Backgroud

Luteinizing hormone (LH) is a glycoprotein hormone secreted by the gonadotropin cells of the anterior pituitary gland, which plays an important role in hormone generation, ovulation promotion and luteinization [1]. Gonadotropin-releasing hormone antagonist (GnRH-ant) protocol has been widely used in assisted reproduction in recent years due to its advantages of non-flare up effect, rapidly and effectively inhibiting the LH surge, reduction of ovarian hyperstimulation syndrome (OHSS) incidence and short treatment period [2]. LH level changes diversitily during COH, most LH levels spontaneously decrease before the administration of antagonist, while about 1/3 LH levels increased, and there are differences in LH level changes between before and after the administration of antagonist [2].

Excessive or insufficient LH levels as well as significant rise or decrease of LH levels will reduce the clinical pregnancy rate [3,4,5]. While the LH level decreased achieved better COH outcomes than increased ones [6]. Meanwhile, there was no difference in clinical outcomes with different LH levels or LH level changes in similar articles [7,8,9].

Up to date, there is no unified conclusion on the effect of LH level changes on COH and fresh ET outcomes in GnRH-ant protocol. Most of published studies have focused on the significant changes of LH level at a single time point, here we study the LH level changes before and after the initiation of antagonist during COH. The purpose of this study was to explore the effect of LH level changes on COH and ET outcomes.

Materials and methods

Subjects

A retrospective study was performed, analyzing data from 721 the first in vitro fertilization (IVF) or intracytoplasmic sperm injection (ICSI) cycles performed at Center for Reproductive Medicine, The Third Affiliated Hospital, Sun Yat-sen University (China, Guangdong) during the year 2019.

The research steps are shown in the Fig. 1.

Fig. 1
figure 1

Schematic diagram of research route

Full size image

The study was approved by the Ethics Committee Review Board of The Third Affiliated Hospital, Sun Yat-sen University (No: [2021]02-246-01).

Inclusion criteria

Patients who underwent ovarian stimulation with GnRH-ant (ganirelix [Ganirest®, 0.25 mg; MSD] or cetrorelix [Cetrotide®, 0.25 mg; Merck]) protocol to suppress the pituitary and patients aged 20–45 years old were enrolled in this study.

Exclusion criteria

Patients who underwent ovarian stimulation with non-GnRH antagonist protocol, with history of hydrosalpinx, uterine malformation and uterine adhesions, endometriosis, endometrial polyps, polycystic ovary syndrome and other diseases or a concurrent medical condition that have to cancel oocyte retrieval were excluded in this study.

Protocol

GnRH-ant protocol was applied in this study. COH began on day 2 or 3 of a menstrual period with initial injections of 150–300IU of gonadotropin (Gn) (Gonal F, 450IU; Merck/ Follistim, 300IU; MSD). Antagonist ganirelix (Ganirest, 0.25 mg; MSD) or cetrorelix (Cetrotide, 0.25 mg; Merck) was administered 4 days later according to fixed GnRH antagonist protocol. Drugs are injected at a fixed time every morning. Recombinant human chorionic gonadotropin (rhCG) (Ovitrelle, 250 ug; Merck) or human chorionic gonadotropin (hCG, 10000IU; Lizhu Pharmaceuticals) was administered once when three leading follicles reached ≥ 17 mm mean diameter or one leading follicle reached ≥ 18 mm mean diameter were observed, oocyte retrieval was performed 34–36 h later. Blood samples were taken to measure hormone levels (Follicle stimulating hormone (FSH), Luteinizing hormone (LH), Estrogen (E2), Progesterone (P)) on day 1 of ovarian stimulation (day 2–3 of menstrual cycle), day 1 of the antagonist injection (day 6–7 of menstrual cycle) and trigger day, meanwhile B-ultrasounds were performed for follicular size and number. After oocyte retrieval, IVF or ICSI was routinely performed and embryo culture was performed after fertilization. According to Veeck’s criteria and Gardner’s scoring criteria, normal fertilized embryos with a cell count of 7, 8 and 9 on the third day after fertilization, and blastocysts with a stage 3 or more in their inner cell mass and trophoblast cell scores ≥ B on the fifth day, or blastocysts with a stage 4 or more in their inner cell mass and trophoblast cell scores ≥ B on the sixth day, were defined as high-quality embryos.

COH process were divided into 2 stages, before (stage 1) and after (stage 2) the GnRH-ant initiation, and each with 5 groups basing on LH levels: LH decreased more than 50% (A1, A2), decreased 25-50% (B1, B2), change less than 25% (C1, C2), increased 25-50% (D1, D2), and increased more than 50% (E1, E2).

Calculation method: LH changes before GnRH-ant initiation: (LH level on the fifth day of ovulation stimulation - LH level on the first day of ovulation stimulation)/ LH level on the first day of ovulation stimulation. LH changes after GnRH-ant initiation: (LH level on trigger day - LH level on the fifth day of ovulation stimulation)/ LH level on the fif day of ovulation stimulation.

