Dysnatremias and survival in adult burn patients: a retrospective analysis.

BACKGROUND/AIMS Dysnatremias have been evaluated in many populations and have been found to be significantly associated with mortality. However, this relationship has not been well described in the burn population. METHODS Admissions to the burn center at our institution from January 2003 to December 2008 were examined. Independent variables included gender, age, percentage total body surface area burned (%TBSA), percentage of third-degree burn, inhalation injury, injury severity score (ISS), Acute Kidney Injury Network (AKIN) stage, hypernatremia, and hyponatremia. They were examined via Cox proportional hazard regression models against death. Moderate to severe hypo- and hypernatremia were defined as serum sodium <130 and >150 mmol/l, respectively. RESULTS In 1,969 subjects with a mean age of 36.3 ± 16.4 years, a median %TBSA of 9 (interquartile range 4-20) and a median ISS of 5 (interquartile range 1-16) hypernatremia occurred in 9.9% (n = 194), while hyponatremia occurred in 6.8% (n = 134) with mortality rates of 33.5 and 13.8%, respectively. Patients without a dysnatremia had a mortality rate of 4.3%. On Cox proportional hazard regression age, %TBSA, ISS, and AKIN stage were found to be significant predictors of mortality. Hypernatremia (HR 2.00, 95% CI 1.212-3.31; p = 0.0066), but not hyponatremia (HR 1.72, 95% CI 0.89-3.34; p = 0.1068) was associated with mortality. CONCLUSIONS In the burn population, hypernatremia, but not hyponatremia, is an independent predictor of mortality.