Determination of the optimal follicle size at the time of trigger in patients with diminished ovarian reserve undergoing ICSI cycle

Abstract

Background

Patients with diminished ovarian reserve (DOR) typically have fewer retrieved oocytes and lower pregnancy rates than normal responders. Therefore, maximizing oocyte yield is particularly important to improve their reproductive outcomes.

Aim

The purpose of this study was to improve reproductive outcomes in assisted reproductive technology (ART) by determining the optimal follicle size to induce final oocyte maturation in patients with DOR.

Methods

A total of 430 follicles from 98 DOR patients were included in this study. Controlled ovarian stimulation was conducted by a GnRH antagonist protocol, with ovulation triggered when the leading follicle reached ≥ 17 mm, followed by oocyte retrieval 36 h later. On the day of oocyte retrieval, in the operating theatre, all follicles were carefully measured, one by one, by a second trained individual just before aspiration. Then, each follicle was aspirated separately and collected in a separate tube. Throughout the laboratory process, oocytes were tracked and managed based on the size of the follicle from which they originated. Based on follicle size, 11–14 mm, 15–17 mm, and ≥ 18 mm, the follicles were divided into three groups. Reproductive outcomes in this study were included Cumulus Oocyte Complex (COC) retrieval rate, metaphase II (MII) oocyte rate, fertilization rate, and top-quality embryo (TQE) rate.

Results

The final analysis included a total of 430 follicles were measured and aspirated. Follicles measuring 15–17 mm exhibited a significantly higher COC retrieval rate (88.1%), MII oocyte rate (83.3%), fertilization rate (97.1%), and TQE rate (65.4%), in comparison to follicles measuring 11–14 mm and ≥ 18 mm (p < 0.05). The COC retrieval rate and the MII retrieval rate were 79.16% and 66.66%, for follicles measuring 14 mm, and were 45.09% and 33.33%, for follicles measuring 13 mm, respectively (p < 0.05). ROC analysis identified an optimal follicle size of 14.5–15.5 mm for COC retrieval. The incidence of premature ovulation was 13.27% and the mean size of the leading follicle on the day of trigger was 18.6 mm in patients with premature ovulation.

Conclusion

Our study suggests that oocyte retrieval, maturation and embryo development are optimized by triggering final oocyte maturation at a follicle size of 15–17 mm. In addition, there is a need to reconsider the standard trigger criteria in these patients.

Background

Diminished Ovarian Reserve (DOR) is generally characterized by a reduced number and declined quality of oocytes, which can lead to a poor response in vitro fertilization (IVF) cycles to ovarian stimulation [1]. Patients with DOR often represent a challenge for assisted reproductive technology (ART) due to the low number of oocytes and premature ovulation. Therefore, there is a higher probability of cycle cancellation, and the pregnancy rate is lower compared to those of normal responders [2, 3]. In IVF, multiple follicles of different sizes are allowed to grow during a controlled ovarian stimulation cycle. Therefore, follicle growth monitoring is critical to obtain a high number of mature oocytes [4]. The relationship between oocyte maturation and follicle size is well known [4]. Determining the optimal follicle size for triggering is crucial because the administration of HCG for triggering causes LH surge, leading to the resumption of meiosis and final oocyte maturation [5].

While several studies demonstrated that follicle size is correlated with IVF outcome in normal responders, it remains uncertain in patients with DOR [6]. Early triggering results in an immature oocyte, whereas a delayed triggering increases the post mature oocyte and premature ovulation [7]. The European Society of Human Reproduction and Embryology (ESHRE) concludes that a lead follicle size of 16–22 mm is appropriate to trigger final oocyte maturation in normal responders, but this threshold is still controversial for DOR patients [8].

In patients with DOR, decreased antral follicle numbers, accelerated follicular recruitment, and higher levels of early follicular phase FSH result in faster follicular development. This leads to early ovulation and shorter follicular phases and menstrual cycles [9]. Therefore, in patients with DOR who have fewer follicles available for oocyte retrieval, the relationship between follicle size and oocyte maturation is critical for optimizing reproductive outcomes [10]. Consequently, it has been proposed that women with normal ovarian reserve are basically different from poor responder women, therefore cannot be assessed in the same way during stimulation of the ovaries [11]. It appears a small follicle size should be used to induce final oocyte maturation in these patients [12].

