Materials and Methods:

This study was conducted at a tertiary level infertility unit in the United Kingdom (UK) performing approximately 1500 oocyte retrievals per year. Data extraction was performed from the electronic database of all oocyte retrievals performed from November 2015 to December 2017. This period was chosen because at the start of November 2015 the culture medium was changed from sequential media to a single step culture media and then in January 2018 day three morphology checks were ceased. The interim period therefore had consistent culture conditions for comparison. Neither specific funding nor ethical approval was sought for this study as it was a retrospective analysis of data. There was no patient and public involvement in the design of this study. Only the first cycle in this study period per woman was included for analysis. The primary outcomes analysed were live birth and cumulative live birth per oocyte retrieval. Details of the number of cycles included and the exclusion criteria have been provided in figure 1.

Ovarian stimulation and oocyte retrieval

Ovarian stimulation with gonadotrophins was performed using either a conventional long agonist protocol or antagonist protocol to prevent a premature Luteinising hormone (LH) surge. Follicular monitoring was performed using transvaginal ultrasound with ovulation triggered using either human chorionic gonadotrophin (hCG) or GnRH (gonadotropin releasing hormone) agonist when at least three follicles were 17mm or larger. Oocyte retrieval was performed 35 hours after ovulation trigger under ultrasound guidance. IVF or ICSI was performed between 39 and 41 hours post ovulation trigger depending on the sperm parameters and previous fertilisation results if applicable.

Embryo culture

The patients had the choice of two culture systems in our unit; SC (BT37 bench-top incubators, Planar, UK or Heracell 150 chamber incubators, Thermo Scientific, UK) or TLI at an additional cost (EmbryoScope®, Vitrolife, DK). Where IVF was performed and TLI was required, up to 12 fertilised zygotes were placed in individual wells of an EmbryoScope® Slide at fertilisation check (at 16-19hpi) and placed into the TLI. For TLI cases where ICSI was performed, injected oocytes were placed immediately into TLI if there were 12 or fewer mature oocytes injected, or alternatively fertilised zygotes were placed into TLI after fertilisation check where there were more than 12 mature oocytes injected. Where there were more than 12 zygotes for a patient the additional zygotes were placed into SC. Cycles where both TLI and SC incubators were used were excluded from the analysis.
A low oxygen (5%) culture environment with 6% carbon dioxide at a temperature of 370C was used in both SC and TLI culture systems. Quinn’s Advantage Fertilisation Medium (Cooper Surgical, Denmark) was used from oocyte retrieval to fertilisation check for IVF cases and until denudation for ICSI cases. This was followed by Sage 1-Step™ (Cooper Surgical, Denmark), with both types of culture medium having an oil overlay (Origio Liquid Paraffin, Cooper Surgical, Denmark). In both TLI and SC incubators, single embryo culture was performed in micro-drops. Embryos in SC incubators were removed from culture on day 3 for morphology assessment, whereas the TLI embryos had undisturbed culture until the day of embryo transfer. TLI embryos were annotated throughout culture as has been described previously4.

Embryo transfer and follow up

Morphological assessment and grading of embryos was performed according to UK guidelines 8. Embryo transfer was performed on either day 3 or day 5 depending on patient and embryo criteria consistent with NICE guidelines 9). Patients who had two or more good quality embryos on day 3 (at least 6 cells, grade 3 or 4 for both evenness and fragmentation) were offered extended culture to day 5. The number of embryos transferred was based on patient age, embryo quality, the number of previous transfers (if any), and considering the unit’s multiple births minimisation protocol10. Morphology grading alone was used to select embryo(s) to transfer in SC cycles. In TLI cycles, morphological grading was the primary selection tool and in cases where there was more than one morphologically good quality blastocyst, avoidance criteria and the KIDScore D5 score (V2.0 algorithm, Vitrolife, Denmark) were used to exclude embryos with lower implantation potential. The avoidance criteria included unevenness at the two-cell stage, multinucleation at four-cell, direct first cleavage and irregular or reverse cleavage as described in detail previously 4 . Any surplus good quality blastocysts on day 5 or 6 were cryopreserved by vitrification (Cryotop® and KT801 vitrification media, Kitazato, Japan). Embryo utilisation rate was defined as the number of embryos transferred or cryopreserved per 100 zygotes.
Pregnancy was established using a urinary pregnancy test 18 days following oocyte retrieval which was followed by a transvaginal ultrasound scan at 7 weeks’ gestation. The International Glossary on Infertility and Fertility Care and the Core Outcome Measure for Infertility Trials (COMMIT) were used to define the outcomes assessed11,12. A clinical pregnancy was confirmed when cardiac activity was visualised on ultrasound. A multiple pregnancy was diagnosed where there was more than one foetal pole with cardiac activity on ultrasound. Pregnancies which did not reach the stage of a clinical pregnancy were defined as biochemical pregnancy loss and loss of a clinical pregnancy prior to 22 completed weeks was classified as a miscarriage. The birth of at least one viable infant after 22 weeks’ gestation was counted as a live birth.

Frozen embryo transfer

The electronic database was checked manually to ascertain the outcomes of frozen embryo transfer (FET) cycles of women who did not have a live birth after fresh embryo transfer but had surplus blastocysts cryopreserved and women who had embryos electively cryopreserved after oocyte retrieval (freeze-all embryos). FET cycles were either artificial or natural and embryos were warmed rapidly following standard protocols (VT802 warming media, Kitazato, Japan). In artificial cycles the endometrium was primed with oestrogen following a long agonist down-regulation protocol with progesterone supplementation when the endometrium measured >7mm on ultrasound scan. In natural cycles, blastocyst transfer was performed six days after an early morning urine LH surge had been detected. If a live birth had occurred after frozen embryo transfer, this was documented in order to calculate the cumulative live birth rate. All live births reported until July 2020 were included as a reasonable time-frame for assessing the short term cumulative live birth rate given that the study period encompassed November 2015 to December 2017.

Statistical analysis:

The number of oocyte retrievals was taken as the unit of denominator for comparison between TLI and SC incubators. As only the first cycle per woman was included in this study, the per-oocyte retrieval data is identical to per-woman data. Continuous variables were described by mean and standard deviation and compared through mean difference (MD) with 95% confidence intervals (CI) and p values obtained through unpaired t-tests. Categorical variables were described using percentages and comparison performed with Chi-Square test and reporting utilised odds ratio (OR) with 95% CI and p values. For the primary outcomes of live birth and cumulative live birth, adjusting for confounding variables (age and number of oocytes retrieved) was performed using binomial logistic regression and adjusted OR reported. Statistical analyses were performed using IBM SPSS Statistics for Windows, Version 21.0.