{"paper_id":"a0e23355-9027-4f48-97b8-b7fc3d0a3474","body_text":"Endometrial receptivity and embryo quality are the two main determinants of pregnancy\nsuccess in IVF/ICSI cycles ( Ma  et\nal. , 2017 ) . Despite technological advances, implantation\nfailure still occurs in approximately 30-50% of transfers using high-quality\nembryos. Implantation failure is considered a distinct and multifactorial process;\nhowever, successful embryo implantation is known to require synchronized\ncommunication between the embryonic tissue and the maternal endometrium. As such,\nendometrial receptivity is regarded as a rate-limiting step in in vitro\nfertilization (IVF) and has garnered increasing attention ( Arian  et al ., 2023 ; Wallach  et\nal ., 1992).\nAmong endometrial factors, endometrial thickness (EMT) remains one of the most widely\nused indicators of receptivity, although its predictive value remains controversial\n( Liu  et al ., 2025 ;  Lam  et al ., 2022 ;  Liu  et al ., 2018 ). Some studies\nassociate EMT with improved implantation ( Liu\n et al. , 2018 ;  Ma\n et al. , 2017 ;  Noyes\n et al. , 1995 ) while others report no correlation\n( Kasius  et al. , 2014 ;\n Lam  et al. , 2022 ).\nTherefore, the present study aimed to evaluate whether EMT measured before\nprogesterone initiation is associated with chemical pregnancy rates in artificial\nFET cycles conducted by a single physician.\n\nThis retrospective analysis was performed using all artificial FET cycles performed\nbetween January 2021 and December 2024 at a private fertility clinic in\nGoiânia, Brazil. Dates were extracted from electronic medical records. The\nstudy was conducted according to institutional policies on data confidentiality and\nreceived a waiver of ethical approval due to its retrospective design.\n(1) patients with indication for Assisted Reproduction treatment who\nunderwent TEC with current endometrial preparation protocols.\n(2) patients with cryopreserved blastocyst stage embryos.\n(1) patients with diagnosed and untreated uterine infertility factors\nsuch as septate uterus, intrauterine adhesions, adenomyosis, and\nendometrial polyps. Endometrial thickness was not used as a criterion\nfor cycle cancellation.\n(2) patients with cryopreserved cleavage stage embryos.\nSince this study did not involve therapeutic interventions or modifications to\nstandard IVF-TEC protocols, no additional approval from the institutional ethics\ncommittee was required.\nEndometrial preparation consisted of oral administration of estradiol valerate\nstarting on the second or third day of menstruation. This regimen stimulated\nendometrial proliferation and thickening and aimed to achieve a thickness of 7\nmm, measured by transvaginal ultrasound. In cases of suboptimal endometrial\nresponse, in addition to increasing the dose, transdermal estradiol was added.\nApproximately 10 to 16 days after estradiol administration, vaginal micronized\nprogesterone was started to prepare the endometrium for embryo transfer.\nSerum progesterone levels were measured approximately 24 hours before embryo\ntransfer, approximately 6±2 hours after the last vaginal dose of 200 mg\nof micronized progesterone. Results were available on the same day.\nPatients with serum progesterone levels <10 ng/mL received an additional\nsubcutaneous dose of 25 mg progesterone (Progestan Dex; Kocak, Turkey) once\ndaily for at least 42 days, starting on the day of embryo transfer. This\nsupplemental dose was administered at the same time each day, coinciding with\nthe time of the first injection. Luteal phase support was continued until\npregnancy was confirmed or discontinued on the day of the pregnancy test if it\nwas negative.\nThe endometrial assessment protocol remained consistent throughout the study\nperiod. Endometrial thickness (EMT) was measured by the second author, a medical\nspecialist at the fertility center, using two-dimensional transvaginal\nultrasonography. EMT was defined as the maximum distance between the echogenic\ninterfaces of the myometrium and endometrium and was measured in the midsagittal\nplane on the day before starting micronized progesterone.\nEndometrial morphology was classified into two main patterns: (1) a trilaminar\n(three-line) pattern, characterized by two outer hyperechoic lines representing\nthe endometrial-myometrial interface on the anterior and posterior uterine\nwalls, and two inner hypoechoic layers separated by a central hyperechoic line\nrepresenting the interface between the endometrial surfaces; and (2) a\nhomogeneous echogenic pattern. EMT referred to the peak thickness measured\nduring the estradiol phase (E2).\nEmbryos were thawed using the Vit Kit - Warm (FUJIFILM Irvine Scientific, USA),\nand between one and three embryos were transferred into the uterus using a\nGuardian Access ET catheter (Cook Incorporated, IN, USA). The best frozen\nblastocysts from each patient were selected for thawing. After embryo transfer,\nhormonal supplementation with progesterone was continued for 14 days until a\nblood pregnancy test was performed. Patients with a positive result continued\nhormonal supplementation until the 12th week of gestation.