Administration of Amino Acids, Not Glucose or Lipids, to Critically Ill Children Explains Harm By Early Parenteral Nutrition

Presentation Number: OR30-6
Date of Presentation: April 4th, 2017

Ilse Vanhorebeek*1, Sascha C Verbruggen2, Michaël P Casaer1, Jan Gunst1, Pieter Wouters1, Jan Hanot1, Gonzalo G Guerra3, Dirk Vlasselaers1, Koen F Joosten2 and Greet Van den Berghe1
1KU Leuven, Leuven, Belgium, 2Erasmus Medical Centre, Rotterdam, Netherlands, 3University of Alberta, Stollery Children's Hospital, Edmonton, AB, Canada


Background: Large randomized controlled trials have shown harm from early parenteral nutrition in adult and pediatric intensive care units (ICUs), with increased risk of acquiring an infection in ICU, delayed weaning from mechanical ventilation and prolonged need of intensive care. Overdosing of energy with too little protein was suggested as potential reason.

Research question: This study analyzed which macronutrient statistically explains the harm by early parenteral nutrition in the critically ill children included in the “Pediatric Early versus Late Parenteral Nutrition In Critical Illness (PEPaNIC)” trial [1].

Methods: Doses of glucose, lipid and protein/amino acids, administered during the first 7 days, were expressed as percentages of doses recommended for age/weight. Independent associations between average doses up to each of these days and likelihood of acquiring an infection in the pediatric ICU (PICU), of earlier live weaning from mechanical ventilation and of earlier live PICU discharge were investigated through multivariable Cox proportional hazard analyses, adjusted for risk factors.

Results: Administration of amino acids explained harm by early parenteral nutrition. With increasing doses, the likelihood of acquiring an infection was higher (adjusted hazard ratios (HRs) per 10% of recommended dose between 1.064-1.136 for day 1-5, P≤0.02), while that of earlier live weaning from mechanical ventilation (HRs 0.939-0.977 day 2-7, P≤0.01) and earlier live PICU discharge (HRs 0.936-0.970 day 1-7, P≤0.01) was lower. In contrast, more glucose during the first 3 days was independently associated with fewer infections (HRs 0.875-0.908, P≤0.008), whereas more lipids independently associated with earlier weaning (HRs 1.003-1.059, P≤0.03) and earlier PICU discharge (HRs 1.008-1.067, P≤0.008). Risk of harm with amino acids started with low doses, linearly increasing up to maximal toxicity with 40%-50% of recommended doses. Higher amino acid doses coincided with higher plasma urea concentrations throughout the first week in PICU.

Conclusions and Interpretation: Early administration of amino acids, but not glucose or lipid, explained harm by early parenteral nutrition in critically ill children. The suppressive effect of amino acids on autophagy as crucial pathway for innate immunity and damage removal, or the burden to liver and kidney of shuttling excessive amino acids to hepatic production and urinary excretion of urea, may have played a role.


Nothing to Disclose: IV, SCV, MPC, JG, PW, JH, GGG, DV, KFJ, GV