The effect of large follicle puncture and aspiration on the outcomes of IVF-ET in patients with asynchronized follicles under the long GnRH-a protocol: a retrospective cohort study

Typically, the human reproductive cycle is characterized by single follicle development and ovulation. During the luteal-follicular transition phase of the natural menstrual cycle, antral follicles sensitive to FSH are recruited for rapid growth due to the elevation of FSH concentrations beyond the FSH threshold window. Afterward, due to the negative feedback induced by inhibin B (INH-B) and estradiol (E₂), FSH begins to decline, and the FSH window closes below the threshold for follicle selection. Except for the most sensitive follicle to FSH, the other follicles stop developing and become atretic. Therefore, the time interval of the FSH threshold window is crucial for selecting a single dominant follicle from the recruited follicles [1]. To improve the success rate of in vitro fertilization-embryo transfer (IVF-ET), it is usually necessary to perform controlled ovarian hyperstimulation (COH) treatment to counter the process of single follicle development and obtain an acceptable number of oocytes for subsequent fertilization and embryonic transfer. In COH cycles, the use of exogenous Gn artificially increases the FSH concentration and prolongs the FSH threshold window period, allowing the follicles that would have been atretic to keep growing. However, the actual size and FSH threshold of each follicle in the cluster differs. Consequently, asynchronization in follicular development occurs, whereby the use of Gn may amplify this asynchronicity [2]. In patients with asynchronous follicles, by secreting high concentrations of INH-B and E₂ and through paracrine and autocrine effects, the asynchronous dominant follicles (DF) inhibit the growth and development of adjacent small follicles, which in turn affects oocyte maturation rate and pregnancy rate [3, 4]. Therefore, improving follicular development synchronicity is critical in clinical practice. The key to improving follicular synchronization lies in controlling FSH levels during the luteal-follicular transition and avoiding the early recruitment and development of follicles before Gn use.

Among the numerous COH protocols, the long GnRH-a protocol is considered one of the most classical superovulation protocol. Pituitary down-regulation with GnRH-a can reduce FSH levels during the luteal-follicular transition, and reduce cycle cancellation rate caused by premature LH surge in COH [5]. This assists in synchronizing follicular development and ultimately achieving satisfactory clinical outcomes. For ovarian stimulation protocols lacking luteal FSH control, such as the short protocol and gonadotropin-releasing hormone antagonist (GnRH-ant) protocol, use of oral contraceptives (OC), estrogen, GnRH-ant and other pretreatment methods can control the FSH level during the luteal-follicular transition. Huirne Judith AF et al. [6] compared the effects of antagonist regimen with or without OC pretreatment on the number of oocytes retrieved in IVF-ET patients. They postulated that OC pretreatment could decrease the initial concentration of FSH, thereby improving follicular homogeneity and eventually obtaining more high-quality embryos. However, both Kolibianakis et al. [7] and Rombauts et al. [8] reported adverse effects such as increased Gn dosage, prolonged Gn stimulation time, and increased early abortion rate after OC pretreatment. Fanchin et al. [9, 10] reported that the GnRH-ant protocol with E₂ pretreatment could improve follicular synchronization and increased the number of available embryos, potentially improving the pregnancy rate. However, E₂ pretreatment only has an inhibitory effect on FSH, and exposure to high levels of LH may adversely affect the implantation rate by altering endometrial receptivity in patients [11]. As GnRH-ant can rapidly inhibit endogenous Gn secretion, premenstrual injection of GnRH-ant is supposed to prevent luteal FSH elevation and promote consistency in follicular development [2]. However, further cumulative data and research are paramount in determining whether this approach can truly improve IVF outcomes.