\nChemical pregnancy was confirmed by β-hcg testing (>25 mIU/ml)\napproximately two weeks after transfer.\nData were analyzed using BioEstat software (Belém-PA, Brazil). All tests\nwere two-tailed and  p <0.05 was considered statistically\nsignificant. Descriptive data are presented as median (range), mean ±\nstandard deviation (SD). Patients were divided into two groups: those who\nachieved a chemical pregnancy (Group A) and those who did not (Group B). A\nstatistical analysis using the Mann-Whitney U test was performed to assess the\nassociation between endometrial thickness and pregnancy rates. Regression\nanalysis was performed to assess the association between endometrial thickness\nand pregnancy rates. To evaluate the ideal cutoff point for endometrial\nthickness in predicting gestational outcomes, the ROC (Receiver Operating\nCharacteristic) curve was used and the endometrial thickness cut-off was\nperformed and statistical comparisons were made by the chi-square test\n(χ 2 ). Additionally, based on scientific evidence\nsuggesting that 7 mm as the minimum threshold for optimal endometrial\nreceptivity, patients were categorized into two groups: those with a thickness\n<7 mm and those with a thickness ≥7 mm and statistical comparisons\nperformed by the chi-square test (χ 2 ).\n\nBetween January 2021 and December 2024, a total of 522 FET cycles were performed with\ncumulative pregnancy rate 71,68%; 44 patients were excluded: seventeen because of\nexclusion criteria and twenty-seven for age matching, resulting in a final sample of\n478 patients. The median age was 34.9 years (range, 23-49) and the cumulative\nchemical pregnancy rate (PR), after the exclusion criteria, was 70.29%. EMT ranged\nfrom 3.6 to 20 mm (mean 8.82±2.4 mm). No statistically significant difference\nwas observed between the groups ( Figure 1 ).\nFigure 1 Distribution of patients according to endometrial thickness and hCG\nresult. Lab-Rep-UFG/Humana Medicina Reprodutiva (2021-2024). Mann-Whitney (U=22421.50)\nDistribution of patients according to endometrial thickness and hCG\nresult. Lab-Rep-UFG/Humana Medicina Reprodutiva (2021-2024).\nA simple linear regression analysis was performed to evaluate the relationship\nbetween the endometrial thickness and pregnancy rates. The model was not\nstatistically significant ( F (1,6)=1.18;  p =0.319),\nindicating no significant linear association between the variables. The coefficient\nof determination ( R 2 =0.1649) showed that only 16.5% of\nthe variance in the PR was explained by the EMT. The regression coefficient was not\nsignificant ( b =-1.205;  p =0.3182; 95%\n CI:  -3.914 to 1.504), suggesting that EMT variations do not\nsignificantly predict variations in the PR ( Figure\n2 ).\nFigure 2 Endometrial thickness linear regression curve. Lab-Rep-UFG/Humana\nMedicina Reprodutiva (2021-2024).\nEndometrial thickness linear regression curve. Lab-Rep-UFG/Humana\nMedicina Reprodutiva (2021-2024).\nTo assess the endometrial thickness cut-off, a ROC curve analysis was performed\n( Figure 3 ) with an area under the curve\n(AUC)=0.55 and sensitivity and specificity values of 0.542 and 0.518, respectively.\nBased on the ROC analysis, an optimal cut-off value of 8.3 mm was identified.\nFigure 3 Endometrial thickness ROC curve. Lab-Rep-UFG/Humana Medicina Reprodutiva\n(2021-2024). Cut-off: 8.3 Distance (d): 0.665\nEndometrial thickness ROC curve. Lab-Rep-UFG/Humana Medicina Reprodutiva\n(2021-2024).\nBased on cut-off, patients were subsequently stratified into two groups: Group 1 with\nendometrial thickness≤8.3 mm and Group 2 with thickness > 8.3 mm. No\nstatistically significant difference in pregnancy rates was observed between the\ngroups ( p =0.4944) as shown in  Table\n1 .\nPregnancy rates below and above 8.3 mm endometrial thickness\nChi-square ( p <0.05)\nConsidering scientific evidence that supports 7 mm as the minimum threshold for\noptimal endometrial receptivity, a chi-square test was performed to compare\npregnancy rates between two groups stratified by endometrial thickness: < 7 mm\nand ≥ 7 mm ( Table 2 ).\nPregnancy rates below and above 7.0 mm endometrial thickness\nChi-square ( p <0.05)\nIn this study population, no statistically significant difference in pregnancy rates\nwas observed between the groups ( p =0.1036).\n\nThe relationship between EMT and pregnancy outcomes has long been a subject of\ndebate. Despite numerous meta-analyses and large-scale studies, the controversy\nremains unresolved. Notably, there is a significant discrepancy between the findings\nof prospective and retrospective studies.