Time-lapse monitoring embryo culture system

EmbryoViewer software (Unisense Fertilitech Aarhus) was used to inspect the embryo images to elucidate the embryo growth after microinjection over time (hours). After microinjection, oocyte development was evaluated in the time-lapse system, including pronuclear fading (Pnf), the moment when both pronuclei disappear, first cleavage, and the times at which the zygote divides into 2–8 cells (t2 to t8, respectively). The second embryonic cell cycle length (EMCC2 = t3–t2) is the time of embryo cleavage from two blastomeres into three blastomeres. The time from this cleavage to a four-blastomere embryo is the second synchrony (s2 = t4–t3). The time for the three-blastomere embryo to split into a five-blastomere embryo is the prolongation of the third cell cycle (cc3 = t5–t3). A previous study determined the optimal morphological values for the best implantation rate for t5 (t5 = 48.8–56.6 h), s2 (s2 < 0.76 h), and cc2 (cc2 < 12 h, 20.5<t5–t2 < 29, and 11<cc3[t5–t3]<18) (Basile et al., Reference Basile, del Carmen Nogales, Bronet, Florensa, Riqueiros, Rodrigo, García-Velasco and Meseguer2014). We also considered these optimal values.

Embryo grading

Embryo quality was graded A, B, C, or D based on morphokinetic parameters as follows: embryos with a duration of t5–t2 > 20.5 h were defined as A or B, while those outside this optimal range were defined as C or D. Embryos with a duration of cc3 between 11 and 18 h were classified as A or C, while those beyond the optimal range were classified as B or D, depending on t5–t2 (Basile et al., Reference Basile, del Carmen Nogales, Bronet, Florensa, Riqueiros, Rodrigo, García-Velasco and Meseguer2014).

Embryo transfer

Embryos were transferred between the second and fifth days of development with a Sure-Pro Ultra Embryo Replacement Catheter with an Obturator (Cooper Surgical Company, The Netherlands) and a medium Sage One Step.

All patients received luteal support with the daily use of a 400 mg vaginal progesterone tablet (Progestan vaginal capsule 400 mg) and 30 mg oral dydrogesterone (Duphaston tablet 10 mg; Abbott Laboratories, Eczacibasi, Turkey).

A biochemical pregnancy is characterized by a serum level of β-hCG>20 IU/I on day 12 following embryo transfer. Clinical pregnancy is established when the activity of the heart of the fetus is detected through ultrasonography 8 weeks post-embryo transfer. Any pregnancy loss that occurs after clinical pregnancy confirmation is classified as a miscarriage.

Statistical analysis

All data statistical analysis were performed using Statistical Package for Social Sciences 26.0 studio programme and p < 0.05 was considered statistically significant. Continuous data were shown as the mean and standard deviation. Independent sample T-test was used for the analysis of the baseline parameters and the embryomorphokinetic parameters. Categorical variables (fertilization, embryo grade A, clinical pregnancy rates) were presented as percentages and were evaluated using the Pearson’s chi-square test. Embryomorphokinetic parameters were additionally analysed using a Linear Mixed Model.

Power analysis

The G*Power 3.1.9 package programme was employed for the power analysis of the study. Following the collection of data, a post-hoc power analysis based on pregnancy outcomes revealed that the study had a power of 0.89 for 412 patients.

Results

In this retrospective cohort study, a total of 412 patients were included. ICSI + AOA group consisted of 225 patients and ICSI group included 187 patients. When female age, male age, BMI, duration of infertility, duration of induction(days), oocyte count, MII count and 2PN count were examined, there were no significant differences between groups, except for male age (Table 1).

Table 1.

Clinical parameters

Table 1. Long description

Baseline clinical characteristics of 225 patients in the ICSI+AOA group and 187 patients in the ICSI group. Variables include female age, male age, body mass index, duration of infertility, duration of ovarian stimulation, oocyte count, MII oocyte count, and 2PN count. Male age was significantly higher in the ICSI+AOA group (p < 0.001),while there were no significant differences between the groups for the remaining variables (p > 0.05).

Note: Independent sample T-test. Data were presented as mean ± standard deviation.

AOA = Artificial Oocyte Activation, BMI = Body Mass Index, n = number of patients included in each group.

When the durations from 2PN to the blastulation time (tB) were examined, no significant differences were found between the two groups in terms of t2, t3, t4, or t5. When the optimal time lapse parameters (t5, cc2, S2, cc3, and t5-t2) were evaluated no significant relation was found between the probability of optimal time lapse parameters and the use of Ca-I (Table 2). The use of Ca-I was found to extend (p < 0.05) time to start of blastulation (Figure 1) and blastulation times (tB) (p < 0.05) (Figure 2). The embryo morphokinetic data were analysed at the embryo level. Since more than one embryo could originate from the same patient and, in some cases, double embryo transfer was performed, observations were not fully independent. To account for intra-patient clustering, we performed an additional analysis using a Linear Mixed Model, with patient identity included as a random effect.