Currently, the POSEIDON (Patient-Oriented Strategies Encompassing Individualized Oocyte Number) grouping system has been introduced for improvement of pregnancy rates in poor responders [13]. Group 1 and 2 patients have sufficient parameters of the ovarian reserve, antral follicular count (AFC) >5 and anti-mullerian hormone (AMH) >1.2 ng/mL, but with an unexpectedly poor or suboptimal response from the ovaries. Group 3 was defined as patients < 35 years of age with AFC < 5 and AMH < 1.2 ng/mL and group 4 patients ≥ 35 years of age with AFC < 5 and AMH < 1.2 ng/mL [14].

In this study, we set out to determine the optimal leading follicle size to induce final oocyte maturation in DOR patients.

Methods

The present single-center cross-sectional study was conducted between September 2023 and December 2024 at the Shahid Beheshti infertility clinic in Isfahan, Iran. A total of 116 patients aged between 22 and 44 years with DOR undergoing Intracytoplasmic Sperm Injection (ICSI)/IVF treatment based on the POSEIDON criteria were included. Patients were in groups 3 and 4 poor responders.

Exclusion criteria included (i) having endocrine disorders: hyperprolactinemia and hypothyroidism or hyperthyroidism, (ii) endometriosis, (iii) male or female chromosomal abnormalities, (iv) Genetic disorder, (v) body mass index more than 29 kg/m2, (vi) azoospermia in male partner.

On day 2 of the cycle, all patients underwent vaginal ultrasound examination to roll out the presence of follicle size more than 10 mm in size. Baseline hormonal profiles were also part of the evaluation. Controlled ovarian induction was performed using an antagonist protocol, with the administration of gonadotropins 300 IU/day of recombinant FSH (Gonal F, Merck, Germany) and/or human menopausal gonadotropin (HMG; Menopur, Merck, Germany) starting on the second day of the cycle. The GnRH antagonist Cetrorelix 0.25 mg/day (Cetrotide, Merck, Germany) was started in a fixed antagonist protocol on day 6 of stimulation. Vaginal ultrasound was used to measure follicles, and the size of follicles was calculated as the mean of the two largest diameters. Administration of 0.2 mg of GnRH agonist (Decapeptyl, Ipsen) and 5,000 IU HCG (Pregnyl; Organon, Vienna, Austria) in a double trigger form was used when at least one follicle had reached 17 mm for ovulation. Hormonal tests including estradiol (E2), luteinizing hormone (LH), and progesterone (P) were assessed on the day of the trigger. Thirty-six hours after trigger administration, ovum pickup (OPU) was performed. Ultrasonography was conducted prior to surgery and OPU has been cancelled due to premature ovulation. Premature ovulation was defined as at least one of the leading follicles ruptured on the day of egg retrieval. In the operating theatre, following the induction of general anesthesia, a transvaginal ultrasound was performed to evaluate the ovaries and assess the number of follicles. All follicles were carefully measured, one by one, by a second trained individual. The size of follicles was calculated using the mean of the largest vertical and horizontal diameters. A 17-gauge double-lumen aspiration needle was then inserted into each follicle. Follicular fluid was aspirated under a pressure of 110 mmHg and collected in a separate tube. The aspirated fluid was immediately transferred to the embryology laboratory to examine to identify the presence of a cumulus-oocyte complex (COC). If a COC was recovered, the next follicle was similarly measured and aspirated into a new tube, if the oocyte was not retrieved with the native follicular fluid, the follicle was flushed with media (Modified Ham’s F-10; Irvine Scientific, Santa Ana, CA, USA). In cases where no oocyte was obtained after flushing, the aspiration system was rinsed with media to recover any oocyte that might have been retained within the lumen of the needle or tubing. This procedure was repeated for all accessible follicles. Each retrieved COC was placed in a pre-labeled 60 × 15 mm culture dish, with each droplet corresponding to the original follicular size. The incubation was done for COCs, 2–3 h before denudation. After removing the cumulus cells, the maturation status of each oocyte was assessed and classified as germinal vesicle (GV), metaphase I (MI), or metaphase II (MII).

Mature (MII) oocytes were injected via ICSI and placed in individually labeled culture droplets, to preserve the identity of each follicle. Fertilization was assessed 16 to 20 h post-injection, with normal fertilization defined by the presence of two polar bodies and two pronuclei (2PN) [15]. Each embryo was cultured and evaluated after 72 h. A top-quality embryo (TQE) was defined as an 8-cell embryo with blastomeres of uniform size and no more than 5% fragmentation [16]. Development of each embryo was carefully assessed to correlate its quality with the size of its originating follicle.

