Abstract
Background:
There is no absolute consensus for the best time for triggering. The aim of this study was to investigate the effect of different proportion of dominant follicles (PDF) on the human chorionic gonadotropin (HCG) day for the clinical outcomes in patients with polycystic ovary syndrome (PCOS) of different ovarian stimulation protocols.
Methods:
A total of 371 cycles of the gonadotropin-releasing hormone (GnRH) agonist long protocol and 347 cycles of GnRH antagonist protocol from January 2014 to December 2016 were included in this retrospective study. Based on the PDF on the day of the HCG administration, the included patients were divided into three groups: Group A (low PDF), PDF <20%; Group B (medium PDF), 20%≤ PDF ≤40%; Group C (high PDF), PDF >40%. The measurements regarding ovarian stimulation characteristics, fertilization rate, top quality embryo rate, clinical pregnancy rate, and ovarian hyperstimualtion syndrome (OHSS) rate were compared in different PDF groups with different protocols.
Results:
In both the GnRH antagonist protocol and GnRH agonist long protocol, the characteristics such as mean age, anti-Mullerian hormone, antral follicle count (AFC), and body mass index were comparable between groups. The number of oocytes retrieved decreased statistically significantly as the PDF and rate of matured oocytes increased. In the GnRH agonist long protocol, the rate of normally fertilized oocytes was highest in Group A (59.74 ± 31.21 vs. 49.70 ± 37.95, 49.67 ± 36.62; F = 3.743, P = 0.025). There were no significant differences in the rate of top-quality embryos and the clinical pregnancy rate between the groups. The clinical pregnancy rate was similar in the three groups (63.6%, 62.5%, 67.5%, respectively, χ2 = 0.989, P = 0.911). The moderate and severe OHSS rate increased statistically significantly when the PDF increased, which was highest in group C (1.4%, 3.1%, 6.7%, respectively, χ2 = 12.014, P = 0.017). In the GnRH antagonist protocol, there were no significant differences in the rate of top-quality embryos, the rate of normally fertilized oocytes, the clinical pregnancy rate, and the moderate and severe OHSS rate between the groups. The clinical pregnancy rate in Group C was higher than that in Group A (57.9% vs. 46.6%, χ2 = 10.850, P = 0.093).
Conclusions:
In the GnRH antagonist protocol, PDF on the HCG day of less than 20% may be unfavorable to the clinical pregnancy rate in PCOS. In the GnRH agonist long protocol, delaying the HCG trigger timing has no good effect on clinical pregnancy and the risk of OHSS might increase in patients with PCOS.
Keywords: Proportion of dominant follicles, GnRH antagonist protocol, GnRH agonist long protocol
Introduction
Polycystic ovary syndrome (PCOS) is one of the most common female reproductive disorders affecting approximately 10% of reproductive age women. This population is at an extremely high risk of ovarian hyperstimualtion syndrome (OHSS). With the development of assisted reproductive technology (ART), there are more and more issues concerned about how to decrease or avoid OHSS. Currently, a gonadotropin-releasing hormone (GnRH) antagonist protocol is strongly recommended in PCOS, which became an important part of the guideline for OHSS prevention from the American Society for Reproductive Medicine as Grade A evidence level.[1,2]
The GnRH antagonist protocol was undertaken for nearly 20 years in China. However, the clinical pregnancy rate was lower than that in the classic GnRH agonist long protocol by 8% to 10%.[3,4] There are many factors which affect the outcome, including ovarian response, protocol selection, trigger timing, and luteal phase support.[5,6] Some professors have proposed that the human chorionic gonadotropin (HCG) administration criteria in the GnRH antagonist protocol should be different from the GnRH agonist long protocol.[7] The question as to whether it is better to delay or put forward the time of HCG administration in the different protocols remains elusive.
HCG administration for triggering the final oocyte maturation is the last and key procedure for controlled ovarian hyperstimulation (COH), but the optimal criteria of the HCG trigger for all patients in a variety of protocols is not available. Previously, HCG was administered when there was one or two leading follicles ≥18 mm or three follicles ≥17 mm in diameter. The number of dominant follicles was preferred as an important item for HCG administration. The number of pre-ovulatory follicles available at the end of ovarian stimulation may be not a reliable reflection of cycle prognosis. Obviously, it is not suitable for all patients with different ovarian reserve especially in PCOS with a high ovarian reserve and in different COH protocols.
PDF was proposed as it can reflect the overall follicle development. This study was based on the previous study after the advantage of PDF on HCG day.[8] We retrospectively analyzed the different PDFs and clinical outcomes in PCOS of the different protocols and explored the role of PDF in the HCG triggering decision.
