The combined oral contraceptive pill (OCP) currently being used by more than 100 million women worldwide1 contains both estrogen and progestin in various combinations. The main mechanism of action of the steroid is inhibition of follicular development and prevention of ovulation by pituitary and hypothalamic suppression of follicle-stimulating hormone and luteinizing hormone.12 Other secondary mechanisms of contraceptive activity include progestogenic induction of hostile cervical mucus and of an endometrial environment that is unfavorable for implantation.3 A predominant progestational effect of combined OCP on the endometrium includes an arrest of glandular proliferation, induction of pseudosecretion, and stromal edema followed by decidualized stroma with granulocytes and thin sinusoidal blood vessels. Prolonged use results in progressive endometrial atrophy.4 This latter property has been used for the treatment of dysfunctional uterine bleeding and endometrial hyperplasia. Currently, pretreatment with combined OCPs also is used for diagnostic and operative hysteroscopy in which the visualization of the endometrial cavity is facilitated by a thin endometrium.5
Proper endometrial development is vital for successful implantation of an embryo. Although controversial, it is generally considered that an ultrasonographic endometrial measurement of 7 mm in cross-section is the minimum thickness required for successful implantation.6 While monitoring women ultrasonographically during assisted reproductive techniques, we occasionally noticed that some women had a persistently thin endometrium with no obvious cause. On closer questioning, we observed a frequent history of long-term combined OCP use in the past in many of these women.
The objective of the present study was to estimate whether there was any association of long-term use of combined OCPs with adverse endometrial growth. We performed a study to examine the duration of combined OCP use and its effect on endometrial growth and thickness in a group of infertile women all undergoing the same endometrial estrogen preparation for frozen embryo transfer.
MATERIALS AND METHODS
The present study was performed at the Toronto Centre for Advanced Reproductive Technology, a University of Toronto–affiliated infertility center. Ultrasound monitoring flow sheets of patients undergoing frozen embryo transfer were reviewed from 2008 to 2010. The study was approved by the Institutional Research and Ethics Review Board at Mount Sinai Hospital, Toronto, Ontario.
A total of 488 frozen embryo transfer cycles was identified. Patients between 30 and 45 years of age were included. The same standard protocol for endometrial preparation was used in 295 cycles consisting of micronized 17 β-estradiol (E2) for simulated follicular phase endometrial development. We excluded cycles in women in whom a different protocol for frozen embryo transfer was used (natural cycle, estrogen patch) or who were supplemented with low-dose acetylsalicylic acid. Each patient was accounted for only once in the first cycle. Thus, we included a total of 207 single cycles in 207 women. We excluded patients who had used combined OCPs within the past 2 years. Patients using contraception other than combined OCPs, eg, injectable hormones, intrauterine devices, spermicidal jells, or barrier contraceptives were excluded. Patients having other associated pathology such as uterine fibroids, Asherman syndrome, previous septum resection, endometriosis, and polycystic ovaries were also excluded. With these inclusion and exclusion criteria, we had a study group of 137 patients undergoing frozen embryo transfer with a history of combined OCP use that was stopped more than 2 years previously (Fig. 1).
All the patients included in the study used the described endometrial preparation protocol. Tablets of micronized E2 (2 mg) were taken orally twice per day starting from day 3 of the menstrual cycle. Once vaginal bleeding stopped, the E2 tablets were inserted vaginally twice daily. On cycle day 8, the E2 dose was increased to two tablets twice daily, ie, a total daily dose of 8 mg vaginally. On cycle day 10, a transvaginal ultrasound scan was performed to assess the endometrial thickness and pattern and was repeated until the endometrial thickness was at least 8 mm with a triple line pattern. Progesterone was then added, 200 mg progesterone vaginal suppositories, three times per day and the E2 dose was reduced to one tablet twice daily (4 mg) orally. Frozen embryo transfer was performed either 5 days after starting progesterone (D3 cryopreserved embryos) or 7 days later (blastocyst cryopreservation).
