The prognostic importance of serum sodium levels at hospital discharge and one-year mortality among hospitalized patients.

Abstract

The optimal range of serum sodium at hospital discharge is unclear. Our objective was to assess the one-year mortality based on discharge serum sodium in hospitalized patients. We analyzed a cohort of hospitalized adult patients between 2011 and 2013 who survived hospital admission at a tertiary referral hospital. We categorized discharge serum sodium into five groups; ≤132, 133-137, 138-142, 143-147, and ≥148mEq/L. We assessed one-year mortality risk after hospital discharge based on discharge serum sodium, using discharge sodium of 138-142mEq/L as the reference group. Of 55901 eligible patients, 4.9%, 29.8%, 56.1%, 8.9%, 0.3% had serum sodium of ≤132, 133-137, 138-142, 143-147, and ≥148mEq/L, respectively. We observed a U-shaped association between discharge serum sodium and one-year mortality, with nadir mortality in discharge serum sodium of 138-142mEq/L. When adjusting for potential confounders, including admission serum sodium, one-year mortality was significantly higher in both discharge serum sodium ≤137 and ≥143mEq/L, compared with discharge serum sodium of 138-142mEq/L. The mortality risk was the most prominent in elevated discharge serum sodium of ≥148mEq/L (HR 3.86; 95% CI 3.05-4.88), exceeding the risk associated with low discharge serum sodium of ≤132mEq/L (HR 1.43; 95% CI 1.30-1.57). The optimal range of serum sodium at discharge was 138-142mEq/L. Both hypernatremia and hyponatremia at discharge were associated with higher one-year mortality. The impact on higher one-year mortality was more prominent in hypernatremia than hyponatremia.