Li Junjie, Shao Yuekai, Zheng Jie, Dai Qiuyu, Yu Kun, Qin Song, Liu Xinxin, Mei Hong
Department of Critical Care Medicine, Affiliated Hospital of Zunyi Medical University, Zunyi, China.
Zunyi Medical University, Zunyi, China.
Front Nutr. 2025 May 14;12:1563311. doi: 10.3389/fnut.2025.1563311. eCollection 2025.
Sepsis is a notable cause of death and poor prognosis in the intensive care unit (ICU). Presents an ambiguous association between advanced lung cancer inflammation (ALI) and short-term lethality in sepsis patients. The purpose of this study is to explore this relationship.
This retrospective study identified sepsis cases from the MIMIC-IV 3.0 dataset. Multivariable Cox regression analysis was used to evaluate the relationship between ALI and the risks of 30-day all-cause mortality (ACM) and ICU mortality. Kaplan-Meier (K-M) curves and log-rank tests were employed for survival analysis. Restricted cubic spline (RCS) regression was employed to explore the nonlinear association between ALI and mortality risk. Subgroup and sensitivity analyses were performed to confirm the reliability of the results and to evaluate the incremental effect of ALI on the prediction of short-term mortality.
A total of 4,147 sepsis cases were included in this study, with a 30-day ACM rate of 26.7% and a 30-day ICU mortality rate of 18.5%. In the completely adjusted Cox model, patients in the highest quartile of log2-ALI had a 38% lower risk of 30-day ACM (HR = 0.62, < 0.001) and a 29% lower risk of 30-day ICU mortality (HR = 0.71, = 0.002) compared to those in the lowest quartile. K-M curves showed that the group with the lowest log2-ALI had the lowest 30-day ACM and ICU survival rates (log-rank < 0.001). RCS showed a nonlinear relationship between log2-ALI and 30-day ACM (P-overall < 0.001, P-nonlinear < 0.05). In all subgroups, the relationship between log2-ALI and outcomes showed no notable heterogeneity (P for interaction > 0.05), and four different sensitivity analyses yielded robust results. The combination of sequential organ failure assessment (SOFA) score and log2-ALI improved the predictive ability for 30-day ACM, with significant increases in C-statistic, Net Reclassification Improvement (NRI), and Integrated Discrimination Improvement (IDI).
This research found that lower levels of ALI were notably linked to higher 30-day ACM and 30-day ICU mortality in sepsis patients, warranting further verification through prospective studies.
脓毒症是重症监护病房（ICU）死亡和预后不良的一个显著原因。晚期肺癌炎症（ALI）与脓毒症患者的短期致死率之间存在模糊的关联。本研究的目的是探讨这种关系。
这项回顾性研究从MIMIC-IV 3.0数据集中识别脓毒症病例。采用多变量Cox回归分析来评估ALI与30天全因死亡率（ACM）和ICU死亡率风险之间的关系。采用Kaplan-Meier（K-M）曲线和对数秩检验进行生存分析。采用受限立方样条（RCS）回归来探讨ALI与死亡风险之间的非线性关联。进行亚组分析和敏感性分析以确认结果的可靠性，并评估ALI对短期死亡率预测的增量效应。
本研究共纳入4147例脓毒症病例，30天ACM率为26.7%，3０天ICU死亡率为18.5%。在完全调整的Cox模型中，与最低四分位数的患者相比，log2-ALI最高四分位数的患者30天ACM风险降低38%（HR = 0.62，< 0.001），30天ICU死亡率风险降低29%（HR = 0.71，= 0.002）。K-M曲线显示，log2-ALI最低的组30天ACM和ICU生存率最低（对数秩< 0.001）。RCS显示log2-ALI与30天ACM之间存在非线性关系（总体P < 0.001，非线性P < 0.05）。在所有亚组中，log2-ALI与结局之间的关系均未显示出明显的异质性（交互作用P > 0.05），四种不同的敏感性分析均得出了可靠的结果。序贯器官衰竭评估（SOFA）评分与log2-ALI的组合提高了30天ACM的预测能力，C统计量、净重新分类改善（NRI）和综合辨别改善（IDI）均显著增加。
本研究发现，较低水平的ALI与脓毒症患者较高的30天ACM和30天ICU死亡率显著相关，需要通过前瞻性研究进一步验证。