The above methods enable FSH-sensitive antral follicles to avoid premature exposure to gradient FSH concentrations during the luteal-follicular transition period in case antral follicles of various sizes appear at the beginning of ovarian stimulation. However, in the actual ovarian stimulation process, considering that each follicle has a different sensitivity to Gn, there are still some patients with asynchronized follicles. During ovarian stimulation, the puncture and aspiration of asynchronized large follicles allow the oocytes and granulosa cells to be aspirated. Thereby the inhibitory effect on adjacent follicles and possible induced endogenous LH surge are elimitated, and the oocyte quality can be enhanced [12]. In this study, the clinical data of 180 patients with asynchronized follicles during GnRH-a protocol COH were analyzed and compared in order to provide a basis for the clinical treatment of patients.

Herein, 180 IVF/ICSI patients with asynchronous follicles who underwent the long GnRH-a protocol were included, of which 81 patients underwent large follicle puncture and aspiration (Group 1), whereas 99 patients did not (Group 2). Table 1 summarizes patients’ baseline characteristics of these two groups of patients. There were no significant differences in age, BMI, duration of infertility and basal endocrine level after pituitary down-regulation between the two groups (P > 0.05). However, the diameter difference between dominant and secondary follicles (4.46 ± 1.26 vs. 4.03 ± 1.08, P = 0.015), Gn usage time (12.57 ± 1.91 vs. 11.89 ± 2.08, P = 0.025) and Gn dosage (2321.30 ± 418.04 vs. 2160.83 ± 436.03, P = 0.013) were significantly higher in Group 1 than in Group 2.

Table 2 illustrates the reproductive outcomes of Group 1 and Group 2. There were no significant differences in the number of oocytes retrieved and the 2PN fertilization rate between the two groups (P > 0.05). However, oocyte maturation rate (92.3% vs. 88.9%, P = 0.009) and high-quality embryo rate (75.2% vs. 65.7%, P = 0.007) were significantly higher in Group 1 than in Group 2. Nevertheless, the final clinical pregnancy rate and live birth rate were not significantly different between the two groups (P > 0.05).

Next, we compared the data of the A1 (n = 33), A2 (n = 47), B1 (n = 48) and B2 (n = 52) subgroups to investigate whether large follicle puncture and aspiration could affect the clinical outcomes of patients with asynchronous follicles at different time points. Tables 3 and 4 compare the baseline information and treatment outcomes of Subgroup A1 and Subgroup A2. When the diameter of dominant large follicles was ≤ 14 mm, the total days of Gn (13.12 ± 2.06 vs. 11.79 ± 1.77, P = 0.003) and the total dosage of Gn (2433.71 ± 359.98 vs. 2246.49 ± 389.04, P = 0.032) were significantly greater in Subgroup A1. The oocyte maturation rate (92.7% vs. 88.1%, P = 0.023), high-quality embryo rate (72.9% vs. 61.8%, P = 0.047), and final live birth rate (54.5% vs. 31.9%, P = 0.043) were also significantly higher in Subgroup A1 compared to Subgroup A2.

Tables 5 and 6 compare the baseline information and treatment outcomes of Subgroups B1 and B2. When the diameter of dominant follicles was > 14 mm, there were no significant differences in all comparisons.

In recent years, assisted reproductive technologies have developed rapidly. In particular, improvements in controlled ovarian stimulation protocols allowed the development of multiple synchronized follicles in the same oocyte retrieval cycle, which helps patients achieve increased high-quality embryo and pregnancy rates. At present, among the numerous COH protocols, the long GnRH-a protocol is considered one of the most classical superovulation protocol. By binding to GnRH receptors, GnRH-a results in a dramatic decrease in the number of receptors on the pituitary surface, whereby the receptors lose response to endogenous and exogenous GnRH. After 14 days of GnRH-a, the pituitary gland reaches a state of down-regulation, avoiding the appearance of early-onset LH surge and promoting the synchronous development of follicles [5]. However, given that the sensitivity to Gn stimulation is not completely consistent during superovulation, growth differences between developing follicles cannot be completely overcome. Therefore, asynchronous follicular development can still be clinically observed in some patients. The determination of the HCG day is complex due to asynchronous follicles and is a challenge regularly faced by doctors. The immature or postmature follicles in patients are frequently collected, which reduces the number of effective embryos and affects the final clinical outcomes.

