Exposure to gonadotropin-releasing hormone agonist in early pregnancy leads to adverse pregnancy outcomes: a retrospective analysis.

Patients included in this study were sourced from the database of the Reproductive Center at Sun Yat-Sen Memorial Hospital, Sun Yat-Sen University. Women aged 20–45 who were advised to undergo IVF/ICSI-ET at our fertility center and later achieved natural pregnancy following down-regulation with GnRH-a, resulting in canceled oocyte retrieval, were designated as the NP group (Natural pregnancy with the use of GnRH-a) from January 2011 to December 2022. Additionally, women aged 20–45 who underwent IVF/ICSI-ET at our fertility center and successfully conceived in a subsequent IVF/ICSI-ET cycle after GnRH-a down-regulation were categorized as the CT group. A long protocol of pituitary down-regulation was performed on days 21–23 of the previous cycle, either with a daily dose of 0.1 mg short-acting GnRH-a (Triptorelin, 0.1 mg; Ferring Pharma, Kiel, Germany) for 14 days, or with a single dose of 0.8–1.875 mg long-acting GnRH-a (Triptorelin, 3.75 mg; Beaufour Ipsen Pharma, Paris, France). The single dose of long-acting GnRH-a was adjusted according to the patients’ weight. All cases with inflammatory or immune diseases, with serious underlying disease (e.g., uncontrolled hypertension, diabetes, liver and kidney dysfunction, cardiopulmonary dysfunction, etc.), and those with missing and unrecoverable data such as medical record number, age, BMI, diagnosis, infertility years, fertility programs, medication regimens, etc. were excluded. Our study was approved by the Ethics committee at Sun Yat-sen Memorial Hospital, Sun Yat-sen University (SYSKY-2024-445-01). Women in both groups were followed up until clinical pregnancy was determined. Covariates included age of both partners, BMI of both partners, years of infertility, fertility-related diagnoses (abnormal glucose metabolism, uterine factors, cervical factors, tubal factors, PCOS, history of adverse pregnancy, endometrial polyps, hyperlipidemia, ovarian masses, pelvic inflammatory disease, endometriosis), the basal FSH, basal LH, duration and total dose of GnRH-a usage. The primary outcome of the study is the ectopic pregnancy rate. The secondary outcomes included spontaneous abortion rate, clinical pregnancy rate, live birth rate, adverse neonatal outcomes such as preterm birth rate, low birth weight birth rate, and malformation rate. R version 4.3.1 was used to analyze the data. Baseline characteristics of the two groups were compared using chi-square tests for categorical variables and t-tests for continuous variables. A 1:4 propensity score matching was conducted, with the matched variables matching those described previously. Ectopic pregnancy rate, spontaneous abortion rate, live birth rate, adverse neonatal outcomes such as preterm birth rate, low birth weight birth rate, and malformation rate were compared between the two groups after matching. We performed logistic regression to investigate the correlation between GnRH-a exposure and whether adverse pregnancy outcomes occurred. The results are reported as odds ratio (OR) with 95% confidence intervals (CI). P  < 0.05 was considered statistically significant.

After deleting cases with missing data, 6602 IVF/ICSI cycles performing long protocols of pituitary down-regulation were enrolled in our research at our center over the past 12 years. The number of annual GnRH-a long protocol-associated IVF/ICSI cycles and the corresponding spontaneous conception cases involving exposure to GnRH-a are summarized in Fig.  1 . Fig. 1 Flow chart of study population Flow chart of study population Among the initial 136 couples in the NP group, 18 couples were lost to follow-up while three couples opted for artificial abortion, leaving 105 (77.20%) couples with available clinical data and are presented in Fig.  1 . A total of 524 IVF/ICSI cycles that met the criteria were finally enrolled in this study, of which 105 cycles were included into the NP group (Natural pregnancy with the use of GnRH-a), while the other 419 cycles were categorized as the control group after propensity score matching (Fig.  1 ). No significant differences were found in other baseline characteristics, but the SMDs for female BMI, male BMI, years of infertility, type of infertility, ovulatory dysfunction, uterus diseases, and female basal FSH were greater than 0.1, necessitating propensity score matching. After 1:4 propensity score matching, there were no significant differences between the NT group and the control group for each baseline variable, and the SMDs were all less than 0.1 (Fig.  2 ). Fig. 2 Standardized mean differences of baseline characteristics before and after propensity score matching. SMD standardized mean differences Standardized mean differences of baseline characteristics before and after propensity score matching. SMD standardized mean differences Among the 105 women with inadvertent pregnancy, 59 (56.2%) gave birth to 59 children (including one case of esophageal