الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 289 |  |  |
| | | from | 289 | | |
Abstract
Compared with GnRH agonist long protocols, the introduction of GnRH antagonist protocols for controlled ovarian hyperstimulation has offered a great opportunity to reduce the duration of treatment and the consumption of gonadotrophins (Al-Inany and Aboulghar, 2001).
However, GnRH antagonist protocols are cycle-dependent (Albano et al., 1997; Olivennes et al., 1998; The ganirelix dose-finding study group, 1998; Huirne and Lambalk, 2001), which makes scheduling IVF or ICSI cycles difficult. Moreover, there is a slight reduction in the number of retrieved oocytes compared to the GnRH agonist protocol (Al-Inany and Aboulghar, 2001),
For these reasons, more attention has been paid to the potential use of steroid pretreatments to schedule IVF cycles, modulate the hormonal environment before starting COH, and thereby synchronize the follicular cohort before stimulation. Indeed, combined oral contraceptive pills (OCP) (Huirne et al., 2006; Rombauts et al., 2006; Cédrin-Durnerin et al., 2007), progestogens alone (Frydman et al., 1986; Wardle et al., 1986; Gerli et al., 1989; Biljan et al., 1998) and natural estrogens alone (De Ziegler et al., 1998) have been largely used for many years to program GnRH antagonist cycles.
Nevertheless, outcomes of OCP pretreatment in GnRH antagonist cycles showed ongoing pregnancy rate was significantly reduced in OCP-pretreated women, with an odds ratio ranging between 0.74 – 0.80. Paradoxically, each of the components of combined OCP in isolation did not show this detrimental effect (Farquhar et al., 2017). Consequentially, in the 2020 ESHRE guideline, combined oral contraceptive pill pretreatment in the GnRH antagonist protocol is not recommended due to reduced efficacy (strong recommendation); whereas pretreatment with either estrogen alone or progesterone alone is probably not recommended for improving efficacy or safety (conditional recommendation), but is probably acceptable for scheduling purposes given the data on efficacy and safety.
Despite being the first-line recommended COH protocol in women with PCOS (Teede et al., 2018; The ESHRE Guideline group on Ovarian Stimulation et al., 2020), data regarding the efficacy or safety of steroid pretreatment for its scheduling in such population is scarce.
Overall data suggest that there is a scope to revisit the concept of scheduling GnRH antagonist cycles by steroid pretreatment, especially in women with PCOS. Hence, the purpose of this prospective observational study was to assess the effect of OCP pretreatment on pregnancy outcomes in women with PCOS undergoing IVF/ICSI using GnRH antagonist protocol.
Polycystic ovary syndrome (PCOS) is the commonest endocrine disorder in women of reproductive age. It is the major cause of anovulatory infertility, menstrual disturbances, and hirsutism. In addition, PCOS is also associated with metabolic disturbances, central to which is peripheral insulin resistance and compensatory hyperinsulinemia (Dunaif, 1997; Ehrmann, 2005). These endocrinal and metabolic abnormalities appear to have an influence on the development of anovulation in women with PCOS, but these factors do not exclude the possibility of an intrinsic abnormality of folliculogenesis in PCOS.
The characteristic follicular feature of polycystic ovaries in anovulatory women is the accumulation of antral follicles in the range of 2–8 mm in diameter (Franks et al., 2000; Franks et al., 2008). The simple explanation for this phenomenon is that serum concentrations of FSH, while rarely frankly low, are suppressed below the ‘threshold’ level required during the early follicular phase to stimulate normal follicle maturation (Hillier, 1994). Incrementing the feedback response for this will raise questions since follicles themselves are failing to develop, and deprived of the pituitary trophic support necessary to derive their steroidogenic machinery; unless the follicles are defective themselves.
Defective follicles or aberrant function of their cells include but are not confined to; theca cells which are characterized by constitutively increased activity of steroidogenic enzymes. Whereas there is little difference in the characteristics of theca cells between follicles obtained from ovulatory compared with anovulatory women with polycystic ovaries; vast differences exist in granulosa cell steroidogenesis according to the ovulatory status.
The effect of PCO on reproduction and fertility is not confined only to anovulation and the metabolic and endocrinal changes accompanied, but also the effect of these on the endometrium.
