Background and aims The first large RCT on tight-glucose-control (TGC) to age-adjusted normoglycemia in the pediatric-intensive-care-unit (PICU) (Vlasselaers 2009) revealed that TGC reduced PICU morbidity and mortality as compared with usual-care, but increased hypoglycemia ≤40 mg/dL. As both hyper- and hypoglycemia may adversely affect the developing brain, an assessment of long-term neurocognitive function was required to exclude harm and validate any short-term benefit of TGC.
Methods Follow-up of all 700 patients included in the original RCT, was performed 4 years after randomization. Death or disability precluding neurocognitive testing were a priori defined as poor outcomes. The primary endpoint was full-scale IQ, assessed with age-adjusted intelligence-tests (Wechsler-IQ-scales). Neurodevelopmental-testing also encompassed a neurological examination, and tests for visual-motor-integration (VMI-Beery-Buktenica-Developmental-Test), attention and executive functions (ANT-Amsterdam-Neuropsychological-Tasks), memory (Children’s-Memory-Scale), and behavior (Child-Behavior-CheckList). For comparison, 216 healthy siblings and unrelated children were tested.
Results At follow-up, TGC in PICU had not increased the incidence of poor outcomes [19% vs.18%, univariable OR for poor outcome with TGC 1.10 (0.76–1.62), P=0.6]. Sixteen percent of the 700 ICU patients declined participation or were not contactable. TGC did not affect full-scale IQ [median 88.0 (IQR 74.0–100.0) vs. 88.5 (74.3–99.0), P=0.7], nor other scores for intelligence, visual-motor-integration, memory and behavior. TGC actually improved motor coordination (all P≤0.03) and cognitive flexibility (P=0.02), the latter up to the level of healthy children. Imputation for missing values confirmed these results. Hypoglycemia evoked by TGC did not negatively affect neurocognitive outcome.
Conclusion Despite hypoglycemia, TGC in PICU did not harm neurocognitive development 4 years later.