On day 10 of the stimulated cycle, the maximal endometrial thickness in the sagittal plane was measured ultrasonographically using a 5-mHz endovaginal probe. Endometrial thickness less than 7 mm was considered thin, and a thickness equal to 7 mm or more was considered adequate. Thus, we divided the patients into two groups based on endometrial thickness and then we reviewed the clinical charts of both the groups and noted past duration of use of combined OCPs. In the literature, Farrow et al7 showed no adverse effect of more than 5 years of combined OCP use on women's fertility. Therefore, we initially defined long-term use as 10 years or more to examine whether there was any effect of combined OCP use on the endometrium. However, we also analyzed the data from 5 years of use. In the women with endometrial thickness less than 7 mm on day 10, we continued E2 at a dose of 8 mg per day for up to 1 week to estimate whether an increase in endometrial thickness could be obtained in this group. If the thickness was still less than 7 mm after another week of estrogen administration, then the cycle was cancelled. Additional demographic factors that were noted and later compared in both the groups were body mass index (calculated as weight (kg)/[height (m)]2), age at menarche, type of infertility (primary or secondary), and duration of infertility (Table 1).
All analyses were performed using the Statistical Package for Social Sciences and GraphPad Prism software and Excel 7. One-sample Kolmogorov-Smirnov test was conducted for testing normality. For continuous parameters in both groups that were normally distributed, parametric t tests were performed, whereas the nonparametric Mann-Whitney test by ranks was used to analyze the nonnormally distributed data. The χ2 and Fisher exact tests were used for categorical variables. A significance level of 0.05 was used for all comparisons. We categorized the patients into four groups: those with endometrial thickness less than 7 mm and those with endometrial thickness more than 7 mm, and those with more than 10 years of combined OCP use or less than 10 years of use. For this purpose, a two×two contingency table was prepared taking endometrial thickness and years of combined OCP use as binomial variables. Assuming a null hypothesis of no difference in the proportion of patients with more than 10 years of combined OCP use in the two endometrial thickness groups, we performed Fisher exact tests to assess significance of any difference (Table 2). Similarly, we used the same strategy to check the significance for 5 years of combined OCP.
We also calculated the mean endometrial thickness in patients using combined OCPs for more than 10 years and less than 10 years. The difference in the means and its significance was calculated. Finally, binary logistic regression was performed and the odds ratio was calculated, taking the years of combined OCP use as the independent variable and endometrial thickness of less than 7 mm and 7 mm or greater as the two possible outcomes (dependent variable). Box plots taking endometrial thickness and years of OCP use were also drawn, highlighting any outliers (Fig. 2).
Of the 488 patients with frozen embryo transfer between 2008 and 2010, 137 patients were within the same age group (30–45 years), had the same protocol for endometrial preparation before embryo transfer, and had a history of combined OCP use more than 2 years before treatment (Fig. 1).
As illustrated in Table 1, 30 out of the 137 patients (21.9%) had an endometrial thickness less than 7 mm on day 10 of the frozen embryo transfer cycle. These patients comprised the study group with mean endometrial thickness of 6.06±0.62 mm. The control group comprised 107 (78.1%) patients who had an endometrial thickness of 7 mm or more. The mean endometrial thickness in this group was 10.03±1.45 mm. The P value for endometrial thickness was highly significant (P<.001; Table 1). The mean age in the less than 7 mm endometrial thickness group was 36.27±3.40 years, and in the other group it was 36.06±3.32 years. Age and body mass index were normally distributed; hence, Student t test was used for comparison. The P values were .63 and .30, respectively (Table 1). Likewise, no significant difference was observed between degree of infertility (primary or secondary) and duration of infertility in both groups (as shown in Table 1). However, the age at menarche was 0.5 years later in the less than 7 mm group (P=.01; Table 1). The mean years of combined OCP use in the less than 7 mm endometrial thickness group was found to be 9.8±4.54, and in more than or equal to 7 mm group it was 5.8±4.52. The difference in means of duration of combined OCP use was highly significant between the two groups as tested by Mann-Whitney test by ranks (P<.001; Table 1). Taking 10 years of use as a threshold for long-term combined OCP use, 63.35% of patients were long-term users in the less than 7 mm endometrial thickness group whereas 28.04% were long-term users in the 7 mm and above endometrial thickness group. This difference was also highly significant (P<.001) by Fisher exact test (Table 2). The odds ratio was 4.43 (95% confidence interval 1.89–10.41), indicating that those patients with a history of combined OCP use of less than 10 years had 4.43-fold odds of having optimum endometrium compared with those who had used OCPs for 10 years or more.
