DOI: https://doi.org/10.21203/rs.3.rs-1965716/v1
Purpose: This randomised clinical pilot study evaluated the effect of the mid-luteal additional single dose of gonadotropin-releasing hormone agonist (GnRH-a) on the clinical outcome of the females subjected to artificial cycle frozen-thawed embryo transfer (AC-FET).
Methods: A total of 129 females were randomized into two groups (70 into the control group and 59 into the intervention group). Both groups received the standard luteal support. The intervention group was given an extra dose of 0.1 mg GnRH-a in the luteal phase. The live birth rate served as the primary endpoint. The secondary endpoints were the positivity of pregnancy tests, the clinical pregnancy rate, the miscarriage rate, the implantation rate, and the multiple pregnancy rate.
Results: There were more positive pregnancy tests, clinical pregnancies, live births, and twinning pregnancies, and less miscarriage observed in the intervention arm comparing to the controls, though no statistical significance was concluded. However, the difference of 12.1 percentage points in live births rate (40.7% vs 28.6%) would be relevant in clinical practice.
Conclusions: Overall, the distinct, however statistically insignificant, improvement of the pregnancy outcome supports the non-inferiority of the luteal phase GnRH-a support in AC-FET. The beneficial effects need to be further validated by larger scale clinical trials.
With the success of improved cryopreservation techniques, pregnancy rates between fresh and frozen-thawed embryo transfer (FET) cycles are close to equal (1). It have been proposed that the ‘freeze all’ policy is challenging the current management practices in assisted reproduction treatment (ART). The artificial cycle (AC) is a prevalent option for endometrial preparation in FET, due to its clinical practicability and less demand for monitoring. Due to the lack of corpus luteum, luteal phase support (LPS) in AC-FET is essential to induce the endometrial receptivity and to maintain the on-going pregnancy. However, luteal phase defect (LPD) may result from the suboptimal responsiveness of the endometrium to EXOGENOUS progesterone" (2).
As the supplementation for luteal phase support, Gonadotropin-releasing hormone agonist (GnRH-a) improves the implantation rate and the pregnancy rate(3). In 2004, Tesarik et al. conducted the first prospective controlled study to evaluate the effects of GnRH-a administration as luteal support on AC for fresh embryo transfers in recipients of donated oocytes(4). Since then, several studies have shown the benefits of the administration of GnRH-a in the luteal phase on the pregnancy outcomes of the fresh cycle(5–9). More recently, A prospective, randomized trial showed that repeated doses of GnRH-a on every other day supplied safe and effective luteal support in GnRH antagonist IVF cycle (10).
There are differences in hormonal regulation and corpus luteum function in different embryo transfer protocols. The existing evidence is based on the studies of the fresh ART cycles. There is a limited number of studies regarding the mid-luteal GnRH-a support in FET cycles. Preliminary data suggest a benefit of the administration of GnRH-a as a luteal support in natural cycles (NC) FET(11, 12); however, the effect of GnRH-a in AC-FET remains controversial(13–16). This study is a prospective randomized clinical pilot trial that evaluated the efficacy of the additional single-dose of GnRH-a at the time of implantation on the pregnancy outcome of patients undergoing hormonally substituted AC for FET.
This study is a prospective clinical pilot trial with randomization, approved by the Ethics Committee of the Reproductive Center of the First Hospital of Changde City. The study complied with the principles of the Declaration of Helsinki. All patients were well-informed and the formal consent was acquired. The females were recruited between July 2018 and December 2018 from the Reproductive Center of the First Hospital of Changde City, China.
A total of 156 patients were recruited for AC-FET. Exclusion criteria included female over 40 years of age or follicle-stimulating hormone (FSH) ≥ 20IU/L, uterine anomalies, intramural myomas (≥ 4 cm), submucous fibroids, endometrium thickness less than 7 mm prior to embryo transfer, patients who had untreated systemic or endocrine disorders such as diabetes mellitus, thyroid dysfunction or hyperprolactinemia and female or male chromosomal abnormality.
