Luteal phase stimulation is associated with lower peak estradiol levels and less GNRH antagonist use while yielding the same outcomes as conventional follicular phase stimulation

Abstract

Listen

Translate article

Elevated levels of estradiol are associated with an increase in the incidence of ovarian hyperstimulation syndrome (OHSS) in women undergoing IVF treatment. GnRH antagonists are frequently used in conventional antagonist protocols to prevent a premature luteinizing hormone (LH) surge, though they are known to be invasive and costly. We hypothesize that an alternative protocol initiating stimulation in the luteal phase, wherein progesterone may serve as a natural inhibitor of the LH surge, in combination with an aromatase inhibitor, may be a comparable protocol for women undergoing IVF who are at risk for OHSS. Additional benefits to using this protocol include fewer medication costs to the patient and decreased chance of cycle cancellation due to OHSS. The purpose of this study was to compare peak estradiol levels, duration of GnRH antagonist use, and oocyte yield in a luteal phase stimulation compared to a conventional follicular phase stimulation protocol. This was a retrospective study in a private IVF program examining 56 patients treated with either a standard follicular phase stimulation or luteal phase stimulation between January and October 2017. We included all women with ovulatory cycles, age < 39, BMI < 30 kg/m2, and CD3 FSH < 9. Luteal phase protocol consisted of daily 250 IU Follistim, 5 IU microdose hcg and 5mg daily letrozole initiated after spontaneous ovulation or inducing with 200mg vaginal progesterone twice daily. GnRH antagonist (ganirelix acetate or cetrorelix acetate) was initiated when leading follicle size >15mm or when E2 levels were > 200 pg/mL. Baseline characteristics were collected for all patients and the primary outcomes measured were oocyte yield, duration of GnRH antagonist use, and peak E2 levels. Baseline characteristics were similar between both protocol groups. Peak estradiol levels were 647 and 4669 pg/mL in luteal and conventional stimulations, respectively (p <0.01). GnRH antagonist was administered for 3.3 days in luteal protocols compared to 5.4 days in a conventional protocol, while the total number of stimulation days were similar (11.3 vs 10.1, p = 0.29). The average oocyte yield between antagonist and luteal stimulations was not significantly different (12.9 vs 13.6, p=0.72) (Table 1). No cases of premature luteinization, ovulation, or OHSS were observed in either group.Tabled 1Table 1ParameterStimulation ProtocolP-valueAntagonist(n=29)Luteal(n=27)Maternal age (range)34.4 (25-39)33.7 (26-39)0.43CD 3 FSH6.96.00.44BMI22.522.30.82Stim duration (d)10.111.3<0.01*GnRH antagonist duration (d)5.43.2<0.01*Stim day at time of GnRHa start5.79.1<0.01*Peak E2 (pg/ml)4669.9656.3<0.01*# oocytes retrieved13.612.70.60*p<0.05 Open table in a new tab *p<0.05 Luteal phase stimulation with use of letrozole is associated with significantly lower peak estradiol level while producing similar competent oocytes compared to conventional follicular phase stimulation. This protocol may be preferred for women at higher risk of developing OHSS or for women with endometriosis. In addition, progesterone exposure during the luteal phase can be used to prevent premature luteinization, therefore requiring fewer GnRH antagonist injections and reducing cost.