Garrel 2003.

Methods	Design: randomised controlled trial. Multi‐centre or single‐centre: single‐centre. Period: July 1998 to May 2001. Sample size calculation: not reported. Generation of allocation: randomisation by random numbers tables stratified by TBSA: 2% to 40%/40% to 60%/60% to 80%. Allocation concealment: Investigator was blinded to randomisation sequence. Blinded assessment of treatment allocation: not reported. Withdrawals: 45 participants were randomly assigned; 3 died in the first 72 hours and 1 could not receive enteral feedings so did not receive glutamine. These 4 participants were included only in the mortality analysis and were not included for other outcomes. Intention‐to‐treat analysis: used for mortality analysis but not for other outcomes. Follow‐up: not reported
Participants	45 recruited; 45 analysed for mortality data and 41 for other outcomes How many enter the study on each arm? 45 participants: 4 excluded (2 control, 2 glutamine); control 22 (21 M,1 F); glutamine 19 (16 M, 3 F) How many finish the study on each arm? 41 participants: control 22 (21 M,1 F); glutamine 19 (16 M,3 F) Mean age: control 38 years; glutamine 39 years Mean total burn surface area (TBSA): control 42%; glutamine 40% Inclusion criteria: 45 adult patients admitted to the burn centre within 24 hours of thermal injury who sustained 2nd‐ or 3rd‐degree burns to at least 20% but less than 80% of total body surface area Exclusion criteria: age > 65; total body surface area burn > 80%; pregnancy; chronic illness: respiratory, cardiac or renal insufficiency; cancer
Interventions	Before 48 hours of admission by nasoenteral tube: Experimental (n = 19): glutamine (Cambridge Neutraceutical, Boston, MA) 4.3 grams 4 times a day (26 grams/d) Control (n = 22): isonitrogenous mixture of aspartic acid, asparagine and glycine All participants received nasoenteral feeding by the following formula: 15% fat, 20% protein, 65% carbohydrate (Sandosource, Novartis, Minneapolis, MN) with no immunonutrient. Total energy requirements were calculated using the Curreri formula and were adjusted for measurement of energy expenditure by indirect calorimetry
Outcomes	Overall mortality: ITT, per‐protocol LOS: “length of care”: time required for participant's wounds to heal (grafted and ungrafted) Positive blood culture: unspecified definition Number of participants with gram‐negative bacteria Wound infection: wound swab
Notes	Other outcomes (biochemical): phagocytosis by blood polymorphonuclear cells, IL‐6, IL2Ra, serum glutamine Erratum (2004 reference) taken into count. Data extracted from 2003 article Number of ventilated participants, number of days with mechanical ventilation, number of participants with LIS < 5, number of days with LIS < 5, LOC days, LOC days/TBSA, number of deaths (ITT), number of blood cultures per participant, number of participants with PBC, number of participants with PBC + Pseudomonas aeruginosa, number of participants with > 2 days of PBC, number of PBC days/participant, % of study time with > 3 antibiotics, number of participants with wound infection
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Random numbers tables stratified by burn severity
Allocation concealment (selection bias)	Low risk	Randomization sequence concealed
Blinding (performance bias and detection bias) All outcomes	Unclear risk	Blinding not stated in Methods section
Incomplete outcome data (attrition bias) All outcomes	Low risk	Intention‐to‐treat analysis when possible
Other bias	Low risk