Balanced and sequential exposure to both estrogen and progesterone yields an orchestrated pattern of gene expression, resulting in a synchronized period of uterine receptivity towards embryonic implantation (Talbi et al., 2006). Conversely, given the central role of chronic anovulation in PCOS-associated infertility; the abnormal anovulatory endocrine milieu was regarded as the main culprit for the aberrant endometrium in PCOS.
It seems that in PCOS, an “endometrial disease phenotype” is promoted by anovulation and aggravated by hyperandrogenism and metabolic and inflammatory changes related to obesity, insulin resistance, and accompanying hyperinsulinemia; all of which are common in PCOS women
The uterus undergoes dramatic physiological and morphological changes during the peri-implantation period, which are orchestrated by ovarian hormones as well as morphogens, growth factors, and transcription factors (Wang and Dey, 2006; Zhang et al., 2013). Thus, the PCOS-related aberrant hormonal milieu, and expression of endometrial steroid receptor and their coactivators, might be implicated in the higher incidence of implantation failure in women with PCOS (Azziz et al., 2009; Fauser et al., 2012).
After learning the hormonal and follicular dynamics of PCO and the effect they have on the endometrium, hence the implantation we need to further understand the effect of both GNRH antagonist protocol and combined oral contraceptives on these domains.
To sum up the GnRH antagonist protocol hormonal dynamics, current concepts of gonadotrophic control of ovarian function and clinical evidence have established that both a ‘threshold’ and a ‘ceiling’ for LH concentrations (framing the so-called LH ‘window’) exist during the follicular phase of menstrual and induced cycles (Hillier, 1994, 2000; Balasch and Fábregues, 2002; Shoham, 2002; Balasch, 2004). Therefore, concentrations of LH should be neither too high nor too low during ovulation induction. During the second half of the follicular phase, as plasma FSH concentrations decline, the LH-dependent phase of pre-ovulatory follicular development proceeds normally only if LH is present at concentrations over the threshold concentration and below the ceiling value. When the ceiling is exceeded at the midcycle surge of LH, further division of granulosa cells ceases as luteinization proceeds.
Follicular dynamics; the lower oocyte yield of GnRH antagonist protocol is hypothesized to originate from the asynchronous growth of the recruited follicular cohort. Optimal synchronization depends on the sustained suppression of endogenous FSH before starting exogenous stimulation (Huirne et al., 2007)
Endometrial dynamics; with the initial reports of reduced implantation and pregnancy outcomes with GnRH antagonists (Al-Inany et al., 2006), impairment of endometrial receptivity was incriminated as a contributing cause; as frozen-thawed embryos yielded higher pregnancy rates compared to fresh ones from the same stimulation cycles, especially those using high-dose GnRH antagonists (Kol et al., 1999). This is thought to be due to an indirect effect of GnRH antagonists on endometrium receptivity might exist, through their effect on the follicular-phase hormonal milieu (gonadotropin and steroid levels); thus, affecting endometrial growth and maturation (Kolibianakis et al., 2002).
Whereas, the effect of OCP on hormonal, follicular, and endometrial dynamics are as the following;
As regard hormonal dynamics; In the early follicular phase, the estrogen component acts in a negative feedback and directly inhibits FSH secretion. This limits the process of FSH-dependent follicle growth, which plays an integral role in the process of follicular recruitment and selection. Should, however, a dominant follicle emerge, the LH surge – and thus ovulation – is reliably prevented through the progestin component. (Mishell et al., 1977; Van Heusden and Fauser, 1999).
On the other hand, the progestin component is responsible for the inhibition of pituitary LH secretion. The secretion of FSH is minimally affected by natural progesterone or by synthetic progestogens used in the doses included in most oral contraceptives.
As regard follicular dynamics; Several studies have documented the development of follicles to diameters of 10 mm or greater in women using OCP containing 20μg to 40μg EE and various progestins (Killick et al., 1987; Grimes et al., 1994; van der Does et al., 1995; Coney and DelConte, 1999; Pierson et al., 2003; Baerwald et al., 2006; D’Arpe et al., 2016). Unfortunately, reaching 10 mm follicle size is the critical point in the process of follicular development; at which follicles generally become physiologically selected for preferential growth (see before), and therefore have the greatest potential for escape ovulation (Ginther et al., 2001)
Finally, the effect on the endometrium demonstrates a mixture of the effects of both exogenous estrogen and progestin, albeit with a dominant progestin effect; or the so-called “pill endometrium” (Dinh et al., 2015):
- In the first few cycles: progestins induce secretory differentiation, with coexistent proliferative and secretory features.
