Relation of graphical follicle models to the cumulative live birth rate in GnRH-antagonist stimulation treatment cycles

doi:10.21203/rs.3.rs-2008137/v1

Download PDF

Research Article

Relation of graphical follicle models to the cumulative live birth rate in GnRH-antagonist stimulation treatment cycles

https://doi.org/10.21203/rs.3.rs-2008137/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background: Previous studies have observed the number and size of follicles are two independent indicators of whether the oocyte is adequate and mature and are used to predict the trigger time in routine clinical practice. However, due to individualized differences, it currently relies mainly on physician experience. It is necesaary to explore an innovative, simple, and graphical model for predicting clinical results in ovary stimulation treatment and facilitating personalized protocol adjustment.

Methods: This retrospective study included 8405 patients who started their first in vitro fertilization cycle with a GnRH-ant protocol including fresh and subsequent frozen-thawed cycles during 2016-2020. Using follicle size and number, we classified follicles recorded on the human chorionic gonadotropin (HCG) day by size (small, medium, large). Producing graphical models, and classified into Inverted-trapezoid (large follicles in dominant proportion), Polygon (moderate follicles in dominant proportion), Trapezoid (small follicles in dominant proportion), and Rectangle (equivalent proportions of the three size categories). The Cochran-Mantel-Haenszel (CMH) and Generalized Linear Model(GLM) were used to evaluate the difference among models about maturity, fertilization, and the number of viable embryos, as well as cumulative live birth rate(CLBR).

Results: In GLM analysis, after adjusting the confounders, there are differences between models of CLBR. The CLBR of the different models was higher in the Polygon and Inverted-trapezoid model than Trapezoid and Rectangle model (42.75%, 39.56%, vs. 37.38%, 28.57%, respectively; all P<0.05). For oocytes derived from very large follicles (>20 mm), the CLBR was lower than that of patients with ≤20 mm follicles [26.10% vs. 42.10%, OR = 1.74 (95% confidence interval 1.52–2.00), P<0.01] in Inverted-trapezoid model, but there was no difference between models.

And the risk of ovarian hyperstimulation syndrome (OHSS) rate of patients with ≤20 mm follicles was lower than that of patients with >20 mm follicles [8.64% vs. 17.89%, OR = 0.57 (95%CI: 0.49-0.65), P<0.01] in Inverted-trapezoid model. Patients who received an adjusted Gn dose(whether Decreased or Increased-dose protocol) showed no difference in CLBR among models (Polygon vs Inverted-trapezoid vs Trapezoid vs Rectangle model: 47.07% vs. 49.21% vs. 47.69% vs. 42.42%, P>0.05); but when patients continued with the same starting dose in Fixed-dose protocol, the CLBR of the Polygon model was higher than that of other models (40.43% vs. 34.32, 31.13% vs. 26.46%, respectively; all P value <0.05).

Conclusion：Our results suggested the follicle models can demonstrate characteristics and the ovarian response of patients. In addition to the patients in very good condition who develop a Polygon model, it is better for other patients to reach the Inverted-trapezoid model with follicles >18 mm and <20 mm. Adjusting the protocol is critical to the outcome. But determining how to improve follicle models by regulating stimulation remains to be further studied.

Cumulative live birth rate

ovarian follicle

controlled ovarian stimulation

gonadotropin releasing hormone antagonist stimulation

The main purpose of assistant reproductive technology (ART) treatment is to achieve a live birth. However, as the retrieval oocyte is considered to be an important prognostic variable[1], treatment protocols aim to optimize the outcome. During controlled ovarian stimulation (COS), different sizes of follicles develop, monitored by transvaginal sonography, and human chorionic gonadotropin (HCG) is usually provided at the end of COS to induce final follicular maturation[2]. Doctors determine whether the follicle size, number, and serum hormone levels meet trigger requirements, so as to prevent the retrieval of immature oocytes. It is known that oocyte maturity is related to follicle size[3]. The most widely applied protocol to start the trigger is when several follicles reach a diameter of >17 or 18 mm[4].

Previous studies have observed the association between the number and size of follicles and in vitro fertilization (IVF) outcomes. Timeva et al.[5] demonstrated that fewer than five oocytes considerably reduced the IVF pregnancy rate, and >15 oocytes did not improve the outcome but increased the risk of ovarian hyperstimulation syndrome (OHSS). A published study showed that the cumulative live birth rate (CLBR) increased with a higher number of oocytes retrieved, up to around 20 oocytes[6]. However, a prospective study found a higher normal fertilization rate of metaphase II (MII) oocytes, and the rate of good-quality embryos in the large follicle group compared to the small follicle group, but the difference was not statistically significant[7]. Other studies have found a significantly higher MII oocyte rate in large follicles than in small ones, but no difference in fertilization rates and blastocyst rates[3]. Ozgur et al [8] observed that the number of retrieved oocytes was not independently predictive of CLBR in one complete freeze-all cycle among 1582 patients in a general infertile population, but conflicting results have been reported regarding the relationship between the number and size of oocytes retrieved and IVF outcome[10-13].

A consensus has been reached on the preferred primary outcome of all infertility treatment trials, which is the CLBR. This requires a long period including fresh and subsequent frozen-thawed cycles from the initial oocyte retrieval, and there are different equations used to calculate the CLBR [14-16]. Moreover, CLBR might help patients to decide on whether to continue with the treatment or to remain childless.

Hence, in this study, we aimed to compare the CLBR of different models graphically represented by the number and size of follicles on the HCG day in the gonadotropin releasing hormone antagonist (GnRH-ant) stimulation cycle. To provide evidence for personalized protocol adjustment.

