Relation Between Serum Sodium Trajectory and Survival in Septic Patients with Cardiopulmonary Bypass Surgery: Based on Medical Information Mart for Intensive Care-IV Database

The influence of serum sodium level changes on septic patient survival after cardiopulmonary bypass surgery is not clear. We attempted to figure out the impact of serum sodium trajectory changes on the 30-day mortality rate of such patients. The Medical Information Mart for Intensive Care (MIMIC)-IV database was searched to gather patients who developed sepsis after cardiopulmonary bypass surgery. A group-based trajectory model (GBTM) was employed to determine the serum sodium trajectory within 72 h of ICU admission. Patients’ survival differences between different trajectory groups were compared using Kaplan–Meier (K–M) survival curves. Cox regression models were further employed to explore the correlation between survival status and serum sodium trajectory. 1,038 eligible patients were involved in this project. GBTM identified 3 serum sodium trajectories, all showing a trend of initial decrease followed by an increase. K–M curve analysis uncovered a notable difference in 30-day survival status between Class 1 and Class 2 (Log-rank <i>p</i> = 0.039), while no obvious differences were observed between other groups. Cox hazard analysis revealed that in the three models adjusting for different covariates, Class 2 was connected with the increased risk of survival (OR &gt; 1, <i>p</i> &lt; 0.05). Higher serum sodium trajectory is linked with elevated 30-day death risk in septic patients following cardiopulmonary bypass surgery. Repressing high levels of serum sodium may be beneficial for patient survival.

体外循环术后脓毒症患者的血清钠水平变化对其生存结局的影响尚未明确。本研究旨在探讨血清钠动态变化轨迹对该类患者30天死亡率的影响。
本研究检索了重症监护医学信息库（Medical Information Mart for Intensive Care, MIMIC-IV），收集了体外循环术后并发脓毒症的患者。采用基于分组的轨迹模型（group-based trajectory model, GBTM）分析患者入住重症监护病房（intensive care unit, ICU）后72小时内的血清钠动态变化轨迹。采用卡普兰-迈耶（Kaplan-Meier, K-M）生存曲线比较不同轨迹分组患者的生存差异，进一步运用Cox回归模型探究患者生存状态与血清钠动态变化轨迹的相关性。
本研究共纳入1038例符合纳入标准的患者。基于分组的轨迹模型共识别出3种血清钠动态变化轨迹，所有轨迹均呈现先下降后上升的趋势。
卡普兰-迈耶生存曲线分析显示，第1组与第2组患者的30天生存状态存在显著差异（Log-rank检验，p=0.039），其余组间未观察到明显差异。Cox风险回归分析结果表明，在校正了不同协变量的3种模型中，第2组患者的生存风险显著升高（比值比（odds ratio, OR）>1，p<0.05）。
体外循环术后脓毒症患者的血清钠轨迹水平越高，其30天死亡风险越高。调控过高的血清钠水平或许有助于改善患者的生存结局。