Background: Hyperglycemia (HG) is common in extremely low birth weight (ELBW: <1000 g) infants who are conventionally treated by parenteral glucose restriction. This policy will limit possible energy intake in babies who have minimal energy reserves.

Methods: We performed a prospective trial of continuous insulin infusion versus glucose restriction (nested within a larger trial of aggressive nutrition - Arch Dis Child 1997; 77; F4-11) for HG, defined as plasma glucose > 200 mg/dL (>11 mmol/L) with glycosuria >+++ (>55 mmol/L). Babies were randomised on admission to graded insulin infusion (commencing at 0.05 iu/kg/hr) or control (glucose restriction).

Results: HG occurred in 42 (55%) of 77 babies receiving standard dextrose or dextrose-amino acid infusion. The insulin group (n=19) had a mean birth weight of 750 g and gestational age of 25.8 wk, compared to the control group (n=23) of 755 g and 25.8 wk respectively. There were non-significant differences in mean postnatal age and mean glucose intake at onset of HG. The mean (SD) glucose intakes during the period of HG were 8.2 (1.7) and 5.1 (1.2) mg/kg/min respectively for insulin group and controls (p<0.001). The mean(SD) duration of HG was 2.9 (1.6) and 7.8 (6.6) days for insulin and control groups (p<0.005). There were no cases of hypoglycemia (glucose <2.2 mmol/L or 40 mg/dL).

Conclusion: We conclude that insulin infusion is a safe means of treatment of HG in ELBW infants, allowing an improved provision of energy and shortening the duration of HG.