Enteral water for hypernatremia and intestinal morbidity in infants less than or equal to 1000 g birth weight.

OBJECTIVE

To evaluate the relationship between enteral water infusion for hypernatremia and significant intestinal morbidity in infants <or=1000 grams.

STUDY

DESIGN

This is a retrospective study of 321 infants <or=1000 g birth weight. Infants were grouped by the highest serum sodium (mmol/l) during the first 14 days of life as follows: <150 (normal control), >or=150 (high sodium control), >or=150 and treated with sterile water (study group). Significant intestinal morbidity was defined as probable or proven necrotizing enterocolitis or spontaneous intestinal perforation. Statistical analysis included Student's t test for continuous variables and chi(2) with Yeats correction for frequency variables. Multivariate logistic regression analysis was then performed to evaluate confounding variables among groups.

RESULTS

The incidence of intestinal morbidity was significantly higher in the high sodium-water treated group compared to each of the other groups (13/33 (38%) for high sodium-water versus 16/100 (16%) for high sodium control and 18/188 (10%) for normal sodium control, P<0.01 chi(2)). Logistic regression analysis indicated that enteral water and hydrocortisone were risk factors for significant intestinal morbidity.

CONCLUSIONS

Enteral sterile water for hypernatremia appears to be associated with significant intestinal morbidity in infants <or=1000 g. Hydrocortisone is also a risk factor.