Clinical outcomes of the short agonist stop protocol versus the long GnRH agonist protocol in poor ovarian responders undergoing IVF/ICSI: a retrospective analysis.

The baseline clinical characteristics of the 103 enrolled patients are presented in Table  1 . Notably, the proportion of patients classified in POSEIDON Group 3 was significantly higher than in the other three groups. Table 1 Baseline clinical characteristics ( n  = 103) Age (years) 33.60 ± 4.79 BMI(Kg/m 2 ) 23.61 ± 3.10 Infertility duration (years) 5.18 ± 2.89 Infertility type  Primary infertility n (%) 49(47.57)  Secondary infertility n (%) 54(52.43) AMH (ng/ml) 0.73 ± 0.46 FSH (mIU/ml) 14.88 + 3.21 LH (mIU/ml) 4.21 ± 4.74 E2 (ng/ml) 57.77 ± 28.51 AFC (n) 3.75 ± 1.62 POSEIDON Classification  Group 1 n (%) 16 (15.53)  Group 2 n (%) 13 (12.62)  Group 3 n (%) 49 (47.57)  Group 4 n (%) 25 (24.27) Categorical variables are presented as number (%). Continuous variables are presented as mean ± standard deviation BMI body mass index, AMH anti-Müllerian hormone, FSH follicle stimulating hormone, LH luteinizing hormone, E2 estradiol, AFC antral follicle count Baseline clinical characteristics ( n  = 103) Categorical variables are presented as number (%). Continuous variables are presented as mean ± standard deviation BMI body mass index, AMH anti-Müllerian hormone, FSH follicle stimulating hormone, LH luteinizing hormone, E2 estradiol, AFC antral follicle count A significantly greater number of Gn stimulation days were observed in the STOP group than in the control group ( P   0.05). Table 2 Characteristics of controlled ovarian hyperstimulation cycles Grouping STOP group Control group p Gn days n 11.67 ± 4.14 10.71 ± 2.61 0.048 Total Gn amount IU 41.02 ± 16.94 42.07 ± 16.22 0.652 Gn initiation day  FSH (mIU/ml) 14.62 ± 2.22 14.21 ± 2.73 0.250  LH (mIU/ml) 3.50 ± 2.15 2.93 ± 1.95 0.047  E2 (ng/ml) 56.21 ± 24.13 49.07 ± 17.56 0.017 Gn day 5  LH (mIU/ml) 2.3 ± 2.0 2.48 ± 2.04 0.529  E2 (ng/ml) 409.89 + 222.06 436.98 + 220.77 0.383  P (ng/ml) 0.59 ± 1.01 0.55 ± 0.62 0.703 HCG day  LH (mIU/ml) 3.81 ± 3.43 2.48 ± 4.82 0.023  E2 (ng/ml) 382.38 ± 253.27 444.09 ± 1079.58 0.575  P (ng/ml) 0.59 ± 0.27 0.88 ± 1.63 0.194 Follicles ≥ 16 mm on HCG administration day n 2.65 ± 1.27 2.99 ± 1.63 0.099 Trigger type 0.578  HCG trigger alone n (%) 84 (81.55) 87 (84.47)  Combined trigger n (%) 19 (18.45) 16 (15.53) Categorical variables are presented as frequency (%), and comparisons were performed using Pearson’s chi-square test. Continuous variable normality was assessed by the Shapiro–Wilk test. Normally distributed continuous variables are presented as mean ± standard deviation and were compared using the independent-samples t -test or one-way ANOVA, as appropriate. Non-normally distributed continuous variables are presented as median (interquartile range) and were compared using the Mann–Whitney U test. p  < 0.05 was considered statistically significant Gn gonadotropin,  FSH follicle stimulating hormone, LH luteinizing hormone, E2 estradiol, P progesterone, hCG human chorionic gonadotropin Characteristics of controlled ovarian hyperstimulation cycles Categorical variables are presented as frequency (%), and comparisons were performed using Pearson’s chi-square test. Continuous variable normality was assessed by the Shapiro–Wilk test. Normally distributed continuous variables are presented as mean ± standard deviation and were compared using the independent-samples t -test or one-way ANOVA, as appropriate. Non-normally distributed continuous variables are presented as median (interquartile range) and were compared using the Mann–Whitney U test. p  < 0.05 was considered statistically significant Gn gonadotropin,  FSH follicle stimulating hormone, LH luteinizing hormone, E2 estradiol, P progesterone, hCG human chorionic gonadotropin On the Gn initiation day, FSH did not differ significantly between groups ( P  > 0.05). However, significantly higher Luteinizing hormone (LH) and E2 levels were observed in the STOP group than in the control group ( P   0.05). On the day of hCG administration, LH levels were significantly higher in the STOP group than in the control group ( P   0.05). The number of follicles ≥ 16 mm on the day of hCG administration and at trigger did not differ significantly between the two groups ( P  > 0.05). Regarding embryology laboratory characteristics, FORT and FOI did not differ between groups ( P  > 0.05). However, the numbers of retrieved oocytes and MII oocytes were significantly higher in the STOP group than in the control group ( P  < 0.05). No significant differences were observed between the STOP and control groups about fertilization method, number of two-pronuclear embryos, embryo formation rate, or blastocyst formation rate following fertilization ( P  > 0.05). There were no significant differences found in the number of cycles without transferable embryos or in the cycle cancellation rate due to unavailable or unusable oocytes between the two groups ( P  > 0.05) (Table  3 ). Table 3 Laboratory characteristics of embryos from controlled ovarian hyperstimulation ( n  = 103) Grouping STOP group ( n  = 103) Control group ( n  = 103) P Retrieved oocytes ( n  = 103) 2.99 ± 1.25 2.24 ± 1.07 0.00 FORT (follicular output rate) (%) 0.86 ± 0.65 1.1 ± 1.13 0.061 FOI (follicle-oocyte index) (%) 0.95 ± 0.67 0.81 ± 0.76 0.173 Number of MII oocytes n 2.46 ± 1.25 1.95 ± 1.04 0.002 Number of 2PN n 1.67 ± 1.33 1.55 ± 0.94 0.472 Good-quality embryo rate (%) 44.19 46.25 0.706 Blastocyst formation rate (%) 83.87 73.53 0.311 Cycles without transferable embryos n (%) 44 (42.72) 47 (45.63) 0.674 Fertilization method 0.418  IVF n (%) 36 (37.50) 39 (43.33)  ICSI n (%) 60 (62.50) 51 (56.67) Cycle cancellation rate n (%) 7 (6.80) 13 (12.62) 0.158 Categorical variables are presented as frequency (%), and comparisons were performed using Pearson’s chi-square test. Continuous variable normality was assessed by the Shapiro–Wilk test. Normally distributed continuous variables are presented as mean ± standard deviation and were compared using the independent-samples t -test or one-way ANOVA, as appropriate. Non-normally distributed continuous variables are presented as median (interquartile range) and were compared using the Mann–Whitney U test. p  < 0.05 was considered statistically significant The blastocyst formation rate was calculated as the total number of blastocysts formed on day 5 or day 6 divided by the number of normally fertilized oocytes (i.e., cleavage-stage embryos on day 2 or day 3), expressed as a percentage. The day 3 good-quality embryo rate was calculated as the number of good-quality day 3 embryos divided by the number of normally fertilized embryos, expressed as a percentage. A good-quality day 3 embryo was defined as an embryo containing 7–9 cells, with a fragmentation rate < 10%, no significant multinucleation, and uniformly sized blastomeres, as assessed on day 3 after fertilization FORT follicular output rate, FOI follicle-oocyte index, IVF in vitro fertilization, ICSI intracytoplasmic sperm injection Laboratory characteristics of embryos from controlled ovarian hyperstimulation ( n  = 103) Categorical variables are presented as frequency (%), and comparisons were performed using Pearson’s chi-square test. Continuous variable normality was assessed by the Shapiro–Wilk test. Normally distributed continuous variables are presented as mean ± standard deviation and were compared using the independent-samples t -test or one-way ANOVA, as appropriate. Non-normally distributed continuous variables are presented as median (interquartile range) and were compared using the Mann–Whitney U test. p  < 0.05 was considered statistically significant The blastocyst formation rate was calculated as the total number of blastocysts formed on day 5 or day 6 divided by the number of normally fertilized oocytes (i.e., cleavage-stage embryos on day 2 or day 3), expressed as a percentage. The day 3 good-quality embryo rate was calculated as the number of good-quality day 3 embryos divided by the number of normally fertilized embryos, expressed as a percentage. A good-quality day 3 embryo was defined as an embryo containing 7–9 cells, with a fragmentation rate < 10%, no significant multinucleation, and uniformly sized blastomeres, as assessed on day 3 after fertilization FORT follicular output rate, FOI follicle-oocyte index, IVF in vitro fertilization, ICSI intracytoplasmic sperm injection Clinical pregnancy outcomes analyzed in this study included data from both fresh and frozen