Introduction
The goal of reproductive medicine specialists is to increase pregnancy rates and achieve singleton live births in a shorter time, while minimizing the number of transferred embryos to a couple undergoing assisted reproductive treatment (ART; Ubaldi et al., Reference Ubaldi, Capalbo, Colamaria, Ferrero, Maggiulli, Vajta, Sapienza, Cimadomo, Giuliani, Gravotta, Vaiarelli and Rienzi2015).
The laboratory environment is one of the crucial influencing elements of the delicate equation to a successful cycle. It has been established that optimal IVF laboratory conditions are a key requirement in optimizing the chances of having a good-quality embryo to transfer (Wale and Gardner, Reference Wale and Gardner2016). Stable gas concentrations and temperature, well designed culture medium, minimum light exposure, controlled pH and osmolarity and elimination of volatile organic compounds (VOC) in the laboratory, are examples of conditions that are in need of constant optimal quality control to ensure proper embryo development (Cairo Consensus Group, 2020). Another important decision in an IVF laboratory is the choice of the incubator. Each incubator has its benefits and negative aspects as well (Vajta et al., Reference Vajta, Parmegiani, Machaty, Chen and Yakovenko2021). Benchtop incubators show efficacy in terms of culture due to the small chamber that allows fast recovery of gases and temperature after lid opening (Swain, Reference Swain2014). Conversely, time-lapse technology (TLT) incubators allow automated images to be captured without embryo disturbance during incubation.
Many studies have been published comparing efficiency (Sciorio et al., Reference Sciorio, Thong and Pickering2018), safety (Nakahara et al., Reference Nakahara, Iwase, Goto, Harata, Suzuki, Ienaga, Kobayashi, Takikawa, Manabe, Kikkawa and Ando2010; Cruz et al., Reference Cruz, Gadea, Garrido, Pedersen, Martınez, Pérez-Cano, Muñoz and Meseguer2011; Kirkegaard et al., Reference Kirkegaard, Hindkjaer, Grøndahl, Kesmodel and Ingerslev2012) and outcome performance (Mascarenhas et al., Reference Mascarenhas, Fox, Thompson and Balen2019; Kalleas et al., Reference Kalleas, McEvoy, Horne, Roberts and Brison2022) between benchtop and TLT incubators. Although TLT incubators have shown better quality embryos when compared with benchtop incubators (Alhelou et al., Reference Alhelou, Mat Adenan and Ali2018; Sciorio et al., Reference Sciorio, Thong and Pickering2018; Kalleas et al., Reference Kalleas, McEvoy, Horne, Roberts and Brison2022), some studies did not find any difference in embryo quality between both incubators (Nakahara et al., Reference Nakahara, Iwase, Goto, Harata, Suzuki, Ienaga, Kobayashi, Takikawa, Manabe, Kikkawa and Ando2010; Cruz et al., Reference Cruz, Gadea, Garrido, Pedersen, Martınez, Pérez-Cano, Muñoz and Meseguer2011; Park et al., Reference Park, Bergh, Selleskog, Thurin-Kjellberg and Lundin2015).
Higher implantation (Alhelou et al., Reference Alhelou, Mat Adenan and Ali2018), pregnancy (Rubio et al., Reference Rubio, Galán, Larreategui, Ayerdi, Bellver, Herrero and Meseguer2014), ongoing pregnancy (Ueno et al., Reference Ueno, Ito, Uchiyama, Okimura, Yabuuchi, Kobayashi and Kato2019) and live birth rates (Mascarenhas et al., Reference Mascarenhas, Fox, Thompson and Balen2019; Kalleas et al., Reference Kalleas, McEvoy, Horne, Roberts and Brison2022) have been observed with the use of TLT incubators, although similar pregnancy outcomes have been reported in many other studies (Cruz et al., Reference Cruz, Gadea, Garrido, Pedersen, Martınez, Pérez-Cano, Muñoz and Meseguer2011; Kahraman et al., Reference Kahraman, Çetinkaya, Pirkevi, Yelke and Kumtepe2012; Kirkegaard et al., Reference Kirkegaard, Hindkjaer, Grøndahl, Kesmodel and Ingerslev2012; Park et al., Reference Park, Bergh, Selleskog, Thurin-Kjellberg and Lundin2015; Barberet et al., Reference Barberet, Chammas, Bruno, Valot, Vuillemin, Jonval, Choux, Sagot, Soudry and Fauque2018). Consequently, a recent Cochrane review (Armstrong et al., Reference Armstrong, Bhide, Jordan, Pacey, Marjoribanks and Farquhar2019) on TLT incubators confirmed the lack of clinical data to choose TLT incubators over benchtop incubators for embryo culture. Interestingly, no publication has explored a possible difference in euploid rate between these two incubators.