On the basis of follicle size, the follicles were separated into three groups; Group 1: mean diameter of 11–14 mm, Group 2: mean diameter of 15–17 mm, and Group 3: mean diameter of ≥ 18 mm. This classification was based on the statistical distribution of the data to enhance the power of the analysis.

The reproductive outcomes of this study included the COC retrieval rate, the MII oocyte rate, the rate of fertilization, and the TQE rate. The rate of COC retrieval was calculated as the number of COCs retrieved to the number of follicles aspirated. The rate of MII oocyte is the result of the number of MII oocytes obtained per aspirated follicles, while the ratio of MII oocyte was determined as the number of MII oocytes to the total number of obtained COCs. The fertilization rate was determined by 2PN fertilized oocytes to the number of MII oocytes. Finally, the rate of TQE is the result of the number of TQE obtained from the total number of 2PN fertilized oocytes. This research was approved by the Ethics Committee (NO. IR.MUI.REC.1403.043). This study was designed as a prospective observational cohort tracking outcomes from standard procedures without therapeutic intervention or randomization; therefore, it was not registered as a clinical trial.

Statistical analysis

Statistical analysis was done using the Statistical Package for the Social Sciences (SPSS version 28, Chicago, IL, USA). For assess the normality of the data distribution, the Kolmogorov–Smirnov test was used. Mean ± standard deviation (SD), and the median and the interquartile range were applied for parametric variables and nonparametric data, respectively. Chi-square test (χ²) was used for comparisons of categorical variables where appropriate. Logistic regression analysis was performed to assess the relationship between follicle size and reproductive outcomes, adjusting for potential confounders. An ROC curve was created to predict COC retrieval. A p-value < 0.05 was considered statistically significant.

Results

Of the 116 patients who were included during the study period, 18 subjects were excluded from the study due to inadequate response to stimulation in 3 patients and premature ovulation in 15 patients. In the final analysis, 98 patients were part of the study and a total of 430 follicles were available for measurement and aspiration. The basic characteristics of 98 patients who were retrieved are shown in Table 1. The average age of the patients was 36.03 ± 5.16 years, ranging from 23 to 43, the mean body mass index (BMI) was 22.45 ± 1.59, the mean AMH level was 0.73 ± 0.28. The mean levels of LH, E2 and P on the trigger day were 4.81 ± 2.61, 922.11 ± 401.65, and 0.68 ± 0.32, respectively. Other variables are summarized in the Table 1.

Table 1.

Baseline characteristics of the study population (N = 98 patients)

Variables	Mean ± SD
Age (years)	36.03 ± 5.16
BMI (kg/m²)	22.45 ± 1.59
Duration of infertility (years)	6.21 ± 3.90
AMH (ng/mL)	0.73 ± 0.28
Basal FSH (mIU/mL)	8.10 ± 2.59
Basal LH (mIU/mL)	5.50 ± 1.48
Basal Estradiol (pmol/L)	31.51 ± 3.61
Basal Progesterone (ng/mL)	0.57 ± 0.15
LH on trigger day (mIU/mL)	4.81 ± 2.61
Estradiol on trigger day (pmol/L)	922.11 ± 401.65
Progesterone on trigger day (ng/mL)	0.68 ± 0.32
Total gonadotropin dose (IU)	3042 ± 559.04
Duration of stimulation (days)	8.17 ± 1.36

The number of aspirated follicles in each group were 155 (36.04%) 11–14 mm (group 1), 168 (39.06%) 15–17 mm (group 2), and 107 (24.88%) ≥ 18 mm (group 3).

The COC retrieval rate was significantly higher in the group 2 (88.1%) in comparison with group 1 (45.8%) and group 3 (66.4%), while the probability of retrieving MII oocytes was lower in group 1 (33.5%) and group 3 (58.9%) compared to group 2 (83.3%). Moreover, the likelihood of retrieving an MII oocyte among patients with COC in the group 2 (94.6%) was more than the group 1 (73.2%) and the group 3 (88.7%). Furthermore, fertilization rate among those who had a MII oocyte, were significantly higher in the group 2 (97.1%) compared with the group 1 (59.6%) and the group 3 (57.1%).

In addition, group 2 (97.1%) compared to group 1 (59.6%) and group 3 (57.1%) had a significantly higher fertilization rate among those with an MII oocyte (Table 2).

Table 2.

Reproductive outcomes based on follicle size.