- After several cycles: the progestin down-regulates the estrogen receptor and a “classic pill endometrium” occurs, which is composed of quiescent, atrophic glandular epithelium against a background of early pseudodecidualized stroma, akin to the normal late secretory phase (Figure 3.5).
- With continued long-term use: atrophy of both glands and stroma occur.
After understanding and learning about the effect of both GnRH antagonist protocol and OCP on the three most important domains of human fertility, the question yet to be answered is to pill or not to pill and the debate behind this question. An unsettled debate continues for over a decade, regarding the harms and the benefits of OCP pretreatment. Several theories were proposed, and opposing evidence was reached. Yet, the latter is still held by many practitioners as unconvincing
Why pill before GnRH antagonist protocol?
a- To schedule the IVF cycle
b- Prevent early follicular phase progesterone elevation
c- To improve the synchronization of the follicular cohort, leading to an increased number of retrieved mature oocytes.
d- To normalize elevated basal endogenous LH levels before starting the GnRH antagonist
e- To suppress the excessive intraovarian androgens in PCOS
Why NOT pill before GnRH antagonist protocol?
a- OCP pretreatment might have a residual adverse effect on endometrial receptivity
b- OCP pretreatment might profoundly suppress endogenous LH levels
For each of the above points, there are pieces of evidence to either support or advice against this act.
So how exactly will we proceed with our study?
The study included 700 women with PCOS undergoing IVF/ICSI using flexible GnRH antagonist protocol, that adopt for scheduling their cycles, either a routine OCP pretreatment strategy (n = 350 women); or a routine no pretreatment strategy (pertaining awaiting spontaneous withdrawal bleeding) (n = 350 women).
- Women were included in the study according to the following inclusion criteria:
a- Age 18-39 years of age.
b- BMI 18-29 Kg/m2
c- Polycystic ovary syndrome (according to Rotterdam ESHRE/ASRM-Sponsored PCOS consensus workshop group, 2004)
d- First trial IVF/ICSI.
e- Informed consent of the patient to follow the routine pretreatment protocol of the IVF center.
- The following patients were excluded from the study:
a- Anatomical abnormalities; e.g., endometrial polyps, fibroids, endometrioma, ovarian cysts or hydrosalpinx.
b- Hormonal abnormalities not attributed to PCOS, e.g., hypothyroidism.
c- Abnormal male or female karyotyping.
d- Abnormal semen analysis parameters according to WHO criteria 2010.
e- Any hormonal or insulin-sensitizing treatment within the last 3 months.
f- Chronic medical disorders, e.g., hypertension, autoimmune disorders, … etc.
g- Women planning for elective freeze-all, preimplantation genetic diagnosis (PGT-M), or preimplantation genetic screening (PGT-A)
h- For women planned to be pretreated with OCP, any category 4 medical condition that contraindicated the usage of OCP according to the WHO Medical eligibility criteria (World Health Organization, 2015), e.g., smoking with age ≥ 35 years and cigarettes ≥ 15 /day, previous history or acute DVT/PE or any known thrombogenic mutations.
Women fulfilling the inclusion criteria were allocated, according the pretreatment protocol of the IVF center, to either two groups:
a. OCP Group:
b. No-pretreatment Group
Ovarian stimulation as per protocol then retrieved oocytes were inseminated either by ICSI or IVF/ICSI insemination, and embryos were cultured up to Day 3 or 5
Hormonal and ultrasound assessment; Vaginal ultrasound for the assessment of follicular and endometrial development, was performed concomitantly with hormonal assessment at each visit, or more frequently as indicated.
Outcomes were measured and monitored, the primary outcome being live birth.
Results of the statistical analysis are presented in the following two sections:
- Section A: The parameters and outcomes of COH in the OCP-pretreated and non-pretreated groups.
- Section B: Exploratory subgroup analysis for the effect of basal LH levels on the outcomes of COH in the OCP-pretreated and non-pretreated groups.