In the natural cycle, the other follicles on the side of the ovary where the dominant follicle is located are smaller than the contralateral follicle, indicating that the presence of the dominant follicle may inhibit the growth of other non-dominant follicles [15, 16]. Follicular dominance consists of two primary components: indirect endocrine action and direct intra-ovarian regulation. The dominant follicle is able to secrete high concentrations of INH-B and E2, leading to small follicles atresia due to the down-regulation of FSH concentrations. The dominant follicle is more sensitive to FSH stimulation due to the increased number of granulosa cells and the increase of FSH receptors, which allows the dominant follicle to survive despite the later decrease in circulating FSH concentration. This process can be regarded as an indirect endocrine effect of the dominant follicle [3, 17]. In addition to this systemic endocrine factor, the dominant follicle can directly inhibit the growth and development of small follicles through paracrine and autocrine effects. The concentration of insulin-like growth factors-1 (IGF-1) is higher in the dominant follicle than in the small follicles, and the IGF-1 system can amplifie FSH and stimulate the growth of follicles. The presence of a dominant follicle allows increased IGF-binding protein (IGFBP) production on small follicles, thus reducing the concentration of available IGF-1. Therefore, the dominant follicle can continue growing despite reduced FSH concentrations, while the adjacent small follicles undergo follicular atresia and die [18, 19]. Additionally, Spears et al. [20] determined that the dominant follicle's effects were established when the follicles were cultured together, while this phenomenon did not arise when follicular clusters were cultured separately under similar conditions. Therefore, it is hypothesized that there is a direct interaction between the follicles, which allows the dominant follicle to influence the fate of the adjacent follicles by enhancing the endocrine dominance effect.

In clinical applications of the down-regulation protocol to induce superovulation, under the action of a high dose of FSH, the follicles developed in follicular clusters may still appear uneven in size. This phenomenon suggests that even if exogenous FSH can attenuate the indirect endocrine effect, the appearance of large dominant follicles will still affect the growth of the remaining small follicles due to internal regulation in the ovaries, which is, in turn, detrimental to clinical outcomes. In the ovine superovulation model, the dominant follicle influenced the number of follicles and embryos obtained from the ovary on the dominant follicle's side in ewes during high-dose FSH treatment and decreased embryonic viability as well [4]. Semra Kahraman et al. [21] studied the impact of follicular dynamics on early human embryo development for the first time. They discovered that for follicles derived from homogenous cycles, the odds of obtaining top or good quality blastocysts were 1.370-fold higher than that of heterogeneous follicles. Therefore, follicular sizes and variable growth rates in IVF-ET cycle could affect early human embryonic development.

In superovulation cycles, puncture and aspiration were performed to remove oocytes and granulosa cell from the dominant large follicles. The dominant follicles disappeared, thereby eliminating its direct inhibition on the adjacent follicles through interfollicular interactions. After the steroid-rich follicular fluid was aspirated, the steroid hormone level in the circulation decreased. This avoids the asynchronous development of endometrial glandular epithelium and stroma caused by the premature rise of LH surge and progesterone, helping to improve the embryo implantation rate [22, 23]. Fisch et al. [24] reported on a 39-year-old woman with a poor response to repeated IVF, who had prematurely developed 27-mm dominant follicles punctured during superovulation treatment and found that the number of oocytes retrieved increased after the puncture, and a healthy pregnancy was eventually attained. Animal experiments have revealed that removing the dominant follicle in cows 48 h before hyperstimulation could promote follicular growth and increase the number of transferable embryos [25]. Herein, we compared the clinical outcomes of large follicle puncture and aspiration in patients with asynchronous follicles during COH cycles in order to investigate whether large follicle puncture and aspiration were beneficial in clinical treatment.

