Effect of Progesterone Supplementation on Natural Frozen-Thawed Embryo Transfer Cycles: A Randomized Controlled Trial

Cryopreserved-thawed embryo transfer began in 1983 and became a popular, vital component of assisted reproduction technology (1). The transfer of a frozen embryo enhances the cumulative pregnancy rate, decreases cost, Is easy to perform and can be fulfilled successfully in a relatively shorter time span in comparison with repeated fresh cycles (2-5). Furthermore, endometrial receptivity can be compromised by controlled ovarian hyperstimulation (COH) protocols (6) and secretory endometrial transformation (7). Endometrial development in frozen-thawed cycles can be controlled more than during COH cycles (8).

Various protocols (gonadotropin/GnRH agonists, clomiphene citrate, or exogenous estrogen and progesterone) have been discussed in literature reviews with regards to the endometrium preparation for frozen-thawed embryo transfer (3, 9). The most prevalent protocol for frozen-thawed embryo transfer is the natural cycle or endometrial preparation with exogenous estrogen and progesterone, with or without the addition of a GnRH agonist (10-12).

Because the natural cycle protocol does not require exogenous hormones‚ it is favored by many patients (13). It has been observed that temporal characteristics of the endometrium such as the formation of pinopodes (markers of endometrial receptivity) are out-of-phase according to measurements in normal females who have been placed on exogenous steroids (14). Thus, the transfer of frozen-thawed embryos in natural cycles is a favored option for women with normal ovulatory menstrual cycles (15).

There is an idea that the endogenous production of progesterone is enough to support implantation in a natural cycle. However, an inadequate progesterone level at the time of implantation or during early pregnancy may happen naturally due to luteal phase deficiency (LPD), which can result in infertility or abortion (16).

The reported frequency of LPD ranges from 3.7% to 20% among infertile patients (17, 18). The frequency has been demonstrated to be approximately 8.1% in natural cycles in normoovulatory patients with primary or secondary infertility (19). Thus, women who undergo frozen- thawed embryo transfers may have sub optimal endometrium during their natural cycles. There is limited information about the effect of luteal phase supplementation on pregnancy rate in natural frozen-thawed embryo transfer cycles. Therefore, we have designed a prospective randomized study to verify if pregnancy rates could be enhanced with progesterone supplementation during the luteal phase and early pregnancy following a frozen-thawed embryo transfer in a natural cycle.

There were no significant differences noted in the fertilization rate between study and control groups (55.4% vs. 64.3%; p=0.16). Of the 102 patients included in this study, 51 received progesterone and the other 51 did not. Table 1 describes the basic characteristics of the patients in the two groups. The demographic parameters were similar in both groups in terms of age, basal FSH levels‚ body mass index (BMI)‚ the number of previous cycles‚ etiology of infertility, and infertility duration. Table 2 compares the previous fresh cycle characteristics in the two groups.

The mean number of oocytes retrieved‚ mean number of mature oocytes and the number of embryos obtained and vitrified did not differ between the groups. There were no significant differences noted in the fertilization rate (55.4% vs. 64.3%; p=0.16). In addition, the previous stimulation protocols and fertilization procedures were similar in the two groups. Only ejaculated sperms had used for conventional IVF or intracytoplasmic sperm injection and percent of sperms with progressive motility and sperms with normal morphology‚ also sperm count were not different in those groups. There was no significant difference observed between the groups regarding the reasons for embryo freezing. Table 3 compares the cycle characteristics of the two groups. Endometrial thickness and estradiol‚ progesterone and LH levels on the day of hCG administration were similar between groups.

The cycle length until the day of hCG administration‚ number of embryos transferred, and the number of good-quality embryos did not differ in the two groups. Table 4 presents a comparison of the pregnancy outcomes of the study groups. Again, no statistically significant differences were observed in the clinical pregnancy rate between the groups (33.3% vs.27.5%, p=0.66). Although there was a trend toward an increased clinical pregnancy rate with luteal supplementation‚ the difference was not significant. There were no differences between the implantation rates (16.6% vs. 15.3%‚ p=0.93) or clinical abortion rates (11.8% vs.14.3%‚ p=0.83). The flowchart of the study is shown in figure 1.

