急性缺血性脑卒中康复无效的风险预测Nomogram模型构建及验证  

Construction of a Nomogram to predict the risk of ineffective rehabilitation of acute ischemic stroke and its verification

作　　者：李晗静 周卫鹏 LI Han-jing;ZHOU Wei-peng

机构地区：[1]郑州卷烟厂康复医院神经康复科,河南郑州450000

出　　处：《中国疗养医学》2025年第1期15-21,共7页Chinese Journal of Convalescent Medicine

摘　　要：目的构建急性缺血性脑卒中康复无效的风险预测Nomogram模型并进行验证。方法回顾性分析2018年10月至2022年10月郑州卷烟厂康复医院收治的396例急性缺血性脑卒中患者的临床资料,按照2∶1的比例将其随机分为建模组(264例)和验证组(132例);根据建模组患者治疗后3个月的康复情况将其进一步分为无效组(63例)和有效组(201例),采用多因素Logistic回归分析法分析急性缺血性脑卒中康复无效的危险因素;采用R 3.4.3软件包绘制Nomogram模型,采用Bootstrap法进行内部验证,并绘制校准曲线评价Nomogram模型的校准度,绘制受试者工作特征(ROC)曲线评价Nomogram的预测效能,采用决策曲线(DCA)验证模型的临床净获益率。结果建模组中无效组心房颤动史、梗死灶最大径>4 cm、意识障碍、出血转化占比以及美国国立卫生院神经功能缺损评分(NIHSS)、血糖水平、糖化血红蛋白水平、D-二聚体(D-D)水平均高于有效组(P<0.05),发病至溶栓时间长于有效组(P<0.05);经多因素Logistic回归分析显示,心房颤动、发病至溶栓时间长、NIHSS评分升高、梗死灶最大径>4 cm、出血转化、血糖水平升高、D-D水平升高均是急性缺血性脑卒中患者康复无效的危险因素(P<0.05);根据以上影响因素构建的Nomogram模型经Bootstrap法验证,建模组与验证组的一致性指数分别为0.830、0.816,校正曲线与标准曲线拟合度较好;ROC曲线结果显示,建模组、验证组Nomogram预测急性缺血性脑卒中患者康复无效的曲线下面积分别为0.873、0.856;DCA曲线显示,建模组和验证组的风险阈值概率分别在2%~99%及9%~90%时有较高的净获益值。结论心房颤动、发病至溶栓时间长、NIHSS评分升高、梗死灶最大径>4 cm、出血转化、血糖水平升高、D-D水平升高均是急性缺血性脑卒中患者康复无效的危险因素,据此构建的Nomogram模型具有较好的临床应用价值。Objective To construct a nomogram to predict the risk of ineffective rehabilitation of acute ischemic stroke and to validate its performance.Methods The clinical data of 396 patients with acute ischemic stroke admitted to Zhengzhou Cigarette Factory Rehabilitation Hospital from October 2018 to October 2022 were retrospectively analyzed.They were randomly divided into modeling group(264 cases)and verification group(132 cases)according to a ratio of 2∶1.According to the rehabilitation of patients in the modeling group 3 months after treatment,they were further divided into ineffective group(63 cases)and effective group(201 cases).Multivariate Logistic regression analysis was used to analyze the risk factors of ineffective rehabilitation of acute ischemic stroke.The R 3.4.3 software package was used to construct a Nomogram to predict the risk of ineffective rehabilitation of acute ischemic stroke and its internal verification was performed by the Bootstrap method.The calibration curve was drawn to evaluate the calibration of the Nomogram,and the receiver operating characteristic(ROC)curve was drawn to evaluate the predictive efficacy of the Nomogram.The decision curve(DCA)was used to verify the clinical net benefit rate of the nomogram.Results In the modeling groups,history of atrial fibrillation,maximum diameter of infarction>4 cm,disturbance of consciousness,proportion of hemorrhagic transformation,the National Institutes of Health stroke scale(NIHSS)score,blood glucose level,glycosylated hemoglobin level and D-dimer(D-D)level in the ineffective group were significantly higher than those in the effective group(P<0.05),and the time from onset to thrombolysis was significantly longer than that in the effective group(P<0.05).Multivariate Logistic regression analysis showed that atrial fibrillation,long time from onset to thrombolysis,elevated NIHSS score,maximum diameter of infarction>4 cm,hemorrhagic transformation,elevated blood glucose level and elevated D-D level were all risk factors for ineffective rehabilita

关 键 词：急性缺血性脑卒中 康复无效 Nomogram模型 风险预测 

分 类 号：R743.3[医药卫生—神经病学与精神病学]