Progestin primed ovarian stimulation yields comparable outcomes to the GnRH antagonist protocol for controlled ovarian hyperstimulation.

The mean age of the mothers in the antagonist group was 36.80 ± 5.43 years compared to 35.87 ± 5.31 years in the PPOS group ( p  = 0.102). The mean age of the fathers in the antagonist group was 39.40 ± 6.32 years and that of the PPOS group was 38.51 ± 5.12 years ( p  = 0.131). The mean BMI of the PPOS group was significantly higher than that of the antagonist group ( p  = 0.003). The percentage of women who smoked in the PPOS group was significantly higher than in the antagonist group ( p  = 0.014). The percentage of those who had not previously undergone COH cycle(s) in the antagonist group was significantly higher than in the PPOS group ( p  = 0.004). While the percentage of parents with unknown and mixed infertility etiology was higher in the PPOS group, the percentage of mothers with DOR as the cause of infertility was significantly higher in the antagonist group ( p  = 0.013). The percentage of those who underwent NGS was significantly higher in the PPOS group ( p  = 0.001). Other examined variables were similar in the two groups. In the GnRH antagonist group, the LBR was 30.64% (72/235), compared to 25.33% (38/150) in the PPOS group ( p  = 0.261). The ongoing pregnancy rate was 41.28% (97/235) in the antagonist group and 38.00% (57/150) in the PPOS group ( p  = 0.522). (Tables 1 , 2 , 3 ). Table 1 Summary of demographic and background related variables with regard to groups. Groups p Antagonist (n = 235) PPOS (n = 150) Age of female 36.80 ± 5.43 35.87 ± 5.31 0.102 † Age of male 39.40 ± 6.32 38.51 ± 5.12 0.131 † Body mass index (kg/m 2 ) 23.44 ± 3.60 24.82 ± 4.84 0.003 † Smoking (Female) 32 (13.62%) 35 (23.33%) 0.014 # Smoking (Male) 77 (32.77%) 57 (38.00%) 0.293 # Duration of infertility (months) 27.5 (12–41) 28 (16–48) 0.212 ‡ Type of infertility Primary 161 (68.51%) 93 (62.00%) 0.189 # Secondary 74 (31.49%) 57 (38.00%) Number of prior COH None 130 (55.32%) 67 (44.67%) 0.004 # One 28 (11.91%) 37 (24.67%) Two or above 77 (32.77%) 46 (30.67%) Infertility etiology Tubal factor 10 (4.26%) 3 (2.00%) 0.013 # Male factor 21 (8.94%) 12 (8.00%) Unknown 37 (15.74%) 37 (24.67%) Anovulation 8 (3.40%) 5 (3.33%) DOR 122 (51.91%) 55 (36.67%) Endometriosis 10 (4.26%) 4 (2.67%) Mixed 24 (10.21%) 31 (20.67%) Uterine factor 3 (1.28%) 3 (2.00%) Descriptive statistics were presented by using mean ± standard deviation for normally distributed continuous variables, median (25th percentile–75th percentile) for non-normally distributed continuous variables and frequency (percentage) for categorical variables. COH controlled ovarian hyperstimulation, DOR diminished ovarian reserve, PPOS progestin-primed ovarian stimulation. † : Student t test, ‡: Mann Whitney U test, #: Chi-square test. Significant values are in [bold]. Table 2 Summary of variables related to ART procedures with regard to groups. Groups p Antagonist (n = 235) PPOS (n = 150) AMH (ng/mL) 1.01 (0.50–2.01) 1.01 (0.40–1.81) 0.451 ‡ AFC 5.5 (3–9) 6 (4–10) 0.209 ‡ Total gonadotropin dose (IU) 3600 (2250–4500) 3600 (2250–4050) 0.202 ‡ Duration of stimulation (days) 9.37 ± 1.16 9.22 ± 1.33 0.231 † Number of oocytes retrieved 7 (3–12) 7 (3–13) 0.825 ‡ Number of MII oocytes 4 (2–8) 5 (2–9) 0.500 ‡ Number of MI oocytes 1 (0–2) 1 (0–1) 0.745 ‡ Number of GV oocytes 1 (0–2) 1 (0–2) 0.666 ‡ Number of fertilizations at Day 1 4 (2–7) 4 (2–7) 0.959 ‡ Number of cleavages 4 (2–7) 4 (2–7) 0.814 ‡ Number of Day 3 embryos 4 (2–7) 4 (2–7) 0.767 ‡ Number of Day 5 embryos 3 (1–5) 3 (2–5) 0.448 ‡ Number of good-quality embryos 2 (1–4) 2 (1–4) 0.158 ‡ Oocyte maturation rate (%) 68.75 (50.00–84.62) 75.00 (58.82–83.33) 0.319 ‡ Fertilization rate (%) 62.97 ± 25.69 58.67 ± 25.85 0.111 † Rate of cleavage (%) 59.92 ± 26.40 56.43 ± 26.16 