The predictive value of the dynamic change slope of red blood cell distribution width to platelet ratio combined with clinical indicators for the mortality outcome of patients with sepsis

Abstract

This retrospective study investigated the predictive value of the dynamic slope of the red blood cell distribution width to platelet ratio (RPR) for in-hospital mortality in patients with sepsis admitted to the intensive care unit (ICU). A total of 154 patients with sepsis admitted to the ICU of the Second Affiliated Hospital of Anhui Medical University between August 2023 and August 2024 were included and classified into non-survivors ( n  = 37) and survivors ( n  = 117) according to in-hospital outcome. Red blood cell distribution width (RDW) and platelet count (PLT) were recorded on days 1–5 after ICU admission, and RPR was calculated as RDW/PLT. Generalized estimating equations demonstrated significant differences in RPR between groups and over time, with a significant group-time interaction, indicating distinct temporal trends between survivors and non-survivors. The RPR in the survivors increased initially and then declined, whereas the RPR in the non-survivors showed a continuous upward trend. Receiver operating characteristic analysis showed that the slope of RPR change had good predictive performance for in-hospital mortality, with an area under the curve of 0.863 (95% CI 0.781–0.946). Optimal cutoff value of 0.017 yielded sensitivity of 86.5% and specificity of 83.8%. Multivariate logistic regression analysis identified high RPR slope (OR = 5.665, 95% CI 1.453–22.084), lactate level, Sequential Organ Failure Assessment (SOFA) score, and mechanical ventilation time as independent risk factors for in-hospital mortality. Furthermore, the combined model incorporating RPR slope, SOFA score, lactate, and mechanical ventilation time showed excellent predictive ability, with an area under the curve of 0.954 (95% CI 0.924–0.983). These findings suggest that dynamic monitoring of RPR provides valuable prognostic information and may improve early risk stratification in patients with sepsis.