老年脓毒症合并急性肾损伤住院死亡预测模型建立与验证  

Establishment and validation of a prediction model for death during hospitalization in elderly patients with sepsis combined with acute kidney injury

作　　者：宋铭慧 王磊[1] 毕迎[1] 宋艳[1] 王慧[1] SONG Minghui;WANG Lei;BI Ying;SONG Yan;WANG Hui(Department of Critical Care Medicine,The First Affiliated Hospital of Harbin Medical University,Harbin,Heilongjiang 150000,China)

机构地区：[1]哈尔滨医科大学附属第一医院重症医学科,黑龙江哈尔滨150000

出　　处：《中国临床研究》2024年第11期1686-1690,共5页Chinese Journal of Clinical Research

基　　金：黑龙江省医学科技创新项目(H2020-L31)。

摘　　要：目的初步建立老年脓毒症合并急性肾损伤(AKI)患者住院期间死亡的预测模型,并进行验证。方法回顾性纳入2019年1月至2023年12月于哈尔滨医科大学附属第一医院治疗的378例老年脓毒症合并AKI患者的临床资料。根据住院期间转归,将其分为死亡组122例(32.28%)和生存组256例(67.72%)。采用多因素logistic回归法筛选脓毒症AKI患者住院死亡的独立危险因素,并据此建立预测模型。采用受试者工作特征(ROC)曲线对模型进行评价。结果多因素logistic回归结果显示,年龄>70岁(OR=1.387,95%CI:1.083~1.776)、AKIⅢ期(OR=2.006,95%CI:1.388~2.899)、SOFA评分>10分(OR=1.791,95%CI:1.266~2.536)、PCT>3.67 ng/mL(OR=1.553,95%CI:1.166~2.068)、APTT>40 s(OR=1.290,95%CI:1.040~1.602)是老年脓毒症合并AKI患者住院期间全因死亡的独立危险因素(P<0.05)。建立脓毒症合并AKI患者住院死亡风险(C-index)的方程,即C-index=-1.722+0.327×(年龄)+0.696×(AKI分期)+0.583×(SOFA)+0.440×(PCT)+0.255×(APTT)。ROC曲线显示,C-index预测老年脓毒症合并AKI患者住院死亡的曲线下面积为0.876(95%CI:0.837~0.915),准确率为81.22%,敏感度为78.69%,特异度为82.42%。结论基于年龄、SOFA评分、PCT、AKI分期、APTT建立的预测模型有助于早期识别老年脓毒症合并AKI患者住院期间死亡高风险人群,进而早期干预并调整治疗策略,有助于患者预后的改善。Objective To initially establish a prediction model for death during hospitalization in elderly patients with sepsis combined with acute kidney injury(AKI)and validate it.Methods Clinical data of 378 elderly patients with sepsis combined with AKI treated in The First Affiliated Hospital of Harbin Medical University from January 2019 to December 2023 were retrospectively included.Based on the outcome during hospitalization,they were divided into death group(122 cases,32.28%)and survival group(256 cases,67.72%).Multivariate logistic regression was used to screen the independent risk factors for hospitalized death in septic AKI patients,and a prediction model was established accordingly.The model was evaluated using receiver operating characteristic(ROC)curves.Results The results of multivariate logistic regression showed that age>70 years(OR=1.387,95%CI:1.083-1.776),AKI stageⅢ(OR=2.006,95%CI:1.388-2.899),sequential organ failure assessment(SOFA)score>10(OR=1.791,95%CI:1.266-2.536),procalcitonin(PCT)>3.67 ng/mL(OR=1.553,95%CI:1.166-2.068),and activated partial thromboplastin time(APTT)>40 s(OR=1.290,95%CI:1.040-1.602)were independent risk factors for all-cause mortality during hospitalization in elderly patients with sepsis combined with AKI(P<0.05).Based on the results of multivariate analysis,an equation for the risk of in-hospital death(C-index)in patients with sepsis combined with AKI was established,C-index=-1.722+0.327×(age)+0.696×(AKI stage)+0.583×(SOFA)+0.440×(PCT)+0.255×(APTT).The ROC curves showed that the C-index predicted in-hospital death in elderly patients with sepsis combined with AKI with an AUC of 0.876(95%CI:0.837-0.915),an accuracy of 81.22%,a sensitivity of 78.69%and a specificity of 82.42%.Conclusion The prediction model based on age,SOFA score,PCT,AKI staging,and APTT can help to identify the high-risk group of elderly patients with sepsis combined with AKI at an early stage of death during hospitalization,and then intervene and adjust the treatment strategy at an early stage,which can

关 键 词：老年 脓毒症 急性肾损伤 活化部分凝血酶时间 序贯器官衰竭评分 

分 类 号：R631[医药卫生—外科学]