This study was approved by the Ethics Committee of 900TH Hospital of the Joint Logistics Support Force. Informed consent was obtained from all participating couples. All patient information was anonymized and strictly confidential.

We retrospectively analyzed the clinical data of patients with elevated basal FSH levels combined with DOR (401 cycles) who underwent IVF/intracytoplasmic monosperm injection (ICSI)-assisted conception treatment at the 900TH Hospital of Joint Logistics Support Force Reproductive Center, Fuzhou, China, from January 2019 to October 2022. Inclusion criteria included (1) age < 45 years; (2) basic FSH ≥ 10 U/L; in addition to at least two of the following three criteria: (3)vaginal ultrasound suggestive of bilateral ovarian antral follicle count (AFC) ≤ 7 [6, 7]; (4)anti-mullerian hormone (AMH) < 1.1 ng/ml [6, 7]; (5)previous cycles with low ovarian response and ≤ 3 eggs obtained by conventional protocol [6]. Exclusion criteria included: (1) chromosomal abnormalities in both or one of the couple; (2) coexistence of relevant diseases affecting IVF pregnancy outcome, such as severe adenomyosis, untreated hydrosalpinx, untreated endometrial lesions, or uterine fibroids ≥ 4 cm; and (3) patients with endocrine metabolic diseases such as polycystic ovary syndrome (PCOS). There are no effective treatments to improve outcomes in these severe cases of DOR. E2 pretreatment may be able to increase oocyte production and improve outcomes. A total of 401 cycles were included and divided into two groups according to whether estrogen pretreatment was performed during the luteal phase: there were 79 patients in the estrogen pretreatment group and 322 patients in the control group (i.e., no estrogen pretreatment). Early patients got estrogen pretreatment (source of 79 experimental cohorts) because we thought that pretreatment with an antagonist regimen could enhance the synchronization of follicular clusters. On the other hand, we discovered that FSH rebound was common in patients with basal FSH ≥ 10 U/L in addition to DOR. Consequently, estrogen therapy was eventually stopped in these individuals (source of 322 controls). The treatment parameters employed by the medical staff were uniform.

All patients received 17β-estradiol (Fentanyl Red Tablets, Abbott Laboratories, Netherlands). Ovulation was monitored one cycle prior to the antagonist regimen for ovulation. The objective of the ultrasound examinations was to evaluate the number and sizes of early antral follicles. 2 mg of oral 17β-estradiol was taken twice daily, starting 7 to 8 days after ovulation and discontinued after the second day of menstruation. The patient was pretreated with estrogen and did not have any washout period prior to initiation of gonadotropins. A potential disadvantage of considering any washout period is the gradual release of endogenous FSH prior to initiation of FSH, which may compromise the coordination required for antral follicle size. All patients had venous blood draws completed between 7am and 7:15am.

On the second day of menstruation, gonadotropin (Gn), including recombinant FSH (Prilosec, Merzadone, Germany; Gonafine, Merck Serono, Switzerland) or urinary-derived FSH (Lishenbao, Zhuhai Lizhu Pharmaceuticals) was administered at a dose of 150–300 IU/d, adjusted according to the responsiveness of the follicles. When the maximum follicle diameter reached 12–13 mm, the E2 level exceeded 400 pg/mL, and/or luteinizing hormone (LH) > 5 U/L, 0.25 mg/day cetrorelix acetate (Cetrotide, Merck Serono, Switzerland) was administered until HCG day. When the diameter of the dominant follicle reached 18 mm, HCG was injected at 6,000–10,000 U that night, and eggs were retrieved by vaginal ultrasound-guided puncture 36–38 h later.

The obtained oocytes were subjected to conventional IVF or ICSI in vitro fertilization, and fertilization was observed 16–18 h after insemination. High-quality embryos included normal fertilization, day 3, and day 5/6 embryos of high quality (day 3 embryos of grade 1–2, 7–9 ovoid spheres, < 20% fragmentation ; blastocysts at least at expansion stage 3 with an endocytic quality score of A or B, and day 5 trophectoderm score of A or B). On the 3rd day after egg retrieval, 1 ∼ 2 embryos with the highest grade were routinely transferred, while the rest were cultured, and blastocysts with grades of 4BC, 4CB, or higher were frozen. All embryo freezing was performed if one of the following conditions was met: progesterone (P) level on the day of HCG > 1.4 ng/ml, endometrial thickness < 6 mm, and the presence of cavity occupancy or uterine adhesions not suitable for fresh embryo transfer.

The frozen–thawed embryo transfer (FET) protocol was performed one to two menstrual cycles after egg retrieval. One blastocyst, or one to two cleavage-stage embryos were transferred depending on the regularity of the patient’s menstrual cycle and the condition of the endometrium; a natural cycle or hormone replacement cycle was selected.

The primary outcomes were the number of high-quality embryos, clinical pregnancy rate of fresh embryo transfer, and cumulative pregnancy rate; secondary outcomes were the number of MII eggs, Gn dose, and duration. Clinical pregnancy was diagnosed by ultrasound detection of a gestational sac 2 weeks after a positive hCG test.

The SPSS 26.0 software package (IBM Corp., Armonk, NY, USA) was used for statistical analysis. Normally distributed data were represented as mean and standard deviation (SD), and skewed data are described as the median and interquartile range (IQR). We used the chi-square test or Fisher’s exact test (when appropriate) to make statistical inferences about the qualitative data. We used the t-test or Mann–Whitney U test to compare continuous variables, as required. A probability (P) value < 0.05 indicated that the difference between the two groups was statistically significant.