Effect of Low-Dose Estradiol on the Clinical Pregnancy Rate of the Hormone Replacement Frozen-Thawed Embryo Transfer Cycle.

This retrospective study included FET cycles conducted at the Reproductive Medicine Center of Nanjing Drum Tower Hospital from January 2022 to June 2025. All patients underwent endometrial preparation using an artificial hormone replacement protocol (Femoston, 2 mg estradiol; 2 mg estradiol combined with 10 mg dydrogesterone, Abbott, USA). Before initiating the HRT-FET cycle, all patients underwent a comprehensive evaluation to rule out contraindications for medication and pregnancy. The participants in this study were aged between 20 and 39 years and all received a single day-5 (D5) blastocyst transfer. The exclusion criteria were as follows ( Figure 1 ): Concomitant use of other hormone replacement medications. More than two previous transfer cycles. Pretreatment with gonadotropin-releasing hormone agonist (GnRHa). Concurrent hydrosalpinx, endometrial pathology, or abnormal uterine environment (eg., submucosal uterine fibroids, intrauterine adhesions). Endometriosis or adenomyosis. Preimplantation genetic testing (PGT) cycles. Figure 1 A flow chart of this study. A flowchart of HRT-FET study exclusions and inclusions from 2022 to 2025. Concomitant use of other hormone replacement medications. More than two previous transfer cycles. Pretreatment with gonadotropin-releasing hormone agonist (GnRHa). Concurrent hydrosalpinx, endometrial pathology, or abnormal uterine environment (eg., submucosal uterine fibroids, intrauterine adhesions). Endometriosis or adenomyosis. Preimplantation genetic testing (PGT) cycles. A flow chart of this study. All patients provided written informed consent, authorizing the use of anonymized medical records for research purposes. This study was approved by the Ethics Committee of Nanjing Drum Tower Hospital, Affiliated Hospital of Nanjing University Medical School (approval number: 2021–384-01). Hormone assays (serum E 2 , progesterone (P)) and transvaginal ultrasound examination were conducted during the early follicular phase (day 2–3) of the patient’s menstrual cycle. If the criteria for the early follicular phase were met, estradiol administration (estradiol tablets, Femoston, Abbott, USA; 2 mg q.d. or b.i.d. or t.i.d) was initiated for 14 days. We generally prioritized high-dose HRT regimens to attain maximal stability. Nevertheless, for patients with pre-existing conditions like endometrial hyperplasia, endometrial polyps, or polycystic ovary syndrome (PCOS, characterized by the absence of natural follicular development), certain physicians may select low-dose HRT regimens according to individual clinical evaluations. Subsequently, hormone assays (serum E 2 , P) and transvaginal ultrasound (to measure endometrial thickness) were repeated. Patients with suboptimal endometrial thickness (< 8 mm) were given additional estradiol tablets (estradiol tablets, Femoston, 2 mg q.d). When the endometrial thickness reached the predefined criterion (usually ≥ 8 mm; although some patients exhibit less than 8 mm, the expected endometrial thickness has still been achieved), a combined oral regimen of estradiol and dydrogesterone (estradiol/dydrogesterone tablets, Femoston, 2 mg estradiol and 10 mg dydrogesterone q.d. or b.i.d. or t.i.d) was started for 6 days. Concurrently, daily intramuscular progesterone injections (Progesterone Injection, XIANJU Pharma, Zhejiang, China; 60 mg q.d.) were administered for 6 days to induce endometrial transformation, followed by thawing and transfer of a single D5 blastocyst. Blastocyst morphological evaluation is based on the Gardner scoring system. 