OBJECTIVE: To study if the GnRH agonist administration in luteal phase improves clinical pregnancy rate of fresh and frozen embryo transfer. Also, this meta-analysis compares the treatment effect of luteal GnRH agonist administration between two ovarian stimulation protocols of fresh cycles (long agonist and antagonist), and between two types of treatment: fresh and frozen embryo transfers.
DESIGN: Systematic review and meta-analysis (according to PRISMA statement). PROSPERO registration number was CRD42017059152.
MATERIALS AND METHODS: For fresh cycles, we include randomised control trials to assess the effects of the GnRH agonist in luteal phase. For frozen embryo transfer cycles, we include randomised control trials and prospective cohort studies (in assessing a possible difference through meta-regression). Studies examining women undergoing assisted reproductive technology (ART) treatment with fresh or frozen embryo transfer cycles (including oocyte recipient cycles) were eligible for the review. Unpublished studies were also included. The intervention was addition of GnRH agonist during the luteal phase. Clinical pregnancy among participating women is the primary outcome. A computerized literature search in PubMed, Cochrane Central Register of Controlled Trials (CENTRAL), EMBASE and RCT registries (clinicaltrials.gov) covering the period up to May 2017 was performed.
RESULTS: For the overall 20 studies (4824 patients), clinical pregnancy rate significantly increased in GnRH agonist administration group (RR 1.25, 95% CI 1.15-1.36, p<0.001). Among fresh cycles, no significant difference was observed between long agonist and antagonist protocol (RR= 1.13, 95% CI 0.94 -1.39, p=0.194). The effect in frozen embryo cycles is homogenous (I2=0%, Q-test, p=0.845) and significantly more benefit than in fresh cycles (RR= 1.22, 95% CI 1.03-1.43, p=0.016).
CONCLUSIONS: GnRH agonist administration in luteal phase may improve clinical pregnancy rate in patients undergoing ART treatment, regardless fresh or frozen embryo transfer cycles. This effect in frozen cycles is homogenous and higher than in fresh cycles. No significant difference of benefit is found between two protocols of fresh cycles.