P2D325: Reduced Dose Soybean Lipid Emulsion Results in a Decreased Rate of Direct Bilirubin Increase Without Adversely Affecting Growth

Background: Parenteral nutrition-associated cholestasis (PNAC) is one of the most serious morbidities resulting from the administration of parenteral nutrition (PN) and can progress to end-stage liver disease. The etiology of PNAC appears to be multifactorial, but the lipid component of PN, particularly soybean lipid emulsion (SLE), has been implicated due to several mechanisms. While fish oil-based and composite lipid formulations are associated with a decreased risk of cholestasis, availability has been negatively affected by supply chain challenges. Protocols using traditional SLE, therefore, continue to be required. Use of a dose reduction strategy of SLE has been shown to prevent PNAC. Concerns about this strategy relate to adverse effects on growth and neurodevelopment. The purpose of this study was to evaluate the efficacy of a reduced dose SLE for the prevention of PNAC as well as effects on growth.

Methods: Surgical neonates at 4 centers were randomized to receive standard SLE (3 gm/kg/day; group A) or reduced dose SLE (1 gm/kg/day; group B) over a 12 week study period. Bilirubin data was collected at baseline and then weekly until 7 days after PN was discontinued. Growth parameters, including Z-scores for weight and height, were measured weekly. A linear mixed effects model was used to estimate the effects of time and group on direct bilirubin levels and growth parameters.

Results: Twenty-one patients were randomized (group A=9, group B=12). Two patients in group A were administratively withdrawn from the study. Over the study time, the mean IV lipid dose in group A was 2.456 gm/kg/day and 0.968 gm/kg/day in group B which was significantly different (p=0.001). The mean total PN calories was not different between the two groups (87.2 kcal/kg/day vs 68.4 kcal/kg/day, p=0.157). In group 1, 3 patients developed PNAC and 1 patient developed severe PNAC. In group B, 2 patients developed PNAC and none developed severe PNAC. There was no difference in the prevalence of PNAC (p=0.617) or severe PNAC (p=0.617). The rate of bilirubin increase was slower in group B and began to decrease earlier at day 27 versus day 63. The rate of bilirubin decrease in group B was greater although this was not statistically significant (p=0.07 at day 84) (Figure 1). No infants in either group developed essential fatty acid deficiency. There was no significant difference in the change in weight (p=0.352 at day 84) or height Z-score (p=0.11 at day 84) over time between the two groups (Figure 2).

Conclusion: Neonates randomized to a reduced dose of SLE had improved trends in bilirubin levels compared to those randomized to a standard dosing regimen This benefit was seen without adverse effects on overall growth. The use of reduced dose SLE remains a reasonable approach to long term PN use in surgical neonates.