0.204 † Rate of blastocyst (%) 44.16 ± 27.11 44.19 ± 24.94 0.991 † Rate of good quality embryo (%) 33.33 (16.67–50.00) 34.31 (20.00–50.00) 0.146 ‡ Rate of blastocyst formation (%) 66.67 (50.00–100.00) 75.00 (50.00–100.00) 0.109 ‡ Rate of good quality blastocyst (%) 90.91 (50.00–100.00) 100.00 (57.14–100.00) 0.179 ‡ NGS application 27 (11.49%) 36 (24.00%) 0.001 # Number of NGS-applied embryos 2 (2–3) 2 (1–4) 0.714 ‡ Number of euploid embryos 1 (1–2) 1 (0–2) 0.349 ‡ Rate of euploid embryo (%) 50.00 (25.00–100.00) 50.00 (0.00–70.83) 0.172 ‡ Treatment outcome No oocyte retrieved 2 (0.85%) 4 (2.67%) 0.159 * Cycle cancelled 22 (9.36%) 10 (6.67%) No viable embryo 11 (4.68%) 13 (8.67%) Embryo transferred 200 (85.11%) 123 (82.00%) Number of embryos transferred None 35 (14.89%) 27 (18.00%) 0.126 # One 114 (48.51%) 83 (55.33%) Two 86 (36.60%) 40 (26.67%) Descriptive statistics were presented by using mean ± standard deviation for normally distributed continuous variables, median (25th percentile–75th percentile) for non-normally distributed continuous variables and frequency (percentage) for categorical variables. AFC antral follicle count, AMH anti müllerian hormone, GV germinal vesicle, M metaphase, NGS next-generation sequencing, PPOS progestin-primed ovarian stimulation. † : Student t test, ‡: Mann Whitney U test, #: Chi-square test, *: Fisher-Freeman-Halton test. Significant values are in [bold]. Table 3 Summary of variables related to ART procedures and outcomes with regard to groups. Groups p Antagonist (n = 235) PPOS (n = 150) Pregnancy outcome Biochemical pregnancy 133 (56.60%) 89 (59.33%) 0.596 # Clinical pregnancy 104 (44.26%) 63 (42.00%) 0.663 # Ongoing pregnancy 97 (41.28%) 57 (38.00%) 0.522 # Multiple pregnancy 14 (5.96%) 5 (3.33%) 0.359 # Miscarriage 55 (23.40%) 46 (30.67%) 0.114 # Live birth 72 (30.64%) 38 (25.33%) 0.261 # Gestational week at live birth 38 (37–38) 38 (38–39) 0.137 ‡ Maternal complication 3 (1.28%) 2 (1.33%) 1.000 § OHSS 3 (1.28%) 3 (2.00%) 0.682 § Fetal complication 2 (0.85%) 3 (2.00%) 0.382 § Descriptive statistics were presented by using mean ± standard deviation for normally distributed continuous variables, median (25th percentile–75th percentile) for non-normally distributed continuous variables and frequency (percentage) for categorical variables. ‡: Mann Whitney U test, #: Chi-square test, *: Fisher-Freeman-Halton test, §: Fisher’s exact test. ART assisted reproductive technologies, OHSS ovarian hyperstimulation syndrome, PPOS progestin-primed ovarian stimulation. Summary of demographic and background related variables with regard to groups. Descriptive statistics were presented by using mean ± standard deviation for normally distributed continuous variables, median (25th percentile–75th percentile) for non-normally distributed continuous variables and frequency (percentage) for categorical variables. COH controlled ovarian hyperstimulation, DOR diminished ovarian reserve, PPOS progestin-primed ovarian stimulation. † : Student t test, ‡: Mann Whitney U test, #: Chi-square test. Significant values are in [bold]. Summary of variables related to ART procedures with regard to groups. Descriptive statistics were presented by using mean ± standard deviation for normally distributed continuous variables, median (25th percentile–75th percentile) for non-normally distributed continuous variables and frequency (percentage) for categorical variables. AFC antral follicle count, AMH anti müllerian hormone, GV germinal vesicle, M metaphase, NGS next-generation sequencing, PPOS progestin-primed ovarian stimulation. † : Student t test, ‡: Mann Whitney U test, #: Chi-square test, *: Fisher-Freeman-Halton test. Significant values are in [bold]. Summary of variables related to ART procedures and outcomes with regard to groups. Descriptive statistics were presented by using mean ± standard deviation for normally distributed continuous variables, median (25th percentile–75th percentile) for non-normally distributed continuous variables and frequency (percentage) for categorical variables. ‡: Mann Whitney U test, #: Chi-square test, *: Fisher-Freeman-Halton test, §: Fisher’s exact test. ART assisted reproductive technologies, OHSS ovarian hyperstimulation syndrome, PPOS progestin-primed ovarian stimulation. According to logistic regression analysis results, longer duration of infertility (OR 0.987, 95% CI 0.977–0.997, p  = 0.011) and DOR (OR 0.279, 95% CI 0.153–0.509, p  < 0.001) were independently associated with lower LBR. High number of good quality embryos (OR 1.169, 95% CI 1.054–1.297, p  = 0.003), high rate of blastocyst (OR 1.016, 95% CI 1.006–1.027, p  = 0.003) and two embryo transfer (OR 2.161, 95% CI 1.297–3.602, p  = 0.003) were independently associated with higher LBR (Table 4 ). Table 4 Odds ratios for live birth, logistic regression analysis results. Univariable Multivariable (1) OR (95% CI) p OR (95% CI) p Age of female 0.911 (0.872–0.951) < 0.001 0.164 Age of male 0.961 (0.924–1.000) 0.051 Body mass index (kg/m 2 ) 0.955 (0.902–1.010) 0.107 Smoking (Female) 0.987 (0.551–1.771) 0.966 Smoking (Male) 0.930 (0.583–1.483) 0.761 Duration of infertility (months) 0.991 (0.982–1.000) 0.045 0.987 (0.977–0.997) 0.011 Type of infertility, Primary 1.647 (1.009–2.687) 0.046 0.066 Number of prior COH 0.725 (0.617–0.852) < 0.001 0.140 Etiology, Tubal factor 6.037 (1.819–20.039) 0.003 0.082 Etiology, Male factor 2.254 (1.092–4.652) 0.028 0.370 Etiology, Unknown 1.161 (0.669–2.015) 0.595 Etiology, Anovulation 1.589 (0.508–4.969) 0.426 Etiology, DOR 0.238 (0.144–0.394) < 0.001 0.279 (0.153–0.509) < 0.001 Etiology, Endometriosis 3.516 (1.191–10.383) 0.023 0.285 Etiology, Mixed 1.673 (0.922–3.035) 0.091 Etiology, Uterine factor 5.151 (0.930–28.540) 0.061 AMH 1.263 (1.059–1.507) 0.010 0.352 AFC 1.096 (1.043–1.152) < 0.001 0.631 Treatment protocol, PPOS 0.768 (0.485–1.218) 0.262 Total dose 1.000 (1.000–1.000) 0.060 Duration of stimulation 0.929 (0.772–1.117) 0.433 Number of oocytes retrieved 1.086 (1.049–1.125) < 0.001 0.634 Number of MII oocytes 1.133 (1.080–1.189) < 0.001 0.415 Number of fertilizations 1.150 (1.088–1.217) < 0.001 0.937 Number of cleavages 1.156 (1.092–1.223) < 0.001 0.959 Number of Day 3 embryos 1.159 (1.093–1.230) < 0.001 0.711 Number of Day 5 embryos 1.264 (1.165–1.372) < 0.001 0.960 Number of good-quality embryos 1.303 (1.188–1.429) < 0.001 1.169 (1.054–1.297) 0.003 Oocyte maturation rate (%) 1.012 (1.002–1.021) 0.016 0.555 Fertilization rate (%) 1.009 (1.000–1.018) 0.061 Rate of cleavage (%) 1.014 (1.005–1.023) 0.003 0.852 Rate of blastocyst (%) 1.015 (1.006–1.024) 0.001 1.016 (1.006–1.027) 0.003 Rate of good quality embryo (%) 1.015 (1.007–1.024) 0.001 0.685 Rate of blastocyst formation (%) 1.015 (1.006–1.023) 0.001 0.934 Rate of good quality blastocyst (%) 1.014 (1.007–1.022) < 0.001 0.361 NGS application 0.603 (0.313–1.161) 0.130 Number of embryos transferred, Two 1.955 (1.236–3.094) 0.004 2.161 (1.297–3.602) 0.003 Nagelkerke R 2 – 0.250 AFC antral follicle count, AMH anti müllerian hormone, CI confidence interval, COH controlled ovarian stimulation, DOR diminished ovarian reserve, M metaphase, OR odds ratio, PPOS progestin-primed ovarian stimulation. (1) Multivariable analysis was performed by using forward selection method. Significant values are in [bold]. Odds ratios for live birth, logistic regression analysis results. AFC antral follicle count, AMH anti müllerian hormone, CI confidence interval, COH controlled ovarian stimulation, DOR diminished ovarian reserve, M metaphase, OR odds ratio, PPOS progestin-primed ovarian stimulation. (1) Multivariable analysis was performed by using forward selection method. Significant values are in [bold].

