冠状动脉搭桥术后医院感染风险预测模型构建

目的构建冠状动脉搭桥术(CABG)后医院感染的风险预测模型。方法选择2017年6月-2020年6月在郑州市第七人民医院接受CABG治疗的冠心病患者121例,根据术后住院期间是否发生医院感染,分为感染组34例和未感染组87例。采集感染患者临床标本进行病原菌分离和鉴定,收集患者年龄、有无糖尿病、血清白蛋白(ALB)、体外循环时间、术后引流量、引流管留置时间、术后气管插管时间和住院时间等。采用Logistic回归和卡方自动交互检测(CHAID)模型分析CABG术后医院感染的危险因素,受试者工作特征(ROC)曲线检测模型的预测效能。结果 121例患者CABG术后有34例患者发生医院感染,感染率为28.10%;术后医院感染患者共分离病原菌29株,主要为肺炎克雷伯菌(27.59%)、大肠埃希菌(20.69%)、铜绿假单胞菌(17.24%);Logistic回归分析显示,年龄>60岁、ALB<30 g/L、体外循环时间>120 min、引流管留置时间>7 d、术后气管插管时间>24 h、住院时间>30 d均为CABG术后医院感染的危险因素;CHAID模型分析显示,引流管留置时间、体外循环时间、年龄及术后气管插管时间均为CABG术后医院感染的危险因素,模型预测的准确性为72.70%(P<0.05);ROC分析显示,Logistic回归模型预测医院感染的AUC为0.808,显著高于CHAID模型预测的0.640(P<0.05)。结论 Logistic回归模型可以有效预测CABG术后医院感染的发生,CHAID模型可以显示各变量的相互关系,可与Logistic回归模型互补应用于临床风险因素分析。

Construction of nosocomial infection risk prediction model after coronary artery bypass grafting

OBJECTIVE To construct the nosocomial infection risk prediction model after coronary artery bypass grafting(CABG). METHODS A total of 121 patients with coronary heart disease(CHD) who underwent CABG in the Zhengzhou Seventh People′s Hospital from Jun 2017 to Jun 2020 were enrolled in the study and divided into the infection group with 34 cases and the non-infection group with 87 cases according to the status of nosocomial infection during the hospital stay.The clinical specimens were collected from the patients with infection, the pathogens were isolated and identified.The clinical data such as age, diabetes mellitus, serum albumin(ALB), time of extracorporeal circulation, postoperative drainage volume, indwelling time of drainage tube, postoperative tracheal intubation time and length of hospital stay were collected from the patients.The risk factors for the postoperative nosocomial infection in the CABG patients were analyzed by means of logistic regression and chi-square automatic interactive detection(CHAID) model, and the predictive efficiency of the model was detected by receiver operating characteristic(ROC) curve. RESULTS Of the 121 patients, 34 had nosocomial infection after CABG, with the infection rate 28.10%.Totally 29 strains of pathogens were isolated from the patients with postoperative nosocomial infection, 27.59% of which were Klebsiella pneumoniae, 20.69% were Escherichia coli, and 17.24% were Pseudomonas aeruginosa.Logistic regression analysis showed that the more than 60 years of age, ALB less than 30 g/L, time of extracorporeal circulation more than 120 min, drainage tube indwelling time more than 7 days, postoperative tracheal intubation time more than 24 hours and length of hospital stay more than 30 days were the risk factors for the postoperative nosocomial infection in the CABG patients.CHAID model analysis indicated that the drainage tube indwelling time, time of extracorporeal circulation, age and postoperative tracheal intubation time were the risk factors for the postoperative nosocomial infection in the CABG patients, and the accuracy of the model was 72.70% in prediction of the infection(P<0.05).ROC analysis showed that the AUC of the logistic regression model was 0.808 in prediction of nosocomial infection, significantly higher than 0.640 of the CHAID model(P<0.05). CONCLUSION Logistic regression model can effective predict the postoperative nosocomial infection in the CABG patients, CHAID model can show the interrelations among the variables, which can be used as a complement of the logistic regression model in analysis of clinical risk factors.