Study design

This was a retrospective real-world data study, linking the information of patients from the Electronic Medical Record System (EMRS) to evaluate women who commenced their first COS with the gonadotropin releasing hormone antagonist (GnRH-ant) protocol (both IVF and Intracytoplasmic Sperm Injection [ICSI]) including fresh and subsequent frozen-thawed cycles, from January 2016 to June 2020.

Participants

All patients included in this study were diagnosed with pelvic environment, tubal factor infertility, ovulatory dysfunction, diminished ovarian reserve, or male factor infertility undergoing their first COS cycle with the GnRH-ant protocol. Patients with preimplantation genetic testing (PGT) cycles, male/female chromosome abnormalities, uterine malformation, endometriosis, and unexplained infertility were excluded from the data analysis. Finally, a total of 8405 patients were analyzed including 8178 patients who either had a live birth or transferred all fresh and frozen embryos, and another 227 patients who underwent embryo transfer at least twice without a live birth were followed-up until research finished but did not return to continue their treatment (regarded as no live birth) [17].

Data sources

(1)GnRH-antagonist protocol

COS was performed using the GnRH-ant protocol, as previously described[18]. The starting dose should be determined by antral follicle count (AFC), age, body mass index (BMI), and other factors. A starting dose of 100–149 IU is recommended for patients with a very high ovarian response, 150–225 IU for those with normal ovarian response, and 225–300 IU for those with a poor ovarian response[19].This was initiated on day 2 or 3 of the cycle with a dose of r-follicle stimulating hormone(rFSH) ranging from 75 to 300 IU. The patients were administered 0.25 mg of GnRH-ant (Orgalutran, Organon, the Netherlands or Cetrorelix, Merck Serono, Switzerland) daily if at least one of the following criteria was fulfilled: (i) at least one follicle of >14 mm; (ii) serum estrogen level > 600 pg/mL; and (iii) serum luteinizing hormone level > 10 IU/L[11]. During the follicular monitoring period, doctors will adjust the Gn dose (≤150 IU) less than two times according to the follicular growth (commonly on days 4–7). The adjustment was performed based on the experience of the physician.

(2) Follicle measurement

Transvaginal ultrasonography was performed by a senior physician every 1–2 days to monitor follicle development during ovarian stimulation and to record the follicle diameter (Hitachi Aloka Medical, Tokyo Japan) until the HCG day. We used the mean of vertical and horizontal ultrasonic measurements of the follicle (orthogonal diameters) as the diameter of a follicle. For the first 4 days after Gn administration, follicles with more than 5-mm diameter were recorded; for days 5–7 after Gn administration, follicles with more than 8-mm diameter were recorded; and for days 8–9 or day 10 and beyond, follicles with more than 10- and 12-mm diameter were recorded. All ultrasound measurements were checked by another physician.

(3) Graphical models of diameter and number of follicles on HCG day

Each follicle size was measured and recorded before oocyte aspiration. Follicles were grouped according to size on the HCG day: small-sized follicles with diameter <16 mm, moderate-sized follicles with 16–18-mm diameter, and large-sized follicles with >18 mm diameter[20]. We ensured that the follicles in each patient were monitored for all three groups. Combining follicle size and number, follicles of the same size were classified and centered to obtain the graphical model. There were four types of models (Inverted-trapezoid, Polygon, Trapezoid, Rectangle) including 25 combinations (Figure 1). If there were at least three follicles measuring >16 mm in diameter, then HCG (Merck Serono, Italy) was administered.

(4) Embryo culture and transfer

After trigger, transvaginal oocyte retrieval was performed at 36 h, and then, embryo transfer was performed on day 3 after oocyte retrieval. Luteal-phase support was started immediately after oocyte retrieval with vaginal combined oral progesterone. Most patients received double embryo transfer unless the patient had only one available embryo for single embryo transfer, according to the national guidelines[21]. Surplus available embryos or all-frozen embryos (owing to OHSS, thin endometrium, abnormal blood biochemical index, or personal reasons of patients) were frozen for later transfer in subsequent frozen embryo transfer (FET) cycles.

Outcome measures

The primary outcome was the CLBR, calculated as the number of live births per ovum pick-up, which was defined as the first live baby born at ≥28 weeks of gestation resulting from the first ART cycle (including fresh and subsequent FETs) divided by the number of patients who started their first ART cycle during the study period. If a live birth occurred, the patient obtains an outcome, regardless of subsequent cycles. According to this definition, multiple deliveries from the same pregnancy or multiple live births were considered one live birth. CLBRs were calculated as the proportion of cycles that achieved a first live birth.

Bias

Ovarian reactivity: ovarian response was not considered as inclusion or exclusion criteria. Control: stratified analysis was performed for Gn administration, adjusted according to ovarian response.
AFC: the AFC was not considered as inclusion or exclusion criteria. Control: The Cochran-Mantel-Haenszel (CMH) analysis and general linear model(GLM) were used to control bias.

Statistical methods

Data that showed normal distribution are presented as mean (standard deviation, SD) and the median is here considered descriptive for data of abnormal distribution, as appropriate. Analysis of variance (ANOVA-Bonferroni) of the difference between two groups. The CLBR in the GnRH-ant groups was compared using the chi-square test. Patient age was categorized into three groups: ≤35, and >35 years, the AFC was categorized into <5 and ≥5 and anti-Müllerian hormone(AMH) was categorized into <1.2ng/ml and ≥1.2ng/ml [22]. The Cochran–Mantel–Haenszel(CMH) test by different ages, AFCs, AMH, and GLM were used to compare the CLBR, MII oocyte, normal fertilization, and viable embryo between models. We performed CMH analysis for comparison of Gn start dose and Gn adjustment between models. P-values < 0.05 were considered to indicate statistical significance. All analyses were conducted using Stata version 15.1 (StataCorp LLC, College Station, TX, USA) and RStudio Team 2015 (RStudio, Inc., Boston, MA, USA).