This retrospective study was conducted at the Reproductive Medicine Center of Acıbadem Kozyatagı Hospital, Istanbul, Turkey, following approval from the Acıbadem University Health Institutions Medical Research Ethics Committee (decision no: 2023-17/570). All procedures adhered to institutional ethical standards and the 1964 Helsinki declaration and its amendments. Participant data were anonymized for privacy. Due to the retrospective nature of the study, the Acıbadem University Health Institutions Medical Research Ethics Committee waived the need of obtaining informed consent. The study included first-time ICSI cycle patients with frozen embryos who were treated at our center between January 2022 and November 2023. This retrospective analysis comprised a total of 385 patients, with 150 undergoing the PPOS protocol and 235 receiving the GnRH antagonist protocol. Inclusion criteria were maternal age 22–48 years, ART indication, and the use of PPOS or GnRH antagonist protocol for controlled ovarian hyperstimulation (COH). Secondarily, only patients who underwent frozen embryo transfer, received blastocyst, and had undergone endometrial preparation via artificial hormonal cycle were included. Exclusion criteria were as follows: use of hormonal contraceptives for pretreatment before the study cycle, presence of a functional ovarian cyst with E2 > 100 pg/mL, cycles using donated oocytes or sperm, congenital or acquired uterine anomaly, abnormal chromosomal karyotype in either partner, endocrine disorders such as abnormal thyroid function and hyperprolactinemia, missing relevant data, any contraindication for ovarian stimulation therapy. Moreover, women were excluded if they had an abnormal uterine cavity shown on hysteroscopy. The demographic information and smoking status of the couples, body mass index (BMI) of the females, information related to infertility, all ART cycles, pregnancy, and childbirth were obtained from hospital computer records, patient files, and through phone calls with the patients. The following detailed data were also collected and recorded: duration of infertility, type of infertility (primary or secondary), number of prior ART cycles, infertility etiology, anti müllerian hormone (AMH) levels, antral follicle count (AFC), total gonadotropin dosage administered during ovarian stimulation, duration of stimulation, and the number of oocytes, retrieved number of metaphase (M) II oocytes, number of MI oocytes, number of germinal vesicle (GV) stage oocytes. Additionally, we recorded counts for fertilization, cleavage, day 3 embryos, day 5 embryos, and good quality embryos. Next-Generation Sequencing (NGS) application was assessed and the number of NGS-applied embryos and the number of euploid embryos were clarified. Finally, treatment outcome (classified as: no oocyte retrieved, cycle cancelled, no viable embryo, and embryo transferred), the number of embryos transferred, pregnancy outcome (classified as: biochemical pregnancy, clinical pregnancy, ongoing pregnancy, multiple pregnancy, miscarriage, and live birth), gestational week at live birth, OHSS, and maternal and fetal complications were recorded. Moreover, oocyte maturation index, fertilization rate, rate of cleavage, rate of blastocyst, rate of good quality embryo, rate of blastocyst formation and rate of good quality blastocyst were calculated. Patients were classified as having diminished ovarian reserve (DOR) if they met at least one of the following criteria: Basal follicle stimulating hormone (FSH) value > 10 IU/L and estradiol levels ≥ 80 pg/mL on cycle day 2 or 3; AFC < 5 or AMH level < 1.1 ng/mL 22 . A rising beta-human chorionic gonadotropin (hCG) level with no ultrasound evidence of a gestational sac is defined as a biochemical pregnancy 23 . A clinical pregnancy was determined by the visualization of a gestational sac or fetal heartbeat on ultrasound in women with a confirmed positive beta-hCG test 12 . A miscarriage was defined as a pregnancy that ended in spontaneous or therapeutic abortion 24 . Ongoing pregnancy was defined as the presence of at least one fetus showing heart pulsations on ultrasound beyond the 10th week of gestation 1 . A live birth was identified as the delivery of any viable infant at 28 weeks of gestation or later 24 . The primary outcome was live birth rate (LBR). Secondary outcomes included oocyte yield, embryo quality, fertilization rates, and other ART related parameters. Controlled ovarian stimulation was initiated on the 2nd or 3rd day of menstrual bleeding. All patients were assessed via transvaginal ultrasound to determine if there were any impediments to commencing ovarian stimulation on the 2nd–3rd day of menstruation. In the antagonist protocol: on the 2nd–3rd day of the menstrual cycle, the initial dose was determined based on the patient’s age, BMI, AMH level, and AFC. Daily recombinant FSH injections (Gonal F, Merck Serono, Switzerland) were started within a dose range of 225–450 IU. When the leading follicle reached a diameter of 12–14 mm based on ultrasound monitoring every 2–3 days, daily 0.25 mg GnRH antagonist injections (Cetrotide, Merck Serono, France) were initiated for the purpose of pituitary suppression and continued until the trigger day. In the PPOS protocol; gonadotropins were initiated on the 2nd–3rd day of the menstrual cycle in a similar manner. Concurrently with gonadotropins, twice-daily 5 mg medroxyprogesterone acetate (Tarlusal, Deva, Turkey) tablets were prescribed