The duration and circumstances of the exposure of the embryos outside the incubator during embryo evaluation and medium changeover, as required with embryos cultured in benchtop incubators, may contribute to the differences observed in embryological and clinical outcomes. Although stable culture conditions – mimicking physiological conditions – should be guaranteed, it has been shown that increased culture condition variations have a considerable negative effect on the number of blastocysts obtained, as well as on pregnancy and implantation rates (Krasnopolskaya et al., Reference Krasnopolskaya, Beketova, Sesina, Сhinchenko, Badalyan, Sudarikova, Bocharova and Zakharchenko2019), thereby questioning their influence on the ploidy status of the generated blastocysts.
Therefore, the objective of this study was to compare the development, useable blastocyst rate, quality and ploidy outcomes of sibling embryos cultured in TLT and benchtop incubators. This was to observe if the differences between both incubation systems had an effect on the final embryo developmental outcome.
Materials and methods
This study was approved by the Research Ethics Committee of ART Fertility Clinics LLC, Abu Dhabi, UAE (REC Reference number – REFA055).
Study design
This retrospective observational sibling oocyte study was performed at ART Fertility Clinics, Abu Dhabi, UAE, between March 2018 and April 2020 and included 796 mature oocytes injected from 42 stimulation cycles. Patients with maternal age between 18 and 45 years old, and who underwent ovarian stimulation for intracytoplasmic sperm injection (ICSI) and preimplantation genetic testing for aneuploidies (PGT-A) through next generation sequencing (NGS) on trophectoderm biopsies due to primary or secondary infertility, were included. Indications to perform PGT-A were the following: advanced maternal age (AMA) if the patient was over 35 years old (9), recurrent miscarriage (5), implantation failures (1), severe male factor (12) and cases in which genetic testing was performed according to the patient’s request (15). With the objective to have comparable groups, only patients with at least 16 fresh mature oocytes subjected to ICSI were included. Oocytes after injection were transferred randomly to the EmbryoScope slide until completion of one full dish; 12 oocytes were cultured in an EmbryoScope™ incubator (Vitrolife, Sweden) and the remaining oocytes were allocated for culture in a K-SYSTEMS G185© incubator (Cooper Surgical, Denmark). Cycles with testicular sperm extraction (TESE) or microTESE were excluded from the study. All included patients had a systemic progesterone level <1.5 ng/ml on the day of trigger for final oocyte maturation.
Research questions
The primary endpoint was to analyze the useable blastocyst rate after culturing sibling oocytes in the two different types of incubators. The useable blastocyst rate was calculated as the number of biopsied blastocysts on Days 5, 6 and 7 per number of normally fertilized zygotes in each arm of the study.
Secondary endpoints comprised fertilization and cleavage rates, embryo qualities on Day 3 and on the day of trophectoderm biopsy (Day 5, 6 or 7), and euploid rate per incubator. Fertilization was calculated as the number of normally fertilized zygotes (2PN) per number of injected oocytes. The cleavage rate was calculated as the number of cleaved embryos on Day 3 per 2PN. The euploid rate was defined as the number of euploid blastocysts per number of biopsied blastocysts in each group.
