Michael P. Diamond, MD

	Ovulatory dysfunction is a major cause of infertility; ovulation induction is often the initial treatment. Traditionally, clinical practice has been that ovulation induction is initiated following administration of progestins to induce menses. This practice has not been supported by direct experimental data and recent findings challenges this routine practice. The analysis revealed that progestin withdrawal adversely affects outcomes of ovulation induction with clomiphene. This review attempts to account for this observation based on prior studies while recognizing the critical need for both prospective assessment of the effect of progestin-induced withdrawal bleeding prior to ovulation induction, and that proposed mechanisms of a detrimental impact require experimental verification.  In addition, while we await definitive determination of whether there is a direct procreational benefit of eliminating progestin withdrawal bleeding, such a practice would potentially allow for increased opportunities to conceive within a given period of time.

	In women with polycystic ovarian syndrome (PCOS), ovulatory dysfunction is one of the most common causes of infertility. Infertility often presents in PCOS in the setting of amenorrhea, irregular menses, and hirsutism.

)  When other significant infertility factors are excluded, ovulation induction is the initial treatment of choice for most anovulatory or oligo-ovulatory infertile women.

   Progestins are often used to induce a withdrawal bleed before ovulation induction, as described by the practice committee of The American Society for Reproductive Medicine.

  However, recent studies suggest that a progestin-induced withdrawal bleed may reduce conception and live birth rates in women undergoing ovulation induction with clomiphene citrate. This finding, if confirmed in appropriately designed and conducted prospective studies, has the potential to radically change our current management of anovulatory infertility, while presenting many important questions:

		Does a clomiphene cycle (or other ovulating inducing agent cycles) need to be preceded by some form of endometrial shedding? 

		If not, should a patient with spontaneous onset of menses wait before starting clomiphene? And how long should the wait last? 

		Also, are such findings relevant to women who do not have PCOS, and are they pertinent in frozen embryo transfer cycles?  Understanding the mechanism of how administration of progestins may negatively affect pregnancy outcome following ovulation induction is essential to answering these questions. 

	Unfortunately, little data exist regarding the effect of endometrial shedding on infertility treatment outcome in anovulatory women.  The traditional practice of inducing progestin withdrawal before initiation of ovulation induction has not been supported by evidence. Neither has the decision to avoid progestin withdrawal, until recently. Hurst et al

 described a novel clomiphene stair-step protocol in which, after cycles with ovulation failure, clomiphene was immediately increased and administered beginning approximately cycle day 20, rather than after induction of a withdrawal bleed.  The authors found that use of this stair-step protocol decreased time to ovulation compared to a traditional protocol. They therefore concluded that menses induction before clomiphene administration was not necessary in nonresponsive (i.e. non-ovulating) patients with PCOS (so-called clomiphene-resistant patients).

 However, this report did not describe conception or live birth outcomes. 

 retrospectively investigated the effect of endometrial shedding on ovulation induction outcomes (both conception and birth rates). A secondary analysis of the Reproductive Medicine Network (RMN), Pregnancy in Polycystic Ovary Syndrome I study sought to determine whether progestin administration to induce a withdrawal bleed immediately before initiation of ovulation induction would affect ovulation, conception, and live birth. Results revealed that a menstrual bleed immediately before the cycle, whether induced or not, is associated with a reduced chance of conception and live birth in a subsequent cycle of ovulation induction.  Although ovulation was observed more often in cycles preceded by spontaneous menses, the highest rates of conception and live birth were observed in cycles with anovulation without progestin withdrawal. Moreover, that relationship was observed when examining each individual treatment arm: clomiphene alone, metformin alone, and clomiphene plus metformin.

	Similarly, in a retrospective study, Dong et al

 sought to determine whether a progesterone-induced endometrial bleed before ovulation induction affected pregnancy rates. The control group included cycles with spontaneous menses. They found that patients in the induced shedding group had significantly thinner peak endometrium (n=184 cycles) compared to the controls (n=57 cycles).  Although, rates of pregnancy and live birth were higher in the progesterone withdrawal group compared to the control group, the difference was not significant.

