[The impact of iatrogenic hypernatremia on the prognosis of critical patients].

OBJECTIVE: To explore the effect of iatrogenic hypernatremia on the prognosis of critical patients, and its risk degree.

METHODS: One thousand five hundred and sixty-eight critical patients of emergency intensive care unit (EICU) and surgical intensive care unit (SICU) of Chaoyang Hospital Affiliated to Capital Medical University from January 2002 to December 2005, were retrospectively analyzed. They were divided into iatrogenic hypernatremia group, non-iatrogenic hypernatremia group and non-hypernatremia group according to their serum sodium content (the definition of hypernatremia was serum sodium > or = 149 mmol/L). The age, gender, ICU length of stsy (LOS), the main reason of admission to the ICU, acute physiology and chronic health evaluation II (APACHE II) score in first 24 hours of all the patients were recorded. Cox Regression models were used for analysis in survivors.

RESULTS: Of 1 568 patients, 361 (23.0%) had hypernatremia, and iatrogenic hypernatremia accounted for 260 (16.6%). Patients in non-iatrogenic hypernatremia group had significantly increased APACHE II score (28.16+/-11.21) than iatrogenic hypernatremia group (17.55+/-14.96) and non-hypernatremia group (16.02+/-10.77), ICU LOS of iatrogenic hypernatremia group [(24.14+/-17.53) days] had significantly prolonged than non-iatrogenic hypernatremia group [(14.07+/-27.88) days] and non-hypernatremia group [(13.14+/- 10.53) days], 28-day mortality in non-iatrogenic hypernatremia group (47.52%) and iatrogenic hypernatremia group (42.31%) was significantly higher than that in non-hypernatremia group (33.64%, P < 0.05 or P < 0.01). In Cox Regression models for survival analysis, iatrogenic hypernatremia was an independent risk factor for mortality of critical patients [in univariate analysis: risk ratio 1.83, 95% confidence interval (CI) 1.27-3.96, P < 0.001; in multivariate analysis: risk ratio 1.55, 95%CI 1.17-3.10, P < 0.001].

CONCLUSION: Iatrogenic hypernatremia maybe an independent risk factor for mortality of critical patients and should be avoided and taken care of timely.