ICU患者医院感染危险因素及列线图模型建立

Risk factors for nosocomial infections in ICU patients and establishment of nomogram model

目的 探索贵州省三家综合医院重症监护病房(ICU)患者医院感染危险因素,并构建可视化列线图模型。方法 选取贵州省人民医院、贵州省第二人民医院和贵州茅台医院三所医院,综合ICU病房2022年5月24日2023年5月24日收治的所有患者共850例为研究对象,其中发生医院感染266例,未发生医院感染584例。采用二分类Logistic回归模型识别患者发生医院感染的危险因素,基于这些危险因素构建临床可视化列线图模型,并采用受试者工作特征(ROC)曲线验证模型的区分性能。结果 共850例患者中有266例发生医院感染(31.3%);医院感染的主要危险因素是:男性、住院时间长、血常规送检多次异常、降钙素原送检多次异常、入院科室为非ICU、多重耐药菌检出次数多和长期使用呼吸机;基于以上风险因素构建的列线图模型能很直观的计算出ICU患者未来可能发生医院感染的风险概率;且根据上述因素构建的列线图模型具有较好的区分价值,曲线下面积为0.872。结论 基于危险因素构建的贵州省ICU患者医院感染列线图模型,能量化医院感染危险因素,有利于更好地采取相应措施,进一步减少医院感染发生。

OBJECTIVE To explore the risk factors for nosocomial infection in the patients of intensive care units(ICUs)of three general hospitals in Guizhou Province and establish a visualized nomogram model.METHODS A total of 850 patients who were treated in comprehensive ICUs of Guizhou Provincial People's Hospital,Guizhou Provincial Second People's Hospital and Keichow Moutai Hospital from May 24,2022 to May 24,2023 were re-cruited as the research subjects,266 of whom had nosocomial infection,and 584 did not have nosocomial infection.The risk factors for the nosocomial infection were identified by using binary logistic regression model,the visual-ized nomogram model was established based on the risk factors,and the differentiating performance of the model was validated by means of receiver operating characteristic(ROC)curves.RESULTS Among the 850 patients,266(31.3%)had nosocomial infection.The major risk factors for the nosocomial infection included the male,long length of hospital stay,many time of abnormal submission of blood routine test,many times of abnormal submis-sion of procalcitonin,stay of non-ICU departments,many times of isolation of multidrug-resistant organisms and long-term use of ventilator.The risk probability of nosocomial infection in the ICU patients could be directly calcu-lated by the nomogram model that was established based on the risk factors,and the model has significant differ-entiating value,with the area under curve 0.872.CONCLUSION The nomogram model that is established based on the risk factors for the nosocomial infection in the ICU patients of Guizhou Province can quantify the risk factors so as to take more effective coping measures and further reduce the incidence of nosocomial infection.