embryo transfer cycles following the respective OS protocols. A total of 64 transfer cycles were performed in the STOP group and 59 in the control group. No significant differences were observed between the STOP and control groups in the number of embryo transfer cycles (fresh and frozen) or in the number of embryos transferred (cleavage-stage embryos or blastocysts) ( P  > 0.05). No significant differences were observed between the two groups in implantation rate, biochemical pregnancy rate, clinical pregnancy rate (including both singleton and twin pregnancies), miscarriage rate, or ectopic pregnancy rate ( P  > 0.05) (Table  4 ). Table 4 Clinical pregnancy outcomes Grouping STOP group ( n  = 64) Control group ( n  = 59) p Number of transfers n (%) 0.901 Number of frozen transfers 46 (71.88) 43 (72.88) Number of fresh transfers 18 (28.12) 16 (27.12) Number of embryos transferred n (%) 0.830  Embryos 46 (63.89) 41 (62.12)  Blastocysts 26 (36.11) 25 (37.88) Embryo implantation rate n (%) 27 (37.50) 15 (22.73) 0.060 Biochemical pregnancy rate n (%) 17 (26.56) 19 (32.20) 0.492 Clinical pregnancy rate n (%) 24 (37.5) 15 (25.42) 0.150 Single fetus 21 (72.41) 15 (75.00) 0.273 Twin fetus 3 (10.34) 0 (0.00) 0.082 Miscarriage rate n (%) 5 (20.83) 4 (26.67) 0.711 Ectopic pregnancy rate n (%) 0 (0.00) 1 (6.60) 0.318 Categorical variables are presented as frequency (%) and were compared using Pearson’s chi-square test. p  < 0.05 was considered statistically significant Clinical pregnancy outcomes Categorical variables are presented as frequency (%) and were compared using Pearson’s chi-square test. p  < 0.05 was considered statistically significant

This retrospective, single-center study was conducted at the Reproductive Center of Shanxi Women and Children Hospital. According to the Bologna criteria for POR established by the ESHRE in 2011, patients with POR who underwent IVF or ICSI between January 2023 and April 2024 and received either the STOP or the standard short-acting long protocol for ovulation induction treatment within six months were included. Differences in various clinical indicators between the two protocols were subsequently compared. Approval for this study was obtained from the Ethics Committee of Shanxi Children’s Hospital (Women and Children Hospital) (Ethics No. IRB-KYSB-2023-024). Written informed consent was obtained from all participants included in the study between January 2025 and March 2025, and patient confidentiality was maintained throughout. Inclusion criteria: Diagnosis of POR according to the Bologna criteria; treatment with both the STOP protocol and the standard short-acting long protocol for OS within a six-month period; completion of embryo transfer (fresh and frozen) for all embryos obtained from both stimulation protocols; and no abnormalities in the uterine cavity or endometrial morphology [ 16 ]. Exclusion criteria: Chromosomal abnormalities in either or both partners; confirmed hydrosalpinx; abnormal uterine cavity morphology; a diagnosis of endometriosis; and a history of recurrent implantation failure (≥ 3 unsuccessful embryo transfer attempts). Pituitary downregulation was initiated during the mid-luteal phase of the preceding cycle (day 20) with triptorelin acetate at a dose of 0.1 mg/day (Changchun Jinsai Pharmaceutical, Jinlei, 1 mL:0.1 mg) until menstruation onset. Downregulation was confirmed on cycle days 2–4 by serum estradiol (E2) levels < 50 pg/mL and antral follicle diameter < 5 mm on transvaginal ultrasound, after which triptorelin administration was discontinued. Controlled OS was subsequently initiated with Gn at 300–450 IU/day, with the dosage determined according to individual BMI and previous ovarian response. The Gn preparations used included urinary follicle-stimulating hormone (FSH; Lizhu Pharmaceutical, 75 IU/vial) or recombinant FSH (Changchun Jinsai Pharmaceutical, 5.5 µg [75 IU]/vial). When the leading follicle reached a diameter of ≥ 13 mm and/or serum E2 levels exceeded 400 pg/mL, a GnRH antagonist (GnRH-Ant) (ganirelix acetate, 0.25 mg/day; Chia Tai Tianqing Pharmaceutical, 0.25 mg/vial) was added using either a fixed or flexible protocol and continued until trigger day. Final oocyte maturation was triggered when at least two follicles reached a