Endometrial thickness and live birth rates after IVF: a systematic review.

There are many factors that influence the success of assisted reproductive technology (ART) programs, including maternal age, endometrial characteristics such as endometrial structure, endometrial thickness (ET), subendometrial blood flow, and the number of embryos available for transfer ( 1 – 3 ). While routine monitoring of ET during ovarian stimulation is not recommended, it can be measured by ultrasound examination during oocyte aspiration or on trigger day to detect a very thin endometrium that may affect the outcome of ART procedures ( 4 ). Over the years, there have been numerous publications on potential sonographic markers of endometrial receptivity ( 5 ). Although it remains controversial, ET is the most commonly used predictor of endometrial receptivity and live birth rate (LBR) in ART programs such as in-vitro fertilization (IVF) or intracytoplasmic sperm injection (ICSI) ( 1 , 6 , 7 ). Decreased endometrial receptivity is associated with decreased likelihood of pregnancy, spontaneous miscarriage, ectopic pregnancy, low birth weight, and low birth weight relative to gestational age ( 8 , 9 ). Pregnancy outcome data were similar in fresh and frozen cycles, including rates of clinical pregnancy, implantation, and fertility ( 10 , 11 ). Nevertheless, ET was a better predictor of endometrial receptivity in fresh in vitro fertilization (IVF) cycles than ET in frozen embryo transfer (FET) cycles ( 12 , 13 ). The most common factors for thin endometrium (up to 60%) have been associated with intrauterine surgery, which can damage the basal layer of the endometrium ( 14 , 15 ). In some women, the endometrium may also be thin in nature ( 16 ) or under the influence of steroid hormones ( 17 ). Transcriptome analysis can predict a thin endometrium by identifying genes with increased activity ( 18 ). The authors of this study found no published articles comparing clinical and epidemiological factors that might be associated with LBR in patients who went through fresh and FET cycles with thin endometrium. This study aims to systematically review the current literature on published studies with data on the clinical and epidemiological data of the patients with thin endometrium who underwent fresh and FET cycles to compare their LBR and possible risk factors.

A comprehensive search of Pubmed, Web of Science, ScienceDirect, Scopus and Google Scholar found 1939 nonduplicative titles. Further screening for relevant articles based on title and abstract yielded 837 records. 81 full-text articles were screened for eligibility, and 20 articles were included in the systematic review. The flowchart of study selection is shown in Figure 1 . The flowchart of study selection. The study design and characteristics of patients with thin endometrium are presented in Table 1 . Most of the studies were conducted in China (eleven). We also included studies from Turkey ( 35 ), Portugal ( 25 , 38 ), Argentina( 34 ), Israel ( 32 ), Switzerland ( 26 ), Canada ( 37 ), South Korea ( 27 ), the United Kingdom ( 2 ), and India ( 36 ). Seven studies were retrospective cohort studies, seven studies were retrospective, two studies were prospective-interventional, two studies were prospective, one study was longitudinal-experimental, and one study was prospective-interventional. Six studies reported fresh embryo transfers and 11 studies reported frozen embryo transfers. Three studies reported both fresh and frozen embryo transfers. All studies had a low risk of bias. Characteristics of the included studies. Abbreviations: BMI - body mass index; CPR – clinical pregnancy rate; ET – endometrial thickness; FET – frozen embryo transfer; hCG - human chorionic gonadotropin; LBR – live birth rate; N – number; PRP – platelet-rich TE – thin endometrium A total of 14 804 patients were included in 20 studies (mean sample size = 306 patients, range = 24 to 3350 patients). The mean age of the subjects ranged from 28.86 to 41.03 years. The mean BMI of the patients ranged from 20.93±0.97 to 26.2±2.7. Reported ET rates ranged from 15 mm. Clinical pregnancy rates ranged from 9.1% to 79.3%, and live birth rates ranged from 4.8% to 48.99%. In the different studies, ET was measured at different time points. ET was most commonly measured on the day of human Chorionic Gonadotropin (hCG) administration regardless of the embryo transfer setting (Fresh or FET). LBR ranged from 4.80% to 48.99% in fresh embryo cycles, and from 6.06% to 39.19% in FET cycles overall. Clinical pregnancy rates ranged from 9.09% to 61.49% in fresh embryo transfer cycles, and from 13.3% to 79.31% in FET cycles. In five of nine studies reporting Fresh cycles, the LBR was less than 20%. In all those five studies the CPR was less than 30%. In six of fourteen studies reporting FET cycles the LBR was less than 20%. There was only one study with a CPR above 30%. In the study with the lowest percentage of live births during fresh cycles, the mean age of the patients was 29.5 + 3.2 years, whereas the mean duration of infertility was relatively short – 3 years. In the study with the lowest percentage of live births during the FET cycles, most patients were 20.93±0.97 years on average. The risk factors and LBR reported by the studies are summarized in Table 2 . The full table with risk factors is presented in Supplemental Table 1 . Risk factors associated with thin endometrium. Abbreviations: CPR – clinical pregnancy rate; ET – endometrial thickness; FET – frozen embryo transfer; FSH - follicle-stimulating hormone; G-CSF - granulocyte colony stimulating factor; GnRH-a - gonadotrophin releasing hormone agonist; hCG - human chorionic gonadotropin; LBR – live birth rate; N – number; PCOS - polycystic ovary syndrome; PGD - preimplantation genetic diagnosis; SD – standard deviation.