21 After embryo transfer, the oral medication was continued at the same dosage, while the intramuscular progesterone was replaced with a sustained-release vaginal progesterone gel (90 mg q.d.) to maintain endometrial receptivity for embryo implantation. Serum β-human chorionic gonadotropin (β-hCG) levels were measured two weeks after transfer to determine biochemical pregnancy. Patients with positive β-hCG underwent a transvaginal ultrasound two weeks later to confirm clinical pregnancy (visualization of a gestational sac). Luteal phase support was typically continued until 8 weeks after transfer for pregnant patients. Rigorous telephone follow-up was carried out to detect any abnormalities during pregnancy. Spontaneous miscarriage occurring before 12 weeks of gestation was defined as early miscarriage. Live birth was defined as the delivery of a live neonate after 28 weeks of gestation. The clinical pregnancy rate and live birth rate were calculated as the ratio of the number of cycles achieving clinical pregnancy or live birth, respectively, to the total number of embryo transfer cycles. The primary outcome measure of this study was the clinical pregnancy rate. All cycle data were stratified into two groups based on the estradiol dosage administered: Group A (high-dose estradiol group), which was initiated with 6 mg of estradiol, and Group B (low-dose estradiol group), which was initiated with 2 mg of estradiol, with a maximum increment not exceeding 2 mg. Univariate analysis was initially performed to assess variables potentially associated with the clinical pregnancy rate. Subsequently, a multivariate logistic regression model was employed to analyze the impact of low-dose estradiol in hormone replacement therapy cycles on the clinical pregnancy rate. Furthermore, propensity score matching (PSM) was applied to mitigate the influence of confounding variables (including female age, male age, Infertility type, baseline luteinizing hormone (LH), antral follicle count (AFC) and anti-Müllerian hormone (AMH)) on the final outcomes. Matching was conducted using a 1:2 without-replacement scheme, with a caliper width set at 0.01 times the standard deviation of the logit of the propensity score. Data are presented as mean ± standard deviation (SD). All analyses were performed using R ( https://www.R-project.org ) and EmpowerStats software ( https://www.empowerstats.com , X&Y Solutions, Inc., Boston, MA). A p-value < 0.05 was considered statistically significant.

A total of 2157 participants were included in the study ( Figure 1 ), with 1955 participants in Group A (high-dose estradiol) and 202 participants in Group B (low-dose estradiol) ( Table 1 ). Female age was significantly higher in Group A compared to Group B (31.15 ± 3.71 years vs. 30.07 ± 3.59 years, P < 0.001). Ovarian reserve markers revealed significant differences between the two groups. AMH levels were significantly lower in Group A (5.12 ± 3.83 ng/mL) compared to Group B (7.07 ± 4.83 ng/mL, P < 0.001). Similarly, basal Follicle-Stimulating Hormone (FSH) levels were higher in Group A (6.64 ± 2.06 IU/L) compared to Group B (6.28 ± 1.61 IU/L, P = 0.018), while basal LH levels were significantly lower in Group A (6.71 ± 4.72 IU/L) than in Group B (7.96 ± 5.01 IU/L, P < 0.001). Additionally, Group A had a lower AFC (24.13 ± 12.86) compared to Group B (29.53 ± 12.21, P < 0.001). A larger proportion of participants in Group B had primary infertility (68.32%) compared to Group A (53.81%), while Group A had more participants with secondary infertility (46.19%) compared to Group B (31.68%). Before endometrial transformation, there were significant differences in hormonal levels between the two groups. Group A had significantly higher E 2 levels (260.89 ± 256.44 pg/mL) compared to Group B (154.10 ± 81.59 pg/mL, P < 0.001). LH levels before endometrial transformation were also higher in Group B (18.23 ± 10.97 IU/L) compared to Group A (12.24 ± 8.84 IU/L, P < 0.001). However, there was no significant difference in progesterone levels (Group A: 0.17 ± 0.14 ng/mL, Group B: 0.17 ± 0.14 ng/mL, P = 0.450). Endometrial thickness was significantly greater in Group B (10.06 ± 1.63 mm) than in Group A (9.59 ± 1.48 mm, P < 0.001). Clinical pregnancy rates were significantly higher in Group B (79.21%) compared to Group A (69.57%, P = 0.004). However, early miscarriage rates did not significantly differ between the two groups (Group A: 12.13%, Group B: 13.75%, P = 0.556). There was no significant difference in the male age between the two groups (Group A: 32.17 ± 4.27 years vs. Group B: 31.52 ± 3.80 years, P = 0.058). The Body Mass Index (BMI) of females in both groups was similar, with Group A at 23.07 ± 3.38 kg/m 2 and Group B at 23.43 ± 3.66 kg/m 2 (P = 0.153). There was no significant difference in infertility duration between the groups, with Group A having a duration of 3.16 ± 2.09 years and Group B having a duration of 3.08 ± 2.18 years (P = 0.607). Fertilization methods did not differ significantly between the groups (P = 0.683). Table 1 Comparison of General Characteristics and FET Outcomes Data Between Different Estradiol Dose Groups Variable Group A (High-Dose Estradiol) Group B (Low-Dose Estradiol) P-value N 1955 202 Female age (years) 31.15 ± 3.71 30.07 ± 3.59 <0.001 Male age (years) 32.17 ± 4.27 31.52 ± 3.80 0.058 Female BMI (kg/m 2 ) 23.07 ± 3.38 23.43 ± 3.66 0.153 AMH (ng/mL) 5.12 ± 3.83 7.07 ± 4.83 <0.001 Basal FSH (IU/L) 6.64 ± 2.06 6.28 ± 1.61 0.018 Basal LH (IU/L) 6.71 ± 4.72 7.96 ± 5.01 <0.001 AFC (n) 24.13 ± 12.86 29.53 ± 12.21 <0.001 Infertility type <0.001 Primary infertility 1052 (53.81%) 138 (68.32%) Secondary infertility 903 (46.19%) 64 (31.68%) Infertility duration (years) 3.16 ± 2.02 3.08 ± 2.18 0.607 COH protocol 0.003   Long-acting GnRHa 1322 (67.62%) 123 (60.89%)   GnRH-Ant 541 (27.67%) 77 (38.12%)   CC/HMG 67 (3.43%) 2 (0.99%)   PPOS 25 (1.28%) 0 (0.00%) Fertilization method 0.683   IVF 1572 (80.41%) 160 (79.21%)   ICSI 383 (19.59%) 42 (20.79%) E 2 level before endometrial transformation (pg/mL) 260.89 ± 256.44 154.10 ± 81.59 <0.001 LH level before endometrial transformation (IU/L) 12.24 ± 8.84 18.23 ± 10.97 <0.001 P level before endometrial transformation (ng/mL) 0.17 ± 0.14 0.17 ± 0.14 0.450 EMT before endometrial transformation (mm) 9.59 ± 1.48 10.06 ± 1.63 <0.001 Clinical pregnancy 1360 (69.57%) 160 (79.21%) 0.004 Early miscarriage 165 (12.13%) 22 (13.75%) 0.556 Abbreviations : BMI, Body Mass Index; AMH, anti-müllerian hormone; FSH, Follicle-stimulating hormone; LH, luteinizing hormone; AFC, Antral follicle count; COH, Controlled Ovarian Hyperstimulation; GnRH, Gonadotropin-Releasing Hormone; CC, Clomifene Citrate; HMG, Human Menopausal Gonadotropin; PPOS, Progestin-primed ovarian stimulation; IVF, in vitro fertilization; ICSI, Intracytoplasmic Sperm Injection; E 2 : estrogen; P, progesterone; EMT, endometrial thickness. Comparison of General Characteristics and FET Outcomes Data Between Different Estradiol Dose Groups Abbreviations : BMI, Body Mass Index; AMH, anti-müllerian hormone; FSH, Follicle-stimulating hormone; LH, luteinizing hormone; AFC, Antral follicle count; COH, Controlled Ovarian Hyperstimulation; GnRH, Gonadotropin-Releasing