连续性肾脏替代治疗脓毒症急性肾损伤患者预后预测模型构建  

Construction of prognostic prediction model for patients with sepsis-induced acute kidney injury treated with continuous renal replacement

作　　者：李雅琳[1] 李东风[1] 王静 李昊 王肖[2] therapy;Li Yalin;Li Dongfeng;Wang Jing;Li Hao;Wang Xiao(Department of Critical Care Medicine,Fuyang People's Hospital,Fuyang 236000,Anhui,China;Department of Urology,Fuyang People's Hospital,Fuyang 236000,Anhui,China)

机构地区：[1]阜阳市人民医院重症医学科,安徽阜阳236000 [2]阜阳市人民医院泌尿外科,安徽阜阳236000

出　　处：《中华危重病急救医学》2024年第12期1268-1272,共5页Chinese Critical Care Medicine

基　　金：安徽省卫生健康科研项目(AHWJ2023A30219);阜阳市临床重点专科建设项目(2023-41)。

摘　　要：目的分析连续性肾脏替代治疗(CRRT)脓毒症急性肾损伤患者预后的影响因素,并构建死亡风险预测模型。方法采用回顾性研究方法,选择2021年2月至2023年9月阜阳市人民医院收治的接受CRRT的脓毒症急性肾损伤患者作为研究对象。收集患者入组24 h内的一般资料、合并症、生命体征、实验室指标、疾病严重程度评分、治疗情况、重症监护病房(ICU)住院天数及28 d预后情况;采用Cox回归模型筛选脓毒症急性肾损伤患者预后的影响因素,并构建预测脓毒症急性肾损伤患者死亡的列线图模型;采用受试者工作特征曲线(ROC曲线)、校准曲线和Hosmer-Lemeshow检验对列线图模型的预测效能进行验证。结果共纳入146例脓毒症急性肾损伤患者,其中治疗28 d存活98例,死亡48例(病死率32.88%)。死亡组患者的血乳酸、白细胞介素-6(IL-6)、血清胱抑素C、急性生理学与慢性健康状况评分Ⅱ(APACHEⅡ)、序贯器官衰竭评分(SOFA)及机械通气比例均明显高于生存组,ICU住院天数明显长于生存组,肾小球滤过率明显低于生存组,差异均有统计学意义。Cox回归分析显示,血乳酸〔优势比(OR)=2.992,95%可信区间(95%CI)为1.023~8.754〕、IL-6(OR=3.522,95%CI为1.039~11.929)、血清胱抑素C(OR=3.999,95%CI为1.367~11.699)、机械通气(OR=4.133,95%CI为1.413~12.092)、APACHEⅡ评分(OR=5.013,95%CI为1.713~14.667)、SOFA评分(OR=3.404,95%CI为1.634~9.959)是脓毒症急性肾损伤患者死亡的危险因素(均P<0.05),肾小球滤过率(OR=0.294,95%CI为0.101~0.860)是脓毒症急性肾损伤患者死亡的保护因素(P<0.05)。ROC曲线分析显示,列线图模型预测脓毒症急性肾损伤患者28 d死亡的敏感度为80.0%(95%CI为69.1%~89.2%),特异度为89.3%(95%CI为83.1%~95.2%)。结论血乳酸、IL-6、机械通气、APACHEⅡ评分、SOFA评分、肾小球滤过率、血清胱抑素C与脓毒症急性肾损伤患者的死亡风险有关,构建列线图模型有助于早期Objective To explore the influencing factors of prognosis in patients with sepsis-induced acute kidney injury undergoing continuous renal replacement therapy(CRRT),and to construct a mortality risk prediction model.Methods A retrospective research method was adopted,patients with sepsis-induced acute kidney injury who received CRRT at Fuyang People's Hospital from February 2021 to September 2023 were included in this study.Collect general information,comorbidities,vital signs,laboratory indicators,disease severity scores,treatment status,length of stay in the intensive care unit(ICU),and 28-day prognosis were collected within 24 hours of patient enrollment.The Cox regression model was used to identify the factors influencing prognosis in patients with sepsis-induced acute kidney injury,and a nomogram model was developed to predict mortality in these patients.Receiver operator characteristic curve(ROC curve),calibration curve,and Hosmer-Lemeshow test were used to validate the predictive performance of the nomogram model.Results A total of 146 patients with sepsis-induced acute kidney injury were included,of which 98 survived and 48 died(with a mortality of 32.88%)after 28 days of treatment.The blood lactic acid,interleukin-6(IL-6),serum cystatin C,acute physiology and chronic health evaluationⅡ(APACHEⅡ),sequential organ failure assessment(SOFA),and proportion of mechanical ventilation in the death group were significantly higher than those in the survival group.The ICU stay was significantly longer than that in the survival group,and the glomerular filtration rate was significantly lower than that in the survival group.Cox regression analysis showed that blood lactic acid[odds ratio(OR)=2.992,95%confidence interval(95%CI)was 1.023-8.754],IL-6(OR=3.522,95%CI was 1.039-11.929),serum cystatin C(OR=3.999,95%CI was 1.367-11.699),mechanical ventilation(OR=4.133,95%CI was 1.413-12.092),APACHEⅡscore(OR=5.013,95%CI was 1.713-14.667),SOFA score(OR=3.404,95%CI was 1.634-9.959)were risk factors for mortality in patients

关 键 词：脓毒症 急性肾损伤 连续性肾脏替代治疗 预后 预测模型 

分 类 号：R459.7[医药卫生—急诊医学] R692[医药卫生—治疗学]