The granulosa cells of the developing follicle generate estradiol in response to gonadotropin stimulation in natural cycles. The endometrium acquires receptivity to embryo implantation by responding to progesterone action on an appropriately primed endometrium. Estrogenic stimulation would result in endometrial proliferation and the induction of progesterone receptors. The endometrium undertakes profound conformational and biochemical changes, from proliferative to secretory, with a concomitant induction of endometrial receptivity and opening of the window of implantation in response to progesterone (20). During the implantation window, the endometrium which is unexpectedly unreceptive towards embryo implantation acquires a functional condition useful to blastocyst reception (21).

The transfer of frozen-thawed embryos has important implications for the management of women undergoing ovarian hyperstimulation for IVF (2). Frozen embryo transfer is reported to be successful during the natural cycle after spontaneous ovulation according to the literature (22). In a study by Morozov et al. a higher pregnancy rate was observed in recipients who underwent natural cryothaw cycles than in hormone replacement treatment cycles. In their study the level of estradiol was greater in the substitution cycles when compared with the natural cycle. Regarding those results, we have supported the theory that the window of uterine receptivity closes earlier at a higher endogenous estrogen level and limits the time for the transferred embryos to implant successfully (9). According to their results,hormone replacement treatment versus the natural cycle for cryothaw embryo transfer was associated with decreased pregnancy rates. In the current study, we have evaluated the outcome of hCG-induced natural cryothawed embryo transfer cycles that were supported during the luteal phase with IM progesterone. We compared this with the outcome of hCG-induced natural cycles in the absence of luteal phase support.

Our hypothesis was that progesterone support has a beneficial effect on pregnancy rate after frozen embryo transfer in natural cycles, but the results did not support our hypothesis. In our study, hCG was used for final oocyte maturation. It was suggested that hCG administered for the final oocyte maturation in stimulated IVF cycles would cause a luteal phase defect by suppressing LH production through a short-loop feedback mechanism (23) although the use of hCG did not down-regulate LH secretion in the luteal phase of regular and unstimulated cycles in women with normal ovulation (24). Additionally, in our study none of the patients developed premature luteinization. Premature LH surge is defined as an LH level of ≥10 IU/L and a progesterone level of ≥1.0 ng/ml on the day of hCG administration (25). An elevated progesterone level advances the endometrium‚ therefore the replacement of day 3 embryos occur in an asynchronous endometrium with subsequent failure of establishing an embryo-endometrium cross-dialog, resulting in implantation failure (26).

Bourgain et al. have reported that progesterone induces a secretory transformation of the endometrium in the luteal phase (27) and by inducing this change after sufficient estrogen priming, progesterone improves endometrial receptivity (28). Progesterone not only supports endometrial development but also maintains embryo survival by shifting the immune system toward the production of non-inflammatory Th2 cytokines (29, 30). In addition, by inducing nitric oxide synthesis in the deciduas‚ they intensify local vasodilatation and uterine repose (31). A study by Orvieto et al. has shown that, in artificial cryothawed embryo transfer cycles, a highdose progesterone supplementation in the luteal phase resulted in a higher clinical pregnancy rate (32).

In contrast to our study, Bjuresten et al. have reported that progesterone supplementation improved the live birth rate after embryo transfer in natural cycles (15). In their study, women received vaginal progesterone at a dose of 400 mg twice a day from the day of the embryo transfer. They attributed the increase in live birth rate to the effects of vaginal progesterone. Vaginal progesterone results in adequate endometrial development, in spite of low serum progesterone levels.

Our study was in agreement with a study by Kyrou et al. that reported luteal phase support did not affect ongoing pregnancy rates in natural hCG-induced frozenthawed embryo transfer cycles (33). A possible reason for our finding was that the women in the present study had a normal ovulatory function; those with ovulatory dysfunction were excluded from the study. Luteal phase defect in stimulated IVF cycles is due to supra physiological levels of steroids which directly inhibit the LH release via negative feedback actions at the hypothalamic- pituitary axis level (34). However it seems that LPD is not a main etiologic factor for implantation failure in natural frozen thawed embryo transfer cycles.

There emerged no significant differences between the two groups in our study with regards to the implantation or clinical pregnancy rates‚ but there was a trend toward an increased clinical pregnancy rate with luteal supplementation. Thus, further studies are needed to confirm our findings.