Randomised controlled trials (RCTs) of women who underwent IVF after ovarian stimulation with gonadotrophins and gonadotrophin releasing hormone (GnRH) analogues, that had a pregnancy outcome were eligible for inclusion. Included studies had to compare patients who received progesterone only with those who received progesterone plus oestrogen (for luteal support). Studies that used pseudo-randomisation methods (including studies in which patients had contributed more than one cycle) were excluded.
Included studies were published between 2006 and 2007. Patient ages ranged from less than 39 to 44 years old where reported. Recombinant gonadotrophins were used in all studies. Urinary human chorionic gonadotrophin (hCG) was used to trigger final oocyte maturation in most of the included studies. Type, dose and route of administration, timing of initiation and duration of luteal support with oestrogen and progesterone varied. Fertilisation methods included were intra-cytoplasmicsperm injection (ICSI) with or without IVF. Primary outcomes were achievement of pregnancy expressed as positive hCG, clinical pregnancy and live birth rate. Secondary outcomes included biochemical miscarriage per patient with positive hCG, clinical miscarriage per patient with a clinical pregnancy and total miscarriage per patient with positive hCG.
Two reviewers independently selected studies. Disagreements were resolved by discussion.