Incremental prognostic value of inflammatory biomarkers for 28-day mortality in patients with sepsis presenting to the emergency department: a retrospective 24-hour landmark cohort study

Background Early identification of patients with sepsis who are at high risk of death is central to emergency care. Procalcitonin (PCT), the neutrophil-to-lymphocyte ratio (NLR), and the monocyte-to-lymphocyte ratio (MLR) are readily available inflammatory markers, but their combined prognostic value is unclear. We evaluated whether a PCT/NLR/MLR-based inflammatory score predicts 28-day mortality in patients with sepsis. Methods This retrospective cohort study included 248 adult patients with sepsis admitted to the emergency department of Suining Central Hospital from January 2024 to April 2026. Demographic data, clinical scores, and first laboratory results within 24 hours of admission were collected. Receiver operating characteristic analysis was used to define cut-off values for PCT, NLR, and MLR and construct a 0–3 point score. Multivariable Cox regression, Harrell's C-index, ROC analysis, calibration plots, and 1000 bootstrap resamples were used for prognostic analysis and internal validation. Results Among 248 patients, 51 (20.6%) died within 28 days. The Youden-derived cut-off values were PCT > = 31.7 ng/mL, NLR > = 28.79, and MLR > = 1.60. In multivariable Cox regression based on 220 complete cases, female sex (hazard ratio [HR] 0.456, 95% confidence interval [CI] 0.239–0.872), albumin (HR 0.927, 95% CI 0.881–0.976), lactate (HR 1.086, 95% CI 1.010–1.167), APACHE II score (HR 1.078, 95% CI 1.042–1.115), and the combined inflammatory score (HR 1.436 per point, 95% CI 1.023–2.017) were associated with mortality. The final model had a C-index of 0.809 (95% CI 0.745–0.874), compared with 0.796 (95% CI 0.728–0.863) for the base model without the inflammatory score. The optimism-corrected C-index was 0.788, and the optimism-corrected calibration slope was 0.878. Conclusions The PCT/NLR/MLR-based inflammatory score was associated with 28-day mortality in patients with sepsis, but its incremental value beyond APACHE II score, lactate, albumin, and sex was limited. Routine inflammatory markers may support early risk stratification but should be interpreted with established indicators of severity, perfusion, and host reserve. Trial registration: Not applicable.