Background: Prolonged use of parenteral nutrition (PN) in neonates can lead to parenteral nutrition-associated liver disease (PNALD), manifested by elevated direct bilirubin concentrations, and in some cases progressing to hepatic failure. When new potential means of preventing PNALD in the neonatal intensive care unit (NICU), such as Omegaven usage, are tested in clinical trials, the studies should enroll neonates at a very high risk of developing PNALD. However, it is not always clear, in the first days of life, which neonates are most likely to develop PNALD. Therefore, preparatory to devising studies of prophylaxis against PNALD, we conducted an evaluation of all NICU patients who received PN for >or=14 day, assessing their likelihood of developing PNALD.
Methods: We performed an historic cohort analysis of all neonates in the Intermountain Healthcare system, receiving PN for 14 days or more during their stay, with dates of birth between 1 January, 2002 and 30 June, 2006.
Results: During the 4(1/2)-year period, 9861 neonates were cared for in the Intermountain Healthcare NICUs. Of these, 9547 (96.8%) survived for at least 28 days, and of these 6543 (68.5%) received PN. Twenty-one percent (1366 patients) of those receiving PN, received it for >or=14 days. PNALD was ascertained in this group by a direct bilirubin >or=2.0 mg/dl. Neonates receiving PN for 14-28 days had a 14% incidence of PNALD, those receiving PN for 29-56 days had a 43% incidence, those receiving PN for 57-100 days had a 72% incidence and those receiving PN for >100 days had a 85% incidence. Groups of patients identifiable on the first day of life as having the highest risk of developing PNALD were birth weight <500 g (odds ratio (OR), 30.7), birth weight 500-749 g (OR, 13.1), gastrochisis (OR, 20.3) and jejunal atresia (OR, 24.0). Among 357 patients who developed PNALD, the highest direct bilirubin concentrations correlated with the highest serum alkaline phosphatase and transaminase concentrations. Deaths after 28 days were much more common in those with the highest direct bilirubin and transaminase concentrations (P<0.0001).
Conclusions: In the first days of life, certain NICU patients can be identified as being at very high risk for developing PNALD. These are patients <750 g birth weight, those with gastrochisis and those with jejunal atresia. We speculate that these groups would be reasonable subjects for including in a PNALD prophylaxis trial, testing new preventative strategies such as Omegaven usage.