Abstract In IVF, blastocysts developed on Day5-6 are typically selected for transfer, cryopreservation, or biopsy, while embryos that do not reach this developmental stage before 144 hours post-insemination (hpi) are often discarded. Recently, there has been an increased adoption of blastocyst culture aimed at enhancing embryo selection potential and supporting a single embryo transfer approach. Consequently, the clinical significance of Day7 blastocysts (embryos reaching this stage beyond 144 hpi) has become a prominent topic of discussion, particularly in the context of treating patients with poor prognosis. Numerous studies worldwide have reported lower competence for these embryos, as evidenced by poorer morphological quality, higher aneuploidy rates, and lower implantation rates following both untested and euploid transfer. Nonetheless, a notable proportion of these blastocysts are euploid and result in healthy live births. While their clinical utilization may entail lower efficiency per transfer, it ultimately increases the cumulative live birth rate per cycle. However, most studies have encountered challenges due to diverse definitions of Day7, varying clinical settings, culture strategies, incubators, and local regulations. Presently in IVF, leveraging time-lapse technology (TLT), artificial intelligence (AI), and preimplantation genetic testing for aneuploidies (PGT-A) could standardize and enhance Day7 blastocyst characterization and clinical assessment. Consequently, we designed a study exclusively in freeze-all PGT-A cycles with vitrified-warmed euploid single embryo transfer after continuous culture in time-lapse incubators, trophectoderm biopsy without zona pellucida drilling on day3, comprehensive chromosome testing, confirming embryo morphological assessments through an AI-powered software, and utilizing this tool also to automate developmental timing annotations. Our analysis confirmed that all blastocysts were biopsied only after reaching an expansion level suitable for the procedure (approximately 24,000 µm²) and consistent across all embryos clustered every 12 hours in 6 groups (<120, 120-132, 132-144, 144-156, 156-168, and >168 hpi). Day7 blastocysts accounted for approximately 15% of all blastocysts obtained over 8 years, derived from older oocytes, exhibiting poorer quality according to both Gardner’s score and AI score, slower development from early stages, and accumulating delays across various intermediate processes like blastocyst expansion from tEB to the time of biopsy. They were less euploid and resulted in a lower live birth rate per euploid transfer. However, only the latter association was significant when adjusted for confounders (e.g., blastocyst quality). Comparing only embryos biopsied before 144 hpi versus embryos biopsied after this threshold, neither the euploidy nor the live birth rate were significantly different, supporting the extension of embryo culture beyond Day6. Overall, we estimated that Day7 embryos contribute to a relative increase of 4.4%, 13.7%, and 5.2% in the number of patients who achieved a live birth, who were not pregnant after the transfer of faster embryos, and who have supernumerary transferable embryos after a live birth, respectively. The main limitations of our investigation were its retrospective nature and the fact that retrospective associations do not provide evidence of causation. There might be a bias towards poor prognosis patients being subject to the transfer of Day 7 embryos, for instance. Lastly, more data are required regarding the putative impact of extended embryo culture including post-natal outcomes (reassuring to date), as well as the cost-effectiveness of this clinical choice. Considering the increasing personalization and patient-centeredness of IVF, where permitted by local regulations, the decision to culture embryos to Day7 should be based on a careful evaluation of couples’ reproductive history and future prognosis, and should be subject to a patient-centered decision-making process. In the future, AI can be harnessed to support IVF practitioners by objectifying the assessment of poor-quality and slow-growing embryos with measurable information and reliable predictions, thereby mitigating the inherent subjectivity of embryologists’ decisions.
Read full abstract