Participants

According to the inclusion and exclusion criteria, a total of 8405 patients were identified as eligible for participation in this analysis. Epidemiological, clinical, and biological characteristics of the study population are summarized in Table 1. Patients with good prognosis made up a high percentage of those who showed a Polygon model, but those with poor prognosis patients were highly represented in the Rectangle model. The characteristics of patients were similar in the Inverted-trapezoid and Trapezoid model.

Table 1 Baseline characteristics of patients receiving IVF treatment

Characteristics	Inverted-trapezoid(95%CI)	Polygon(95%CI)	Trapezoid(95%CI)	Rectangle(95%CI)
Age(years)	31.88±4.89(31.59-32.17)a	33.68±5.29(33.5-33.86)b	32.88±4.99(32.7-33.07)c	34.61±5.02(34.32-34.89)d
Basal FSH (mIU/ml)	12.82±6.99(12.41-13.24)a	14.35±6.18(14.14-14.56)b	14.16±6.09(13.93-14.38)b	16.37±7.49(15.95-16.8)c
AFC	9.89±5.34(9.58-10.2)a	7.35±4.79(7.19-7.52)b	7.82±4.87(7.64-7.99)c	5.43±3.53(5.23-5.63)d
BMI(kg/m²)	22.19±3.05(22.01-22.37)	22.33±2.95(22.23-22.43)	22.36±3.03(22.25-22.47)	22.41±2.94(22.25-22.58)
AMH(ng/ml)	5.26±5.04(4.96-5.56)a	3.45±4.14(3.30-3.59)b	3.73±4.3(3.57-3.89)b	1.94±2.77(1.78-2.1)c
Fertilization
IVF	76.12%(74.57-77.60)^a	74.55%(71.90-77.03)^a	73.56%(71.85-75.20)^a	78.24%(75.82-80.48)^b
ICSI	23.88%(22.40-25.43)^a	25.45%(22.97-28.10)^a	26.44%(24.80-28.15)^a	21.76%(19.52-24.18)^b

*Different superscript letters show statistical differences of mean values within same parameter (column) (P<0.05).

*a vs b vs c, P value <0.05, respectively; a,b vs a, a,b vs b, same letter means no difference.

Details regarding embryology and CLBR of different models

In general linear model for univariate(GLM-Univariate) analysis, controlling the confounder of age, FSH, AFC, BMI, AMH, and Fertilization, it was observed that the number of MII oocytes, normal fertilization, retrieval oocyte, and viable embryos were significantly higher in the Polygon model(11.37±7.73, 8.55±6.34, 12.98±8.46, 4.66±3.80) than in other models (all P value<0.001), and it showing that were lower in Rectangle model(4.95±4.11, 3.68±3.30, 5.76±4.61, 2.28±1.90) than other models (all P value<0.001). There was no difference observed between the Inverted-trapezoid and Trapezoid model of these variables(Table 2).

In GLM analysis, control the confounder of age, FSH, AFC, BMI, AMH, Fertilization, retrieval oocyte and viable embryo, there are differences between models of CLBR. The CLBR was higher in the Polygon and Inverted-trapezoid model than in the Trapezoid and Rectangle models (42.75%, 39.56% vs. 37.38%, 28.57%, respectively; all P value <0.05). Considering characteristics, we stratified patients by age, AFC, and AMH into subgroups. In Age<35, AFC≥5, and AMH≥1.2 group, the CLBR of the Inverted-trapezoid model was higher than in the Polygon and Trapezoid model(54.15% vs. 48.66%, 46.81%, respectively; all P value <0.05); In Age<35, AFC<5 and AMH<1.2 group, the CLBR of Inverted-trapezoid model was higher than in Trapezoid model(37.88% vs. 25.51%; P value <0.05). And in AFC<5, AMH≥1.2, and AFC≥5, AMH<1.2 groups, it was found that CLBR was higher in the Polygon model than in other models though there was no statistical difference(Table 2, Figure 2A).

Table 2 CLBR rate and related COS outcomes between models

			Polygon model (n=1115)	Inverted-trapezoid model (n=3258)	Trapezoid model (n=2828)	Rectangle model (n=1204)	P value
Retrieval oocyte			12.98±8.46^a	9.12±6.66^b	9.58±6.62^b	5.76±4.61^c	<0.001
MII oocyte			11.37±7.73^a	7.87±5.9^b	7.8±5.98^b	4.95±4.11^c	<0.001
Normal fertilization			8.55±6.34^a	5.93±4.73^b	5.87±4.87^b	3.68±3.30^c	<0.001
Viable embryo			4.66±3.8^a	3.29±2.64^b	3.34±2.94^b	2.28±1.90^c	<0.001
CLBR			42.75%^b	39.56%^b	37.38%^a	28.57%^a	<0.001
Age<35	AFC≥5	AMH≥1.2	326/670(48.66%)^a	672/1241(54.15%)^b	587/1254(46.81%)^a	123/254(48.43%)	0.005
Age≥35	AFC≥5	AMH≥1.2	46/150(30.67%)	110/402(27.36%)	96/340(28.24%)	26/107(24.3%)	NA
Age<35	AFC<5	AMH<1.2	15/50(30%)	111/293(37.88%)^b	62/243(25.51%)^a	58/192(30.21%)	0.007
Age≥35	AFC<5	AMH<1.2	11/63(17.46%)	89/514(17.32%)	54/322(16.77%)	46/300(15.33%)	NA
Age<35	AFC<5	AMH≥1.2	15/34(44.12%)	50/108(46.3%)	37/115(32.17%)	16/53(30.19%)	NA
Age≥35	AFC<5	AMH≥1.2	8/24(33.33%)	28/161(17.39%)	26/149(17.45%)	15/84(17.86%)	NA
Age≥35	AFC≥5	AMH<1.2	16/55(29.09%)	80/308(25.97%)	44/197(22.34%)	19/109(17.43%)	NA
Age<35	AFC≥5	AMH<1.2	35/69(50.72%)	101/231(43.72%)	89/208(42.79%)	41/105(39.05%)	NA

*In general linear model for univariate(GLM-Univariate) analysis, control the confounder of age, FSH, AFC, BMI, AMH and Fertilization, there are differences between models of the number of retrieval oocyte, MII oocyte, normal fertilization embryo and viable embryo.

