Engmann 2008.
| Methods | Open‐label, parallel, university‐based tertiary fertility centre, RCT | |
| Participants | 66 women were included. Inclusion criteria: age 20 to 39 years, basal FSH concentration ≤ 10.0 IU/L and undergoing first cycle of IVF with PCOS or PCOM, or undergoing subsequent cycle with a history of high response in previous IVF cycles. Exclusion criteria: women with hypogonadotropic hypogonadism Baseline characteristics: 32.0 ± 3.7 vs 33.1 ± 3.6 years |
|
| Interventions |
Ovarian stimulation: Control group: OCP + 112 to 225 IU recFSH on CD2 + midluteal 1 mg leuprolide acetate (SC). Study group: OCP + 112 to 225 IU recFSH on CD2 + flexible GnRH antagonist protocol (SC) Intervention: SC leuprolide in a dose of 1 mg approximately 12 hours after last dose of ganirelix vs SC hCG (Profasi; Serono, Randolph, MA) in a dose ranging from 3300 to 10,000 IU, depending on follicular response Number of embryos transferred: GnRH agonist group vs HCG group (mean ± SD: 2.0 ± 0.2 vs 2.2 ± 0.6) LPS: study group: 50 mg IM P in oil + 0.1 mg transdermal E2 patches every other day, starting the day after oocyte retrieval. Both doses were adjusted according to E2 and P levels on the day of embryo transfer and 1 week after oocyte retrieval. Control group: 0 mg IM P in oil |
|
| Outcomes | Primary outcome measures: OHSS occurrence assessed 1 week after oocyte retrieval and implantation rate assessed at 7 weeks' gestation Secondary outcome measures: clinical pregnancy rate assessed at time of ultrasound, mature oocytes assessed at time of retrieval and ovarian volume assessed 1 week after oocyte retrieval | |
| Notes | Supported in part by an unrestricted educational grant from Organon USA, Roseland, New Jersey | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | 1:1 by means of computer‐generated random numbers with separate randomisation for women undergoing first cycle and for women with a previous high response by the use of stratified randomised blocks |
| Allocation concealment (selection bias) | Low risk | Research nurse by using a series of consecutively numbered sealed opaque envelopes (1 for each category of previous cycle) |
| Blinding (performance bias and detection bias) FOR OHSS OUTCOME | High risk | Not blinded. Risk applies to assessment of OHSS |
| Incomplete outcome data (attrition bias) All outcomes | Low risk | Most randomly assigned women were analysed using per‐protocol (PP) and intention‐to‐treat analysis (ITT) |
| Selective reporting (reporting bias) | Unclear risk | Protocol was available and outcomes were prespecified; OHSS, implantation rate (IR), MII, CPR, ovarian volume 1 week after oocyte retrieval. Study reported extra outcomes not stated in the protocol, such as serum luteal phase E2, P levels, fertilisation rate (FR). Live birth rate not reported |
| Other bias | Low risk | No other source of potential bias was identified |