To determine the effect of early continuous insulin therapy on survival of premature infants, European investigators conducted a randomized controlled trial of continuous insulin infusion with concurrent 20% dextrose infusion compared with standard neonatal care for very-low-birth-weight (VLBW) infants.
Infants younger than 24 hours were enrolled if their birth weight was less than 1,500 grams and they required intensive care. While glucose levels were examined in all infants using continuous subcutaneous monitoring, clinical care was directed by standard glucose observations. The outcomes of interest were mortality and morbidity at the expected date of delivery as well as glucose control.
There were 195 infants in the early insulin (EI) group and 194 in the standard treatment (ST) group. Over the seven days of the study, the mean daily glucose levels in infants in the EI group were significantly lower than those in the ST group.
The proportion of infants with more than 10% of their glucose measurements >180 mg/dL was significantly lower in the EI group (21% vs 33%), but these infants also had significantly more documented episodes of hypoglycemia (glucose <47 mg/dL) (29% vs 17%). The increase in hypoglycemia was significant only for infants weighing >1 kg. Infants in the EI group received significantly more intravenous carbohydrates (51 vs 43 kcal/kg/day) but similar amounts of protein and lipids.
There was no significant difference in overall rate of mortality (14% for those in the EI vs 9% in the ST group; P=.2), but the mortality at 28 days of age was greater in the EI group (12% vs 6%; P=.04). There were no differences in rates of sepsis, necrotizing enterocolitis, retinopathy of prematurity, chronic lung disease, or intracranial disease. However, an increased incidence of brain parenchymal lesions in the EI group led to the suspension of the study.
The authors concluded that early insulin therapy was not beneficial in VLBW infants and was associated with hypoglycemia, which can be harmful to brain development in premature infants. They speculated that real time glucose measurements may allow safer glycemic control.
Dr. Bratton has disclosed no financial relationship relevant to this commentary. This commentary does not contain a discussion of an unapproved/investigative use of a commercial product/device.
This study of early insulin treatment for VLBW infants did not show a survival benefit, and a secondary mortality end point suggested that treatment with early insulin was associated with early mortality, as well as increased risk of hypoglycemia. These results again confirm that studies from adult patients often cannot be directly translated to infants and children.
The pathogenesis of hyperglycemia may differ for preterm infants compared to adults and older children. Infants have decreased insulin production in part due to decreased processing of proinsulin to insulin. They also do not suppress hepatic glucose production in response to glucose infusions, and have less muscle and adipose mass (thus...