Figure 1.

Time to start of blastulation in the ICSI + AOA and ICSI groups.

Figure 2.

Blastulation times in the ICSI + AOA and ICSI groups.

Table 2.

Embryomorphokinetic parameters

Table 2. Long description

Embryomorphokinetic parameters were compared between 381 embryos in the ICSI+AOA group and 314 embryos in the ICSI group. Early developmental parameters, including 2PNf, t2, t3, t4, t5, cc2, s2, t5–t2 and cc3, were similar between groups. Time to start of blastulation (tsB) and blastulation time (tB) were significantly longer in the ICSI+AOA group (p<0.05).

^* p < .05.

Note: Independent sample T-test. Data were presented as mean ± standard deviation.

t2: time to the appearance of the 2-cell stage, t3: time to the appearance of the 3-cell stage t4: time to the appearance of the 4-cell stage, t5: time to the appearance of the 5-cell stage, cc2: second cell cycle duration, s2: Synchronization time of the second cell cycle, t5t2: Time interval between the 5-cell stage and the 2-cell stage, cc3: third cell cycle duration, tsB: time to start of blastulation tB: Blastulation time, n = number of embryos analysed in each group.

According to this analysis, no statistically significant differences were observed between the ICSI + AOA and only ICSI groups in early morphokinetic parameters, including t1–t5, cc2, s2, t5t2, and cc3 (all p > 0.05). In contrast, significant delays were detected in later developmental parameters, particularly time to start of blastulation (tsB) (p = 0.002) and time to blastocyst formation (tB) (p = 0.003) (Table 3).

Table 3.

Mixed model results for morphokinetic parameters (ICSI + AOA vs. ICSI)

Table 3. Long description

To account for intra-patient clustering, embryomorphokinetic parameters were additionally analyzed using a linear mixed model with patient identity included as a random effect. No significant differences were observed in early morphokinetic parameters between the ICSI+AOA and ICSI groups, whereas significant delays were detected in time to start of blastulation (tsB; p = 0.002) and blastulation time (tB; p = 0.003) in the ICSI+AOA group.

According to the fertilization results, there were no significant differences between the groups that received and that did not receive AOA treatment (69.87% vs. 74.4%, respectively). When it comes to embryo grades, the grade A embryo rates were not significantly different between ICSI + AOA and ICSI groups (75.3% vs. 71.7%, respectively). There were also no significant differences in the clinical pregnancy outcomes (46.2% vs. 54.6%, respectively) (Table 4).

Table 4.

Pregnancy outcomes

Table 4. Long description

Fertilization rate, Grade A embryo rate, and clinical pregnancy rate were compared between the ICSI+AOA and ICSI groups. No statistically significant differences were observed for fertilization (69.9% vs 74.4%), Grade A embryos (75.3% vs 71.7%), or clinical pregnancy (46.2% vs 54.6%) (all p>0.05).

Note: Pearson’s chi-square test. Data were presented as mean ± standard deviation. Embryo grades were determined using the morphokinetic parameters described by Basile et al.*.

Discussion

We investigated the effects of AOA application with Ca-I on embryomorphokinetic parameters, fertilization, and clinical pregnancy outcomes. To the best of our knowledge, our sample population is the largest of any study conducted on this subject to date. Some clinical studies and meta-analyses have reported that AOA may help improve fertilization, embryonic development, and clinical pregnancy outcomes in many clinical conditions (Murugesu et al., Reference Murugesu, Saso, Jones, Bracewell-Milnes, Athanasiou, Mania, Serhal and Ben-Nagi2017; Chen et al., Reference Chen, Zhao, Lv, Dong, Zhao, Sui, Cui, Liu and Wu2022). In line with our findings, the Ca ionophore group appeared to exhibit a tendency toward a poorer baseline prognosis, although most differences did not reach statistical significance. For example, male age was significantly higher in the Ca ionophore group, while female age and duration of infertility were also numerically higher, albeit without statistical significance. These baseline differences may have contributed, at least in part, to the observed outcomes. Also, because changes in intracellular Ca²⁺ levels influence the entire development process from oocyte to embryo transition, we examined the correlations between intracellular Ca²⁺ (in the form of Ca-I) and morphokinetic parameters.

The use of AOA is controversial, as synthetic activating agents do not fully replicate the natural Ca²⁺ signalling and physiological mechanisms observed in mammalian zygotes (Vanden Meerschaut et al., Reference Vanden Meerschaut, D’Haeseleer, Gysels, Thienpont, Dewitte, Heindryckx, Oostra, Roeyers, Van Lierde and De Sutter2014; Martinez et al., Reference Martinez, Durban, Santaló, Rodríguez and Vassena2021). Activation of the embryo by Ca-I may lead to variations in cellular kinetic patterns and division patterns that differ from normal processes. Therefore, this technique is still experimental, and gathering data from a larger number of patient embryos could yield valuable insights.