Outcomes	11–14 mm	15–17 mm	≥ 18 mm	P-value
COC retrieval rate, %	71/155 (45.8%)	148/168 (88.1%)	71/107 (66.4%)	0.001
MII oocyte rate, %	52/155 (33.5%)	140/168 (83.3%)	63/107 (58.9%)	0.001
MII/COC ratio, %	52/71 (73.2%)	140/148 (94.6%)	63/71 (88.7%)	0.001
Fertilization rate, %	31/52 (59.60%)	136/140 (97.1%)	36/63 (57.1%)	0.001
TQE rate, %	13/31 (38.7%)	89/136 (65.4%)	21/36 (58.3%)	0.001

In the adjusted multivariable analysis, these statically significant differences were also observed, as shown in table (Table 3)

Table 3.

Multivariable analysis of stimulation outcomes based on follicle size

Outcomes	Follicle size	OR, 95% CI, p-value (crude)	Adjusted OR, 95% CI, p-value (multivariable)
COC	11–14 mm	-	-
	15–17 mm	8.75 (4.98–15.38, P = 0.000)	8.51 (4.79–15.11, P = 0.000)
	≥ 18 mm	2.33 (1.40–3.88, P = 0.001)	2.24 (1.33–3.79, P = 0.002)
MII	11–14 mm	-	-
	15–17 mm	9.90 (5.85–16.74, P = 0.000)	10.02(5.82–17.25, P = 0.000)
	≥ 18 mm	2.83 (1.70–4.72, P = 0.000)	2.69 (1.59–4.54, P = 0.000)
2PN	11–14 mm	-	-
	15–17 mm	14.25 (6.69–30.35, P = 0.000)	15.66 (7.13–34.40, P = 0.000)
	≥ 18 mm	1.29 (0.66–2.51, P = 0.446)	1.30 (0.66–2.58, P = 0.444)
TQE	11–14 mm	-	-
	15–17 mm	6.84 (3.45–13.57, P = 0.000)	6.47 (3.31–13.72, P = 0.000)
	≥ 18 mm	1.90 (0.86–4.19, P = 0.108)	1.87 (0.83–4.21, P = 0.128)

Ref group was: 11–14

In addition, an analysis was conducted on Group 2 to determine the COC retrieval rate, MII oocyte rate, fertilization rate and TQE rate for follicles measuring 15, 16, and 17 mm. The findings are summarized in Table 4. The findings revealed no statistically significant differences within this group.

Table 4.

Reproductive outcomes in follicles 15–17 mm in size (group 2)

Outcomes	15 mm	16 mm	17 mm	P-value
COC retrieved rate, %	53/56 (94.6%)	49/55 (89.1%)	46/57 (80.7%)	0.07
MII retrieved rate, %	48/56 (85.7%)	46/55 (83.6%)	46/57 (80.7%)	0.77
MII/COC ratio, %	48/53 (90.56%)	46/49 (93.8%)	46/46 (100%)	0.82
Fertilization rate, %	46/48 (95.83%)	44/46 (95.65%)	46/46 (100%)	0.72
TQE rate, %	30/46 (65.21%)	31/44 (70.45%)	28/46 (60.86%)	0.86

Furthermore, follicles 11–14 mm in diameter were analyzed separately. MII oocyte rate were 4.54% for follicles 11 mm in size and were 5.88%, for follicles 12 mm in size (P > 0.05), but none of the oocytes retrieved from follicles resulted in successful fertilization or TQE. In addition, the COC retrieval rate and MII oocyte rate were 79.16% and 66.66%, for follicles 14 mm in size and were 45.09% and 33.33%, respectively, for follicles 13 mm in size (p ≤ 0.05), but the rate of fertilization and rate of TQE were not significant.

We conducted subgroup analysis to compare outcomes between POSEIDON group 3 (less than 35 years) and POSEIDON group 4 (35 years and older). There were no statistically significant differences in subgroup analysis of POSEIDON 3 and 4 based on the age groups. we found that optimum follicle size for good outcomes (15–17 mm) was similar in both groups. More details are shown in Table 5.

Table 5.

