预测重症肺炎患者发生鲍曼不动杆菌多重耐药的风险列线图模型建立  

Study on the risk of Acinetobacter baumannii multi-drug resistance and the establishment of a Nomogram model for patients with severe pneumonia

作　　者：王鹏[1] 宋秋鸣[1] Wang Peng;Song Qiu-ming(Hefei Third Clinical School of Medicine,Anhui Medical University(Hefei Third People's Hospital),Hefei 230022)

机构地区：[1]安徽医科大学合肥第三临床医学院(合肥市第三人民医院),合肥230022

出　　处：《中国抗生素杂志》2022年第4期399-404,共6页Chinese Journal of Antibiotics

摘　　要：目的探讨重症肺炎患者发生鲍曼不动杆菌多重耐药(Acinetobacter baumannii multi-drug resistance,ABMR)的风险列线图模型建立。方法选取2015年5月至2020年5月我院ICU收治的鲍曼不动杆菌阳性的重症肺炎150例,根据是否发生鲍曼不动杆菌多重耐药分为发生组和未发生组。调取所有患者一般资料,采用Logistic多因素回归分析筛选出影响重症肺炎患者发生ABMR的危险因素,利用R软件建立相应的列线图预测模型,并对模型的预测性与准确度进行验证。结果Logistic回归分析结果显示,年龄、机械通气、APACHEⅡ评分、抗生素种类、ICU时间、碳青霉烯类用药史为重症肺炎患者发生ABMR的危险因素(P<0.05)。基于以上6项指标建立的重症肺炎患者发生ABMR列线图预测模型,模型的验证结果显示,该模型的一致性指数(C-index)为0.863,受试者工作特征曲线下面积显示AUC为0.831(95%CI 0.769~0.893),提示该列线图模型具有精准的预测能力。结论对于重症肺炎患者需要及时考虑年龄、机械通气、APACHEⅡ评分、抗生素种类、ICU时间、碳青霉烯类用药史等因素综合评估重症肺炎患者发生ABMR的发生率,以提高重症肺炎患者发生ABMR的诊断效能,具有较高的临床价值。Objective To explore the establishment of a Nomogram model for the risk of Acinetobacter baumannii resistance(ABMR)in patients with severe pneumonia.Methods 150 cases of severe pneumonia admitted to ICU in our hospital from May 2015 to May 2020 were selected and divided into the occurrence group and the nonoccurrence group according to the occurrence of multidrug resistance of Acinetobacter baumannii.The general data of all patients were collected,and the risk factors affecting the occurrence of ABMR in patients with severe pneumonia were screened out by Logistic multivariate regression analysis.R software was used to establish the corresponding prediction model of the training line,and the prediction and accuracy of the model were verified.Results According to the results of the Logistic regression analysis,age,mechanical ventilation,the APACHEⅡ score,antibiotics types,ICU time,and carbapenems drug history were risk factors for ABMR in patients with severe pneumonia(P<0.05).Based on the above six indicators in patients with severe pneumonia,ABMR Nomogram prediction model was established.The verification results of the model show that the consistency index(C-index)of the model is 0.863,and the area under the subject's working characteristics curve shows that the AUC is 0.831(95%CI 0.769~0.893),indicating that the nomogram model has accurate prediction ability.Conclusion For patients with severe pneumonia,we need to consider their age,mechanical ventilation,the APACHEⅡ score,types of antibiotics,ICU time,carbon penicillium alkene medication history as comprehensive evaluation to predict the incidence of ABMR,in order to improve the diagnosis efficiency of the drug resistance in patients with severe pneumonia Acinetobacter baumannii,which has high clinical value.

关 键 词：重症肺炎 鲍曼不动杆菌 多重耐药 列线图 风险因素 

分 类 号：R978.1[医药卫生—药品]