Section A includes:
I. Analysis of the parameters of COH in the two study groups
a. Pharmacologic parameters of COH: among women who completed COH, the total rFSH dose was statistically significantly higher in the OCP-pretreated group compared to the non-pretreated group. On the other hand, no significant difference could be detected between both groups regarding the type of trigger (hCG or agonist trigger) used before oocyte retrieval
b. Hormonal dynamics throughout COH (serum LH, estradiol and progesterone): OCP pretreatment lowered serum LH levels significantly throughout COH. This effect varied with the progression of time through COH. Serum estradiol (E2) levels were also statistically significantly different between OCP-pretreated and non-pretreated groups. Whereas serum E2
c. Endometrial dynamics throughout COH: endometrial thickness varied significantly between the two groups throughout the course of COH.
d. Folliculometric parameters of follicular recruitment and growth: a follicular cohort of the OCP group was more homogenous, compared to the no pretreatment group,
e. Parameters of cycle scheduling: the proportion of women, who successfully started the COH cycle using the assigned pretreatment protocol, was significantly higher with OCP pretreatment compared to the no pretreatment approach
ii. Analysis of the outcomes of COH and IVF/ICSI in the two study groups
d. Oocyte and embryologic outcomes: Among women who completed COH, the number of oocytes retrieved was statistically significantly higher in the OCP group compared to the no pretreatment group.
e. Pregnancy outcomes: an intention-to-treat basis, no statistically significant difference could be detected between the two groups regarding the incidence of biochemical pregnancy, clinical pregnancy, and live birth.
f. Adverse events related to COH: no statistically significant differences in the incidence of COH-related adverse events, were detected between the two groups.
In brief, our results suggested that OCP-pretreated PCOS women had more retrieved oocytes, albeit, on the expense of increased gonadotropin consumption. However, this increased oocyte yield did not translate to a significant effect on conception rates among intended-to-treat PCOS woman receiving OCP pretreatment. Conversely, it shifted among OCP-pretreated women who actually started COH cycle, to a significant 18% reduction in the probability of successful pregnancy.
An unsettled controversy has arisen over the last two decades, as contradictory data have been published on whether the use of OCP in the preceding cycle might have a deleterious effect on the pregnancy outcome of the subsequent IVF cycle. The debate between the proponents of either sides of the controversy is longstanding, and to-date is yet unsettled. A glimpse of this debate was outlined on the pages of the Reproductive Biomedicine Online journal over several issues (Griesinger et al., 2015; Garcia-Velasco and Fatemi, 2015a; Garcia-Velasco and Fatemi, 2015b).
A multitude of studies did not find a reduction in the probability of achieving successful pregnancy in GnRH antagonist cycles by OCP pretreatment in PCOS women (Ozmen et al., 2013; Kalem et al., 2017; Song et al., 2019), in normo-ovulatory infertile women (Kolibianakis et al., 2006; Rombauts et al., 2006; Cédrin-Durnerin et al., 2007; Tavmergen et al., 2009; Shahrokh Tehrani Nejad et al., 2018), and in all-cause-of-infertility women (including IVF/ICSI for male factor) (Bellver et al., 2007; Montoya-Botero et al., 2020).
On the other side of the controversy, the Cochrane systematic review (Farquhar et al., 2017) and the meta-analysis by Griesinger et al. (2010) analyzed the pooled data of the previous studies of normo-ovulatory women. According to the former, the rate of live birth or ongoing pregnancy was lower with OCP pretreatment, compared to no pretreatment (OR 0.74, 95% CI 0.58 to 0.95; 6 RCTs; 1335 women; I2 = 0%; moderate quality evidence). These figures suggest that, if the chance of a live birth or ongoing pregnancy following no pretreatment was assumed to be 27%, the chance following pretreatment with COCP would be between 18% and 26%. This translates to a number-needed-to-treat in order to harm one woman of approximately 20 women (95%CI: 10 – 100).
Thus, to sum up, whereas one side of the debate rely on several underpowered studies and two large retrospective observational studies, with the caveat of the burden of inherent bias; the other side of the debate (in line with the per protocol conclusion of our study) rely on two large meta-analyses and a secondary analysis of a large landmark clinical trial.
The simple judgment of the strength of evidence of each side is rather obvious; and it would be of interest to solidify the conclusions of our study by the undoubtful merits of meta-analyses, and the deleterious bias-prone nature of observational studies. However, it is not wise to ignore a debate between top-ranked practitioners, without careful analysis of the opposing view.