Comparison between POSEIDON group 3 and 4

Outcomes	POSEIDON group III (age < 35)	POSEIDON group IV (age ≥ 35)	P-value
Number of patients	35(35.72%)	63(64.28%)
Number of follicles	164(38.14%)	266(61.86%)
COC retrieval rate, %	113/164(68.9%)	218/266(81.9%)	0.61
MII oocyte rate, %	99/164(60.3%)	156/266(58.6%)	0.72
Fertilization rate, %	79/99(79.7%)	124/156(79.4%)	0.75
TQE rate, %	49/79(62%)	78/124(62.9%)	0.91

The area under the curve (AUC) of the ROC curve was 0.647 (95%CI: 0.580–0.713, p < 0.001). The optimal cut-off point for follicle size was found to be 14.5 mm, giving a sensitivity of 75.5% and a specificity of 60.0%. Alternatively, a cut-off of 15.5 mm resulted in a sensitivity of 57.2% and a specificity of 62.1%. The ROC curve showed that an optimal balance between sensitivity and specificity was observed for follicles 14.5–15.5 mm in size. These results suggest that follicle size can be a moderate predictor of COC retrieval (Fig. 1).

Fig. 1.

The ROC curve analysis results for predicting retrieval of COC

Among those who were scheduled for oocyte retrieval, 15 (13.27%) had premature ovulation. More details about the comparisons between the patients with premature ovulation and the controls are shown in Table 6. In the premature ovulation group, serum FSH levels were significantly higher (14.30 ± 1.98 vs. 7.10 ± 2.59), while AMH was significantly lower compared to control group (0.43 ± 0.20 vs. 0.93 ± 0.28). The mean serum progesterone and LH levels on the day of trigger were significantly higher in the premature ovulation group compared to control subjects (Table 6). The mean size of the leading follicle on the day of trigger was 18.6 mm in premature ovulation group.

Table 6.

Comparison of baseline parameters between premature ovulation and control groups

Variables	Control group N = 98	Premature ovulation N = 15	p-value
Age (years)	36.03 ± 5.16	38.07 ± 5.04	0.26
BMI (kg/m2)	22.45 ± 1.59	22.65 ± 1.53	0.78
Duration of Infertility (years)	6.21 ± 3.90	6.72 ± 1.70	0.93
AMH (ng/ml)	0.93 ± 0.28	0.43 ± 0.20	< 0.05
Basal FSH (mIU/mL)	7.10 ± 2.59	14.30 ± 1.98	< 0.05
Basal LH (mIU/mL)	5.50 ± 1.48	7.22 ± 2.20	0.89
Basal Estradiol (pmol/L)	31.51 ± 3.61	33.60 ± 6.65	0.64
Basal Progesterone (ng/mL)	0.57 ± 0.15	0.68 ± 0.16	0.77
LH on trigger day (mIU/mL)	4.81 ± 2.61	20.27 ± 2.93	< 0.05
Progesterone on trigger day (ng/mL)	0.68 ± 0.32	1.32 ± 0.28	< 0.05
Estradiol on trigger day (pmol/L)	922.11 ± 401.65	861.93 ± 282.73	0.60
Duration of Stimulation (day)	8.17 ± 1.36	7.47 ± 0.64	0.79
Total gonadotropin dose (IU)	3042.73 ± 559.04	2941.65 ± 239.98	0.18
GnRh Antagonist duration (day)	2.76 ± 0.90	1.93 ± 0.70	0.69

Discussion

DOR patients typically have fewer retrieved oocytes and lower pregnancy rates than normal responders. Therefore, maximizing oocyte yield is particularly important to improve their reproductive outcomes [3]. On the other hand, one of the main concerns in these patients is premature ovulation, which can lead to cycle cancelation. Consequently, determining the optimal leading follicle size for triggering final oocyte maturation in DOR patients is critical to maximize oocyte yield and prevent premature ovulation. Therefore, we conducted this study to determine the optimal leading follicle size for triggering final oocyte maturation in these patients based on POSEIDON groups 3 and 4 poor responders.

Our study shows that the probability of rate of COC retrieval, MII oocyte, fertilization, and TQE were significantly higher in follicle size 15–17 mm in DOR patients. Also, follicles 13 and 14 mm in diameter were significant for the COC retrieval rate and the MII oocyte rate, but not for the fertilization rate and the TQE rate. In addition, we demonstrated that among all aspirated follicles, the highest sensitivity and specificity for COC were observed at 14.4–15.5 mm by ROC curve analysis.

In ART cycles, accurate measurement of follicles and the timing of triggering for maturation of final oocyte is crucial to obtain the maximum number of mature oocytes and TQE [17]. The current protocol is that when 2 or 3 lead follicles have reached a diameter of 17–18 mm, final oocyte maturation is induced in normal responders [18]. In a normal responder population, the ESHRE guidelines recommend that a lead follicle size of 16–22 mm be used for triggering [19].

