脑室出血患者脑室外引流术后颅内感染的风险预测模型构建与验证OA

Objective To establish a risk prediction model for intracranial infection after external ventricular drainage(EVD)in patients with intraventricular hemorrhage,so as to provide a reference for preventing postoperative intracranial infection.Methods Using the convenient sampling method,a total of 215 patients with ventricular hemorrhage who under-went EVD in the Department of Neurosurgery,Yichun People's Hospital from December 2019 to December 2024 were selected as the research objects.According to the order of surgery,the patients were divided into a modeling set(150 cases)and a validation set(65 cases),and the patients were divided into the intracranial infection group and the non-intracranial infection group according to whether they had intracranial infection.The general data of the patients were collected.Univariate analysis and multivariate logistic regression analysis were used to screen the risk factors of intracranial infection after EVD in patients with intraventricular hemorrhage,and a prediction model was established.ROC analysis and Hosmer-Lemeshow test were used to evaluate the discrimi-nation ability and calibration degree of the prediction model,and the overall accuracy of the prediction effect of the model was calculated.Results Among the 150 patients in the modeling set,18 patients(12.00%)had intracranial infection after operation and were included in the intracranial infection group.The 132 patients without intracranial infection were included in the non-intracranial infection group.The results of univariate analysis showed that there were statistically significant differences in Glasgow coma scale(GCS)score,preoperative use of antibiotics,subcutaneous tunnel length,cerebrospinal fluid leakage,frequency of cerebrospinal fluid sampling,catheter indwelling time,catheterization method,and ICU length of stay between the two groups(P＜0.05).The results of multivariate logistic regression analysis showed that the length of subcutaneous tunnel,the frequency of cerebrospinal fluid sampling,cerebrospinal fluid leakage,and catheter indwelling time were independent influencing factors for intracranial infection after EVD in patients with ventricular hemorrhage(P＜0.05),and the prediction model was constructed based on the above four factors.ROC analysis results showed that the AUC of the model in the modeling set was 0.975(95%CI:0.946-1.000),the Youden index was 0.764,the sensitivity was 0.951,and the specificity was 0.813.Hosmer-Lemeshow test results showed P=0.811.The validation set was used to verify the model,and the accuracy of the actual application was 95.38%.Conclusion The model established by sub-cutaneous tunnel length,cerebrospinal fluid sampling frequency,cerebrospinal fluid leakage and catheter indwelling time has a good effect on predicting the risk of intracranial infection after EVD in patients with ventricular hemorrhage,which is helpful for medical staff to take timely intervention measures.