Early versus late parenteral nutrition in critically ill, term neonates: a preplanned secondary subgroup analysis of the PEPaNIC multicentre, randomised controlled trial.

BACKGROUND: Previous randomised studies showed that withholding parenteral nutrition for 1 week of critical illness was superior to early initiation (<24-48 h) of parenteral nutrition in children and adults. However, neonates are considered more susceptible to macronutrient deficits. We investigated the effect of withholding parenteral nutrition for 1 week in critically ill, term neonates.

METHODS: We previously did a randomised, controlled study (PEPaNIC) of children aged up to 17 years admitted to paediatric intensive-care units (ICUs) in three hospitals in Belgium, Canada, and the Netherlands randomly assigned (1:1) to either standard care of parenteral nutrition initiated early within 24 h of admission to an ICU or late parenteral nutrition (where supplemental parenteral nutrition was withheld for 1 week after admission to the ICU). In this preplanned, secondary subanalysis of PEPaNIC, we looked at data from critically ill, term neonate participants (gestational age ≥37 weeks) aged up to 28 days (studied in overlapping age groups of ≤4 weeks, ≤1 week, and <1 day-ie, age at admission). In both the early parenteral nutrition and late parenteral nutrition groups, enteral nutrition was initiated as soon as possible and increased according to local protocols. Outcome assessors and investigators not directly involved in the paediatric ICU were not informed of treatment allocation. The primary endpoints were incidence of new infections and duration of paediatric ICU dependency (quantified as the number of days in the paediatric ICU and likelihood of earlier live discharge from the ICU), analysed based on intention to treat. Multivariable analyses were adjusted for the following risk factors: centre, Paediatric Logistic Organ Dysfunction score, Paediatric Index of Mortality 2 score, diagnosis group, and weight-for-age Z scores on admission. Secondary safety outcomes were mortality (at 90 days, during the intervention, in the paediatric ICU, and in the hospital) and hypoglycaemic incidents during the intervention. All patients in the respective groups were included in the safety analysis.

FINDINGS: Between June 18, 2012, and July 27, 2015, we included 209 participants in this substudy, 145 of whom were aged up to and including 1 week and 45 aged younger than 1 day. In neonates aged up to and including 4 weeks, late parenteral nutrition increased the likelihood of earlier live discharge from the paediatric ICU compared with early parenteral nutrition (adjusted hazard ratio [HR] 1·61, 95% CI 1·19-2·20; p=0·0021) but did not affect the risk of infection. The risk of infection in neonates aged up to and including 1 week and younger than 1 day was lower with late parenteral nutrition than with early parenteral nutrition (adjusted odds ratios [OR] 0·36, 95% CI 0·15-0·83, p=0·017; and 0·10, 0·01-0·64, p=0·015, respectively). For neonates aged up to and including 1 week, the likelihood of an earlier live discharge from the ICU was higher with late parenteral nutrition (adjusted HR 1·69, 95% CI 1·16-2·46; p=0·0063). For neonates younger than 1 day, adjusted HR was 1·95 (95% CI 0·93-4·12; p=0·078). Mortality at all studied timepoints was similar between the groups for all ages; however, in neonates aged up to and including 4 weeks and aged up to and including 1 week, the risk of hypoglycaemia was higher with late parenteral nutrition (23% vs 14%; adjusted OR 3·05, 95% CI 1·27-7·35, p=0·013; and 24% vs 14%; 3·57, 1·23-10·45, p=0·019, respectively.

INTERPRETATION: In critically ill, term neonates, withholding parenteral nutrition for 1 week was clinically superior to standard care of initiating parenteral nutrition within 24 h for short-term outcomes. However, withholding parenteral nutrition for 1 week significantly increased the risk of developing hypoglycaemia, which necessitates long-term follow-up of these children before late parenteral nutrition can be confidently recommended for this vulnerable patient group.

FUNDING: Flemish Agency for Innovation through Science and Technology, Methusalem-Programme Flemish Government, European Research Council, Fonds NutsOhra, Stichting Agis-Zorginnovatie, and the Sophia Research-Foundation.