A prospective randomized controlled trial investigating the effect of artificial shrinkage (collapse) on the implantation potential of vitrified blastocysts.

STUDY QUESTION What is the effect of artificial shrinkage by laser-induced collapse before vitrification on the implantation potential after transfer of vitrified-warmed blastocysts? SUMMARY ANSWER The artificial shrinkage by laser-induced collapse did not significantly increase the implantation rate per transferred collapsed blastocyst (37.6%) compared with non-collapsed blastocysts (28.9%) [odds ratio (OR): 1.48, 95% confidence interval (CI): 0.78-2.83]. WHAT IS KNOWN ALREADY Retrospective studies have demonstrated that artificial shrinkage of the blastocyst prior to vitrification can have a positive effect on blastocyst survival after warming. A recent study found a similar survival rate but higher implantation rate for collapsed blastocysts. So far, no randomized controlled trial has been conducted to investigate the implantation potential of collapsed blastocysts. STUDY DESIGN, SIZE, DURATION Prospective randomized trial. Patients were recruited from December 2011 until April 2014 and warming cycles were included until July 2014. Patients were randomized in the fresh cycle if blastocysts were available for vitrification and were allocated to the study or control arm according to a computer-generated list. In the study group, blastocysts underwent laser-induced collapse before vitrification. In the control group, blastocysts were vitrified without collapsing. PARTICIPANTS/MATERIALS, SETTING, METHODS In total, 443 patients signed informed consent and 270 patients had blastocysts vitrified. One-hundred and thirty-five patients were allocated to the study group and 135 to the control group. Sixty-nine patients from the study group and 69 from the control group returned for at least one warming cycle in which 85 and 93 blastocysts were warmed in the first cycle, respectively. Primary outcome was implantation rate per embryo transferred in the first warming cycle. Secondary outcomes were survival and transfer rates, blastocyst quality after warming, clinical pregnancy rate and implantation rate per warmed blastocyst. Blastocysts were vitrified-warmed one by one using closed vitrification and one or two blastocysts were transferred per warming cycle. MAIN RESULTS AND THE ROLE OF CHANCE We calculated that the group sample sizes of 80 embryos in the collapse group and 80 embryos in the control group were needed to achieve 80% power to detect a difference between the group proportions of +20% with P < 0.05. In the study group, 69 first warming cycles resulted in 69 transfers with 1.2 blastocysts (n = 85) transferred. In the control group, an average of 1.3 blastocysts (n = 83) were transferred in 67 out of 69 warming cycles. Implantation rates per embryo transferred in the first warming cycle were not different between both groups (38 versus 29%, OR: 1.48; 95% CI: 0.78-2.83), neither was the implantation rate per warmed embryo (38 versus 26%, OR: 1.74; 95% CI: 0.92-3.29). When all warming cycles were considered (n = 135 in each group), survival rate after collapse was significantly higher compared with the control group (98.0 versus 92.0%, OR: 4.25; 95% CI: 1.19-15.21). Furthermore, a higher percentage of high-quality blastocysts (36.3 versus 23.5%, OR: 1.86; 95% CI: 1.12-3.08) and hatching blastocysts (19.2 versus 5.4%, OR: 4.18; 95% CI: 1.84-9.52) were found compared with the control group. LIMITATIONS, REASONS FOR CAUTION The study lasted more than 2.5 years since fewer patients than expected returned for a warming cycle because of the high ongoing pregnancy rates in the fresh IVF/ICSI cycle. WIDER IMPLICATIONS OF THE FINDINGS Although no significant higher implantation rate was found after collapse, the better survival and post-warm embryo quality convinced us to recognize a clinical benefit of artificial shrinkage and to implement it in routine vitrification practice. TRIAL REGISTRATION NUMBER NCT01980225, www.clinicaltrials.gov. The first patient was included November 2011 and the study was registered October 2013.