Use of donor sperm in addition to oocyte donation after repeated implantation failure in normozoospermic patients does not improve live birth rates.

STUDY QUESTION Does switching to donor semen after at least three failed oocyte donation (OD) cycles with the partner normozoospermic semen increase the live birth rate in a subsequent OD cycle? SUMMARY ANSWER Switching to donor semen after at least three failed OD cycles with the partner normozoospermic semen does not increase the live birth rate. WHAT IS ALREADY KNOWN In some patients, a viable pregnancy cannot be achieved after several OD cycles, despite normal diagnostic findings for the couple. The ESHRE Capri Workshop Group indicates that, in order to improve reproductive outcomes, a semen donation can be offered after three failed ICSI cycles. STUDY DESIGN, SIZE, DURATION A retrospective cohort analysis of fourth and fifth OD cycles with either the partner's normozoospermic semen (OD) or double-donation cycles (DD), performed between January 2011 and December 2014 in a private fertility center. These couples did not have a known male factor. PARTICIPANTS/MATERIALS, SETTING, METHOD The study included 228 cycles (159 OD and 69 DD). The fertilization method was ICSI in all cycles and embryos were transferred fresh. Fertilization rates were compared between groups using ANOVA while pregnancy outcomes were compared using Chi-square tests. Effect of DD on pregnancy outcomes was further analyzed using a logistic regression model adjusted for recipient's age and BMI, number of embryos transferred, day of embryo transfer and morphological embryo quality score. MAIN RESULTS AND THE ROLE OF CHANCE There was no difference in live birth rate between the DD and OD groups (38.2 versus 35.8%, P = 0.73), even after adjustment for confounding factors (odds ratio 1.41, 95% confidence interval 0.72, 2.76; P = 0.31). Rates of biochemical pregnancy (52.2 versus 54.1%, P = 0.79), clinical pregnancy (41.2 versus 45.9%, P = 0.51) and ongoing pregnancy (38.2 versus 37.1%, P = 0.87) were not different between the DD and the OD groups, as well as fertilization rate (75.3 versus 75.2%, P = 0.97). The DD and OD groups were comparable at baseline in all demographic and cycle variables analyzed (recipient's BMI, number of transferred embryos and embryo quality) with the exception of recipient's age (42.3 in DD versus 44.1 in OD, P = 0.005), and day of embryo transfer (56.5% of DD and 83.6% of OD embryo transfers were performed on blastocyst stage, P < 0.001); both variables were adjusted for in the multivariate analysis. LIMITATIONS, REASONS FOR CAUTION The main limitations of this study are its retrospective nature, the relatively small sample size, the transfer of embryos of different developmental stages and the lack of extensive molecular testing, such as sperm DNA fragmentation test, in normozoospermic patients. WIDER IMPLICATIONS OF THE FINDINGS After excluding several causes for the failed OD cycles, the partner's normozoospermic semen was a common factor in all of them. Nevertheless, the change to a donor's semen does not seem to improve the reproductive outcomes in the subsequent cycle. STUDY FUNDING/COMPETING INTERESTS No extra-mural funding was obtained for this study. There are no conflicts of interest to declare. TRIAL REGISTRATION NUMBER NA.