NEONATAL SOCIETY ABSTRACTS
Are asymptomatic low blood glucose levels in babies of less than 32 weeks gestation really damaging?
Presented at the Neonatal Society 2004 Autumn Meeting (programme).
Tin W (for the Northern Neonatal Nursing Initiative Trial Group)
Background: The blood glucose level of the preterm baby often varies widely in the first few days and weeks of life. An important observational study involving 543 preterm babies, published in 1988 (1) suggested that recurrent, asymptomatic, moderate hypoglycaemia (defined as a plasma glucose of <2·6 mmol/l) had a serious adverse effect on later motor and cognitive development, but no other study has, as yet, ever attempted to replicate this observation.
Aim: The aim of this study was to compare the neurodevelopmental outcome of babies who had frequent low blood glucose levels (<2·6 mmol/l) in the first 10 days of life with that of matched controls.
Methods: This study was a prospective observational study of all babies born before 32 weeks gestation in 1990 and 1991 to mothers resident in the north of England. The laboratory blood glucose level of every baby was measured daily for the first 10 days of life at a standard pre-specified time. The surviving children with a whole blood glucose of less than 2·6 mmol/l on three or more days were matched (for hospital of early care, gestation and birthweight) with a control who never had a documented blood glucose level this low. All were then seen and assessed at two years by a single assessor, blind to any knowledge of the laboratory results, using the Griffiths Mental Development Scale. The general development quotient as well as its sub-quotients were calculated, after correction for prematurity, using the version of the Griffiths Scale originally published in 1970.
Results: 48 of the 566 children born before 32 weeks gestation who survived to two years of age had a blood glucose of less than 2·6 mmol/l on three or more days in the first 10 days of life. One child with severe myotonic dystrophy was excluded from the analysis. The mean general development quotient (DQ) of the other 47 “index cases” was 108·0, compared to 108·2 in the 47 controls. The mean paired difference was -0·2 (95% CI -6·0 to 5·5). Four index children and three control children had disabling cerebral palsy. When the comparison was limited to the 21 matched pairs where the index child had a low blood glucose on at least four separate days, the mean paired difference was +1·1 (95% CI -7·1 to 9·3), and when limited to the seven matched pairs where the index child had a low blood glucose on five separate days, the mean paired difference was -1·9 (95% CI -26·3 to 22·5). The mean DQ of the 183 survivors who never had a low blood glucose was 107·7. Further analysis using all the laboratory blood glucose samples collected during the first 10 days of life did not alter these conclusions. Low blood glucose levels were common in the first five days of life, and less frequently seen after that, and the incidence of “hypoglycaemia” varied inversely with total daily fluid intake.
Conclusion: The definition of what constitutes functional hypoglycaemia remains elusive. The neurodevelopmental outcome in the babies who had what many would classify as an unacceptably low blood glucose level on three or more days in the first 10 days of life in this observational study did not differ from that of matched controls. It will be important to see whether any difference in cognitive ability, adaptive skill, or behaviour, emerges as these children become older.
1. Lucas A, Morley R, Cole TJ. Adverse neurodevelopmental outcome of moderate neonatal hypoglycaemia. BMJ 1988; 297: 1304–8. (See also 318: 194–5.)