急性脑梗死溶栓治疗发生恶性脑水肿列线图预测模型的构建及验证  

Construction and verification of a nomogram prediction model for the occurrence of malignant cerebral edema following thrombolytic treatment of acute cerebral infarction

作　　者：李其玲 刘敏 李军[1] 陈洪涛 郭怀杰 陈建霞[1] LI Qi-ling;LIU Min;LI Jun;CHEN Hong-tao;GUO Hua-ijie;CHEN Jian-xia(Department of Neurology,Luxian County People's Hospital,Luzhou 646100,China)

机构地区：[1]泸县人民医院神经内科,四川泸州646100

出　　处：《中国临床神经外科杂志》2024年第11期670-674,共5页Chinese Journal of Clinical Neurosurgery

基　　金：2021年泸县经济信息科学技术项目(LXYJKF-2021-12)。

摘　　要：目的 探讨急性脑梗死(ACI)溶栓治疗发生恶性脑水肿的危险因素,并构建列线图预测模型。方法 回顾性分析2019年10月至2022年9月采用溶栓治疗的233例ACI的临床资料。溶栓治疗后72 h内意识水平下降或神经功能恶化,头颅MRI或CT显示脑梗死范围超过1/2大脑中动脉供血区伴中线移位>5 mm诊断为恶性脑水肿。采用多因素logistic回归模型分析恶性脑水肿的危险因素,采用R软件包构建列线图模型,采用Bootstrap法验证;绘制受试者工作特征(ROC)曲线评估列线图模型的预测效能。结果 233例中,50例发生恶性脑水肿,发生率为21.5%。多因素logistic回归分析显示,年龄(OR=1.238;95%CI1.081~1.419;P=0.002)、脑梗死面积(OR=1.912;95%CI 1.115~3.280;P=0.019)、发病至溶栓时间(OR=9.828;95%CI 1.829~52.815;P=0.008)、白细胞计数(OR=2.289;95%CI 1.376~3.809;P=0.001)、溶栓前NIHSS评分(OR=6.503;95%CI 2.318~18.245;P<0.001)是ACI溶栓治疗发生恶性脑水肿的独立危险因素。基于多因素logistic回归分析结果,R软件成功构建列线图预测模型,Bootstrap法验证显示列线图模型校正曲线趋近于理想曲线(一致性指数为0.746;P=0.459);ROC曲线分析显示列线图模型预测ACI溶栓治疗发生恶性脑水肿风险曲线下面积为0.977(95%CI 0.951~0.999),预测效能高。结论 年龄、发病至溶栓时间、脑梗死面积、白细胞计数、溶栓前NIHSS评分是影响ACI溶栓治疗并发恶性脑水肿的独立危险因素。基于这些危险因素构建的列线图风险预测模型对ACI溶栓治疗发生恶性脑水肿有较高的预测效能。Objective To explore the risk factors of malignant cerebral edema(MCE)following thrombolytic treatment for acute cerebral infarction(ACI)and to construct a nomogram prediction model.Methods The clinical data of 233 patients with ACI who underwent thrombolytic treatment from October 2019 to September 2022 were retrospectively analyzed.MCE was defined as a decline in the level of consciousness or deterioration of neurological function within 72 hours after thrombolytic treatment,and the area of cerebral infarction exceeding half of the territory supplied by the middle cerebral artery with a midline shift>5 mm as shown by head MRI or CT.A multivariate logistic regression model was employed to analyze the independent risk factors for MCE.A nomogram prediction model was constructed using the R software package and internally validated by the Bootstrap method.The receiver operating characteristic(ROC)curve was plotted to assess the predictive efficacy of the model.Results Of the 233 patients,50 patients(21.5%)developed MCE.Multivariate logistic regression analysis revealed that age(OR=1.238;95%CI 1.081~1.419;P=0.002),the area of cerebral infarction(OR=1.912;95%CI 1.115~3.280;P=0.019),the time from onset to thrombolysis(OR=9.828;95%CI 1.829~52.815;P=0.008),white blood cell(OR=2.289;95%CI 1.376~3.809;P=0.001),and the NIHSS score before thrombolysis(OR=6.503;95%CI 2.318~18.245;P<0.001)were independent risk factors for MCE after thrombolytic treatment for ACI.Based on the above findings,a nomogram prediction model was successfully constructed.The Bootstrap method validation indicated that the calibration curve of the model approached the ideal curve(the concordance index was 0.746;P=0.459).The ROC curve analysis demonstrated that the area under the curve for predicting the risk of MCE after thrombolytic treatment for ACI was 0.977(95%CI 0.951~0.999),indicating high predictive efficacy.Conclusions Age,the time from onset to thrombolysis,the area of cerebral infarction,white blood cell,and the NIHSS score before thrombolysis

关 键 词：急性脑梗死 溶栓治疗 恶性脑水肿 危险因素 列线图模型 

分 类 号：R743[医药卫生—神经病学与精神病学]