RESUMEN
PURPOSE: Sperm play an essential role in embryonic genome activation and embryonic progression to blastocyst. In the present work, we focus on development of embryos created as a result of ICSI with testicular or epididymal sperm from azoospermic males and compare this to outcomes from normospermic males. The objective of this study was to determine if sperm origin influences clinical outcomes, the kinetics of embryo development, or the incidence of cleavage anomalies and multinucleation. METHODS: A total of 93 consecutive intracytoplasmic sperm injection cycles (ICSI) performed for 83 couples were included in this study. Observations were made on 594 fertilized oocytes cultured in the EmbryoScope using time-lapse microscopy (TLM). Epididymal sperm (n = 29) cycles or surgically retrieved sperm from the testis (TESE; n = 37 cycles) of men with either obstructive (OA) or non-obstructive azoospermia (NOA) were used to inject oocytes. A further 27 ICSI cycles were performed using ejaculated sperm from normospermic males, designated as our control sperm (CS) group. Kinetic data and cycle outcomes were retrospectively analyzed. RESULTS: The clinical pregnancy rate was not different between the three groups (TESE 51.4%, PESA 57.7%, and CS 59.3%). A non-significant decrease was observed in both implantation (30.9%) and live birth rate (43%) with TESE as compared to PESA (35.3%, 58%, respectively) and CS groups (45.1%, 56%, respectively). Failure to compact was significantly higher amongst TESE-NOA embryos (35.2%; P < 0.001) as compared to TESE-OA (4%), PESA (9%), and CS (3.8%) embryos. The two points at which TESE-derived embryos (both NOA and OA) behaved most differently from PESA and CS embryos was at cc2 (t3-t2; time to initiation of the second cell cycle) and tSB (time to start of blastulation). A significantly lower percentage of TESE embryos exhibited kinetics typically ascribed to high quality embryos with the greatest developmental potential. Finally, the incidence of direct uneven cleavage (DUC) was observed to be significantly higher after ICSI with sperm retrieved from azoospermic males. CONCLUSIONS: TLM allowed a more in depth comparison of paternal influence on embryo morphokinetics and helped to identify specific differences in cell cycle kinetics. TESE-NOA embryos exhibited a higher incidence of compaction failure.