Hypernatremia in the intensive care unit: An indicator of quality of care?

Objective: To assess the frequency of hypernatremia in patients who were ad mitted to an intensive care unit (ICU) and to determine the correlation of hypernatremia with the clinical outcomes, durations of the patients' stays in the ICU, and other clinical variables. Design: Retrospective survey. Setting: University teaching hospital. Patients: All patients (total, 389) who were admitted to the medical ICU of the department of internal medicine during 1 yr. Measurements: The database of our hospital's mainframe computer was searche d for sodium values greater than or equal to 150 mmol/L that were registere d in the year 1995. These data were then matched with the registration numb ers of all patients who were admitted to our medical ICU between January 1 and December 31, 1995. In this way, we identified all patients in whom hype rnatremia was present at admission or those who developed hypernatremia in the course of their stay in our ICU. The prevalence and duration of hyperna tremia (defined as a serum sodium concentration of greater than or equal to 150 mmol/L or more) were determined; the correlation of hypernatremia with clinical outcome, duration of ICU stay, Acute Physiology and Chronic Healt h Evaluation II scores, and other clinical variables were evaluated; and ch anges in fluid administration in response to hypernatremia and fluid regime ns in the period preceding hypernatremia were examined. Main Results: Of a total of 389 patients who were admitted in 1995, hyperna tremia was present at admission in 34 patients (8.9%). The average duration of hypernatremia in these patients was 16.2 (range, 4-56) hrs. A total of 22 patients (5.7%) developed hypernatremia in the course of their stay in t he ICU. The average duration of hypernatremia in this group was 34.7 (range , 4-89) hrs. Moderately elevated levels of sodium had been detected in most of these patients (n = 21) in the days before the development of severe hy pernatremia; however, adjustments in fluid infusion aimed at preventing the occurrence of hypernatremia were either lacking (n = 7) or inadequate (n = 11). Hospital-acquired hypernatremia vs. hypernatremia present at admissio n to the ICU was associated with a higher mortality rate (32% vs. 20.3%, re spectively; p < .01). Conclusions: Despite frequent measurement of sodium levels in patients in t he ICU, hypernatremia is a relatively common occurrence. Initial treatment of hypernatremia is often inadequate, and sometimes treatment is delayed. T he development of hypernatremia is associated with adverse outcomes for pat ients developing hypernatremia in the ICU. Hypernatremia could potentially be used as an indicator of quality of care in the medical ICU.