Association between endometrial echo on transfer day and pregnancy outcomes in thawed embryo transfer: a retrospective cohort study across different preparation protocols

Abstract

Objective

This study aimed to investigate the relationship between endometrial echo and pregnancy outcome in patients undergoing thawed embryo transfer and explore the effect of different endometrial preparation schemes on endometrial echo.

Methods

A retrospective analysis was conducted on data from 2910 patients who underwent freeze-thaw embryo transfer (FET)-assisted pregnancy in the reproductive medicine centre of our hospital from January 2019 to March 2024. Based on the endometrial echo on the transplantation day, the patients were divided into two groups: the endometrial echo uniform group and the endometrial echo uneven group. Based on the endometrial preparation protocol, they were divided into the natural cycle (NC) group, hormone replacement cycle (HRT) group, and downregulated combined HRT (GnRHa+HRT) group. The general data, pregnancy outcome and endometrial echo of those undergoing different endometrial preparation protocols were compared.

Results

The clinical pregnancy rate in the homogeneous endometrial echo group was significantly higher than that in the non-homogeneous endometrial echo group. The equalisation rate of endometrial echo in the NC group and GnRHa+HRT group was significantly higher than that in the HRT group (p<0.05). Binary logistic regression analysis revealed that homogeneous endometrial echo was associated with a significantly improved clinical pregnancy rate of patients with FET. After adjusting for confounding factors, we found that compared with the HRT endometrial preparation protocol, NC and GnRHa+HRT endometrial preparation protocol significantly improved homogeneous endometrial echo.

Conclusion

Non-homogeneous endometrial echo on the transplantation day was associated with a decreased clinical pregnancy rate of FET. The endometrial preparation protocol of the NC and GnRHa+HRT cycle can improve homogeneous endometrial echo. In FET, clinicians should develop a personalised endometrial preparation protocol based on patients’ situations.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• This was a single-centre retrospective study based on factors affecting pregnancy outcomes after thawed embryo transfer.

• More subjects were included in this study, with high clinical relevance.

• The effect of confounding factors was controlled using the multivariate logistic regression method.

Introduction

The incidence of infertility is nearly 18% in China.¹ The development of human-assisted reproductive technology (ART) has helped many infertile couples to obtain offspring. Freeze-thaw embryo transfer (FET) technology has become an important component of ART.² Embryo quality and endometrial condition are two major factors affecting the outcome of pregnancy.³ The endometrium is the site of implantation and embryo development. Endometrial receptivity is an important factor affecting embryo implantation and development. In clinical practice, endometrial thickness and endometrial echo are commonly used to evaluate endometrial receptivity. However, Gingold et al⁴ showed that endometrial thickness has a low predictive ability for clinical pregnancy. Endometrial echo under colour ultrasound can reflect endometrial proliferation and stromal decidualisation. In the late stage of proliferation, the endometrium shows a ‘three-line sign’ (a strong echo line in the outer layer and centre, and a low echo area between the outer layer and uterine line). In the secretory stage, the endometrial gland cavity is full of secreted glycogen under the action of progesterone, during which the membrane transforms into uniformly high echo, the uterine line disappears and the outline becomes clear.^{5 6} Endometrial echo may represent endometrial blood perfusion, hormone receptor distribution, and other factors, reflecting the functional status of the endometrium. Homogeneous endometrial echo usually indicates that the endometrium is more receptive and more conducive to the adhesion, invasion and implantation of the embryo. A suitable endometrial preparation programme can improve endometrial receptivity, which is the key factor for the success of FET. However, there is no consensus on the endometrial preparation programme of FET.⁷ Previously, most studies have focused on the effect of endometrium morphology on pregnancy outcomes on the human chorionic gonadotropin (hCG) day of the ovarian stimulation cycle, and few studies have focused on the effect of endometrial morphology on pregnancy outcomes on the transplantation day of FET patients. We compared the pregnancy outcome of patients with different endometrial echoes on the transplantation day to investigate the relationship between endometrial echo on the transplantation day and pregnancy outcome. We also compared the endometrial echo of patients undergoing different endometrial preparation protocols with FET to explore the effect of different endometrial preparation protocols on endometrial echo.

