神经外科重症患者多药耐药菌感染危险因素及其预测模型构建

Risk factors for multidrug-resistant organisms infections in critically ill patients of neurosurgery department and establishment of prediction model

目的分析神经外科重症患者多药耐药菌感染病原菌分布特点,并构建多药耐药菌感染预测模型.方法回顾性收集江南大学附属医院2020年1月-2022年6月收治的204例神经外科重症患者临床资料,根据患者是否多药耐药菌感染分为多药耐药菌感染组71例和非多重耐药菌感染组133例,统计神经外科重症患者检出多药耐药菌分布,多因素Logistic回归分析多药耐药菌感染危险因素,并构建预测模型,受试者工作特征(ROC)曲线分析模型的预测价值.结果神经外科重症患者检出的多药耐药菌主要为鲍氏不动杆菌、铜绿假单胞菌和金黄色葡萄球菌;意识障碍、泌尿道插管、中心静脉插管、低蛋白血症、机械通气时间、抗菌药物使用种类3种及以上、抗菌药物使用时间较长均为神经外科重症患者多药耐药菌感染的独立危险因素(P<0.05);ROC曲线分析结果显示,通过上述危险因素构建的模型预测神经外科重症患者多药耐药菌感染的曲线下面积(AUC)为0.829,敏感度为66.20%,特异度为85.71%.结论神经外科重症患者感染的多药耐药菌主要为鲍氏不动杆菌、铜绿假单胞菌、金黄色葡萄球菌,多药耐药菌感染危险因素较多,据此构建的模型预测多药耐药菌感染具有较高的预测价值.

OBJECTIVE To observe the distribution of pathogens isolated from the critically ill patients with multidrug-resistant organisms(MDROs)infection in neurosurgery department and establish the prediction model for MDROs infections.METHODS The clinical data were retrospectively collected from 204 critically ill patients of neurosurgery department who were treated in the Affiliated Hospital of Jiangnan University from Jan 2020 to Jun 2022.The patients were divided into the MDROs infection group with 71 cases and the non-MDROs infection group with 133 cases according to the status of MDROs infection.The distribution of MDROs isolated from the critically ill patients was statistically analyzed,multivariate logistic regression analysis was performed for the risk factors for the MDROs infection,the prediction model was established,and the predictive value of the model was analyzed by means of receiver operating characteristic(ROC)curve.RESULTS Acinetobacter baumanni,Pseudomonas aeruginosa and Staphylococcus aureus were dominant among the MDROs isolated from the critically ill patients of neurosurgery department.Disturbance of consciousness,urinary tract intubation,central venous cathe-terization,hypoproteinemia,duration of mechanical ventilation,use of no less than 3 types of antibiotics and long time use of antibiotics were the independent risk factors for the MDROs infection in the critically ill patients of neurosurgery department(P<0.05).The result of ROC curve analysis showed that area under curve(AUC)of the model that was established based on the above risk factors was O.829 in prediction of the MDROs infection,with the sensitivity 66.20%,the specificity 85.71%.CONCLUSION A.baumannii,P.aeruginosa and S.aureus are the predominant species of MDROs isolated from the critically ill patients of neurosurgery department.There are a variety of risk factors for the MDROs infection.The model that is established based on the risk factors has high value in prediction of the MDROs infection.