Empty Follicle Syndrome Due to Failure of Gonadotropin-Releasing Hormone Agonist Trigger: A Retrospective Study of Possible Factors That Affect the Outcome of a Rescue Human Chorionic Gonadotropin Protocol.

Introduction Empty follicle syndrome (EFS) may occur in in vitro fertilization (IVF) cycles that use a gonadotropin-releasing hormone (GnRH) antagonist protocol with a single GnRH agonist trigger. We evaluated whether a rescue human chorionic gonadotropin (hCG) trigger can salvage oocyte retrieval in these cases, and identified factors associated with obtaining a transferable embryo. Methods Among 4,778 cycles triggered with a GnRH agonist alone between June 2014 and February 2019, we retrospectively identified 43 cycles, in 43 patients, in which no mature oocyte was obtained at the initial retrieval, and a rescue hCG protocol with repeat ovum pick-up was performed. Cycles were divided into an embryo-transfer (ET)-positive group, in which a transferable embryo was obtained (n = 30), and an ET-negative group (n = 13). Baseline and peri-trigger hormone levels were compared, and thresholds for baseline luteinizing hormone (LH) and follicle-stimulating hormone (FSH) were derived by receiver operating characteristic analysis and evaluated in contingency tables. Results The incidence of GnRH agonist trigger failure was 0.9% (43/4,778 cycles). Oocytes were collected in 37 of 43 cycles, and a transferable embryo was obtained in 30 (69.8%); frozen-thawed transfer resulted in 18 live births, a live birth rate of 41.9% per rescue cycle. A baseline LH threshold of 1.9 mIU/mL (area under the curve (AUC) 0.83) and a baseline FSH threshold of 5.4 mIU/mL (AUC 0.85) distinguished the two groups (P < 0.001 for both); the corresponding odds of failing to obtain a transferable embryo were 33.0 (95% CI 3.7-296) and 60.0 (95% CI 6.3-572), with confidence intervals reflecting the small number of events. All 43 patients developed ovarian hyperstimulation syndrome (OHSS): 28 mild and 15 moderate. No severe OHSS requiring hospitalization occurred. Conclusions A rescue hCG trigger is a workable management strategy for EFS following a GnRH agonist trigger, yielding live births in cycles that would otherwise have been cancelled. All patients in this series developed mild or moderate OHSS, and none developed severe OHSS, although with 43 cycles the series is too small to establish the risk of severe OHSS. Baseline LH and FSH, together with the change in estradiol and progesterone between the first and second triggers, may help estimate the probability of obtaining a transferable embryo, but these thresholds are exploratory and require prospective validation.