IVF outcomes related to the thickness of the endometrium remain an urgent problem in the field of reproductive medicine. Inadequate endometrial receptivity accounts for two-thirds of failed implantations, while only one-third of failures depend on embryo quality( 39 ). In a meta-analysis by Kasius et al. ( 1 ) the rate of clinical pregnancy was found to be lower when the ET is <7.0 mm, at the same time it was not associated with the rate of sustained pregnancy and live birth. A meta-analysis by Gao et al. ( 40 ) examined the significant role of ET in pregnancy outcomes after IVF. The results of this study showed that women with lower ET had lower pregnancy and live birth rates than women with higher thickness, regardless of whether a fresh or frozen cycle was performed. In the study by Mahutte et al. ( 9 ) an increase in live births was also observed in fresh embryo transfer cycles with an increase in ET up to 10-12 mm, and in frozen cycles, LBR increased after 7-10 mm. According to a study by Liu et al. ( 41 ) rates of clinical pregnancy and live births decrease with each millimeter ET for fresh IVF cycles below 8 mm, and for FET IVF cycles below 7 mm. However, in this review, we see that there are studies in which rates for clinical pregnancies and live births remain quite relevant in patients with an ET of 4 to 6 mm. Analysis of the study showed that the relationship between ET and LBR, in patients who underwent a fresh or FET cycle depended on study design, ET thresholds, significant risk factors such as age, BMI, type of infertility, reproductive loss history, surgical interventions in the cavity of the small pelvis, ineffectiveness of IVF programs, hormonal profile including initial level of follicle-stimulating hormone (FSH) in blood serum, and ovulatory reserve. Possible reasons and limitations of the present study: only English-language studies were included; most studies were from the China region; retrospective study design was used in most studies; different ET thresholds, which in turn could significantly alter the correlation with pregnancy outcomes; different IVF procedure protocols in fresh or FET cycles. There are many studies in the databases, and many conflicting results, but a variety of risk factors have shown that ET can be used as a predictor of pregnancy and live birth in fresh or frozen cycles. Among risk factors for thin endometrium, a history of recurrent reproductive losses associated with surgical abortions, endometriosis in women with secondary infertility, as well as tubal factors of infertility have drawn attention.

IVF outcomes in patients with impaired endometrial receptivity depend not only on the state of the endometrium. Risk factors and ET significantly affect LBR in Fresh and FET cycles. According to the results of the systematic review, in women with impaired endometrial receptivity, it is important to apply a personalized approach with high-quality diagnostics and effective treatment methods to achieve adequate thickness with improved endometrial receptivity. Due to the methodological weaknesses of the included studies, further research is needed to evaluate the independent importance of ET, structure, treatment protocols and other factors.

The study protocol is registered with the PROSPERO International prospective register of systematic reviews (ID: CRD42022359149). The PROSPERO database was searched to identify the registry of studies that evaluated EMT and outcome after IVF treatment, and no similar studies were found. Therefore, the PubMed, Web of Science, ScienceDirect, Google Scholar, and Open Grey databases were searched between July 2022 and October 2022 for this review. The full search strategy for the databases is provided in the supplementary materials. After the initial selection of the studies from the database search, we performed a hand search of the literature based on the list of references of the selected articles from the database search. The Preferred Reporting Items for Systematic Reviews and Meta-Analyzes (PRISMA) methods were used in this review. Inclusion criteria for studies were: 1) cohort studies or randomized clinical trials; 2) included subfertile childbearing women undergoing IVF or ICSI (aged ≥18 years); 3) endometrial thickness was measured by transvaginal ultrasound; 4) reported either the proportion or number of patients with outcomes of interest: Live birth rates and clinical/epidemiological data of the patients; 5) were published in English from January 2014 to October 2022. The exclusion criteria: 1) publications that lacked the required information; 2) publications that repeated previously reported study results; 3) had a high risk of bias; 4) donor oocytes were used in the study. Database searches, selection of studies based on their inclusion/exclusion criteria, and data extraction were performed by two independent researchers (IK and GM) according to PRISMA guidelines ( 19 ). The authors were contacted by email up two times to obtain missing data. The list of required information was prepared and approved by all authors and included: first author, year, country, study design, setting, transfer of fresh or frozen oocytes, prevalence of patients with thin endometrium, factors associated with thin endometrium, LBR, ET in millimeters. The electronic databases Pubmed, Web of Science, ScienceDirect, Scopus, and Google Scholar were searched between July 2022 and October 2022 for articles describing EMT and LBR after IVF treatment. The search strategy included the following keywords: “thin endometrium IVF”; “thin endometrium invitro fertilization”; “thin endometrium pregnancy”; “thin endometrium live birth”. Search results were restricted using filters to publications published between 2014 and 2022. No restrictions were applied to the type of publication. The titles and abstracts of all identified studies were screened, and the full texts of the initially selected articles were read by two researchers (G.M. & I.K.). Both researchers independently extracted data from the full texts of the articles using a prepared data extraction form. A narrative review was conducted to summarize the results. According to the Cochrane Collaboration Network Risk Assessment Tool, biases include selection bias, performance bias, discovery bias, attrition (discontinuation) bias, and reporting bias ( 20 ). A bias risk assessment was performed independently by each of the two reviewers. When this was not possible, a discussion was held with the third author. The following assessments were used: low risk, high risk, or unclear (lack of information or uncertainty regarding the potential for bias).