Several theories have been proposed to explain the mechanism by which lack of endometrial shedding in anovulatory women could exert a positive effect on subsequent conception and live births. Potential mechanisms include the direct and indirect effect of a progestin on the hypothalamic-pituitary-ovarian axis, a direct effect on endometrial receptivity, and a direct or indirect effect related to endometrial thickness.

 Endometrial receptivity and morphological appearance is indicative of the functional status of the hypothalamus-pituitary-ovary axis; understanding the morphological changes and biomarkers expressed by the endometrium in response to progestin is critical in explaining potential mechanisms.

  Moreover, hypothalamic pituitary dysfunction (HPD) can result in luteal phase deficiency,

 which has been postulated to affect fertility, although that could not be demonstrated in an earlier RMN study.

	 The endometrium undergoes a complex series of organized proliferative and secretory changes in each menstrual cycle, leading to creation of a favorable environment for implantation.

 The window of implantation is a short period of endometrial receptivity that is primarily considered to be coordinated by estradiol and progesterone.  Human endometrial stromal decidualization is required for embryo receptivity, angiogenesis, and placentation.  The steroid hormones progesterone (P4) and estradiol (E2), through binding of estrogen and progesterone receptors (ER and PR), also play an important role in regulating decidua formation during pregnancy.  Moreover, functional studies indicate that decidual angiogenesis is mainly regulated by vascular endothelial growth factor A (VEGF-A) secreted from decidual stromal cells expressing PR; thus, greater expression of VEGF-A corresponds to high PR expression.

  In fact, first-trimester miscarriage, preeclampsia, placental failure and intrauterine growth restriction may all result from vascular disturbances when these signaling pathways and cellular coordination are disrupted.

  In addition, progesterone is one of the factors that regulate trophoblastic invasion, which is an important process in placentation and establishment of a successful pregnancy. Progesterone-induced binding factor (PIBF) and leptin were found to control trophoblastic invasion, aided by differential expression of PR.

  In fact, in the human endometrium, progestins suppresses leptin receptor mRNA expression via the PRs.

 Furthermore, Leptin receptor expression is lower in patients with PCOS and endometriosis and higher in patients with recurrent implantation failure, indicating the importance of regulation of endometrial expression.

 Expression of adipokines such as leptin and resistin is influenced by estrogen and progesterone and both have been noted to increase in the midluteal phase of the menstrual cycle during the window of implantation.

 Early progestin exposure may lead to altered expression of PR, or to PR resistance, resulting in disrupted vascular remodeling and placentation, and thus impact pregnancy outcome.  Moreover, a thin endometrium following a progesterone-induced withdrawal bleed may have impaired proliferation due to clomiphene citrate’s antagonistic properties at remaining endometrial ERs.  This could theoretically be in contrast to the effect of clomiphene when there has been no endometrial shedding, with a near normal endometrial thickness.  In such situations, endometrial ER antagonism by clomiphene may be less consequential, due to the pre-existing endometrial thickness.

	Current literature indicates that not only anovulation, but also endometrial dysfunction, contributes to infertility in women with PCOS.  Even when ovulation is restored, these patients exhibit a higher rate of implantation failure and spontaneous miscarriage.

 In a retrospective study, Bromer et al. evaluated the base endometrial development in a heterogeneous infertility population. Significant decrease in endometrial thickness was observed in women with PCOS despite correction for estradiol levels.  The authors also found that endometrium growth plateaus around day 9 regardless of treatment regimen, drug, underlying diagnosis or success.  This suggests that the endometrium can tolerate long estrogen exposure and optimal endometrial thickness precedes peak estradiol level.

	Gene expression analysis indicate that patients with PCOS who are treated with clomiphene to induce ovulation show evidence of progesterone resistance in the endometrium.

 In fact, biopsies performed on women during the secretory phase of a normal cycle and in women with PCOS before and after treatment with micronized progesterone revealed a lower number of glands and thicker luminal epithelium in the secretory phase in the women with  PCOS.