Ovarian stimulation protocol and oocyte retrieval
Ovarian stimulation was performed by standard protocols, either gonadotropin-releasing hormone (GnRH) agonist or GnRH antagonist protocols, using recombinant follicle-stimulating hormone (recFSH) or human menopausal gonadotropin (HMG) as stimulation medication. The dosage of the stimulation medication was chosen according to the ovarian reserve parameters (La Marca and Sunkara, Reference La Marca and Sunkara2014). Final oocyte maturation was achieved by administration of either 5000 or 10,000 IU hCG, 0.3 mg of GnRH agonist (Triptorelin) or dual trigger (hCG and GnRH analogue), as soon as ≥3 follicles ≥17 mm were present. Oocyte retrieval (OR) was scheduled for 36 h after the trigger for the final oocyte maturation. Oocytes were collected in Quinn’s Advantage Medium with HEPES, (SAGE, Målov, Denmark) supplemented with human serum albumin (HSA; Vitrolife, Göteborg, Sweden; HTF-HSA) and washed in Global Total LP medium for fertilization (CooperSurgical Inc., USA) after which they were cultured at 37°C, 6% CO2 and 5% O2 until denudation.
Insemination and embryo culture
Obtained cumulus–oocyte complexes (COCs) were denuded (Hyase, Vitrolife) 3 h after OR and ICSI were performed on all mature (MII) oocytes 40 h after trigger (Palermo et al., Reference Palermo, Joris, Devroey and Van Steirteghem1992; Vandenberghe et al., Reference Vandenberghe, Santos-Ribeiro, De Munck, Desmet, Meul, De Vos, Van de Velde, Racca, Tournaye and Verheyen2021). After injection, sibling oocytes were cultured in an EmbryoScope or K-SYSTEMS (37°C, 6% CO2, 5% O2). According to our clinical practice, only one EmbryoScope slide (∼12 oocytes) is used per patient, and the remaining oocytes are cultured in a K-SYSTEMS incubator. For the EmbryoScope, the first 12 injected oocytes were placed individually on the EmbryoScope slide with 12 culture drops of 25 μl overlaid with 1.4 ml of oil (Ovoil®; Vitrolife), and for the K-SYSTEMS, oocytes were placed individually in a culture dish containing eight culture drops of 25 μl, overlaid with 8 ml of oil. Single Step Medium (Global®Total®LP, CooperSurgical Inc.) was used for both arms of the study. The average ICSI time per case was 30 min.
For the EmbryoScope, Embryo Viewer Software© was used to evaluate zygotes and embryos without removing the dish from optimal conditions. The fertilization (16–18 h post-injection) and embryo evaluation on Day 3 (68 h), and Days 5, 6 and 7 (116, 140 and 164 h, respectively) were assessed under an inverted microscope for embryos cultured in the K-SYSTEMS incubator and according to the Vienna Consensus. In both groups, zygote/embryo evaluation was performed at the same time and by the same operator, as well as medium refreshment on Day 3 of embryo development, to sustain development up to Day 7. For the EmbryoScope, 20 µl of culture medium was removed from each culture drop and replenished with 20 µl overnight pre-equilibrated Global Total LP medium. For the K-SYSTEMS, embryos were transferred individually to a fresh GPS® Dishware (CooperSurgical LLC) that was pre-equilibrated overnight. Embryos were cultured until the blastocyst stage (Day 5, 6 or 7) and biopsied once the blastocyst was graded ≥ BL3CC (Gardner and Schoolcraft, Reference Gardner, Schoolcraft, Jansen and Mortimer1999). On Days 5/6/7, embryos from both incubators were checked twice a day at the same time; blastocysts incubated in the EmbryoScope were verified on the screen of the incubator, via EmbryoViewer and blastocysts in the K-SYSTEMS were removed from the incubator and evaluated under a microscope. Morphokinetics were not considered for blastocyst biopsy selection in the EmbryoScope arm.