diameter of ≥ 17 mm with appropriate peripheral E2 levels, using either human chorionic gonadotropin (hCG) alone (Lizhu Pharmaceutical, 2000 IU/vial) or a dual-trigger regimen combining hCG with a GnRH agonist (triptorelin acetate, 0.1 mg; Changchun Jinsai Pharmaceutical). Oocyte retrieval was performed 36 h after trigger, followed by standard luteal phase support. Pituitary downregulation was initiated during the mid-luteal phase of the preceding cycle (day 20) with triptorelin acetate at a dose of 0.1 mg/day (Changchun Jinsai Pharmaceutical, Jinlei, 1 mL:0.1 mg) and continued until the onset of menstruation. Downregulation was confirmed on cycle days 2–4 by serum E2 levels < 50 pg/mL and antral follicle diameter < 5 mm on transvaginal ultrasound. Controlled OS was subsequently initiated with Gn at 300–450 IU/day, with the dosage determined according to individual body mass index and previous ovarian response, using urinary FSH (Lizhu Pharmaceutical, 75 IU/vial) or recombinant FSH (Changchun Jinsai Pharmaceutical, 5.5 µg [75 IU]/vial), in combination with continued administration of triptorelin acetate at 0.1 mg/day until the day of trigger. Triggering, oocyte retrieval, and luteal phase support were performed as described for the STOP protocol. Embryo morphology was graded by number, size, and uniformity of blastomeres, the degree of fragmentation, and the presence of multinucleated blastomeres. A top-quality embryo was defined as an embryo with at least three cleavage-stage blastomeres on day 2 and at least seven cleavage-stage blastomeres on day 3, relatively homogeneous blastomeres, no multinucleation, and < 20% fragmentation [ 10 ]. On days 5 and 6, blastocyst morphology was assessed using the Gardner and Schoolcraft grading system [ 17 ]. Embryo transfer strategies were determined by a multidisciplinary team based on multiple clinical considerations. Fresh transfer was performed either at the cleavage stage (day 3) or the blastocyst stage (day 5/day 6). All available embryos not selected for fresh transfer due to factors such as elevated progesterone levels, abnormal endometrial morphology, or inadequate endometrial thickness were cryopreserved for subsequent frozen embryo transfer cycles. In frozen embryo transfer cycles, endometrial preparation was performed using either natural or artificial cycle protocols. One or two embryos were transferred according to individual patient characteristics and patient preference. All embryo transfers were performed under transabdominal ultrasound guidance in accordance with departmental protocols [ 18 ]. Pregnancy outcomes were assessed according to the following criteria. Pregnancy was determined 10 days after embryo transfer by measurement of serum β-hCG levels. Biochemical pregnancy was defined as a positive β-hCG test result in the absence of ultrasound evidence of a gestational sac. Clinical pregnancy was confirmed by transvaginal ultrasound demonstrating a gestational sac with fetal cardiac activity [ 16 ]. The follicular output rate (FORT) was calculated as follows: (number of follicles with a diameter ≥ 16 mm on the day of hCG administration)/AFC × 100% [ 19 ]. The follicle-to-oocyte index (FOI) was calculated as follows: number of retrieved oocytes/AFC × 100% [ 20 ]. Primary outcomes included the number of retrieved oocytes, good-quality embryo rate, blastocyst formation rate, biochemical pregnancy rate, and clinical pregnancy rate. Blastocyst formation rate was defined as the total number of blastocysts formed on day 5 or day 6 divided by the number of normally fertilized oocytes, represented by the number of cleavage-stage embryos on day 2 or day 3. Statistical analyses were performed using SPSS version 26.0 software. Statistical methods were selected according to the nature and distribution of the data. For categorical variables, Pearson’s chi-square test was used, and the results were presented as numbers and percentages. The normality of continuous variables was assessed using the Shapiro–Wilk test. For normally distributed continuous variables, Student’s t -test or one-way analysis of variance was used. Non-normally distributed continuous variables were analyzed using the Mann–Whitney U test. A two-sided P  < 0.05 was considered statistically significant.

