P–251 To collapse or not to collapse blastocysts before vitrification? A matched case-control study on single vitrified-warmed blastocyst transfers

Study question Does laser-induced artificial blastocoel collapse result in better blastocyst cryopreservation survival and higher live birth rate (LBR) in comparison with intact counterparts? Summary answer Compared to vitrification of intact blastocysts, collapsed blastocysts resulted in higher survival and for 5% higher LBR. Neonatal outcomes were comparable in both groups. What is known already Blastocysts have long been considered a stage that is suboptimal for freezing-thawing procedures due to their high fluid content and different cell types. The development of a modified vitrification technique has enabled blastocysts to better survive cryopreservation compared to a slow freezing procedure. Many studies on the optimization of cryopreservation of blastocysts have mentioned the need for artificial collapsing of the blastocoel prior to cryopreservation, thereby reducing the risk of intracellular ice-crystals formation. However, the effectiveness of artificial collapsing on blastocyst survival rate, single vitrified-warmed blastocyst transfer (SVBT) outcome and on safety of such intervention remains to be confirmed. Study design, size, duration A retrospective matched case-control study of transfers of single blastocysts being artificially collapsed (case) or intact (control) before vitrification. A sample size of 306 cycles in both arms was needed to achieve 80% power to detect a difference between the groups of 10% with P < 0.05. Controls were matched to cases on a 1:1 ratio by female age, parity, fresh and frozen cycle protocol, blastocyst age and quality, getting 309 pairs of cases and controls. Participants/materials, setting, methods Artificial collapsing was introduced into clinical practice gradually. In fresh IVF cycles (performed in university clinic from 2012 until 2014) with supernumerary blastocysts, half of the blastocysts were randomly selected before vitrification for laser-induced artificial collapsing. The other half was vitrified in intact form. Only the first transfers of a single vitrified-warmed blastocyst (n = 818) were included in the study. By matching, 309 pairs of collapsed (study) and intact (control) SVBTs were identified. Main results and the role of chance Both groups were comparable by their characteristics in indications, female age, type and length of ovarian hyperstimulation, insemination method in fresh cycle, protocol for warmed blastocyst transfer, blastocyst quality and day of blastocyst vitrification. Survival rates in case and control group ((309/316) 97.8% and (309/323) 95.7%; P = 0.13) were comparable, but optimal survival rates (100% survival and re-expansion after warming) was significantly higher in artificial collapse group ((247/316) 78.2% and (225/323) 69.7%; P = 0.01). Clinical pregnancy rates ((120/309) 38.8% and (110/309) 35.6%; P = 0.4), miscarriage rates ((15/120) 12.5% and (24/110) 21.8%; P = 0.06) and LBR per transfer ((100/309) 32.4% and (85/309) 27.5%; P = 0.19) or LBR per warmed blastocyst ((100/316) 31.6% and (85/323) 26.3%; P = 0.14) were not statistically different between case and control groups. Since the study was powered to detect a 10% difference, the possibility of type 2 error cannot be excluded. Perinatal outcomes were available for 175 live births. There were 10.5% (10/95) preterm births in the study group vs. 16.3% (13/80) in control group (P > 0.05). Birth weights (3,308 g (SD 592 g) vs 3,308 g (SD 738 g) and sex ratio (50.7% vs 49.2% boys) were also comparable between both groups (P > 0.05). There were no major malformations detected in the study population. Limitations, reasons for caution The research is retrospective, but the cycles from both groups were performed in the same time period. The groups were balanced according to all possible confounders. Blastocysts for vitrification were first categorized by quality groups and embryos from each category were randomized for collapsing or for remaining intact. Wider implications of the findings: No significant difference was found in live births by this sample size. Nevertheless, increasing the success by 5% with the introduction of artificial collapsing can be an important step towards optimizing of blastocyst cryopreservation. To confirm a 5% improvement in results, a sample size of > 2500 cases would be needed. Trial registration number The study has been approved by the National Ethics Committee of the Republic of Slovenia (0120–204/2016–2).