atresia found after birth). For the remaining mothers (43.8%), they had to terminate their pregnancies on account of ectopic pregnancies (14.3%) or spontaneous abortion (29.5%). There were no significant differences in preterm birth rates, low birth weight rates, or malformation rates between the two groups. Notably, the spontaneous abortion rate, ectopic pregnancy rate of the NP group were significantly higher than the CT group ( P  < 0.05) and the live birth rate of the NP group was thus lower than the CT group ( P  < 0.05) (Table 1 ). Table 1 Pregnancy outcome of NP group and Control group NP group (Natural pregnancy with the use of GnRH-a) ( n = 105) Control group ( n = 419) P N . Percentage N . Percentage Spontaneous abortion (%) 31 29.5 57 13.6 <0.05 Ectopic pregnancy (%) 15 14.3 13 3.1 <0.05 Live birth (%) 59 56.2 349 83.3 <0.05 Preterm birth (%) 5/59 8.5 12/349 3.4 0.150 Low birth weight (%) 3/59 5.1 16/349 4.6 0.944 Malformation (%) 1/59 1.7 2/349 0.6 0.913 Pregnancy outcome of NP group and Control group A logistic regression model was composed, in which clinical spontaneous abortion rate and ectopic pregnancy rate were served as the dependent variable, while GnRH-a exposure served as independent variables (Table 2 and 3 ). GnRH-a exposure was found independently associated with spontaneous abortion [OR 2.66, 95%CI 1.61–4.40, P  < 0.05] and ectopic pregnancy [OR 5.21, 95% CI 2.39–11.32, P  < 0.05]. Table 2 Logistic regression model for spontaneous abortion rate OR a 95%CI b P Lower Upper GnRH-a exposure 2.66 1.61 4.40 <0.05 a OR odds ratio b CI Confidence interval Table 3 Logistic regression model for ectopic pregnancy rate OR a 95%CI b P Lower Upper GnRH-a exposure 5.21 2.39 11.32 <0.05 a OR odds ratio b CI Confidence interval Logistic regression model for spontaneous abortion rate a OR odds ratio b CI Confidence interval Logistic regression model for ectopic pregnancy rate a OR odds ratio b CI Confidence interval

GnRH-a is a synthetic GnRH analog with a D-amino acid replacing the 10th glycine position of GnRH and an ethylamide replacing the 10th glycine position. This structural modification gives it a higher affinity for the GnRH receptor and enhances its resistance to protein hydrolysis explication, resulting in better stability, longer half-life, and a 50–200-fold enhancement of biological effects [ 12 ]. In assisted reproductive technology, it is common to combine it with gonadotropins in order to complete an ovulation induction cycle. For infertile couples who are seeking assisted reproduction technology, natural conception can be difficult. Though most of them suffer from subfertility, rather than absolute infertility. Some fertility centers may recommend that they attempt pregnancy while their pituitary gland is downregulated, which may help to alleviate their anxiety regarding infertility to some extent and assist in facilitating a pregnancy [ 13 ]. Multiple studies have been conducted to confirm the role of GnRH-a in the promotion of pregnancy. Since there is currently no clinically effective means of detecting pregnancy within 1 week of ovulation, it is possible to detect a natural pregnancy before the ovulation induction cycle promoted with gonadotropins. After 2–3 days of GnRH-a administration, there will be a temporary increase in FSH and LH levels, which improves the luteal function of the ovary and briefly boosts the production of estradiol and progesterone from the ovary. The elevated E2 and P may act on the endometrium, which is conducive to the implantation of the embryo and the maintenance of pregnancy. GnRH-a binds to GnRH receptors on the endometrium and improves endometrial tolerance and facilitates embryo implantation by inhibiting the local inflammatory response [ 14 ], modulating the development of cytosolic synapses [ 15 ], delaying the time to implantation window closure [ 16 ], improving sub-endometrial blood perfusion, and stimulating endometrial thickening. Additionally, recent research has revealed the presence of GnRH receptors on the embryo, where GnRH-a binding triggers the expression of N-calmodulin, facilitating trophoblast cell invasion [ 17 , 18 ]. Studies on spontaneous pregnancy after down-regulation are few and mostly were clinical case reports, and there is a lack of comparison of pregnancy outcomes between infertile women with spontaneous pregnancy after down-regulation and normal couples or infertile couples conceived through IVF/ICSI. Wilshire et al. [ 19 ] reported observing 5 early miscarriages in 18 natural pregnancies diagnosed by hCG positivity after down-regulation, while Platteau et al. [ 20 ] observed 21 miscarriages in 73 natural pregnancies after down-regulation. Cahill et al. [ 21 ] found 22 live births in 25 natural pregnancies after down-regulation and concluded that progesterone supplementation did not have a positive effect on pregnancy outcomes. Arie Herman et al. [ 22 ] concluded that timely progesterone supplementation may be critical to the success of GnRH-a-exposed pregnancies by examining changes in