Embryo quality
Embryo quality on Day 3 was evaluated according to the number of blastomeres, symmetry, fragmentation, multinucleation and presence of vacuoles. The scoring grade was determined according to a combination of all five classical morphological assessments described above and divided into four categories of embryo qualities: excellent or top quality [embryo quality 1 (EQ1)]: ≥7 blastomeres, fragmentation ≤10%, equal blastomere size, no vacuoles or multinucleation; moderate quality (EQ2): ≥6 blastomeres, fragmentation until 20%, similar blastomere size, and multinucleation and vacuoles observed in <50% of the blastomeres; poor quality (EQ3): embryos with ≥4 blastomeres, fragmentation between 21–50%, unequal blastomere size, several normal nuclei or smaller fragmented nuclei visible in <50% of the blastomeres and vacuolated blastomeres, with ≥50% small vacuoles; and bad quality (EQ4): embryos with >50% of fragmentation, unequal cell size, multinucleation and large vacuoles seen in >50% of the blastomeres.
Biopsied blastocysts were graded based on the inner cell mass (ICM) and trophectoderm (TE) morphology and expansion stage (Gardner and Schoolcraft, Reference Gardner, Schoolcraft, Jansen and Mortimer1999). The scoring grades were divided into four categories of embryos: excellent or top quality (EQ1): AA, AB, BA; good quality (EQ2): BB; moderate quality (EQ3) AC, BC, CA, CB, CC; and poor quality (EQ4) AD, BD, CD, DA, DB, DC, DD. For statistical purposes, embryo quality was dichotomized: good quality (GQ = EQ1 + EQ2) and poor quality (PQ = EQ3 + EQ4).
Blastocyst biopsy and NGS
Trophectoderm biopsy was performed on expanded blastocysts (≥BL3) on Day 5, 6 or 7. The biopsy technique and assessment of ploidy status were performed as previously described (De Munck et al., Reference De Munck, El Khatib, Abdala, El-Damen, Bayram, Arnanz, Melado, Lawrenz and Fatemi2020). Blastocysts were vitrified 1 h after the biopsy procedure according to the manufacturer’s protocol (Kitazato© Vitrification).
Statistical analysis
Data analysis was performed using the R Computational Statistical Framework [Ref1: https://www.r-project.org/] with IDE RStudio [Ref2: https://rstudio.com/]. Cycle characteristics were described using mean ± standard deviation (SD), minimum and maximum values. Outcomes were inspected initially using Bland–Altman plots, as well as computing and visualizing Spearman’s coefficients and their statistical significance. The Mann–Whitney statistical test was applied to selected outcome variables of mean per cent per cycle to compare the two incubators (EmbryoScope versus K-SYSTEMS), using the ‘wilcox.test’ function. Differences in embryo development and outcome in each arm were analyzed using the chi-squared test (or Fisher’s exact test). All the significance tests were two-tailed and the significance level was set at P < 0.05. Figures were generated using R library ggplot2 (https://ggplot2.tidyverse.org/) ‘geom_smooth’ function, suppressing the representation of standard error (SE).
This retrospective study included 796 embryos of which 63.2% were placed in an EmbryoScope (n = 503) and 36.8% in K-SYSTEMS (n = 293) incubator considering a 1:1.7 sample ratio. The euploid rates using both types of incubators were assumed to be 60% vs 70% and 50% vs 60% with 0.10 effect size, considering a 0.05 margin of error (Zα/2). Our study calculated power was 0.8131 and 0.7812, respectively, that is acceptable to prove the significant results (Serdar et al., Reference Serdar, Cihan, Yücel and Serdar2021).
Results
In total, 796 mature sibling oocytes from 42 patients were assigned for culture in two different incubators (EmbryoScope and K-SYSTEMS) and their embryo development was analyzed. Table 1 describes the cycle characteristics of all patients included in this study. The indications for PGT-A were AMA (21.4%), repeated implantation failure (2.4%), recurrent miscarriages (11.9%), male factor (28.6%) and elective PGT-A in 35.7% of the cases.
