Serum Osmolality and Stroke Mortality in the ICU: A U-Shaped Risk Pattern and Its Clinical Implications

Background: Serum osmolality (SOSM) is a simple and objective tool for assessing hydration status and has been demonstrated in several studies to hold significant prognostic value in patients with cardiovascular and cerebrovascular diseases. This study aims to evaluate the association between SOSM and clinical outcomes in patients with stroke. Methods: This study evaluated the prognostic relevance of SOSM in stroke patients using data from the MIMIC-IV database. Eligible participants were divided into four quartile groups based on their SOSM: Q1, 277.62-296.30 mOsm/kg H₂O; Q2, 296.31-301.60 mOsm/kg H₂O; Q3, 301.61-307.74 mOsm/kg H₂O; and Q4, 307.75-327.50 mOsm/kg H₂O. This study used Cox proportional hazards regression, subgroup analysis, and restricted cubic spline analysis to examine the association between SOSM and mortality. Results: This study included 6005 stroke patients. The 30-day and 365-day mortality rates were 13.49% and 15.84%, respectively. After adjusting for relevant confounders, multivariate Cox regression analysis showed that higher SOSM was independently associated with an increased risk of 30-day death (HR: 1.83; 95% CI: 1.48-2.27; p < 0.001) and 365-day death (HR: 1.71; 95% CI: 1.41-2.08; p < 0.001). Analyses using restricted cubic splines (RCSs) and threshold effect modeling demonstrated a clear U-shaped relationship between SOSM and both short-term and long-term risk of death in stroke patients. Furthermore, subgroup analyses confirmed the stability of this association across diverse patient profiles. Conclusions: SOSM is independently associated with both 30-day and 365-day mortality in stroke patients. Our findings indicate that SOSM may be an effective indicator for stratifying high-risk patients who might benefit from targeted interventions, ultimately improving prognostic outcomes.