Effect of oocyte donor stimulation on recipient outcomes: data from a US national donor oocyte bank.

study questionHow does ovarian stimulation in an oocyte donor affect the IVF cycle and obstetric outcomes in recipients? SUMMARY ANSWER: Higher donor oocyte yields may affect the proportion of usable embryos but do not affect live birth delivery rate or obstetric outcomes in oocyte recipients. WHAT IS KNOWN ALREADY: In autologous oocyte fresh IVF cycles, the highest live birth delivery rates occur when ~15-25 oocytes are retrieved, with a decline thereafter, perhaps due to the hormone milieu, with super-physiologic estrogen levels. There are scant data in donor oocyte cycles, wherein the oocyte environment is separated from the uterine environment. STUDY DESIGN, SIZE, DURATION: This was a retrospective cohort study from 2008 to 2015 of 350 oocyte donors who underwent a total of 553 ovarian stimulations and oocyte retrievals. The oocytes were vitrified and then distributed to 989 recipients who had 1745 embryo transfers. The primary outcome was live birth delivery rate, defined as the number of deliveries that resulted in at least one live birth per embryo transfer cycle. PARTICIPANTS/MATERIALS, SETTING,

methodsThe study included oocyte donors and recipients at a donor oocyte bank, in collaboration with an academic reproductive endocrinology division. Donors with polycystic ovary syndrome and recipients who used gestational carriers were excluded. The donors all underwent conventional ovarian stimulation using antagonist protocols. None of the embryos underwent pre-implantation genetic testing. The average (mean) number of embryos transferred to recipients was 1.4 (range 1-3). MAIN RESULTS AND THE ROLE OF CHANCE: Per ovarian stimulation cycle, the median number of oocytes retrieved was 30 (range: 9-95). Among the 1745 embryo transfer cycles, 856 of the cycles resulted in a live birth (49.1%). There were no associations between donor oocyte yield and probability of live birth, adjusting for donor age, BMI, race/ethnicity and retrieval year. The results were similar when analyzing by mature oocytes. Although donors with more oocytes retrieved had a higher number of developed embryos overall, there was a relatively lower percentage of usable embryos per oocyte warmed following fertilization and culture. In our model for the average donor in the data set, holding all variables constant, for each additional five oocytes retrieved, there was a 4% (95% CI 1%, 7%) lower odds of fertilization and 5% (95% CI 2%, 7%) lower odds of having a usable embryo per oocyte warmed. There were no associations between donor oocyte yield and risk of preterm delivery (<37 weeks gestation) and low birthweight (<2500 g) among singleton infants. LIMITATIONS, REASONS FOR CAUTION: Ovarian stimulation was exclusively performed in oocyte donors. This was a retrospective study design, and we were therefore unable to ensure proportional exposure groups. These findings may not generalizable to older or less healthy women who may be vitrifying oocytes for planned fertility delay. There remain significant risks to aggressive ovarian stimulation, including ovarian hyperstimulation. In addition, long-term health outcomes of extreme ovarian stimulation are lacking. Lastly, we did not collect progesterone levels and are unable to evaluate the impact of rising progesterone on outcomes. WIDER IMPLICATIONS OF THE

findingsLive birth delivery rates remain high with varying amounts of oocytes retrieved in this donor oocyte model. In a vitrified oocyte bank setting, where oocytes are typically sent as a limited number cohort, recipients are not affected by oocyte yields. STUDY FUNDING/COMPETING INTEREST(S): Additional REDCap grant support at Emory was provided through UL1 TR000424. Dr. Audrey Gaskins was supported in part by a career development award from the NIEHS (R00ES026648).