In order to establish the biochemical basis for dietary interventions, we investigated the fatty acid composition of plasma lipid classes in patients with inactive inflammatory bowel disease. In this cross-sectional study thirty patients with ulcerative colitis (UC), twenty-one with Crohn disease (CD) and twenty-four controls were investigated (mean age: UC, 40·8 (sd 12·1); CD, 37·6 (sd 11·0); control, 31·5 (sd 8·4) years). Fatty acid composition of plasma lipids was determined by high-resolution capillary GLC. In plasma phospholipids, significantly higher values of eicosapentaenoic (20 : 5n-3), docosapentaenoic (22 : 5n-3) and γ-linolenic (18 : 3n-6) acids were found in control patients and patients with UC as compared to patients with CD [median % (weight by weight), control v. UC v. CD : 20 : 5n-3, 0·09 (interquartile range (IQR) 0·05) v. 0·14 (IQR 0·10) v. 0·16 (IQR 0·10), P < 0·05; 22 : 5n-3, 0·14 (IQR 0·10) v. 0·27 (IQR 0·16) v. 0·31 (IQR 0·10), P < 0·001; 18 : 3n-6, 0·02 (IQR 0·02) v. 0·03 (IQR 0·02) v. 0·05 (IQR 0·03), P < 0·05]. When compared to the control, values of the principal n-3 and n-6 long-chain PUFA, arachidonic acid (20 : 4n-6) and DHA (22 : 6n-3) were significantly higher in patients with UC but not in patients with CD [median % (w/w), UC v. control: 20 : 4n-6, 8·43 (IQR 3·23) v. 6·92 (IQR 2·96), P < 0·05; 22 : 6n-3, 1·22 (IQR 0·56) v. 0·73 (IQR 0·39), P < 0·05]. As seen there are considerable differences between the long-chain PUFA status of patients suffering from UC or CD. The data obtained in the present study do not support the concept of eicosapentaenoic acid or DHA deficiency in patients with either UC or CD.