Materials and methods

Participants

This study was started in July 2024. In total, 2910 patients undergoing FET-assisted pregnancy were collected from the reproductive medicine centre of our hospital from January 2019 to March 2024. Based on the endometrial echo of transplantation day, patients were divided into the following groups: uniform endometrial echo group (1970 cases) and uneven endometrial echo group (940 cases). Based on the endometrial preparation protocol, patients were divided into the natural cycle (NC) group (723 cases), hormone replacement cycle (HRT) group (1147 cases) and downregulated combined HRT group (1040 cases). The inclusion criteria were as follows: (1) undergoing FET-assisted pregnancy, (2) complete clinical data, (3) age ≤40 years, (4) no serious organic diseases and (5) no immune diseases. The exclusion criteria were as follows: (1) chromosomal abnormality on either side, (2) repeated implantation failure, (3) endometriosis, adenomyosis, uterine fibroids, (4) intrauterine polyps and intrauterine adhesions, (5) uterine malformation, (6) hydrosalpinx, (7) recurrent abortion and (8) intrauterine peristalsis on the transplantation day. This study was approved by the ethics committee of our hospital (KY2024292).

Research methods

Endometrial preparation protocol

Natural cycle group

Vaginal ultrasound was conducted on the 9–11 days of the menstrual cycle to monitor follicles and endometrium, or the patient self-measured luteinising hormone (LH) in urine. Follicle monitoring was conducted when the colour was red. When the diameter of the dominant follicle was >16 mm, vaginal ultrasound was conducted daily to monitor follicles, and blood oestradiol (E2), progesterone (P), and LH levels were monitored at the same time. When follicle diameter was >18 mm and endometrial thickness was >8 mm in vaginal ultrasound, subcutaneous injection of recombinant hCG 250 ug (Aizer, Germany, Merck Serono AG) was conducted, and vaginal ultrasound was repeated the next day. The day of ovulation was determined, and luteal support was conducted on the day of ovulation. Ovulation day is considered day 0, and the recovery and transplantation of cleavage embryos were conducted on the third day after ovulation, while the transplantation of blastocysts was conducted on the fifth day after ovulation.⁸ In the HRT group, vaginal ultrasonography was conducted on the second to fourth day of the menstrual cycle, and oestradiol valerate tablets were taken orally (Benjale, Bayer, Germany) 3 mg, two times per day after administration, and vaginal ultrasonography and E2 and P examination were conducted 14 days after administration. If the endometrium was type A, thickness ≥8 mm, E2 level ≥150–200 pg/mL, the endometrial transformation was performed with luteal support. The transformation lasted for 4 days, and the recovery and transplantation of cleavage embryo was conducted on the fifth day. The transformation lasted for 6 days, and the transplantation of blastocyst was conducted on the seventh day. In the GnRHa+HRT group, patients with no sinus follicles >8 mm in both ovaries were subcutaneously given 3.75 mg of GnRHa (Boenokang, Beijing, Boente Pharmaceutical) on the first or second day of the menstrual cycle, and re-examination was conducted 28 days later. After meeting the standard of reduction adjustment (E2<50 pg/mL, FSH<5 mIU/mL, LH<5 mIU/mL, etc). P <1 ng/mL, endometrial thickness <5 mm, HRT was used for endometrial preparation, using the same method as HRT.

Freeze-thaw embryo transfer

The vitrification freezing technique was applied to freeze embryos, and when the endometrial preparation reached the standard, the frozen embryos were resuscitated and their survival and quality were assessed. Cleavage stage embryo grading standard reference Huang Guoning, Editor in Chief of the Assisted Reproduction Laboratory Technology, Assisted Reproductive Technology Institute (Society for Assisted Reproductive Technology, SART2010), AlphaExecutive and ESHRE Embryology Professional (2011). As for the scoring method of embryo blastomere stage in group B, high-quality embryos were regarded as embryos with grade I and grade II blastomeres with 7–9 cells on day 3 from the 2PN source. The Gardner scoring system was used to score blastocysts,⁹ and high-quality embryos were regarded as blastocysts of 4BB grade and above. Transplantation was conducted 2–4 hours after resuscitation under the guidance of conventional ultrasound.