Hormone; CC, Clomifene Citrate; HMG, Human Menopausal Gonadotropin; PPOS, Progestin-primed ovarian stimulation; IVF, in vitro fertilization; ICSI, Intracytoplasmic Sperm Injection; E 2 : estrogen; P, progesterone; EMT, endometrial thickness. As shown in Table 2 , the age of females was significantly associated with clinical pregnancy, with older females exhibiting a lower clinical pregnancy rate (OR = 0.919, 95% CI: 0.896–0.943, P < 0.001). The age of males also correlated positively with clinical pregnancy (OR = 0.953, 95% CI: 0.932–0.973, P < 0.001). However, female BMI did not show a significant association (OR = 0.980, 95% CI: 0.954–1.007, P = 0.149). AMH was significantly associated with clinical pregnancy, with higher AMH levels having a positive effect (OR = 1.084, 95% CI: 1.055–1.115, P < 0.001). Similarly, basal LH was positively associated with clinical pregnancy (OR = 1.053, 95% CI: 1.029–1.078, P < 0.001), while basal FSH did not show a significant association (OR = 0.967, 95% CI: 0.925–1.011, P = 0.139). AFC was positively associated with clinical pregnancy (OR = 1.026, 95% CI: 1.018–1.034, P < 0.001). Participants with secondary infertility had a lower likelihood of clinical pregnancy (OR = 0.761, 95% CI: 0.632–0.916, P = 0.004), whereas the OR for primary infertility was 1 (reference). The fertilization method did not significantly affect clinical pregnancy, with ICSI (OR = 0.993, 95% CI: 0.787–1.253, P = 0.954) showing no significant association. Table 2 Univariate Analysis of Estradiol Dosage in HRT-FET Cycles Variable Clinical Pregnancy OR 95% CI P-value Female age (years) 31.05 ± 3.71 0.919 (0.896, 0.943) <0.001 Male age (years) 32.11 ± 4.23 0.953 (0.932, 0.973) <0.001 Female BMI (kg/m 2 ) 23.11 ± 3.41 0.980 (0.954, 1.007) 0.149 AMH (ng/mL) 5.31 ± 3.98 1.084 (1.055, 1.115) <0.001 Basal FSH (IU/L) 6.60 ± 2.03 0.967 (0.925, 1.011) 0.139 Basal LH (IU/L) 6.83 ± 4.76 1.053 (1.029, 1.078) <0.001 AFC (n) 24.63 ± 12.90 1.026 (1.018, 1.034) <0.001 Infertility type   Primary infertility 1190 (55.17%) 1   Secondary infertility 967 (44.83%) 0.761 (0.632, 0.916) 0.004 Infertility duration (years) 3.15 ± 2.10 0.981 (0.940, 1.025) 0.401 COH protocol   Long-acting GnRHa 1445 (66.99%) 1   GnRH-Ant 618 (28.65%) 1.293 (1.046, 1.598) 0.018   CC/HMG 69 (3.20%) 0.957 (0.570, 1.607) 0.868   PPOS 25 (1.16%) 0.796 (0.349, 1.816) 0.588 Fertilization method   IVF 1732 (80.30%) 1   ICSI 425 (19.70%) 0.993 (0.787, 1.253) 0.954 Estradiol dosage   High-dose 1955 (90.63%) 1   Low-dose 202 (9.37%) 1.667 (1.171, 2.373) 0.005 Abbreviations : BMI, Body Mass Index; AMH, anti-müllerian hormone; FSH, Follicle-stimulating hormone; LH, luteinizing hormone; AFC, Antral follicle count; COH, Controlled Ovarian Hyperstimulation; GnRH, Gonadotropin-Releasing Hormone; CC, Clomifene Citrate; HMG, Human Menopausal Gonadotropin; PPOS, Progestin-primed ovarian stimulation; IVF, in vitro fertilization; ICSI, Intracytoplasmic Sperm Injection. Univariate Analysis of Estradiol Dosage in HRT-FET Cycles Abbreviations : BMI, Body Mass Index; AMH, anti-müllerian hormone; FSH, Follicle-stimulating hormone; LH, luteinizing hormone; AFC, Antral follicle count; COH, Controlled Ovarian Hyperstimulation; GnRH, Gonadotropin-Releasing Hormone; CC, Clomifene Citrate; HMG, Human Menopausal Gonadotropin; PPOS, Progestin-primed ovarian stimulation; IVF, in vitro fertilization; ICSI, Intracytoplasmic Sperm Injection. In the unadjusted analysis, low-dose estradiol was associated with a higher clinical pregnancy rate when compared to high-dose estradiol (OR = 1.667, 95% CI: 1.171–2.373, P = 0.005, Table 2 ). However, after