Hyperglycemia has been associated with poor outcome in children with head injuries and burns. However, there has not been a correlation noted between hyperglycemia and infections in severely injured children. The trauma registry of a Level I trauma center was queried for injured children <13 years admitted between July 1, 1999 and August 31, 2003. The records of severely injured children [Injury Severity Score (ISS) > 15] were examined for survival, age, weight, ISS, infection, length of stay (LOS), and maximum glucose levels within the first 24 hours of injury (D1G). Statistical analysis was performed using a t test, Fisher's exact test, a Mann-Whitney Rank Sum test, or Kendall's Tau where appropriate. Eight hundred and eighty eight children under 13 years of age were admitted. One hundred and nine had an ISS > 15, and 57 survived to discharge with measured D1G. Patients excluded were those who died in less than 72 hours or had an LOS less than 72 hours. The survivors were divided into high glucose (> or =130 mg/dL; n = 48) and normal glucose (<130 mg/dL; n = 9). There was no difference between the groups with respect to age, weight, incidence of head injury, and ISS. An elevated D1G correlated with an increased risk of infection (P = 0.05) and an increased LOS (P = 0.01). These data suggest that severely injured children are often hyperglycemic in the first 24 hours after injury. Hyperglycemia in this study population correlated with an increased incidence of infection and increased length of stay. This suggests that strict control of hyperglycemia in injured children may be beneficial.