progesterone and hCG levels in pregnant women after down-regulation. In their study, the rate of early miscarriage in spontaneous pregnancies after down-regulation in infertile women did not differ from that of the normal population, thus they concluded that contraception during the down-regulation cycle was not necessary. However, in a study by Tan et al. [ 23 ], it was believed that the pregnancy safety of GnRH-a exposure during pregnancy is still questionable and infertile couples are advised to use contraception during the month immediately preceding the anticipated start of GnRH-a. However, there are current theories supporting the notion that GnRH-a is detrimental to the maintenance of pregnancy, and our findings appear to be consistent with this notion. Our report represents 105 cases represents a significant number of pregnancies which involved exposure to GnRH-a. Although there was no significant difference in adverse neonatal outcomes between the NP group and the control group, adverse pregnancy outcomes such as the rate of ectopic pregnancies and the rate of spontaneous abortions were significantly higher in the NP group. We found that GnRH-a exposure was associated with an increased risk of ectopic pregnancy and spontaneous abortion when logistic regression was used to explore the effect of GnRH-a exposure on spontaneous abortion and ectopic pregnancy rate in women in the NP group at the time of descending tonus. It is known that the FDA classification of GnRH-a is X and GnRH-a is contraindicated in pregnancy. The first 6 weeks of gestation are pre-embryonic organogenesis, and the effects of drugs during this period are characterized as all-or-none, i.e., lethal effects or no effects. In studies of developmental and reproductive toxicology in animals, daily administration of GnRH-a to pregnant women resulted in maternal toxicity and embryo-fetal toxicity, including pregnancy loss. Since GnRH-a can easily cross the placenta, which provides a basis for explaining the higher rate of spontaneous abortion in the NP group in relation to the toxic effects of GnRH-a on the embryo. In addition to the potential embryotoxic effects of the drug, reduced luteal support due to GnRH-a seems to explain the failure to maintain pregnancy. Progesterone is important for maintaining pregnancy. However, after about 5–7 days of GnRH-a administration, LH and FSH levels decrease as a result of pituitary down-regulation, leading to reduced estrogen and progesterone production by the ovaries. McNeely et al. [ 24 ] have found that decreased LH levels could lead to luteolysis. Outside the hypothalamus, the ovary also has GnRH-a receptors, which bind to GnRH-a and inhibit progesterone secretion in granulosa cells [ 25 ]. Insufficient levels of progesterone can jeopardize pregnancy maintenance and potentially result in early miscarriage. It is estimated that ectopic pregnancies account for 2–3% of all spontaneous conceptions and can be as high as 11% in women with a history of tubal infertility [ 26 ]. Additional factors that may lead to tubal pregnancy include chronic pelvic inflammatory disease, endometriosis, uterine abnormalities, and advanced age [ 27 ]. In our research, we conducted propensity matching to balance risk factors like uterine malformations, history of tubal surgery, and history of tubal pregnancy that could contribute to tubal pregnancy in both the NP and control groups. Following this analysis, we discovered a higher incidence of ectopic pregnancy in women with natural conceptions after down-regulation compared to those who conceived through IVF/ICSI. Compared to infertile women with IVF after normal COS, those with natural pregnancies after down-regulation were exposed to GnRH-a and its subsequent hormonal changes during gestation. Ectopic pregnancy (EP) is a condition in which a fertilized embryo is deposited outside of the uterine cavity, with the most common site of extrauterine attachment being the tubal uterus [ 28 ]. The leading causes of maternal death in the first trimester of pregnancy are intra-abdominal hemorrhage and hemorrhagic shock resulting from tubal rupture [ 29 ]. Minimizing the rate of ectopic pregnancies and early detection and management of ectopic pregnancies are important in reducing adverse pregnancy complications and outcomes in pregnant women. IVF-ET is identified as a significant risk factor for ectopic pregnancy, Keegan D et al. [ 30 ] suggest that the ET technique may affect the rate of ectopic attachment by forcing the embryo through the tubal orifice through hydrostatic pressure or the use of a large amount of transfer media. For women with normal menstrual cycles and normal luteal function, after the oocyte is ovulated from the ovary, it stays in the fallopian tube for about 2–3 days for fertilization. The embryo is gradually moved towards the uterus in the fallopian tube under the influence of the ciliary activity in the fallopian tube, the contraction of the muscles, and the flow rate and direction of the fluid in the tube. On the 7th day