目的 评估重症急性肾损伤(acute kidney injury,AKI)患者启动连续性肾脏替代治疗(continuous renal replacement therapy,CRRT)后48 h液体超负荷变化与28 d死亡率之间的关系.方法 利用2008年-2019年MIMIC-Ⅳ数据库中的数据进行回顾性队列研究.纳入在重症监护病房入院后14d内因AKI接受CRRT治疗超过24 h的患者.暴露变量为CRRT启动后48 h液体超负荷改变比例(proportion of change of fluid overload,△FO％),其定义为体重标准化的液体入量与出量之差;研究终点是28 d死亡率.使用广义可加线性回归模型和logistic回归模型分析暴露因素和研究终点的关系.结果 研究纳入911例患者,△FO％的中位数(下四分位数,上四分位数)为-3.27％(-6.03％,0.01％),28 d死亡率为40.1％.广义可加线性回归模型显示CRRT启动后48 h的△FO％与28d死亡呈J型曲线关系.在校正其他变量后,与第2个四分位组相比,第1个四分位组的死亡风险无显著增高[比值比(odds ratio,OR)=1.23,95％置信区间(confidence interval,CI)(0.81,1.87),P=0.338],但第 3 个四分位组[OR=1.54,95％CI(1.01,2.35),P=0.046]和第 4 个四分位组[OR=2.05,95％CI(1.32,3.18),P=0.001]的 28d死亡率均显著增高.使用同样方法发现CRRT启动后第1个24 h △FO％四分位分组与28 d死亡无关(P＞0.05),但CRRT启动后第2个24 h △FO％与28 d死亡呈线性关系,即△FO％越大,死亡率越高[每增加1％的OR=1.10,95％CI(1.04,1.16),P＜0.001].结论 重症AKI患者启动CRRT后48 h内△ FO％过大与28 d死亡风险增高独立相关,且CRRT液体管理目标可能是动态变化的.
Association between the early change of fluid overload during continuous renal replacement therapy and mortality in critically ill patients with acute kidney injury
Objective To assess the relationship between the change in fluid overload at 48 h after initiation of continuous renal replacement therapy(CRRT)and 28-day mortality in critically ill patients with acute kidney injury(AKI).Methods A retrospective cohort study was performed using data from the MIMIC-Ⅳ database from 2008 to 2019.Patients who received CRRT for AKI for more than 24 h within 14 d of admission to the intensive care unit were included.The exposure variable was the proportion of change of fluid overload(△FO％,defined as the difference between body weight normalized fluid input and output)at 48 h after CRRT initiation,and the endpoint was 28-day mortality.Generalized additive linear regression models and logistic regression models were used to determine the relationship between the exposure and endpoint.Results A total of 911 patients were included in the study,with a median(lower quartile,upper quartile)△FO％of-3.27％(-6.03％,0.01％)and a 28-day mortality of 40.1％.Generalized additive linear regression model showed that the △FO％at 48 h after CRRT initiation was associated with a J-shaped curve with 28-day mortality.After adjusting for other variables,as compared with the second quartile of △FO％group,the first quartile group[odds ratio(OR)=1.23,95％confidence interval(CI)(0.81,1.87),P=0.338]was not associated with higher risk of 28-day mortality,while the third quartile group[OR=1.54,95％CI(1.01,2.35),P=0.046]and the fourth quartile group[OR=2.05,95％CI(1.32,3.18),P=0.001]were significantly associated with higher risk of 28-day mortality.There was no significant relationship between △FO％groups and 28-day mortality in the first 24-hour after CRRT initiation(P＞0.05),but there was a linear relationship between △FO％and 28-day mortality in the second 24-hour after CRRT initiation,the larger the △FO％,the higher the mortality rate[OR=1.10,95％CI(1.04 1.16),P＜0.00l for per 1％increase].Conclusion In critically ill patients with AKI,the △FO％greater than-3.27％within 48 h after CRRT initiation is independently associated with an increased risk of 28-day mortality,and the goals of CRRT fluid management may be dynamical.
NETL
NSTL
万方数据
