基于超声特征、临床危险因素和全身炎症标志物的多模态列线图预测颈动脉易损斑块的开发与验证
前瞻性纳入2024年4月至2025年5月行颈动脉内膜剥脱术的146例患者,分训练队列100例和验证队列46例,用LASSO筛选8个独立预测因子构建列线图;训练队列AUC 0.911(95%CI 0.851-0.972),验证队列AUC 0.889(95%CI 0.786-0.992),校准曲线、决策曲线分析和临床影响曲线显示模型预测与观测一致且临床效用良好。
要点
- 列线图纳入NLR、斑块内新生血管分级、吸烟史、中性粒细胞计数、BMI、斑块回声、CRP和斑块形态共8个预测因子。
- 训练队列AUC 0.911(95%CI 0.851-0.972),验证队列AUC 0.889(95%CI 0.786-0.992)。
- 校准曲线显示预测与观测一致,DCA和CIC确认模型临床效用良好。
核心结论
- 该多模态列线图能准确预测颈动脉易损斑块、辅助患者风险分层,并支持对最高危患者启动针对性预防治疗。
研究设计与样本
- 前瞻性研究,纳入2024年4月至2025年5月行颈动脉内膜剥脱术的患者146例,分为训练队列100例和验证队列46例;候选变量包括临床特征、实验室参数和多模态超声斑块特征,采用LASSO回归筛选变量并构建列线图,以AUC、校准曲线、决策曲线分析和临床影响曲线评估性能。
关键影像表现/征象
- 超声斑块特征中,斑块内新生血管分级、斑块回声和斑块形态被LASSO回归选为列线图的独立预测因子。
- 斑块内新生血管分级、斑块回声、斑块形态与临床危险因素及全身炎症标志物共同构成多模态列线图。
诊断标准与关键数值
- 列线图在训练队列的AUC为0.911(95%CI 0.851-0.972),在验证队列的AUC为0.889(95%CI 0.786-0.992)。
- LASSO回归识别出8个独立预测因子:中性粒细胞与淋巴细胞比值、斑块内新生血管分级、吸烟史、中性粒细胞计数、体重指数、斑块回声、C反应蛋白和斑块形态。
临床应用与阅片要点
- 作者认为该列线图可准确预测颈动脉易损斑块、辅助患者风险分层,并支持对最高危患者启动针对性预防治疗。
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
OBJECTIVES: This study aimed to develop and validate a multimodal nomogram that integrates ultrasound features, clinical risk factors, and systemic inflammatory markers to predict carotid vulnerable plaques (VPs), assess its precision in recognizing high-risk plaques associated with stroke, and offer personalized clinical interventions.
METHODS: A total of 146 patients who underwent carotid endarterectomy (CEA) between April 2024 and May 2025 were enrolled in this prospective study and subsequently divided into a training cohort (n = 100) and a validation cohort (n = 46). The candidate variables comprised clinical characteristics, laboratory parameters, and multimodal ultrasound plaque features. The least absolute shrinkage and selection operator (LASSO) regression was utilized for robust variable selection to mitigate overfitting. A nomogram was developed based on the selected features, and its performance was evaluated using the area under the receiver operating characteristic curve (AUC), calibration curves, decision curve analysis (DCA), and clinical impact curve (CIC).
RESULTS: The LASSO regression analysis identified 8 independent predictors for the nomogram: neutrophil-to-lymphocyte ratio (NLR), intraplaque neovascularization (IPN) grade, smoking history, neutrophil count, body mass index (BMI), plaque echogenicity, C-reactive protein (CRP), and plaque morphology. The nomogram exhibited excellent discriminative capacity, with an AUC of 0.911 (95% CI: 0.851-0.972) in the training cohort and 0.889 (95% CI: 0.786-0.992) in the validation cohort. Calibration curves demonstrated good concordance between predictions and observations, whereas DCA and CIC confirmed the model's favorable clinical utility.
CONCLUSIONS: The developed multimodal nomogram accurately predicts carotid VPs, facilitates patient risk stratification, and supports the initiation of targeted preventive therapies for those at the highest risk.
引用
- Liu Y, Zhang L, Xiong J, et al. Development and Validation of a Multimodal Nomogram for Carotid Vulnerable Plaques Based on Ultrasound Features, Clinical Risk Factors, and Systemic Inflammatory Markers. J Ultrasound Med. 2025. PMID: 41410304. doi:10.1002/jum.70144
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。