摘要
目的探讨基于主成分分析的回归模型对重度创伤性脑损伤去骨瓣术后脑积水的评估价值。方法选取邯郸市中心医院2022-01—2023-10收治的92例重度创伤性脑损伤患者,均接受去骨瓣术治疗。根据患者术后是否发生脑积水分为发生组(20例)及未发生组(72例),比较2组临床资料,并对其进行主成分分析,然后提取主成分作为自变量,采用主成分分析-Logistic回归分析重度创伤性脑损伤去骨瓣术后脑积水的危险因素,采用受试者工作特征曲线评估模型对患者发生脑积水的预测效能。结果单因素分析显示,发生组脑震荡、脑内血肿、术后颅内感染、术后大面积脑梗死、手术前后脑室系统出血占比、受伤距离手术时间、手术前后CGS评分、白细胞计数均高于未发生组(P<0.05)。主成分分析-Logistic回归分析显示,合并脑内血肿、颅内感染、术后大面积脑梗死、手术前后脑室系统出血、手术前后GCS评分低是重度创伤性脑损伤去骨瓣术后发生脑积水的独立危险因素(P<0.05)。主成分分析-Logistic回归模型评估重度创伤性脑损伤去骨瓣术后发生脑积水的灵敏度为90.00%,特异度为94.44%,AUC为0.958。结论主成分分析-Logistic回归模型能够有效评估重度创伤性脑损伤去骨瓣术患者的预后情况,脑内血肿、颅内感染、术后大面积脑梗死、脑室系统出血、GCS评分等是影响患者发生脑积水的重要因素。
Objective To investigate the value of regression model based on principal component analysis in evaluating hydrocephalus after bone flap removal for severe traumatic brain injury.Methods Totally 92 patients with severe traumatic brain injury treated in the Handan Central Hospital from January 2022 to October 2023 were selected,all of whom were treated by bone flap removal.According to whether the patients developed hydrocephalus after operation,they were divided into the occurrence group(20 cases)and the non-occurrence group(72 cases).The clinical data of the two groups were compared,and principal component analysis was performed on them.After principal component analysis,principal component was extracted as the independent variable,and the risk factors for hydrocephalus after bone flap removal for severe traumatic brain injury were analyzed by principal component analysation-Logistic regression.The predictive efficacy of the model for hydrocephalus was evaluated by receiver operating characteristic curve.Results Univariate analysis showed that the proportion of concussion,intracranial hematoma,postoperative intracranial infection,massive cerebral infarction after operation,ventricular hemorrhage after operation,distance from injury to operation,CGS score before and after operation,and white blood cell count were all higher than those in the non-occurrence group(P<0.05).Principal component analysis-Logistic regression analysis showed that intracerebral hematoma,intracranial infection,massive cerebral infarction after operation,ventricular hemorrhage before and after surgery,and low GCS score before and after surgery were independent risk factors for hydrocephalus after bone flap removal for severe traumatic brain injury(P<0.05).The sensitivity,specificity and AUC of principal component analysis-Logistic regression model were 90.00%,94.44%and 0.958,respectively in evaluating the incidence of hydrocephalus after bone flap removal for severe traumatic brain injury.Conclusion Principal component analysis-Logistic regression model can effectively evaluate the prognosis of patients with severe traumatic brain injury after bone flap removal.Intracerebral hematoma,intracranial infection,massive cerebral infarction after operation,ventricular system hemorrhage,and GCS score are important factors affecting the occurrence of hydrocephalus in patients.
作者
晁艳艳
张钧
李海红
王月然
CHAO Yanyan;ZHANG Jun;LI Haihong;WANG Yueran(Handan Central Hospital,Handan 056008,China)
出处
《中国实用神经疾病杂志》
2024年第10期1293-1297,共5页
Chinese Journal of Practical Nervous Diseases
基金
邯郸市科学技术研究与发展计划项目(编号:19422083009-3)。
