Bauer 2000.

Methods	Randomised clinical trial (blocks of 10), France
Participants	120 hospitalised adults admitted to the ICU for more than 2 days, at nutritional risk due to being in the ICU Male:Female = 82:38 Mean age: 54 years Exclusion criteria: elective surgery or presenting a contraindication to enteral or parenteral support, or both, having a previous history of allergy to vitamins
Interventions	Experimental group: received parenteral nutrition. Treatment consisted of a 3‐in‐1 solution of carbohydrates, fat, and protein, Vitrimix KV and hydrosoluble vitamins, Soluvit. (n = 60)Control group: received placebo. Treatment consisted of sodium chloride 0.9% with Intralipid 20% (50 ml/l) and Soluvit (10 ml/l), stable for 24 hrs Treatment and placebo were administered in the same type of plastic bags (1 ± 2 l), at a concentration of 1 kcal/ml in the treatment group. The solution was administered through a central line (960 mOSm/l) that was not inserted solely for nutritional purposes. The rate of intravenous administration was increased to 120 ml/hr for 18 ± 24 hrs. (n = 60) Co‐intervention: both groups received enteral support: Participants were bolus‐fed every 4 hrs, 5 times a day with a standard, noncommercial, modular polymeric diet. The composition of the solution was protein (20%), polyunsaturated fats (30%), carbohydrates (50%), non‐soluble fibres, sodium chloride (2 g/l), potassium chloride (3 g/l), and a standard solution of hydro‐ and lipo‐soluble vitamins; the concentration of the solution was 1 kcal/ml. A typical 70‐kg participant would receive 100 ml initially, with an increased amount in 50‐ml steps to a maximum of 350 ml every 4 hrs 5 times a day.
Outcomes	Levels of retinol‐binding protein and prealbumin, morbidity, mortality, cost
Study dates	Not stated
Notes	No contact information could be obtained.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Not described
Allocation concealment (selection bias)	Unclear risk	The envelopes were described as sealed but it was uncertain if the envelopes were opaque.
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Neither the healthcare providers nor the participants were aware of the treatment given.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Although the statistician was blinded to the allocation of treatment until all events had occurred, it is not stated clearly who performed the outcome assessment.
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	6/60 early dropouts in the experimental group and 7/60 in the control group They stated that they used intention‐to‐treat analysis, but did not fully describe how they dealt with missing participants.
Selective reporting (reporting bias)	Low risk	The trial reported all‐cause mortality and serious adverse events. No protocol could be found.
For‐profit bias	Unclear risk	It was unclear how the trial was funded.
Other bias	Low risk	The trial appeared to be free of other components that could put it at risk of bias.