Xinyuan Ding , Shangzhong Chen , Lihong Zhu , Chengcheng Zheng , Hang Mou , Guolong Cai , Yanfei Shen
{"title":"Dynamic inflammatory trajectories and 28-day mortality in patients in the intensive care unit: An exploratory observational study","authors":"Xinyuan Ding , Shangzhong Chen , Lihong Zhu , Chengcheng Zheng , Hang Mou , Guolong Cai , Yanfei Shen","doi":"10.1016/j.jointm.2025.12.011","DOIUrl":null,"url":null,"abstract":"<div><h3>Background</h3><div>Inflammatory dynamics profoundly influence outcomes in critically ill patients; however, the prognostic significance of early inflammatory trajectories remains unclear. This study aimed to characterize the distinct trajectories of inflammatory mediators and evaluate their association with 28-day hospital mortality in patients in the intensive care unit (ICU).</div></div><div><h3>Methods</h3><div>A retrospective cohort of 511 critically ill patients in the ICU (July 2018–June 2024) was analyzed using group-based multitrajectory modeling for longitudinal measurements of C-reactive protein, procalcitonin (PCT), interleukin (IL)-2, IL-4, IL-6, and IL-10 during the first five ICU days (log-transformed as <em>Y</em>=ln[mediator + 1] for standardization). Multivariable Cox regression, inverse probability weighting, and treatment interaction analyses were performed to assess mortality risk and therapeutic responses across trajectories.</div></div><div><h3>Results</h3><div>Three inflammatory trajectories were identified. The mortality rate was 9.5% in the first trajectory (Traj. 1), characterized by persistent mild inflammation (18.6%). In the second trajectory (Traj. 2), a shift toward moderate inflammation was observed with subsequent resolution (55.6%), leading to a mortality rate of 12.3%. Compared with Traj. 1, this trajectory did not demonstrate a substantial increase in death risk (adjusted hazard ratio [HR]=1.14, 95% confidence interval [CI]: 0.54–2.42). The third trajectory (Traj. 3) exhibited sustained hyperinflammation (25.8%), resulting in a mortality rate of 28.0%, which was significantly associated with an elevated risk of death compared with Traj. 1 (adjusted HR=3.41, 95% CI: 1.57 to 7.41). Inverse probability weighting analysis confirmed that compared to Traj. 1, the absolute mortality increase for Traj. 2 was 4.3% (<em>P</em>=0.234), whereas that for Traj. 3 was 10.5% (<em>P</em>=0.031). Notably, corticosteroid use reduced mortality, specifically in Traj. 3 patients (<em>P</em>=0.031), whereas vasopressors and ulinastatin showed no trajectory-specific benefits.</div></div><div><h3>Conclusions</h3><div>Early inflammatory trajectories robustly predict ICU mortality, with sustained hyperinflammation (Traj. 3) conferring the highest risk. Dynamic biomarker profiling may facilitate the development of targeted immunomodulatory strategies in high-risk subgroups.</div></div>","PeriodicalId":73799,"journal":{"name":"Journal of intensive medicine","volume":"6 2","pages":"Pages 166-174"},"PeriodicalIF":0.0000,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of intensive medicine","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2667100X26000332","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2026/2/28 0:00:00","PubModel":"Epub","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Background
Inflammatory dynamics profoundly influence outcomes in critically ill patients; however, the prognostic significance of early inflammatory trajectories remains unclear. This study aimed to characterize the distinct trajectories of inflammatory mediators and evaluate their association with 28-day hospital mortality in patients in the intensive care unit (ICU).
Methods
A retrospective cohort of 511 critically ill patients in the ICU (July 2018–June 2024) was analyzed using group-based multitrajectory modeling for longitudinal measurements of C-reactive protein, procalcitonin (PCT), interleukin (IL)-2, IL-4, IL-6, and IL-10 during the first five ICU days (log-transformed as Y=ln[mediator + 1] for standardization). Multivariable Cox regression, inverse probability weighting, and treatment interaction analyses were performed to assess mortality risk and therapeutic responses across trajectories.
Results
Three inflammatory trajectories were identified. The mortality rate was 9.5% in the first trajectory (Traj. 1), characterized by persistent mild inflammation (18.6%). In the second trajectory (Traj. 2), a shift toward moderate inflammation was observed with subsequent resolution (55.6%), leading to a mortality rate of 12.3%. Compared with Traj. 1, this trajectory did not demonstrate a substantial increase in death risk (adjusted hazard ratio [HR]=1.14, 95% confidence interval [CI]: 0.54–2.42). The third trajectory (Traj. 3) exhibited sustained hyperinflammation (25.8%), resulting in a mortality rate of 28.0%, which was significantly associated with an elevated risk of death compared with Traj. 1 (adjusted HR=3.41, 95% CI: 1.57 to 7.41). Inverse probability weighting analysis confirmed that compared to Traj. 1, the absolute mortality increase for Traj. 2 was 4.3% (P=0.234), whereas that for Traj. 3 was 10.5% (P=0.031). Notably, corticosteroid use reduced mortality, specifically in Traj. 3 patients (P=0.031), whereas vasopressors and ulinastatin showed no trajectory-specific benefits.
Conclusions
Early inflammatory trajectories robustly predict ICU mortality, with sustained hyperinflammation (Traj. 3) conferring the highest risk. Dynamic biomarker profiling may facilitate the development of targeted immunomodulatory strategies in high-risk subgroups.