Criteria for adjudicating pregnancy outcome

hCG positive: 2 weeks after transplantation, serum hCG (β-hCG) was detected, β-hCG >5 IU/L, hCG positive rate = the number of hCG-positive patients/the number of transplant cycle patients×100%; Clinical pregnancy: the pregnancy sac (intrauterine or extrauterine) was detected using vaginal ultrasound 4–5 weeks after transplantation. Clinical pregnancy rate = the number of patients with clinical pregnancy/the number of patients in transplant cycle×100%. Early abortion: termination of pregnancy within 12 weeks of gestation when the embryo or fetus is not still viable. Early abortion rate = the number of spontaneous abortion cycles/the number of clinical pregnancy cycles×100%.

Criteria for endometrial morphology

In this study, the endometrial echo on the transplantation day was verified by two senior physicians who were experts at colour ultrasonography. After FET transformation, the endometrial secretory phase was changed under the action of progesterone, and the endometrial glandular cavity was full of secreted glycogen. In this phase, the endometrial layer exhibited uniform high echo, the uterine line disappeared, and the contour was clear, which was included in the endometrial homogeneity group (figure 1). On the contrary, it belongs to the uneven group (such as local intrauterine low echo area (figure 2), enhanced echo area (figure 3), mixed echo area (figure 4).

Figure 1. Endometrial homogeneous hyperecho: the endometrium is homogeneously hyperechoic, the uterine line disappears and the outline is clear.

Figure 2. Endometrial echo is locally weakened.

Figure 3. Endometrial midline enhanced echo.

Figure 4. Endometrial mixed echo: the endometrium presents a mixed echo of hyperecho, locally weakened echo and enhanced echo.

Statistical analysis

SPSS V.26.0 software was used for data analysis. Qualitative data are presented as numbers (percentages), and measurement data are presented as mean±SD (${\displaystyle \overline{x}\pm s}$). Qualitative data were analysed using the χ² test, and differences in measurement data were analysed using Analysis of Variance (ANOVA). The correlation between each factor and pregnancy outcome was analysed using binary logistic regression. p<0.05 was considered statistically significant.

Patient and public involvement

Patients and/or the public were not involved in this study.

Comparison of general data and pregnancy outcomes

The clinical pregnancy rate in the homogeneous endometrial echo group was significantly higher than that in the non-heterogeneous endometrial echo group (p<0.05). There were no significant differences regarding other general conditions and pregnancy outcomes between the two groups (p>0.05) (table 1).

Table 1. Comparison of general conditions and pregnancy outcomes between the two groups.

Category		Endometrial echo in the homogeneous group	Endometrial echo in the non-uniform group	P value
Number of cycles		1970	940
Age (years)		31.61±4.15	31.21±4.31	0.187
Infertility (years)		4.37±3.33	4.33±3.37	0.895
Infertility type	Primary infertility	857 (43.5%)	1113 (56.5%)	0.388
	Secondary infertility	393 (41.8%)	547 (58.2%)
Graft intima thickness (mm)		10.09±2.30	10.18±2.36	0.104
Body mass index (kg/m2)		22.43±3.18	22.42±3.14	0.909
Single cleavage embryo transfer rate (%)		92 (4.7%)	45 (4.8%)	0.889
Double cleavage embryo transfer rate (%)		756 (38.1%)	383 (40.7%)	0.175
Polyblastic embryo transfer rate (%)		12 (0.6%)	8 (0.9%)	0.460
Single blastocyst transfer rate (%)		177 (9.0%)	85 (9.0%)	0.959
Double blastocyst transfer rate (%)		939 (47.7%)	419 (44.6%)	0.118
Quality embryo transfer rate (%)		1514 (76.9%)	742 (78.9%)	0.208
Human chorionic gonadotropin positive rate (%)		1262 (64.1%)	581 (61.8%)	0.238
Clinical pregnancy rate (%)		1059 (53.8%)	465 (49.5%)	0.030
Early abortion rate (%)		231 (21.8%)	91 (19.6%)	0.328

BMI, weight (kg)/height (m2)

Logistic regression analysis for the effect of endometrial echo on pregnancy outcomes

Logistic regression analysis showed that compared with homogeneous endometrial echo, non-homogeneous endometrial echo did not significantly alter the chance of hCG-positivity (OR 0.908 (95% CI 0.773 to 1.066), p value: 0.238) and early abortion (OR 0.873 (95% CI 0.666 to 1.146), p value: 0.328), but decreased the chance of clinical pregnancy (OR 0.842 (95% CI 0.721 to 0.984) p value: 0.030). The results indicated that the clinical pregnancy rate in the uniform endometrial echo group was significantly higher than that in the non-uniform endometrial echo group (p<0.05) (table 2).