\nMany authors have reported significant differences in pregnancy rates above and below\nthreshold EMT values ranging from 7 to 10 mm ( Liu\n et al ., 2025 ). Several studies suggest that an\nadequately thick endometrium is associated with improved implantation potential,\nwhereas thin endometrial linings are often correlated with reduced implantation\nrates and lower pregnancy success ( Tomic  et\nal. , 2020 ). Conversely, excessively thick endometria have\nalso been linked to unfavorable reproductive outcomes ( Xu  et al ., 2022 ). Our findings suggested that\ndifferent EMT intervals revealed no statistically significant differences,\nsuggesting that EMT was not a reliable predictor of pregnancy in this study\npopulation ( Lam  et al. , 2022 ;\n Noyes  et al. , 1995 ;\n Shakerian  et al. , 2021 ;\n AYUSTAWATI  et al. ,\n2002 ).\nSome studies report a linear positive association between EMT and pregnancy outcomes\n( Al-Lamee  et al. , 2024 ;\n Liao  et al. , 2022 ), while\nothers propose a nonlinear relationship ( Lam\n et al. , 2022 ;  Shakerian  et al. , 2021 ). Inconsistent findings,\nreliance on arbitrary cutoff values (e.g., <6, <7, or <8 mm) and a lack of\nstandardized definitions further complicate the understanding of EMT’s role in\npredicting pregnancy outcomes. Despite various cutoff values have been proposed to\ndefine a “thin” endometrium, with most studies setting this threshold at less than\n7-8 mm on the day of ovulation trigger. However, reported cutoff values have ranged\nfrom < 6 mm ( Mathyk  et al. ,\n2023 ) to >14 mm ( El-Toukhy  et\nal. , 2008 ).\nOur findings suggest that EMT alone may not be a reliable predictor of pregnancy in\nartificial FET cycles. This aligns with recent meta-analyses reporting weak or no\ncorrelation between EMT and live birth rates ( Lam\n et al. , 2022 ;  Shakerian  et al. , 2021 ). Although a minimal threshold\nof 6-7 mm is often cited, our data indicate satisfactory pregnancy outcomes even in\npatients with thinner linings.\nIn contrast, a study involving 743 frozen-thawed embryo transfers using autologous\nembryos found a lower pregnancy rate in women with an EMT between 7 and 8 mm\ncompared to those with an EMT greater than 8 mm ( Martel  et al. , 2021 ). Similarly,  El-Toukhy  et al.  (2008)  analyzed 768\nconsecutive medicated frozen embryo replacement cycles and demonstrated that EMT\nbetween 9 and 14 mm was associated with higher implantation and pregnancy rates\ncompared to ET of 7-8 mm. Ning-Zhao Ma and collaborators ( Ma  et al. , 2017 ) stratified patients into\nthree groups based on EMT (group A: ≤8 mm; group B: 9-14 mm; group C:\n≥15 mm) and observed that live birth rate, clinical pregnancy rate, early\nmiscarriage rate, and ectopic pregnancy rate were all influenced by EMT at the day\nof hCG administration, with outcomes improving as EMT increased.\nNational data from autologous in IVF and FET cycles in Canada showed that live birth\nrates plateau after an EMT of 7-10 mm. However, EMT below 6 mm was clearly\nassociated with significantly reduced live birth rates in both fresh and frozen\nembryo transfer cycles ( Mahutte  et\nal. , 2022 ). Interestingly, in our study, patients with EMT\n≤6 mm still exhibited good pregnancy rates, this suggests that, in the\npresent analysis, endometrial thickness did not play a determining role in pregnancy\noutcomes.\nAs previously discussed, several studies have suggested that EMT below certain\ncut-off negatively predicts pregnancy, often leading to cancellation or postponement\nof embryo transfer cycles. However, cut-off values commonly used to define a thin\nendometrium are largely arbitrary and lack robust biological justification. As a\nresult, the determination of an optimal threshold remains uncertain, thereby\ncomplicating comparisons across studies. Current evidence regarding the relationship\nbetween EMT and pregnancy outcomes remains conflicting and insufficient to justify\ndefinitive clinical decisions. Such decisions carry significant clinical, financial,\nand ethical implications ( Shakerian  et\nal. , 2021 ). Our data indicate EMT alone may not be a\nreliable predictor of pregnancy rates and that “thin” endometrium can still be\nassociated with good pregnancy rates. Therefore, the common clinical practice of\ncanceling embryo transfers when EMT is less than 7 mm may be unwarranted, as it\npotentially denies patients the opportunity for pregnancy. Nevertheless, confounding\nvariables such as embryo quality, serum progesterone, and endometrial morphology\nwere not controlled, which may influence outcomes. Prospective studies are needed to\nclarify whether EMT independently affects implantations.\n\nIn conclusion, endometrial thickness was not identified as a significant predictor of\nchemical pregnancy in FET cycles. Our data indicate that embryo transfers should not\nbe canceled solely based on EMT measurements, as favorable outcomes can occur even\nin cases with thinner endometria.","source_license":"CC-BY-4.0","license_restricted":false}