红细胞分布宽度/血清白蛋白比值对ICU全因病死率的预测价值  

The predictive value of red blood cell distribution width/serum albumin ratio for ICU all-cause mortality

作　　者：冯浠镰 张伟[2] 杨柳柳[2] Feng Xilian;Zhang Wei;Yang Liuliu(Emergency Department,Guangdong Integrated Traditional Chinese and Western Medicine Hospital,Foshan 528200,China)

机构地区：[1]广东省中西医结合医院急诊科,广东佛山528200 [2]广州中医药大学第一附属医院呼吸与危重症医学科,广东广州510405

出　　处：《中国急救医学》2025年第3期205-210,共6页Chinese Journal of Critical Care Medicine

摘　　要：目的分析红细胞分布宽度/血清白蛋白比值(RAR)与重症监护病房(ICU)全因病死率的联系,探讨RAR的预测价值。方法于重症监护医学信息数据库Ⅲ(MIMICⅢ)中提取ICU住院患者资料,采用四分位数方法,将RAR分为4个区间,分析不同区间RAR的生存曲线;采用Cox回归模型联合三次样条回归及亚组分析探讨RAR与ICU全因病死率的关系,并计算各亚组的交互P值,制作森林图;采用受试者工作特征(ROC)曲线评价RAR的预测性能;采用决策曲线分析(DCA)评价临床应用价值。结果研究共纳入27473例ICU患者,RAR分成4个区间,RAR≤3.51有6738例,3.51<RAR≤4.17有6979例,4.17<RAR≤5.21有6862例,RAR>5.21有6894例;Cox回归显示,RAR是ICU全因病死率的独立危险因素;三次样条回归显示,RAR与ICU全因病死率呈非线性关系;ROC曲线下面积(AUC)为0.7071,高于单独应用红细胞分布宽度(RDW,AUC=0.6518);DCA显示,RAR具有临床应用价值;亚组分析结果显示,在心脏外科监护病房(CSRU)、外科监护病房(SICU)及创伤外科监护病房(TSICU)中,亚组具有交互作用,风险比(HR)及95%置信区间(CI)分别为CSRU 1.53(1.43~1.63)、SICU 1.06(1.01~1.11)、TSICU 1.02(0.96~1.09)。结论RAR是ICU全因病死率的独立危险因素,对ICU全因病死率具有预测意义,但在TSICU中并无价值。Objective To investigate the association of red blood cell distribution width/serum albumin ratio(RAR)with all-cause mortality in the intensive care unit(ICU)and the predictive value of RAR.Methods The data of ICU inpatients were extracted from medical information mart for intensive careⅢ(MIMICⅢ)medical database.RAR was divided into four intervals by quartile method,and the survival curves of RAR in different intervals were analyzed.Cox regression models combined with restricted cubic spline and subgroup analysis were used to study the association between RAR and ICU all-cause mortality.The P value for interaction of each subgroup was calculated and a forest map was drawn.Receiver operating characteristic(ROC)curve was used to evaluate the predictive performance of RAR.Decision curve analysis(DCA)was used to evaluate the clinical application value.Results A total of 27473 ICU patients were included in the study.6738,6979,6862 and 6894 patients fell into the first interval(RAR≤3.51),the second interval(3.51<RAR≤4.17),the third interval(4.17<RAR≤5.21)and fourth interval(RAR>5.21),respectively.In the Cox regression model adjusted for different confounding variables,the RAR was an independent risk factor for all-cause mortality in ICU.The restricted cubic spline showed a non-linear association between RAR and all-cause mortality in ICU.The area under the ROC curve(AUC)was 0.7071,which was higher than the AUC of the red blood cell distribution width(RDW,AUC=0.6518).DCA showed that RAR had clinical application value.The results of subgroup analysis showed that the P for interaction was less than 0.05 in the cardiac surgery recovery unit(CSRU),surgical intensive care unit(SICU)and trauma-surgical intensive care unit(TSICU).The hazard ratio(HR)and 95%confidence interval(CI)of the CSRU,SICU and TSICU were 1.53(1.43-1.63),1.06(1.01-1.11)and 1.02(0.96-1.09),respectively.Conclusions RAR is an independent risk factor for ICU all-cause mortality,which has predictive significance for ICU all-cause mortality,but has no

关 键 词：红细胞分布宽度 血清白蛋白 红细胞分布宽度/血清白蛋白比值 重症监护病房 全因病死率 重症监护医学信息数据库Ⅲ(MIMICⅢ) 

分 类 号：R459.7[医药卫生—急诊医学]