Methods
Ethical approval
This study was approved by the Institutional Review Board at Sun Yat-sen Memorial Hospital (2016, No. 44). The data in this study were collected from the reproductive center of Sun Yat-sen Memorial Hospital between January, 2014 and December, 2016. There were a total of 371 cycles of the GnRH agonist luteal long protocol and 347 cycles of GnRH antagonist protocol in patients with PCOS. The PCOS diagnosis was according to the Rotterdam criteria in 2003. The baseline characteristics were compared including age, antral follicle count (AFC), anti-Mullerian hormone, baseline serum follicle stimulation hormone (FSH). The items such as the number of leading follicles ≥18 mm, the number of follicles ≥10 mm on HCG day, the mature oocytes rate, the fertilization rate, the clinical pregnancy rate, and moderate/severe OHSS rate were compared. We excluded cycles cancelled before HCG triggering.
GnRH agonist long protocol
Long-acting GnRHa (0.93–1.25 mg intramuscular [i.m.] once; IPSEN, Paris, France) or a short-acting GnRHa (0.1 mg i.m. per day until HCG day; IPSEN) was usually administered on the 18th to 22nd days of menstrual cycle (5–7 days after ovulation) after the corpus luteum detected on the vaginal ultrasound, if necessary, we measured serum progesterone level to determine the ovulation. Two to 3 weeks later, the vaginal ultrasound and blood hormones test (FSH, luteinizing hormone [LH], estradiol [E2]) were necessary. Gonadotropin 100 to 300 U/day can be administered when the results fully reached the pituitary complete down-regulation criteria. Every 3 to 5 days, the ultrasound and blood hormones were monitored to evaluate follicular development, when at least one follicle ≥18 mm or 3 follicles ≥17 mm in diameter emerges, HCG (Livon, China) 4000 to 10,000 U or GnRHa (IPSEN) 0.2 mg can be used as a trigger. Thirty-six to 38 h later, the vaginal oocyte retrieval was performed, embryos were cultured until the 3rd to 5th day, and then the embryo transfer was performed. Corpus luteal support began from the oocyte recovery with a progesterone intramuscular injection 60 mg/day or progesterone capsules 600 mg/day trans-vaginal until the day of pregnancy test.
GnRH antagonist protocol
On the 2nd or 3rd day of menstrual cycle, the sex hormones FSH, LH, E2, and the vaginal ultrasound were performed to evaluate the bilateral ovarian follicular condition. Gonadotropin 75 to 300 U/day was administered for ovarian stimulation, every 3 to 5 days to monitor ultrasound and blood hormones to evaluate follicular growth. GnRH antagonist was added according to the flexible or fixed protocol. When at least one follicle ≥18 mm or 3 follicles ≥17 mm in diameter merged, HCG (Livon) 4000 to 10,000 U or GnRHa (IPSEN) 0.2 mg was administered to trigger the oocyte final maturation. The oocyte retrieval, embryo transfer, and a luteal support regime were the same as in the GnRH agonist long protocol.
Measurements
PDF was obtained by the number of ≥18 mm follicles/number of ≥10 mm follicles on the HCG day. We divided the cycles into three sub-groups according to PDF in the GnRH agonist long protocol and the GnRH antagonist protocol, respectively. The tertile borderline of ≥18/≥10 mm follicles proportion was 20% and 40%, respectively. Proportion of leading follicles more than 18 mm on HCG day more than 40% was set high PDF, and medium PDF: between 20% and 40%, low PDF was less than 20%.
The primary outcome of interest was the clinical pregnancy rate. The secondary measurements were the mature oocyte rate, the normally fertilized rate, and the rate of moderate and severe OHSS. We also analyzed the individual characteristics of cycles.
Statistical analysis
The data was subjected to a normality test using the one-sample Kolmogorov-Smirnov method. The variance homogeneity test was performed on the measurement data. Normal distribution data were presented as mean ± standard deviation (SD), and non-normal distribution data were presented by median (Q1–Q3). The variance homogeneity data were analyzed by analysis of variance (ANOVA); the comparison between the two was analyzed by least significant difference and Tamhane method. The Chi-square test is used for qualitative data and is expressed as a percentage. Baseline and demographic characteristics were analyzed by ANOVA, the results were showed with mean ± SD, and a Chi-square test was used for categorical data such as clinical pregnancy rate. Non-normal distribution data were analysed by Kruskal-Wallis test. Logistic regression analysis, forward Likelihood ratio (LR), and indicators related to clinical pregnancy were observed. All data were tested at the two-sided level of 0.05. Analysis was performed using SPSS version 19.0 (IBM® SPSS® Statistics, International Business Machines Corp, Chicago, IL, USA).
