Background: Tight blood glucose control with intravenous insulin reduces morbidity and mortality in adult surgical intensive care patients. This has never been investigated in premature infants weighing </=1500 g. We investigate the relationship between blood glucose levels repeatedly elevated >150 mg/dL and median blood glucose levels in the first week of life on one hand, and morbidity and mortality in premature infants weighing </=1500 g on the other.
Patients and methods: The following data were collected from 252 premature infants weighing </=1500 g at birth: demographic data, blood glucose levels at three set times during the day (capillary and arterial) in the first week of life, actual and relative elevation of blood glucose level >/=150 mg/dL, median blood glucose level, allocation of patients into groups according to number of elevated blood glucose levels >/=150 mg/dL (0, 1-3 or >/=4 incidents), and median blood sugar level in relation to mortality and morbidity like IVH, ROP, and sepsis.
Results: A significant increase in mortality (P<0.0001) with increasing median blood glucose level and repeated (>/=4) incidents of blood glucose levels >/=150 mg/dL and in infants with low gestational age (<27 weeks) were observed. There was no correlation between blood glucose level and morbidity.
Conclusion: Premature infants with low gestational age (<27 weeks), elevated median blood glucose levels and/or repeatedly elevated blood glucose levels >/=150 mg/dL have a significantly increased mortality. However, further prospective studies considering the gestational age should determine the relationship between tight glucose control and mortality.