All participants were assigned to the intervention or control group according to the random number table on the day of embryo transfer. Each female was enrolled and studied only once. Patients were administered estradiol valerate range 6 to 8 milligrams daily from day 2 or day 3 of the cycle, followed by daily progesterone injections (total dose 80 mg) when the endometrial thickness reached 7mm at least. The thickness of the endometrium was assessed with a transvaginal ultrasound. Randomization was done on the day of embryo transfer: A single dose of 0.1 mg GnRH-a (Decapepty l; Ferring, Germany) was injected subcutaneously to patients in the intervention group (n = 59) when the age of embryo was six. The patients in the control group (n = 70) did not receive it. Both groups received daily progesterone injections (total dose of 80 mg) and estradiol valerate 6–8 mg. The serum β-subunit of human chorionic gonadotropin (β-hCG) concentration was used to diagnose for chemical pregnancy on 14 days after the day-three embryo transfer and 12 days after the day-five blastocyst transfer. The clinical pregnancy was confirmed by the detection of fetal heart activity using transvaginal ultrasound on the 14 days after a positive hCG test. Hormone replacement therapy was administered until either the negative pregnancy test or the tenth week of gestation.
The endpoint was live birth. The main outcome evaluations included the rates of the positive pregnancy test, the clinical pregnancy, the miscarriage rate, and the live birth rate. The rates for implantation, chemical pregnancy, and multiple pregnancies were documented. The serum β-hCG level of the patients and the weight of the newborns were recorded as well.
The statistical analysis was performed using Graphpad, version 6.0. T-tests and Chi-square test were used for comparing differences between categorical data. P < 0.05 was considered statistically significant.
A total of 129 patients who met the inclusion criteria were enrolled (59 in the intervention group and 70 in the control group). The baseline clinical characteristics of the two study groups are shown in Table 1. No significant difference was observed in age, body mass index (BMI), FSH, duration of infertility, primary infertility rate, endometrial thickness at transfer day, number of transferred embryos, completely survived embryos rate and number of good quality embryos (Table 1).
Variable | Intervention group (n = 59) | Control group (n = 70) | P value | |
---|---|---|---|---|
Female age at FET | 30.14 ± 0.43 | 30.21 ± 0.42 | 0.897 | |
BMI (kg/m2) | 22.67 ± 0.46 | 23.55 ± 0.50 | 0.206 | |
FSH (mIU/mL) | 8.13 ± 0.34 | 7.36 ± 0.31 | 0.094 | |
Duration of infertility (year) | 3.64 ± 0.28 | 4.10 ± 0.37 | 0.326 | |
Primary infertility (%) | 29 (49.2) | 35 (50) | 0.924 | |
Etiology for infertility (%) | 0.724 | |||
Female only | 50 (84.7) | 59 (84.3) | ||
Male | 4 (6.8) | 3 (4.3) | ||
Combined | 5 (8.5) | 8 (11.4) | ||
Endometrial thickness at transfer day (mm) | 9.79 ± 0.22 | 11.25 ± 1.51 | 0.343 | |
Number of transferred embryos/FET cycle | 2.02 ± 0.07 | 2.00 ± 0.07 | 0.863 | |
Completely survived embryos (%) | 124(98.4) | 164(98.2) | 0.750 | |
Number of good quality embryos | 1.32 ± 0.12 | 1.31 ± 0.11 | 0.962 | |
Number of FET cycles with blastocyst (%) | 3 (5.1) | 5 (7.1) | 0.726 | |
Values are mean ± SD unless stated otherwise. BMI: body mass index; FET: frozen embryo transfer. |
A slightly higher β-hCG level and increased implantation rate were observed in the intervention group comparing to the controls, although the differences were not statistically significant. There was a higher rate of positive pregnancy tests in the intervention group than the rate of the control group (57.6% vs 42.9%). Correspondently, the clinical pregnancy rate was 9.2 percentage points higher in the intervention group comparing to the controls (49.2% and 40.0%); Females in the intervention group had a 12.1 percentile higher live birth rate (LBR) versus the control group (40.7% vs 28.6%). Correspondently, the miscarriage rate was 14.7 percentage points lower (10.3% vs 25%). The differences were notable in the clinic although no statistical significance was concluded. From the intervention group, 59 females gave birth to 30, while only 21 births were counted in 70 females in the control group. The number of macrosomia in both the intervention group and the control group is similar. No difference was observed in the numbers of the chemical pregnancy, the ectopic pregnancy and the twining pregnancy between the two groups. (Table 2).