We also analyzed the data taking 5 years of combined OCP use as the threshold based on a previous study using this definition7 and found that 5 years of use also had a significant adverse effect on endometrial thickness. In the 7 mm or greater thickness group, 56.1% had used the OCP for more than 5 years compared with 86.6% in the less than 7 mm endometrial thickness group (P=.002 by Fisher exact test; Table 3). Alternatively, to consolidate these findings further, we calculated the mean endometrial thickness on cycle day 10 in patients using combined OCP for less than 10 years and for 10 years or more. The mean thickness was 9.54±1.88 mm and 8.48±2.33 mm, respectively, with P=.007. The mean endometrial thickness for those who had used combined OCP for less than 5 years was 9.72±1.69 mm, whereas for those with 5 or more years of use, the mean thickness was 8.81±2.23 mm (P=.008). Figure 2 shows the box plot with 5 years of combined OCP use as the cutoff. The 50th percentile of endometrial thickness in less than 5 years of use was 8.5–11 mm, whereas it was 6.7–10.75 mm in 5 or more years of OCP use. There were no outliers in the 5 or more years of use group. In the less than 5 years group, there was one high outlier and three low outliers.
As secondary outcome measures, we calculated cycle cancellation and pregnancy rates. We found that 7 of 30 cycles (23%) were cancelled in the study group because of persistently thin endometrium despite continuing estrogen stimulation for 7 more days. The rest of these women completed the cycle with prolonged treatment with estrogen. The mean day of starting progesterone in the 7 mm or greater endometrial thickness group was 10.06, whereas it was 14.43 in the less than 7 mm endometrial thickness group. Thus, an extra 4 days of estrogen was needed in the thin endometrium group to reach the required 7-mm-thick endometrium before starting progesterone (P<.001; Table 1). In the 7 mm or greater endometrial thickness group, four cycles were cancelled for various reasons other than thin endometrium. The overall cycle cancellation rate was 4% (4 of 107), which is significantly less than the thin endometrium group in which the cycle cancellation rate was 23% (7 of 30; P=.002; Table 1). The pregnancy rate in the less than 7 mm endometrial thickness group (including patients in whom up to 7 more days of estrogen was required to thicken the endometrium) was 13% (4 of 30 patients), whereas it was 27% (29 of 107 patients) in the 7 mm or greater endometrial thickness group (P=.15; Table 1).
In the present study, the endometrial thickness was recorded in women of similar ages who received the same endometrial preparation for frozen embryo transfer. All had a history of combined OCP use of varying duration, and all had stopped the combined OCP 2 years before frozen embryo transfer.
The only literature we could find addressing long-term combined OCP use (defined as 5 years or more) and fertility suggests that there is no adverse effect on endometrial growth or pregnancy outcome.7 In fact, the authors observed a reduced time to pregnancy in women using combined OCPs for 5 or more years compared with less than 5 years or no use at all. The authors speculated that the combined OCP may prevent the possibility of endometriosis progression by minimizing endometrial proliferation and menstrual bleeding, thereby improving the chance to conceive. It is known that there are other health benefits of combined OCP use, such as a reduced risk of endometrial,8 ovarian,8 and colorectal cancer.9 The ovarian and endometrial benefits appear to persist for more than 15 years after stopping OCPs. The reduced risk of endometrial cancer was attributed to the antiproliferative actions of progestin on the endometrium.8,9 A later study hypothesized that the 50%–60% reduced endometrial cancer risk with the use of combined OCP or progestin-releasing intrauterine device was partly attributable to negative selection for subclinically mutated glands (phosphatase and tensin homolog mutated glands).10 This mechanism cannot explain a suppressed endometrial growth in normal endometrium. However, the observation that the protective effect against endometrial cancer persists for a long time in combined OCP users might support our findings that long-term combined OCP use in the past somehow hampers optimal endometrial growth.
