Fanchin 2003.

Methods	Parallel group study. Number of women randomized: 100 (number of women per group not reported). Number of withdrawals: 10 (4 due to personal reasons and 6 due to major protocol violation). Number of women analyzed: 90 (47 in intervention group; 43 in control group). Duration of study: 1 IVF‐ET cycle, from day 20 of the previous cycle to day of hCG administration (information obtained from contact person).
Participants	Country of authors: France. Inclusion criteria: aged ≤ 38 years; regular, ovulatory menstrual cycles every 25‐35 days; both ovaries present; no current or past diseases affecting ovaries or gonadotrophin or sex steroid secretion, clearance or excretion; BMI 18‐27 kg/m2; no hormone therapy during the past 6 weeks; adequate visualisation of both ovaries in transvaginal ultrasound scans. Exclusion criteria: not reported. Median age (range): intervention group: 33 (26‐38) years; control group: 33 (25‐38) years.
Interventions	Intervention: micronised 17β‐E2 (4 mg/day, PO), started cycle day 20 until day 2 of the next cycle + rFSH (225 IU/day, SC) on cycle days 3‐7 + GnRH antagonist (cetrorelix acetate 3 mg single dose, SC) when ≥ 1 follicle > 13 mm in diameter. Control: rFSH (225 IU/day SC) on cycle days 3‐7 + GnRH antagonist (cetrorelix acetate 3 mg single dose, SC) when ≥ 1 follicle > 13 mm in diameter. rFSH dose adjustments according to follicle growth determined by serum oestradiol levels and ultrasound monitoring.
Outcomes	Days of GnRH antagonist administration. Day of hCG administration. Dose of gonadotrophins. Number of mature follicles. Number of embryos transferred. Clinical pregnancy rates per cycle (cycles equivalent to number of participants); presence of a gestational sac with foetal heart activity at 6 weeks on ultrasound scan.
Notes	Power calculation performed: yes. ITT analysis performed: no Our data analysis in this review includes 90 women with full follow‐up. We did not include all randomized women because it is unclear how many women were randomized to each group before dropouts. The study publication reports very low measures of variability which can be to be assumed SEs and which we have converted to SDs
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "Women randomly received...", "...according to a computer‐generated, blocked randomization list."
Allocation concealment (selection bias)	Low risk	Quote: "Treatment allocation was decided by an independent person."
Blinding (performance bias and detection bias) All outcomes	Unclear risk	No blinding, information obtained from contact person.
Incomplete outcome data (attrition bias) All outcomes	High risk	Unclear how many withdrew in each group and data not analyzed on the basis of ITT.
Selective reporting (reporting bias)	High risk	All planned outcomes not reported.
Other bias	Low risk	No difference in baseline characteristics with regard to age, indication for IVF‐ET, duration of infertility, rank of the current IVF‐ET attempt, menstrual cycle length, day 3 serum FSH and oestradiol.