期刊文献+

耐碳青霉烯类肺炎克雷伯菌感染风险预测模型的构建 被引量:2

Construction of risk prediction model for carbapenem-resistant Klebsiella pneumoniae infection
下载PDF
导出
摘要 目的 探讨耐碳青霉烯类肺炎克雷伯菌(CRKP)感染的危险因素及应用Rstudio统计学软件建立列线图预测模型,并进行验证。方法 回顾性分析2021年1月至2022年4月入住襄阳市第一人民医院发生肺炎克雷伯菌感染的494例患者的临床资料,在二元Logistic回归分析的基础上建立风险预测模型,并通过Rstudio统计软件建立列线图模型,采用受试者工作特征(ROC)曲线、校准曲线、临床决策曲线评价模型并通过验证组验证模型。结果 抗菌药物使用时间≥4 d、机械通气≥96 h、住院时间>30 d、入院时已经存在肺部感染以及碳青霉烯类药物的使用均是住院患者发生CRKP感染的独立危险因素(P<0.05)。CRKP风险评估模型在建模组和验证组数据中ROC曲线的曲线下面积(AUC)分别为0.775(95%CI:0.720~0.830)和0.766(95%CI:0.710~0.822)。校准曲线显示该模型具有较好的校准度。临床决策曲线显示该模型有较好的临床效用。结论 基于上述5个危险因素构建的风险预测模型能较为准确地预测住院患者CRKP感染风险,可对CRKP感染高危人群进行有效识别。 Objective To explore the risk factors of carbapenem-resistant Klebsiella pneumoniae(CRKP) infection,and to construct a nomogram prediction model with the Rstudio statistical software and conduct the verification.Methods The clinical data of 494 inpatients with Klebsiella pneumoniae infection treated in Xiangyang Municipal First People′s Hospital from January 2021 to April 2022 were retrospectively analyzed.The nomogram predictive model was built by the Rstudio statistical software on the basis of binary Logistic regression analysis.The receiver operating characteristic(ROC) curve,calibration curve and clinical decision curve were adopted to evaluate the model and the model was verified by the validation group.Results The antimicrobial drugs use time ≥4 d,mechanical ventilation ≥96 h,hospitalization >30 d,lung infection existence at admission and carbapenem drugs use were the independent risk factors for CRKP infection occurrence.The area under the curve(AUC) of ROC curve of the risk assessment model in the data of the model construction group and validation group were 0.775(95%CI:0.720-0.830)and 0.766(95%CI:0.710-0.822) repsectively.The calibration curve showed that the model has a better calibration degree and the clinical decision curve showed the good clinical utility of the model.Conclusion The risk prediction model constructed based on the above mentioned five risk factors could accurately predict the risk of CRKP in the inpatients and could effectively identify the people with high risk of CRKP infection.
作者 熊琴 王震宇 宋涛 XIONG Qin;WANG Zhenyu;SONG Tao(School of Medicine,Wuhan University of Science and Technology,Wuhan,Hubei 430065,China;Department of Clinical Laboratory,Xiangyang Municipal First People′s Hospital,Xiangyang,Hubei 441000,China)
出处 《检验医学与临床》 CAS 2023年第6期771-775,共5页 Laboratory Medicine and Clinic
关键词 肺炎克雷伯菌 碳青霉烯类耐药 危险因素 预测模型 列线图 Klebsiella pneumoniae carbapenem-resistant risk factor predictive model nomogram
  • 引文网络
  • 相关文献

参考文献6

二级参考文献33

共引文献420

同被引文献18

引证文献2

二级引证文献2

相关主题

;
使用帮助 返回顶部