老年急性缺血性脑卒中后认知功能障碍预测模型的初步建立
Preliminary establishment of a predictive model for cognitive dysfunction in the elderly after acute ischemic stroke
张竣婷 1关卓 1滕海艳 1王恩石 1魏然1
作者信息
- 1. 150001 哈尔滨医科大学附属第一医院神经内科
- 折叠
摘要
目的 建立老年急性缺血性脑卒中(Acute ischemic stroke,AIS)后认知功能障碍(Post-stroke cognitive impairment,PSCI)的预测模型,并对其预测效能进行初步评价.方法 回顾性选取2020年1月-2022年12月哈尔滨医科大学附属第一医院治疗的236例老年AIS患者,根据6个月内是否发生PSCI,将其分为PSCI组114例(48.31％)和非PSCI组122例(51.69％);通过电子病历系统收集所有患者的临床资料,采用多因素Logistic回归法筛选老年AIS患者发生PSCI的危险因素,并据此构建预测模型,采用受试者工作特征(Receiver operating characteristi,ROC)曲线评估模型预测PSCI的效能.结果 与非PSCI组比较,PSCI组年龄＞75岁、糖尿病、衰弱的比例以及美国国立卫生研究院卒中量表(National institute of health stroke scale,NIHSS)评分、纤维蛋白原(Fibrinogen,Fib)、同型半胱氨酸、C-反应蛋白(C-reactive protein,CRP)升高,血红蛋白降低(P＜0.05);多因素Logistic回归显示,年龄＞75岁、Fib,CRP、衰弱、NIHSS评分是老年AIS患者发生PSCI的独立危险因素;基于多因素Logistic回归分析显示,老年AIS患者发生PSCI的风险指数Cindex=-1.834+0.469 ×(年龄)+0.278 ×(Fib)+0.331 ×(CRP)+0.392 ×(衰弱)+0.503 ×(NIHSS);ROC曲线显示,Cindex预测老年AIS患者发生PSCI的曲线下面积(Area under the curve,AUC)为0.885(95％CI=0.842～0.928),敏感度为81.58％,特异度为84.43％,准确度为83.05％.结论 年龄、Fib,CRP、衰弱、NIHSS评分是老年AIS患者发生PSCI的独立危险因素,以此构建的模型对于PSCI的识别能力良好,有助于临床实现PSCI高风险人群的筛查.
Abstract
Objective To establish a prediction model for poststroke cognitive impairment(PSCI)in elderly patients with acute ischemic stroke(AIS)and to conduct a preliminary evaluation of its predictive effi-cacy.Methods A total of 236 elderly AIS patients treated at the First Hospital of Harbin Medical University from January 2020 to December 2022 were retrospectively selected and divided into 114 patients(48.31％)in the PSCI group and 122 patients(51.69％)in the non-PSCI group according to whether PSCI occurred within 6 months.The clinical data of all patients were collected through the electronic medical records system,and multifactorial logistic regression was used to screen for risk factors for the occurrence of PSCI in elderly pa-tients with AIS.A prediction model was constructed,and the efficacy of the model in predicting PSCI was as-sessed using receiver operating characteristic(ROC)curves.The results are expressed as the area under the curve(AUC).Results Compared with those in the non-PSCI group,the proportions of individuals aged＞75 years,with diabetes mellitus,and debility,as well as the National Institutes of Health Stroke Scale(NIHSS)score,fibrinogen(Fib),homocysteine,and C-reactive protein(CRP),were significantly greater,and the he-moglobin level was significantly lower in the PSCI group(P＜ 0.05).Multivariate logistic regression revealed that age＞75 years,Fib,CRP,frailty status,and NIHSS score were independent risk factors for the occur-rence of PSCI in elderly AIS patients.Based on the Results of multifactorial logistic regression analysis of the risk index of PSCI in elderly AIS patients,Cindex=-1.834+0.469 ×(age)+0.278 ×(Fib)+0.331 ×(CRP)+0.392 ×(debilitation)+0.503 ×(NIHSS).The ROC curves revealed that the C-index had an AUC of 0.885(95％CI=0.842～0.928)for predicting the occurrence of PSCI in elderly patients with AIS,with a sensitivity of 81.58％,a specificity of 84.43％and an accuracy of 83.05％.Conclusion Age,Fib,CRP,frailty status,and NIHSS score are independent risk factors for the occurrence of PSCI in elderly pa-tients with AIS.The model constructed in this way has good potential for identifying PSCI and is helpful for the clinical screening of high-risk groups for PSCI.
关键词
急性缺血性脑卒中/脑卒中后认知功能障碍/预测模型/老年Key words
Acute ischemic stroke/Poststroke cognitive impairment/Prediction model/Elderly引用本文复制引用
基金项目
黑龙江省卫生健康委科研项目(2020-104)
出版年
2024
NETL
NSTL
万方数据