The ultimate goal of preimplantation genetic screening and assisted reproductive treatment is to select one to two of the most competent embryos with normal chromosome compositions for transfer in order to maximize the chances of a successful pregnancy with delivery of a healthy baby while minimizing the incidence of miscarriages in each treatment cycle. Aneuploidy rates are extremely high in IVF patients, especially in those with unexplained recurrent pregnancy loss , repeated implantation failure  and/or previous aneuploid conceptions . Recent studies with array CGH screening have demonstrated a significant improvement in pregnancy outcomes for PGS patients [37, 64–66]. Meanwhile, recent advances in time-lapse culture and monitoring have provided new morphokinetic markers for selecting competent embryos for transfer [26, 34]. In the current study, we have combined these two advanced technologies available in our IVF clinics to provide the advantage of selecting competent blastocysts for transfer and thereby maximizing the chances of a successful pregnancy for our PGS patients. There were significant differences in clinical pregnancy rates between the time-lapse system (Group A) and the conventional incubator (Group B) (71.1% vs. 45.9%, respectively, p = 0.037). Moreover, the implantation rate was higher in Group A compared to Group B (66.2% vs. 42.4%, respectively, p = 0.011). A significant difference in the ongoing pregnancy rate was also observed between Group A and Group B (68.9% vs. 40.5%, respectively, p = 0.019). Collectively, our data show the distinct benefits of combining time-lapse monitoring and array CGH testing to select competent blastocysts for transfer for patients undergoing preimplantation screening. A recent retrospective analysis of a large number of IVF treatment cycles (n = 7305) also concluded that monitoring and selecting embryos in the time-lapse system significantly improved clinical pregnancy and implantation rates compared to the conventional incubator .
Compared to previous reports, our current study has multiple advantages with regard to studying the clinical benefits of combining time-lapse monitoring and array CGH testing to select competent blastocysts for transfer in PGS patients. First, ploidy was determined with array CGH testing, and selection of embryos for transfer was primarily based on the array CGH results in both time-lapse system and conventional incubator groups in order to ensure that only euploid embryos were selected for transfer to patients. In the time-lapse system group, the morphokinetic markers within the most predictive parameters were the secondary criterion for selection when multiple euploid blastocysts were recognized from individual patients. In the conventional incubator group, morphological grading by microscopic evaluation was the secondary criterion for selection when multiple euploid blastocysts were available. However, in previous studies comparing the time-lapse system and the conventional incubator, ploidy of the transferred embryos had not been determined before the embryos were selected for transfer. Lack of chromosomal screening may lead to transfer of euploid and/or aneuploid embryos to patients, producing inconsistent data and conflicting pregnancy outcomes [32–34, 36]. Second, in our prospective study, a sibling oocyte model was designated so that the patients served as their own control, and much larger numbers of MII oocytes (n = 1163) were included in order to draw a firmer statistical conclusion compared to the earlier time-lapse studies with sibling oocytes [32, 33]. Moreover, a relatively larger number (n = 138) of younger patients (at a mean age of 36.6 ± 2.4 year ranging 28 to 39 years) was included in the present study in order to avoid the effects of advanced maternal age on morphokinetic parameters and chromosomal status of embryos when compared to previous research exploring the relationship between morphokinetic parameters and aneuploidy . It has been well documented that the aneuploidy rate increases with maternal age [37–41, 44–46], especially at advanced maternal ages [11, 70]. Recent studies have also revealed that maternal age is one of the major confounding factors affecting clinical outcomes as related to morphokinetic parameters of human embryos that were cultured and monitored in time-lapse systems [34, 36]. Furthermore, in the present study, the time-lapse system was closely monitored and constantly operated with reduced oxygen tension (5%). In the previous studies comparing embryo culture in the time-lapse system and the conventional incubator, however, embryos were entirely cultured under atmospheric oxygen concentration (20%) and the pregnancy and implantation outcomes were not optimized in the time-lapse system group [32, 33]. The significance of culturing oocytes and embryos under low oxygen tension has been well documented in mammalian species including humans [16, 18, 81–83]. Studies with various species of mammals have demonstrated that the concentration of oxygen inside the uterus and oviduct usually falls in the ranges of 2-8%. Improved clinical pregnancy, implantation and live birth rates have also been reported after the use of reduced oxygen tension for embryonic culture to the blastocyst stage [16, 82, 84]. These results are associated with a reduction of the harmful effects of reactive oxygen species (ROS). The increase in the generation and accumulation of ROS is associated with various types of cell damage including DNA fragmentation, altered gene expression, and organelle and membrane disturbances in oocytes and embryos [81, 82]. Consequently, interrupted or delayed embryonic development, apoptosis or health impairment during pregnancy can be observed in embryos cultured under atmospheric conditions [83, 84]. In the current study, clinical pregnancy, implantation and ongoing pregnancy rates were significantly improved in the time-lapse system with reduced oxygen concentration compared to the conventional incubator with atmospheric oxygen concentration. Collectively, our data suggest that the use of time-lapse culture and monitoring with low oxygen tension may improve clinical and implantation outcomes for PGS patients. Finally, the temperature was strictly monitored and controlled in the time-lapse system during the entire period of the current study. In addition, all fertilized oocytes were cultured to the blastocyst stage in the continuous single culture medium (CSC, Irvine Scientific, Irvine, USA) to avoid sudden changes in culture conditions, especially temperature fluctuation. Adverse effects of temperature fluctuation on the meiotic spindle have been well documented in various mammalian species . Transient cooling to room temperature can cause irreversible disruption of the meiotic spindle in human oocytes and embryos . Such disruption may, in turn, result in the elevated levels of aneuploidy in human oocyte and embryos, especially when embryos are handled outside of the incubator during medium change and evaluation, where the earlier PGS studies were performed [53–57].