adjusting for female age, male age, infertility type, AMH, basal LH, and AFC, no significant difference was found between the two groups (adjusted OR = 1.422, 95% CI: 0.983–2.055, P = 0.061, Table 3 ). Considering the differences between the two groups in cycle numbers and baseline characteristics ( Table 1 ), we used propensity score matching to identify cycle cohorts with similar baseline features. However, no statistically significant differences were noted in clinical pregnancy rates or early miscarriage rates between the two groups ( Table 4 ). Table 3 Multivariate Analysis for the Effect of Low-Dose Estradiol on the Clinical Outcome of HRT-FET Cycles GROUP Clinical Pregnancy aOR* 95% CI P-value A (High-dose estradiol) 1 B (Low-dose estradiol) 1.422 (0.983, 2.055) 0.061 Notes:* Adjust for: Female age; Male age; Infertility type; AMH; Basal LH; AFC. Abbreviations : aOR, odds ratio; CI, confidence interval. Table 4 Characteristics of HRT-FET Cycles with Different Estradiol Dosage After Propensity-Score Matching Variables GROUP A2 GROUP B2 Standardized Difference P-value N 384 192 Female age (years) 30.21 ± 3.46 30.04 ± 3.50 0.047 0.593 Male age (years) 31.42 ± 4.02 31.45 ± 3.63 0.008 0.934 Female BMI (kg/m 2 ) 23.12 ± 3.33 23.46 ± 3.69 0.097 0.266 AMH (ng/mL) 6.23 ± 4.05 6.66 ± 4.30 0.103 0.241 Basal FSH (IU/L) 6.45 ± 2.34 6.30 ± 1.60 0.072 0.441 Basal LH (IU/L) 7.48 ± 5.65 7.91 ± 4.86 0.082 0.365 AFC (n) 27.60 ± 13.46 29.16 ± 11.86 0.123 0.173 Infertility type 0.228 0.174   Primary infertility 242 (63.02%) 132 (68.8%)   Secondary infertility 142 (36.98%) 60 (31.2%) Infertility duration (years) 3.18 ± 1.85 3.11 ± 2.21 0.033 0.699 COH protocol 0.161   Long-acting GnRHa 244 (63.5%) 121 (63%) 0.011   GnRH-Ant 124 (32.3%) 69 (35.9%) 0.077   CC/HMG 9 (2.3%) 2 (1%) 0.101   PPOS 7 (1.8%) 0 (0) 0.193 Fertilization method 0.099 0.303   IVF 319 (83.1%) 152 (79.2%)   ICSI 65 (16.9%) 40 (20.8%) E 2 level before endometrial transformation (pg/mL) 235.06 ± 150.19 156.47 ± 83.71 0.301 <0.001 LH level before endometrial transformation (IU/L) 13.16 ± 9.10 18.36 ± 11.01 0.515 <0.001 P level before endometrial transformation (ng/mL) 0.18 ± 0.13 0.16 ± 0.14 0.077 0.377 EMT before endometrial transformation (mm) 9.98 ± 1.53 9.98 ± 1.55 0.005 0.953 Clinical pregnancy 293 (76.3%) 151 (78.6%) 0.056 0.599 Early miscarriage 30 (10.2%) 21 (13.9%) 0.107 0.251 Abbreviations : BMI, Body Mass Index; AMH, anti-müllerian hormone; FSH, Follicle-stimulating hormone; LH, luteinizing hormone; AFC, Antral follicle count; COH, Controlled Ovarian Hyperstimulation; GnRH, Gonadotropin-Releasing Hormone; CC, Clomifene Citrate; HMG, Human Menopausal Gonadotropin; PPOS, Progestin-primed ovarian stimulation; IVF, in vitro fertilization; ICSI, Intracytoplasmic Sperm Injection; E 2 : estrogen; P, progesterone; EMT, endometrial thickness. Multivariate Analysis for the Effect of Low-Dose Estradiol on the Clinical Outcome of HRT-FET Cycles Notes:* Adjust for: Female age; Male age; Infertility type; AMH; Basal LH; AFC. Abbreviations : aOR, odds ratio; CI, confidence interval. Characteristics of HRT-FET Cycles with Different Estradiol Dosage After Propensity-Score Matching Abbreviations : BMI, Body Mass Index; AMH, anti-müllerian hormone; FSH, Follicle-stimulating hormone; LH, luteinizing hormone; AFC, Antral follicle count; COH, Controlled Ovarian Hyperstimulation; GnRH, Gonadotropin-Releasing Hormone; CC, Clomifene Citrate; HMG, Human Menopausal Gonadotropin; PPOS, Progestin-primed ovarian stimulation; IVF, in vitro fertilization; ICSI, Intracytoplasmic Sperm Injection; E 2 : estrogen; P, progesterone; EMT, endometrial thickness.

