Johansson 2015.

Methods	RCT; parallel‐group (phase III). Undertaken in a single centre in Denmark
Participants	Patients > 18 years scheduled to undergo elective CABG surgery who were not anaemic according to WHO criteria were included in the study. Data extracted by authors from primary study for iron‐deficient patients (defined as ferritin < 100 mcg/L or 100‐300 mcg/L where TSAT < 20%)
Interventions	Single dose of study drug infused prior to surgery (iron isomaltoside, 20 mg/kg) or equivalent volume of placebo (0.9% sodium chloride)
Outcomes	Primary outcome was a lower decrease in Hb after cardiac surgery. Secondary end points were requirement for allogeneic blood transfusion, iron‐related parameters and safety endpoints.
Study funding arrangements	Study was funded by Pharmacosmos A/S.
Author conflicts of interest	Single (senior) study author is an employee of Pharmacosmos A/S. No further conflicts of interest were declared.
Sample size	60 participants were randomised 1:1 to receive the intervention (n = 30) or the control (n = 30) treatment.
Notes	Iron deficiency was not considered an inclusion criteria for this study, but we contacted study authors and extracted data for participants who were iron deficient as per the inclusion criteria for this review. Study was registered with a public trials register (NCT 01563367)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "Permuted block randomisation with a block size of 4 was used to randomise the patients. The randomisation list was prepared centrally by a contract research organization, Max Neeman International Data Management Centre, using a validated computer program (Statistical Analysis Software [SAS] 9.1.3; SAS Institute Inc., Cary, NC) PROC PLAN procedure. An interactive web response system method was used to randomise the eligible patient to the treatment groups. When the patient data had been entered into the interactive web response system, a unique randomisation number was generated for the patient, identifying which treatment the patient was allocated to."
Allocation concealment (selection bias)	Low risk	Quote: "The screening and enrolment of the patients were performed by the investigator at the site, whereas the entering of the patient data into the interactive web response system generating the randomisation number was typically performed by the trial nurse or trial coordinator."
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Quote: "Patients in the iron isomaltoside 1000 group received iron isomaltoside 1000 as a single‐dose infusion of 1000 mg over 15 min with a maximum single dose of 20 mg/kg. Patients in the placebo group received saline (Natriumklo‐ rid 9 mg/ml; Fresenius Kabi, Copenhagen, Denmark) as a single‐dose infusion of 100 ml over 15 min." Comment: no comment is made in the article on attempts to conceal the characteristic colour of the iron infusion.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Comment: no discussion made on involvement of investigators with outcome assessment and administration of presumably unblinded study drug.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Comment: data for NAID participants was provided in full by study authors on request. All outcomes were fully reported.
Selective reporting (reporting bias)	Low risk	Comment: Data for NAID participants was provided in full by study authors on request. All outcomes were fully reported.
Other bias	High risk	Comment: study was conceived and run by Pharmacosmos A/S.