期刊文献+

脑卒中后肺部多药耐药菌感染危险因素及预测模型构建 被引量:3

Analysis of risk factors for pulmonary multi-drug resistant bacteria infection after stroke and the construction of prediction model
原文传递
导出
摘要 目的探讨脑卒中后肺部多药耐药菌(MDRB)感染的危险因素,构建风险预测模型。方法回顾性分析2018年3月-2020年10月在南阳南石医院就诊的脑卒中患者100例,根据患者是否发生肺部MDRB感染,分为感染组27例和未感染组73例。比较两组患者的临床资料,并将有统计学意义的影响因素纳入多因素Logistic回归分析并建立预测模型。结果有糖尿病史(OR=5.184,P=0.018)、有抗生素预防应用(OR=6.617,P=0.007)、有意识障碍(OR=6.657,P=0.010)、有侵入性操作(OR=6.911,P=0.013)以及重症加强护理病房(ICU)住院时间≥72h(OR=8.056,P=0.003)均为影响脑卒中后肺部MDRB感染的独立危险因素;建立预测模型P=1/[1+e^(-9.983+1.646×(糖尿病史)+1.890×(抗生素预防应用)+1.896×(意识障碍)+1.933×(侵入性操作)+2.086×(ICU住院时间))],Hosmer-Lemeshowχ^(2)=6.441,P=0.598,模型预测脑卒中后肺部MDRB感染的AUC为0.924,95%CI为0.865~0.983,具有良好的拟合度和预测效能。结论临床可重点关注有糖尿病史、抗生素预防应用、意识障碍、侵入性操作、ICU住院时间≥72h的患者。 OBJECTIVE To explore the risk factors for pulmonary multi-drug resistant bacteria(MDRB)infection after stroke,and construct the risk prediction model.METHODS A retrospective analysis was performed on 100stroke patients who were treated in the Nanyang Nanshi Hospital from Mar 2018to Oct 2020.According to whether pulmonary MDRB infection occured,they were divided into the infection group of 27cases and an uninfected group of 73cases.The clinical data between the two groups were compared,the influencing factors with significance being included in the multivariate Logistic regression analysis,and a predictive model was established.RESULTS Diabetes history(OR=5.184,P=0.018),prophylactic usage of antibiotics(OR=6.617,P=0.007),disturbance of consciousness(OR=6.657,P=0.010),invasive operation(OR=6.911,P=0.013)and stay time in intensive care unit(ICU)≥72h(OR=8.056,P=0.003)were all independent risk factors for pulmonary MDRB infection after stroke.The expression of the risk prediction model was as the following:P=1/[1+e^(-9.983+1.646×(diabeters history)+1.890×(prophylactic usage of antibiotics)+1.896×(disturbance of consciousances)+1.933×)invasive operation(+2.086×(atay time in ICU))],Hosmer-Lemeshowχ^(2)=6.441,P=0.598.AUC and 95%CI of the model for predicting pulmonary MDRB infection after stroke were 0.924and 0.865-0.983,respectively,showing good fit and predictive efficiency.CONCLUSION Patients with diabetes history,prophylactic usage of antibiotics,disturbance of consciousness,invasive operation and stay time in ICU≥72hneed close attentions in clinic.
作者 王蕾 邱胜利 曹俊 朱帅 李靖 WANG Lei;QIU Sheng-li;CAO Jun;ZHU Shuai;LI Jing(Nanshi Hospital of Nanyang,Nanyang,Henan 473000,China)
出处 《中华医院感染学杂志》 CAS CSCD 北大核心 2022年第7期990-993,共4页 Chinese Journal of Nosocomiology
基金 河南省科研基金资助项目(2019SD21414)。
关键词 脑卒中 多药耐药菌 危险因素 LOGISTIC回归分析 肺部感染 预测模型 Stroke Multi-drug resistant bacteria Risk factor Logistic regression analysis Pulmonary infection Predictive model
  • 相关文献

参考文献10

二级参考文献126

共引文献9573

同被引文献53

引证文献3

二级引证文献1

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部