Insulin therapy is not associated with improved clinical outcomes in critically ill infants with stress hyperglycemia.

The aim of the present study was to examine the benefits of insulin use and non-use in critically ill infants with stress-induced hyperglycemia. The present retrospective study used clinical data from 302 critically ill infants with stress hyperglycemia admitted to pediatric intensive care units (PICUs). The patients were recruited randomly and divided into three groups: The tight glycemic control, conventional insulin therapy and control groups. Correlations between insulin therapy and improved clinical outcomes were assessed according to key parameters (length of PICU stay, total length of stay, occurrence of organ dysfunction and mortality). Correlations between blood glucose level and these parameters in the three groups were also examined. Blood glucose levels following insulin therapy were not correlated with the length of PICU stay, total length of stay, mortality, secondary coma, or secondary hepatic or renal dysfunction in the three groups. At 96 h following PICU admission, blood glucose levels were statistically similar (5.0±1.2, 4.9±1.3 and 5.1±0.9 mmol/l, respectively; P>0.05). Insulin therapy was revealed to have no benefit on the length of hospitalization, the occurrence of organ dysfunction or mortality in critically ill pediatric patients with stress hyperglycemia. Even with no insulin use, the blood glucose level could spontaneously return to normal, with no associated risk of organ dysfunction or fatality.