Cumulative live birth rates of 31 478 untested embryos from 11 463 women challenge traditional recurrent implantation failure definitions.

Study question: Is there evidence of a plateau in the cumulative live birth rate (cLBR) after a certain number of consecutive transfers of untested embryos?

Summary answer: In our cohort of 11 463 women, the cLBR continues to increase with each additional transfer of an untested embryo, reaching 68.3% after six blastocyst transfers and 78.0% after 10 blastocyst transfers.

What is known already: While cumulative success rates in ART are rising, implantation failure remains a persistent challenge. The actual frequency of recurrent implantation failure (RIF) and whether RIF surpasses the inherent implantation potential of transferred embryos remains a matter of debate. A recent study reported a cLBR of 98% after five euploid blastocyst transfers, suggesting that most implantation failures are likely embryonic rather than endometrial. However, it remains unclear how these findings can be extrapolated to patients who did not undergo preimplantation genetic testing for aneuploidy (PGT-A). While theoretical models estimate cumulative implantation probabilities based on published blastocyst euploidy rates by female age, real-world data on cumulative success in routine clinical practice remain limited.

Study design, size, duration: This non-interventional retrospective cohort study included records of all completed IVF/ICSI cycles (including thus fresh and frozen-thawed transfers of one oocyte retrieval cycle) in women who underwent IVF/ICSI at the Ghent University Hospital between January 2010 and December 2022. After excluding treatments involving PGT, oocyte donation and surrogacy, or a mix of cleavage stage and blastocyst stage transfers, our dataset consisted of 11 463 women who underwent a total of 19 378 IVF/ICSI cycles, resulting in a total of 31 478 embryo transfers.

Participants/materials, setting, methods: The number of embryos transferred ('time') until achieving live birth ('event') was analysed using a Kaplan-Meier approach with inverse probability weighting (IPW). Additionally, logistic regression analysis was conducted to assess the predictive value of the number of previously transferred embryos on the live birth rates (LBRs) of the second and subsequent transfers, adjusting for female age, quality of previously transferred embryos, and stage of embryos transferred (cleavage stage versus blastocyst stage).

Main results and the role of chance: Kaplan-Meier estimates using an IPW approach showed cLBRs increasing from 51.1% (95% CI: 49.2-53.0%) after a third, up to 68.3% (95% CI: 64.6-72.0%) after a sixth and even as high as 78.0% (95% CI: 69.5-86.5%) after a tenth blastocyst transfer, respectively. As maternal age increases, higher numbers of blastocysts are required to achieve the same cLBR. Moreover, no age category achieves an 80% cLBR until after the transfer of eight blastocysts. Maternal age has a considerable effect, as illustrated by cLBR after the fourth blastocyst transfer of 68.9% (95% CI: 65.8-71.8%) for <35 years; 57.6% (95% CI: 50.4-64.8%) for 35-37 years; 42.9% (95% CI: 37.5-48.4%) for 38-40 years; 16.3% (95% CI: 10.7-21.8%) for 41-42 years; and 13.5% (95% CI: 3.2-23.7%) for >42 years, respectively. In the adjusted logistic regression analysis, the odds for achieving live birth are estimated to decrease for each additional embryo transferred; however, this effect is not statistically significant (OR = 0.91; 95% CI: 0.86-1.07). Female age, as expected, was a significant predictor of implantation rate with subsequent transfers (OR = 0.92; 95% CI: 0.91-0.93). Additionally, being assigned to a blastocyst transfer strategy rather than to a cleavage stage transfer strategy was also a significant predictor (OR = 1.34; 95% CI: 1.20-1.51), as was proportion of embryos classified as excellent or good quality based on predefined morphological criteria out of the total number of embryos previously transferred (OR = 1.21; 95% CI: 1.06-1.38). Implantation rate is also correlated with the response to stimulation (P = 0.016) and the blastocyst formation rate (P < 0.0001). There was no significant difference in LBR after an equal number of previously unsuccessful blastocyst transfers, depending on how many oocyte collection cycles it took to reach that number of blastocysts.

Limitations, reasons for caution: The results are limited by the observational retrospective design, and while regression analyses were adjusted for potential confounding factors, residual confounding may persist, particularly given the considerable heterogeneity in treatments. We did not exclude patients with factors associated with less favourable reproductive outcomes because our goal was to evaluate actual clinical practice.

Wider implications of the findings: Our data emphasize the potential for successful live birth even after multiple unsuccessful transfers. Factors such as age, embryo quality, response to ovarian stimulation, and rate of blastocyst formation influence outcomes. By addressing these multifaceted influences, our research provides valuable insights and a hopeful outlook for patients undergoing fertility treatment.

Study funding/competing interest(s): L.D. received a grant from the Agency for Innovation through Science (IWT SB-141441). The funder did not have any role in the study design; data collection, data analysis, and interpretation of data; the writing of the report; nor the decision to submit the paper for publication. The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Trial registration number: N/A.