Hypernatremia in the neurologic intensive care unit: how high is too high?

Abstract

Hypernatremia is associated with increased mortality in hospitalized patients and in medical/surgical intensive care units. This relationship has not been studied in neurologic/neurosurgical intensive care units (NNICUs), where hypernatremia is often a component of treatment of cerebral edema. We performed a retrospective analysis of prospectively collected data in patients admitted to the NNICU over a 6.5-year period. Hypernatremia (serum sodium >150 mEq/L) was seen in 339 patients (7.9%) and was more common (24.3%) in patients who were treated with mannitol. Hypernatremic patients had a lower median admission Glasgow Coma Scale score (8 vs 14, P < .001), higher initial Acute Physiology and Chronic Health Evaluation II probability of death (34.9% vs 19.1%, P < .001), higher incidence of mechanical ventilation (80.5% vs 41.1.5%, P < .001), higher mortality (30.1% vs 10.2%, P < .001), and higher incidence of renal failure (10.3% vs 0.9%, P < .001). Mortality increased with increasing hypernatremia; however, only severe hypernatremia (serum sodium >160 mEq/L) was independently associated with increased mortality. Other factors independently associated with mortality were age, mechanical ventilation, initial Acute Physiology and Chronic Health Evaluation II probability of death or low admission Glasgow Coma Scale score, and a diagnosis of cerebrovascular disease. In conclusion, hypernatremia is common in the NNICU, more so in patients treated with mannitol. In this population, severe (but not mild or moderate) hypernatremia is independently associated with increased mortality.