基于几何形态学参数预测左前降支冠状动脉动脉粥样硬化异质性病因的列线图
本研究利用冠状动脉CT血管造影(CCTA)提取左前降支(LAD)的曲率、扭转及迂曲指数等几何形态参数及周围附属结构解剖特征,结合临床与血清学指标构建逻辑回归模型预测LAD中度至重度狭窄(MSS-LAD)。单中心回顾性研究中,训练集115例CCTA病例中49例为MSS-LAD;基于BMI、中间支-对角支(RI-D)分型、脂肪体积(FV)及近中段平均扭转(AT-PM)构建的列线图在训练集AUC为0.839(95%CI 0.762-0.961),验证集AUC为0.790(95%CI 0.665-0.915)。
要点
- 基于CCTA的几何形态参数(扭转、迂曲指数)及周围脂肪体积可辅助预测LAD中度至重度狭窄。
- 列线图整合BMI、RI-D分型、FV及AT-PM四个指标,训练集AUC 0.839、验证集AUC 0.790。
- 作者认为该列线图可考虑纳入未来冠心病高危人群的风险评估与管理方案。
核心结论
- 冠状动脉几何构型与动脉粥样硬化斑块形成之间存在复杂的相互作用;结合周围附属结构解剖特征与临床指标构建的风险列线图能有效预测MSS-LAD,作者认为可考虑将其纳入未来冠心病高危人群的风险评估与管理方案。
研究设计与样本
- 单中心回顾性研究,训练集115例CCTA病例(其中49例为MSS-LAD),基于CCTA提取曲率、扭转、迂曲指数(TI)及周围附属结构解剖特征,结合临床指标与血清学结果构建逻辑回归模型并验证列线图。
关键影像表现/征象
- MSS-LAD组与非MSS-LAD组在中间支-对角支(RI-D)分型、脂肪体积(FV)、平均扭转(AT)及近中段平均扭转(AT-PM)上差异有统计学意义(p<0.05)。
诊断标准与关键数值
- 多因素逻辑回归显示BMI、RI-D分型、FV及AT-PM为MSS-LAD的显著预测因子(p<0.05)。
- 基于上述四指标构建的列线图预测MSS-LAD的AUC在训练集为0.839(95%CI 0.762-0.961),在验证集为0.790(95%CI 0.665-0.915)。
临床应用与阅片要点
- 作者认为该列线图可考虑纳入未来冠心病高危人群的风险评估与管理方案。
局限性
- 原文未报告局限性。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
英文原文(摘要)· 点击展开
AIM: This study aims to analyse the correlation between the geometric morphology of the left anterior descending (LAD) artery and the severity of arterial stenosis in order to explore potential causes of atherosclerotic heterogeneity.
MATERIALS AND METHODS: Geometric morphological parameters, including curvature, torsion, and tortuosity index (TI), along with anatomical features of surrounding accessory structures, were derived from coronary computed tomography angiography (CCTA). By integrating clinical indicators and serological results, a logistic regression model was developed to predict moderate to severe stenosis of the LAD (MSS-LAD). A novel nomogram was constructed based on significant influencing factors identified in the model, and its predictive performance was validated.
RESULTS: In this single-centre retrospective study, out of 115 CCTA cases in the training set, 49 were identified with MSS-LAD. Significant differences were observed in the classification of ramus intermedius-diagonal branch (RI-D), fat volume (FV), average torsion (AT), and AT of the proximal and mid (AT-PM) between groups (p<0.05). Multivariate logistic regression analysis indicated that body mass index (BMI), RI-D classification, FV, and AT-PM were significant predictors of MSS-LAD (p<0.05). A nomogram for predicting MSS-LAD was constructed based on these four indicators, demonstrating good predictive performance in both the training set (area under the curve [AUC]: 0.839; 95% confidence interval [CI]: 0.762-0.961) and the validation set (AUC: 0.790; 95% CI: 0.665-0.915).
CONCLUSION: There is a complex interaction between the geometric configuration of coronary arteries and atherosclerotic plaque formation. The risk nomogram established by combining anatomical features of surrounding accessory structures and clinical indicators effectively predicts MSS-LAD. It may be considered for inclusion in future risk assessment and management protocols for populations at high risk of coronary heart disease.
引用
- Yu X, Gao QX, Sun LG, et al. A nomogram for predicting heterogeneous etiology of atherosclerosis in the left anterior descending coronary artery using geometric morphological parameters. Clin Radiol. 2025. PMID: 40043631. doi:10.1016/j.crad.2025.106842
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。