摘要
目的建立原发性颅脑恶性肿瘤患者术后颅内感染预测模型,并分析颅内感染对患者预后的影响。方法回顾性分析2015年1月-2022年12月临汾市人民医院神经外科收治的1352例原发性颅脑恶性肿瘤手术患者的临床资料,根据患者术后是否发生颅内感染,将患者分为颅内感染组(n=52)和对照组(n=1300),比较两组患者临床特征,同时分析原发性颅脑恶性肿瘤患者术后颅内感染的危险因素,根据相关危险因素,建立Nomogram预测模型,同时分析颅内感染组预后情况。结果多因素Logistic回归分析显示,糖尿病、开颅手术、手术时间、术后脑脊液漏、术后脑出血是原发性颅脑恶性肿瘤患者术后颅内感染的独立影响因素。将数据集随机分为训练集和验证集,将糖尿病、开颅手术、手术时间、术后脑脊液漏、术后脑出血纳入预测模型,绘制列线图、临床决策曲线、校准曲线和受试者工作特征(ROC)曲线,训练集ROC曲线下面积(AUC)为0.849(95%CI=0.763~0.934),验证集ROC的AUC为0.838(95%CI=0.732~0.943),在验证集里对模型进行拟合优度检验(χ2=14.399,P=0.072),这说明本模型具有良好的可信度。与对照组比较,颅内感染组患者院内死亡率显著增高(分别为15.38%和1.54%,P<0.001)。结论本研究建立的预测模型可以精准识别原发性颅脑恶性肿瘤患者术后颅内感染的高危人群。
Objective To establish a predictive model for postoperative intracranial infection in patients with brain tumors and analyze the impact of intracranial infection on the prognosis of patients.Methods The clinical data of 1352 patients with intracranial tumor surgery admitted to Linfen People s Hospital from January 2015 to December 2022 were analyzed retrospectively.According to whether the patients developed intracranial infection after surgery or not,they were divided into intracranial infection group(n=52)and control group(n=1300).The clinical characteristics of the two groups of patients were compared,and the risk factors for postoperatively intracranial infection in patients with craniocerebral tumor were analyzed.Based on the relevant risk factors,a Nomogram prediction model was established.Meanwhile,the prognosis of the intracranial infection group was analyzed.Results Multivariate logistic regression analysis showed that diabetes,craniotomy,operation time,postoperative cerebrospinal fluid leakage and postoperative cerebral hemorrhage were independent influencing factors of postoperative intracranial infection in patients with brain tumors.The data set was randomly divided into a training set and a verification set.Diabetes,craniotomy,operation time,postoperative cerebrospinal fluid leakage and postoperative cerebral hemorrhage were included in the prediction model.Nomogram,clinical decision curves,calibration curves and receiver operating characteristic(ROC)curves were drawn.The area under curve(AUC)of ROC in the training set was 0.849(95%CI=0.763-0.934),and the AUC of ROC in the validation set was 0.838(95%CI=0.732-0.943).In the validation set,the model was subjected to the Hosmer-Lemeshow Goodness-of-Fit Test,with a chi square value of 14.399 and a P value of 0.072,which indicated that this model had good reliability.Compared with the control group,the hospital mortality rate in the intracranial infection group was significantly higher(15.38%vs 1.54%,P<0.001).Conclusions The prediction model established in this study can accurately identify patients at high risk of intracranial infection in patients with intracranial tumors after surgery.
作者
杜临强
梁建荣
皇甫斌
DU Linqiang;LIANG Jianrong;HUANG Fubin(Department of Neurosurgery,Linfen People's Hospital,Linfen 014000,China)
出处
《临床神经外科杂志》
2024年第5期568-571,576,共5页
Journal of Clinical Neurosurgery
基金
山西省基础研究计划项目(202103021224386)。
关键词
颅脑恶性肿瘤
颅内感染
预测模型
预后
malignant brain tumor
intracranial infection
prediction model
prognosis
作者简介
通信作者:梁建荣。
