Résumé : Purpose: Oocytes containing smooth endoplasmic reticulum aggregates (SERa) have been associated with reduced fertilization and clinical pregnancy rates as well as compromised neonatal outcomes. It was therefore recommended by an Alpha-ESHRE Consensus to discard oocytes presenting this dysmorphism. The data in the literature are nevertheless conflicting and healthy babies have recently been obtained from affected oocytes. The objectives of this study were to compare clinical outcomes between ICSI cycles with and without oocytes affected by smooth endoplasmic reticulum aggregates and to confirm whether affected oocytes can produce healthy babies. Methods: A prospective observational study was performed comparing 714 SERa− ICSI cycles to 112 SERa+ cycles. Among the SERa+ cycles, 518 SERa− oocytes and 213 SERa+ oocytes were analyzed. Fertilization, embryo quality, and pregnancy rates as well as neonatal outcomes were compared between SERa+ and SERa− cycles as well as between SERa+ and SERa− oocytes. Results: The presence of SERa was not associated with an adverse effect on embryological, clinical or neonatal data for SERa+ cycles and oocytes. Seven healthy babies were born from embryos originating from SERa+ oocytes. Conclusions: These results are encouraging and might contribute in the future to a revision of the Alpha-ESHRE Consensus. Larger studies, including a correlation between frequency and size of SERa, clinical outcomes and malformation rates, as well as the follow-up of babies born are nevertheless necessary. In the meantime, the currently conflicting data requires caution when considering transfers of embryos affected by SERa.