 Before treatment, the endometrium of patients with PCOS exhibited a slightly increased number of glands that transformed to histological delay after treatment (less developed glands and higher concentration of stromal component).

 The authors concluded that progesterone administration for 10 days failed to correct the dysfunctional endometrium; however, progesterone in fact may have an adverse effect on the endometrium as indicated by the lower number of glands in the secretory phase after therapy.

Moreover, androgen receptor and steroid receptor co-activators are over-expressed in the endometrium of women with PCOS, with a rise of AR expression observed in the late proliferative phase.

 AR expression is up-regulated by both circulating estrogens and androgens, and down- regulated by progesterone.

 In preparation for implantation, both ER and PR are down-regulated in the epithelial compartment, and AR is almost undetected in the secretory phase of normally cycling women. Also, biomarkers of endometrial receptivity such as αvβ3 integrin are down-regulated when epithelial AR is overexpressed.

 Although progesterone seems to down regulate AR expression, early exposure may alter the hypothalamic-pituitary-ovarian axis, resulting in a delay in the progesterone effect, and thereby leading to abnormal endometrial receptivity.

	In editorial remarks in response to the RMN secondary analysis

, Casper proposed that the decrease in pregnancy rate can be explained simply by the thinner endometrium in spontaneous or induced menses.

 The author argued that clomiphene, acting predominantly as an ER antagonist, is associated with thin endometrium. Thus, endometrial shedding results in a thin endometrium that is more likely to become ER depleted, and thus unresponsive after clomiphene treatment.  Consistent with this hypothesis, multiple studies have reported a positive correlation between endometrial thickness and pregnancy rates. Dickey et al. showed that endometrial thickness was negatively related to clomiphene, and an endometrium measurement of 9 mm or more was associated with higher pregnancy rates.

 These observations suggest that it is more likely that progesterone administration and endometrial shedding is able to adversely affect an anovulatory endometrium that is deregulated and unresponsive at the molecular level.  In addition, thick endometrium doesn’t always ensure successful implantation, and pregnancy can occur with thin endometrium.

 Sonographic detection of adequate endometrial thickness doesn’t always represent functional endometrium.

	As noted previously, the traditional practice of administering progestin to provoke a withdrawal bleed has no foundation in evidence-based practice. In addition, recent data raise important questions regarding the conventional wisdom that induction of a withdrawal bleed in anovulatory women prepares the endometrium for pregnancy initiation in a subsequent cycle. In fact, results suggest that there may be an advantage in avoiding initiation of progestin withdrawal. Because this discovery presents important implications in the practice of ovulation induction, the findings need to be definitively confirmed by appropriately designed and conducted clinical trials.  Although multiple mechanisms have been proposed to account for this observation, suggested molecular events at the endometrial level require further elucidation and experimental confirmation, which is likely to reveal involvement of more than one mechanism.  If in fact progestin-induced withdrawal bleeds in anovulatory/oligo-ovulatory women had a detrimental effect on pregnancy outcome following clomiphene ovarian stimulation, further studies will also be needed with other ovulation induction agents, as well as in regularly ovulating women.  Furthermore, the effect of progestin administration in frozen embryo transfer cycles in ovulatory or anovulatory women represents another clinical paradigm with a potential impact on pregnancy outcome.

	Dr Diamond reports that this paper was supported in part by HD 39005 to him.

	1. Practice Committee of the American Society for Reproductive M. Use of clomiphene citrate in women. 

Clinical gynecologic endocrinology and infertility.

 Philadephia: Lippincott Williams & Wilkins; 2005.

	3. Hurst BS, Hickman JM, Matthews ML, Usadi RS, and Marshburn PB. Novel clomiphene "stair-step" protocol reduces time to ovulation in women with polycystic ovarian syndrome. 

	4. Diamond MP, Kruger M, Santoro N, Zhang H, Casson P, Schlaff W, et al. Endometrial shedding effect on conception and live birth in women with polycystic ovary syndrome. 

