基于炎症因子水平和颅内动脉易损斑块参数构建老年短暂性脑缺血发作后进展为脑梗死的预测模型
A prediction model of progression to cerebral infarction after transient ischemic attack in the elderly based on inflammatory factors and vulnerable intracranial artery plaque parameters
韩娟 1李旭 1韩悦 1陈博 1许超峰 1王民珩 1陈亚伦1
作者信息
- 1. 473000 河南省南阳市第二人民医院
- 折叠
摘要
目的 基于炎症因子水平和颅内动脉易损斑块参数构建老年短暂性脑缺血发作(Tansient is-chemic attack,TIA)后进展为脑梗死的预测模型,并完成预测价值验证.方法 回顾性选取2020年10月-2023年10收治的老年TIA患者300例为研究对象,按照7:3的分配比例分为建模组210例和验证组90例;建模组患者根据住院7d随访记录进展为脑梗死与否分为非脑梗死组与脑梗死组;收集建模组患者包括炎症因子指标与颅内动脉易损斑块参数在内的临床基线资料;进行非脑梗死组与脑梗死组的组间单因素分析以确定收集信息中影响患者TIA后进展为脑梗死的相关因素;对相关因素行二元Logistic回归分析以筛选独立影响因素,并以独立影响因素为基础,结合验证组资料收集,通过R软件完成列线图预测模型构建、验证与价值分析.结果 建模组老年TIA患者进展为脑梗死47例,进展率为22.38％(47/210);患者年龄、合并高血压病的占比、血清白细胞介素-6(Interleukin-6,IL-6)、肿瘤坏死因子-α(Tumor necrosis factor-α,TNF-α)、C反应蛋白(C-reaction protein,CRP)水平、颈动脉狭窄程度与颅内动脉易损斑块的重构、负荷、强化程度等均是影响患者TIA后进展为脑梗死的相关因素(P均＜0.05),其中血清IL-6,CRP水平与颅内动脉易损斑块的负性重构、负荷、强化程度等为影响患者TIA后进展为脑梗死的独立危险因素(P均＜0.05);列线图预测模型验证显示,校准图显示预测模型性能良好;建模组受试者工作特征(Receiver operating characteristic,ROC)曲线的线下面积(Area under curve,AUC)为0.907;验证组ROC的AUC为0.950,校正曲线拟合度良好(P均＜0.05).结论 基于炎症因子水平和颅内动脉易损斑块参数构建的列线图预测模型对老年TIA后进展为脑梗死风险具有较高预测价值,或可通过模型筛选脑梗死高风险患者,开展针对性治疗方案的制定,预防脑梗死发生,改善患者预后.
Abstract
Objective To establish a predictive model for cerebral infarction after transient ischemic at-tack(TIA)based on inflammatory factors and vulnerable intracranial artery plaque parameters,and validate its predictive value.Methods A total of 300 elderly TIA patients admitted from October 2020 to October 2023 were retrospectively selected as the study objects,and were divided into the modeling group(210 cases)and the validation group(90 cases)in a 7:3 allocation ratio.Patients in the modeling group were divided into nor-mal group and cerebral infarction group according to 7d follow-up records.Clinical baseline data,including in-flammatory factor indicators and intracranial artery vulnerable plaque parameters,were collected in the model group.An intergroup univariate analysis was performed between the normal group and the cerebral infarction group to identify the relevant factors affecting the progression of patients to cerebral infarction after TIA.Bi-nary Logistic regression analysis was performed on relevant factors to screen independent influencing factors.R software was applied to complete the construction,verification and value analysis of the nomogram predic-tion model on the basis of independent influencing factors and in combination with the data collection of the verification group.Results In the modeling group,47 elderly TIA patients developed cerebral infarction,with a progression rate of 22.38％(47/210).Factor analysis found that Patient's age,the proportion of combined hypertension,serum interleukin-6(IL-6),tumor necrosis factor-α(TNF-α),and C-reaction protein(CRP)level,the degree of carotid artery stenosis and the remodeling,loading and strengthening degree of vulnerable intracranial artery plaque were all relevant factors affecting the patients'progression to cerebral infarction after TIA(all P＜0.05).Serum IL-6,CRP and the negative remodeling,load and enhancement degree of intracra-nial artery vulnerable plaque were independent risk factors for the progression to cerebral infarction after TIA(all P＜0.05).The calibration diagram shows that the prediction model has good performance.The area un-der curve(AUC)of receiver operating characteristic(ROC)curve in modeling group was 0.907.The AUC of ROC in the verification group was 0.950,and the calibration curve fit was good(all P＜0.05).Conclusion The nomogram prediction model constructed based on inflammatory factors and intracranial artery vulnerable plaque parameters has a high predictive value for the risk of cerebral infarction progression after TIA in the elderly.The model may be applied to screen high-risk patients with cerebral infarction,develop targeted treat-ment plans,prevent the occurrence of cerebral infarction,and improve the prognosis of patients.
关键词
短暂性脑缺血发作/脑梗死/炎症因子/易损斑块/因素分析/列线图Key words
Tansient ischemic attack/Cerebral infarction/Inflammatory factors/Vulnerable plaque/Factor analysis/Nomograph引用本文复制引用
基金项目
河南省医学科技攻关计划联合共建项目(LHGJ20230962)
出版年
2024
NETL
NSTL
万方数据