Eneroth 2004.

Study characteristics
Methods	Study design: prospective RCT Setting/location: diabetic foot‐care clinic at the Department of Internal Medicine, Lund University Hospital, Sweden Study duration: 6 months
Participants	Sample size: n = 53 participants Inclusion criteria: aged ≥ 60 years; diabetes mellitus; Wagner grade 1 or 2 foot ulcer of at least four weeks' duration; distal blood pressure must have been measured in the previous 3 months; agreed to participate in the study. Exclusion criteria: active chronic inflammatory intestinal disease; malignancy; immunosuppressive treatment; decreased kidney function; severe heart disease; psychiatric, addictive or any other disorder compromising the patient’s ability to participate in the study or to give truly informed consent Participant characteristics: men: 40 (75%); women: 13 (25%) median age: 74 years (min 59 years‐max 88 years) duration of diabetes: median 15 years (min 1 year‐max 51 years) Nutritional status: no difference between the groups at baseline; the supplement was given randomly irrespective of the presence or absence of protein‐energy malnutrition at inclusion wound duration (weeks): intervention: 25 (4–100) control: 22 (4–105); P = 0.9 baseline wound size (median wound size in cm²): intervention: 1.2 control: 2.2
Interventions	Intervention: 20 g protein per 200 mL bottle, 1 kcal/mL, nutritional supplement with added vitamins, minerals and trace elements (n = 26); taken orally, once daily Control: 400 mL placebo daily (n = 27); taken orally, once daily
Outcomes	Complete wound healing: 12/26 intervention; 10/27 placebo; P > 0.05 Amputation: 2/26 intervention, none reported in the placebo group Death: 1/26 intervention and 1/27 control
Notes	This study was sponsored by a grant from Nutricia AB, Netherlands
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	High risk	Statistically significant differences between the groups at baseline for the following parameters: palpable pulses: dorsalis pedis or tibialis: control: 10/27 (37%); placebo: 3/23 (13%); P = 0.05 PEM present: control: 12/27 (44%); intervention: 5/26 (19%); P = 0.05
Allocation concealment (selection bias)	Unclear risk	Not stated
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Quote: "each set of packages used for one patient had a number on it, indicating whether it was a placebo or intervention. Neither the patient nor the individual responsible for handing out the package knew what the numbers represented."
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Quote "the participants and investigator were blinded to treatment until the trial end."
Incomplete outcome data (attrition bias) All outcomes	Low risk	Wound healing outcomes were provided for the participants who dropped out of the study
Selective reporting (reporting bias)	Low risk	Reported: complete wound healing, amputation and death
Other bias	Low risk	None detected