*In generalized linear model(GLM), control the confounder of age, FSH, AFC, BMI, AMH, Fertilization, retrieval oocyte and viable embryo, there are differences between models of CLBR.

*Different superscript letters show statistical differences of mean values within same parameter (column) (P<0.05).

*a vs b vs c vs d, P value <0.05, respectively; a,b vs a, a,b vs b, same letter means no difference.

For oocytes derived from very large follicles (>20 mm), the CLBR was lower than that of patients with ≤20 mm follicles [26.10% vs. 42.10%, OR = 1.74 (95% CI 1.52–2.00), P<0.01] in Inverted-trapezoid model, but there was no difference between models. For follicles≤20 mm, the CLBR of Polygon and Inverted-trapezoid model was higher than Trapezoid model, and the lowest CLBR of Rectangle model(43.03%, 42.10% vs. 35.15% vs. 28.65%, respectively; all P value <0.05). In GLM analysis, control the confounder of age, FSH, AFC, BMI, AMH, Fertilization, retrieval oocyte and viable embryo, the follicles >20mm increased the OHSS rate[β=0.71(0.50~0.92), P<0.001] . And the OHSS rate of patients with ≤20 mm follicles was lower than that of patients with >20 mm follicles [8.64% vs. 17.89%, OR = 0.57 (95%CI: 0.49-0.65), P<0.01] in Inverted-trapezoid model.(Figure 2C&D).

Factors influencing follicle models

After performing stratified analysis by age, AFC, and AMH, only in Age<35, AFC≥5 and AMH≥1.2 group, Gn start dose of Rectangle model was higher than Polygon and Inverted-trapezoid model (202.12±61.92 vs. 176.5±52.5, 180.45±54.17, respectively; all P value <0.05). But in Age<35, AFC≥5, AMH<1.2 group, it was lower in Inverted-trapezoid model than in other models(269.21±39.61 vs. 280.25±32.54, 277.7±34.04, 281.19±30.58, respectively; all P value <0.05).

About Gn dose and Gn days, it was higher in the Polygon and Inverted-trapezoid model than Trapezoid and Rectangle model in most subgroups, in particular, the statistical differences were pronounced between Inverted-trapezoid and Trapezoid model, shown in Table 3.

Table 3 Factors influencing follicle models (Gn start dose/Gn total dose/Gn days).

	Stratified	Polygon model	Inverted-trapezoid model	Trapezoid model	Rectangle model
Gn start dose	Age<35; AFC≥5; AMH≥1.2	176.5±52.5^a	180.45±54.17^a	185.53±55.76	202.12±61.92^b
	Age≥35; AFC≥5; AMH≥1.2	261.42±47.3	266.39±43.54	267.98±45.41	277.1±39.28
	Age<35; AFC<5; AMH<1.2	293±21.45	294.28±18.72	296.71±13.62	294.92±17.26
	Age≥35; AFC<5; AMH<1.2	300±0	296.55±19.22	297.83±12.29	297.67±15.4
	Age<35; AFC<5; AMH≥1.2	284.56±36.42	272.57±49.93	272.28±51.72	285.85±28.8
	Age≥35; AFC<5; AMH≥1.2	300±0	295.89±21.09	297.32±13.36	296.43±16.53
	Age≥35; AFC≥5; AMH<1.2	293.64±20.54	290.58±25.85	291.31±24.64	296.79±13.65
	Age<35; AFC≥5;AMH<1.2	280.25±32.54^a	269.21±39.61^b	277.7±34.04^a	281.19±30.58^a
Gn dose	Age<35; AFC≥5; AMH≥1.2	1653.98±617.08	1699.43±608.76	1706.44±652.92	1829.48±628.85
	Age≥35; AFC≥5; AMH≥1.2	2325.33±582.07	2495.04±565.08^a	2337.92±554.08^b	2483.88±558.75
	Age<35; AFC<5; AMH<1.2	2759±598.74^a	2799.57±563.24^a	2572.33±656.28^b	2669.92±534.02^a
	Age≥35; AFC<5; AMH<1.2	2800±647.95^a	2842.27±561.62^a	2598.68±677.39^b	2705.92±557.99^a
	Age<35; AFC<5; AMH≥1.2	2697.06±553.66^a	2657.76±677.7^a	2393.04±725.38^b	2546.7±636.23
	Age≥35; AFC<5; AMH≥1.2	2893.75±683.07	2952.95±531.27^a	2712.28±642.94^b	2768.21±624.73^b
	Age≥35; AFC≥5; AMH<1.2	2700.46±500.43^a	2684.98±497.04	2525.25±518.79^b	2625.69±526.27
	Age<35; AFC≥5;AMH<1.2	2497.1±599.84	2491.99±526.32	2434.2±515.76	2416.91±529.72
Gn days	Age<35; AFC≥5; AMH≥1.2	8.83±1.41^b	9.06±1.39^a	8.55±1.45^b	8.68±1.5^b
	Age≥35; AFC≥5; AMH≥1.2	8.76±1.21^b	9.31±1.36^a	8.63±1.38^b	8.87±1.49^b
	Age<35; AFC<5; AMH<1.2	9.36±1.9	9.48±1.76^a	8.63±2.13^b	8.99±1.65
	Age≥35; AFC<5; AMH<1.2	9.32±2.14^a	9.55±1.8^a	8.7±2.22^b	9.07±1.76^a
	Age<35; AFC<5; AMH≥1.2	9.35±1.43^a	9.6±1.72^a	8.58±1.88^b	8.83±1.83
	Age≥35; AFC<5; AMH≥1.2	9.63±2.2	9.97±1.56^a	9.11±2.08^b	9.31±1.97^b
	Age≥35; AFC≥5; AMH<1.2	9.18±1.58^a	9.22±1.44	8.62±1.63^b	8.83±1.67
	Age<35; AFC≥5;AMH<1.2	8.81±1.73	9.13±1.28^a	8.61±1.48^b	8.51±1.5^b

