DOI:
江苏卫生事业管理:2023,Vol.>>Issue(11):1537-1541
本文二维码信息
码上扫一扫!
老年TBI患者医院感染危险因素探讨与预测模型构建△
徐幼桥,钱 静*,练 敏,陆 婷,罗琳娜,徐翠平
(南京大学医学院附属鼓楼医院感染管理办公室;江苏省军区南京第二十六离职干部休养所门诊部)
Risk Factors and Predicting Modelof Nosocomial Infectionin Elderly Patients with Traumatic Brain Injury△
XU Youqiao,QIAN Jing*,LIAN Min,LU Ting,LUO Linna,XU Cuiping
(Department of Infection Management,Nanjing Drum Tower Hospital,The Affiliated Hospital of Nanjing University Medical School;Department of Outpatient,Nanjing 26th Retired Cadre Rest Center,Jiangsu Military Region)
摘要
图/表
参考文献
相似文献
本文已被:浏览 889次   下载 602
    
中文摘要: 目的:探讨老年颅脑创伤(TBI)患者发生医院感染的危险因素,建立并验证预测模型。方法:纳入南京大学医学院附属鼓楼医院神经外科2019年6月至2022年6月开颅手术治疗的270例老年TBI患者,按8:2比例分为模型组(n=216)和验证组(n=54)。回顾性分析模型组患者在院期间临床资料,单因素及多因素分析医院感染危险因素,构建感染预测模型,并基于验证组数据检验其预测效能。结果:270例患者发生医院感染59例,占比21.85%。模型组单因素分析显示,年龄、TBI严重程度、低蛋白血症、气管切开、机械通气、卧床时间和留置尿管时间与医院感染相关。Logistic多因素回归分析显示,年龄>75岁、重型TBI、气管切开、机械通气、卧床时间>12天和留置尿管时间>7天是老年TBI患者医院感染独立危险因素(P < 0.05)。进一步构建感染预测模型:Z=-7.341+1.955×年龄+2.665×TBI严重程度+1.616×气管切开+1.833×机械通气+1.610×卧床时间+1.392×留置尿管时间。模型受试者工作特征曲线下面积为0.938(95%CI0.901~0.974),Hosmer-Lemeshow检验χ2=9.400(P > 0.05)。最大Youden指数为0.740,对应最佳截断值为0.502,敏感度为0.822,特异性为0.941。代入验证组数据显示模型预测准确率为0.944,敏感度为0.833,特异性为0.976。结论:高龄、重型TBI、气管切开、机械通气、长期卧床和长期留置尿管增加老年TBI患者医院感染风险,基于此构建的模型预测效能和拟合度良好,可为潜在感染患者的早期发现和治疗提供依据。
Abstract:Objective:To investigate the risk factors ofnosocomial infectioninelderly patientswithtraumatic brain injury(TBI)and establish apredictingmodelfollowed by clinicalvalidation. Methods:A total of 270 elderly TBIpatients whoaccepted neurosurgicaloperation in Nanjing Drum TowerHospitalfrom June 2019 to June 2022 were involvedin thisstudy. Thepatients were divided into model group(n=216)and validation group(n=54)at a ratio of 8:2. The characteristicsofuninfectedandinfected patientsinmodel group were comparedusingunivariate analysis,andthe significantfactors associated with nosocomial infectionwere evaluatedbylogistic multivariate regression analysis. Then,thepredictivemodel fornosocomial infection in elderlyTBIpatients was constructed based onthe above statistical analysis,thepredicting powerwasassessedbyreceiver operating characteristic(ROC)curveand wereverified invalidation group. Results:59 out of 270 patients developed nosocomial infection(21.85%). Univariate analysis showed that age,TBI severity,hypoalbuminemia,tracheotomy,mechanical ventilation,lying-bed period,and time ofindwelling urethralcatheter were associated withnosocomial infectionin elderly TBI patients. Logistic regression analysisrevealed thatage>75 years,severe TBI,tracheotomy,mechanical ventilation,lying-bed period> 12 days,and indwelling urethral catheter period>7dayswere the risk factorsofnosocomial infection(P < 0.05). The predictingmodelwas:Z=-7.341+1.955×age+2.665×TBIseverity+1.616×tracheotomy+1.833×mechanical ventilation+1.610×lying-bed period+1.392×time ofindwelling urethralcatheter..The areaunder ROC curve was 0.938(95%CI 0.901~0.974),Hosmer-Lemeshow test χ2=9.400 (P > 0.05)and the maximum Youden index was 0.740,the correspondingcut-off value,sensitivity and specificity was 0.502,0.822 and 0.941. Inaddition,thepredicting accuracy,sensitivity andspecificity of the model invalidation groupwere 0.944,0.833 and 0.976 respectively. Conclusion:Advanced age,severe TBI,tracheotomy,mechanical ventilation,long termbed restandindwelling urethralcatheter increasethe risk ofnosocomial infectioninelderly patientswith TBI. The predicting modelhas goodaccuracy and maybe helpfulfor early diagnosis and therapyof potentially infected patients.
文章编号:     中图分类号:    文献标志码:
基金项目:南京大学中国医院改革发展研究院面上项目(NDYG2022020) 南京大学中国医院改革发展研究院面上项目(NDYG2022020)
引用文本:
徐幼桥,钱 静*,练 敏,陆 婷,罗琳娜,徐翠平.老年TBI患者医院感染危险因素探讨与预测模型构建△[J].江苏卫生事业管理,2023,34(11):1537-1541.
XU Youqiao,QIAN Jing*,LIAN Min,LU Ting,LUO Linna,XU Cuiping.Risk Factors and Predicting Modelof Nosocomial Infectionin Elderly Patients with Traumatic Brain Injury△[J].Jiangsu Health System Management,2023,34(11):1537-1541.

用微信扫一扫

用微信扫一扫