A successful pregnancy of in vitro fertilization (IVF) treatment depends on the number and quality of oocytes retrieved. Ovarian stimulation (OS) could obtain multiple oocytes in one treatment cycle (1). During OS, gonadotropin-releasing hormone (GnRH) antagonist is used to inhibit the luteinizing hormone (LH) surge to prevent the premature ovulation. Except for the trigger of final maturation of oocyte and ovulation, LH also plays an essential role in the development of oocyte, and oversuppressed LH level might be related to inferior embryo quality and early pregnancy loss (2–4). Therefore, LH supplementation during OS treatment is proposed, even though the threshold of LH for normal follicular development remains controversial. The expert consensus in the Asia-Pacific region suggested that LH supplementation might benefit patients aged ≥35 years with poor or suboptimal response to standard OS protocol (5).

Because of the homology between human chorionic gonadotropin (hCG) and LH, hCG has been widely used as an alternative to endogenous LH (3, 6). Compared with recombinant LH (rLH), hCG possesses a longer half-life and a higher affinity to LH/hCG receptor (6). Serafini et al. reported that addition of low-dose hCG in late follicular phase in patients with low LH level could improve IVF outcomes by generating a high number of top-quality embryos (7). Drakakis et al. demonstrated that daily administration of 100 IU of hCG throughout the early follicular phase improved implantation and clinical pregnant rates in women aged 35–40 years (8). However, in their early studies, they could not obtain any improvement in clinical pregnancy rate by adding rLH in mid-late follicular phase during OS in women aged >35 years (2, 9). Another prospective randomized controlled trial (RCT) reported that compared with a daily injection of 50 and 100 IU, 150 IU of hCG could retrieve a significantly higher number of oocytes (6). Therefore, the dose and time of adding hCG to optimize the development of oocytes during OS are still controversial.

The purpose of this retrospective study was to define a possible optimal dose and timing for the addition of hCG to routine GnRH antagonist protocol in ‘‘unpredictable” poor responding patients.

Sixty-seven patients aged from 23 to 40 years who underwent two cycles of OS and IVF/ICSI were recruited (Figure 1). Thirty-five women who received the first hCG (150 IU) injection concomitantly with rFSH on stimulation day 1 were classified as Group A; 32 women who received the first injection of hCG on stimulation day 6 were grouped as Group B. No statistically significant differences were observed in female age, body mass index (BMI), duration of infertility, basal sex hormone level, AMH, AFC, starting dose of rFSH, and IVF/ICSI indication between the two groups (Table 1).

OS results were compared between the first and second IVF/ICSI cycles (Tables 2, 3). After the supplement of hCG on different days of OS during the second cycle, eight patients in Group A and 21 patients in Group B still showed POR, as ≤3 oocytes were retrieved. ET in nine patients in Group A and seven patients in Group B was canceled because of diminished number of available embryos in the second IVF/ICSI treatment cycle.

In Group A, the daily addition of hCG (150 IU) at the beginning of stimulation significantly improved the total number of ≥14-mm follicles (8.1 ± 3.2 vs. 3.3 ± 1.4) and oocytes retrieved (6.1 ± 3.4 vs. 1.9 ± 1.4) compared with that at the first cycle. Accordingly, retrieved oocyte rate (76% vs. 57%) and the number of mature oocytes in ICSI cycle (4.1 ± 2.5 vs. 1.4 ± 1.1) were also significantly increased after the hCG addition. Mean serum E₂ (1,131.3 ± 444.4 vs. 378.1 ± 111.7), LH (2.8 ± 1.1 vs. 1.2 ± 0.7), and hCG (6.1 ± 3.4 vs. 1.9 ± 1.4) level on the day of hCG trigger were significantly higher in the second cycle compared with the first. There was no significant difference in E₂ level on stimulation day 6, progesterone concentration on the day of hCG trigger, total dosage of rFSH used, and days of stimulation between the two cycles (Table 2).

Details referring to OS and embryological results of Group B are summarized in Table 3. The same as Group A, Group B showed a higher serum level of E₂ (946.3 ± 337.6 vs. 410.2 ± 198.9), LH (2.5 ± 1.1 vs. 1.2 ± 0.7), and hCG (6.4 ± 2.5 vs. 0.4 ± 0.2) on hCG trigger day compared with the first cycle. But the number of ≥14-mm follicles, oocytes retrieved, mature oocytes in ICSI cycle, and retrieved oocyte rate showed no difference in Group B as hCG added on the stimulation day 6 compared without it. E₂ level on stimulation day 6, progesterone concentration on the day of hCG trigger, total dosage of rFSH used, and days of stimulation also showed no differences between two cycles.

