Chen Zhu-Hong, Yang Guan, Zhang Chi, Su Dan, Li Yu-Ting, Shang Yu-Xuan, Zhang Wei, Wang Wen
Functional and Molecular Imaging Key Lab of Shaanxi Province, Department of Radiology, Tangdu Hospital, Air Force Medical University, Xi'an, Shaanxi, China.
Department of Medical Imaging, Gansu Corps Hospital of Chinese Armed Police Force, Lanzhou, Gansu, China.
Front Neurol. 2025 Jan 21;16:1511252. doi: 10.3389/fneur.2025.1511252. eCollection 2025.
This study aimed to develop and validate a robust predictive model for accurately identifying migraine without aura (MWoA) individuals from migraine patients.
We recruited 637 migraine patients, randomizing them into training and validation cohorts. Participant's medical data were collected such as demographic data (age, gender, self-reported headache characteristics) and clinical details including symptoms, triggers, and comorbidities. The model stability, which was developed using multivariable logistic regression, was tested by the internal validation cohort. Model efficacy was evaluated using the area under the receiver operating characteristic curve (AUC), alongside with nomogram, calibration curve, and decision curve analysis (DCA).
The study included 477 females (average age 46.62 ± 15.64) and 160 males (average age 39.78 ± 19.53). A total of 397 individuals met the criteria for MWoA. Key predictors in the regression model included patent foramen ovale (PFO) ( = 2.30, = 0.01), blurred vision ( = 0.40, = 0.001), dizziness ( = 0.16, < 0.01), and anxiety/depression ( = 0.41, = 0.02). Common symptoms like nausea ( = 0.79, = 0.43) and vomiting ( = 0.64, = 0.17) were not statistically significant predictors for MWoA. The AUC values were 79.1% and 82.8% in the training and validation cohorts, respectively, with good calibration in both.
The predictive model developed and validated in this study demonstrates significant efficacy in identifying MWoA. Our findings highlight PFO as a potential key risk factor, underscoring its importance for early prevention, screening, and diagnosis of MWoA.
本研究旨在开发并验证一种强大的预测模型，用于从偏头痛患者中准确识别无先兆偏头痛（MWoA）个体。
我们招募了637名偏头痛患者，将他们随机分为训练队列和验证队列。收集参与者的医学数据，如人口统计学数据（年龄、性别、自我报告的头痛特征）以及包括症状、触发因素和合并症在内的临床细节。使用多变量逻辑回归开发的模型稳定性通过内部验证队列进行测试。使用受试者操作特征曲线下面积（AUC）、列线图、校准曲线和决策曲线分析（DCA）评估模型效能。
该研究纳入了477名女性（平均年龄46.62±15.64）和160名男性（平均年龄39.78±19.53）。共有397名个体符合MWoA标准。回归模型中的关键预测因素包括卵圆孔未闭（PFO）（β = 2.30，P = 0.01）、视力模糊（β = 0.40，P = 0.001）、头晕（β = 0.16，P < 0.01）以及焦虑/抑郁（β = 0.41，P = 0.02）。恶心（β = 0.79，P = 0.43）和呕吐（β = 0.64，P = 0.17）等常见症状并非MWoA的统计学显著预测因素。训练队列和验证队列中的AUC值分别为79.1%和82.8%，两者校准良好。
本研究中开发并验证的预测模型在识别MWoA方面显示出显著效能。我们的研究结果突出了PFO作为潜在关键风险因素的重要性，强调了其在MWoA早期预防、筛查和诊断中的重要性。
