A secondary outcome analysis of a randomized trial using a mixed lipid emulsion containing fish oil in infants with extremely low birth weight: Cognitive and behavioral outcome at preschool age

Objective: To evaluate the impact of a parenteral lipid emulsion containing fish oil compared with a soybean oil based-lipid emulsion on the cognitive outcome and behavior of preschool children with extremely low birth weight. Study design: This was a retrospective secondary outcome analysis of a randomized controlled trial performed between June 2012 and June 2015. Infants with extremely low birth weight received either a mixed (soybean oil, medium chain triglycerides, olive oil, fish oil) or a soybean oil-based lipid emulsion for parenteral nutrition. Data from the Kaufman Assessment Battery for Children II, the Child Behavior Checklist 1.5-5, and anthropometry were collected from medical charts at 5.6 years of age. Results: At discharge, 206 of the 230 study participants were eligible. At 5 years 6 months of age, data of 153 of 206 infants (74%) were available for analysis. There were no significant differences in Kaufman Assessment Battery for Children II scores for Sequential/Gsm, Simultaneous/Gv, Learning/Glr, and Mental Processing Index (mixed lipid: median, 97.5 [IQR, 23.5]; soybean oil: median, 96 [IQR, 19.5]; P =.43) or Child Behavior Checklist 1.5-5 scores for internalizing problems, externalizing problems, or total problems (mixed lipid: median, 37 [IQR, 12.3]; soybean oil: median, 37 [IQR, 13.5]; P =.54). Conclusions: A RandomForest machine learning regression analysis did not show an effect of type of lipid emulsion on cognitive and behavioral outcome. Parenteral nutrition using a mixed lipid emulsion containing fish oil did not affect neurodevelopment and had no impact on child behavior of infants with extremely low birth weights at preschool age. Trial registration: ClinicalTrials.gov: NCT01585935.