Because most of the effects of estrogen on the uterus are mediated via estrogen receptors,11 there is a possibility that there is deficiency of estrogen and progesterone receptors in a suboptimal endometrium. However, it also has been established that estrogen or progesterone receptor concentrations were not related to endometrial thickness and receptivity, and that the only significant correlation is with the relative concentrations of these receptors.12 For example, the ratio of progesterone receptor to estrogen is decreased in endometrium with a supposedly favorable pattern for implantation compared with endometrium with a suspected adverse pattern on ultrasonography. Clomiphene citrate, a selective (estrogen) modulator with both estrogen agonist and antagonist properties, has been demonstrated to result in an endometrium that is thinner than that seen with other stimulated cycles,13,14 most likely because of depletion of estrogen receptor levels in the endometrium. Clomiphene citrate inhibits estrogen-induced endometrial epithelial cell proliferation and estrogen response element transactivation, thereby inhibiting the recruitment of steroid receptor coactivator-1 to estrogen receptor α.15 There may be a similar mechanism explaining the long-term effects of combined OCP use, although at present there are insufficient data to validate this suggestion.
Adequate endometrial growth is important for embryo implantation. Most investigators suggest that an endometrial thickness of 7 mm or more is preferable for successful implantation during fertility treatments.13,16–18 Some studies state that a minimum endometrial thickness of 6 mm18 is acceptable as a prerequisite for pregnancy, and yet another study reported pregnancy with endometrium as thin as 4 mm.19 In a previous study of infertile women receiving estrogen replacement, an endometrial thickness of 7 mm or more was more likely to have an in phase endometrium compared with an endometrial thickness of less than 7 mm.20 In our study, we found that the average endometrial thickness in the thin endometrium group was 6.06±0.62 mm, whereas it was 10.03±1.45 mm in the 7 mm or greater endometrial thickness group. With longer duration of treatment with estrogen, the thin endometrium group may catch up and frozen embryo transfer is possible, although further along in the cycle. A total of seven cycles had to be cancelled in the less than7 mm group because of persistently thin endometrium.
Currently, much focus is on endometrial stem cells and their important role as the “regenerator” of the human endometrium. Although endometrial growth and differentiation are primarily regulated by estrogen and progesterone, it is postulated that the clonogenic activity of the endometrial epithelial and stromal cells is actually responsible for the remarkable regenerative capacity of the human endometrium. Several growth factors have been identified to support clonogenicity.21 Kjiana et al21 demonstrated for the first time to our knowledge that inactive endometrium, ie, endometrium of women using oral contraceptive therapy and postmenopausal women, also contains clonogenic epithelial and stromal cells. The endometrium regenerates on cessation of oral contraceptive therapy and, in postmenopausal women, when hormone replacement therapy is started. However, they also demonstrated that the cloning efficiencies were lower in inactive endometrium than either the proliferative or the secretory phase endometrium. This suggests that the number of clonogenic cells increases with cycling21 and supports the possibility that with a prolonged period of inactivity there is significant suppression of the stem cells, so that the endometrium fails to recover even with estrogen stimulation or needs a much prolonged or higher-dose stimulation for optimal functioning.
We acknowledge that the present study is limited by several factors. The sample size is small and we also lack information on the exact composition of the combined OCP (dose and type of estrogen and progestin) used. At present, we do not have knowledge of the mechanism of this potential adverse effect of combined OCPs and a biologic explanation for this phenomenon awaits further research. However, we believe our findings of an association between long-term use (5 years or more) of combined OCPs and persisting thin endometrium in women undergoing frozen embryo transfer cycles demonstrates an unusual and, until now, unidentified side effect of combined OCPs, leading to a higher cancellation rate and more days of stimulation. These findings may be of importance when counseling women considering long-term contraception.
1. Trussel J. Contraceptive efficacy. In: Hatcher RA, editor. Contraceptive technology. 19th revised ed. New York (NY): Ardent-Media; 2007.
2. Speroff L, Darney PD. Oral contraception. In: A clinical guide for contraception. 4th ed. Philadelphia (PA): Lippincott Williams & Wilkins; 2005. p. 21–138.