Table 2. Binary logistic regression analysis with endometrial echo as the dependent variable.

Pregnancy outcome	OR	95% CI	P value
Human chorionic gonadotropin positivity	0.908	0.773 to 1.066	0.238
Clinical pregnancy	0.842	0.721 to0.984	0.030
Early abortion	0.873	0.666 to 1.146	0.328

The values that are statistically significant are mentioned in bold.

Relationship between the endometrial preparation regimens of different FETs and endometrial echo on the transplantation day

The endometrial echo homogeneous rate of the NC group and GnRHa+HRT group was significantly higher than that of the HRT group (p<0.05), but there was no significant difference between the NC group and the GnRHa+HRT group (p>0.05) (table 3).

Table 3. Endometrial echo in the three groups.

FET endometrial preparation protocol	NC	HRT	GnRHa+HRT	P value
Homogeneous endometrial echo	508 (70.3%)	713 (62.2%)	749 (72.0%)	<0.001*
Non-homogeneous endometrial echo	215 (29.7%)	434 (37.8%)	291 (28.0%)

Note: GnRHa+HRT group, NC group, vs HRT group, ✱P; The values that are statistically significant are mentioned in bold.

^*p<0.05.

FETfreeze-thaw embryo transferHRThormone replacement cycleNCnatural cycle

Independent risk factors for endometrial echo

Univariate logistic regression analysis showed that endometrial preparation protocol and age significantly affected endometrial echo on the transplantation day (p<0.05). Multivariate logistic regression analysis was conducted on the above variables. After adjusting for relevant confounding factors, the corrected OR values of the HRT group compared with the NC group and the GnRHa+HRT group were 1.416 (95% CI 1.159 to 1.730) and 1.535 (95% CI 1.280 to 1.842). The results indicated that the endometrial echo homogeneous rate of the NC group and the GnRHa+HRT group was significantly higher than that of the HRT group (p<0.05) (table 4).

Table 4. The results of binary logistic regression analysis with endometrial echo as the dependent variable.

Variable	OR	95% CI	P value
Endometrial preparation protocol (the HRT group compared with the natural cycle (NC) group)	1.416	1.159 to1.730	0.001
Endometrial preparation protocol (the GnRHa+HRT group compared with the NC group)	0.922	0.748 to 1.137	0.449
Endometrial preparation protocol (the HRT group compared with the GnRH-a+HRT group)	1.535	1.280 to1.842	<0.001
Age (years)	0.983	0.965 to 1.002	0.074

The values that are statistically significant are mentioned in bold.

HRThormone replacement cycle

Discussion

FET is an important means to increase the success rate of pregnancy in ART, which refers to the resuscitation and transplantation of frozen embryos. It can improve the cumulative pregnancy rate and reduce the effect of high oestrogen levels on fresh embryo transplantation, endometrial susceptibility and the incidence of ovarian hyperstimulation syndrome.² Embryo quality and endometrial condition are two major factors that affect the outcome of pregnancy after FET. In this study, the biochemical pregnancy rate in the homogeneous endometrial echo group was significantly lower than that in the heterogeneous endometrial echo group. In addition, the clinical pregnancy rate was significantly higher than that in the heterogeneous endometrial echo group, which was consistent with the results of Hou et al¹⁰ The endometrium morphology changes periodically during the menstrual cycle. The endometrium echo is divided into three types: low echo, equal echo and high echo. In the luteal phase, the endometrium becomes decidualised under the action of progesterone, transforming the endometrium from the proliferative phase to the secretive phase. After deciduating, the endometrium can produce various bioactive substances, such as prolactin (PRL) and insulin-like growth factor binding protein-1 (IGFBP-1), to promote the adhesion and invasion of the embryo. At this stage, the endometrium exhibits uniformly high echo, the uterine cavity line disappears, and the contour becomes clear. The endometrial echo may be uneven under colour ultrasound when the endometrial decidualisation is incomplete.^{11 12} Endometrial decidualisation is the key to the successful implantation of embryos.¹³ Endometrial decidualisation can reduce endometrial receptivity, increase the difficulty of embryo implantation, reduce the clinical pregnancy rate, and even affect embryonic development after implantation, which is related to infertility and recurrent abortion.14,17 In the fully developed luteal phase, endometrial glands are more enlarged and curved, and the glycogen reserves in the endometrial gland are more adequate, which is more conducive to the implantation and development of fertilised eggs and the maintenance of pregnancy.¹⁰

