Does an FSH surge at the time of hCG trigger improve IVF/ICSI outcomes? A randomized, double-blinded, placebo-controlled study

STUDY QUESTION Does an artificially induced FSH surge at the time of hCG trigger improve IVF/ICSI outcomes? SUMMARY ANSWER An additional FSH bolus administered at the time of hCG trigger has no effect on clinical pregnancy rate, embryo quality, fertilization rate, implantation rate and live birth rate in women undergoing the long GnRH agonist (GnRHa) protocol for IVF/ICSI. WHAT IS KNOWN ALREADY Normal ovulation is preceded by a surge in both LH and FSH. Few randomized clinical trials have specifically investigated the role of the FSH surge. Some studies indicated that FSH given at hCG ovulation trigger boosts fertilization rate and even prevents ovarian hyperstimulation syndrome (OHSS). STUDY DESIGN, SIZE, DURATION This was a randomized, double-blinded, placebo-controlled trial conducted at a single IVF center, from June 2012 to November 2013. A sample size calculation indicated that 347 women per group would be adequate. A total of 732 women undergoing IVF/ICSI were randomized, using electronically randomized tables, to the intervention or placebo groups. Participants and clinical doctors were blinded to the treatment allocation. PARTICIPANTS/MATERIALS, SETTING, METHODS Patients aged <= 42 years who were treated with IVF/ICSI owing to tubal factor, male factor, unexplained, endometriosis and multiple factors were enrolled in this trial. Subjects all received a standard long GnRHa protocol for IVF/ICSI and hCG 6000-10000 IU to trigger oocyte maturation. A total of 364 and 368 patients were randomized to receive a urinary FSH (uFSH) bolus (6 ampules, 450IU) and placebo, respectively, at the time of the hCG trigger. The primary outcome measure was clinical pregnancy rate. The secondary outcome measures were FSH level on the day of oocyte retrieval, number of oocytes retrieved, good-quality embryo rate, live birth rate and rate of OHSS. MAIN RESULTS AND THE ROLE OF CHANCE There were no significant differences in the baseline demographic characteristics between the two study groups. There were also no significant differences between groups in cycle characteristics, such as the mean number of stimulation days, total gonadotrophin dose and peak estradiol. The clinical pregnancy rate was 51.6% in the placebo group and 52.7% in the FSH co-trigger group, with an absolute rate difference of 1.1% (95% CI -6.1% to 8.3%). The number of oocytes retrieved was 10.474.52 and 10.745.01 (P=0.44), the rate of good-quality embryos was 37% and 33.9% (P=0.093) and the implantation rate was 35% and 36% (P=0.7) in the placebo group and the FSH co-trigger group, respectively. LIMITATIONS, REASONS FOR CAUTION This was a single-center study, which may limit its effectiveness. The use of uFSH is a limitation, as this is not the same as the natural FSH. We did not collect follicular fluid for further study of molecular changes after the use of uFSH as a co-trigger. WIDER IMPLICATIONS OF THE FINDINGS Based on previous data and our results, an additional FSH bolus administered at the time of hCG trigger has no benefit on clinical pregnancy rates in women undergoing the long GnRHa protocol in IVF/ICSI: a single hCG trigger is sufficient.