With the increasing patient demands for scheduling flexibility and comfort, frozen-thawed embryo transfer (FET) has gained growing popularity. Its primary advantage lies in achieving pregnancy rates comparable to those of fresh embryo transfer while significantly reducing the risk of ovarian hyperstimulation syndrome (OHSS). 1–7 No significant superiority has been demonstrated among the various endometrial preparation protocols for FET cycles. The hormone replacement therapy (HRT) cycle is widely adopted due to its scheduling convenience and stable clinical pregnancy outcomes. 8 , 9 However, growing evidence suggests that HRT-FET may increase risks of pregnancy complications (eg., hypertensive disorders, placenta accreta, intrahepatic cholestasis) as well as low birth weight and SGA infants, possibly linked to abnormal estrogen (E 2 ) levels during HRT-FET cycles. 10 , 11 Previous studies indicate that endometrial receptivity, uterine spiral artery remodeling, and placental development are closely associated with serum E 2 levels, 10 , 12 and that high serum E 2 prior to fresh IVF transfer may reduce embryo implantation rates 13 and potentially increase the incidence of preeclampsia, fetal growth restriction, and low birth weight. 14 In HRT-FET cycles, we typically administer relatively high doses of exogenous estrogen to achieve stable serum E 2 levels and maintain adequate endometrial growth, while preventing cycle cancellation due to follicular development or premature ovulation. 15 , 16 Previous studies suggest that lower serum E 2 levels in HRT-FET cycles may negatively affect clinical pregnancy outcomes. Concurrently, our earlier research indicated that patients with lower serum E 2 levels (below 300 pg/mL) prior to endometrial transformation actually had higher clinical pregnancy rates. 17–19 Additionally, our previously published study showed that the dominant follicle development during HRT-FET cycles did not affect clinical pregnancy rates, early miscarriage rates, or live birth rates. 20 We recognize that clinical outcomes are influenced by many factors; therefore, multivariate analysis is essential. Hormone replacement therapy cycles are extensively utilized in our reproductive medicine center. In certain patients, the serum E 2 levels prior to endometrial transformation are notably higher than the natural physiological levels. Can we regulate the serum E 2 levels by adjusting the dosage of exogenous estradiol in HRT-FET cycles, thus minimizing the potential risks related to high-estrogen exposure without undermining the clinical pregnancy rates? However, current study on the application of low-dose estradiol during the HRT-FET cycle remains severely inadequate. We hypothesized that a lower starting dose of estradiol would not reduce clinical pregnancy rates compared with the conventional regimen. To tackle this issue, we carried out a retrospective analysis of the HRT-FET cycles conducted in recent years at the Reproductive Medicine Center of Nanjing Drum Tower Hospital, with the aim of exploring the impact of reducing the exogenous estrogen dosage in HRT cycles on clinical pregnancy outcomes. Notably, we innovatively investigated the impact of estradiol dosage on the clinical outcomes of HRT-FET, rather than merely examining the relationship between serum E 2 levels and clinical outcomes.