Several studies have found that size of follicle is related to oocyte maturity. Mehri et al. have shown that follicles larger than 18 mm at the time of retrieval predominantly contain a greater number of mature oocytes and have a higher rate of fertilization [20]. Another research by Rosen et al. showed that in an unselected IVF population, follicles 16–18 mm had significantly lower mature oocytes rates and fertilization than follicles larger than 18 mm [21]. Other studies have indicated that MII oocytes are more likely to be found in medium and large follicle groups, suggesting that follicles larger than 15 mm are most likely to contain mature oocytes [4, 22, 23]. It is important to note that an early trigger will result in the yielding of immature oocytes and a fewer mature oocytes [7].Therefore, in order to obtain a larger follicular pool and more viable embryos, it is reasonable to prolong the stimulation period [24].

On the contrary, Kolibianakis et al. reported that in IVF cycle, large follicles (≥ 17 mm) may result in follicular luteinization, leading to a lower ongoing pregnancy rate [25]. On the other hand, delayed triggering results in more follicle growth and subsequent high levels of P, which negatively influence pregnancy rates in fresh embryo transfer cycles [26]. However, all of mentioned studies mainly focused on normal responders, which is in contrast to the results obtained in the present study in patients with DOR. It seems that in patients with DOR, due to physiological changes, postponement of triggering, cause premature ovulation and cycle cancellation [27]. Our results revealed that follicles measuring 18 mm or larger were associated with poorer outcomes in DOR patients. Specifically, both fertilization rates and TQE formation were significantly lower in the ≥ 18 mm group compared to the 15–17 mm group.

Distinctive features can be observed when comparing the ovarian physiology of patients with DOR and normal responders. Pre-antral and small antral follicles produce AMH, which inhibit FSH-induced aromatase activity and regulate follicular recruitment [28]. In DOR patients, reduced in the antral follicles and AMH level results in a decrease in luteal phase inhibition of FSH-induced aromatase activity, recruitment of early dominant follicle, and accelerated follicular development [24]. This progression leads to early ovulation, a shorter follicular phase, and a shorter menstrual cycle length [29]. Increasing basal FSH levels in women with DOR has a critical role in shaping follicular development. High FSH levels promote earlier maturation of granulosa cells, which leads to premature expression of LH receptors. As a result, follicles in DOR patients may be competent for ovulation at smaller sizes in comparison with those seen in women with normal ovarian reserve [30]. This altered physiology suggests that follicles in these patients may achieve functional maturity sooner, which emphasize the need for adjusted timing when triggering final oocyte maturation. Recognizing this shift is essential in tailoring individualized stimulation and trigger strategies to optimize clinical outcomes in this challenging patient population.

Therefore, considering the pathophysiological changes in patients with DOR, adjusting the trigger time to a smaller than normal follicle size may be advisable. This approach could potentially better mimic the natural physiology of these patients [12].

Few studies have examined relationship between follicle size and oocyte maturity in patients with DOR. Akgün and colleagues demonstrated that in order to prevent premature ovulation and optimize cycle outcomes in poor responders in POSEIDON groups 3 and 4, triggering should be done when the leading follicle reaches 16.5–17 mm [31]. Similarly, we indicated that follicles with a size of 15–17 mm had better reproductive outcomes.

Conversely, Tian and colleagues reported that the rate of fertilization was dramatically higher in follicles sized 16–17 mm compared to those sized 12–15 mm or ≥ 18 mm. However, other indices, such as the number of oocytes retrieved and the number of good-quality embryos, showed no significant variation between groups [6]. In the current study, rate of fertilization in follicles with a size of 15–17 mm was significantly higher than other groups. Lawrenz et al. found that the probability of oocyte retrieval was comparable between follicles larger and smaller than 15 mm. However, the highest maturation rates were observed in the 13–17 mm range, with a decrease in maturation rates at both extremes of follicle size [12]. Another study showed that among patients older than 35 years of age, there was a positive effect on the mature oocyte rate and fertilization rate in medium group (16–18 mm) [18]. Our data suggest that oocytes tend to mature earlier in women with DOR, resulting in accelerated follicular growth. This suggests that a follicle size of 15 to 17 mm, rather than 18 mm, may be the optimum size for triggering ovulation. Although the ROC analysis identified an optimal cut-off of 14.5–15.5 mm for predicting COC retrieval, our results showed that follicles measuring 15–17 mm yielded significantly higher rate of mature oocyte (MII), fertilization, and high-quality embryo formation. These findings suggest that while follicles in a size of 14.5 mm can be successfully retrieved, the likelihood of obtaining a mature and developmentally competent oocyte is highest in the 15–17 mm range.

