Research News

Cognitive function is unaffected by tight glucose control in paediatric intensive care

BMJ 2012; 345 doi: (Published 24 October 2012) Cite this as: BMJ 2012;345:e7065

In 2009, researchers completed a randomised trial that tested tight glucose control against usual care for children admitted to one paediatric intensive care unit in Belgium. Children treated to achieve lower targets of blood glucose concentration had fewer infections and a shorter stay on the unit than controls, but they also had significantly more episodes of hypoglycaemia. These extra episodes of hypoglycaemia had no discernible effect on children’s intelligence four years later, according to the latest follow-up study.

Children treated to achieve lower targets in the original trial did no worse than controls in a battery of neurocognitive tests at a mean age of 5 years. The two groups had comparable mortality and rates of disability. Hypoglycaemia was not associated with worse neurocognitive performance in fully adjusted observational analyses combining both groups. Most of the 469 children in this follow-up report joined the original trial after cardiac surgery for congenital defects.

These findings are reassuring, says a linked editorial (doi:10.1001/jama.2012.14151). Tight glucose control did no harm to children in the medium term, despite a high incidence of brief hypoglycaemia. We still don’t know, however, whether tight control does any good in this setting. The evidence from published trials is conflicting. More trials are already under way.

In the meantime, we should also investigate why children in the new report were significantly less intelligent at follow-up than age matched peers who had never been admitted to intensive care (average IQ 88 v 103; mean difference in fully adjusted analyses, 9 points, 95% CI 6 to 12).


Cite this as: BMJ 2012;345:e7065