3. Rivera R, Yacobson I, Grimes D. The mechanism of action of hormonal contraceptives and intrauterine contraceptive devices. Am J Obstet Gynecol 1999;181:1263–9.
4. Deligdisch L. Hormonal pathology of the endometrium. Mod Pathol 2000;13:285–94.
5. Grow DR, Iromloo K. Oral contraceptives maintain a very thin endometrium before operative hysteroscopy. Fertil Steril 2006;85:204–7.
6. Oliveira JB, Baruffi RL, Mauri AL, Petersen CG, Borges MC, Franco JG Jr. Endometrial ultrasonography as a predictor of pregnancy in an in-vitro fertilization programme after ovarian stimulation and gonadotrophin-releasing hormone and gonadotrophins. Hum Reprod 1997;12:2515–8.
7. Farrow A, Hull MG, Northstone K, Taylor H, Ford WC, Golding J. Prolonged use of oral contraception before a planned pregnancy is associated with a decreased risk of delayed conception. Hum Reprod 2002;17:2754–61.
8. Collaborative Group on Epidemiological Studies of Ovarian Cancer, Beral V, Doll R, Hermon C, Peto R, Reeves G. Ovarian cancer and oral contraceptives: Collaborative reanalysis of data from 45 epidemiological studies including 23,257 women with ovarian cancer and 87,303 controls. Lancet 2008;371:303–14.
9. Cogliano V, Grosse Y, Baan R, Straif K, Secretan B, El Ghissassi F, et al.. Carcinogenicity of combined oestrogen-progestagen contraceptives and menopausal treatment. Lancet Oncol 2005;6:552–3.
10. Lin MC, Burkholder KA, Viswanathan AN, Neuberg D, Mutter GL. Involution of latent endometrial precancers by hormonal and nonhormonal mechanisms. Cancer 2009;115:2111–8.
11. Weihua Z, Saji S, Makinen S, Cheng G, Jensen EV, Warner M, et al.. Estrogen receptor (ER) beta, a modulator of ERalpha in the uterus. Proc Natl Acad Sci U S A 2000;97:5936–41.
12. Ohno Y, Fujimoto Y. Endometrial oestrogen and progesterone receptors and their relationship to sonographic appearance of the endometrium. Hum Reprod Update 1998;4:560–4.
13. Randall JM, Templeton A. Transvaginal sonographic assessment of follicular and endometrial growth in spontaneous and clomiphene citrate cycles. Fertil Steril 1991;56:208–12.
14. Gonen Y, Casper RF. Prediction of implantation by the sonographic appearance of the endometrium during controlled ovarian stimulation for in vitro fertilization (IVF). J In Vitro Fert Embryo Transf 1990;7:146–52.
15. Amita M, Takahashi T, Tsutsumi S, Ohta T, Takata K, Henmi N, et al.. Molecular mechanism of the inhibition of estradiol-induced endometrial epithelial cell proliferation by clomiphene citrate. Endocrinology 2010;151:394–405.
16. Friedler S, Schenker JG, Herman A, Lewin A. The role of ultrasonography in the evaluation of endometrial receptivity following assisted reproductive treatments: A critical review. Hum Reprod Update 1996;2:323–35.
17. Kovacs P, Matyas S, Boda K, Kaali SG. The effect of endometrial thickness on IVF/ICSI outcome. Hum Reprod 2003;18:2337–41.
18. Coulam CB, Bustillo M, Soenksen DM, Britten S. Ultrasonographic predictors of implantation after assisted reproduction. Fertil Steril 1994;62:1004–10.
19. Sundstrom P. Establishment of a successful pregnancy following in-vitro fertilization with an endometrial thickness of no more than 4 mm. Hum Reprod 1998;13:1550–2.
20. Hofmann GE, Thie J, Scott RT Jr, Navot D. Endometrial thickness is predictive of histologic endometrial maturation in women undergoing hormone replacement for ovum donation. Fertil Steril 1996;66:380–3.
21. Kjiana E, Schwab R, Wah SC, Gargett CE. Putative stem cell activity of human endometrial epithelial and stromal cells during the menstrual cycle. Fertil Steril 2005;84(Suppl 2):1124.