重症监护室病人多重耐药菌感染预测模型的构建  

Construction of a prediction model for multidrug-resistant organism infection in intensive care unit patients

作　　者：王亚喜 张玲慧 曹文[2] 辛维青[3] 王忠心[3] 庞旭峰[3] WANG Yaxi;ZHANG Linghui;CAO Wen;XIN Weiqing;WANG Zhongxin;PANG Xufeng(Department of Critical Medicine,The Affilialed Hospital of Qingdao University,Qingdao 266021,China)

机构地区：[1]青岛大学附属医院重症医学科,山东青岛266021 [2]青岛大学附属医院病案室,山东青岛266021 [3]青岛大学附属医院护理院感部,山东青岛266021

出　　处：《青岛大学学报（医学版）》2024年第1期115-119,共5页Journal of Qingdao University(Medical Sciences)

基　　金：青岛市民生科技计划项目(19-6-1-10-nsh)。

摘　　要：目的构建及验证重症监护室(ICU)病人多重耐药菌(MDRO)感染预测模型。方法回顾性分析2021年8月—2022年1月入住青岛市某三甲医院ICU 688例病人的临床资料,根据留置ICU期间是否发生MDRO分为MDRO感染组和非MDRO感染组,基于最小绝对收缩与选择算子算法和逐步回归筛选危险因素、构建MDRO感染预测模型,绘制列线图,并进行内部验证。结果在纳入的ICU病人中,MDRO感染率为15.70%,危险因素包括住院时间、留置ICU时间、长期卧床、入ICU前使用抗生素、入ICU前进行侵入性操作、合并慢性肺部疾病、低蛋白血症、急性生理和慢性健康评分系统评分、抗生素的使用数量。基于此构建的预测模型受试者工作特征曲线下面积为0.896,特异度、灵敏度、准确度分别为76.85%、87.41%、85.76%。Hosmer-Lemeshow拟合优度检验P=0.659,模型的校准度良好。结论构建的ICU病人MDRO感染预测模型的预测效果较好,可为临床筛选MDRO感染高危人群提供参考。Objective To construct and validate a prediction model of multidrug-resistant organism(MDRO)infection in intensive care unit(ICU)patients.Methods We retrospectively analyzed the clinical data of 688 patients admitted to the ICU of a grade A tertiary hospital in Qingdao from August 2021 to January 2022.According to whether MDRO infection occurred or not during stay in the ICU,they were divided into MDRO infection group and non-MDRO infection group.The risk factors were determined using the least absolute shrinkage and selection operator method and stepwise regression analysis to construct an MDRO infection prediction model.A nomogram was created and internally verified.Results The MDRO infection rate was 15.70%among the included ICU patients.The risk factors included the length of hospital stay,the length of ICU stay,long-term bed rest,the use of antibiotics before ICU,invasive operation before ICU,chronic lung disease,hypoproteinemia,Acute Physiology and Chronic Health Evaluation score,and the consumption of antibiotics.The area under the receiver operating characteristic curve for the constructed prediction model was 0.896,with the specificity,sensitivity,and accuracy being 76.85%,87.41%,and 85.76%,respectively.The Hosmer-Lemeshow test demonstrated a good fit of the model(P=0.659).Conclusion The MDRO infection prediction model for ICU patients has good performance,which can provide a reference for identifying patients at high risk for MDRO infection.

关 键 词：重症监护病房 抗药性 多种 细菌 危险因素 列线图 

分 类 号：R197.323[医药卫生—卫生事业管理]