until the trigger day. Both protocols involved monitoring ovarian response through sequential transvaginal scanning, with or without hormonal monitoring. When three or more follicles reached a diameter exceeding 18 mm and a significant number of follicles were over 14 mm, the administration of 200 µg of triptorelin acetate (Gonapeptyl, Ferring GmbH, Kiel, Germany) and 250 µg of recombinant hCG (Ovitrelle, Merck-Serono, Italy) subcutaneously was conducted to prompt final oocyte maturation. After triggering ovulation, transvaginal oocyte retrieval was carried out (around 34–35 h later). In all cases, ICSI was performed for oocyte fertilization. Embryos were cultured until they reached the blastocyst stage using SAGE 1-Step HSA (Origio, Cooper Surgical Company, Denmark). Fertilization was confirmed by the presence of 2 pronuclei (2PN), and embryo quality was evaluated based on cleavage, blastomere number and regularity, and degree of embryonic fragmentation on the 2nd or 3rd day of culture. On day 5, all embryos meeting a grade of ≥ 2BC according to the Gardner and Schoolcraft system were classified as good quality and subjected to vitrification 25 . Vitrification and thawing were done according to manufacturer instructions (Vitrification Media VT601, Thawing Media VT602, Kitazato BioPharma Corporation, Japan). After thawing, embryos that survived for at least 2 h under laboratory conditions and expanded were considered suitable for transplantation. Indications for Preimplantation Genetic Testing for Aneuploidy (PGT-A) included advanced maternal age, recurrent miscarriages, repeated ART failure, severe sperm factors, or at the family’s request. A standardized biopsy and PGT-A protocol were employed. Trophectoderm biopsy was conducted before embryos were vitrified on day 5. The approach to endometrial preparation was the same for the two groups. All patients underwent office hysteroscopy with endometrial cavity scratching immediately after their menstrual period ended, and received 3.75 mg subcutaneous leuprolide acetate (Lucrin, AbbVie Medical, Turkey) on the same day. In the following cycle, oral 4 mg estradiol hemihydrate (Estrofem, Novo Nordisk A/S, Denmark) was started on days 2–4 of the menstrual cycle and gradually increased to a maximum of 8 mg/day until the 14th day, depending on endometrial thickness. Once the endometrial thickness reached ≥ 8 mm, the dosage of estradiol hemihydrate was maintained, and vaginal micronized progesterone (Progestan, Koçak Farma Pharmaceuticals, Turkey) at a dosage of 800 mg/day and subcutaneous progesterone (Progestan Dex, Koçak Farma Pharmaceuticals, Turkey) at a dosage of 25 mg twice daily were added. On the 14th day of the cycle, methylprednisolone (Prednol tablets, Mustafa Nevzat Pharmaceutical Industry, Turkey) 4 mg once daily and doxycycline (Monodoks tablets, Deva Holding, Turkey) 100 mg twice daily were started and continued until the day of transfer. Blastocyst transfers were conducted on the 6th day of progesterone administration. The number of embryos transferred was determined by maternal age, ART indication, and embryo quantity and quality. All embryo transfers and ultrasound follicle tracking were carried out by the same skilled infertility physician under transabdominal ultrasound guidance. The doses of estrogen and progesterone remained constant until blood beta-hCG was assessed 12 days post-transfer. If pregnancy persisted, estrogen and progesterone doses were gradually reduced from week 8 until discontinuation at week 12. AMH and other hormone levels were measured in the certified biochemistry laboratory of our hospital using calibrated standard measuring devices and according to the manufacturer’s recommendations. All cycles were conducted under standardized laboratory conditions, using the same culture media and protocols. Embryo assessments were performed by the same team of experienced embryologists, ensuring consistency across all cases. According to descriptive statistics (effect size = 0.479) in the study by Zhuo-ni Xiao et al. 26 , sample size of 93 for each group (186 in total) achieve 90% power at the two-sided 0.05 significance level. Sample size was calculated using “t tests—Means: Difference between two independent means (two groups)” function of G*Power software 27 . All analyses were performed on IBM SPSS Statistics for Windows, Version 25.0 (IBM Corp., Armonk, NY, USA). Two-tailed p-values of less than 0.05 were considered statistically significant. Histogram and Q-Q plots were used to determine whether variables are normally distributed. Descriptive statistics were presented by using mean ± standard deviation for normally distributed continuous variables, median (25th percentile–75th percentile) for non-normally distributed continuous variables, and frequency (percentage) for categorical variables. Normally distributed variables were analyzed with the Student t-test. Non-normally distributed variables were analyzed with the Mann–Whitney U test. Categorical variables were analyzed with the chi-square test or the Fisher’s exact test (or Fisher-Freeman-Halton depending on group count). Logistic regression analyses were performed to determine significant factors independently associated with LBR. Variables were analyzed with the univariable regression analysis and those with univariate significance were included into the multivariable model. Multivariable logistic regression analysis was performed with the forward selection method.

