Background: Hyponatremia (serum sodium concentration <135 mEq/L) is the most common electrolyte abnormality among hospitalized patients. Our aim was to study the epidemiology of hyponatremia in hospitalized patients, as well as the short-term mortality rates, the length of stay (LOS), and associated hospital costs.
Methods: This retrospective cohort study included 6,539 hospitalizations in the internal medicine ward of a Swiss tertiary-care teaching hospital between January 1, 2012, and December 31, 2018 (42.7% women, mean age 69 years). Using serum sodium concentration, we identified hospitalizations with hyponatremia and calculated the prevalence of overall hyponatremia, admission hyponatremia (AH), hospital-acquired hyponatremia (HAH), and persistent hyponatremia (PH) at discharge. We also studied the impact of hyponatremia on 30-day readmissions, in-hospital and 30-day mortality, and hospital LOS and costs, using multivariable logistic regression and Cox proportional hazards models, with normal natremia as reference.
Results: Prevalence of overall hyponatremia was 32.5% [95% confidence interval (CI), 31.3–33.6%], while prevalence of PH among hospitalizations with AH and HAH was 33.7% (31.7–35.8%). After multivariable adjustment, hyponatremia was associated with increased hospital costs (CHF 19,025 ± 485 vs. 14,962 ± 341, p < 0.001) and
LOS (13.4 ± 0.2 vs. 10.7 ± 0.2 days, p < 0.001). Increased severity of hyponatremia was associated with higher hospital costs and LOS (p for trend <0.001). There was a trend toward more frequent 30-day readmissions associated with hyponatremia [adjusted odds ratio (OR), 1.15 (1.01–1.31), p = 0.032], mainly with PH: adjusted OR = 1.41 (1.17–1.71), p < 0.001. No association was found between severity of hyponatremia and readmissions. Hyponatremia was associated with an increase of in-hospital [adjusted OR = 1.94 (1.49–2.53), p < 0.001] and 30-day mortality: adjusted OR = 1.80 (1.44–2.24), p < 0.001. Increased severity of hyponatremia was associated with higher in-hospital and 30-day mortality (p for trend < 0.001).
Conclusions: Hyponatremia is highly prevalent among hospitalized patients and associated with an increase of LOS, early hospital readmission, in-hospital and 30-day mortality, and hospital costs. PH was associated with a substantial increase of the risk of early hospital readmission and 30-day mortality.