Older adults are at high risk of sepsis, but the immune-inflammatory features distinguishing sepsis from nonsevere infection and their prognostic relevance remain incompletely defined. We aimed to characterize immune-inflammatory profiles across the older adult infection spectrum and evaluate their longitudinal association with in-hospital mortality. In this multicentre prospective observational cohort, 1,851 participants aged ≥ 60 years were enrolled, including healthy controls, patients with nonsevere infection, and patients with sepsis. Immune-cell subsets, cytokines, complement components, and routine laboratory indices were assessed. Adjusted cross-sectional comparisons were performed using multivariable linear regression. Among patients with sepsis, model-estimated longitudinal profiles from intensive care unit (ICU) day 1 to day 7 were analysed using linear mixed-effects models among available scheduled measurements, and day-1 biomarkers associated with in-hospital mortality were evaluated using Cox regression. The cohort included 524 healthy controls, 511 patients with nonsevere infection, and 816 patients with sepsis. Across the infection spectrum, sepsis was characterized by progressively lower circulating lymphocyte, T-cell, cluster of differentiation 8 (CD8 + ) T-cell, and natural killer (NK)-cell counts, together with higher interleukin (IL)-6, IL-8, and IL-10 levels. These differences remained after adjustment for age, sex, and major comorbidities. Among patients with available follow-up measurements, non-survivors had consistently higher Sequential Organ Failure Assessment (SOFA) scores, IL-6, and IL-10 and lower lymphocyte, T-cell, CD8⁺ T-cell, NK-cell, and complement component 3 (C3) levels. No time-by-outcome interaction remained significant after false-discovery-rate correction. In multivariable Cox models, higher day-1 SOFA score and IL-10 were associated with increased in-hospital mortality, whereas higher NK-cell count was associated with lower mortality. In older adults, sepsis was associated with an adjusted circulating immune-inflammatory pattern characterized by lower cellular immune markers and higher cytokine levels. This separation was evident on ICU day 1 and remained observable among patients with available follow-up measurements. Given the observational design, these associations should not be interpreted as causal.
Wang et al. (Sat,) studied this question.