Our analyses show that in ICSI cycles performed by day-5 transfer of cryopreserved embryos into artificially prepared endometrium, the use of the PPOS and GnRH antagonists protocols resulted in no significant differences in terms of variables related to oocytes, fertilization, embryos, and pregnancy outcomes, as well as maternal and fetal outcomes. Longer duration of infertility and the presence of DOR were independent factors for low LBR rate; whereas, high number of good quality embryos, high rate of blastocysts, and greater number of embryos transferred were independent factors for high LBR rate. PPOS appears to be an effective procedure for ovarian stimulation in ICSI cycles, offering comparable outcomes to the GnRH antagonist protocol. Supporting this, a recent meta-analytical review highlights that poor ovarian responders, patients with PCOS, women of advanced age, and oocyte donors derive greater benefit from PPOS protocols compared to conventional stimulation methods 37 . Nonetheless, treatment strategies should be individualized based on each patient’s clinical profile to optimize outcomes.

Efforts to mitigate the adverse effects of early LH surge on pregnancy outcomes typically involve the use of GnRH analogs 1 . However, recent advancements in vitrification technology and its increased adoption have expanded the use of progestins in ovarian stimulation to prevent early LH surge 9 . The PPOS protocol offers considerable flexibility and potential benefits in clinical practice; however, there remains a lack of sufficient evidence regarding its effectiveness and safety concerning fertilization and pregnancy outcomes 17 . Elevated serum progesterone levels during oocyte maturation have been shown to adversely affect embryo quality and LBR 2 , 28 , 29 . On the other hand, several studies comparing PPOS protocols using different types of progestins, such as medroxyprogesterone acetate and desogestrel, to antagonist protocols have reported no significant differences in fertilization and pregnancy outcomes, suggesting that progestins may offer comparable ART success rates with a potentially more favorable safety profile 19 , 20 , 30 . In the present study, there were no significant differences in total gonadotropin dose, duration of stimulation, oocyte, embryo and fertilization outcomes, biochemical pregnancy rate (BPR), clinical pregnancy rate (CPR), ongoing pregnancy rate (OPR), multiple pregnancy rate, miscarriage rate, LBR, gestational week at live birth, and maternal and fetal complications in patients receiving either protocol. Furthermore, multiple regression analysis revealed that the type of protocol was not associated with LBR. Duration of infertility and presence of DOR were the two factors that were independently associated with decreased LBR; whereas, higher number of good quality embryos, high rate of blastocyst, and two embryo transfer were independently associated with greater LBR. In a study involving oocyte donation cycles, no difference was detected in the number of MII oocytes retrieved when these protocols were compared. MII oocytes were fertilized with partner sperm in 76% of cycles in the medroxyprogesterone acetate group compared to 74% in the ganirelix group. Despite similar recipient and cycle characteristics, reproductive outcomes were poorer in progestin recipients, as demonstrated by significantly lower BPR, CPR, OPR, and LBR 13 . This may be due to the negative effect of progestin on oocyte quality in the follicular phase or oocyte freeze–thaw procedure. The success of the use of PPOS in combination with oocyte cryopreservation should be questioned in comprehensive studies. However, similar to our results, a randomized controlled trial involving individuals with poor ovarian response demonstrated that premature LH surges were less likely in the PPOS group compared to the GnRH antagonist group. In the PPOS group, the average counts of oocytes and viable embryos were comparable to those observed in the GnRH antagonist group, and the LBR showed no significant difference between the two groups 20 . In another trial including infertile women with polycystic ovary syndrome (PCOS), the number of mature oocytes, maturity rate, number of oocytes with 2PN, and serum estradiol levels on trigger day were significantly lower in PPOS recipients compared to the antagonist group. Additionally, serum LH levels on the trigger day were higher in the PPOS group than in the antagonist group. Although mild and moderate OHSS occurred less frequently in the PPOS group, both the BPR and CPR values were statistically similar in the two groups—despite relatively better results in the antagonist group 31 . In a study by Huang et al., which involved women with poor ovarian response, there were no significant differences observed in the number of oocytes retrieved between the two groups. However, the rates of MII oocytes, fertilization, and formation of good-quality embryos were notably higher in the PPOS group compared to the antagonist group. Additionally, analyses for subsequent frozen-thawed embryo transfer showed that both CPR and LBR were significantly higher in the PPOS group compared to the antagonist group 32 , indicating superiority of outcomes for PPOS recipients. For secondary results, the retrospective cohort study by Turkgeldi et al. focusing on women with DOR, showed that the median duration of stimulation, pituitary suppression, the median number of cumulus-oophorous complexes, MII oocytes, total number of cryopreserved oocytes, and oocyte maturation rates were comparable between the protocols 33 . A meta-analysis of randomized controlled trial (RCT) examining the PPOS protocol revealed that the CPR, LBR and OPR did not differ significantly from those observed among recipients of the GnRH antagonist protocol. However, in the subgroup analysis of patients with DOR, the PPOS protocol was associated with a lower incidence of premature LH surge and OHSS. Secondary outcomes indicated that, in both