	5. Dong X, Zheng Y, Liao X, Xiong T, and Zhang H. Does progesterone-induced endometrial withdrawal bleed before ovulation induction have negative effects on IUI outcomes in patients with polycystic ovary syndrome? 

	6. Casper RF. Detrimental effect of induced or spontaneous menses before ovulation induction on pregnancy outcome in patients with polycystic ovary syndrome. 

	7. Bergeron C. Morphological changes and protein secretion induced by progesterone in the endometrium during the luteal phase in preparation for nidation. 

	8. Valdez-Morales FJ, Gamboa-Dominguez A, Vital-Reyes VS, Cruz JC, Chimal-Monroy J, Franco-Murillo Y, et al. Changes in receptivity epithelial cell markers of endometrium after ovarian stimulation treatments: its role during implantation window. 

	9. Coutifaris C, Myers ER, Guzick DS, Diamond MP, Carson SA, Legro RS, et al. Histological dating of timed endometrial biopsy tissue is not related to fertility status. 

	10. Revel A. Defective endometrial receptivity. 

	11. Kim M, Park HJ, Seol JW, Jang JY, Cho YS, Kim KR, et al. VEGF-A regulated by progesterone governs uterine angiogenesis and vascular remodelling during pregnancy. 

	12. Miko E, Halasz M, Jericevic-Mulac B, Wicherek L, Arck P, Arato G, et al. Progesterone-induced blocking factor (PIBF) and trophoblast invasiveness. 

	13. Koshiba H, Kitawaki J, Ishihara H, Kado N, Kusuki I, Tsukamoto K, et al. Progesterone inhibition of functional leptin receptor mRNA expression in human endometrium. 

	14. Dos Santos E, Serazin V, Morvan C, Torre A, Wainer R, de Mazancourt P, et al. Adiponectin and leptin systems in human endometrium during window of implantation. 

	15. Asimakopoulos B, Milousis A, Gioka T, Kabouromiti G, Gianisslis G, Troussa A, et al. Serum pattern of circulating adipokines throughout the physiological menstrual cycle. 

	16. Gamundi-Segura S, Serna J, Oehninger S, Horcajadas JA, and Arbones-Mainar JM. Effects of adipocyte-secreted factors on decidualized endometrial cells: modulation of endometrial receptivity in vitro. 

	17. Shang K, Jia X, Qiao J, Kang J, and Guan Y. Endometrial abnormality in women with polycystic ovary syndrome. 

	18. Bromer JG, Aldad TS, and Taylor HS. Defining the proliferative phase endometrial defect. 

	19. Savaris RF, Groll JM, Young SL, DeMayo FJ, Jeong JW, Hamilton AE, et al. Progesterone resistance in PCOS endometrium: a microarray analysis in clomiphene citrate-treated and artificial menstrual cycles. 

	20. Lopes IM, Maganhin CC, Oliveira-Filho RM, Simoes RS, Simoes MJ, Iwata MC, et al. Histomorphometric Analysis and Markers of Endometrial Receptivity Embryonic Implantation in Women With Polycystic Ovary Syndrome During the Treatment With Progesterone. 

	21. Apparao KB, Lovely LP, Gui Y, Lininger RA, and Lessey BA. Elevated endometrial androgen receptor expression in women with polycystic ovarian syndrome. 

	22. Slayden OD, Nayak NR, Burton KA, Chwalisz K, Cameron ST, Critchley HO, et al. Progesterone antagonists increase androgen receptor expression in the rhesus macaque and human endometrium. 

	23. Dickey RP, Olar TT, Taylor SN, Curole DN, and Matulich EM. Relationship of endometrial thickness and pattern to fecundity in ovulation induction cycles: effect of clomiphene citrate alone and with human menopausal gonadotropin. 

	24. Paulson RJ. Hormonal induction of endometrial receptivity. 

Articles

Recent studies suggest that a progestin-induced withdrawal bleed may reduce conception and live birth rates in women undergoing ovulation induction with clomiphene citrate.

Author | Michael P. Diamond, MD

Articles

Is menses induction necessary before ovulation induction?