To compare the incidence of cholestasis in very low birth weight infants receiving cycled versus continuous parenteral nutrition, and to determine factors that predispose to parenteral nutrition–associated cholestasis (PNAC).

Preterm infants weighing ≤1250 g (n = 70) at birth were randomly assigned within the first 5 postnatal days to either cycle (n = 34) or continuous (n = 36) parenteral nutrition. Liver function tests were obtained at baseline, and sequentially thereafter. Cholestasis was defined as direct bilirubin >2 mg/dL. Infants with major congenital anomalies, congenital hepatic disease, clinically apparent congenital viral infection, and those who required major abdominal surgery were excluded.

The incidence of PNAC was similar in the 2 groups (cycle 32% vs continuous 31%; P = 1.0). Bilirubin and transaminases were similar in both groups by repeated measures of ANOVA. Gestational age, birth weight, and Apgar scores were significantly lower, and Clinical Risk Index for Babies II scores were significantly higher in infants who developed PNAC. Using backward selection logistic regression, bronchopulmonary dysplasia, duration of parenteral nutrition, and days to full enteral nutrition emerged as factors independently associated with PNAC.

Early prophylactic parenteral nutrition cycling in very low birth weight infants in this study did not reduce cholestasis. Time to full feedings is a significant predictor for PNAC in very low birth weight infants. Preterm infants with bronchopulmonary dysplasia are more likely to have PNAC as a comorbidity. The Clinical Risk Index for Babies II score may help identify those preterm infants who might benefit from future prospective prevention trials.