In prior practice, a relatively high fixed dose of exogenous estradiol was typically administered during HRT-FET cycles to attain stable serum estradiol levels and endometrial conditions. However, in our study, patients receiving a lower estradiol dose showed a higher clinical pregnancy rate in HRT-FET cycles than those receiving a higher dose. After adjusting for potential confounders, the clinical pregnancy rate was not significantly associated with the estradiol dosage. These findings suggest that a lower-dose strategy may be comparable to the conventional higher-dose protocol in terms of clinical pregnancy rate. Since the first successful case in 1983, FET has gained increasingly widespread application in assisted reproductive technology (ART). 22 The clinical pregnancy outcome of FET does not seem to be significantly affected by the choice of different endometrial preparation protocols. 23 Hormone replacement therapy cycles are more commonly used, mainly because they eliminate the need to consider the regularity of the patient’s menstrual cycle, enabling flexible scheduling of examinations and embryo transfers. 24 , 25 However, studies suggest that natural cycle FET may reduce the risk of maternal obstetric complications, such as hypertensive disorders of pregnancy, intrahepatic cholestasis of pregnancy, placenta accreta, and macrosomia. 10 , 11 , 23 , 26 This potential benefit might be due to the avoidance of exogenous estrogen and the presence of endogenous luteal function. A study showed that, compared to natural cycle FET, women undergoing estrogen replacement had significantly increased thrombin generation, which might increase their thrombotic risk. 27 Therefore, can we minimize the dose of estradiol in HRT cycles to potentially reduce related risks while achieving comparable clinical outcomes? Concerns regarding the reduction of estradiol dosage in HRT-FET cycles primarily focus on two potential issues: a drop in serum estradiol levels and the spontaneous development of a dominant follicle. Remohi J et al 18 reported that in oocyte donation FET cycles, a lower serum estradiol level (< 100 pg/mL) was sufficient to support normal embryo implantation. Conversely, excessively high serum E 2 levels during controlled ovarian stimulation in IVF cycles may lead to reduced clinical pregnancy rates 28 or adverse pregnancy outcomes. 29 A retrospective study by Fritz et al 17 involving 110 HRT-FET cycles showed that the mean serum E 2 level in cycles resulting in ongoing pregnancy/live birth was significantly lower than in those that did not, suggesting that elevated serum estradiol levels might negatively impact ongoing pregnancy and live birth rates. Sita et al 30 conducted a retrospective analysis of 509 HRT-FET cycles. Patients were classified into six groups according to their serum E 2 levels before endometrial transformation: 500 pg/mL. The results showed that the implantation and clinical pregnancy rates were generally comparable across the groups. However, a slight decrease in both the embryo implantation rate and the clinical pregnancy rate was noted in patients with estrogen levels surpassing 500 pg/mL. In a previous large-scale retrospective cohort study from our center, which included 10,209 HRT-FET cycles, smoothed curve fitting along with threshold effect analysis initially suggested that the clinical pregnancy rate significantly declined as the serum E 2 levels before endometrial transformation increased. Specifically, serum E 2 levels above 300 pg/mL were related to a substantial reduction in the clinical pregnancy rate. 31 Furthermore, a possible issue with lower estradiol doses is the emergence of a dominant follicle during the HRT cycle. A study demonstrated that the clinical pregnancy rate in the group with dominant follicle development was slightly higher, although not statistically significant, compared to the group without follicle development, while the miscarriage rate was significantly lower. 32 In line with this, a recently published retrospective study from our center also concluded that the development of a dominant follicle during HRT-FET cycles has no impact on the clinical pregnancy rate, early miscarriage rate, or live birth rate. 19 , 20 Collectively, the presented evidence supports the justification for reducing the estradiol dosage in HRT cycles. Previous studies have shown that in bilaterally oophorectomized rhesus monkeys, progesterone alone suffices to maintain endometrial receptivity and support normal pregnancy, suggesting that estrogen is not essential for luteal-phase receptivity. 33 Endometrial proliferation requires a certain level of E 2 ; however, once serum progesterone rises during the secretory phase, additional E 2 becomes unnecessary. Serum E 2 exhibits a threshold effect, exerting a “permissive action” rather than a dose‑dependent “regulatory action” on the endometrium. 34 High E 2 levels may impair endometrial function, leading to defective transformation, delayed glandular maturation, premature stromal changes, and aberrant expression of implantation‑related genes (eg., Lif, Ptgs2 ). 35 Moreover, elevated E 2 may restrict trophoblast invasion into uterine spiral arteries, compromising blood supply for implantation and placentation. 