Background and aims: Data in critically ill patients on the effect of intravenous lipid emulsions (LEs), containing omega-3 polyunsaturated fatty acids (PUFAs), in parenteral nutrition (PN) are scarce and conflicting. This study compared the effects of a four-oil LE (30% soybean oil, 30% medium-chain triglycerides, 25% olive oil and 15% fish oil (FO)) (SMOFlipid®) to those of a 100% soybean oil-based LE in critically ill adult intensive care unit (ICU) patients. Methods: In this double-blind, randomised study, patients (n ¼ 75) predicted to need PN for more than 5 days were randomised to receive either a four-oil LE (Study Group (SG)) or a 100% soybean oil LE (Control Group (CG)). Isocaloric, isonitrogenous PN was administered continuously for 5 days. FO was provided at a dose of 0.09e0.22 g/kg body weight. Measurements included biochemical parameters and sequential organ failure assessment (SOFA) score daily and plasma total phospholipid fatty acids (FAs) and cytokine levels on days 1, 3, 6. Days on mechanical ventilation, length of stay and mortality were also recorded. ANOVA was used to compare response variables between the two groups over the time and Pearson correlation was used to measure relationships between continuous variables. Results: 68 patients completed the study (n ¼ 35 SG, n ¼ 33 CG), with male predominance (66% SG, 56% CG). Average age was 60.8 ± 13.9 years (SG) versus 55.7 ± 14.8 (CG) (p ¼ 0.143). The majority were surgical admissions (85% SG versus 91% CG) followed by medical. Plasma phospholipid oleic acid (p ¼ 0.022) and alpha-linolenic acid (p<0.0005) increased in both groups. In the SG, plasma phospholipid EPA and DHA increased (both p<0.001), whereas the omega-6:omega-3 PUFA (n-6:n-3 PUFA) ratio decreased (p < 0.001). Aspartate aminotransferase (AST), alanine aminotransferase (ALT) and bilirubin decreased in both treatment groups. Considering only the change from day 1 to day 6 there was a bigger decrease in AST, ALT and bilirubin levels in the SG. Concentrations of TNF-a decreased from day 1 to day 6 in the SG, whereas they increased in the CG, but the change was not statistically significant (p ¼ 0.112). A significant negative correlation was found between EPA provision on day 3 and the SOFA score (r ¼ À0.4047, p ¼ 0.018). Days on mechanical ventilation (1.24 ± 0.83 days in SG versus 0.88 ± 1.63 days in CG, p ¼ 0.385) and ICU LOS (9.5 ± 7.09 days in SG versus 10.7 ± 7.6 days in CG, p ¼ 0.490) were not different between groups. Conclusion: PN containing a four-oil LE increased plasma EPA and DHA, decreased n-6:n-3 PUFA ratio, and was safe and well tolerated. The negative relationship between day 3 EPA and SOFA score seems promising, but EPA intake and effects may have been diluted by enteral nutrition which was started in more than half of patients on day 4. There was no significant difference in terms of other biochemical