Abstract
Study question
Can time-lapse imaging (TLI) identify morphokinetic events impacted by high sperm DNA fragmentation index (DFI), irrespective of conventional morphological embryo assessment and development rate?
Summary answer
Embryo morphokinetic parameters are negatively impacted by high DFI, whereas conventional morphological embryo assessment and blastocyst development rate are not related to DNA integrity.
What is known already
Paternal genome activation occurs late in the embryo, and therefore, the negative impact of sperm factors on embryo development is more often observed in the outcomes of pregnancy, such as embryonic implantation and pregnancy loss, than in the potential for embryonic development itself, such as successful development to the blastocyst stage. With the development of TLI technology, and the possibility of assessing complete embryonic development, we hypothesized that sperm factors related to DNA fragmentation may interfere with the speed and pattern of cell divisions, leading to slower embryos; something that would not be detected by conventional morphological embryo assessment.
Study design, size, duration
The study included 978 zygotes cultured until day five in a TLS incubator between March/2019 and August/2020, derived from 118 patients undergoing ICSI in a private university-affiliated IVF center. Kinetic markers from the point of insemination were recorded. Generalized linear models adjusted for potential confounders, followed by Bonferroni post hoc were used to compare timing of specific events in patients with low (<30%) of high (≥30%) DFI. The post hoc achieved power was > 90%.
Participants/materials, setting, methods
Recorded kinetic markers were: timing to pronuclei appearance and fading (tPNa and tPNf), timing to two (t2), three (t3), four (t4), five (t5), six (t6), seven (t7), and eight cells (t8), and timing to morulae (tM), start of blastulation (tSB) and blastulation (tB). Durations of second and third cell cycles (cc2 and cc3) and timing to complete synchronous divisions s1, s2, and s3 were calculated. The KIDScore ranking was also recorded.
Main results and the role of chance
Blastocyst development (53.1% ± 1.3 vs. 55.1% ± 1.5, p = 0.380) and high-quality rates (87.9% ± 2.9 vs. 86.2% ± 3.6, p = 0.749) were similar between embryos derived from sperm samples with <30% DFI (n = 592) and ≥30% DFI (n = 386), respectively. Embryos derived from sperm samples with ≥30% DFI showed significantly slower divisions compared to those from <30% DFI: tPNa (6.1h ± 0.2 vs. 6.8h ± 0.2, p = 0.030), tPNf (23.0h ± 0.3 vs. 24.2h ± 0.3, p = 0.009), t2 (25.4h ± 0.3 vs. 26.9h ± 0.3, p = 0.002), t3 (34.8h ± 0.3 vs. 37.3h ± 0.4, p > 0.001), t4 (37.5h ± 0.4 vs. 39.3h ± 0.4, p = 0.003), t5 (46.2h ± 0.5 vs. 49.5h ± 0.6, p < 0.001), t6 (49.7h ± 0.5 vs. 52.8h ± 0.6, p = 0.001), t7 (52.4h ± 0.6 vs. 55.6h ± 0.7, p = 0.001), t8 (56.2h ± 0.7 vs. 58.9h ± 0.8, p = 0.017), tSB (97.5h ± 1.5 vs. 105.9h ± 1.7, p = 0.002), tB (108.6h ± 0.8 vs. 112.4h ± 1.2, p = 0.016). The KIDScore ranked significantly different between embryos derived from samples with <30% or ≥ 30% DFI (4.5 ± 0.1 vs. 3.9 ± 0.2, p = 0.033, respectively). A significant difference was observed in implantation rate (<30% DFI: 51.5% ± 2.2 vs. ≥30% DFI: 30.5% ± 1.3, p < 0.001).
Limitations, reasons for caution
Retrospective nature of this study and the small sample size may be a reason for caution, despite adequate power has been achieved.
Wider implications of the findings
Increasing DFI correlates with delayed cell cleavage and blastulation, leading to reduced implantation rates, without compromising blastulation rate and quality. This finding highlights the importance of TLI for the identification and de-selection of slow-growing embryos for transfer, in cycles with high DFI.
Trial registration number
Not applicable
Morphokinetic parameters comparison between embryos from couples with high or low sperm DNA fragmentation index