慢性脑供血不足患者颈内动脉重度狭窄的影响因素及预测模型构建与验证

Influencing factors and construction and validation of a prediction model for severe internal carotid artery stenosis in patients with chronic cerebral circulatory insufficiency

摘要:

目的 分析慢性脑供血不足（CCCI）患者颈内动脉重度狭窄的影响因素，进而构建预测模型并验证该模型。方法 纳入2024年6月至2025年2月保定市第二医院收治的375例CCCI患者（建模组），根据颈内动脉狭窄程度分为轻中度狭窄组（230例）和重度狭窄组（145例）。收集2组临床资料，通过多因素Logistic回归分析评估CCCI患者颈内动脉重度狭窄的影响因素，并建立预测模型。另采用抽样法纳入该院2025年3月至2025年6月收治的125例CCCI患者，开展模型验证（验证组）。结果 多因素Logistic回归分析显示，吸烟史、饮酒史、高血压、糖尿病、同型半胱氨酸（Hcy）>15 μmol/L、D-二聚体>0.5 mg/L是CCCI患者颈内动脉重度狭窄的影响因素（P<0.05）。建模组受试者操作特征曲线及曲线下面积（AUC）为0.915（95%CI：0.878～0.953），灵敏度为86.8%，特异度为80.7%；验证组AUC为0.892（95%CI：0.831～0.953），灵敏度为80.2%，特异度为85.8%。校准曲线显示模型拟合良好（建模组χ²=9.328，P=0.315；验证组χ²=8.642，P=0.471）。决策曲线分析显示，在合理阈概率区间内该模型均显示正向净获益。结论 吸烟史、饮酒史、高血压、糖尿病、Hcy>15 μmol/L、D-二聚体>0.5 mg/L是CCCI患者颈内动脉重度狭窄的影响因素。据此构建的预测模型区分度、拟合度及临床实用性良好，验证表现佳，可为临床早期识别高危患者提供参考。

Abstract:

Objective To analyze the influencing factors for severe internal carotid artery stenosis in patients with chronic cerebral circulatory insufficiency （CCCI）， and construct and validate a prediction model accordingly.Methods A total of 375 CCCI patients admitted to The Second Hospital of Baoding between June 2024 and February 2025 were included as the modeling group. According to the degree of internal carotid artery stenosis， they were divided into a mild/moderate stenosis group （n=230） and a severe stenosis group （n=145）. Clinical data were collected from both groups. Multivariable logistic regression analysis was used to identify the influencing factors for severe internal carotid artery stenosis in CCCI patients， and a prediction model was established. Additionally， 125 CCCI patients admitted to the hospital between March 2025 and June 2025 were enrolled using a sampling method for model validation （validation group）.Results Multivariable logistic regression analysis showed that smoking history， drinking history， hypertension， diabetes mellitus， homocysteine >15 μmol/L， and D-dimer >0.5 mg/L were influencing factors for severe internal carotid artery stenosis in CCCI patients （all P<0.05）. In the modeling group， the area under the receiver operating characteristic curve was 0.915 （95% confidence interval： 0.878-0.953）， with a sensitivity of 86.8% and a specificity of 80.7%. In the validation group， the area under the receiver operating characteristic curve was 0.892 （95% confidence interval： 0.831-0.953）， with a sensitivity of 80.2% and a specificity of 85.8%. Calibration curves demonstrated good model fit （modeling group： χ²=9.328， P=0.315； validation group： χ²=8.642， P=0.471）. Decision curve analysis showed positive net benefits within a reasonable threshold probability range.Conclusions Smoking history， drinking history， hypertension， diabetes mellitus， homocysteine >15 μmol/L， and D-dimer >0.5 mg/L are influencing factors for severe internal carotid artery stenosis in CCCI patients. The prediction model constructed based on these factors demonstrates good discriminative ability， calibration， and clinical utility， with satisfactory validation performance， and may serve as a reference for early identification of high-risk patients in clinical practice.