*Different superscript letters show statistical differences of mean values within same parameter (column) (P<0.05).

*a vs b vs c, P value <0.05, respectively; a,b vs a, a,b vs b, same letter means no difference.

Starting dose, Gn adjustments and follicular models

During COS, different Gn starting doses and subsequent adjustments in FSH administration could have relationships with follicular models.

We studied patients who had undergone Fixed-dose protocols (N=6406), Increased-dose protocols(N=1453), and Decreased-dose protocols(N=546), we arranged them by Gn starting doses. A higher proportion of Decreased-dose protocols in Inverted-trapezoid was observed in Gn starting doses of <150IU and >225IU , that in Inverted-trapezoid model was higher than other models(<150IU: 17.55% vs. 10.91%vs. 6.52%, 7.44%; all P value<0.05; >225IU: 6.17% vs. 1.75%, 1.21%, 2.09%; all P value<0.05); that in Inverted-trapezoid model was higher than Trapezoid model in 150-225IU(14.62% vs. 4.74%; P<0.05). The proportion of patients showed a higher proportion of Increased-dose protocols in the Trapezoid model than in other models when starting dose<150 IU (44.02% vs. 33.85%, 31.31%, 38.84%, P<0.05) and a lower proportion of Increased-dose protocols in the Rectangle model than in other models when starting dose>225 IU (4.17% vs. 7.66%, 5.13%, 6.75%, P<0.05)(Table 4).

Table 4. Gn starting dose and Gn adjustment protocol between follicular models

Gn starting doses	Gn adjustment	Polygon model	Inverted-trapezoid model	Trapezoid model	Rectangle model
<150IU	Fixed-dose	248(55.23%)	405(51.14%)	372(49.47%)	65(53.72%)
	Decreased	49(10.91%)^a	139(17.55%)^b	49(6.52%)^c	9(7.44%)^a,c
	Increased	152(33.85%)^a	248(31.31%)^a	331(44.02%)^b	47(38.84%)^a,b
150-225IU	Fixed-dose	133(63.33%)	332(59.93%)	319(63.04%)	79(63.71%)
	Decreased	19(9.05%)^a,b	81(14.62%)^b	24(4.74%)^a	11(8.87%)^a,b
	Increased	58(27.62%)	141(25.45%)	163(32.21%)	34(27.42%)
>225IU	Fixed-dose	414(90.59%)^a,b	1695(88.7%)^b	1445(92.04%)^a	899(93.74%)^a
	Decreased	8(1.75%)^a	118(6.17%)^b	19(1.21%)^a	20(2.09%)^a
	Increased	35(7.66%)^a	98(5.13%)^a,b	106(6.75%)^a	40(4.17%)^b

*Different superscript letters show statistical differences of mean values within same parameter(column) (P<0.05).

*a vs b vs c, P value <0.05, respectively; a,b vs a, a,b vs b, same letter means no difference.

However, patients who received an adjusted Gn dose(whether Decreased or Increased-dose protocol) showed no difference in CLBR among models (Polygon vs Inverted-trapezoid vs Trapezoid vs Rectangle model: 47.07% vs. 49.21% vs. 47.69% vs. 42.42%, P>0.05); but when patients continued with the same starting dose in Fixed-dose protocol, the CLBR of the Polygon model was higher than that of other models (40.43% vs. 34.32, 31.13% vs. 26.46%, respectively; all P value <0.05)(Figure 2B).

This study is first report on this novel two-dimensional parameter model, which combined follicular number and size during COS monitoring. Our results showed a relationship between the follicle models and CLBR in the first IVF cycle. The higher chance of having a live birth was associated with the Polygon and Inverted-trapezoid models.

Before this study, number and size of follicles are two independent indicators of whether the oocyte is adequate and mature are used to predict the trigger time in routine clinical practice. Doctors monitor the follicular size and administer the trigger of oocyte maturation once the appropriate number and size of follicles are reached [9]. Triggering is administered when 2–3 ovarian follicles are at least 16–18 mm in diameter [23]. However, physicians often become confused when making clinical decisions using these two factors, so an explicit parameter that can not only predict ART results but also facilitate the selection of optimal COS treatment is needed.

