开颅手术后颅内感染风险预测模型构建及效果评价

Construction and effect evaluation of risk prediction model for intracranial infection after craniotomy

摘要:

目的 运用logistic回归分析构建神经外科开颅手术后颅内感染风险预测模型并进行效果评价。方法选取某院神经外科2019年1月-2021年6月行开颅手术的患者为研究对象，根据术后是否发生颅内感染分为病例组和对照组，采用logistic回归分析开颅手术后颅内感染发生的危险因素并构建风险预测模型，通过Hosmer-Lemeshow拟合优度检验和受试者工作特征(ROC)曲线对其效果进行综合评价。 结果 共纳入778例开颅手术患者，121例发生术后颅内感染，发病率为15.55%；logistic多因素回归分析结果显示，幕下手术、脑室引流时间≥3 d、使用明胶海绵≥3片、出血量≥300 mL、切口脑脊液漏是开颅手术后颅内感染的独立危险因素(均P < 0.05)；开颅手术后颅内感染的风险预测模型为：logit(P)=5.408+0.833×(幕下手术)+0.083×(脑室引流时间)+1.059×(使用明胶海绵)+0.456×(出血量)+2.821×(切口脑脊液漏)；Hosmer-Lemeshow拟合优度检验结果显示颅内感染的预测概率和实际发病率比较，差异无统计学意义(P=0.768)；logistic回归风险预测模型验证准确率为86.00%，ROC曲线下面积为0.847，95%CI为0.814~0.878。 结论 幕下手术、脑室引流时间≥3 d、使用明胶海绵≥3片、出血量≥300 mL、切口脑脊液漏是神经外科开颅手术后颅内感染的独立危险因素，运用logistic回归分析构建的风险预测模型对术后颅内感染的预测效果较好。

Abstract:

Objective To construct the risk prediction model for intracranial infection (ICI) after craniotomy in neurosurgery department by logistic regression analysis, and evaluate its effect. Methods Patients who underwent craniotomy from January 2019 to June 2021 in the neurosurgery department of a hospital were selected as research objects, according to whether ICI occurred after the operation, they were divided into case group and control group, logistic regression analysis was used to analyze the risk factors for ICI after craniotomy, the risk prediction model was constructed, and the effect was comprehensively evaluated by means of Hosmer-Lemeshow goodness-of-fit test and receiver operating characteristic (ROC) curve. Results A total of 778 patients undergoing craniotomy were included, 121 of whom had post-operative ICI, incidence was 15.55%; logistic multivariate regression analysis showed that subtentorial surgery, ventricular drainage time ≥3 days, use of gelatin sponge ≥3 pieces, bleeding volume ≥300 mL, cerebrospinal fluid leakage of incision were independent risk factors for ICI after craniotomy (all P < 0.05); the risk prediction model of ICI after craniotomy was logit(P)=5.408+0.833×(subtentorial surgery)+0.083×(ventricular drainage time)+1.059×(use gelatin sponge)+0.456×(bleeding volume)+2.821×(incision cerebrospinal fluid leakage); Hosmer-Lemeshow goodness-of-fit test showed that there was no significant difference in the predicted probability and the actual incidence of intracranial infection (P=0.768); the validation accuracy of logistic regression risk prediction model was 86.00%, the area under ROC curve was 0.847, and 95%CI was 0.814-0.878. Conclusion Subtentorial operation, ventricular drainage time ≥3 days, use of gelatin sponge ≥3 pieces, bleeding volumn ≥300 mL and cerebrospinal fluid leakage of incision are independent risk factors for ICI after neurosurgical craniotomy, the risk prediction model constructed by logistic regression analysis has a good prediction effect on post-operative ICI.