摘要
目的探讨脓毒症相关急性肾损伤(SA-AKI)不良预后的危险因素并建立列线图预测模型。方法回顾性分析2019年1月至2022年9月山东第一医科大学附属省立医院重症医学科收治的SA-AKI患者的临床数据,包括人口学资料,诊断SA-AKI 24 h内的血细胞计数及血生化指标最差值,住院期间是否接受肾脏替代治疗(RRT)、机械通气、血管升压药物治疗,诊断24 h内的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)、纤维蛋白原/白蛋白比值(FAR),以及急性肾损伤(AKI)分期、总住院时间、重症监护病房(ICU)住院时间等。根据28 d结局将患者分为存活组和死亡组,进行组间指标比较。采用单因素和多因素Logistic回归分析筛选SA-AKI患者死亡的危险因素,根据得出的危险因素构建预测SA-AKI预后的列线图模型;绘制受试者工作特征曲线(ROC曲线)和校准曲线,评估列线图模型对SA-AKI预后的预测价值。结果共纳入SA-AKI患者113例,其中存活组67例,死亡组46例,SA-AKI患者28 d病死率为40.7%。组间比较显示,年龄≥65岁、AKI分期、机械通气、血管升压药物、RRT、ICU住院时间及实验室指标胱抑素C(Cys C)、纤维蛋白原(Fib)、FAR差异均有统计学意义。多因素Logistic回归分析显示,年龄≥65岁〔优势比(OR)=7.967,95%可信区间(95%CI)为1.803~35.203,P=0.006〕、Cys C(OR=7.202,95%CI为1.756~29.534,P=0.006)、FAR(OR=2.444,95%CI为1.506~3.968,P<0.001)、RRT(OR=7.639,95%CI为1.391~41.951,P=0.019)是SA-AKI患者死亡的独立危险因素。ROC曲线分析显示,年龄≥65岁、Cys C、FAR、RRT预测SA-AKI患者死亡的ROC曲线下面积(AUC)分别为0.713、0.856、0.911、0.701。以年龄≥65岁、Cys C、FAR、RRT构建SA-AKI患者预后风险的列线图预测模型,ROC曲线分析显示该模型预测预后的AUC为0.967(95%CI为0.932~1.000),校准曲线提示实际概率与预测概率的一致性良好。结论年龄≥65岁、Cys C、FAR、RRT是SA-AKI患者死亡的独立危险因素,基于以上4个因素构建的列线图预测模型能较准确地预测SA-AKI患者预后,有助于医生及时调整治疗策略,改善患者预后。
Objective To explore the risk factors for poor prognosis in sepsis-associated acute kidney injury(SA-AKI)and establish a nomogram predictive model.Methods The clinical data of patients with SA-AKI admitted to the department of critical care medicine of Shandong Provincial Hospital Affiliated to Shandong First Medical University from January 2019 to September 2022 were retrospectively analyzed,including demographic information,worst values of blood cell counts and biochemical indicators within 24 hours of SA-AKI diagnosis,whether the patient received renal replacement therapy(RRT),mechanical ventilation,vasopressor therapy during hospitalization,acute physiology and chronic health evaluationⅡ(APACHEⅡ),sequential organ failure assessment(SOFA),fibrinogen-to-albumin ratio(FAR)within 24 hours of diagnosis,acute kidney injury(AKI)staging,total length of hospital stay,length of intensive care unit(ICU)stay,and others.According to the 28-day outcome,the patients were divided into survival group and death group,and the indicators between the two groups were compared.Univariate and multivariate Logistic regression analyses were used to screen for risk factors associated with mortality in SA-AKI patients.A nomogram predictive model for SA-AKI prognosis was constructed based on the identified risk factors.Receiver operator characteristic curve(ROC curve)and calibration plots were generated to evaluate the predictive value of the nomogram model for SA-AKI prognosis.Results A total of 113 SA-AKI patients were included,with 67 in the survival group and 46 in the death group.The 28-day mortality among SA-AKI patients was 40.7%.The comparison between the two groups showed that there were statistically significant differences in age≥65 years,AKI stage,mechanical ventilation,vasopressors,RRT,length of ICU stay,and laboratory indicators cystatin C(Cys C),fibrinogen(Fib),and FAR.Multivariate Logistic regression analysis showed that age≥65 years[odds ratio(OR)=7.967,95%confidence interval(95%CI)was 1.803-35.203,P=0.006],cystatin C(OR=7.202,95%CI was 1.756-29.534,P=0.006),FAR(OR=2.444,95%CI was 1.506-3.968,P<0.001),and RRT(OR=7.639,95%CI was 1.391-41.951,P=0.019)were independent risk factors for mortality in SA-AKI patients.ROC curve analysis showed that the area under the ROC curve(AUC)for age≥65 years,cystatin C,FAR,and RRT in predicting SA-AKI patient mortality were 0.713,0.856,0.911,and 0.701,respectively.A nomogram predictive model for SA-AKI patient prognosis was constructed based on age≥65 years,cystatin C,FAR,and RRT,with an AUC of 0.967(95%CI was 0.932-1.000)according to ROC curve analysis.The calibration plot indicated good consistency between predicted and actual probabilities.Conclusions Age≥65 years,cystatin C,FAR,and RRT are independent risk factors for mortality in SA-AKI patients.The nomogram predictive model based on these four factors can accurately predict SA-AKI patient prognosis,helping physicians adjust treatment strategies in a timely manner and improve patient outcomes.
作者
赵丽
刘岩
陈曼
陈丽
周升琳
白雪
张继承
Zhao Li;Liu Yan;Chen Man;Chen Li;Zhou Shenglin;Bai Xue;Zhang Jicheng(Department of Critical Care Medicine,Shandong Provincial Hospital Affiliated to Shandong First Medical University,Jinan 250021,Shandong,China;Department of Pulmonary and Critical Care Medicine,China-Japan Friendship Hospital,Beijing 100029,China)
出处
《中华危重病急救医学》
CAS
CSCD
北大核心
2023年第12期1255-1261,共7页
Chinese Critical Care Medicine
基金
中华国际医学交流基金会心血管多学科整合思维研究基金项目(Z-2016-23-2001-41)
中华国际医学交流基金会中青年医学研究专项基金项目(Z-2018-35-2004)。
关键词
脓毒症相关急性肾损伤
列线图
预后
预测模型
Sepsis-associated acute kidney injury
Nomogram
Prognosis
Prediction model
