Ziebe 2013.

Study characteristics
Methods	Study design: multicentre RCT (14 different clinics) Country: study co‐ordinated in the Netherlands. Participants were recruited from clinics in Sweden and Denmark. Setting: fertility clinics Sponsorship source: supported by ORIGIO, Maløv, Denmark Trials registry number: NCT00565747 (prospectively registered)
Participants	Inclusion criteria: women aged 25 to 39 years, women who had a regular menstrual cycle of 21 to 35 days, women treated with a standard gonadotrophin‐releasing hormone(GnRH) agonist or antagonist protocol, women with 3 or more follicles with a diameter of >/= 14 mm on the day of human chorionic gonadotrophin (hCG) administration, including a leading follicle of >/= 17 mm Exclusion criteria: previous participation in the study, use of assisted hatching, use of non‐ejaculated sperm, medical conditions or genetic disorders prohibiting in vitro fertilisation(IVF)/ intracytoplasmic sperm injection (ICSI) or interfering with interpretation of results, use of investigational drugs within 30 days before oocyte retrieval, severe chronic disease of relevance for reproduction, and oocyte donation Number of participants randomised to each arm of the study: 654 women randomised to receive GM‐CSF‐supplemented culture medium, and 678 women randomised to the control culture medium Group differences: no significant between‐group differences in baseline characteristics Fresh or frozen cycle?: fresh IVF or ICSI?: IVF and ICSI Length of time exposed to intervention medium: from fertilisation through to embryo transfer Trade name and concentration of GM‐CSF in intervention medium: ORIGIO 2 ng/mL Trade name of control medium: EmbryoAssist Day of embryo transfer: day 3 Maxiumum number of embryos transferred: 2
Interventions	Intervention: GM‐CSF‐supplemented culture medium Control: standard IVF medium
Outcomes	Live‐birth rate Miscarriage rate Clinical pregnancy rate Multiple gestation rate Preterm birth rate Birth defect rate Aneuploidy rate Stillbirth rate
Notes	Email correspondence with all additional data outlined in Appendix 13.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"Randomisation (blocks of four) was computer‐generated individually for each clinic to maintain balance between the treatment groups at each site."
Allocation concealment (selection bias)	Low risk	Following email correspondence: "Randomization was performed by ORIGIO a/s, and based on a randomisation list per clinic generated automatically using www.randomization.com. Each study site received a list of study specific consecutive patient ID numbers (e.g. clinic 1: 01001, 01002, 01003, 01004 etc.) and a corresponding number of identically looking randomized bottles of test and control media individually labelled with the corresponding study specific patient ID numbers. On site the lowest number on the list was always allocated to any new patient recruited, at the time of informed consent signature. Therefore, both the clinician, embryologist and patient, was blinded to the treatment allocation. The master randomisation list was held by ORIGIO a/s. All data analysis was performed externally by a Clinical Research Organization (CRO). During the interim analysis the statistician was blinded at all times to the media, which were presented as Medium A and Medium B. For the final statistical analyses, the codes for blinding were broken after database lock and patient classification."
Blinding of participants and personnel (performance bias) All outcomes	Low risk	"Participants and investigators were blinded to treatment allocation. Study media were packaged unidentifiably and labelled only with the randomization number. For each new patient recruited, the lowest available randomization number was used."
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Following email correspondence: "The clinicians performing the ultrasound scans were blinded to the treatment allocation at all times."
Incomplete outcome data (attrition bias) All outcomes	High risk	All dropouts accounted for, however the reasons given were not included in the predefined exclusion criteria.
Selective reporting (reporting bias)	Low risk	The authors confirmed through email correspondence that all outcomes were published.
Other bias	High risk	There is a potential conflict of interest regarding the funding and administration of this study as it was granted by the manufacturer of the intervention culture medium. However, the study authors have been very forthcoming with information regarding the trial, and there has been transparency. We assessed this trial as at high risk of other bias because the data provided in private email correspondence differ from the published data.