Table 1. Cycle characteristics (n = 42)
Results are presented as mean ± SD per cycle. AMH: anti-Müllerian hormone (ng/ml); BMI: body mass index (kg/m2); COC: cumulus–oocyte complex; P4: progesterone.
Out of 796 mature oocytes included in the study, 503 were cultured in an EmbryoScope (63.2%) and the remaining 293 sibling oocytes were cultured in K-SYSTEMS (36.8%; Table 2). The fertilization (79.3% vs 78.8%, P = 0.932) and cleavage rates (98.5% vs 99.1%, P = 0.676) were not different between both incubators. However, the total useable blastocyst rate was significantly higher for embryos cultured in the EmbryoScope (64.8% vs 49.6%, P < 0.001), mainly because a higher percentage of blastocysts was biopsied on Day 5 from the EmbryoScope (67.8% vs 57.0%, P = 0.037). There was no difference in the total euploid rate between the EmbryoScope and K-SYSTEMS (59.9% vs 50.4%, P = 0.314); however, a significantly higher euploid rate was observed for blastocysts cultured and biopsied on Day 5 from the EmbryoScope (63.5% vs 37.4%, P = 0.001). The distribution of the useable blastocyst rate is shown in Figure 1. The useable blastocyst rate was also calculated per age in both the EmbryoScope (P = 0.005, aR2: 0.1606) and K-SYSTEMS (P = 0.028, aR2: 0.093; Figure 2). Advanced maternal age showed a negative correlation with the useable blastocyst rate in both incubators. For all age categories, the EmbryoScope showed a highly statistically increased useable blastocyst rate compared with K-SYSTEMS (P < 0.0001).
Table 2. Fertilization and embryo development
Calculation of useable blastocyst rate (number of biopsied blastocysts per number of normally fertilized oocytes), euploid rate (number of euploid blastocysts per number of biopsied blastocysts) and euploid rates per day of biopsy (5, 6 or 7). BL – blastocyst.
a Wilcoxon test (per cycle data).
b Chi-squared test or Fisher’s exact test or post-hoc chi-squared test (embryo data).
* SD for mean % (per cycle), SE for proportion % (per embryo).
Figure 1. Useable blastocyst rate per cycle. Useable blastocyst rate (blastocysts biopsied/normally fertilized oocytes) per cycle, stratified per incubator (every dot represents one stimulation cycle) with 95% confidence limits. ES: EmbryoScope; KS: K-SYSTEMS.
Figure 2. Useable blastocyst rate per age in EmbryoScope (ES) and K-SYSTEMS (KS). Distribution of useable blastocyst rate in both incubators per age (years) with darker area representing 95% CI.
Embryo quality at cleavage and biopsy stage is presented in Table 3. On Day 3, as well as for all biopsied blastocysts, no difference was observed in the embryo quality (GQ or PQ) between both incubators (P = 0.543 and P = 0.151, respectively). A significant difference in good-quality and poor-quality blastocysts biopsied on Day 5 and Day 6 was noted between embryos cultured in the EmbryoScope versus K-SYSTEMS, P = 0.008. The chance of good-quality or poor-quality cleavage stage embryos (Day 3) to develop into useable blastocysts, was calculated per incubator. Comparing the total useable blastocyst development, no difference was found between embryos originating from good-quality (P = 0.083) or poor-quality (P = 0.111) cleavage stage embryos. However, when stratifying according to the day of blastocyst development, poor-quality embryos on Day 3 showed a statistically significant superior blastocyst formation on Day 5 when cultured in the EmbryoScope (64.1% vs 39.1%), which was shifted to Day 6 for embryos cultured in K-SYSTEMS (35.9% vs 60.9%; P = 0.005). This difference in the day of blastocyst development was not observed for GQ cleavage stage embryos (P = 0.916).
Table 3. Embryo quality and blastocyst formation
Analysis of cleavage stage embryo quality and blastocyst quality at the time of biopsy between embryos cultured in the EmbryoScope and K-SYSTEMS. Day 7 development is not present due to insufficient data. D: day, GQ: good quality, PQ: poor quality, BL: blastocyst;
chi-squared test or Fisher’s exact test or post-hoc chi-squared test.