Poor ovarian response (POR) among women undergoing assisted reproductive technology has increased, affecting approximately 10%–24% of this population [ 1 ]. Controlled ovarian hyperstimulation (COH) is central to in vitro fertilization and intracytoplasmic sperm injection (IVF/ICSI) but in patients with POR, obtaining enough oocytes represents a significant challenge [ 2 ]. Although various ovarian stimulation (OS) protocols are available for patients with POR, an optimal treatment protocol has not yet been established [ 3 – 6 ]. In 2011, the European Society of Human Reproduction and Embryology (ESHRE) defined POR using the Bologna criteria, requiring at least two of the following three criteria: (1) advanced maternal age (≥ 40 years); (2) a previous poor response to conventional OS (≤ 3 oocytes retrieved); and (3) abnormal ovarian reserve test results, including an antral follicle count (AFC) < 5–7 or anti-Müllerian hormone (AMH) level < 1.1 ng/mL [ 7 ]. In 2016, the POSEIDON criteria classified POR into four groups. Group 1 includes patients younger than 35 years with AFC ≥ 5 and/or AMH ≥ 1.2 ng/mL; Group 2 comprises patients aged 35 years or older with AFC and AMH values similar to those of Group 1; Group 3 includes patients younger than 35 years with AFC < 5 and/or AMH < 1.2 ng/mL; and Group 4 consists of patients aged 35 years or older who meet the AFC and AMH criteria of Group 3 [ 8 ]. In 1998, Faber et al. introduced the stop GnRH agonist/GnRH antagonist (STOP) protocol for POR [ 9 ]. The STOP protocol initiates GnRH agonist (GnRH-Ag) downregulation in the luteal phase, continues until menstruation onset, then administers high-dose gonadotropins (Gn). This protocol has demonstrated significant benefits in patients with POR with multiple IVF failures, including increased oocyte yield, higher good-quality embryo rates, and improved pregnancy outcomes [ 10 ]. GnRH-Ag combined with Gn enhances follicular recruitment and ovarian responsiveness in patients with POR, yielding more oocytes and high-quality embryos, thereby improving the likelihood of pregnancy [ 11 – 13 ]. Consequently, the STOP protocol has emerged as a promising therapeutic strategy for the management of POR. However, evidence suggests that neither the timing of GnRH-Ag initiation (luteal versus follicular phase) nor discontinuation versus continuation of GnRH-Ag at OS onset significantly influences clinical outcomes [ 14 ]. As the STOP protocol is not included in the ESHRE guidelines on OS for IVF/ICSI and evidence remains limited regarding its use, this study aimed to compare OS characteristics and pregnancy outcomes between the conventional long GnRH agonist protocol and the STOP protocol in patients with POR undergoing IVF/ICSI treatment [ 15 ]. By evaluating differences in the numbers of retrieved oocytes and metaphase II (MII) oocytes, embryo development, and clinical pregnancy outcomes, this study aimed to provide additional clinical evidence on the STOP protocol’s role in POR management.