after fertilization, the trophoblast layer of the embryonic follicle attaches to the endometrium and gradually implant into the uterine cavity, finally completing the implantation. Although most studies have concluded that GnRH-a improves endometrial tolerance, embryo implantation also requires synchronization of endometrial and embryo development. The “flair-up” effect of GnRH-a induces transiently high levels of estrogen and progesterone at the peri-implantation stage. High levels of progesterone affect the formation of the pinopodes, while high levels of estrogen interfere with the development of the pinopodes, which in turn affects the window of implantation (WOI). GnRH-a can inhibit the expression of aro-matase cytochrome P450 and neuroprotein-1 to reduce endometrial microvascularization [ 31 ], which in turn reduces endometrial perfusion, resulting in asynchronous development of the embryo and the endometrium, and consequently, and ultimately leading to implantation failure. After the failure of implantation, the embryo is free in the uterine cavity. In addition to the causes mentioned above, signaling appears to be an important cause of ectopic pregnancy. Jia-Rong Z et al. [ 32 ] concluded that controlled ovarian stimulation (COS) and subsequent hormonal alterations affect normal biological interactions between the endometrium, fallopian tube and embryo. The fallopian tube competes with the uterine to transmit signals consisting of L-selectin, myotrophic proteins, cytokines, chemokines, and adhesion molecules. The inability of the embryo to properly accomplish localization, adhesion, and invasion in the endometrium leads to migration and implantation of the embryo in the fallopian tube. One can easily surmise that short-term use of GnRH-a in early pregnancy significantly reduces the clinical pregnancy rate. The risk of medical abortion and evacuation due to miscarriage becomes higher, and the probability of a laparoscopic operation due to ectopic pregnancy increases. Corresponding operations can further lead to adverse pregnancy factors such as uterine adhesions, pelvic adhesions, chronic endometritis and pelvic inflammatory disease [ 33 ]. Previous clinical case studies have concluded that the rate of early miscarriage in spontaneous pregnancies in infertile women after down-regulation does not differ from that of the normal population. A research from Wu H et al. [ 11 ] found that pregnancy complications in women inadvertently exposed to GnRH-a in early pregnancy were similar to those in the normal population and did not appear to have adverse effects on long-term neurodevelopment of the child. Most of the studies above concluded that contraception during pituitary down-regulation cycles is not necessary based on their observations. Some reproductive centers even advocate that these infertile couples actively prepare for pregnancy while ovulation is being promoted. However, there is still a lack of guidelines for the management of pregnancy in women exposed to GnRH-a during down-regulation. Contrary to popular beliefs, we advocate for clear guidance advising women contemplating ovulation-induction cycles to refrain from pregnancy and utilize barrier contraception in the month leading up to the intended start of GnRH-a pituitary down-regulation based on our findings. Our research revealed that there was no notable discrepancy in obstetric outcomes between women who were exposed to GnRH-a and those in the control group. Thus, for women who conceive spontaneously after exposure to GnRH-a during pituitary down-regulation, termination may not be necessary after ultrasound confirmation of intrauterine pregnancy. For patients after embryo transfer, we routinely use intramuscular, vaginal and oral progesterone luteal support therapy until 10–12 weeks of gestation. As for patients with spontaneous pregnancies after down-regulation, in addition to GnRH-a, their progesterone levels in early pregnancy are usually lower. Exposure to GnRH-a can down-regulate estrogen and progesterone levels via the hypothalamic-pituitary–gonadal(H-P-O) axis. Lemay A et al. [ 34 ] suggested that the use of GnRH-a during the luteal phase promotes luteolysis and reduces progesterone levels. Inadequate progesterone levels can compromise the maintenance of a pregnancy and lead to early miscarriage. Thus, a higher progesterone supplementation during early pregnancy is recommended. Due to the time span and the fact that many of our center’s patients return to their local area to give birth after successful assisted reproduction, many cases cannot be traced back to their regimen in the early stages of pregnancy. A prospective study of the differences in pregnancy outcomes between regular progesterone support and higher progesterone support in women with spontaneous pregnancies after down-regulation is being conducted at our center. Since there are no clear international recommendations for the management of these infertile women, our study may provide a basis for recommendations on pregnancy management.

