Luteal phase support for in vitro fertilization/intracytoplasmic sperm injection fresh cycles: a systematic review and network meta-analysis

BACKGROUND: Various luteal phase supports (LPSs) have been proven to increase the pregnancy rate in fresh cycles of in vitro fertilization or intracytoplasmic sperm injection; however, there is still significant debate regarding the optimal use of LPS. METHODS: A systematic review with the use of a network meta-analysis was performed via electronic searching of Ovid MEDLINE, the Cochrane Library, Embase, Web of Science, ClinicalTrials.gov and Google Scholar (up to January 2021) to compare the effectiveness and safety of various LPSs, as well as to evaluate the effects of different initiations of LPSs on pregnancy outcomes. The primary outcomes included live birth and ongoing pregnancy, with the results presented as odds ratios (ORs) with 95% confidence intervals (CIs). RESULTS: Eighty-nine randomized controlled trials with 29,625 women comparing 14 interventions or placebo/no LPS treatments were included in the meta-analyses. No significant differences were found in terms of the pregnancy outcomes when LPS was started within 48 h after oocyte retrieval versus a delayed initiation between 48 h and 96 h after oocyte retrieval. The addition of gonadotropin-releasing hormone (GnRH) agonists to progesterone vaginal pessaries showed a significant benefit in terms of live birth (OR 1.39, 95% CI 1.08 to 1.78). Only human chorionic gonadotropin (HCG) was found to be more efficacious than the placebo/no LPS treatment in terms of live birth (OR 15.43, 95% CI 2.03 to 117.12, low evidence). Any active LPSs (except for rectal or subcutaneous progesterone) was significantly more efficacious than the placebo/no LPS treatment in terms of ongoing pregnancy, with ORs ranging between 1.77 (95% CI 1.08 to 2.90) for the vaginal progesterone pessary and 2.14 (1.23 to 3.70) for the intramuscular progesterone treatment. Among the comparisons of efficacy and tolerability between the active treatments, the differences were small and very uncertain. CONCLUSION: Delays in progesterone supplementation until 96 h after oocyte retrieval does not affect pregnancy outcomes. The safety of GnRH agonists during the luteal phase needs to be evaluated in future studies before the applications of these agonists in clinical practice. With comparable efficacy and acceptability, there may be several viable clinical options for LPS. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12958-021-00782-5.