Thirty-six adult severe head injury and cerebral stroke patients in four intensive-care units were randomized to receive one of three enteral diets for 21 days. These diets, which supplied 45% of calories from fat, differed only in lipid composition. Diet A was comprised of 100% soybean oil, diet B contained a 50:50 (wt/wt) mixture of soybean oil and medium-chain triglycerides (MCTs), and diet C contained 42.5% MCT, 50% soybean oil, and 7.5% blackcurrant seed oils. Plasma phosphatidylcholine and fatty acid composition of plasma total phospholipids were determined before initiating treatment (day 0) and weekly throughout the study. Results indicated that at the start of the study, all patients had low linoleic acid (18:2 omega 6) levels compared with healthy subjects. Emulsion A disturbed the balance between several fatty acids of the omega 6 series, as exemplified by the significant increase in 18:2 omega 6 proportions. In contrast, both emulsions B and C introduced a less-pronounced rise in 18:2 omega 6 associated for emulsion C with a significant increase in dihomo-gamma-linolenic acid (20:3 omega 6) and docosapentaenoic acid (22:5 omega 3) in plasma phospholipids. Furthermore, 18:3 omega 6 change was significantly different between groups A and C and that of 20:3 omega 6 between group A and both groups B and C. Throughout the study, arachidonic acid (20:4 omega 6) exhibited remarkable steady-state levels regardless of the diet. This study shows that providing the injured body with high amounts of 18:2 omega 6 does not lead to high levels of its upper derivatives in plasma phospholipids.(ABSTRACT TRUNCATED AT 250 WORDS)