BackgroundIntravenous lipids are critical to the care of extremely premature and other high-risk infants. ObjectiveThe study evaluated safety and efficacy of parenteral nutrition (PN) with composite intravenous lipid emulsion (CO-ILE) with fish oil compared to pure soybean oil lipid emulsion (SOLE). MethodsRandomized, controlled, double-blind, multicenter study (NCT02579265) in neonates/ infants anticipated to require ≥28 days of PN due to gastrointestinal malformations or injury. Duration of the initial and extended treatment phase was 28 days and 84 days (for patients with PN indication after day 28). Results83/ 78 patients (mean postnatal age: 11.4/ 8.3 days, 54/ 59 preterm) received CO-ILE and SOLE, respectively. 33 patients per group completed 28 days on treatment. Risk of having conjugated bilirubin values > 2 mg/dL confirmed by a second sample 7 days after the first during the initial treatment phase (primary outcome) was 2.4% (2 of 83) with CO-ILE and 3.8% (3 of 78) with SOLE (risk ratio 0.59 [95% CI: 0.09, 3.76]). Between days 29 and 84, the number of patients with confirmed conjugated bilirubin values > 2 mg/dL did not increase in the CO-ILE group (n=2) and increased in the SOLE group (n=9). At the end of the initial treatment phase, conjugated bilirubin concentrations were 45.6% lower under CO-ILE than under SOLE (p=0.006). There was no clinical or laboratory evidence of essential fatty acid deficiency in patients in the CO-ILE group. Median time to discharge alive was 56.7 and 66.4 days with CO-ILE and SOLE, respectively (hazard ratio: 1.16; 95% CI: 0.81, 1.68). ConclusionsCO-ILE was associated with a possible lower risk of cholestasis and significantly lower conjugated bilirubin at the end of the initial treatment phase in high-risk neonates and infants as compared to patients treated with SOLE. In summary, these data indicate that CO-ILE can be considered safe and may be preferable over SOLE in high-risk neonates.Clinical Trial Registry number: Clinicaltrials.gov, study ID NCT02579265