To assess whether lipid infusion could be a risk factor for parenteral nutrition-associated cholestasis (PNAC) in low birth weight neonates, 22 newborns with cholestasis (29.8 +/- 1.6 weeks, 1298 +/- 217 g) were compared with 22 without cholestasis (29.5 +/- 1.7 weeks, 1286 +/- 363 g). The mean level of peak direct bilirubin for the cholestasis group was 4.6 mg/dl compared to 1.2 mg/dl for the noncholestasis group. A univariate analysis revealed that PNAC was significantly related to duration of fasting (p = 0.008) and parenteral nutrition (p < 0.0001), days of antibiotics use (p = 0.025), positive C-reactive protein (p = 0.018) or gastric culture (p = 0.018), and feeding intolerance (p < 0.0001). Total amino acid amount (p < 0.0001), total lipid amount (p < 0.0001), and average daily lipid amount (p = 0.002) were significantly higher in the cholestasis group than in the noncholestasis group. Conversely, prenatal administration of dexamethasone was a significant protective factor of PNAC (p = 0.008). Logistic regression analysis revealed that the cumulative amount of lipid infusion was an independent risk factor for PNAC (p = 0.041; OR 1.174; CI 1.007-1.369). We suggest that decreasing the cumulative load of amino acids and intralipids with early trophic feeding, control of infection, and prenatal administration of dexamethasone could possibly attenuate the severity of PNAC.