DOR patients and those with normal ovarian reserve, the PPOS protocol resulted in higher numbers of retrieved oocytes, MII oocytes, and viable embryos compared to the antagonist protocol 34 . In a recent study, it was found that the rate of cycle cancellations was notably lower among GnRH antagonist recipients when compared to the PPOS group. While the implantation rate, CPR, OPR, and LBR did not differ significantly, the cumulative LBR after one complete in vitro fertilization (IVF) cycle was significantly higher in the GnRH antagonist group. Additionally, and perhaps more importantly, the average time required to achieve live birth was significantly shorter in the GnRH antagonist group than in the PPOS group, indicating a critical advantage for the antagonist protocol. Furthermore, ongoing pregnancies leading to live births were significantly more frequent among GnRH antagonist recipients compared to the corresponding value in the PPOS group. Ultimately, women undergoing the antagonist protocol were found to be 2.32 times more likely to achieve a live birth compared to those using the PPOS protocol 1 . According to a very recent systematic review and meta-analysis involving women with PCOS, PPOS recipients had lower OHSS risk and required higher gonadotropin dosage compared to those undergoing the GnRH antagonist protocol. However, crucially, no significant differences were found regarding oocyte quality and pregnancy outcomes 35 . In a recent prospective study involving repeated ovarian stimulation cycles, the PPOS protocol yielded a significantly higher number of oocytes, MII oocytes, and 2PN compared to the GnRH antagonist protocol. Regarding euploidy rates per biopsied embryo, the PPOS and GnRH antagonist groups had statistically similar outcomes, with frequencies of 29% and 35%, respectively 21 . In the current study, euploid embryo rates were 50% in both groups. In their investigation of three protocols, Zhou et al. found that the PPOS regimen was associated with lower LBR compared to the GnRH antagonist and GnRH agonist protocols. Similar to aforementioned literature, the authors also reported shorter time to live birth among GnRH antagonist recipients compared to PPOS recipients. When compared to recipients of GnRH analogues, PPOS recipients had lower BPR and CPR, and fewer high-quality blastocysts. Perinatal outcomes, however, were similar with all three protocols 4 . The varied outcomes reported in the literature suggest uncertainties regarding the utility and reliability of PPOS in ovarian stimulation, a crucial stage of ART. The differing results could stem from variations in the selected patient populations. Some studies included specific patient subsets (such as those undergoing only frozen embryo transfer, donor oocyte transfer, patients with DOR, poor ovarian response, PCOS or endometriosis, as well as those undergoing IVF or ICSI), while others examined all patients undergoing ART procedures. The exact day of embryo transfer may also influence the outcomes. Another potential reason could be differences in the number of participants. In the current study, to minimize the potential impact of these factors, only ICSI procedures involving day-5 frozen embryos transferred to the artificially-prepared endometrium were assessed. In endometrial preparation, hormone replacement following pituitary desensitization procedure was used. Endometrium was prepared after stimulation of the endometrium with scratching and hysteroscopy confirmed that the cavity was normal. This procedure is quite different from the endometrial preparation procedures used in current publications. Our results, consistent with the majority of previous studies, demonstrated comparable in fertilization and pregnancy outcomes between the PPOS and the GnRH antagonist protocols. Additionally, it was determined that the most influential factors on LBR were the duration of infertility, the presence of DOR, the number of good quality embryos, the blastocyst rate, and the number of embryos transferred, consistent with previous research. The significant baseline imbalances observed between groups represent a major limitation of this retrospective study. While multivariable logistic regression was used to adjust for these differences, statistical techniques such as propensity score matching or inverse probability weighting would have provided stronger control for confounding and enhanced causal inference. The higher BMI and smoking rates in the PPOS group, combined with higher DOR prevalence in the antagonist group, create potential for unmeasured confounding that our analytical approach cannot fully address. Although LH surge was not investigated in our study, it is important that there was no significant difference between PPOS and GnRH antagonist groups in terms of cycle and pregnancy outcomes. PPOS has the potential to provide the same pregnancy outcomes with a lower rate of adverse side effects. Therefore, future studies should be planned with consideration of all these confounding factors, potentially yielding more reliable results regarding the safety and efficacy of PPOS in ovulation stimulation. Beyond clinical outcomes, the PPOS protocol offers practical advantages including oral administration and potentially lower costs compared to GnRH antagonists. In our experience, patients often prefer oral medications due to convenience and reduced injection discomfort. While formal cost-effectiveness analysis was beyond this study’s scope, the comparable efficacy combined with these practical benefits suggests PPOS may be a patient-friendly alternative in clinical practice. Although our study has notable advantages, such as the high patient counts, the fact that embryo transfer and patient follow-up were done by the same person and the analysis of only frozen embryos transferred on day 5 into artificial-prepared endometrium, it also has some limitations. This is a single-center study and the external validity is limited and results could