Table 4 compares the outcomes of the second cycle between Groups A and B. E₂ level on day of hCG trigger (1,131.3 ± 444.4 vs. 946.3 ± 337.6), number of oocytes retrieved (6.1 ± 3.4 vs. 4.9 ± 2.3), and number of available embryos (2.5 ± 1.4 vs. 1.8 ± 0.8) were significantly higher in patients receiving 150 IU of hCG on the first day of stimulation (Group A) than on stimulation day 6 (Group B). Implantation rate, clinical pregnancy rate, and live birth rate showed an increasing trend in Group A than in Group B, with no statistically significant difference.

The constitution of different-diameter follicles during OS on different stimulation days are shown in Figure 2. The number of large follicles (diameter >14 mm) on the day of hCG administration and the number of the medium follicles (diameter 10–14 mm) on stimulation day 6 were significantly higher in the patients receiving 150 IU of hCG on first stimulation day than in other groups. The number of small-diameter follicles (diameter < 10 mm) throughout the whole OS showed no difference among the four groups.

POR generally resulted from diminished ovarian reserve. However, in some cases, poor ovarian responders may present as normal ovarian reserve with AFC >5 and/or AMH >1.1 ng/ml (13, 14). In this retrospective analysis, we evaluated the effects of low-dose hCG supplementation on women with normal AFC and AMH levels but showed a POR to rFSH stimulation in their first cycle. Our results showed that the addition of 150 IU of hCG in those “unpredictable” POR patients could improve their response to OS. For the first time, we compared the effects of different timing of hCG administration on OS outcome. Our results suggested that the addition of 150 IU of hCG in early follicle phase could provide a better clinical outcome compared with that in mid-late phase addition.

Our results were in line with previous studies that the addition of LH/hCG could increase oocyte number and E₂ production (15, 16). A meta-analysis confirmed that the use of rLH/hCG increased the peak E₂ levels and oocyte number and improved embryo quality and, therefore, enhanced implantation rate (17). Enrico et al. found that unexpected poor ovarian responders could benefit from addition of 150 IU rLH in subsequent cycle (14). Drakakis et al. demonstrated a better clinical outcome of early hCG administration during OS in women with previous IVF failure compared with rLH (8).

Without LH, FSH alone is inadequate to induce normal follicular development (18). In general, approximately 1% of activated LH receptors are sufficient to maintain follicular development and steroid production. There were also studies that reported that adding rLH/hCG to OS may reduce the number of developing follicles and retrieved oocytes. They explained that a high level of LH activity restricted the development of small ovarian follicles (<10 mm) and inhibited follicular recruitment in patients with normal ovary response (19, 20). Our data showed that the number of medium follicles (10–14 mm) on stimulation day 6 was significantly higher in “unpredictable” POR patients who received 150 IU of hCG at the beginning of rFSH administration than other groups. The reason behind this improvement is that LH binds to the receptor on the theca cells, which increases androgen secretion (21). Androgen improves granulose cell function and increases the size of the follicle of the cohort recruited (22, 23). Therefore, patients with poor response that resulted from deficiency of LH receptor activity may benefit from LH supplementation (24).

Follicular recruitment occurs in the early follicular phase during OS (25). The size of this follicular cohort determines the number of retrieved oocytes in accordance with the number of useable embryos. We found that a higher number of retrieved oocytes and useable embryos were seen in “unpredictable” poor responders who employed 150 IU of hCG in early follicular phase during the second GnRH antagonist stimulation protocol. But the same dose of hCG addition in mid-late follicular phase was not able to increase the number of oocytes and embryos. Therefore, different timing for hCG addition could be a possible reason to explain different conclusions drawn by previous studies. Based on our findings, we proposed the early follicular phase of hCG addition during OS.

In addition, on the day of hCG trigger, endometrial thickness and the progesterone level showed no significant differences between the hCG addition and no additional cycle. Furthermore, the timing of hCG administration, on either the first day or sixth day of OS, also had no effects on endometrial receptivity. The similar implantation and clinical pregnant rate between Group A and Group B in this study have proved it.

In conclusion, this study firstly showed that early follicular phase addition of hCG significantly increased the number of retrieved oocytes and embryos in patients with “unpredictable” poor response to GnRH antagonist regimens. As a result, the implantation rate, clinical pregnancy rate, and ongoing pregnancy rate were also observed at an increased trend in the “unpredictable” POR patients receiving early hCG supplement. Potential disadvantages of this study include retrospective design and relatively small sample size. There remains a lack of well-designed, randomized, controlled trials evaluating this protocol, and additional research is warranted.

The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding author.

This study was approved by Peking University Shenzhen Hospital Ethics Committee. Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements.

CZ and WQ designed the study and performed all ovum pick-up operation; CZ, WQ, BS, and CC performed the ovarian stimulation; CZ, FW and ZW collected and analyzed the data, also wrote the manuscript. All authors contributed and approved the submitted version.

This study received funding from National Key Research and Development Program (2018YFC1002104).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

The reviewer QX declared a shared affiliation, with no collaboration, with the authors to the handling editor at the time of review.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.