儿童热性惊厥日后癫痫发作的危险因素及列线图预测模型构建  

Risk factors for epileptic seizures after febrile seizures in children andconstruction of a nomogram prediction model

作　　者：李清峰 刘欢 董改琴[1] LI Qingfeng;LIU Huan;DONG Gaiqin(Department of Pediatrics,Affiliated Hospital of Yangzhou University,Yangzhou,Jiangsu 225000,China)

机构地区：[1]扬州大学附属医院儿科,江苏扬州225000

出　　处：《中华全科医学》2024年第1期86-88,153,共4页Chinese Journal of General Practice

基　　金：江苏省卫生健康委员会基金资助项目(M2021076)。

摘　　要：目的 分析儿童热性惊厥日后癫痫发作的危险因素,并构建个体化预测癫痫发作的列线图模型。方法 选取扬州大学附属医院2019年9月—2022年10月期间296例热性惊厥儿童为研究对象,根据惊厥日后是否癫痫发作分为癫痫组(54例)和非癫痫组(242例)。经单因素及多因素logistic回归分析确定儿童热性惊厥日后癫痫发作的影响因素,采用R语言软件包建立预测儿童热性惊厥日后癫痫发作的列线图模型,绘制校准曲线及ROC曲线评估列线图模型的预测效果。结果 癫痫组初次发作年龄为3个月~1岁、首次惊厥次数≥4次、复杂性热性惊厥、有脑电图异常、有癫痫家族史比例均显著高于非癫痫组(P<0.05)。初次发作年龄为3个月~1岁(OR=3.425,P<0.05)、首次惊厥次数≥4次(OR=2.974,P<0.05)、复杂性热性惊厥(OR=4.621,P<0.05)、有脑电图异常(OR=2.836,P<0.05)、有癫痫家族史(OR=5.572,P<0.05)均为影响儿童热性惊厥日后癫痫发作的独立危险因素。校准曲线趋近于理想曲线,列线图模型预测儿童热性惊厥日后癫痫发作的预测值与实际值基本一致。ROC曲线下面积为0.896(95%CI:0.851~0.940),预测效能中等。结论 本研究基于多因素logistic回归分析构建的预测模型可较好地预测儿童热性惊厥日后癫痫发作,对防治癫痫发作有一定的指导意义。Objective To analyze the risk factors for epileptic seizures in children after febrile seizures and to develop an individual nomograph model to predict epileptic seizures.Methods A total of 296 children with febrile seizures in the Department of Pediatrics,Affiliated Hospital of Yangzhou University from September 2019 to October 2022 were selected as study subjects,the patients were divided into epileptic group(54 cases)and nonepileptic group(242 cases)according to whether they had epileptic seizure after the day of seizure.The factors influencing epileptic seizures in children after febrile seizure were determined by single-factor and multiple-factor logistic regression analyses,a nomogram model was constructed using the R language software package to predict seizures after febrile seizure in children,and a calibration curve and ROC curve were drawn to evaluate the predictive ability of the nomogram model.Results The proportions of patients with the age of first seizure between 3 months and 1 year,number of first seizures≥4 times,complex febrile seizures,abnormal electroencephalogram(EEG)and family history of epilepsy in the epileptic group were significantly higher than those in the nonepileptic group(P<0.05).Age of first seizure between 3 months and 1 year(OR=3.425,P<0.05),number of first seizures≥4 times(OR=2.974,P<0.05),complex febrile seizures(OR=4.621,P<0.05),abnormal EEG(OR=2.836,P<0.05)and family history of epilepsy(OR=5.572,P<0.05)were all independent risk factors for epileptic seizure after febrile seizure in children.The calibration curve was close to the ideal curve,and the predictive value of the nomogram model for predicting epileptic seizures after febrile seizure in children was basically in agreement with the actual value.The area under the ROC curve was 0.896(95%CI:0.851-0.940),indicating moderate predictive efficiency.Conclusion The prediction model based on multifactor logistic regression analysis in this study can better predict seizures after childhood febrile seizures,and has some guid

关 键 词：热性惊厥 儿童 癫痫发作 危险因素 列线图预测模型 

分 类 号：R729[医药卫生—儿科] R742.1[医药卫生—临床医学]