癫痫并发偏头痛风险的列线图预测模型构建与验证  

Construction and validation of a nomogram prediction model for the risk of epilepsy complicated with migraine

作　　者：魏瑞花[1] 洪丽[1] 王献[1] Wei Ruihua;Hong Li;Wang Xian(Department of Neurology,Anyang Regional Hospital,Anyang 455000,China)

机构地区：[1]安阳地区医院神经内科,河南安阳455000

出　　处：《临床内科杂志》2025年第2期113-117,共5页Journal of Clinical Internal Medicine

摘　　要：目的探讨癫痫并发偏头痛的风险因素,并据此构建列线图预测模型。方法将372例癫痫患者按照2∶1的比例随机分为建模组(248例)和验证组(124例)。根据有无偏头痛情况,将建模组再分为偏头痛组(54例)和无偏头痛组(194例)。采用多因素logistic回归分析评估癫痫并发偏头痛的危险因素,构建列线图预测模型。采用Bootstrap法进行内部验证,采用校准曲线、受试者工作特征(ROC)曲线评估模型校准度及区分度,采用决策曲线(DCA)评估模型临床净获益率。结果与无偏头痛组比较,偏头痛组癫痫病程更长,女性、高血压史、偏头痛家族史、夜间睡眠不佳、焦虑情绪患者比例及发作频率、血清C反应蛋白(CRP)、IL-6、肿瘤坏死因子(TNF)-α水平均更高(P<0.05)。多因素logistic回归分析结果显示,女性、癫痫病程长、发作频率高、偏头痛家族史、夜间睡眠不佳、焦虑情绪、高血压史及高水平血清CRP、IL-6、TNF-α均是癫痫并发偏头痛的独立危险因素(P<0.05)。建立的列线图预测模型经内部验证显示一致性指数(C-index)为0.821(95%CI 0.789~0.855),建模组与验证组ROC曲线下面积(AUC)分别为0.833和0.819,校准曲线显示模型拟合度良好,DCA分析表明当阈值概率在6%~79%、81%~92%时可获得临床净收益。结论女性、病程、发作频率、偏头痛家族史、夜间睡眠不佳、焦虑情绪、高血压史及血清CRP、IL-6、TNF-α水平均是癫痫并发偏头痛的危险因素,基于上述危险因素建立的列线图预测模型预测效能较好,有助于临床早期筛选高风险患者。Objective To explore the risk factors of epilepsy complicated with migraine,and to construct a nomogram prediction model.Methods A total of 372 patients with epilepsy were randomly divided into modeling group(248 cases)and validation group(124 cases)according to the ratio of 2∶1.According to the presence or absence of migraine,the modeling group was further divided into migraine group(54 cases)and non⁃migraine group(194 cases).Multivariate logistic regression analysis were used to evaluate the risk factors of epilepsy complicated with migraine,and a nomogram prediction model was constructed.The Bootstrap method was used for internal validation,calibration curve and receiver operating characteristic(ROC)curve were used to evaluate the calibration and discrimination of the model,and decision curve analysis(DCA)was used to evaluate the net clinical benefit rate of the model.Results Compared with non⁃migraine group,migraine group had significantly longer duration of epilepsy,significantly higher proportion of women,history of hypertension,family history of migraine,poor sleep at night,anxiety and seizure frequency,as well as significantly higher serum levels of C⁃reactive protein(CRP),IL⁃6 and tumor necrosis factor(TNF)⁃α(P<0.05).Multivariate logistic regression analysis showed that female,long duration of epilepsy,high frequency of attack,family history of migraine,poor sleep at night,anxiety,history of hypertension and high levels of serum CRP,IL⁃6,TNF⁃αwere independent risk factors for epilepsy complicated with migraine(P<0.05).A nomogram prediction model was constructed and internally validated,showing a C⁃index was 0.821(95%CI 0.789⁃0.855).The areas under the ROC curve(AUC)for the modeling and validation groups were 0.833 and 0.819 respectively,with calibration curves indicating good model fit.DCA showed that net clinical benefits could be obtained when the threshold probability was 6%⁃79%and 81%⁃92%.Conclusion Female,course of disease,attack frequency,family history of migraine,poor sl

关 键 词：癫痫 偏头痛 风险因素 列线图预测模型 

分 类 号：R742.1[医药卫生—神经病学与精神病学] R747.2[医药卫生—临床医学]