Provision of a soy-based intravenous lipid emulsion at 1 g/kg/d does not prevent cholestasis in neonates.

BACKGROUND: One of the most common and severe complications of long-term parenteral nutrition (PN) is PN-associated cholestasis. The soybean oil-based lipid emulsion administered with PN has been associated with cholestasis, leading to an interest in lipid reduction strategies. The purpose of this study was to determine whether the provision of a soybean oil-based lipid emulsion at 1 g/kg/d compared with 2-3 g/kg/d is associated with a reduced incidence of cholestasis.
METHODS: Retrospective review of neonates admitted between 2007 and 2011 with a gastrointestinal condition necessitating ≥ 21 days of PN support. Neonates were divided into 2 groups based on the intravenous lipid emulsion dose: 1-g group (1 g/kg/d) and 2- to 3-g group (2-3 g/kg/d). The primary outcome measure was the incidence of cholestasis.
RESULTS: Sixty-one patients met inclusion criteria (n = 29, 1-g group; n = 32, 2- to 3-g group). The 2 groups did not differ in any baseline characteristics other than associated comorbidities that were more common in the 2- to 3-g group. The duration of PN, the number of operative procedures and bloodstream infections, and enteral nutrition (EN) were similar between groups. The incidence of cholestasis was not different between groups (51.7%, 1-g group; 43.8%, 2- to 3-g group; P = .61), and there was no difference between groups in the time to cholestasis (32.6 ± 24.1 days, 1-g group; 27.7 ± 10.6 days, 2- to 3-g group; P = .48). Overall, 44.8% of patients with cholestasis were transitioned to full EN, and 55.2% were transitioned to a fish oil-based lipid emulsion after which the direct bilirubin normalized in all patients.
CONCLUSION: Lipid reduction to 1 g/kg/d does not prevent or delay the onset of cholestasis in neonates.