Discussion
As the exposure of embryos outside the incubator may affect their development, this study aimed to evaluate the embryo development and ploidy rates when sibling oocytes were cultured in the EmbryoScope and K-SYSTEMS: a significantly higher total useable blastocyst rate (64.8% vs 49.6%, P < 0.001) and euploid rate on Day 5 was observed for embryos cultured in the EmbryoScope (63.5% vs 37.4%, P = 0.001). In addition to this, an increased blastocyst quality on Day 5 was observed after culture in the EmbryoScope (P = 0.008).
The optimization of ART outcomes is a long and complex journey that includes the couple’s previous medical history, stimulation strategy, cycle plan, insemination method and embryo culture conditions. In the IVF laboratory, constant efforts are made to improve culture conditions (Wale and Gardner, Reference Wale and Gardner2016) and embryo selection with the objective of improving cycle outcomes. The incubator’s choice is one relevant decision that possibly affects embryo development. Benchtop incubators are proven to be effective due to the fast recovery of temperature and gases, but a major limitation is a need to remove the culture dish from the incubator to perform embryo assessment (Swain, Reference Swain2014). Although excluding Day 2 and Day 4 evaluations, the need to check fertilization, as well as performing a Day 3 changeover, induces a considerable amount of time outside the incubator during which time the embryos are exposed to suboptimal culture conditions (Zhang et al., Reference Zhang, Li, Peng, Guo, Heng and Tong2010). When TLT is considered, the continuous frames captured by a camera attached to the incubator, and subsequent video access by the embryologist, offer significantly reduced exposure of embryos to this suboptimal environment (Rubio et al., Reference Rubio, Galán, Larreategui, Ayerdi, Bellver, Herrero and Meseguer2014).
Previous retrospective (Ueno et al., Reference Ueno, Ito, Uchiyama, Okimura, Yabuuchi, Kobayashi and Kato2019) and randomized controlled trials (Alhelou et al., Reference Alhelou, Mat Adenan and Ali2018; Barberet et al., Reference Barberet, Chammas, Bruno, Valot, Vuillemin, Jonval, Choux, Sagot, Soudry and Fauque2018) have described higher useable blastocyst rates in a TLT incubator compared with conventional incubators (51.0% vs 46.6%, P < 0.05; Ueno et al., Reference Ueno, Ito, Uchiyama, Okimura, Yabuuchi, Kobayashi and Kato2019), the total proportion of cryopreserved embryos (29.5% vs 24.8%, P = 0.027; Barberet et al., Reference Barberet, Chammas, Bruno, Valot, Vuillemin, Jonval, Choux, Sagot, Soudry and Fauque2018) and blastocyst formation (52.1% vs 46.3%, P = 0.0022; Alhelou et al., Reference Alhelou, Mat Adenan and Ali2018). These publications are in line with the results obtained in this study, as embryos cultured in EmbryoScope had a significantly higher chance to develop into blastocysts and be biopsied than the ones cultured in a K-SYSTEMS incubator (64.8% vs 49.6%, P < 0.001), especially on Day 5 (67.8% vs 57.0%, P = 0.037). The criteria for cycle inclusion in our study (≥16 mature oocytes for injection), resulted in the selectin of a ‘good prognosis population’ with a mean maternal age of 31.4 ± 6.3 years and a mean anti-Müllerian hormone (AMH) level of 6.03 ± 4.2 ng/ml, which may explain the generally increased blastocyst formation in our study compared with previous publications. Despite the fact that previous research using sibling oocytes failed to demonstrate the increased performance of TLT compared with conventional incubators for fertilization rate (57.5% vs 57.5%), GQ embryos on cleavage stage (36.0% vs 36.0%; Nakahara et al., Reference Nakahara, Iwase, Goto, Harata, Suzuki, Ienaga, Kobayashi, Takikawa, Manabe, Kikkawa and Ando2010), blastocyst rate (54.8% vs 50.6%; Cruz et al., Reference Cruz, Gadea, Garrido, Pedersen, Martınez, Pérez-Cano, Muñoz and Meseguer2011), implantation rate (38.1% vs 30.4%, P = 0.75) and clinical pregnancy rate (36.8% vs 33.3%, P = 1.0; Kirkegaard et al., Reference Kirkegaard, Hindkjaer, Grøndahl, Kesmodel and Ingerslev2012), the TLT incubator was proven to be safe, as no differences were found (Nakahara et al., Reference Nakahara, Iwase, Goto, Harata, Suzuki, Ienaga, Kobayashi, Takikawa, Manabe, Kikkawa and Ando2010; Cruz et al., Reference Cruz, Gadea, Garrido, Pedersen, Martınez, Pérez-Cano, Muñoz and Meseguer2011; Kirkegaard et al., Reference Kirkegaard, Hindkjaer, Grøndahl, Kesmodel and Ingerslev2012).
