Live birth and perinatal outcomes using cryopreserved oocytes: an analysis of the Human Fertilisation and Embryology Authority database from 2000 to 2016 using three clinical models

Abstract STUDY QUESTION Are live birth (LB) and perinatal outcomes affected by the use of frozen own versus frozen donor oocytes? SUMMARY ANSWER Treatment cycles using frozen own oocytes have a lower LB rate but a lower risk of low birth weight (LBW) as compared with frozen donor oocytes. WHAT IS KNOWN ALREADY A rising trend of oocyte cryopreservation has been noted internationally in the creation of donor oocyte banks and in freezing own oocytes for later use in settings of fertility preservation and social egg freezing. Published literature on birth outcomes with frozen oocytes has primarily utilised data from donor oocyte banks due to the relative paucity of outcome data from cycles using frozen own oocytes. STUDY DESIGN, SIZE, DURATION This was a retrospective cohort study utilising the anonymised database of the Human Fertilisation and Embryology Authority, which is the statutory regulator of fertility treatment in the UK. We analysed 988 015 IVF cycles from the Human Fertilisation and Embryology Authority (HFEA) register from 2000 to 2016. Perinatal outcomes were assessed from singleton births only. PARTICIPANTS/MATERIALS, SETTING, METHODS Three clinical models were used to assess LB and perinatal outcomes: Model 1 compared frozen own oocytes (n = 632) with frozen donor oocytes (n = 922); Model 2 compared frozen donor oocytes (n = 922) with fresh donor oocytes (n = 24 706); Model 3 compared first cycle of fresh embryo transfer from frozen donor oocytes (n = 917) with first cycle of frozen embryo transfer created with own oocytes and no prior fresh transfer (n = 326). Preterm birth (PTB) was defined as LB before 37 weeks and LBW as birth weight <2500 g. Adjustment was performed for confounding variables such as maternal age, number of embryos transferred and decade of treatment MAIN RESULTS AND THE ROLE OF CHANCE The LB rate (18.0% versus 30.7%; adjusted odds ratio (aOR) 0.61, 95% CI 0.43–0.85) and the incidence of LBW (5.3% versus 14.0%; aOR 0.29, 95% CI 0.13–0.90) was significantly lower with frozen own oocytes as compared with frozen donor oocytes with no significant difference in PTB (9.5% versus 15.7%; aOR 0.56, 95% CI 0.26–1.21). A lower LB rate was noted in frozen donor oocyte cycles (30.7% versus 34.7%; aOR 0.69, 95% CI 0.59–0.80) when compared with fresh donor oocyte cycles. First cycle frozen donor oocytes did not show any significant difference in LB rate (30.1% versus 19.3%; aOR 1.26, 95% CI 0.86–1.83) or PTB, but a higher incidence of LBW (17.7% versus 5.4%; aOR 3.77, 95% CI 1.51–9.43) as compared with first cycle frozen embryos using own oocytes. LIMITATIONS, REASONS FOR CAUTION The indication for oocyte freezing, method of freezing used (whether slow-freezing or vitrification) and age at which eggs where frozen were unavailable. We report a subgroup analysis of women using their own frozen oocytes prior to 37 years. Cumulative LB rate could not be assessed due to the anonymous nature of the dataset. WIDER IMPLICATIONS OF THE FINDINGS Women planning to freeze their own eggs for fertility preservation or social egg freezing need to be counselled that the results from frozen donor egg banks may not completely apply to them. However, they can be reassured that oocyte cryopreservation does not appear to have a deleterious effect on perinatal outcomes. STUDY FUNDING/COMPETING INTEREST(S) No specific funding was sought for the study. The authors have no relevant conflicts of interest. TRIAL REGISTRATION NUMBER N/A