Statistical analysis

Continuous variables were presented as means ± standard deviation (SD), categorical variables were presented as percentage. For quantitative data, Kruskal - Wallis tests were used for statistical analysis, while for qualitative data, Fisher’s Exact test or Chi-Square test was used for statistical analysis. For multivariate studies, multiple linear regression models were used for statistical analysis. The significance of all statistical results was set at P < 0.05, and statistical analysis was performed using SPSS software (SPSS Version 25.0.0.0, IBM Corp., USA).

Results

The effect of LH level changes before the initiation of GnRH-ant on the outcome of COH and ET

There was a total of 721 first IVF/ICSI patient-cycles with 395 cycles (54.8%) in LH decreased more than 50% before the initiation of GnRH-ant (A1), 179 cycles (24.8%) in LH decreased 25-50% (B1), 99 cycles (13.7%) in LH change less than 25%(C1), 19 cycles (2.6%) in LH increased 25-50% (D1) and 29 cycles (4.0%) in LH increased more than 50%(E1). Tables 1, 2, 3 and 4 show the patients baseline characteristics, protocol related parameters and cycle outcomes, such as COH outcomes, Embryo transfer and pregnancy outcomes. There was significant difference regarding infertility years among groups A1, B1, C1, D1, E1 (P<0.05) with the longest in group E1. At the same time, no significant differences were detected in age, BMI, infertility factors, AMH level (Roche fully automatic electrochemical luminescence detector model e602; Roche Diagnostic GmbH, Germany) and antral follicle number (P > 0.05).

Table 1 Baseline Characteristics in LH level changes before administration of GnRH-ant
Full size table
Table 2 Ovarian Stimulation Parameters in LH level changes before administration of GnRH-ant
Full size table
Table 3 COH outcomes in LH level changes before administration of GnRH-ant
Full size table
Table 4 Embryo transfer and pregnancy outcomes in LH level changes before administration of GnRH-ant
Full size table

The results showed that although duration of stimulation and the total dose of Gn in 5 groups had no statistical differences, the total dose of GnRH-ant and duration of antagonist in group D1 were lower than those in other groups (P < 0.05). There were significant differences in FSH and E2 levels among all groups during ovarian stimulation (P<0.05). while there were no statistical differences in progesterone (P) level and endometrial thickness in 5 groups on HCG trigger day (P>0.05).

It was found that there was no significant correlation between LH change before the antagonist initiation and COH outcome items among all groups (P > 0.05).

There were no statistically significant differences regarding the ET outcomes among all groups (P > 0.05).

The effect of LH level changes after the initiation of GnRH-ant on the outcome of COH and ET

Tables 5, 6, 7 and 8 show a total of 721 first IVF/ICSI patient-cycles with 100 cycles (13.9%) in LH decreased more than 50% after the initiation of GnRH-ant (A2), 110 cycles (15.3%) in LH decreased 25-50% (B2), 178 cycles (24.7%) in LH change less than 25%(C2), 62 cycles (8.6%) in LH increased 25-50% (D2) and 271 cycles (37.6%) in LH increased more than 50%(E2). According to Kruskal-Wallis test, there were statistically significant differences in age, BMI, infertility type, AMH and AFC among all groups, with age and BMI higher in group E2 than those in other groups (P < 0.05) but AMH and AFC lower in group E2 (P < 0.001).

Table 5 Baseline Characteristics in LH level changes after initiation of GnRH-ant
Full size table
Table 6 Ovarian Stimulation Parameters in LH level changes after initiation of GnRH-ant
Full size table
Table 7 COH outcomes in LH level changes after initiation of GnRH-ant
Full size table
Table 8 Embryo transfer and pregnancy outcomes in LH levels after initiation of GnRH-ant
Full size table

The days of ovarian stimulation and the total amount of Gn in group A2 were lower than those in other groups (P = 0.004, P < 0.001). And FSH level in group A2 was the lowest on the initiation day of stimulation and trigger day (P < 0.001).

Among the 5 groups, the number of oocytes retrieved, mature oocytes, 2PN fertilized oocytes, embryo cleavage and the numbers of embryo available in group A2 were significantly higher than those in other groups (P < 0.05). The group A2 had the highest cancellation rate (77.0%) (P < 0.001), while there were no significant differences in ET outcomes among all groups (P > 0.05). According to multiple linear regression analysis, the number of oocytes retrieved were significantly affected by patients’ age, AMH and AFC. While the LH level changes after the addition of antagonists had no significant impact on the number of oocytes retrieved. However, patients’ age, AMH, AFC and the LH level changes after the addition of antagonists had no significant impact on the number of available embryos (Supplementary Tables 1–2). According to binary logistic analysis, LH level changes after the initiation of antagonist did not have a significant effect on clinical pregnancy, but age had a significant effect on clinical pregnancy (Supplementary Table 3).

Discussion

It has been demonstrated that < 1% of LH receptors being occupied is enough to elicit a normal steroidogenic response [10]. However, there is no consensus on the optimal clinical LH threshold range for COH in assisted reproductive technology (ART) [11]. In the study by Bosch E and coworkers [12], no differences were observed between the number of oocyte retrived or the fertilization, implantation, and pregnancy rates of different LH concentrations on days 3, 6, and 8 of stimulation and on the day of hCG, what was also supported by other similar studies [13].