For patients undergoing IVF, age is an important indicator of their ovarian reserve, which is indicated by the AFC and AMH; additionally, the AFC, AFC/age ratio, and AMH decrease gradually with age [24]. However, a recent study on the predictive effect of AMH showed a declining trend in the effect with increased age [25]. It is known that oocyte maturation progresses during follicular development [26]. Follicles that are either “too small” or “too large” are less likely to yield mature oocytes [27,28]. In our follicle models, the Polygon and Inverted-trapezoid models had the highest CLBR in comparison with the other models, and the proportion of mature oocytes in these models was higher than that in the other models. This is consistent with a study suggesting that the proportion of MII oocytes was significantly lower in small follicles (<13 mm) than in large follicles (≥16 mm), but with no difference in fertilization [29]. Previous studies have also reported lower oocyte recovery, reduced MII oocyte rates, and normal fertilization rates in small follicles [3,30,31]. This was also verified in our results. The highest number of viable embryos in the Polygon model may be due to the higher proportion of mature oocytes. This is supported by previous findings [29-31] that the proportion of mature oocytes is correlated with follicular size. In the Polygon model, moderate follicles (diameter 16–18 mm) accounted for the largest group, and there was less likelihood of receiving “post-mature” oocytes than in the Inverted-trapezoid model, which has 25% follicles that are too large (>20 mm). A related study reported that appropriately sized follicles on the day of triggering are most likely to yield mature oocytes [23]. Overall, follicles >22 mm in size on the day of triggering are more likely to contain post-mature oocytes [32]. Abbara et al. [23] found that follicles 12–19 mm in size on the day of triggering made a greater contribution to the number of embryos than larger follicles.

The quality of retrieved oocytes is strongly associated with IVF outcome. Wirleitner et al. [3] found a trend toward a higher LBR after transfer of blastocysts derived from small (8–12 mm) follicles compared with moderate (13–23 mm) or large ((≥ 24 mm) follicles (54.5%, 42.0%, and 41.7%, respectively). They also identified a trend toward a lower rate of blastocysts from small follicles than moderate or large follicles (15.3% vs. 27.9% vs. 26.1%, respectively, P<0.01). However, that study was limited to cycles in which patients underwent ICSI and lacked data for IVF. A recent study ²⁹ indicated that both the blastocyst formation rate and the pregnancy rate did not differ according to follicular size. These studies were focused on the outcome of only one blastocyst transfer cycle. It should be noted that only 30% of cleavage-stage embryos can develop to the blastocyst stage [33], and developmental potential cleavage-stage embryos might be lost. This was not confirmed by our data. According to our data, the CLBR in the Polygon model was higher than that in other models, suggesting the potential of moderately sized follicles (16–18 mm) is good enough for patients in relatively good condition.

Ovarian stimulation with exogenous Gn is a key procedure in IVF, but standardizing an effective and safe ovarian stimulation dosage is difficult owing to the substantial heterogeneity among patients [35]. The starting dose of Gn is administered on the basis of the woman’s age, AFC, BMI, and ovarian response to stimulation [34], and the influencing factors are complex during the process of stimulation. Our data showed a relationship between Gn adjustment and the follicle models. For patients whose protocols were adjusted, the higher proportion of the Decreased-dose protocol in the Inverted-trapezoid and Polygon model showed better ovarian response than that of the Trapezoid and Rectangle model. And the higher proportion of the Increased-dose protocol in the Trapezoid and Rectangle model showed poorer ovarian response than that of the Inverted-trapezoid and Polygon model. After Gn adjustment indicated that it benefitted patients associated with all models, especially when the patient's condition was not as good as in the Polygon model.

Strengths and limitations

The strengths of this study included it was the first time to propose the concept of two-dimensional parameter graphical follicle models, providing innovative, simple, and practiced parameters for predicting clinical results in ovary stimulation treatment, and facilitating personalized protocol adjustment to support visual follicular analysis. The limitations of this study include its retrospective nature. Because we analyzed real-world data, the follicular corresponding outcome following measurement was not obtained. Besides, the majority of patients received 2 cleavage stage embryos transfer, this approach may impact the external validity of follicle model. A well-designed, multicenter study is still warranted to further support our results.

Our results suggested the follicle models can demonstrate characteristics and the ovarian response of patients. In addition to the fact that patients in very good condition who develop a Polygon model, it is better for other patients to reach the Inverted-trapezoid model with follicles >18 mm and <20 mm. Adjusting the protocol is critical to the outcome. But determining how to improve follicle models by regulating stimulation remains to be further studied.

ART: assistant reproductive technology

HCG: human chorionic gonadotropin

CMH: Cochran-Mantel-Haenszel

GLM: Generalized Linear Model

CLBR: cumulative live birth rate

OHSS: ovarian hyperstimulation syndrome

COS: controlled ovarian stimulation

IVF: in vitro fertilization

GnRH-ant: gonadotropin releasing hormone antagonist

EMRS: Electronic Medical Record System

ICSI: Intracytoplasmic Sperm Injection

PGT: preimplantation genetic testing

AFC: antral follicle count

BMI: body mass index

FSH: follicle stimulating hormone

FET: frozen embryo transfer

P: Polygon model

IT: Inverted-trapezoid model

T: Trapezoid model

R: Rectangle model

Acknowledgements

We would like to thank all the embryologists and nurses of Chongqing Reproductive and Genetics Institute for embryo culture, data recording, and other clinical assistance.

Authors' contributions

The present work was designed by XZ and GH. Data extraction and analysis were performed by JY, XL and HY. JY, XL, and FL participated in the data collection. GH, HY and LZ made revisions to the article. All authors have read and approved the final manuscript.

Funding

This study was supported by the Social Guarantee and Technical Innovation Foundation of Chongqing Province, China (cstc2017shms-zdyfx0020; cstc2016shmszx130026)

Availability of data and materials

The data underlying this article will be shared on reasonable request to the corresponding author.

Ethics approval

This study was reviewed and approved by the Ethics Committee Review Board of Chongqing Health Center for Women and Children (approval number: 2020-RGI-13) for retrospective analysis and clinical data reporting. Informed consent was waived by the committee because of the retrospective nature of the study.

Competing interests

The author(s) declare that they have no potential conficts of interest.

