Neurocognitive development of children 4 years after critical illness and treatment with tight glucose control: a randomized controlled trial

Abstract

Context: A large randomized controlled trial revealed that tight glucose control (TGC) to age-adjusted normoglycemia (50-80 mg/dL at age <1 year and 70-100 mg/dL at age 1-16 years) reduced intensive care morbidity and mortality compared with usual care (UC), but increased hypoglycemia (≤40 mg/dL) (25% vs 1%).

Objective: As both hyperglycemia and hypoglycemia may adversely affect the developing brain, long-term follow-up was required to exclude harm and validate short-term benefits of TGC.

Design, setting, and patients: A prospective, randomized controlled trial of 700 patients aged 16 years or younger who were admitted to the pediatric intensive care unit (ICU) of the University Hospitals in Leuven, Belgium, between October 2004 and December 2007. Follow-up was scheduled after 3 years with infants assessed at 4 years old between August 2008 and January 2012. Assessment was performed blinded for treatment allocation, in-hospital (83%) or at home/school (17%). For comparison, 216 healthy siblings and unrelated children were tested.

Main outcome measures: Intelligence (full-scale intelligence quotient [IQ]), as assessed with age-adjusted tests (Wechsler IQ scales). Further neurodevelopmental testing encompassed tests for visual-motor integration (Beery-Buktenica Developmental Test of Visual-Motor Integration); attention, motor coordination, and executive functions (Amsterdam Neuropsychological Tasks); memory (Children's Memory Scale); and behavior (Child Behavior Checklist).

Results: Sixteen percent of patients declined participation or could not be reached (n = 113), resulting in 569 patients being alive and testable at follow-up. At a median (interquartile range [IQR]) of 3.9 (3.8-4.1) years after randomization, TGC in the ICU did not affect full-scale IQ score (median [IQR], 88.0 [74.0-100.0] vs 88.5 [74.3-99.0] for UC; P = .73) and had not increased incidence of poor outcomes (death or severe disability precluding neurocognitive testing: 19% [68/349] vs 18% [63/351] with UC; risk-adjusted odds ratio, 0.93; 95% CI, 0.60-1.46; P = .72). Other scores for intelligence, visual-motor integration, and memory also did not differ between groups. Tight glucose control improved motor coordination (9% [95% CI, 0%-18%] to 20% [95% CI, 5%-35%] better, all P ≤ .03) and cognitive flexibility (19% [95% CI, 5%-33%] better, P = .02). Brief hypoglycemia evoked by TGC was not associated with worse neurocognitive outcome.

Conclusion: At follow-up, children who had been treated with TGC during an ICU admission did not have a worse measure of intelligence than those who had received UC.

Trial registration: clinicaltrials.gov Identifier NCT00214916.