Hyponatremia as a predictor of cognitive deterioration in hospitalized post-stroke patients.

ABSTRACT

PURPOSE: Evidence is scarce regarding the association between hyponatremia and alterations in cognitive function among hospitalized older patients. We aimed to investigate the associations between hyponatremia and the baseline cognitive status, as well as the improvement in cognitive function, in hospitalized post-stroke patients. METHODS: This retrospective cohort study included consecutive hospitalized post-stroke patients. Serum sodium concentrations were extracted from medical records based on blood tests performed within 24 h of admission, with hyponatremia defined as a serum sodium concentration < 135 mEq/L. The main outcomes included admission and discharge scores for cognitive levels, assessed through the cognitive domain of the Functional Independence Measure (FIM-cognition), as well as the score changes observed during the hospitalization period. Multivariate linear regression analyses were used to determine the association between hyponatremia and outcomes of interest, adjusted for potential confounders. RESULTS: Data from 955 patients (mean age 73.2 years; 53.6 % men) were included in the analysis. The median baseline blood sodium level was 139 [137, 141], and 84 patients (8.8 %) exhibited hyponatremia. After full adjustment for confounders, the baseline hyponatremia was significantly and negatively associated with FIM-cognition values at admission (ß = -0.009, p = 0.016), discharge (ß = -0.038, p = 0.043), and the gain during hospital stay (ß = -0.040, p = 0.011). CONCLUSION: Baseline hyponatremia has demonstrated a correlation with decline in cognitive level over the course of rehabilitation in individuals after stroke. Assessing hyponatremia at the outset proves to be a pivotal prognostic indicator.