Sunkara SK, Rittenberg V, Raine-Fenning N, Bhattacharya S, Zamora J, Coomarasamy A. Association between the number of eggs and live birth in IVF treatment: an analysis of 400 135 treatment cycles. Hum Reprod. 2011; 26(7):1768-74.
Ben-Haroush, A, Sapir, O, Salman, L,Altman, E, Garor, R, Margalit, T, Shufaro, Y, Oron, G. Does 'Dual Trigger' Increase Oocyte Maturation Rate? J OBSTET GYNAECOL. 2019; 40 (6): 860-862.
Simonetti S, Veeck LL, Jones HW Jr. Correlation of follicular fluid volume with oocyte morphology from follicles stimulated by human menopausal gonadotropin. Fertil Steril. 1985 Aug;44(2):177-80.
Wirleitner B, Okhowat J, Vištejnová L, Králíčková M, Karlíková M, Vanderzwalmen P, Ectors F, Hradecký L, Schuff M, Murtinger M. Relationship between follicular volume and oocyte competence, blastocyst development and live-birth rate: optimal follicle size for oocyte retrieval. Ultrasound Obstet Gynecol. 2018 Jan;51(1):118-125.
Timeva T, Milachich T, Antonova I, Arabaji T, Shterev A, Omar HA. Correlation between number of retrieved oocytes and pregnancy rate after in vitro fertilization/intracytoplasmic sperm infection. ScientificWorldJournal. 2006 Jun 21;6:686-90.
Magnusson A, Kallen K, Thurin-Kjellberg A, Bergh C. The number of oocytes retrieved during IVF: a balance between efficacy and safety. Human reproduction. 2018; 33(1):58-64.
Mehri S, Levi Setti PE, Greco K, Sakkas D, Martinez G, Patrizio P. Correlation between follicular diameters and flushing versus no flushing on oocyte maturity, fertilization rate and embryo quality. J Assist Reprod Genet. 2014 Jan;31(1):73-7.
Ozgur K, Bulut H, Berkkanoglu M, Donmez L, Coetzee K. Prediction of live birth and cumulative live birth rates in freeze-all-IVF treatment of a general population. J Assist Reprod Genet. 2019 Apr;36(4):685-696.
Robertson I, Chmiel FP, Cheong Y. Streamlining follicular monitoring during controlled ovarian stimulation: a data-driven approach to efficient IVF care in the new era of social distancing. Hum Reprod. 2021 Jan 1;36(1):99-106.
A.L. Nivet,1 M.C. Léveillé,2 A. Leader,2 and M.A. Sirard.Transcriptional characteristics of different sized follicles in relation to embryo transferability: potential role of hepatocyte growth factor signalling. Mol Hum Reprod. 2016; 22 (7): 475-84.
Carpintero NL, Suarez OA, Mangas CC, Varea CG, Rioja RG. Follicular steroid hormones as markers of oocyte quality and oocyte development potential. Journal of human reproductive sciences. 2014; 7 (3): 187-93.
Hu X, Luo Y, Huang K, Li Y, Xu Y, Zhou C, Mai Q. New Perspectives on Criteria for the Determination of HCG Trigger Timing in GnRH Antagonist Cycles. Medicine (Baltimore). 2016 May;95(20):e3691.
Irani M, Canon C, Robles A, Maddy B, Gunnala V, Qin X, Zhang C, Xu K, Rosenwaks Z. No effect of ovarian stimulation and oocyte yield on euploidy and live birth rates: an analysis of 12 298 trophectoderm biopsies. Hum Reprod. 2020 May;35(5):1082-1089.
Maheshwari A, McLernon D, Bhattacharya S. Cumulative live birth rate: time for a consensus? Hum Reprod. 2015 Dec;30(12):2703-7.
De Neubourg D, Bogaerts K, Blockeel C, Coetsier T, Delvigne A, Devreker F, Dubois M, Gillain N, Gordts S, Wyns C. How do cumulative live birth rates and cumulative multiple live birth rates over complete courses of assisted reproductive technology treatment per woman compare among registries? Hum Reprod. 2016 Jan;31(1):93-9.
Luke B, Brown MB, Wantman E, Lederman A, Gibbons W, Schattman GL, et al. Cumulative birth rates with linked assisted reproductive technology cycles. The New England journal of medicine. 2012; 67 (11): 711-712.
van Hoogenhuijze NE, Mol F, Laven JSE, Groenewoud ER, Traas MAF, Janssen CAH, Teklenburg G, de Bruin JP, van Oppenraaij RHF, Maas JWM, Moll E, Fleischer K, van Hooff MHA, de Koning CH, Cantineau AEP, Lambalk CB, Verberg M, van Heusden AM, Manger AP, van Rumste MME, van der Voet LF, Pieterse QD, Visser J, Brinkhuis EA, den Hartog JE, Glas MW, Klijn NF, van der Meer S, Bandell ML, Boxmeer JC, van Disseldorp J, Smeenk J, van Wely M, Eijkemans MJC, Torrance HL, Broekmans FJM. Endometrial scratching in women with one failed IVF/ICSI cycle-outcomes of a randomised controlled trial (SCRaTCH). Hum Reprod. 2021 Jan;36(1):87-98.
Yang J, Zhang X, Ding X, Wang Y, Huang G, Ye H. Cumulative live birth rates between GnRH-agonist long and GnRH-antagonist protocol in one ART cycle when all embryos transferred: real-word data of 18,853 women from China. Reprod Biol Endocrinol. 2021 Aug 12;19(1):124.
Expert Consensus Compilation Group of Reproductive Medicine Committee of China Medical Women's Association.Expert consensus on standardized application of antagonist protocol in assisted reproductive technology. Chinese Journal of Reproduction and Contraception. 2022; 42:109-116. DOI:10.3760/cma.j.cn101441-20211108-00495.