During the ovarian stimulation process of GnRH-ant protocol, 55%(395/721) of the patients showed a spontaneously significant decrease in LH level before the addition of the antagonist, and only a few patients (37/721) had a significant increase in LH level. The significant decrease in LH level before the addition of antagonists may be related to the production of gonadotrophin surge-attenuating factor (GnSAF). Previous studies [14, 15] have found that in the process of assisted reproduction using exogenous gonadotropin to promote follicle development, exogenous gonadotropin was found to stimulate the ovarian production of an uncharacterized hormone known by its specific effect of reducing pituitary responsiveness to GnRH. This hormone has been called gonadotrophin surge-attenuating factor (GnSAF). It regulates LH secretion by reducing the sensitivity of the pituitary to GnRH and antagonizing the stimulatory effects of oestradiol on GnRH-induced LH secretion. The main role of GnSAF is probably the negative regulation of pulsatile LH secretion, mainly during the first half of the follicular phase. What’s more, the results of this study showed that the change of LH level before the addition of the antagonist did not affect the outcome of the COH and pregnancy outcome. It was basically consistent with the study of Vanetik et al. [16], they proposed that there was no statistical difference in pregnancy outcome when LH level increased or decreased on the 5th day compared with that on the initiation day during COH. While the results of a retrospective study involving 2116 fresh ET cycles showed that before antagonist addition, the oocytes retrieved rate and fertilization rate in the group with LH increased were lower than those in the group with LH decreased [17], but there was no statistical difference in pregnancy-related outcomes among the groups.

LH level changes after the addition of antagonists had a significant difference in the outcome of ovarian stimulation with the number of oocyte retrieved and available embryos in group A2 significantly higher than those in other groups. But, no significant difference was found between LH level changes and embryo transfer result. It was revealed in our study the total rFSH dose was significantly lower in people with LH decreased ≥ 50% after the addition of antagonists during COH, while their E2 level was relatively higher on trigger day. We consider that it was associated with a better ovarian response in people with significantly decreased LH level after antagonist addition, for AMH and AFC were higher in them, also the age was younger in them. Existing studies have demonstrated that ovarian responsiveness is generally assessed by ovarian markers such as antral follicle count (AFC) and anti-Müllerian (AMH), in conjunction with age, in order to predict poor, normal or hyper-response [18]. It has been proved that with the increase of estrogen level (E2 ≤ 4 800 pg/mL) in the GnRH-ant protocol, both the number of oocyte retrieved and embryos obtained increased [19]. Similarly, we found people who with significantly decreased LH had a higher average level of E2, therefore had a relatively better COH outcomes. Besides, the number of oocyte retrieved, mature oocytes, 2PN fertilized oocytes, embryos cleavage and the numbers of embryo available were negatively correlated with LH level changes. In the study of Scheffer and coworkers [20], they found that age was negatively correlated with the quality of D3 and D5 embryos, but AMH and AFC were positively correlated with the quality of D3 and D5 embryos. Here in this study we also found that the number of high-quality embryos and high-quality embryonic rate were lower in patients with significantly increased LH level, which was related to that patient with advanced age, low AMH and low AFC was more prone to appear LH significantly increase.

There are many factors affecting the outcomes of COH and pregnancy. Such as E2, LH and P levels on trigger day as well as age and BMI all have certain influence on the outcomes of ovulation stimulation and pregnancy, while LH level on trigger day is negatively correlated with the number of oocytes retrieved, but has no significant correlation with the outcome of pregnancy [21].The conclusion is consistent with Ji Hui and coworkers’ results [22], they proposed that low LH levels on trigger day can predict higher the number of oocyte retrieved, but no effect on early miscarriage rate or clinical pregnancy rate. Similarly, we observed that there were no significant differences in pregnancy outcomes among different LH change groups after GnRH-ant addition (P > 0.05). However, Younis JS and coworkers found that a decrease of LH level > 50% significantly reduced the embryo transfer outcome [23]. And in the study by Geng Y and coworkers [24], the clinical pregnancy rates were reduced in high ovarian response patients with LH surge during ovulation stimulation in GnRH-ant protocol.

Conclusion

In conclusion, this study demonstrates that LH level changes before GnRH-ant initiation had no effect on COH and ET outcomes. The number of oocyte retrieved, mature oocytes, fertilized oocytes, embryos cleavage, high quality embryos, high-quality embryonic rate and the numbers of embryo available have significant differences with the change of LH after GnRH-ant initiation, but without influence on the pregnancy outcome. Therefore, the LH level changes in GnRH antagonist protocol can predict the outcomes of ovulation stimulation to a certain extent, but it cannot be used to predict the outcome of clinical pregnancy. What’s more, the limitation of this study lies in the fact that it is a retrospective study, which is influenced by unnoticed bias or confounding factors. Meanwhile, on account of the unequal sample size among the groups, the sample may not accurately reflect the overall situation. Hence, more comparative studies are needed in the future.