Deshpande 2014.

Methods	Design: single‐centre double‐blind RCT Setting: regional tertiary NICU of King Edward Memorial Hospital for Women, Perth, Australia Study enrolment: January 2010 to June 2011 I. Allocation concealment: yes II. Blinding of intervention: yes III. Blinding of outcome measurement(s): yes IV. Complete follow‐up: yes
Participants	Inclusion criteria: preterm neonates, < 30 weeks' gestation admitted to NICU requiring PN providing > 75% of energy expenditure requirements for 7 days and postnatal age < 7 days. Exclusion criteria: blood culture‐positive sepsis; thrombocytopenia (platelet count < 150 × 109 cells/L); unconjugated hyperbilirubinaemia (requiring exchange transfusion); metabolic disorders including lactic or uncompensated acidosis (or both); no parenteral consent; administration of intravenous lipid infusion before study; postnatal age > 7 days; bleeding disorder.
Interventions	34 infants were randomised, 30 initially and then 2 more in each arm Group 1 (n = 17): OS‐LE; 20% ClinOleic Group 2 (n = 17): MOFS‐LE; 20% SMOFlipid TPN protocol: dose protocol was day 1, 1 g/kg; day 2, 2 g/kg; day 3, 3 g/kg and days 4 to 7, 3 g/kg. Duration of study was 7 days, after which all of the participants received ClinOleic LE, which was the standard of practice in the nursery. Intravenous lipids were continued as long as PN support was deemed necessary by the attending neonatologist. The emulsions were dispensed in amber‐coloured coded syringes and amber‐coloured infusion lines suitable for infusion pumps and infused intravenously through a central or peripheral line.
Outcomes	Primary outcomes: levels of LC‐PUFA (mean and SD) in red cell membrane and lipid peroxidation status measured by plasma F2‐isoprostane levels (mean and SD) as picomole per litre. Secondary outcomes: weight, head circumference and length at birth at study entry, exit and at discharge; enteral vs PN proportion; number of episodes of blood culture‐positive sepsis; IVH; duration of hospital stay, mechanical ventilation and PN support; mortality and vitamin E levels.
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "The coordinating pharmacist who was not directly involved in patient care randomised neonates (by using a computer‐generated randomisation list)."
Allocation concealment (selection bias)	Low risk	Quote: "Pharmacist prepared coded ready‐to‐use syringes of either OO (20% ClinOleic Baxter, S.A. Belgium) or FO [fish oil] (20% SMOFlipid Fresenius Kabi, Pymble, Australia) lipid emulsion. Given this strategy and the identical appearance of the coded, ready‐to‐use identical syringes, the researcher and other team members were blinded to the allocation status and the content of syringes."
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Quote: "similar appearance of lipid emulsions with code broken after analysis;" "amber colored coded syringes." Comment: done
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Quote: "The data were analyzed without breaking the code to ensure masking of statistical analysers" and "similar appearance of lipid emulsions with code broken after analysis" " amber colored coded syringes." Comment: done
Incomplete outcome data (attrition bias) All outcomes	Low risk	Outcomes described for the whole group and ITT analysis performed.
Selective reporting (reporting bias)	Low risk	Quote: "Clinical trial registration number: ACTRN 12609001017213" Comment: not identified.
Other bias	Unclear risk	55 participants could not be approached on weekends due to resource limitations in the study. Funding: partial funding from Fresenius Kabi and Baxter Health Care for 2 similar studies. Though author mentions that both the companies had no influence on any aspect of the study.