Our results failed to show statistically significant differences for any measures of IVF treatment outcome between pretreatment with OCP and E2 for COH with the GnRH antagonist protocol. To the best of our knowledge, this is the first study to directly compare these two methods of cycle scheduling in GnRH antagonist cycles.
Cédrin-Durnerin et al. showed a significantly higher consumption of gonadotropins in OCP-pretreated cycles as compared to either pretreatment with E2 or no pretreatment, with no significant differences in treatment outcomes . However, their study was mainly aimed at examining changes in the hormonal environment and antral follicles during the washout period after discontinuation of different methods of steroid pretreatment, and it included an even lower number of patients per treatment arm. Their data showed that steroid pretreatments differently affect the hormonal environment before the start of stimulation. According to their findings, it took 5 days after stopping OCP for FSH and LH to return to baseline levels from a strong suppression, suggesting this was an optimal washout period in cycles pretreated with OCP. In contrast, E2 pretreatment did not reduce FSH levels significantly but rather avoided the increase of FSH during the luteal-follicular transition, and the rapid FSH rebound after stopping natural estrogen intake argues for a short, 1–2 day washout interval. Indeed, we administered E2 until the day before the initiation of ovarian stimulation in order to optimize the synchronization of endogenous and exogenous FSH stimuli.
In a more recent study , the same group of researchers hypothesized that a 1-day washout period was too short to allow for complete recovery of baseline FSH levels, and that this was responsible for the increased gonadotropin consumption as compared to cycles without any pretreatment. However, no clinical studies have been performed with COH started 2 days after stopping E2 administration.
Although the length of steroid exposure varied considerably both within each pretreatment group and between the two groups, previous studies have shown that the number of pretreatment days – at least within the range of days used in our study – has no impact on COH outcomes in cycles pretreated with either E2 or OCPs . There is, however, considerable difference between OCPs in terms of their suppressive effect, depending on their progestin component . A recent, small-scale, retrospective cohort study, performed in egg donors, suggested that OCPs containing androgenic (estrane- and gonane-derived) progestins lead to a more profound suppression of follicular development than those with anti-androgenic activity, resulting in lower oocyte yields . The authors hypothesized that OCPs including an androgenic progestin would inhibit gonadotropin support for the growing follicle but maintain androgen exposure, which would lead to initial androgen-driven follicle growth but, in the end, to atresia of growing follicles because of the lack of FSH support, whereas in patients treated with anti-androgenic OCPs, lacking the androgen-driven growth of small growing follicles and the growth support from FSH, very small follicles would fail to grow but would not undergo atresia either. Upon suspending OCPs, and starting COH, these small follicles would then still have the ability to resume growth and development, leading to ultimately larger oocyte yields than with androgenic HCs. However, this hypothesis, based on retrospective comparison of two relatively small groups, each receiving several types of OCPs, needs to be confirmed by prospective studies. In fact, these findings would not affect the results of our study, as we used a single type of progestin (levonorgestrel) in all our OCP-pretreated patients.
That weekend oocyte retrievals can be significantly reduced by using OCPs is well known . Using this method ensures that there is no need to prolong the follicular phase by delaying administration of hCG, a practice that has been shown to result in lower pregnancy rates if triggering is delayed by 2 days . However, as shown by Tremellen and Lane , a 1-day advancement or delay of hCG administration does not adversely impact IVF live birth outcomes. Earlier studies demonstrated that E2 pretreatment can also be used for optimal cycle scheduling by reducing weekend retrievals [18, 24]. Moreover, extending E2 pretreatment beyond the menses has no deleterious effect on COH outcomes . Our results are in line with these findings, as the proportion of oocyte retrievals performed on weekend days was significantly reduced using either method (Figure 2). Although the reduction in weekend retrievals was even more pronounced in a study by Blockeel et al.  (1/37 in the E2 pretreatment group vs. 8/39 in patients without steroid pretreatment), our aim was not to completely avoid weekend pickups, but rather to distribute pickups evenly among the days of the week, taking into consideration the lower number of personnel working on weekends. In fact, Blockeel et al.  initiated stimulation between Thursday to Sunday, whereas our practice is to start gonadotropin administration between Wednesday and Sunday. This one-day difference might account for the more pronounced reduction in weekend pickups in their study.
A major limitation of our study is its sample size. In fact, with 50 patients in each arm of the study, only a difference of >26% could have been detected with 80% power, at a 0.05 significance level. However, as mentioned earlier, our aim was to contribute our clinical results to a future meta-analysis on the subject. An interesting future direction would be to extend our study to women with decreased ovarian reserve, as these are the patients in whom an increase in oocyte yield—due to the hypothetical beneficial effect of steroid pretreatment on follicular synchronization—could more easily be demonstrated. As far as the use of OCPs in poor responders is concerned, a modification to the GnRH antagonist protocol has been proposed by Orvieto et al.  The so-called "ultrashort GnRH agonist/GnRH antagonist protocol" is supposed to combine the beneficial effects of OCP pretreatment with that of the gonadotropin flare induced by the GnRH agonist, which is administered at the beginning of COH. The latter could circumvent the alleged detrimental effect of the OCP pretreatment on endogenous LH levels, which in turn could impair oocyte competence or endometrial receptivity. This protocol has been used with success in poor responder patients . However, no prospective studies have compared so far the efficacy of this combined approach to other stimulation protocols.