Objective: Cholestatic liver disease in premature infants has been linked to numerous etiologies however, the specific cause is yet to be determined. The objective of this study was to identify risk factors associated with cholestasis (direct bilirubin greater than 1.5 mg/dl) in premature infants.

Methods: A retrospective chart review was conducted on infants with birthweight between 500-1250 grams born at MetroHealth Medical Center from the years 1993-1995. A total of 195 charts were reviewed. Information about demographics, hospital diagnosis, and nutritional status (days on total parenteral nutrition-TPN, days without feeds-NPO, number of days til full feeds, day of first enteral feed) was obtained.

Results: Infants with and without cholestasis were compared. Thirty-six infants presented with cholestasis (18.5%). Significance was determined by the Mann Whitney U test. Infants with cholestasis: (1) spent a longer time on TPN,(2) spent more days without oral feeds, and (3) started feeds at a later day of life. After stratification patients receiving less than 28 days of TPN were older when oral feeds were initiated (6.9 +/- 4.1 vs10.9 +/- 5.7, p value 0.018) and were without enteral feeds for a longer period (7.3 +/- 5.0 vs 11.8+/- 7.4, p value 0.024). In those infants receiving more than 28 days of TPN, being NPO was associated with cholestasis (17.5 +/- 8.5 vs 30.7 +/- 15.3, p value 0.001) but the day of life enteral feeds was initiated was not significant. Gestational age was not a significant factor for the development of cholestasis in infants receiving TPN less than or greater than 28 days.

Conclusion: Infants with cholestasis: (1) spent more days on TPN, (2) spent more days without enteral feeds, and (3) started enteral feeds later in life. Those infants on TPN for less than 28 days with cholestasis spent more days without enteral feeds and started enteral feeds later in life. Infants on TPN greater than 28 days spent more days without enteral feeds but the day of life feeds was initiated was not significant.