By combining these two advanced technologies, this prospective study extends prior research where either time-lapse monitoring or array CGH screening alone was used for evaluation and selection of competent embryos for transfer. To the best of our knowledge, this is the first prospective study with sibling oocytes to apply both time-lapse monitoring and array CGH testing to select competent blastocysts for transfer in patients undergoing preimplantation genetic screening. Our research contributes new array CGH and time-lapse evaluation data, assuring the importance of selecting competent embryos for transfer in the PGS patients with various clinical indications. The extent of aneuploidy in human embryos can be extensive [37–46], although this rate is typically lower in embryos at blastocyst stage [43, 60]. This prospective study provides further evidence of substantial chromosomal abnormalities in apparently normal blastocysts inside or outside of range of the most predictive morphokinetic parameters, including monosomy, trisomy, dual and complex aneuploidy [11, 12, 20, 60]. Our data also confirmed the previous observation that morphological evaluation should not be solely relied upon in the selection of competent embryos for transfer [11, 12]. Moreover, there were no significant differences in any of the morphokinetic parameters of the early embryonic development between euploid and aneuploid embryos, although there was a slight delay in some of the morphokinetic parameters at the late stage of embryonic development in aneuploid embryos compared to euploid embryos. Additionally, there was a non-significant trend in which clinical pregnancy and implantation rates increased in the euploid blastocysts with early initiation of blastulation compared to the euploid blastocysts with delayed initiation of blastulation. These data suggest that ploidy of the transferred blastocysts may be likely the primary factor for determining the clinical pregnancy and implantation outcomes in patients undergoing preimplantation genetic screening, while morphokinetic markers of the last stages of embryonic development (e.g. tIB) may be used as a complementary system  to array CGH for embryonic selection. Thus, the combination of time-lapse monitoring and array CGH testing should be recommended for PGS patients to maximize the chances of successful pregnancies and to minimize the incidences of harmful miscarriages.
Several limitations in our prospective study should be addressed. First, although the combination of time-lapse evaluation and array CGH screening displays distinct benefits for many patients undergoing preimplantation genetic screening, this approach is not for all IVF patients with various clinical indications, especially those with diminished ovarian reserve or poor stimulation responders. The improved implantation and ongoing pregnancy rates in the time-lapse monitoring group noted here may not necessarily apply to patients in all age groups, especially those over 40 years old. Moreover, the observed difference in results between array CGH testing of the trophectoderm cells and the cytogenetic analysis of the products of conception suggests that mosaicism may be the cause of the misdiagnosis of a small proportion of human embryos at the blastocyst stages [37, 85], although this mosaicism rate is generally believed to be lower than that of embryos at cleavage stages [37, 43, 60, 85]. Additionally, there was a non-significant trend in which the rate of pregnancy loss decreased in the time-lapse system compared to the conventional incubator (3.1% vs. 11.8%, respectively, p = 0.273). This observation may be due to the cumulative sample size being insufficient to detect a significant difference in this category. Finally, potential epigenetic effects as related to external factors such as stimulation protocol, culture media, light exposure, incubation conditions and manipulation of embryos remain relatively unknown [86, 87]. Further prospective clinical trials with a larger scale of randomized samples may be helpful in clarifying these issues.