Although the STOP protocol was associated with significantly greater retrieved oocytes and MII oocyte numbers than the standard protocol, the oocyte yield increase was not accompanied by improvements in fertilization rate, embryo quality, blastocyst formation rate, implantation rate, or clinical pregnancy outcomes. These findings suggest that greater oocyte quantity alone may be insufficient to improve reproductive outcomes in patients with POR. Therefore, the clinical value of the STOP protocol in this population remains uncertain. Further prospective studies with larger sample sizes are warranted to clarify its potential role in specific POR subgroups.

In this study, 103 patients with POR who underwent the standard long GnRH agonist protocol or STOP protocol within six months demonstrated that the STOP protocol significantly increased retrieved oocyte and MII oocyte numbers. Although no significant differences were observed between the two protocols with respect to the high-quality embryo rate, blastocyst formation rate, or clinical pregnancy outcomes, the STOP protocol broadens the treatment options available for patients with POR. In patients with POR, obtaining sufficient high-quality oocytes is critical for improving live birth rates (LBRs) after IVF/ICSI. The selection of an appropriate OS protocol is therefore central to achieving this objective. Currently, clinical pregnancy rates per fresh cycle in patients with POR range from approximately 15% to 30%, with considerable variation across age groups [ 21 ]. An analysis of 144,018 IVF cycles performed between 2003 and 2007 by Nelson demonstrated that the retrieval of one additional oocyte per stimulated cycle was associated with an approximately 25% increase in the cumulative LBR per cycle [ 22 ]. Among women aged 35–37 years, LBRs were approximately 5–13% in those with 1–2 oocytes retrieved, compared with 13–18% in those with 2–3 oocytes retrieved [ 22 ]. In a study of 1,099 women aged 18–40 years undergoing their first IVF cycle at the Centre for Reproductive Medicine, Universitair Ziekenhuis Brussel, between January 2009 and December 2013, patients were stratified by oocyte yield (1–3, 4–9, 10–15, or > 15 oocytes). Following transfer of all fresh and frozen embryos, cumulative LBRs were significantly lower in patients with 1–3 oocytes than in those with 4–9 oocytes [ 23 ]. A randomized study involving 70 patients with POR assigned to either a short GnRH agonist protocol or a long GnRH agonist protocol found that the short protocol significantly increased the number of mature oocytes retrieved but did not significantly improve clinical pregnancy or implantation rates [ 24 ]. An analysis of 63 patients with POR comparing the STOP protocol with a conventional stimulation protocol presented that the STOP protocol significantly increased retrieved and mature oocyte numbers but did not significantly improve usable embryo numbers or clinical pregnancy rate, consistent with the present findings [ 16 ]. In contrast, Hazout et al. reported that although a short GnRH agonist protocol did not significantly increase oocyte yield versus a long-acting long protocol, but was associated with improved embryo morphology and more usable embryos [ 11 , 12 ]. Follicular synchronization is effectively achieved with long protocols through prolonged and potent suppression of endogenous FSH by GnRH-Ag [ 25 ]. However, continued GnRH-Ag administration may excessively suppress ovarian GnRH receptors, reducing ovarian sensitivity to FSH. Consequently, a longer duration of OS and higher doses of Gn may be required. Evidence suggests that increased Gn doses may be necessary in patients with POR to achieve satisfactory clinical pregnancy rates. For example, in conventional IVF protocols for patients with POR, the clinical pregnancy rate per cycle was only 4% with standard COH doses [ 26 ]. In contrast, the LBR per cycle increased to 7.7% when a daily Gn dose of 450 IU was administered [ 27 ]. The STOP protocol initiates downregulation with GnRH-Ag in the luteal phase, then discontinues at menstruation onset. This brief ovarian suppression promotes the release of FSH in the early follicular phase through the flare-up effect of GnRH-Ag, enhancing ovarian responsiveness without requiring increased medication dosages. Following discontinuation of GnRH-Ag, hypothalamic–pituitary suppression may persist for up to 10 days. Combined with a GnRH-Ant, this may prevent premature LH surges and improve embryo quality [ 12 , 28 ]. The combination may suppress LH surges during follicular stimulation, reducing premature ovulation risk [ 29 ]. In patients classified as POSEIDON Group 4, the STOP protocol with letrozole increases FSH sensitivity by recruiting two consecutive follicular