Even though the same numbers of embryos were cryopreserved in both incubation systems, Sciorio and colleagues (Sciorio et al., Reference Sciorio, Thong and Pickering2018), using sibling oocytes, observed improved embryo quality on Day 2 and Day 3, when embryos were cultured in the EmbryoScope. This difference in Day 3 embryo quality was not observed in our study; however, a difference was observed in the potential of poor-quality Day 3 embryos to develop to the blastocyst stage. Poor-quality embryos on Day 3 developed significantly faster to blastocysts on Day 5, if cultured in the EmbryoScope rather than in K-SYSTEMS (on Day 5, 64.13% vs 39.13% and Day 6, 35.87% vs 60.87%, P = 0.005), confirming the advantage of a culture system with ‘reduced disturbance’ (Krasnopolskaya et al., Reference Krasnopolskaya, Beketova, Sesina, Сhinchenko, Badalyan, Sudarikova, Bocharova and Zakharchenko2019). In addition to the difference in developmental speed, the EmbryoScope also showed an increased proportion of biopsied blastocysts with GQ on Day 5 compared with Day 6 (78.21% vs 62.5% for Day 5, 21.79% vs 37.5% for Day 6; P = 0.008).
Although the total euploid rate did not show any difference between the two incubators (59.9% vs 50.4%, P = 0.314), once the euploid rate was stratified per day of the biopsy, the EmbryoScope had a significantly higher euploid blastocyst rate on Day 5 compared with K-SYSTEMS (63.5% vs 37.4%, P = 0.001). This result can be ascribed to the fact that also a higher proportion of GQ blastocysts was observed on Day 5 in the EmbryoScope, as a negative correlation between embryo development and chromosomal content had previously been described (Magli et al., Reference Magli, Gianaroli, Ferraretti, Lappi, Ruberti and Farfalli2007; Wang et al., Reference Wang, Kort, Behr and Westphal2018). The advantage of culture in the EmbryoScope may become evident upon transfer, as previous meta-analyses confirmed the improved outcomes in cycles in which blastocysts were transferred on Day 5 of development rather than Day 6 (Bourdon et al., Reference Bourdon, Pocate-Cheriet, Finet de Bantel, Grzegorczyk-Martin, Amar Hoffet, Arbo, Poulain and Santulli2019; Li et al., Reference Li, Wang, Sun, Lv, Ge, Li and Zhou2020).
Advanced maternal age is one of the main infertility causes. Aged oocytes experience chromosome segregation errors during meiosis that can result in aneuploidy and poor oocyte quality (Mikwar et al., Reference Mikwar, MacFarlane and Marchetti2020). Accordingly, maternal age plays a drastic influence on embryo development (Klein and Sauer, Reference Klein and Sauer2001). From our data, the EmbryoScope showed improved blastocyst development and euploid rates compared with K-SYSTEMS for all age ranges, reinforcing the advantage of TLT even in patients with AMA (Figure 2). Krasnopolskaya and colleagues (Krasnopolskaya et al., Reference Krasnopolskaya, Beketova, Sesina, Сhinchenko, Badalyan, Sudarikova, Bocharova and Zakharchenko2019) pointed to the harmful effects on blastulation and pregnancy rate, mainly in patients with AMA, when unnecessary long manipulation of the culture was made on Day 3.
