Serum Sodium Variation Associated with Increased Mortality in Hospitalized Patients with Moderate to Severe Hyponatremia

Background: Correction of severe hyponatremia (serum sodium ≤125 mEq/L) must be approached cautiously given potential complications of overcorrection. Data on the relationship between serum sodium variation (ΔNa) and mortality is limited. Methods: From an 8.6-million-encounter acute-care warehouse (January 2011-June 2024) we identified 7184 admissions (6393 adult patients) whose first serum Na was ≤ 125 mEq/L and who had a repeat measurement within 24 h. ΔNa represented the absolute inpatient range. Primary outcomes were in-hospital, 30-day, and 365-day mortality. Logistic regression and Cox proportional-hazards models adjusted for age, demographics, admission serum sodium, sodium correction rate 24 hour, and Charlson Comorbidity Index (CCI) were performed. Results: Median age was 64 years (IQR 53–75); 54 % were female and 84 % White. Overall mortality was 5.6 % in-hospital, 10.4 % at 30 days, and 23.1 % at 1 year. ΔNa was associated with increased in-hospital, and 30-day morality. Unadjusted OR per 1 mEq/L increase in ΔNa is 1.06, and 1.02 respectively. Cox model (2590 deaths) ΔNa was a predictor of morality (HR 1.01 per mEq/L, 95 % CI 1.00–1.01, p < 0.005) after multivariable adjustment. Age (HR 1.02 per year) and CCI (HR 1.11 per point) were also associated with higher risk, whereas female sex conferred lower mortality (HR 0.84). Concordance of the final model was 0.67 indicating a modest performance. Conclusion: Among adults hospitalized with moderate-to-severe hyponatremia, greater variability in serum sodium levels during admission was associated with increased mortality, even after adjusting for comorbidities and the adequacy of initial 24-hour sodium correction. This finding raises questions about the physiological consequences of sodium fluctuations, independent of the net change over time. This underscores the importance of tightly controlled, stepwise sodium correction strategies.