33 Remohi J et al 18 reported that in oocyte donation FET cycles, a relatively low serum E 2 level (<100 pg/mL) is sufficient to support normal embryo implantation. To date, research on low-dose estrogen replacement protocols remains limited. Davar et al 36 found that clinical outcomes in FET cycles were comparable between constant 2 mg and 6 mg estrogen regimens, suggesting that a lower preparation dose does not adversely affect outcomes. Similarly, another study in HRT cycles observed no significant differences in endometrial thickness or clinical pregnancy rates across different estrogen dose groups. 19 Our retrospective study aligns with these findings, showing that the clinical pregnancy rate in HRT‑FET cycles with a lower estradiol dose is comparable to that observed in patients receiving a higher dose. As previously noted, while the development of a dominant follicle during an HRT cycle does not affect the final clinical pregnancy outcome, it does increase the complexity of cycle monitoring and reduce the inherent convenience advantage of the HRT protocol. Our previous research analyzed the baseline characteristics of patients prone to dominant follicle development in HRT cycles. The results indicated a close association with the patient’s ovarian reserve parameters: patients with higher basal FSH levels, a lower AFC, and a shorter menstrual cycle length were more likely to experience spontaneous dominant follicle development. 20 Furthermore, previous studies have confirmed that in older patients with diminished ovarian reserve (manifested as a low AFC), reduced inhibin secretion leads to weakened negative feedback on FSH, thus facilitating the premature initiation of follicular development. 37 Synthesizing these findings, a low-dose estradiol protocol may be more suitable for relatively younger women or those with normal to high ovarian reserve. This approach aims to maintain favorable pregnancy outcomes while maximizing the convenience of the HRT cycle. This highlights the significant value of implementing individualized, stratified estrogen dosing strategies in clinical practice based on the patient’s ovarian reserve status. This study has several limitations. First, we only included HRT-FET cycles using Femoston and did not include patients receiving other forms of exogenous estrogen. Second, many patients at our center undergo pretreatment with long-acting GnRHa before receiving oral exogenous estrogen. The relationship between reducing exogenous E 2 dosage and clinical outcomes in this specific population requires further investigation. Secondly, since some patients included in the study did not complete the final follow-up, we lack relevant data on the live birth rate. We used only the clinical pregnancy rate as the primary study outcome, which slightly decreased the reliability of the outcome metric. Third, the choice of estradiol dose was not randomized but based on physician or center preference, introducing potential confounding by indication. Fourth, the sample sizes between the lower-dose and higher-dose groups were markedly disparate (1955 vs. 202), which may have reduced the statistical power to detect small but clinically meaningful differences. Furthermore, we did not conduct a sample power test, which carries the potential risk of reduced result reliability due to insufficient sample size. Additionally, while we strictly limited the study to patients receiving single D5 blastocyst transfer, specific embryo grading scores were not incorporated. Due to constraints within our center’s data system, it was challenging to directly retrieve data related to the corresponding oocyte retrieval cycle outcomes (such as the number of oocytes retrieved, number of usable embryos, etc)., which may have adversely affected the control of confounding factors in this study. Furthermore, we were unable to directly obtain data on pregnancy-related complications (such as hypertensive disorders of pregnancy, placenta accreta, neonatal birth weight, etc)., which may be associated with the estradiol dosage used. Moreover, despite using PSM to adjust for measured confounders, the possibility of residual confounding due to unmeasured or imperfectly measured factors remains. A primary limitation of this study lies in its retrospective design. Moreover, our final results are presented using the OR value, a relative effect metric that carries potential risks of exaggeration. These findings should be considered hypothesis-generating and require validation in prospective randomized trials. Higher-quality, larger-scale randomized controlled trials are needed to clarify the impact of low-dose estradiol administration during HRT cycles on clinical pregnancy outcomes (live birth rate, pregnancy complications, neonatal complications, etc). This would help confirm whether a strategy of using lower doses of exogenous estrogen can avoid excessive medication and associated risks while maintaining high clinical pregnancy rates.

In this retrospective analysis of HRT cycles, reducing the dosage of exogenous estradiol did not negatively affect the clinical pregnancy rate of single Day-5 blastocyst FET. These findings should be considered hypothesis-generating and require validation in prospective randomized trials.