Impact of gonadotropin selection on risk of ovarian hyperstimulation syndrome in predicted high responders: a Menopur in Gonadotropin-releasing hormone Antagonist Single Embryo Transfer-High Responder trial analysis.

Objective: To evaluate the risk of ovarian hyperstimulation syndrome (OHSS) with highly purified human menopausal gonadotropin (HP-hMG) or recombinant follicle-stimulating hormone (rFSH) for controlled ovarian stimulation in patients predicted to be high responders. Design: Post hoc analysis of a randomized, open-label, assessor-blind, parallel-group, noninferiority trial conducted at 31 US fertility centers. Subjects: 620 women with serum anti-Müllerian hormone (AMH) ≥5 ng/mL. Intervention: Controlled ovarian stimulation with HP-hMG or rFSH in a gonadotropin-releasing hormone antagonist assisted reproductive technology cycle. Human chorionic gonadotropin trigger and fresh transfer of a single blastocyst was performed unless ovarian response was excessive; in this case, subjects received gonadotropin-releasing hormone agonist trigger and all embryos were cryopreserved. Subjects could undergo frozen blastocyst transfers within 6 months of randomization. Main Outcome Measures: Demographic differences between subjects who developed OHSS and those who did not; incidence of OHSS based on trigger type; interaction between baseline AMH and oocyte yield by treatment group; and OHSS incidence based on number of oocytes retrieved by treatment group. Results: Subjects who developed early OHSS were significantly younger, with lower weight and body mass index, shorter duration of infertility, and lower baseline estradiol than those who did not develop early OHSS. Among subjects who developed early OHSS, those treated with rFSH had a lower weight than those treated with HP-hMG; all other baseline demographics were similar between treatment groups. The rate of early OHSS was not significantly different based on trigger type. A statistically significant interaction was observed between baseline AMH and treatment group on the number of oocytes retrieved. Odds of early OHSS increased by 1.06 times (95% confidence interval [CI]: 1.04, 1.08) for each additional oocyte retrieved, independent of treatment group. Subjects with high oocyte yields (>25) had lower early OHSS rates when treated with HP-hMG compared with rFSH (difference: 24%, 95% CI: 8.3, 36.2). Holding the number of oocytes constant, the odds of early OHSS in HP-hMG-treated subjects was 0.46 times (95% CI: 0.26, 0.82) that in rFSH-treated subjects. Conclusion: In predicted high-responder patients, HP-hMG stimulation was associated with significantly diminished OHSS rates compared with rFSH, adjusted for oocyte yield.