waves, yielding more retrieved oocytes and an improved clinical pregnancy rate [ 30 ]. However, patients with POR often require higher Gn doses and longer stimulation under the STOP protocol to achieve greater oocyte yield, more good-quality embryos, and improved pregnancy rates [ 10 ]. In this study, although no significant difference in total Gn dose was observed between groups, the duration of Gn administration was significantly longer in the STOP group than in the control group. As follicular development and maturation require adequate levels of LH, the significantly higher LH levels in the STOP group on the day of hCG administration may have contributed to the considerably greater retrieved oocyte and MII oocyte numbers in this group. LH levels on the day of Gn initiation were significantly higher in the STOP group. LH surge suppression was achieved during subsequent OS through continued GnRH-Ag administration or the introduction of a GnRH-Ant when required. Continuous administration of GnRH-Ag appeared to be more effective in maintaining LH at a stable, low level, whereas the use of a GnRH-Ant was associated with greater LH fluctuations. This may have contributed to the higher LH levels observed on the day of hCG administration in the STOP group. These differences did not appear to affect fertilization or clinical pregnancy outcomes, as no significant differences were observed between groups. No significant differences in OS outcomes were observed between patients who received hCG alone for triggering and those who received a combined GnRH-Ag and hCG trigger. Although the STOP protocol permits GnRH-Ag for oocyte maturation induction, follicular maturation disorders may occur in some patients due to insufficient LH approximately 12 h after triggering. A dual-trigger combining low-dose hCG with GnRH-Ag has been proposed [ 29 ]. Prior studies have shown that dual triggering with GnRH-Ag and hCG improves final oocyte maturation in patients with POR, thereby improving IVF outcomes [ 31 – 33 ]. Variability in outcomes may reflect differences in ovarian responsiveness to exogenous FSH and cycle-to-cycle antral follicle count variation, including differences in granulosa cell FSH receptor expression and sensitivity [ 34 ]. Despite these variations, a delay of up to 180 days between OS cycles has been reported not to affect pregnancy outcomes in patients with diminished ovarian reserve [ 35 ]. An interval of less than six months between cycles was selected in this study for assessment. Prolonged OS may impair the endometrium at embryo transfer, thereby reducing the success rate of IVF/ICSI procedures [ 36 – 38 ]. The total Gn dose has been identified as a primary factor influencing the LBR [ 39 ]. In patients with total Gn dose exceeding 5000 IU, cancellation of fresh embryo transfer and cryopreservation of embryos is generally recommended. Prolonged OS and a high total Gn dose may be unfavorable for fresh embryo transfer. One study reported that the LBR in patients with POR ranged from 1% to 10% per cycle [ 40 ]. In contrast, cumulative LBR per cycle in POSEIDON patients was approximately 29.4% in Group 3 and 12.5% in Group 4 [ 41 ]. Pregnancy rates were comparable between the discontinued GnRH-Ag protocol and the long GnRH agonist protocol [ 35 ]. Faber et al. reported a pregnancy rate of 24% per embryo transfer, whereas Pantos et al. reported that the pregnancy rate in the discontinued GnRH-Ag group was nearly twice that of the long protocol group (35.2% vs. 19.4%) [ 9 , 42 ]. This study has several limitations. First, the retrospective single-center design carries an inherent selection bias risk. Enrollment was limited by the relatively low prevalence of POR, the requirement for both protocols within six months, and exclusion of cases with missing data or loss to follow-up, yielding a small sample size. Subgroup analyses by transfer type and POSEIDON category could not therefore be performed, and potentially meaningful subgroup differences may have gone undetected. Second, although the within-patient crossover design reduces inter-individual confounding, generalizability may be limited, as patients who completed both protocols within the specified timeframe may not be fully representative of the broader POR population. Third, LBR, which represents the most clinically relevant endpoint in assisted reproduction, was not assessed because the number of patients achieving live birth was too small for meaningful statistical analysis within the available sample. Future studies with larger samples should include LBR as a primary outcome and incorporate subgroup analyses according to transfer type and POSEIDON category to more comprehensively assess the STOP protocol’s comparative effectiveness in this patient population.