BACKGROUND: Hyponatraemia is a prognostic marker of increased mortality and morbidity in selected groups of hospitalised patients. The aim of the present study was to examine the prevalence and prognostic significance of hyponatraemia at hospital admission in an unselected population with a broad spectrum of medical and surgical diagnoses.

METHODS: Consecutive patients >40 years of age admitted to a general district hospital in Greater Copenhagen between 1 April 1998 and 31 March 1999. Median follow-up time was 5.16 years (range 0-4372 days). Plasma sodium measurements were available in 2960 patients, and hyponatraemia defined as P-Na(+) <137 mmol/L at hospital admission was present in 1105 (37.3 %) patients.

RESULTS: One-year mortality was higher for hyponatraemic patients than for normonatraemic patients: 27.5% versus 17.7%. Moreover, hyponatraemia was an independent predictor of short and long-term all-cause mortality after 1 year and after the entire observation period respectively: hazard ratio (HR) 1.6 (95 % confidence interval (CI) 1.4-1.9, P < 0.0001) and HR 1.4 (95 % CI 1.3-1.6, P < 0.0001). Patients with hyponatraemia had longer hospitalisations than patients with normonatraemia: 7.6 (±0.38) days vs 5.6 (±0.21) days, P < 0.001. There was no interaction between hyponatraemia at admission and any admission diagnoses (P > 0.05 for all interaction analyses).

CONCLUSION: Hyponatraemia is associated with increased all-cause mortality and longer admission length independently of diagnosis and clinical variables.