Barmat 2005.

Methods	RCT, multi‐centre (4 US centres), open‐label, parallel design
Participants	80 women undergoing IVF/ICSI Inclusion criteria: < 39 years of age, day 3 FSH level of < 10, E2 level of < 60 pg/mL, AFC > 5 with a menstrual cycle range of 26 ‐ 34 days, and no more than one previous failed IVF or IVF/ICSI cycle. BMI 19 ‐ 32 kg/m2 no hydrosalpinx present by hysterosalpingogram, laparoscopy, or ultrasound within the past year Male factor infertility cases could be included (ICSI and/or frozen sperm) with the exception of nonobstructive azoospermia. Only one study cycle was allowed Exclusion criteria: history of previous poor response (< 4 follicles and/or an E2 level of < 500 pg/mL on the day of hCG), had taken infertility medications (clomiphene and/or gonadotrophins) within the past month, or had failed to consent to taking OCs, GnRH‐analogues, or gonadotrophins.
Interventions	GnRH antagonist (n = 40): OC (Desogen; Organon USA) on cycle days 2 to 4 for 14 to 28 days + 300 IU/day rFSH SC (adjusted) + 250 µg ganirelix was initiated when a lead follicle obtained a mean diameter of 12 to 14 mm (flexible) GnRH agonist (n= 40): leuprolide (GnRH‐agonist group), 0.5 mg per day during the mid‐luteal phase with approximately a 5‐day overlap with the OCs. Once adequate pituitary desensitisation was achieved the dose of GnRH agonist was reduced to 0.25 mg per day + 300 IU/day rFSH SC (adjusted) Oocyte maturation triggering: at follicular diameter 16 ‐ 18 mm, 5000 to 10,000 IU of hCG (Pregnyl). In cases at risk of ovarian hyperstimulation syndrome, the physician could give a dose of 5000 IU of hCG Oocyte retrieval: 35 to 36 hours later Embryo transfer: at 3 or 5 days Luteal phase support: progesterone, one centre treated women with P, 25 mg IM, on the day of retrieval, followed by P, 50 mg IM daily, with some women being supplemented with hCG 2,500 IU on days 3 and 6 after retrieval. The other centres prescribed luteal support with a daily dose of P (50 mg IM).
Outcomes	Participants to oocyte retrieval (n = 77) Days from OCP to oocyte retrieval Days on OC Stimulation day 1 E2 (pg/mL) Recombinant FSH (IU) Days of recombinant FSH Stimulation day of ganirelix start Days of leuprolide or ganirelix LH day hCG (IU/L) E2 day hCG (pg/mL) P4 day hCG (pg/mL) No of follicles Follicle sizes Number of oocytes retrieved Number of mature oocytes Number of 2 pronuclear embryos Number of embryos transferred Percentage of women with cryopreservation Embryos cryopreserved/participant with cryopreservation Number of pregnancies/embryos transferred (%) Number of pregnancies/cycle started (%) Number of ongoing pregnancies/embryos transferred (%) Number of ongoing pregnancies/cycle started (%) Number of implanted embryos (%) Number of ongoing twin gestations (%) Delivered pregnancies
Notes	Women who continued to have elevated E2 levels (> 60 pg/mL) and a cyst were removed from the study. If the E2 level was < 60 pg/mL and the cyst was still present, it could be aspirated and the participant would remain enrolled in the study and begin their recombinant FSH administration on Friday, along with a reduction of the GnRH agonist dose to 0.25 mg per day Women who had a serum E2 level of > 60 pg/mL or a cyst > 20 mm were continued on the same leuprolide dose for another week In women randomised to the GnRH‐antagonist group who had an E2 level of < 60 pg/mL, they could begin recombinant FSH on that Friday (5th day after OC). If they had a cyst > 20 mm, they were withdrawn from the study Number of participants at randomisation: 80 (ganirelix: 40/leuprolide acetate: 40) Number of participants at stimulation: 79 (ganirelix: 38/leuprolide acetate: 41) Number of participants at OPU: 77 (ganirelix: 36/leuprolide acetate: 41)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Dark sealed envelopes (true), randomisation: 1:1 (ganirelix acetate: leuprolide acetate) ratio
Allocation concealment (selection bias)	Low risk	Dark sealed envelopes (true)
Blinding (performance bias and detection bias) All outcomes	High risk	Not reported clearly
Incomplete outcome data (attrition bias) All outcomes	Low risk	No missing outcomes data
Selective reporting (reporting bias)	Low risk	The study protocol was not available, but it is clear that the published reports included most expected outcomes
Other bias	Low risk	The study appears to be free from other sources of bias