Infertility is characterized by a woman who has been trying to conceive without success for at least 12 months [ 1 ]. Infertility is a global public health issue that impacts around 10–15% of couples worldwide [ 2 ]. The social, economic, cultural and health status of countries and people’s lifestyle choices have a huge impact on infertility [ 3 ]. In 1984, Porter et al. [ 4 ] first reported the successful application of GnRH-a for IVF-ET pituitary down regulation, and since then, GnRH-a has been routinely used in controlled ovulation hyperstimulation (COH). During the initial period of administration, GnRH-a binds to GnRH receptors, causing a “flare-up” of follicle-stimulating hormon (FSH) and luteinizing hormone (LH). The increase in FSH and LH secretion is similar to the FSH and LH peaks that occur before natural ovulation. After about 5–7 days of continuous non-pulsatile administration, FSH and LH levels gradually decrease to below basal values. After 10–14 days, follicular development in the ovary is inhibited, and estrogen(E2) decreases to early follicular stage or even postmenopausal levels [ 5 ]. After pituitary down-regulation, multiple follicles in the ovary can be synchronized for development and maturation, and thus increases the number of acquired oocytes and improve the clinical pregnancy rate [ 6 ]. Furthermore, since follicle development relies entirely on external gonadotropin (Gn) stimulation, the natural LH peak in the pituitary gland is suppressed. This in turn prevents premature luteinization of follicles and premature oocyte maturation, resulting in a decrease in cycle cancellation rate from 15–20% to 2% [ 7 ]. Currently, ovulation promotion regimens routinely applied in clinical practice for GnRH-a down-regulation include: the long regimen, the ultra-long regimen, the short regimen, and the ultra-short regimen. Among them, the long program is the most classic and widely used COH. However, inadvertent natural pregnancy may occur during the luteal phase down-regulation, exposing the embryo to maternal use of GnRH-a in its very early stage. The rate of unintended pregnancy following down-regulation therapy during COH has been previously documented to fall approximately between 0.19% and 1%, mirroring the rate of natural pregnancy resulting from tubal factors, male semen abnormalities, and unexplained prolonged causes of infertility [ 8 ]. This natural pregnancy may occur before pituitary down-regulation or may be conceived after spontaneous ovulation without contraception during the down-regulation phase. However, there is no effective clinical method to detect the presence of pregnancy within 7 days of ovulation, and thus we are unable to detect early pregnancy in time before or during the luteal phase of down-regulation. Several clinical cases of early pregnancy detected during the ovarian hyperstimulation and oocyte pick-up phase have been reported internationally, including an infertile woman who presented with menstrual-like bleeding and later confirmed to be preeclampsia in early pregnancy [ 9 ]. In clinical practice, early pregnancy is diagnosed and the cycle will be canceled if a non-menstrual period is detected before ovarian stimulation with Gn (34 d of menopause) or if transvaginal ultrasound is suggestive of a dark area of fluid in the uterine cavity, thickening of the endometrium (≥ 9 mm), and sonographic evidence of corpus luteum in the ovaries, or if a urine pregnancy test is performed or an elevated blood HCG is measured. Studies on natural pregnancies after down-regulation are few and mostly were clinical case reports. It is uncertain whether exposure to GnRH-a impacts pregnancy success. GnRH-a is highly susceptible to passing through the placenta [ 10 ] and its FDA pregnancy classification is X. It is well known that the first 6 weeks of gestation is the pre-existing stage of embryonic organ formation, and the effects of drugs on the embryo during this period are characterized by all-or-none effects, i.e., lethal effects on the embryo or the embryo is unaffected by the drug. There is no definitive guidance on whether contraception is necessary during the down-regulation phase. No effects of GnRH-a on fetal teratogenicity and long-term complications have been reported in the literature [ 11 ], and there is no clear evidence whether women exposed to GnRH-a during pregnancy at this time need to terminate their pregnancies. Therefore, our study aims to further evaluate the effects of GnRH-a exposure on pregnancy to provide a new basis for the management of women with infertility during the down-regulation and for the treatment of early pregnancy in women with spontaneous pregnancies after exposure to GnRH-a.

Below is the link to the electronic supplementary material. Supplementary file1 (DOCX 18 KB) Supplementary file1 (DOCX 18 KB)