Abstract Body (Do not enter title and authors here): Background. Outcomes of patients with acute decompensated heart failure (ADHF) remain suboptimal. Loop diuretics are first-line therapy in ADHF patients and low urinary spot sodium (spot-U_Na) after intravenous (IV) diuretic can identify a poor natriuretic response and predict poor prognosis. A cutoff of < 50 or < 70 mmol/L has been suggested as a threshold in different studies. We examine the validity of these two cutoffs in a large real-world dataset.
Objective. (1) Characterize temporal trends of spot-U_Na utilization and (2) assess association of low spot-U_Na with 90-day mortality rates in HF patients.
Methods. Using multicenter electronic health records (Epic Cosmos) (Jan 2005–May 2025) from 1,715 hospitals (~300 M patients), we identified HF patients using ICD-10 code I50* who were admitted to a hospital and received IV loop diuretics (bolus or infusion). Each hospital encounter was paired with the lowest spot-U_Na value collected within 4 hours post diuretic (median time 1.75 hours [IQR 0.73-2.7 hours]). Frequency of use was calculated by dividing the number of U_Na over no. of encounters that year. A threshold of U_Na50 (<50 mmol/L) and U_Na70 (<70 mmol/L) was used to categorize groups and 90-day mortality measured. Kaplan–Meier curves were compared between low and non-low U_Na groups and multivariable Cox models were used, adjusted for age, sex, race, and ethnicity.
Results. There were 11.4 M HF patients (median age 72 y [IQR 62-82]; 52 % male, 76.2% White, 16.0% Black, 2.0% Asian; 5.2% Hispanic, 88.9% non-Hispanic). 4.9 M had at least one hospital admission with a total of 9.5 M admissions receiving IV loop diuretic. U_Na was measured in 192,119 encounters, rising from 1.0% of admissions in 2005 to 2.8 % in 2025 (p < 0.001) (Fig 1). Low-U_Na50 occurred in 51,378 encounters (26.7 %) with 90-day mortality of 29.5 % versus 19.2 % for ≥50 mmol/L (adjusted HR 1.64, 95 % CI1.60–1.67)(Fig 2A). Low-U_Na70 occurred in 82,042 encounters (42.7 %), with 90-day mortality 26.8 % vs 18.4 % for ≥70 mmol/L (adjusted HR 1.50, 95 % CI 1.47–1.53) (Fig 2B).
Conclusions. Despite current guideline recommendations, U_Na utilization remains low in ADHF patients. Both < 50 mmol/L and < 70 mmol/L thresholds predict 90-day mortality, but later cut-off could flag additional high-risk patients without substantive loss of specificity.