Results
Comparison of baseline data of patients
In GnRH agonist long protocol and GnRH antagonist protocol, the basic clinical data such as age, body mass index, basal FSH, LH, testosterone, and AFC were compared between the three groups. There were no statistical differences in these basic characteristics and serum hormones on the HCG day among the three groups of different PDFs. Only in GnRH antagonist protocol, serum E2 on the HCG day in low PDF group was statistically significantly higher than the other groups [Tables 1 and 2].
Table 1.
Analysis of basic characteristics of patients in GnRH agonist long protocol.
Table 2.
Analysis of basic characteristics of patients in GnRH antagonist protocol.
Comparison of embryo culture and outcome measurements
In GnRH agonist long protocol, the rate of normally fertilized oocytes was highest in group A (59.74 ± 31.21 vs. 49.70 ± 37.95, 49.67 ± 36.62; F = 3.743, P = 0.025). There was no significant difference in the rate of top-quality embryos, the clinical pregnancy rate among the groups. The clinical pregnancy rate was similar in three groups (63.6%, 62.5%, 67.5%, χ2 = 0.989, P = 0.911). While the implantation rate was comparable in the three groups, the implantation rate in Group C was lower than those in other two groups (35.8% vs. 47.7%, 46.7%, χ2 = 3.600, P = 0.165). The moderate and severe OHSS rate increased statistically significantly with an increase in PDF, which was highest in group C [Table 3].
Table 3.
Analysis of outcomes of patients in GnRH agonist long protocol.
In GnRH antagonist protocol, there was no significant difference in the rate of top-quality embryos, the rate of normally fertilized oocytes, the clinical pregnancy rate, the implantation rate, and moderate and severe OHSS rate between groups. The clinical pregnancy rate in Group C was higher than that in Group A (57.9% vs. 46.6%, χ2 = 10.850, P = 0.093) [Table 4].
Table 4.
Analysis outcomes of patients in GnRH antagonist protocol.
Logistic regression analysis of factors associated with clinical pregnancy rate
In the GnRH agonist long protocol, only the total dose of the HCG trigger and AFC were associated with the clinical pregnancy rate, but it was not statistically significantly [Table 5].
Table 5.
Logistic regression on clinical pregnancy rate in GnRH agonist long protocol (n = 371).
In the GnRH antagonist protocol, the number of available embryos was associated with the clinical pregnancy rate (odds ratio = 1.320) [Table 6].
Table 6.
Logistic regression on clinical pregnancy rate in GnRH antagonist protocol (n = 347).
Discussion
In the ovulation cycle, HCG was administered to induce ovulation when the follicles had reached 18 to 22 mm in diameter; the administration time was based on the follicle diameter to determine the oocyte maturity. The timing of the administration of HCG plays a vital part in in vitro fertilization/intra-cytoplasmic sperm injection cycles. Premature administration can lead to follicular atresia, delayed administration can cause follicular aging and a decrease in endometrial receptivity, and it may increase the rate of OHSS. Patients with PCOS were characterized by a high ovarian reserve and a good ovarian response to FSH. In COH of patients with PCOS, a large cohort of medium size follicles are easily seen on HCG day, if the trigger time is postponed and PDF increases, the serum E2 level and OHSS risk increase statistically significantly accordingly. However, early triggering may affect the oocyte maturity. The clinical phenomenon shows that large follicles (diameter greater than 23 mm) yield a low oocyte recovery rate. Medium size follicles (between 16 and 18 mm) yield the most transferable (transferred or cryopreserved) embryos compared with the small or large size follicular groups.[9] The traditional criteria is according to 2 to 3 follicles ≥17 to 18 mm on HCG day to determine the timing of the HCG injection for all patients. However, there may be a bias and limitation. If young patients with a high ovarian reserve such as in PCOS undertake the same criteria for HCG injection, the follow-up of the pregnancy outcome in different ovarian stimulation protocols may be completely different. Hence, the proportion of dominant follicles which is used to determine the time of HCG injection may be not suitable for PCOS, in all the protocols. It is necessary to explore the item for the individual HCG injection timing.
