RESUMO
AIM: We evaluated a strict strategy that aimed to avoid fluctuations in glucose infusion rates (GIRs) and assessed the independent effects of maximal daily GIRs on the hyperglycaemia risk among extremely low birth weight (ELBW) infants receiving early enhanced parenteral nutrition. METHODS: This study comprised all ELBW infants admitted to the neonatal intensive care unit of Oslo University Hospital Rikshospitalet, Norway, before (2007-2009) and after (2012-2013) implementing a strict GIR strategy. Severe hyperglycaemia was defined as two consecutive blood glucose values over 12 mmol/L. Maximum daily GIRs (mg/kg/min) were categorised into low (<5.1), intermediate (5.1-7.0) or high (>7.0). Mixed effects logistic regression modelling for repeated measurements was applied to investigate independent determinants of hyperglycaemia. RESULTS: We included 1293 treatment days for 195 infants. The maximum daily GIR decreased (6.3 versus 5.8 mg/kg/min), while mean daily glucose and energy intakes were maintained in the post-strategy period. The prevalence of severe hyperglycaemia (48% versus 23%), insulin use (39% versus 16%) and mortality (26% versus 10%) fell. Intermediate GIR (odds ratio 2.11) and high GIR (odds ratio 2.85) were significant independent predictors of severe hyperglycaemia compared to low GIR. CONCLUSION: A strict GIR strategy reduced the risk of severe hyperglycaemia and adverse outcomes.