Platelet count has a nonlinear association with 30-day in-hospital mortality in ICU End- stage kidney disease patients: a multicenter retrospective cohort study

Abstract

Abstract Background: This study addresses the relationship between platelet count and 30-day in-hospital mortality in End-Stage Kidney Disease (ESRD) patients in the intensive care unit (ICU), a topic with limited existing evidence. Methods: Utilizing data from the US eICU-CRD v2.0 database (2014-2015), a retrospective cohort study was conducted involving 3700 ICU ESRD patients. We employed binary logistic regression, smooth curve fitting, and subgroup analyses to explore the association between platelet count and 30-day in-hospital mortality. Results: The 30-day in-hospital mortality rate was 13.27% (491/3700), with a median platelet count of 188×109/L. After adjusting for covariates, we observed an inverse relationship between platelet count and 30-day in-hospital mortality (OR = 0.98, 95% CI: 0.97, 0.99). Subgroup analyses supported these findings, particularly in non-diabetic patients (P for interaction 0.0002). Additionally, a nonlinear association surfaced, revealing an inflection point at 222×109/L. The effect sizes (OR) on the left and right sides of the inflection point were 0.94 (0.92, 0.96) and 1.03 (1.00, 1.05), respectively. Conclusion: This study identified a curved relationship between platelet count and 30-day in-hospital mortality rate in ICU ESRD patients. The saturation effect suggests that maintaining a platelet count around 222×109/L can reduce the in-hospital mortality rate of these patients for 30 days.