Abstract: Objective To investigate the risk factors that affect secondary epilepsy in children with brain injury,construct a nomogram model for predicting secondary epilepsy in patients with brain injury,and evaluate the distinction and consistency of the model. Methods A total of 388 children with craniocerebral injury admitted to our hospital from December 2015 to October 2020 were selected as the research objects.And according to whether secondary epilepsy occurred within 6 months after operation,they were divided into 52 cases in the secondary epilepsy group and 336 cases in the non-secondary epilepsy group.The clinical data of patients were collected,and the single factor and multivariate logistic regression analyses were used to screen the risk factors that affected the occurrence of secondary epilepsy in children with craniocerebral injury;R software was used to construct a nomogram model for predicting secondary epilepsy in children with head injury,and ROC curve and calibration curve were used to verify the discrimination and consistency of the nomogram model. Results The incidence rate of secondary epilepsy in children with craniocerebral injury was 13.40% (52/388).There were statistically significant differences in the type of trauma,degree of trauma,injury site,skull fracture,brain contusion and laceration,and traumatic intracerebral hematoma between the secondary epilepsy group and the non-secondary epilepsy group (P＜0.05).Multivariate Logistic regression analysis showed that the injury site of temporal lobe and frontal lobe,brain contusion and laceration,and traumatic intracerebral hematoma were risk factors affecting secondary epilepsy in children with craniocerebral injury (P＜0.05).The constructed nomogram prediction model had good discrimination (the area under the ROC curve was 0.768 (95%CI=0.701~0.835)) and consistency (Hosmer-Lemeshow goodness of fit testχ²=7.677,P=0.466). Conclusion The nomogram model constructed in this study to predict the risk of secondary epilepsy in children with craniocerebral injury has good discrimination and consistency.

Key words: C raniocerebral injury, Secondary epilepsy, Nomogram mode

摘要： 目的 探讨影响颅脑损伤患儿继发性癫痫的危险因素,构建预测颅脑损伤患者继发性癫痫的列线图模型,并评估模型的区分度和一致性。方法 选取2015年12月~2020年10月本院收治的颅脑损伤患儿388例作为研究对象。并根据术后6个月内是否出现继发性癫痫分为继发性癫痫组52例和未继发性癫痫组336例。收集患者临床资料,采用单因素及多因素Logistic回归分析筛选影响颅脑损伤患儿发生继发性癫痫的危险因素;采用R软件构建预测颅脑损伤患儿继发性癫痫的列线图模型,并使用ROC曲线及校准曲线验证列线图模型的区分度和一致性。结果 颅脑损伤患儿继发性癫痫发生率为13.40%(52/388)。继发性癫痫组、未继发性癫痫组患者外伤类型、外伤程度、受伤部位、颅骨骨折、脑挫裂伤、外伤性脑内血肿比较差异有统计学意义(P＜0.05)。多因素Logistic回归分析结果显示,颞叶及额叶受伤部位、脑挫裂伤、外伤性脑内血肿是影响颅脑损伤患儿继发性癫痫的危险因素(P＜0.05)。构建的列线图预测模型具有较好的区分度(ROC曲线下面积为0.768(95%CI=0.701~0.835))和一致性(Hosmer-Lemeshow拟合优度检验χ²=7.677,P=0.466)。结论 本研究构建的预测颅脑损伤患儿继发性癫痫发生风险的列线图模型,具有较好的区分度和一致性。

关键词: 颅脑损伤, 继发性癫痫, 列线图模型