have been impacted by experience with the methods and other secondary factors. Being a retrospective cohort, patients were non-randomly assigned to different protocols, and there were significant differences between the groups in terms of some important confounding variables (e.g., BMI, smoking, previous ART history, etiology of infertility, NGS application). Most critically, the significant baseline imbalances between groups introduce substantial confounding that limits our ability to draw definitive conclusions about protocol effectiveness. Future studies should employ randomized controlled trial design or advanced statistical methods such as propensity score matching to minimize these biases. Additionally, the sample sizes of the groups were not equal, which may have affected the analyses and potentially impacted the study’s statistical power. A RCT would provide stronger evidence for comparing the efficacy of the PPOS and GnRH antagonist protocols. Second we included women up to 48 years of age in our study. It is well-documented that IVF success rates decline significantly with advanced maternal age, particularly after 45 years. In our clinic, we thoroughly counsel all patients, especially those over 45, about the low probability of success and the potential risks associated with IVF treatment. We emphasize that donor oocytes may offer a more realistic chance of success in this age group. However, in Turkey, donor oocyte programs are not legally permitted, and many women are reluctant to consider donor eggs, preferring to exhaust all options using their own oocytes. Additionally, there is no legal upper age limit for IVF treatment in Turkey. As clinicians, we believe that every woman with viable oocytes should have the opportunity to pursue pregnancy, provided she is fully informed about the low likelihood of success. When comparing between protocols, infertility causes were examined together; thus, the differing impacts of specific infertility causes could not be analyzed. Crucially, hormone levels on the trigger day and LH surge information were not included in the study due to concerns regarding data reliability. This represents a significant limitation as LH and estradiol levels on trigger day are crucial indicators of protocol effectiveness in preventing premature LH surge. The absence of this data limits our ability to assess the primary mechanism by which these protocols differ and may mask important physiological differences that could influence oocyte quality and treatment outcomes. Finally, the cost-effectiveness and patient satisfaction of the protocols and the potential differences between different progestin treatments (since all patients received the same protocols) were not assessed, and it is possible that progestin types could have differing effects 36 . Lastly, by including only patients who achieved blastocyst-stage embryos for transfer, our study introduces a selection bias that excludes cycles with poorer outcomes (e.g., no oocytes retrieved, cycle cancellation, or embryo arrest before day 5). This selection criterion likely overestimates the success rates of both protocols and limits generalizability to the broader IVF population, particularly patients who may not reach blastocyst stage due to protocol-specific effects on embryo development.

Ovarian stimulation is essential in assisted reproductive techniques (ART), primarily aimed at inducing follicle production, usually with gonadotropins 1 . However, a rapid estrogen increase can trigger excessive luteinizing hormone (LH) release, leading to early ovulation, which can negatively impact the number of oocytes retrieved and, consequently, affect cycle success and pregnancy rates 1 , 2 . To prevent LH surges during ART, physicians often use gonadotropin-releasing hormone (GnRH) agonists, antagonists, and progestin 1 , 3 . However, the comparative efficacy and reliability of these methods are not well-established 2 , 4 . GnRH agonists, while effective, complicate follicle synchronization, increase costs, and pose risks such as ovarian hyperstimulation syndrome (OHSS) 5 , 6 . The GnRH antagonist protocol is commonly preferred in ART due to its convenience, safety, and efficacy compared to agonists 5 , 7 . Despite its benefits, GnRH antagonists are costly, require parenteral administration, and have been linked to higher rates of cycle cancellation and premature LH surge 8 , 9 . The ongoing challenges have led to the development of safer protocols, including the introduction of Progestin-primed ovarian stimulation (PPOS) in 2015 10 . Progesterone and its derivatives, which can be orally administered, have emerged as an alternative for preventing LH surges during the follicular phase 11 . The PPOS protocol controls the hypothalamic-pituitary-ovarian axis, allowing for oocyte maturation without the risk of moderate or severe OHSS 12 – 14 . This method is reported to effectively prevent premature LH surges without compromising oocyte competence, particularly in cycles followed by embryo cryopreservation 11 , 15 , 16 . Although PPOS is suggested to yield comparable oocyte retrieval and pregnancy rates to the GnRH antagonist protocol 14 , 17 , the evidence is limited and inconclusive, particularly regarding embryo transfers 1 , 2 . Current literature comparing PPOS and antagonist protocols often neglects factors like progesterone’s potential drawbacks in fresh embryo transfers 12 , 18 , 19 , timing of embryo transfer 13 , 18 , 20 , ART type 19 , and methods of endometrial preparation 13 . Furthermore, studies addressing these aspects often have small sample sizes 21 , complicating a fair comparison between PPOS and the GnRH antagonist protocol. This study aims to compare the GnRH antagonist and PPOS protocols in intracytoplasmic sperm injection (ICSI) cycles with frozen blastocyst embryos, specifically in the context of artificial endometrial preparation, focusing on oocyte outcomes, fertilization rates, embryo quality, and pregnancy outcomes.