Franco Jr 2003.

Methods	Parallel group study. Number of women recruited: 22. Number of women randomized: 22 (16 in intervention group; 6 in control group). Number of withdrawals: 1 (both in intervention group, due to spontaneous pregnancies). Number of women analyzed: 20.
Participants	Country of authors: Brazil. Inclusion criteria: women without specific ovulatory dysfunction, aged ≤ 37 years, who would be submitted to ovarian stimulation. Exclusion criteria: not reported. Mean age ± SD: intervention group: 32.2 ± 2.1 years; control group: 31.8 ± 1.9 years.
Interventions	Intervention: oestradiol valerate (4 mg/day) for 14 days, started cycle day 21 + rFSH (150‐300 IU) (fixed dose for 5 days), started post‐treatment day 1 + GnRH antagonist (ganirelix acetate 0.25 mg/day), started when follicular diameter ≥ 15 mm. Control: GnRH agonist (nafarelin acetate 200 μg twice daily, nasal), started cycle day 21 + rFSH (150‐300 IU) (fixed dose for 5 days), started stimulation day 14. Both rFSH and GnRH analogues continued until hCG injection (5000‐10,000 IU), administered when ≥ 2 follicles were ≥ 17 mm in diameter.
Outcomes	Clinical pregnancy rate: ≥ 1 foetal heart beats confirmed with ultrasound, performed at least 4 weeks after embryo transfer. Ovarian cyst: intraovarian sonolucent structure with mean diameter of > 14 mm, time of measurement not reported. OHSS: not defined. Values of LH, oestradiol, progesterone. Dose of FSH. Number of collected oocytes. Number of oocytes in metaphase II. Fertilisation rate. Number of transferred embryos. Embryo implantation rate. Gestation rate per embryo transfer.
Notes	Power calculation performed: no. ITT analysis performed: no
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Randomisation by drawing lots after constructing a table of distribution. 2:1 randomization (intervention:control).
Allocation concealment (selection bias)	High risk	After drawing lots, clinicians and participants could see in the table to which treatment they were assigned to.
Blinding (performance bias and detection bias) All outcomes	Unclear risk	Insufficient information to make a conclusive judgement.
Incomplete outcome data (attrition bias) All outcomes	High risk	There was imbalance in the proportions of withdrawals between the 2 groups and data analyzed on the basis of ITT.
Selective reporting (reporting bias)	High risk	All planned outcomes not reported.
Other bias	Low risk	No significant difference in baseline characteristics with regard to age.