A study has evaluated the relationship between AOA and the morphokinetic behaviour of the embryo. Martinez et al. reported that AOA had no effect, with the exception of second polar body extrusion (tPB2) and time to second cell division (t3); however, they concluded that these changes did not seem to affect the morphokinetic pattern of the preimplantation embryo and ascribed the findings to a rapid transient nonphysiological increase in free intracytoplasmic Ca²⁺ (Martinez et al., Reference Martinez, Durban, Santaló, Rodríguez and Vassena2021). Similarly, Shebl et al. performed a sibling oocyte study in patients with previous fertilization problems. Half of the oocytes were treated with calcimycin. They found that ionophore-treated oocytes underwent significantly earlier pronuclei transformation and had a better synchronized third cell cycle (Shebl et al., Reference Shebl, Trautner, Enengl, Reiter, Allerstorfer, Rechberger, Oppelt and Ebner2021). Ca-I treatment significantly improved the overall rates of fertilization and blastocyst formation. In contrast to these findings, in our study pronuclei transformation or third cell cycle synchronization were not different and we found a prolonged tB in AOA cycles. The patient populations evaluated in the aforementioned studies primarily consisted of couples with fertilization failure and severe male factor infertility. However, our cohort was more heterogeneous, including not only cases with male-factor fertilization failure but also patients with low fertilization rates (< 30%) and poor embryological developmental patterns in previous cycles. Therefore, the differences between our findings and previous reports may reflect differences in patient populations. A possible explanation for these discrepant findings is that calcium ionophores may affect early and late developmental events differently. Although previous studies have reported accelerated or improved early cleavage events in certain patient populations, altered calcium signalling dynamics may have different effects on later developmental processes, such as blastulation.

Research on humans has highlighted the potential of tB as a determining factor for selecting embryos, surpassing other components of embryo morphological grading (Ho et al., Reference Ho, Arrach, Rhodes-Long, Salem, McGinnis, Chung, Bendikson, Paulson and Ahmady2018; Moustafa et al., Reference Moustafa, Rosen, Boylan and Mersereau2020; Huayhua et al., Reference Huayhua, Rodríguez, Vega, Briones, Rodriguez-Alvarez and Mellisho2023). Lee et al. and Moustafa et al. noted delayed tB in embryos with chromosome abnormalities (Lee et al., Reference Lee, Chen, Huang, Cheng, Chen, Ho and Lee2019; Moustafa et al., Reference Moustafa, Rosen, Boylan and Mersereau2020). The clinical importance of these changes in embryo morphokinetic parameters is not clear. The ICSI + AOA process results in non-physiological Ca²⁺ signalling patterns within the oocyte. In normal physiological processes, oocyte activation is characterized by recurrent Ca²⁺ oscillations from internal stores, orchestrated by sperm-borne PLCzeta over several hours. However, in AOA via Ca-I induces a single, transient, high-amplitude Ca²⁺ peak. This situation may be related to the faster expulsion of the second polar body in the initial fertilization and faster cleavage stages (t3) mentioned in previous reports, while the lack of long-lasting oscillations may be associated with the prolongation of the tsB and tB periods (Putney Reference Putney1986; Yoon et al., Reference Yoon, Eum, Lee, Lee, Kim, Han, Won, Park, Shim, Lee, Fissore, Lee and Yoon2012). Previous studies and our study results may suggest the importance of taking caution when applying Ca-I and the prolonged tsB and tB durations may serve as a warning for future clinical practice. It would be more appropriate to consider its utilization in cases where there is a clear indication rather than opting for elective use.