Rosen MP, Shen S, Dobson AT, Rinaudo PF, McCulloch CE, Cedars MI. A quantitative assessment of follicle size on oocyte developmental competence. Fertil Steril. 2008 Sep;90(3):684-90.
Sun Y J, Huang GN, Sun HX, Fan LQ, Shen HH, Liu P, Lu WH, Zhang YS, Sun YP, Feng Y . Chinese expert consensus on numbers of embryos transferred. Journal of Reproductive Medicine. 2018; 27(10):940-945.
Ferraretti AP, La Marca A, Fauser BC, Tarlatzis, B, Nargund, G, Gianaroli, L. ESHRE consensus on the definition of 'poor response' to ovarian stimulation for in vitro fertilization: the Bologna criteria. Hum Reprod. 2011, 26 (7): 1616-24.
Abbara A, Vuong LN, Ho VNA, Clarke SA, Jeffers L, Comninos AN, Salim R, Ho TM, Kelsey TW, Trew GH, Humaidan P, Dhillo WS. Follicle Size on Day of Trigger Most Likely to Yield a Mature Oocyte. Front Endocrinol (Lausanne). 2018; 9:193. DOI: 10.3389/fendo.2018.00193.
Zhou SJ, Zhao MJ, Li C, Su X. The comparison of evaluative effectiveness between antral follicle count/age ratio and ovarian response prediction index for the ovarian reserve and response functions in infertile women. Medicine (Baltimore). 2020; 99(36):e21979.
Depmann M, Eijkemans MJC, Broer SL, Tehrani FR, Solaymani-Dodaran M, Azizi F, Lambalk CB, Randolph JF Jr, Harlow SD, Freeman EW, Sammel MD, Verschuren WMM, van der Schouw YT, Mol BW, Broekmans FJM. Does AMH relate to timing of menopause? Results of an Individual Patient Data meta-analysis. J Clin Endocrinol Metab. 2018 Jul 18. DOI: 10.1210/jc.2018-00724. Epub ahead of print. PMID: 30032277.
Alam MH, Miyano T. Interaction between growing oocytes and granulosa cells in vitro. Reprod Med Biol. 2019 Aug 22;19(1):13-23.
Re C, Mignini Renzini M, Rodriguez A, Dal Canto M, Buccheri M, Sacchi S, Bartolucci S, Fadini R, La Marca A. From a circle to a sphere: the ultrasound imaging of ovarian follicle with 2D and 3D technology. Gynecol Endocrinol. 2019 Mar;35(3):184-189.
Mohr-Sasson A, Orvieto R, Blumenfeld S, Axelrod M, Mor-Hadar D, Grin L, Aizer, A, Haas, J. The association between follicle size and oocyte development as a function of final follicular maturation triggering. Reproductive biomedicine online. 2020; 40(6): 887-893.
Tamura I, Kawamoto-Jozaki M, Fujimura T, DOI-Tanaka Y, Takagi H, Shirafuta Y, Mihara Y, Taketani T, Tamura H, Sugino N. Relationship between follicular size and developmental capacity of oocytes under controlled ovarian hyperstimulation in assisted reproductive technologies. Reprod Med Biol. 2021; 20(3):299-304.
Shah N, Romanski PA, Bortoletto P, Rosenwaks Z, Kligman I. Determining the optimal follicle size at trigger in patients undergoing ovarian stimulation with a clomiphene citrate plus gonadotropin and GnRH antagonist protocol for in-vitro fertilization. Minerva Obstet Gynecol. 2022; 74(1):75-82.
Nogueira D, Friedler S, Schachter M, Raziel A, Ron-El R, Smitz J. Oocyte maturity and preimplantation development in relation to follicle diameter in gonadotropin-releasing hormone agonist or antagonist treatments. Fertility and sterility. 2006; 85(3): 578-83.
Revelli A, Martiny G, Delle Piane L, Benedetto C, Rinaudo P, Tur-Kaspa I. A critical review of bi-dimensional and three-dimensional ultrasound techniques to monitor follicle growth: do they help improving IVF outcome? Reprod Biol Endocrinol. 2014; 12:107. DOI: 10.1186/1477-7827-12-107. PMID: 25420733; PMCID: PMC4255967.
Dumoulin, Cm J, Coonen, E, Bras, M, van Wissen, L C, Ignoul-Vanvuchelen, R,; Bergers-Jansen, J M, Derhaag, J G, Geraedts, J P, Evers, J L. Comparison of in-vitro development of embryos originating from either conventional in-vitro fertilization or intracytoplasmic sperm injection. Human Reproduction. 2000; 15 (2): 402-9.
Deng L, Li XL, Ye DS, Blockeel C, Zhou XY, Chen SL, Chen X. A Second Dose of GnRHa in Combination with Luteal GnRH Antagonist May Eliminate Ovarian Hyperstimulation Syndrome in Women with ≥30 Follicles Measuring ≥11 mm in Diameter on Trigger Day and/or Pre-trigger Peak Estradiol Exceeding 10 000 pg/mL. Curr Med Sci. 2019; 39(2):278-284.
Bissonnette F, Minano Masip J, Kadoch IJ, Librach C, Sampalis J, Yuzpe A. Individualized ovarian stimulation for in vitro fertilization: a multicenter, open label, exploratory study with a mixed protocol of follitropin delta and highly purified human menopausal gonadotropin. Fertil Steril. 2021 Apr;115(4):991-1000.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Relation of graphical follicle models to the cumulative live birth rate in GnRH-antagonist stimulation treatment cycles

Status:

Version 1

Abstract

Figures

Background

Methods

Results

Discussion

Conclusion

Abbreviations

Declarations

References

Additional Declarations

Status:

Version 1