Variable | Intervention group (n = 59) | Control group (n = 70) | P value |
---|---|---|---|
Positive pregnancy (%) | 34/59(57.6) | 30/70(42.9) | 0.095 |
β-hCG (IU/L) | 1338 ± 234.8 | 976.5 ± 130.8 | 0.184 |
Implantation rate (%) | 36/119(30.3) | 31/140(22.1) | 0.138 |
Clinical pregnancy (%) | 29/59(49.2) | 28/70(40) | 0.374 |
Live birth | 24/59(40.7) | 20/70(28.6) | 0.208 |
Chemical pregnancy (%) | 4/34(11.8) | 2/30(6.7) | 0.433 |
Miscarriage (%) | 3/29(10.3) | 7/28(25) | 0.269 |
Ectopic pregnancy (%) | 2/29(6.9) | 1/28(3.6) | 0.975 |
Twinning pregnancy (%) | 8/29(27.6) | 3/28(10.7) | 0.162 |
Macrosomia (%) | 3/30(10) | 2/21(9.5) | 0.673 |
This prospective randomized clinical pilot trial investigated the effects of GnRH-a as an addition to progesterone luteal support on implantation, clinical pregnancy and LBR, and reported the live birth and the perinatal outcomes following the administration of single-dose GnRH-a in the luteal phase of AC-FET. Recently, Ye et al. provided evidence showing that GnRH-a administration in AC-FET cycles did not increase clinical or ongoing pregnancy. However, the result showed the implantation rate was significantly higher in 35 ~ 37 years old females with GnRH-a. They suggested that GnRH-a add-up could improve the implantation rate in the peri-implantation window in aging females (about 37 years old) through a direct effect on the embryo and improving embryo developmental potential(16). In our study, better pregnancy outcomes were achieved in the intervention group. The positive pregnancy rate, implantation rate, and clinical pregnancy, as well as live birth, were higher, although the differences ranged from 8.2 to 14.7 percentage points were not statistically significant. At the same time, we found that the miscarriages were less frequent with additional GnRH-a, which accounted for the increase of LBR. Apart from the risk of miscarriage, it should be noted that the rate of twin pregnancy in the intervention group was slightly higher than that in the control group. Zhou et al. conducted a retrospective analysis to investigate the effects and safety of mid-luteal GnRH-a support. The result indicated that the GnRH-a-added group had a slightly higher twin pregnancy rate and a significantly higher rate of premature delivery, but no obvious long-term effect on the neonates(17). Therefore, with the precautions taken to control the number of implanted embryos and reduce the incidence of twinning pregnancy, mid-luteal GnRH-a administration is relatively effective and safe.
