P–761 Can modified luteal support in fresh cycle after agonist trigger “rescue” the cumulative live birth rate (CLBR) in high responders

Study question Can modified luteal support in fresh cycle “rescue” the cumulative live birth rate (CLBR) in high responders who receive agonist trigger? Summary answer Live birth rate in high responders who had agonist trigger in fresh cycle was significantly reduced despite modified luteal support. What is known already Previous studies, including small randomised controlled trials, claimed that good live birth rate could be achieved at fresh transfer in “high responders” who had GnRHa trigger with modified luteal phase support. However, majority of these studies exclude the true high responders (patients with 20 and above oocytes) and average number of collected eggs reported in many of these studies in the range of 9 to 12. The data on outcome of fresh transfer in true high responders is very limited. Study design, size, duration A prospective observational study was conducted in 407 patients, aged 23–42 years who were expected to be at risk of high response (AFC>18, AMH>20 pmol/l) undergoing controlled ovarian stimulation between 2014–2019 triggered either with HCG or GnRH agonist. Live birth rate (LBR) in a fresh and subsequent 3 frozen transfers were compared in groups with different triggers and freeze all. Participants/materials, setting, methods Patients were stimulated in short antagonist protocol. The trigger was chosen based on the background characteristics, peak oestradiol and clinician discretion. Triggering was achieved either with 0.5 mg buserelin (GnRHa) 0.5mgin 230 patients (A) or with 250 mcg of hCG(H) in 177 patients. Modified luteal support included vaginal progesterone, oral oestrogen and 1500 iu of hCG on the day of egg retrieval. The later was omitted with more than 20 oocytes. Main results and the role of chance The mean age, AFC, number of previous cycles, number of embryos transferred were 33.3, 22.4, 0.26 and 1.2 respectively and did not have significant difference between different triggers. Whereas AMH (53 pmol/l (A) vs 43.1 pmol/l (H), P = 0.003), peak oestradiol (15140.74 (A) vs 9738.59 (H), P = 5.59X10–14), and number of oocytes collected (21 vs.17, P = 5.63X10–7) were significantly higher in GnRHa group. Seventeen patients in buserelin group had elective freeze all. Ovarian Hyperstimulation Syndrome (OHSS) rate was 3.9% in buserelin group (more then half of these cases had a bolus of hCG at egg collection; most were mild/moderate). On the other hand, hCG group had 2.5% of OHSS (all severe). Live birth rate in fresh cycle was 31% in hCG and 21% in GnRHa groups. If freeze all was undertaken in fresh cycle after GnRHa trigger, then LBR in the first frozen cycle of this group was 53% (P = 0.003, fresh vs frozen GnRHa group). CLBR was not different between GnRHa and hCG groups (51%). However, this was significantly lower than CLBR in GnRHa trigger freeze all group 76% (P = 0.03) Limitations, reasons for caution The limitation of this study is its non-randomised nature. However, since it is one of the biggest studies for high responders it has a power to minimise bias by adjusting for multiple variables. Wider implications of the findings: Proceeding with fresh transfer in high responders after GnRHa trigger with modified luteal support not only maintains the risk of OHSS (equivalent to hCG group) but also significantly impairs the LBR not compensated even after 3 subsequent frozen embryo transfers. Therefore, freeze-all approach should be preferred in this group. Trial registration number NA