基于无创血流动力学参数的脓毒症预测模型构建  

A New Model for Predicting Sepsis in Emergency Department was Developed and Validated by Ultrasound continue cardiac output monitor(USCOM)

作　　者：张弘 叶志澄 李敏[1] 胡佳松 李桂芬[1] 江慧琳[1] 林珮仪[1] 莫均荣[1] ZHANG Hong;YE Zhi-cheng;LI Min;HU Jia-song;LI Gui-fen;JIANG Hui-lin;LIN Pei-yi;Mo Jun-rong(Emergency Department,The second Affiliated Hospital of Guangzhou Medical University,510260)

机构地区：[1]广州医科大学附属第二医院急诊科,510260

出　　处：《岭南急诊医学杂志》2024年第3期223-226,共4页Lingnan Journal of Emergency Medicine

摘　　要：目的:探讨超声心排血量监测(USCOM)联合临床指标对脓毒症的预测价值。方法:收集2017年2月至2018年5月广州医科大学附属第二医院EICU疑似感染患者214例为研究对象。记录患者的一般资料和实验室检查结果,并使用USCOM测量患者入院时血流动力学数据。根据脓毒症3.0诊断标准,分为脓毒症组与非脓毒症组,应用Logistic回归分析构建脓毒症的预测模型。采用受试者操作特征曲线ROC曲线下面积检验模型预测效能。结果:共收录214例疑似感染患者,其中非脓毒症组129人,脓毒症组85人。最终纳入脓毒症预测模型的预测因子分别为:每搏功(SW)、外周血管阻力指数(SVRI)、GCS评分、降钙素原(PCT)、白细胞计数(WBC)。ROC曲线下面积(AUC)为0.895,95%CI(0.837-0.938),敏感性为83.1%,特异性为83.3%,Youden指数0.6643。结论:SW、SVRI、WBC、PCT、GCS评分是预测EICU疑似感染患者的脓毒症发生的影响因素,由其构建的模型对脓毒症的发生有较高的预测价值。Objective:To explore the predictive value of non-invasive ultrasonic cardiac output monitoring(USCOM)combined with clinical indicators in sepsis.Methods:A total of 214 patients with suspected infection in EICU of the Second Affiliated Hospital of Guangzhou Medical University from February 2017 to May 2018 were col-lected.The general information and laboratory test results of the patients were recorded,and the non-invasive hemody-namic data of the patients were collected and measured using USCOM at admission.According to the diagnostic criteria of sepsis 3.0,the patients were divided into sepsis group and non-sepsis group.Logistic regression analysis was used to construct the risk prediction model of sepsis.The area under the receiver operating characteristic curve(ROC)was used to test the predictive performance of the model.Results:A total of 214 patients with suspected infection were included,including 129 in the non-sepsis group and 85 in the sepsis group.The predictors finally involved in the sepsis prediction model were:stroke work(SW),systemic vascular resistance index(SVRI),Glasgow coma scale(GCS),procalcitonin(PCT)and white blood cell count(WBC).The area under the ROC curve(AUC)was 0.895,95%CI(0.837-0.938),sensitivity was 83.1%,specificity was 83.3%,Youden index was 0.643.Conclusion:SW,SVRI,WBC,PCT,GCS scores are the influencing factors for predicting the occurrence of sepsis in patients with suspected infection in EICU,and the model constructed by them has a high predictive value for predicting the occurrence of sepsis.

关 键 词：超声心输出量监测 脓毒症 预测模型 外周血管阻力 

分 类 号：R459.7[医药卫生—急诊医学]