Late-stage vacuolization as a compaction-dependent prognostic marker for day 4 embryo selection.

graphical abstractABSTRACT: This study aimed to evaluate the prognostic value of late-stage vacuolization in day 4 embryos relative to compaction status. We retrospectively analyzed 9,779 embryos from 2,409 patients undergoing fresh D4 transfers between 2018 and 2024. Embryos were stratified by compaction status (full, partial, and non-compaction) and assessed for vacuolization. Predictive performance for blastocyst formation was evaluated using ROC curves and logistic regression. The results showed that late-stage vacuolization correlated with higher maternal and paternal age, increased gonadotropin (Gn) dosage, and lower anti-Müllerian hormone (AMH)/estradiol (P < 0.001), but comparable oocyte yield and intracytoplasmic sperm injection (ICSI) rates (P > 0.05). False-compaction subgroup (77.6% of vacuolized embryos) showed reduced transferable blastocyst (38.7 vs 53.3%, P = 0.021), high-quality blastocyst (9.8 vs 23.4%, P = 0.001), and day 5 blastocyst rates (20.6 vs 35.1%, P = 0.007). A significant interaction was observed between late-stage vacuolization and compaction status affecting both transferable and high-quality blastocyst formation (P for interaction < 0.001). In particular, late-stage vacuolization significantly increased the rates of transferable (OR = 3.34, 95% CI: 2.54-4.39, P < 0.001) and high-quality blastocyst formation (OR = 2.51, 95% CI: 1.60-3.96, P < 0.001) in the non-compacted subgroup. Conversely, it decreased the high-quality blastocyst formation rate in the fully compacted subgroup (OR = 0.37, 95% CI: 0.15-0.89, P = 0.026), while showing no significant effect on transferable blastocyst formation in this group (OR = 0.83, 95% CI: 0.28-2.48, P = 0.740). In conclusion, late-stage vacuolization is a compaction-dependent prognostic marker, detrimental in compacted embryos yet potentially beneficial in non-compacted ones. We hypothesize that its effects may be mediated through mechanical stress-induced pseudo-compaction, suggesting that it should be integrated into D4 embryo grading systems in a context-specific manner. LAY SUMMARY: This study looked at over 9,700 embryos used in IVF. We found that tiny, bubble-like structures (called 'vacuolization') inside a 4-day-old embryo can be either good or bad news, depending on whether the embryo's cells are tightly or loosely packed together. In tightly packed embryos, these bubbles are harmful and reduce the chance of forming a high-quality embryo. In loosely packed embryos, the same bubbles can be helpful and increase the chance of successful development. This is different from other embryo problems, which are always bad. By checking for these bubbles and how packed the cells are, we can predict which embryo is most likely to lead to a baby 10% more accurately than before. To use this new method, embryos need to be grown for 4 days instead of 3 days.