A higher probability of ongoing pregnancy is present by using the fixed as compared to the flexible GnRH antagonist protocol

Abstract Study question Is the probability of ongoing pregnancy higher in the fixed as compared to the flexible gonadotrophin releasing hormone (GnRH) antagonist protocol? Summary answer A higher probability of ongoing pregnancy is present by using the fixed as compared to the flexible GnRH antagonist protocol. What is known already Although fixed GnRH antagonist initiation is a simpler protocol that requires less monitoring compared to the flexible one, the latter might avoid unnecessary antagonist use, by administering GnRH antagonist only when there is a real risk for premature luteinizing hormone (LH) rise. A meta-analysis of four randomised controlled trials (RCTs) published in 2005, comparing fixed versus flexible GnRH antagonist administration, showed a lower non-significant probability of clinical pregnancy in the flexible compared with the fixed protocol. Following that meta-analysis, four additional RCTs comparing the fixed versus the flexible GnRH antagonist protocol have been performed necessitating an update on this topic. Study design, size, duration A literature search was performed until 12/2021, in order to identify RCTs comparing the fixed versus the flexible GnRH antagonist protocol in patients undergoing ovarian stimulation for in-vitro fertilization (IVF) using gonadotrophins. The main outcome measure was achievement of ongoing pregnancy. Secondary outcome measures included duration of stimulation, incidence of premature LH rise, estradiol levels on the day of triggering final oocyte maturation and the number of cumulus-oocyte-complexes (COCs) retrieved. Participants/materials, setting, methods Seven RCTs comparing fixed versus flexible protocol, published between 2002 and 2021, were identified (1007 patients). A meta-analysis was performed using the fixed or random effects method depending on the presence of statistically significant heterogeneity. For dichotomous data, estimates were expressed as risk ratio (RR) with 95% confidence intervals (CIs). For continuous data, differences were pooled across resulting in a weighted mean difference (WMD) with 95% CI. Main results and the role of chance Ovarian stimulation was performed using recombinant gonadotrophins in all studies. In the fixed protocol initiation of GnRH antagonist was performed either on Day 6 (six studies) or on Day 5 (single study) of stimulation. In the flexible protocol antagonist was initiated using either ultrasound criteria (four studies) or by combination of ultrasound or hormonal criteria (three studies). No significant difference was present between the fixed and flexible groups in the duration of gonadotrophin stimulation (WMD:+0.23 days, 95% CI: -0.08 to + 0.53, random protocol) and in the incidence of premature LH rise (RR:0.86, 95% CI: 0.48-1.53, fixed protocol). Significantly lower estradiol levels were present on the day of triggering final oocyte maturation (WMD:-237.74, 95% CI: -463.87 to -11.61, random protocol) and significantly less COCs were retrieved (WMD:-1.83, 95% CI: -3.11 to -0.54, random protocol) in the fixed compared with the flexible protocol. A significantly higher probability of ongoing pregnancy was observed in the fixed as compared to the flexible protocol, both per intention to treat (RR:1.28, 95% CI: 1.02-1.61, I2:0%, fixed protocol, six studies, 898 patients) as well as per protocol analysis (RR:1.30, 95% CI: 1.04-1.62, I2:0%, fixed protocol, six studies, 757 patients). Limitations, reasons for caution In the current meta-analysis, the number of studies and patients included does not allow to perform subgroup analyses based on the time of fixed antagonist initiation or on the criteria used for flexible antagonist initiation. Wider implications of the findings Although in theory flexible antagonist initiation is an attractive protocol that leads to less antagonist administration, the current meta-analysis suggests that it is associated with a lower probability of ongoing pregnancy. Further RCTs examining the two antagonist protocols for additional outcomes, including live birth or cumulative live birth, are necessary. Trial registration number N/A