Our findings indicated that in DOR patients, the optimal follicle size for triggering oocyte maturation and supporting embryo development consistently falls within the 15–17 mm range, regardless of chronological age. We found no significant differences in clinical outcomes within this follicle size range in comparison between POSEIDON group 3 (< 35 years) and group 4 (≥ 35 years). Rates of oocyte competence, including COC retrieval and MII maturation, along with fertilization (2PN) and TQE formation, were remarkably similar across both age groups. These findings suggest that, in the context of low ovarian reserve, the biological behavior of follicles, characterized by accelerated granulosa cell maturation and earlier LH receptor expression, may play a more decisive role than age alone in determining follicular readiness for ovulation. Therefore, in DOR patients, relying solely on chronological age to guide the timing of final oocyte maturation may be inadequate. Instead, individualized monitoring of follicular development offers a more precise and effective approach to optimize clinical outcomes.

One important limitation of the current study was that follicular size was recorded on the day of OPU, rather than at the time of trigger. However, based on retrospective chart reviews, we observed that follicular diameter typically increased by approximately 1 to 1.5 mm between the trigger and OPU.

The most favorable outcomes, particularly regarding mature oocyte yield, were seen in follicles 15–17 mm on the day of retrieval. This showed that the size of follicles was likely around 13.5–15.5 mm at the time of trigger. As such, the findings imply that optimal trigger timing in women with DOR may need to occur earlier than current standards suggest. Nonetheless, given the relatively small sample size in this study, we cannot definitively conclude that the ideal trigger size is 13.5–15.5 mm. Larger and prospective researches are needed to validate this hypothesis and refine trigger protocols for this population. Our interpretation is further supported by the study of Lawrenz et al. They showed that even when triggers were administered at smaller follicle sizes (≤ 15 mm), mature oocytes were still retrieved and outcomes were improved in DOR patients [12]. All in all, these results highlight the potential clinical benefit of earlier and individualized trigger strategies, especially in poor responders, and are in line with our observations when accounting for the expected follicular growth between trigger and retrieval.

Premature ovulation remains a key concern in patients with DOR. In the current study, 13.27% of patients experienced premature ovulation, characterized by significantly higher basal FSH, lower AMH, and elevated LH and progesterone on trigger day. Also, the mean follicular size in these patients was 18.6 mm on trigger day. This result was consistent with results of Akgün et al., who reported that patients with premature ovulation had significantly larger leading follicles, averaging 19.8 ± 2.4 mm on trigger day and the mean serum progesterone and LH levels on the day of hCG trigger were significantly higher in the premature ovulation group compared to control subjects [31]. Similarly, Lawrenz et al. demonstrated that triggering ovulation at smaller follicular sizes (≤ 15 mm) was associated with significantly lower rates of premature ovulation and higher oocyte retrieval (OPU) rates, compared to triggering at sizes >15 mm [12].

In a normal ovarian cycle, granulosa cells within developing follicles initially express only the FSH receptor. The expression of the LH receptor typically occurs much later, in a tightly regulated manner, and is confined to preovulatory follicles under the synergistic influence of FSH and estrogen. This precise timing is crucial for preparing the follicle to respond to the mid-cycle LH surge and initiate ovulation [28]. In patients with DOR, this physiological sequence is disrupted. DOR is characterized by a compensatory rise in basal FSH. Since one of the established roles of FSH is to induce LH receptor expression in granulosa cells, the elevated FSH exposure in DOR delivers a disproportionately strong signal, leading to the premature expression of LH receptor. Consequently, follicles enter a luteinized state at a smaller diameter, leading to premature progesterone rise and an increased risk of ovulation before or shortly after the scheduled trigger [30]. Concurrently, low AMH reduces the inhibition on follicular maturation, exacerbating this accelerated development [28]. This altered follicular physiology necessitates a paradigm shift in trigger timing in DOR patients. Therefore, considering earlier trigger at slightly smaller follicle sizes may improve outcomes and reduce premature ovulation and cycle cancellations in these patients.