In 2014, we proposed the use of HCG on the proportion of mature follicles (the number of ≥18/≥10 mm follicles) to study the effect of different proportions on oocyte development and pregnancy outcome on the HCG day. And this retrospective study analyzed outcomes in different PDF of PCOS with different protocols, and further explored the most suitable PDF for HCG injection in PCOS. In our previous study concerned male factor infertility patients excluding PCOS, we found that the high oocyte maturation rate was statistically significantly increased in the high proportion group,[8] but the embryo quality, implantation rate, and pregnancy rate decreased. GnRHa long protocol takes advantage of complete inhibition of hypothalamic-pituitary-ovarian (HPO)axis, the number of oocytes and pregnancy outcomes are closely related to postpone HCG injection.[10] The impact of low luteinization on the endometrium is limited. The effect of the antagonist protocol on the HPO axis is small. The oocyte growth is faster. The time of HCG injection may affect the number of oocytes and the subsequent embryo development. A delayed HCG injection, increased of mature oocyte rate, but did not increase the clinical pregnancy rate and may have increased the OHSS rate. Studies have suggested that it may be better for the clinical outcome to increase the proportion of mature follicles in the antagonist protocol.[11] The results suggest that it is good for the clinical outcome to increase the proportion of dominant follicles in PCOS.
Actually the oocytes retrieved are from a cohort of follicles not only just from 2 to 3 leading follicles, the proposal of PDF was based on the target cohort follicles. As the total number of follicles on HCG day, including some small follicles, such as number of ≥10 mm follicles has a big individual difference on the protocol differences. The dominant follicle usually begins at a size of 10 mm during the stimulation, so the 10 mm follicles are chosen as the denominator of the PDF because it can better reflect the target cohort of follicle's development. PDF can be a good item for HCG trigger time decision. In order to optimize the advantages of PDF, we retrospectively analyzed the large sample of clinical data in different stimulation protocols to guide clinical decision making on HCG triggering in PCOS.
The retrospective study showed that the clinical outcome was not affected in the different PDF groups in the GnRH agonist long protocol. Delaying the HCG trigger has no positive effect on the clinical pregnancy of fresh cycles, but the risk of OHSS may increase in patients with PCOS. In the GnRH antagonist protocol, the number of ≥18/≥10 mm follicles on HCG day less than 20% was unfavorable to the clinical pregnancy rate of fresh cycles in the patients with PCOS.
In order to balance the OHSS risk and clinical pregnancy in fresh cycles, a PDF of 20% to 40% may be recommended in PCOS either in GnRH agonist long protocol or GnRH antagonist protocol. However in GnRH antagonist protocol, results showed that the clinical pregnancy rate decreased in low PDF group. Based on the sub-analysis, we proposed the following reasons:
First, the rate of cycle cancelled due to OHSS risk was highest in the low PDF group. There may be a bias in patients selection in this retrospective study. Second, the rate of GnRHa trigger was highest in the low PDF group. It could not exclude the negative effect of GnRHa trigger on luteal phase function and endometrial receptivity. Last, earlier trigger may affect the maturity synchronization of the oocyte nucleus and cytoplasm, as HCG/LH/GnRHa trigger is closely related to the maturity of oocyte nucleus.
The PDF on HCG day can reflect the synchronization of follicular development; a delayed HCG injection can increase the rate of mature oocytes, but early progesterone elevation may appear, and endometrial receptivity is affected. The appropriate PDF on the HCG day, can make the appropriate number of oocytes, synchronized maturation of the cytoplasm and the nucleus, the number of high-quality embryos to ensure the outcome of pregnancy. Previously, some scholars have proposed the concept of follicle output rate.[12–14] In our view, the number and quality of follicles on HCG day is the key factor that affects the outcome of the ART. At present, the relevant research is very rare, how to seize the specific population such as PCOS, the optimum PDF on the HCG day to ensure the prognosis of patients is one of the purposes of this study.
We clearly know the limitations of this study. It was a retrospective study, so some confounding bias may exist. The PDF was based on the measurements by ultrasound, which may be affected by different operators. A large size, multi-center and strictly designed randomized study is needed.
Funding
This study was supported by the grants from the National Natural Science Funding (No. 81671523), the Natural Science Funding of Guangdong Province (No. 4210015092 and No. 2017A030313895), and the Funding of Yat-sen Scholarship for Young Scientist.
Conflicts of interest
None.
