Cholestatic liver disease is a common complication associated with long term parenteral nutrition (PN). PN associated liver disease (PNALD) is much more common in premature infants and the incidence increases with duration of PN. The use of intravenous omega-3 long chain polyunsaturated fatty acids (ω3PUFA) or fish oil has recently shown promise in the treatment of PNALD. We hypothesize that there are early markers for PNALD that precede the increase in total and direct bilirubin. We further hypothesize that patients with PNALD who receive enteral ω3PUFA supplementation will have an improvement in PNALD or reversal of PNALD. These hypotheses will be tested by a two part study that includes an initial observation period when markers for PNALD are evaluated, followed by a randomized, controlled trial of enteral ω3PUFA supplementation for treatment of PNALD. Infants expected to be on PN for 4 weeks or longer will be eligible for enrollment in this study. The observational part of the study will entail periodic assessment of potential markers for PNALD. Markers will be evaluated for inflammatory cytokines (IL-1, IL-6, TNF-alpha), oxidative stress (8-isoprostane, 8-hydroxydeoxyguanosine, glutathione peroxidase), liver fibrosis (TIMP-1), endogenous steroid production (glucagon and ACTH), total serum bile acids, essential fatty acid profiles, and calprotectin, a novel marker of gut inflammation. Patients will be observed for 6 months duration. Patients enrolled in the study who develop PNALD will be randomized to either the current standard of care (control group) or enteral ω3PUFA supplementation (treatment group). Once able to take oral medications, treatment group patients will receive enteral ω3PUFA 1 g/kg/day for 12 weeks. At the end of the 12 weeks, the protocol will be open label in which any patients who continue to have PNALD in either group will receive enteral ω3PUFA. All patients enrolled in the study (whether or not they develop PNALD or receive ω3PUFA supplementation) will be followed for a total of 6 months. The results of this study will increase our knowledge of the pathogenesis of PNALD, as well as potentially confirm the effectiveness of a novel therapy for this costly and debilitating disease.

Research Design and Methods Study Design. This is a two part study that includes an initial observation period when markers for PNALD are evaluated, followed by a randomized, controlled trial of enteral ω3PUFA supplementation for treatment of PNALD. Patients that have not developed PNALD will be eligible for the first part of the study evaluating early markers and can then continue with part two of the study if PNALD develops. If a patient is transferred to a study site and has already developed PNALD, the patient will not be eligible for part one, but may be included in part two of the study. The investigators will be responsible for receiving informed consent for enrollment into the study protocol from the patient's parent or legal guardian.

The first part of this study will be observational to assess early markers leading to PNALD. As a part of the research study, subjects will have baseline and serial evaluations (every 2 weeks for the first month followed by monthly evaluations for 6 months) of inflammatory cytokines (IL-1, IL-6, TNF-alpha), oxidative stress markers (8-isoprostane, 8-hydroxydeoxyguanosine, glutathione peroxidase), markers for liver fibrosis (TIMP-1), markers for endogenous steroid production (glucagon and ACTH), and total serum bile acids, essential fatty acid profiles, and calprotectin (see Appendix III). Other laboratory results that are drawn per standard of care for all PN patients, including chemistry panels, complete blood counts, c-reactive protein, and direct bilirubin, will also be recorded. All blood samples collected for the study will be scheduled with standard of care blood draws, so there are no additional needle sticks for study participants. Nutrition data including amount of protein, lipids, and dextrose provided by PN, and amount of enteral feeding will be recorded weekly for the duration of the study.

When a subject develops PNALD (three consecutive direct bilirubin concentrations > 2 mg/dL), and is able to tolerate at least trophic feeds (1 mL Q12h), the subject will be randomized to either the control group (our current hospital practice including advancement of enteral feeding, ursodiol, and cyclic PN, but not ω3PUFA) or the active treatment group (current hospital practice plus the addition of enteral ω3PUFA supplementation of 1 g/kg/day with a maximum dose of 4 g/day for a 12 week period). All patients will be started at 1 g/kg/day with a maximum dose of 4 g/day. The dose of 1 g/kg/day is the same dose used by investigators at Children's Hospital of Boston in the investigations of intravenous ω3PUFA. They also used a maximum dose of 4 g/day for patients weighing ≥4 kg. There will not be any dose titration. Doses may be divided so that patients only receive 1 g per dose. Therefore if a patient weighs 3 kg they will receive 1 g three times daily. The maximum frequency will be 4 times daily. All ω3PUFA will be supplied as Lovaza (give specifics on manufacturer) 1 g liquid filled capsules. Therefore a patient dose will be 1 capsule per kg weight with a maximum of 4 capsules per day. As patients grow, the Lovaza will be maintained at 1 g/kg/day rounded to the nearest capsule and not to exceed 4 g/day. In order to administer the dose the nurse will withdraw the liquid from a Lovaza l g capsule and give to the child via feeding tube or orally if the child does not have a feeding tube placed.

Randomization will take place in the Le Bonheur Pharmacy Department based on a blocked list of random numbers (i.e., patients receiving odd enrollment numbers will be randomized to one group and those with even numbers will be randomized to the alternative group). At the end of the 12 week period, the study will become open label, whereby control group subjects who continue to have PNALD will receive enteral ω3PUFA supplementation of 1 g/kg/day. These patients will also have a maximum dose of 4 g/day if they weigh ≥ 4 kg. There will be no dose titration. ω3PUFA supplementation will continue until PNALD resolves (direct bilirubin < 2 mg/dL) or it is determined by the investigators that the patient is not receiving any benefit from ω3PUFA supplementation. There will be no additional blood draws for this phase of the study.

All patients will be followed in the study for 6 months or until discontinuation of PN, or discharge from the hospital without the development of PNALD. If any study patient with PNALD is discharged from the hospital during the study period, the patient will be encouraged to continue in the study and receive Lovaza at home and have the remaining monthly study evaluations in the the affiliated Pediatric Gastroenterology Clinic.

Each study evaluation will include laboratory monitoring where an additional 2 mL of blood will be drawn in conjunction with routine clinically necessary laboratory tests. This blood draw should take less than 5 minutes. All other study information can be gathered from clinically needed patient assessments (i.e. weight, feeding history, medications, PN formulation) and should not require additional time for study involvement.

Inclusion Criteria:

• Neonates / infants < 1 year of age (there is no minimum age for enrollment and subjects may be male or female)

• Enrolled prior to the development of PNALD

• Anticipated duration of PN of 4 weeks or greater including patients with:

  • Short bowel syndrome resulting from surgical management of NEC, congenital bowel defects (omphalocele and gastroschisis), intestinal atresias, midgut volvulus, and other intestinal processes
  • Functional short bowel syndrome

• Subjects must be deemed clinically stable with a life expectancy of at least 6 months before enrollment

Exclusion Criteria:

• Bleeding risk (platelets count < 50 thousand units/μL)

• Receiving aspirin or other anticoagulation agent

• Patient's who are deemed clinically unstable:

  • Severe multi-system disease
  • Genetic disorders
  • DNR