基于列线图模型的多指标检测对缺血性脑卒中预后预测价值分析  

Value of a nomogram model based on multi-index detection in predicting the prognosis of ischemic stroke

作　　者：宋亚君[1] 王翔 陈武[1] SONG Yajun;WANG Xiang;CHEN Wu(Sinopharm Dongfeng General Hospital,Hubei University of Medicine,Shiyan,Hubei 442000,China)

机构地区：[1]湖北医药学院附属国药东风总医院,湖北十堰442000

出　　处：《国际神经病学神经外科学杂志》2023年第6期13-18,共6页Journal of International Neurology and Neurosurgery

基　　金：2019年湖北省知识创新专项(自然科学基金)项目(2019CFb680)。

摘　　要：目的构建多指标列线图模型以预测缺血性脑卒中患者的预后。方法对2019年1月至2021年6月湖北医药学院附属国药东风总医院接诊的126例缺血性脑卒中患者的资料进行回顾性分析。根据格拉斯哥预后评分将患者分为2组:1~3分者为预后不良组,4~5分者为预后良好组。收集可能影响缺血性脑卒中患者预后的因素,比较2组患者各预后因素,并进行多因素分析,根据多因素分析结果构建列线图模型。结果126例患者中,45例预后不良。多因素分析结果显示,有吸烟史、入院时美国国立卫生研究院脑卒中量表评分升高、低密度脂蛋白胆固醇水平升高、神经肽P物质水平升高为缺血性脑卒中预后不良的危险因素(P<0.05);高密度脂蛋白胆固醇水平升高为保护性因素(P<0.05)。根据多因素分析结果构建列线图模型,受试者操作特征曲线下面积为0.892,灵敏度为93.1%,特异度为68.2%,95%CI为0.836~0.949。计算机模拟充分采样法内部验证结果显示,平均绝对误差为0.03,模型表现与理想模型基本拟合,提示模型预测准确度较高。结论缺血性脑卒中患者的预后与吸烟、入院时美国国立卫生研究院脑卒中量表评分、低密度脂蛋白胆固醇水平、神经肽P物质水平等因素有关。根据上述因素构建的列线图模型用于缺血性脑卒中患者预后预测具有较高的准确度与区分度。Objective To establish a nomogram model based on multi-index detection in predicting the prognosis of patients with ischemic stroke.Methods A retrospective analysis was performed for the data of 126 patients with ischemic stroke who were treated in Sinopharm Dongfeng General Hospital,Hubei University of Medicine,from January 2019 to June 2021.According to the Glasgow Prognostic Score,they were divided into poor prognosis group(1-3 points)and good prognosis group(4-5 points).Factors that might affect the prognosis of patients with ischemic stroke were collected,and each prognostic factor was compared between the two groups.A multivariate analysis was performed,and a nomogram model was constructed based on the results of the multivariate analysis.Results Among the 126 patients,45 had poor prognosis.The multivariate analysis showed that a history of smoking,an increase in National Institutes of Health Stroke Scale(NIHSS)score on admission,elevated low-density lipoprotein cholesterol(LDL-C)level,and elevated neuropeptide P substance level were risk factors for poor prognosis of ischemic stroke(P<0.05),while elevated high-density lipoprotein cholesterol level was a protective factor(P<0.05).The nomogram model constructed based on the results of the multivariate analysis had an area under the receiver operator characteristic curve of 0.892(95% confidence interval:0.836-0.949),with a sensitivity of 93.1% and a specificity of 68.2%.Internal validation based on the Bootstrap method showed a mean absolute error of 0.03,and the performance of the model was basically fitted to that of the ideal model,suggesting that the model had a high accuracy in prediction.Conclusions The prognosis of patients with ischemic stroke is associated with the factors such as smoking,NIHSS score on admission,and the levels of LDL-C and neuropeptide P substance,and the nomogram model based on the above factors has relatively high accuracy and discriminatory ability in predicting the prognosis of patients with ischemic stroke.

关 键 词：缺血性脑卒中 多因素分析 列线图 预后预测 

分 类 号：R743.3[医药卫生—神经病学与精神病学]