Abstract Study question Does semen quality affect preimplantation development and clinical outcomes in ICSI cycles using donor oocytes? Summary answer Poor semen parameters led to altered morphokinetic parameters during preimplantation development but did not compromise blastocyst quality nor clinical outcomes in oocyte donation cycles. What is known already The quality of semen used for in vitro fertilization (IVF) may compromise the kinetics and morphology of developing embryos, ultimately impairing implantation and reducing live birth rates. Such effects are particularly pronounced when testicular or epididymal spermatozoa are used for IVF treatment. However, current studies assessing the impact of poor semen parameters on embryo development and clinical outcomes are often confounded by the use of oocytes from infertile patients. Performing such evaluations in oocyte donation cycles will provide more definitive insights into the role of semen during preimplantation development. Study design, size, duration This retrospective study included 616 embryos from 99 oocyte donation ICSI cycles, performed between February 2018 and July 2019. We compared embryo morphokinetics and clinical outcomes (cumulative pregnancy and live birth rates) across three groups with variable semen parameters. These included donor semen (control group, DD, n = 294), partner sperm obtained by testicular sperm extraction (TESE group, n = 72) and partner semen with altered parameters (including poor concentration, motility and morphology, ALT group, n = 250). Participants/materials, setting, methods We assessed several morphokinetic parameters, including second polar body extrusion (tPB2), pronuclei appearance (PN1a, PN2a), 2-3-4-5-8-cell stages (t2 to t8), start of blastulation (tSB) and blastocyst formation (tB). Kaplan-Meier survival curves were built for each parameter per study group. Survival curves were analysed by Cox regression. Qualitative parameters (even size PN and blastomeres, percent fragmentation, abnormal cleavages, blastocyst grade) were assessed using a Chi-Squared test, while Fisher’s exact test was used to evaluate clinical outcomes. Main results and the role of chance Interestingly, TESE embryos reached several developmental stages faster than the DD group: tPB2 (3.03 vs. 3.08, p < 0.001), PN1a (5.55 vs. 6.19, p < 0.001), PN2a (6.78 vs. 7.36, p < 0.001), tPNf (21.30 vs. 22.20, p = 0.001), t3 (34.02 vs. 35.79, p = 0.016), t4 (35.90 vs. 36.92, p = 0.015), t5 (46.70 vs. 48.07, p = 0.036), t8 (52.48 vs. 55.28, p = 0.049), tSB (85.53 vs. 94.62, p < 0.001) and tB (104.09 vs. 106.60, p = 0.002). ALT embryos were faster in early stages: PN1a (6.63 vs. 6.19, p = 0.005), PN2a (7.84 vs. 7.36, p < 0.001), t5 (48.12 vs. 48.07, p = 0.028), t8 (55.57 vs. 55.28, p = 0.032), but slower in the late stages: tSB (93.71 vs. 94.62, p = 0.001), tB (105.60 vs. 106.60, p = 0.002). Compared to DD, ALT and TESE embryos showed a lower rate of even pronuclei (ALT p = 4.6x10-06; TESE p = 8x10-05) and even blastomeres at the 2-cell (ALT p < 0.001; TESE p = 0.006) and 4-cell stage (ALT p = 0.004; TESE p = 0.005). Moreover, embryos derived from testicular sperm showed significantly higher fragmentation rates at the 8-cell stage (p = 0.005), while no significant differences in the frequency of irregular divisions nor in the number of top-quality blastocysts were detected in ALT and TESE. Importantly, cumulative live birth rates per cycle were similar across all study groups (DD 31/53, ALT 31/51, TESE 8/11; p > 0.05). Limitations, reasons for caution The main limitation resides in the retrospective nature of this study and the limited number of embryos in the TESE group. We also note variability in the testicular biopsy (obstructive and non-obstructive azoospermia) and non-normozoospermic diagnoses (oligozoospermia, teratozoospermia or asthenozoospermia), which may confound the analysis. Wider implications of the findings Although poor semen parameters significantly altered embryo morphokinetics, they did not compromise embryo quality, pregnancy or cumulative live birth rates, resulting in similar clinical outcomes as with double donation. Overall, our findings support the use of autologous sperm with donor oocytes, even in severe cases of male factor infertility. Trial registration number not applicable
Read full abstract