Long-chain polyunsaturated fatty acids are important for the growth and early development of the central nervous system. Cholestatic infants suffer from fat malabsorption and disturbed lipid metabolism and therefore may be at risk of developing polyunsaturated fatty acid depletion. The aims of this study were to determine essential fatty acid status in cholestatic infants and to study the relationship to disease severity, degree of undernutrition, antioxidant status and mode of feeding. Twenty-four-hour dietary records were obtained in 34 cholestatic infants, and measurements were taken of skin fold thicknesses, bilirubin levels, activities of serum alanine aminotransferase, alkaline phosphatase, gamma-glutamyl transpeptidase, prothrombin time, serum concentrations of albumin, bile acids, total lipids, phospholipids, cholesterol, vitamins A and E, the fatty acid composition of plasma phospholipids and plasma lipid peroxides expressed as thiobarbiturate reactive substance (TBARS). Plasma phospholipid fatty acids and TBARS were also determined in 12 age-matched healthy control infants. The cholestatic patients had very low percentage values of phospholipid essential fatty acids, particularly linoleic acid ( 18:2omega-6, median 14.74% vs 20.76% in controls, p < 0.001) and its major metabolite arachidonic acid (20:4omega-6, 6.80 vs 7.87%, p=0.04). The patients' essential fatty acid depletion was reflected by increased levels of the non-essential fatty acids, Mead acid (20:3omega-9, 0.74 vs 0.21%, p < 0.001) and palmitoleic acid (16:1omega-7, 2.20 vs 0.43%, p < 0.001). Polyunsaturated fatty acid profiles did not differ between infants with biliary atresia (n=13) and those with intrahepatic cholestasis (n=21), or between 17 infants with severe malnutrition (all skin folds < 10th percentile) and mild malnutrition (at least two skin folds > 10th percentile). TBARS were significantly higher in cholestatic patients than in controls (2.74 vs 0.85 nmol ml(-1), p < 0.001) and correlated with direct (r=0.41, p=0.02) and total bilirubin. The daily dietary intake of linoleic acid (per 100 kcal) correlated with plasma phospholipid linoleic acid (r=0.38,p=0.037) and total omega-6 fatty acids (r=0.38,p=0.036). Breastfed cholestatic infants (n=6) had higher values of the omega-3 long-chain polyunsaturated fatty acids docosapentanoic acid (22:5omega-3, 0.47 vs 0.28%, p=0.0006) and docosahexanoic acid (22:6omega-3, 2.39 vs 1.73%, p=0.01) than formula-fed infants, while disease severity was similar in the two groups. In conclusion, cholestatic infants are at high risk of essential fatty acid depletion, which appears to be related to fat malabsorption, hepatic essential fatty metabolism, enhanced lipid peroxidation and dietary intake.