文像聊天一样搜索,搜遍4000万医学文献。AI智能推荐,让科研检索更轻松。

立即免费搜索

文件翻译

保留排版,准确专业,支持PDF/Word/PPT等文件格式,支持 12+语言互译。

免费翻译文档

深度研究

AI帮你快速写综述,25分钟生成高质量综述,智能提取关键信息,辅助科研写作。

立即免费体验