Does endometrial thickness impact live birth rate following a frozen embryo transfer: outcomes of 30 676 euploid single embryo transfers

Abstract

STUDY QUESTION: Does endometrial thickness (ET) impact live birth rate (LBR) in patients undergoing single euploid frozen embryo transfer (FET)? SUMMARY ANSWER: Patients with the thinnest endometrial lining exhibit a decline in LBR at all centers, but the magnitude of the decline and the ET threshold at which it is identified is variable by center and cycle type. WHAT IS KNOWN ALREADY: Thin endometrium is thought to be an impediment to FET cycle success, and it is widely believed that pregnancy and LBRs are improved when the ET is >6-8 mm. However, evidence for this is limited and some studies report contradictory findings which indicate that ET does not significantly impact LBR. STUDY DESIGN, SIZE, DURATION: This is an international multicenter retrospective cohort study conducted between January 2017 and December 2022 at 25 different IVF centers in 3 countries and including a total of 30 676 cycles. PARTICIPANTS/MATERIALS, SETTING, METHODS: All FET cycles involved a single euploid blastocyst created with autologous oocytes. Endometrial preparation protocols were selected at the discretion of the physicians and the patients, and included programmed cycles, natural, and modified natural cycles (NC and mNC). The primary outcome was the LBR stratified by ET and cycle type. The distribution of ET measurements was assessed with histograms and quantile-quantile plots. Conditional density plots (CDPs) were utilized to determine the associations between ET measurement and LBR. Adjusted effect estimates of ET on LBR were assessed with multivariable logistic regression analyses, and receiver operating characteristics curves (ROC) were used to assess the performance of ET for predicting live birth. MAIN RESULTS AND THE ROLE OF CHANCE: There were 24 097 (78.6%) programmed cycles, 759 (2.5%) NCs, and 5820 (19.0%) mNCs included in the analyses. The median ET among all cycle types at all centers was 8.9 mm (9.0 mm in the USA, 8.7 mm in Spain, 8.0 mm in the UAE). When cycles from all centers were grouped together, CDPs showed a decline in LBR for ET <7 mm in both programmed and mNCs. Regression analyses demonstrated that in cycles with a lining <7 mm undergoing programmed cycles and mNC, odds of LBR were reduced by 22% [aOR 0.78 (95% CI 0.70-0.87), P <= 0.001] and 41% [aOR 0.59 (95% CI 0.49-0.72), P < 0.001], respectively. In patients undergoing NC, there was no ET threshold at which LBR was impacted and regression analysis demonstrated that LBR is not significantly impacted by ET <7 mm in patients undergoing NC [aOR 0.85 (95% CI 0.58-1.25), P = 0.41]. Sensitivity analyses were consistent with the overall analysis, Q-Q plots demonstrated that the distribution of ET measurements varied between the centers and the percentage of programmed cycles with ET <7 mm was lowest in the USA (2.6%) and Spain (5.2%), compared with the UAE (12%). Two models were developed to determine the prognostic value of ET for predicting live birth. The performance of the model with endometrial thickness (AUC: 0.597, interquartile range (IQR): 0.593-0.601) was not significantly higher (P = 0.052) than the model without it (AUC: 0.591, IQR: 0.586-0.595), suggesting that ET is not a strong predictor of LBR. LIMITATIONS, REASONS FOR CAUTION: The retrospective nature of the study design restricts the ability to establish causal relationships and fully control for variables not included in the analysis. WIDER IMPLICATIONS OF THE FINDINGS: This is a large multicenter study including single euploid FET cycles in order to effectively evaluate the impact of ET on LBR. These results provide new insight into the nuances of the relationship between ET and LBR at various international centers and within the context of three major FET protocols.