应激性血糖升高比值与重症急性肾损伤患者院内死亡风险的关系

The relationship between stress hyperglycemia ratio and in-hospital mortality of critically ill patients with acute kidney injury

目的 探讨应激性血糖升高比值(stress hyperglycemia ratio,SHR)对重症急性肾损伤(acute kidney injury,AKI)患者院内死亡风险的影响。方法 本研究数据来源于美国重症医学信息数据库Ⅲ(MIMICⅢ),选择首次入重症监护病房(intensive care unit,ICU)的593例成年AKI患者。采用PostgresSQL语言提取患者临床资料、实验室检查相关指标、生理功能评分和结局事件等数据,所有入选患者根据SHR水平分为SHR≥1.1组和SHR<1.1组。构建Logistic多元回归模型,探讨SHR对重症AKI患者院内死亡的影响。应用受试者工作特征(receiver operator characteristic,ROC)曲线分析SHR对重症AKI患者院内死亡风险的预测能力。结果 SHR<1.1组患者308例,SHR≥1.1组患者285例。SHR≥1.1组患者院内死亡比例明显高于SHR<1.1组(19.6%vs.10.7%,P=0.003)。单因素、多因素Logistic回归模型均提示SHR升高导致重症AKI患者院内死亡风险增加(P<0.05),其中全调整模型3提示,SHR每升高0.1,院内死亡风险升高9%(OR 1.09,95%CI1.03~1.14,P=0.002)。SHR联合简化急性生理功能评估系统Ⅱ(SAPSⅡ)评分预测重症AKI患者院内死亡风险的曲线下面积(area under curve,AUC)为0.742(95%CI0.686~0.798),预测能力明显高于SHR(P=0.001),但与SAPSⅡ评分比较差异无统计学意义(P=0.170)。结论 SHR是重症AKI患者短期预后的独立危险因素,联合SAPSⅡ评分能增加重症AKI患者预后的判断能力。

Objective To investigate the effect of stress hyperglycemia ratio(SHR) on in-hospital mortality of critically ill patients with acute kidney injury(AKI).Methods Data of this study were from Medical Information Mart for Intensive Care Ⅲ(MIMIC Ⅲ) database,593 adult patients with AKI firstly admitted into intensive care unit(ICU) were included in this study.Postgres SQL language was used to extract clinical and lab data,physiological function score and outcome events of patients,and all included patients were divided into SHR≥1.1 group and SHR<1.1 group.Logistic regression models were established to investigate the impact of SHR on in-hospital mortality of critically ill AKI patients.Receiver operating characteristic(ROC) curve analysis was performed to evaluate the predicting ability of SHR on in-hospital mortality of critically ill AKI patients.Results Finally 308 patients were included in SHR<1.1 group and 285 patients in SHR≥1.1 group.The in-hospital mortality was significantly higher in SHR≥1.1 group than in SHR<1.1 group(19.6% vs.10.7%,P=0.003).Univariate and multivariate Logistic regression analysis showed that the increase of SHR was resulted was elevated in-hospital mortality risk of critically ill AKI patients(P<0.05),each 0.1 increment of SHR was associated with 9% increase of the risk of in-hospital mortality in the fully adjusted model 3(OR 1.09,95%CI 1.03-1.14,P=0.002).The area under ROC curve(AUC) of the combination of SHR and the simplified acute physiology scoreⅡ(SAPSⅡ) score was 0.742(95%CI 0.686-0.798)in predicting the risk of in-hospital mortality of critically ill AKI patients,which was significantly higher than that of SHR solely(P=0.001),but there was no significant difference compared with SAPSⅡ score(P=0.170).Conclusions SHR is the independent risk factor of short-term prognosis in critically ill AKI patients,and improves the prognostic ability for critically ill AKI patients in combination with SAPSⅡ score.