Footnotes
How to cite this article: Lin HY, Li Y, Wang WJ, Qiu Q, Zhang QX, Li Y. Role of the proportion of dominant follicles in patients with polycystic ovary syndrome undergoing in vitro fertilization-embryo transfer. Chin Med J 2019;00:00–00. doi: 10.1097/CM9.0000000000000246
References
- 1.Pfeifer S, Butts S, Dumesic D, Fossum G, Gracia C, La Barbera A, et al. Prevention and treatment of moderate and severe ovarian hyperstimulation syndrome: a guideline. Fertil Steril 2016; 106:1634–1647. doi: 10.1016/j.fertnstert.2016.08.048. [DOI] [PubMed] [Google Scholar]
- 2.Royal College of Obstetricians, Gynaecologists (RCOG). The management of ovarian hyperstimulation syndrome: green top guideline number 5. RCOG 2006; 1–11. [Google Scholar]
- 3.Lambalk CB, Banga FR, Huirne JA, Toftager M, Pinborg A, Homburg R, et al. GnRH antagonist versus long agonist protocols in IVF: a systematic review and meta-analysis accounting for patient type. Hum Reprod Update 2017; 23:560–579. doi: 10.1093/humupd/dmx017. [DOI] [PubMed] [Google Scholar]
- 4.Wang R, Lin S, Wang Y, Qian W, Zhou L. Comparisons of GnRH antagonist protocol versus GnRH agonist long protocol in patients with normal ovarian reserve: a systematic review and meta-analysis. PLoS One 2017; 12:e0175985.doi: 10.1371/journal.pone.0175985. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Chi HB, Liu NN, Li R, Tao LY, Chen LX, Qiao J. Comparison of vaginal gel and intramuscular progesterone for in vitro fertilization and embryo transfer with gonadotropin-releasing hormone antagonist protocol. Chin Med J 2018; 131:1557–1561. doi: 10.4103/0366-6999.235106. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Zhang HJ, Song XR, Lü R, Xue FX. Modified super-long down-regulation protocol improves fertilization and pregnancy in patients with poor ovarian responses. Chin Med J 2012; 125:2837–2840. [PubMed] [Google Scholar]
- 7.Kolibianakis EM, Albano C, Camus M, Tournaye H, Van Steirteghem AC, Devroey P. Prolongation of the follicular phase in in vitro fertilization results in a lower ongoing pregnancy rate in cycles stimulated with recombinant follicle-stimulating hormone and gonadotropin-releasing hormone antagonists. Fertil Steril 2004; 82:102–107. doi: 10.1016/j.fertnstert.2004.01.027. [DOI] [PubMed] [Google Scholar]
- 8.Li Y, Li RQ, Ou SB, Zhang NF, Wang WJ, Li Y, et al. Proportion of dominant follicles as the guidance of HCG triggering and its effect on the oocyte development competence (in Chinese). J Sun Yat-sen Univ (Med Sci) 2014; 35:552–558. [Google Scholar]
- 9.Nivet AL, Léveillé MC, Leader A, Sirard MA. Transcriptional characteristics of different sized follicles in relation to embryo transferability: potential role of hepatocyte growth factor signalling. Mol Hum Reprod 2016; 22:475–484. doi: 10.1093/molehr/gaw029. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Vandekerckhove F, Gerris J, Vansteelandt S, De Baerdemaeker A, Tilleman K, De Sutter P. Delaying the oocyte maturation trigger by one day leads to a higher metaphase II oocyte yield in IVF/ICSI: a randomised controlled trial. Reprod Biol Endocrinol 2014; 12:31.doi: 10.1186/1477-7827-12-31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Hu X, Luo Y, Huang K, Li Y, Xu Y, Zhou C, et al. New perspectives on criteria for the determination of HCG trigger timing in GnRH antagonist cycles. Medicine 2016; 95:e3691.doi: 10.1097/md.0000000000003691. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Gallot V, Berwanger da Silva AL, Genro V, Grynberg M, Frydman N, Fanchin R. Antral follicle responsiveness to follicle-stimulating hormone administration assessed by the follicular output rate (FORT) may predict in vitro fertilization-embryo transfer outcome. Hum Reprod 2012; 27:1066–1072. doi: 10.1093/humrep/der479. [DOI] [PubMed] [Google Scholar]
- 13.Genro VK, Matte U, De Conto E, Cunha-Filho JS, Fanchin R. Frequent polymorphisms of FSH receptor do not influence antral follicle responsiveness to follicle-stimulating hormone administration as assessed by the follicular output rate (FORT). J Assist Reprod Genet 2012; 29:657–663. doi: 10.1007/s10815-012-9761-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Zhang N, Hao CF, Zhuang LL, Liu XY, Gu HF, Liu S, et al. Prediction of IVF/ICSI outcome based on the follicular output rate. Reprod Biomed Online 2013; 27:147–153. doi: 10.1016/j.rbmo.2013.04.012. [DOI] [PubMed] [Google Scholar]