Patients presenting with severe hypotonic hyponatremia: etiological factors, assessment, and outcomes.

Abstract

Although hospital-acquired hyponatremia is well described, severe community-acquired hyponatremia has been studied less extensively.To assess characteristics and outcomes of patients admitted with severe hypotonic hyponatremia (SHH) (defined as serum sodium ≤ 120 mmol/L).All patients with serum sodium of ≤ 120 mmol/L who were admitted to 2 large teaching hospitals from January 2000 to August 2007 were identified, and data were obtained from medical records. Main outcome measures were incidence of osmotic demyelination and mortality.Two hundred fifty-five patients were admitted who had SHH (female to male ratio 2:1), and the mean age was 72 ± 14 years. The most common etiological factors were thiazide/indapamide diuretics (41%), syndrome of inappropriate antidiuretic hormone secretion (38%), and hypovolemia (24%). Inappropriately rapid correction of serum sodium (> 12 mmol/L over the first 24 hours) occurred in 37 patients (15%), with 4 patients (11%) developing osmotic demyelination. Patients who developed osmotic demyelination were more likely to be younger, abuse alcohol (3 of 4 patients), and have lower serum potassium levels. One patient had a hypoxic-anoxic episode at presentation. The patients also had a mean serum sodium increase in the first 24 and 48 hours of 21 ± 5 mmol/L and 28 ± 8 mmol/L, respectively. None of the patients with osmotic demyelination received hypertonic saline. None of the patients in whom the serum sodium increment was limited to ≤ 12 mmol/L developed osmotic demyelination. Overall, mortality was 10% and was not related to sodium level at presentation.Patients treated with thiazide or indapamide (particularly elderly women) may benefit from monitoring of serum sodium levels. Inappropriately rapid serum sodium correction is associated with osmotic demyelination, particularly in patients with risk factors for this condition. In contrast to what has been reported for hyponatremia in hospitalized patients, severity of hyponatremia on admission did not predict increased mortality in our patient population.