In this study, compared with Group 2, the differences between the diameter of the dominant and secondary follicles in Group 1 were more significant, and for this reason, clinicians often prefer large follicle puncture and aspiration for such patients. After the removal of the dominant follicles by puncture, the remaining small follicles became the target of Gn medication. Therefore, patients in Group 1 needed longer Gn usage time and Gn dosage to enable small follicles to reach the standard size for the HCG day. Herein, it was determined that dominant follicles tended to be those that developed faster than the remaining adjacent small follicles. However, their development rate did not correspond to the Gn usage days. The short Gn administration time of the dominant follicles can affect the maturation of the cytoplasm. The expression of the nucleus mainly depends on the regulation of cytokines in the cytoplasm, and consequently, the immaturity of the cytoplasm can lead to defects in the quality and quantity of cytokines. Finally, the quality of follicles and embryos are badly affected [26]. In addition, it was challenging to decide the HCG day in Group 2 due to the presence of dominant follicle. On the day of oocyte retrieval, only a few follicles met the diameter criterion for mature follicles. In contrast, the number of small follicles was high, implying that the oocyte maturation rate and high-quality embryo rate were lower than that of Group 1. However, there were no statistically significant differences between the two groups in the 2pn fertilization rate, clinical pregnancy rate, and live birth rate. This may be due to embryo formation and development is a complex and dynamic process. During the whole process of embryonic development, multiple factors, namely, the ultrastructure of oocytes and embryos, embryonic development potential, endometrial receptivity, etc., can affect normal fertilization, embryonic development, and ultimately, clinical pregnancy.

Son WY et al. [27] compared the pregnancy outcomes of IVM cycles in 160 patients and determined that in hCG-primed IVM cycles, ipsilateral immature oocytes were adversely affected when the diameter of dominant follicles exceeded 14 mm, and better clinical outcomes could be achieved by retrieving oocytes before the dominant follicles' diameter exceeded 14 mm. To determine whether large follicle puncture and aspiration affected the outcomes when the dominant follicles appeared at different time points, the 180 patients were tentatively divided into 4 subgroups based on whether the diameter of DF exceeded 14 mm on the day of their appearance and whether large follicle puncture and aspiration were performed. The results revealed that Subgroup A1 had a higher Gn dosage and total Gn usage time compared to Subgroup A2. Moreover, the oocyte maturation rate, high-quality embryo rate, and live birth rate were significantly better in Subgroup A1 than Subgroup A2. However, no significant differences were observed in Subgroup B1 than Subgroup B2. The results demonstrated that when the diameter of the dominant follicles that appeared did not exceed 14 mm, clinicians could perform large follicle puncture and aspiration to improve the pregnancy outcomes. For patients with dominant follicles greater than 14 mm, no special treatment was required, and the hyperovulation protocol could proceed. Large follicle puncture and aspiration did not result in practical benefits for these patients and might instead increase their economic and psychological burdens. However, this results contradict the findings of Son WY. If ipsilateral immature oocytes were adversely affected when the dominant follicles exceeded 14 mm, the large follicle puncture would be more beneficial for such patients. However, our experiment led to the opposite conclusion. Given the paucity of research on this subject and the small sample size included in this experiment, future studies should focus on expanding the sample size to provide more accurate results.

In conclusion, this study found that there may be patients with asynchronous follicles in GnRH-a long protocol COH. For these patients, large follicle puncture and aspiration can improve oocyte maturation and high-quality embryo rates. However, the corresponding Gn dosage will also increase, and clinical pregnancy and live birth rates may not significantly improve. In addition, if the diameter of the dominant follicles does not exceed 14 mm on the day of its appearance, large follicles puncture and aspiration might have a significant impact on improving the oocyte maturation rate, high-quality embryo rate, and live birth rate. Since the small sample size is the main limitation of this retrospective study, this conclusion needs to be further validated in large sample size RCTs in the future.