The enhanced culture environment due to the decreased exposure of embryos to a suboptimal environment outside the incubator during evaluation and medium changeover, is possibly one of the reasons for the improved results seen for embryos cultured in the EmbryoScope. Wale and Gardner (Reference Wale and Gardner2016) reviewed previously that multiple stress factors of the in vitro culture system, such as pH and temperature oscillations and exposure to atmospheric oxygen (20%) may affect embryo development. Considering these stress factors, and even though TLT microwells were mainly designed for optimal optical visibility (Vajta et al., Reference Vajta, Parmegiani, Machaty, Chen and Yakovenko2021), the new generation of TLT incubators could be favourable for embryo development, as minimum manipulation of the culture dish is required (Wang et al., Reference Wang, Meng, Hackett, Odenbourg and Keefe2001). In the present study, medium refreshment was performed on Day 3 for both incubators but decreased time outside the incubator was obtained for the EmbryoScope group as no microscopic evaluations were needed to check fertilization, cleavage and blastocyst development. Furthermore, to avoid the average ICSI duration and the possible effect on the developmental delay observed in one of the groups, embryo development was checked according to the order of injection. Despite the reduced time out of the incubator for embryos cultured in TLT, it must be highlighted that different incubators, culture dishes and quantity of oil were used in the two sibling arms, which may also contribute to the observed differences in blastocyst development, rather than the difference in time out of the incubator alone. Also, it is important to point out that the use of the Embryo Viewer in the EmbryoScope arm gives more time to the embryologist to evaluate blastocyst morphology and their real potential to be biopsied, which can explain the higher percentage of biopsied embryos on Day 5 and their improved quality.
The introduction of TLT incubators for embryo development and selection is still controversial due to non-concurring data on clinical and ongoing pregnancy, miscarriage and live birth between standard incubation and TLT (Armstrong et al., Reference Armstrong, Bhide, Jordan, Pacey, Marjoribanks and Farquhar2019). Unfortunately, the outcome and performance of the euploid blastocysts were not followed in this retrospective study, as the number of patients was too low, and some patients performed ‘mixed’ double embryo transfers with blastocysts cultured in both incubators. Although no clinical benefit has been shown so far (Armstrong et al., Reference Armstrong, Bhide, Jordan, Pacey, Marjoribanks and Farquhar2019), more data from randomized controlled trials is needed to compare clinical outcomes after the transfer of euploid blastocysts, obtained from different incubators. This retrospective study just starts to analyze important data that can be explored with future larger prospective randomized studies.
In summary, the presented data add valuable information for the improvement of blastocyst culture by decreasing embryo stress. Additionally, this study provides increased evidence to support the introduction of TLT incubators into the clinical routine. To conclude, exposure of embryos to a suboptimal environment when outside the incubator during evaluation and medium changeover, may negatively effect in vitro blastocyst development and euploid rates on Day 5.
Acknowledgements
We would like to thank Leyla Depret Bixio, Gioia Altobelli and Rachana Patel for statistical analysis and interpretation of the results presented.
Author contributions
N.N., N.D.: conception and design of the study; N.N., N.D., I.E., A.A., A.B., A.Ar.: oocyte retrieval, ICSI procedure, embryo evaluations, blastocyst biopsy and vitrification; N.N.: construction of the database, data interpretation and drafting the manuscript; and N.D., I.E., A.A., A.D., A.B., A.Ar., B.L., H.F.: review of the manuscript.
Financial support
No funding was obtained for this study.
Competing interest
There are no conflicts to declare.
Ethical standards
This study was approved by the Research Ethics Committee of ART Fertility Clinics LLC, Abu Dhabi, UAE (REC Reference number - REFA055).
