Dynamic neutrophil-to-lymphocyte-platelet ratio trajectories predict 30-day and 1-year mortality in sepsis: a retrospective cohort study based on MIMIC-IV 2.2

Background: This study investigated the relationship between the neutrophil-to-lymphocyte ratio (N/LPR) and its variability ratio (N/LPRR) with 30-day and 1-year mortality outcomes. Methods: A total of 7,443 patients from the MIMIC-IV 2.2 database were included, with 1,765 having multiple N/LPR measurements. Mortality at 1 year and within 30 days served as the primary endpoints. Patients were stratified into four groups according to baseline N/LPRR quartiles. Receiver operating characteristic (ROC) curves assessed the predictive value of N/LPR and N/LPRR for mortality. Kaplan-Meier analysis estimated the risk of mortality events, while restricted cubic spline (RCS) analysis explored the non-linear associations between N/LPR, N/LPRR, and mortality. Cox proportional hazards regression identified the relationship between N/LPRR and all-cause mortality. Results: A total of 792 cases of 1-year mortality (44.9%) were recorded, with 437 deaths (24.8%) occurring within 30 days. ROC analysis revealed that N/LPRR outperformed N/LPR in predicting adverse outcomes. Higher N/LPR and N/LPRR were associated with increased mortality rates. RCS analysis indicated significant non-linear relationships between N/LPR, N/LPRR, and mortality risk (both p-values for nonlinearity < 0.001). Subgroup analyses confirmed the robustness of these findings. Conclusion: In conclusion, elevated N/LPR and N/LPRR are linked to 30-day and 1-year mortality in patients with sepsis. N/LPRR, with its heightened sensitivity, offers clinicians valuable prognostic information on sepsis severity and progression.