Abstract: Background　Drug resistant epilepsy (DRE) in children is a common neurological disorder during childhood. Approximately one-third of pediatric epilepsy patients fail to achieve sustained seizure freedom despite receiving two appropriate and well-tolerated antiepileptic drugs. Long-term, recurrent seizures or status epilepticus can severely impact a child's cognition, memory, quality of life, psychosocial development, and physical growth. Additionally, DRE causes multifaceted irreversible damage to a child's neurological development. Objective　To explore risk factors contributing to DRE in children and its impact on the development of different energy regions in pediatric patients. Methods　This study enrolled 250 newly diagnosed pediatric epilepsy patients from the Department of Pediatric Neurology at the Third Affiliated Hospital of Zhengzhou University between January 2023 and October 2024. Based on drug treatment efficacy, they were categorized into a drug-resistant group (n=94) and a non-drug-resistant group (n=156). A control group of 80 typically developing children who underwent the Chinese version of the Griffiths Mental Development Scale (GDS-C) assessment at the same hospital during the same period served as the normal control group. We collected clinical data and GDS-C assessment results for pediatric patients, evaluating six domains: motor, personal and social, language, eye-hand coordination, performance, and practical reasoning. Statistical analysis was performed using SPSS 25.0 software. Multivariate Logistic stepwise regression analysis was employed to explore risk factors for DRE in children.To plot the ROC curve for predicting childhood DRE using high-risk factors and composite indicators. Results　The drug-resistant group and non-drug-resistant group differed significantly in body weight, age at onset, initial trigger for first seizure, seizure pattern changes, genetic etiology, structural etiology, unknown etiology, pre-treatment seizure duration, pre-treatment seizure frequency, status epilepticus, efficacy of first antiepileptic drug, developmental delay, history of febrile seizures, birth history, fibrinogen, blood ammonia, ceruloplasmin, vitamin D, cranial MRI, initial EEG, EEG after 6 months of treatment, multifocal epileptiform discharges, chromosomal and genetic testing. Differences were statistically significant (P<0.05). Results of the multivariate Logistic stepwise regression analysis revealed that poor efficacy of the first antiepileptic drug (OR=18.928, 95%CI=8.392-42.693, P<0.001), prolonged seizure duration prior to treatment (OR=1.089, 95%CI=1.006-1.180, P=0.036), developmental delay (OR=3.415, 95%CI=1.504-7.754, P=0.003), and and multifocal epileptiform discharges (OR=5.265, 95%CI=2.335-11.873, P<0.001) were identified as high-risk factors for DRE in children. ROC curve analysis revealed that the combined indicators achieved an area under the curve (AUC) of 0.920 (95%CI=0.887-0.953) for predicting childhood DRE, with an optimal cutoff value of 0.316, sensitivity of 0.904, and specificity of 0.801. Comparisons of developmental quotients across six functional domains among children in the control group, refractory group, and non-refractory group revealed statistically significant differences (P<0.05). Specifically, children in the non-refractory group demonstrated higher developmental quotients in motor skills, personal and social skills, language, and hand-eye coordination than those in the refractory group, but lower than those in the control group (P<0.05). Conclusion　Poor response to initial antiepileptic drugs, prolonged seizure duration prior to treatment, developmental delay, and multifocal epileptiform discharges are high risk factors for childhood DRE. The constructed combined indicator prediction model holds clinical reference value; children with DRE exhibit significant developmental lag across multiple energy region.

Key words: Epilepsy, Drug resistant epilepsy, Child, High risk factors, Griffiths mental development assessment scale (Chinese version), Energy region, Developmental quotient

摘要： 背景　儿童耐药性癫痫（DRE）是儿童期常见神经系统疾病。约1/3癫痫患儿经应用正确且能耐受的两种抗癫痫药物仍未能达到持续无发作。癫痫长期、反复发作或癫痫持续状态会对患儿的认知、记忆、生活质量、社会心理及儿童的生长发育等造成严重影响。同时DRE对儿童神经系统发育造成多方面不可逆损害。目的　探讨导致儿童DRE的高危因素及对患儿不同能区发育的影响。方法　本文选取2023年1月—2024年10月由郑州大学第三附属医院小儿神经内科新诊断的250例癫痫患儿为研究对象，根据药物治疗效果分为耐药组94例，非耐药组156例，选取同期于该院行Griffiths精神发育评估量表中文版（GDS-C）评估的正常体检儿童80例为正常对照组。收集患儿临床资料及GDS-C评估结果，评估内容包括运动、个人与社会、语言、手眼协调、表现、实际推理6个能区。运用SPSS 25.0软件进行统计学分析，采用多因素Logistic逐步回归分析探讨儿童DRE的高危因素。绘制高危因素及联合指标预测儿童DRE的ROC曲线。结果　耐药组与非耐药组患儿体质量、起病年龄、首次发病诱因、发作形式改变、遗传病因、结构病因、病因未明、治疗前发作持续时间、治疗前发作次数、癫痫持续状态、首次抗癫痫药效果、发育迟缓、热性惊厥史、出生史、纤维蛋白原、血氨、铜蓝蛋白、维生素D、颅脑MRI、首次脑电图、治疗半年后脑电图、多灶性癫痫样放电、染色体及基因检测比较，差异有统计学意义（P<0.05）。多因素Logistic逐步回归分析结果显示，首次抗癫痫药效果欠佳（OR=18.928，95%CI=8.392~42.693，P<0.001）、治疗前发作持续时间长（OR=1.089，95%CI=1.006~1.180，P=0.036）、发育迟缓（OR=3.415，95%CI=1.504~7.754，P=0.003）、多灶性癫痫样放电（OR=5.265，95%CI=2.335~11.873，P<0.001）是儿童DRE的高危因素。ROC曲线分析结果显示，联合指标预测儿童DRE的ROC曲线下面积（AUC）为0.920（95%CI=0.887~0.953），最佳截断值为0.316，灵敏度为0.904，特异度为0.801。对照组、耐药组与非耐药组3组儿童6个能区发育商比较，差异均有统计学意义（P<0.05）；其中非耐药组儿童运动、个人与社会、语言、手眼协调、表现能区发育商高于耐药组而低于对照组（P<0.05）。结论　首次抗癫痫药效果欠佳、治疗前发作持续时间长、发育迟缓、多灶性癫痫样放电是儿童DRE的高危因素。构建的联合指标预测模型对临床具有参考价值；DRE患儿多个能区的发育均明显落后。

关键词: 癫痫, 耐药性癫痫, 儿童, 高危因素, Griffiths 精神发育评估量表中文版, 能区, 发育商