Kidney Week

Abstract: FR-PO0576

Predictors of 24-Hour Sodium Correction in Moderate to Severe Hyponatremia

Session Information

• Fluid, Electrolyte, and Acid-Base Disorders: Clinical - 2
  November 07, 2025 | Location: Exhibit Hall, Convention Center
  Abstract Time: 10:00 AM - 12:00 PM

Category: Fluid, Electrolytes, and Acid-Base Disorders

• 1102 Fluid, Electrolyte, and Acid-Base Disorders: Clinical

Authors

• Ismail, Ibrahim, University of Minnesota Twin Cities, Minneapolis, Minnesota, United States

Ibrahim Ismail

• Nachman, Patrick H., University of Minnesota Twin Cities, Minneapolis, Minnesota, United States

Patrick H. Nachman, MD, FASN

• Thorne, Peter E., University of Minnesota Twin Cities, Minneapolis, Minnesota, United States

Peter E. Thorne, MD

• Patel, Milen Saurin, University of Minnesota Twin Cities, Minneapolis, Minnesota, United States

Milen Saurin Patel

• Safadi, Sami, University of Minnesota Twin Cities, Minneapolis, Minnesota, United States

Sami Safadi, MD

Background

Moderate to severe hyponatremia is a common occurrence in acute-care settings and must be managed with caution due to the risk of complications from overly rapid correction. This study assessed the prevalence and predictors of 24-hour serum sodium change in a large U.S. health system cohort.

Methods

Using an acute-care data warehouse (January 2011 to June 2024; 8.6 million encounters), we identified adult inpatients with an initial serum sodium level of ≤125 mEq/L who had a repeat sodium measurement within 24 hours (n = 6,393 patients). Data collected included demographics, comorbidities, and intravenous fluid use. Sodium correction after 24 hours (ΔNa) was classified as: Undercorrection: <6 mmol/L, Appropriate correction: 6-12 mmol/L, and Overcorrection: >12 mmol/L.

Results

The median age of patients was 64 years; 54% were female and 84% were White. Appropriate sodium correction occurred in 36% of patients (n = 2,307), while 54% experienced overcorrection (n = 3,427), and 10% remained undercorrected (n = 659).
Age and Charlson Comorbidity Index (CCI) were inversely associated with the rate of sodium correction, ΔNa -0.0729 mmol/L per year of age (p < 0.001) and ΔNa -0.1827 mmol/L per CCI point (p < 0.001). Lower initial sodium levels were predictive of less rapid correction, ΔNa -0.408 mmol/L per mmol/L (p < 0.001).
Administration of normotonic and hypertonic IV fluids had no significant impact on sodium correction rates. In contrast, hypotonic fluid administration was weakly associated with a higher correction rate (ΔNa 0.001 mmol/L change per mL; p < 0.001), though this likely reflects correlation rather than causation.
Multivariate linear regression yielded modest performance (Train R²= 0.3134; Test R² = 0.3163; Train MSE = 22.39; Test MSE = 22.90). A nonlinear model using gradient boosting showed slightly better predictive performance (Train R² = 0.5370; Test R² = 0.3615; Train MSE = 15.0953; Test MSE = 21.3844).

Conclusion

Inappropriate sodium correction remains a significant issue in inpatient care. Only 36% of patients achieved guideline-concordant sodium correction (6-12 mmol/L) within the first 24 hours. Predicting the rate of sodium correction remains challenging. These results underscore the need for advanced predictive models and standardized treatment protocols to enhance hyponatremia management.

Digital Object Identifier (DOI)

• doi: 10.1681/ASN.20250jx18kpk