Clinically, the following endometrial preparation schemes are used for FET: NC, excretion promotion cycle, HRT and downregulation combined with HRT.⁷ NC, HRT and downregulation combined with HRT are the most commonly used endometrial preparation schemes in our hospital. In this study, the endometrial echo equalisation rate in the NC and GnRHa+HRT groups on the transplantation day was significantly higher than that in the HRT group (p<0.05); however, there was no significant difference between the NC group and the GnRHa+HRT group (p>0.05). The endometrial echo equalisation rate in the GnRHa+HRT group on the transplantation day was higher than that in the NC group. The NC programme endometrial development is the physiological state of the body. Through endogenous hormone regulation, the NC programme can better grasp the endometrial implantation window, leading to endometrial and embryonic development synchronisation. The HRT programme uses exogenous oestrogen and progesterone to simulate the bodily changes associated with sex hormones during an NC to control endometrial development. The drug programme is flexible and convenient, and the embryo transfer time can be controlled effectively. However, this programme cannot guarantee the complete inhibition of the pituitary gland, and early LH peak and increased progesterone levels may affect endometrial decidualisation and decrease endometrial receptivity.¹⁸ A higher endometrial uniformity rate was observed in the GnRHa+HRT endometrial preparation scheme because the scheme uses GnRHa to regulate the pituitary gland, and GnRHa can directly affect the endometrium by regulating the expression of enzymes and cytokines.¹⁹ An J et al¹⁵ found that the receptivity of endometrium markers (leukaemia inhibitory factor, vascular endothelial growth factor, homeoboxA10 (HOXA10), HOXA11, integrin β3, osteopontin) and endometrial decidualisation markers (PRL and IGFBP-1) were upregulated in patients who used GnRHa during endometrial preparation of FET, suggesting that endometrial transition from the proliferative phase to the secretory phase was greater and endometrial receptivity was higher in patients treated with GnRHa. The pinocytoid process is a specific morphological marker for endometrial receptivity or opening of the planting window.²⁰ GnRHa can enhance the development and maturation of the pinocytoid process and improve endometrial receptivity.²¹ Other studies have found that GnRHa binds to GnRH receptors in the endometrium and improves endometrial receptivity by upregulating the expression of adhesion molecules in endometrial cells.²² In addition, GnRHa can downregulate inflammatory factors in the pelvic microenvironment, enabling better implantation of embryos.^{23 24}

In summary, this study indicated that the non-homogeneous endometrial echo on the transplantation day can decrease the clinical pregnancy rate of FET. Furthermore, this study revealed that the endometrial preparation programme of the NC and GnRHa+HRT cycle can increase the homogeneous rate of endometrial echo and improve endometrial receptivity. For patients with FET-assisted pregnancy and endometrial third-line signs on the conversion date, NC and GnRHa+HRT cycle should be considered for endometrial preparation. In the presence of sufficient time and funds, the GnRHa+HRT cycle can be considered to better prepare the endometrium and optimise the outcome of pregnancy. This study was a retrospective single-centre study, and the included subjects all showed third-line endometrial signs on the conversion date. A multicentre prospective randomised controlled study can be conducted in the future, and patients who do not have third-line endometrial signs on the conversion date should be included in such studies to more comprehensively analyse the effects of different endometrial preparation programmes on the endometrium of patients with FET. Basic experiments can also be conducted to unravel the specific molecular mechanisms of different endometrial preparation programmes in improving endometrial echo, providing new ideas for optimising the outcome of pregnancy among patients receiving FET.

Data availability statement

Data are available upon reasonable request.