The recent meta-analysis, which includes 13 RCTs with 3,584 cycles, indicated that the females in IVF/ICSI (Intracytoplasmic Sperm Injection Cycles) received GnRH-a for luteal support had a significantly higher implantation rate and higher rates of pregnancy, clinical pregnancy, and live birth(18). At present, the exact mechanism of the beneficial effects of GnRH-a in luteal phase support is still not completely understood. GnRH-a plays a role in the treatment of LPS in fresh cycles. As novel luteal phase support, GnRH-a may act on the corpus luteum, the endometrium, and the embryo(19). GnRH-a stimulates the secretion of LH by pituitary gonadotropin cells and promotes the corpus luteum function(5). LH release stimulates angiogenetic growth factors and cytokines(20, 21). The expression of GnRH-a and its receptor were found in tissues including endometrium, ootheca, testis, placenta, and myometrium(22). Endometrium expresses GnRH and GnRH-receptor mRNA throughout all phases of the menstrual cycle, with the most intense expression during the luteal phase(23). In human embryonic implantation, there is possibly a close interaction between the endocrine and immune systems through the GnRH and its receptor(24). Experiments in vitro have shown that GnRH-a can regulate the synthesis and secretion of hCG in the preimplantation embryo and placenta and improve the development of cultured embryos(25). Two studies showed that the implantation, pregnancy, and LBR increased with mid-luteal GnRH-a administration on oocyte donation (OD) cycles. The data has implied that embryo development was potentially enhanced, which might be benefited from a GnRH-a direct effect on the embryo(4, 26). Based on this, the GnRH-a administration has also been adopted in FET cycles. However, oocyte donation cycles and autologous frozen embryo transfers were not identical because of a difference in the immunological milieu. Haas et al. reported that the addition of two injections of recombinant hCG and GnRH-a might increase clinical pregnancy rates on the day of transfer and 4d later, respectively, in the NC-FET(11). Seikkula et al. found a higher number of clinical pregnancies and live births in NC-FET with GnRH-a, although the statistical power was too low to show significance(12). Also, as the endogenous luteal activity is lacking in hormonally substituted cycles, the effect of midluteal GnRH-a can be different in natural and artificial FET cycles.
In 2015, Davar et al. designed the first prospective randomized study on GnRH-a administration in AC-FET cycles including 200 patients. On the day of the embryo transfer, the patients in the GnRH-a group were given 0.1 mg triptorelin 3 days after ET. No statistically significant difference was observed between the GnRH-a group and the controls in terms of clinical and ongoing pregnancy rates(13). While in another trial of 220 patients with AC-FET cycles, the ongoing pregnancy rate was significantly increased in the group received GnRH-a at the timepoints of day 2 embryos and vitrified blastocysts(14). Seikkula et al. found that LBR was 9.8 percentage points higher in the GnRH-a group due to the lower number of miscarriages, while the clinical pregnancy rates were similar in both groups(15). The authors called for further studies to confirm the effect of GnRH-a on trophoblast– endometrial interaction. In line with the study of Seikkula, our data showed a distinct but statistically insignificant difference in miscarriage (10.3% vs 25%) and LBR (40.7% vs 28.6%) between the patients with or without GnRH-a on the basis of the standard luteal support. But in clinical practice, a difference of 12.1 percentage points in LBR would be relevant. GnRH-a administration in the luteal phase may possibly enhance the endometrial receptivity and the embryo-endometrium dialogue through activating endocrine-paracrine mechanisms.
This pilot study showed the observable, however statistically negative, differences in the rates of implantation, clinical pregnancy and live birth, and miscarriage rate between the GnRH-a arm and the controls in AC-FET. This pilot study was a prospective randomised trial with strict inclusion criteria. The sample size possibly accounted for the statistic interpretation of the results. However, these results offered informative references for further analysis and studies. The possible beneficial effect of GnHR-a in FET needs to be confirmed by further larger scale clinical trials.
Ethical Approval
This study protocol was reviewed and approved by the Ethics Committee of the Reproductive Center of the First Hospital of Changde City, approval number 202102. All enrolled patientshave given written consent. Our study has been performed in compliance with the Helsinki Declaration.
Competing interests
The authors have nothing to declare.
Authors' contributions
Yanghong Liu was the principal investigator, designed the study, performed the statistical evaluations, wrote the first draft, took part in discussions regarding the results, and edited it in all its revisions. Cheng Chen, Li Wen, Min Lei, Yabin Gou and Bin Tang participated in designing the study, edited and proofread the paper, and took part in discussions regarding the results. Kaishu Huang participated in designing the study, retrieved the data, assisted in writing the paper, edited and proofread the paper, and took part in discussions regarding the results. All authors read and approved the final manuscript.
Funding
This paper was not supported by specific funding.
Availability of data and materials
All data generated or analysed during this study are included in this article and its supplementary material files. Further enquiries can be directed to the corresponding author.