One of the major challenges in DOR patients is low antral follicle count, which results in fewer retrieved oocytes and lower pregnancy rates compared to normal responders [32]. For these patients, maximizing of the number of oocytes retrieved is critical. In standard clinical IVF practice, follicles >13–14 mm are usually aspirated during ovum pickup, while follicles smaller than 12 mm are often ignored [3]. However, some studies have suggested that follicles 12 mm in size may yield mature oocytes and TQE suitable for transfer, which help aspirate of follicles larger than 12 mm in DOR patients [12, 33]. On the contrary, Yagüe-Serrano et al. showed that follicles in size of 11.5 mm was able to produce mature oocytes, but only those that were larger than 13.5 mm produced transferable embryos [3]. Similarly, in our study, we observed that follicles 11 and 12 mm produced mature (MII) oocytes, but none of the oocytes retrieved from follicles smaller than 13 mm resulted in successful fertilization or TQE. This finding suggests that although these oocytes have achieved nuclear competence, nuclear maturation alone is not sufficient to improve IVF outcomes [34]. It has been established that both nuclear maturation and cytoplasmic maturation are prerequisites for full oocyte competence [35]. The absence of developmental potential in these mature oocytes, as seen in our study, suggests that they lack cytoplasmic competence. As a result, these oocytes are unable to undergo chromosomal rearrangements, epigenetic modifications, fertilization, and subsequent embryonic development [36]. Based on these findings, the lower rate of oocytes fertilization obtained from follicles ≤ 12 mm in size may be explained by nuclear and cytoplasmic maturation disrupted [37]. Therefore, for mature oocytes obtained from follicles ≤ 12 mm in size, IVM protocols aimed at improving oocyte quality and competence may be appropriate [38]. Our finding suggests that to optimize reproductive success in patients with DOR, all developed ovarian follicles should be punctured.

The novelty of this study was the separate incubation of the collected oocytes and the labeling of the MII oocytes,2PN and TQE based on follicle of origin. This methodological approach allowed the assessment of oocyte fertilization potential and embryo quality according to follicle size. Limitations of the current study were the small sample size, measurement of follicles size on day of OPU, and also the lack of information on pregnancy outcomes due to the short duration of the study.

Further investigation is needed to determine the most effective trigger strategy for women with DOR. In poor responders who do not meet standard criteria for ovulation induction, these findings may help guide clinical decision making and avoid cycle cancellation. In addition, to provide evidence that can inform guidelines and enable more personalized care for these patients, studies focused on low-responder populations are essential.

Conclusion

Our study suggested that in women with DOR, triggering oocyte maturation when follicles measure 15–17 mm leads to the best outcomes in terms of oocyte yield, maturity, fertilization, and embryo quality. These findings were consistent across both younger and older DOR patients, indicating that age alone is not a reliable guide for trigger timing. Instead, the altered follicular physiology in DOR groups such as early LH receptor expression may require earlier trigger decisions. Notably, follicle measurements were taken on the day of oocyte retrieval, and retrospective data suggest a growth of 1–1.5 mm between trigger and retrieval. Thus, optimal trigger size may actually be closer to 13.5–15.5 mm. Given the study’s limitations, including sample size and timing of measurements, larger prospective studies are needed. Still, these results support more individualized and earlier trigger strategies in women with DOR. We propose a hypothesis to be tested in a prospective trial, comparison of trigger at lead follicle of 17 mm versus 15 mm.

Abbreviations

DOR

Diminished ovarian reserve

ART

Assisted reproductive technology

IVF

In vitro fertilization

ICSI

Intracytoplasmic Sperm Injection

ESHRE

European Society of Human Reproduction and Embryology

POSEIDON

Patient-Oriented Strategies Encompassing Individualized Oocyte Number

AFC

Antral follicular count

AMH

Anti-mullerian hormone

OPU

Ovum pickup

COC

Cumulus oocyte complex

MII

Metaphase II

2PN

Two pronuclei

TQE

Top-quality embryo

BMI

Body mass index

FSH

Follicle stimulating hormone

LH

Luteinizing hormone

P

Progesterone

E2

Estradiol

HCG

Human chorionic gonadotropin

HMG

Human menopausal gonadotropin

AUC

Area under the curve

Authors’ contributions

*Elham Naghshineh: * Validation, Conceptualization, Methodology, Writing - Original Draft. *Hatav Ghasemi Tehrani: * Data curation, Methodology, Software. *Ferdous Mehrabian: * Software, Visualization, Investigation. *Farahnaz Mardanian: * Software, Writing - Original Draft. *Marzieh Rezaei: * Supervision, Validation, Writing- Reviewing and Editing.

Funding

Not applicable.

Data availability

The datasets used and/or analyzed during the current study could become available through the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

The study was approved by the ethics committee of Isfahan University of Medical Sciences, Isfahan, Iran under the ethics code of IR.MUI.REC.1403.043. In addition, all subjects signed an informed consent form to participate in the study and a consent form for publication.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.