Morphokinetic behaviour has also been used to predict euploidy rates in different studies (Basile et al., Reference Basile, del Carmen Nogales, Bronet, Florensa, Riqueiros, Rodrigo, García-Velasco and Meseguer2014; Patel et al., Reference Patel, Shah, Kotdawala, Herrero, Rubio and Banker2016) because aneuploidy can potentially occur at any stage from the disjunction of chromatids to the mitotic divisions within the embryo (Hassold and Hunt Reference Hassold and Hunt2001; Wang and Sun Reference Wang and Sun2006; Chavez et al., Reference Chavez, Loewke, Han, Moussavi, Colls, Munne and Reijo Pera2012; Campbell et al., Reference Campbell, Fishel, Bowman, Duffy, Sedler and Hickman2013). We used the morphokinetic embryo grading system that predicts aneuploidy devised by Basile et al., and did not find any differences between the two groups. The literature contains little information on embryo aneuploidy and Ca-I activation. However, our results are in line with a recent study that found that Ca-I application did not increase the risk for aneuploidy and there were no developmental differences in live births as a result of transfers made with Ca-I administration (Zhang et al., Reference Zhang, Yao, Zhang, Hu, Yang, He, Qina, Fan, Bai and Sun2023). Thus Ca²⁺ ionophores have the capability to bind Ca²⁺ cations, and due to their hydrophobic properties, they form a complex at the lipid bilayer of the membrane. This structure enables ionophores to transfer Ca²⁺ across the membrane and release them into the cytosol (Brasseur et al., Reference Brasseur, Notredame and Ruysschaert1983). Consequently, ionophores themselves may not necessarily enter the oocyte, potentially explaining the absence of a detectable effect of ionophores on chromosomal segregation (Capalbo et al., Reference Capalbo, Ottolini, Griffin, Ubaldi, Handyside and Rienzi2016). Reports have indicated that implanted embryos after IVF, using Ca-I as an artificial oocyte activator, do not exhibit any chromosome abnormalities (Nakagawa et al., Reference Nakagawa, Yamano, Moride, Yamashita, Yoshizawa and Aono2001; Lu et al., Reference Lu, Chen, Gao, Ma, Li, Hu and Li2006). Moreover, no perinatal morbidity or mortality was reported among the patients involved in these studies. Comparable outcomes have also been observed in children aged 3–10 years, who were born following ICSI cycles using Ca-I. No differences have been noted compared to their peers (Vanden Meerschaut et al., Reference Vanden Meerschaut, D’Haeseleer, Gysels, Thienpont, Dewitte, Heindryckx, Oostra, Roeyers, Van Lierde and De Sutter2014; Deemeh et al., Reference Deemeh, Tavalaee and Nasr-Esfahani2015).

Furthermore, ionomycin and calcimycin may induce different intracellular Ca²⁺ signalling dynamics and embryological responses (Quintana-Vehí et al., Reference Quintana-Vehí, Martínez, Zamora, Rodríguez, Vassena, Miguel-Escalada and Popovic2023). Since calcium oscillation patterns regulate key events in embryo development (Malcuit et al., Reference Malcuit, Kurokawa and Fissore2006; Neri et al., Reference Neri, Lee, Rosenwaks, Machaca and Palermo2014), these differences may influence embryo morphokinetics and developmental competence. In the present study, calcimycin (CultActive) was used for oocyte activation. Previous studies using calcimycin (A23187)-based protocols, including CultActive, have reported effects on fertilization, embryo developmental outcomes, and embryo morphokinetics, as well as no increased risk of aneuploidy or adverse developmental outcomes (Capalbo et al., Reference Capalbo, Ottolini, Griffin, Ubaldi, Handyside and Rienzi2016; Murugesu et al., Reference Murugesu, Saso, Jones, Bracewell-Milnes, Athanasiou, Mania, Serhal and Ben-Nagi2017; Shebl et al., Reference Shebl, Trautner, Enengl, Reiter, Allerstorfer, Rechberger, Oppelt and Ebner2021; Tsai et al., Reference Tsai, Lin, Lian, Li, Vitale, Mikuš, Su, Tsai, Tsui and Lin2022; Zhang et al., Reference Zhang, Yao, Zhang, Hu, Yang, He, Qina, Fan, Bai and Sun2023). Although our study did not demonstrate differences in early cleavage parameters, the prolonged tsB and tB observed in the AOA group may further support the potential influence of calcimycin on embryo developmental kinetics. Likewise, ionomycin-based studies have also demonstrated alterations in embryo morphokinetics without adverse long-term developmental outcomes (Vanden Meerschaut et al., Reference Vanden Meerschaut, D’Haeseleer, Gysels, Thienpont, Dewitte, Heindryckx, Oostra, Roeyers, Van Lierde and De Sutter2014; Deemeh et al., Reference Deemeh, Tavalaee and Nasr-Esfahani2015; Martinez et al., Reference Martinez, Durban, Santaló, Rodríguez and Vassena2021).

The principal limitation of our study lies in its retrospective design. Consequently, definitive conclusions are challenging to establish. However, it is important to acknowledge that this limitation is mitigated by our substantial sample size. In addition, due to the retrospective design and incomplete follow-up, live birth outcomes were not available; thus, findings should be interpreted within the scope of embryomorphokinetic outcomes. Our findings may help understanding the embryomorphokinetic behaviour and embryo selection in ICSI + AOA cycles in future applications.

Conclusion

In our study, we did not find significant differences in most of the embryomorphokinetic parameters except prolongation of tB and time to start of blastulation. However, the clinical significance of these changes is controversial because of the lack of correlations between these parameters and clinical outcomes. More studies with larger populations are needed on this subject.