Consistent and predictable delivery rates after oocyte vitrification : an observational longitudinal cohort multicentric study

Background: An efficient method for cryopreservation of human oocytes may offer solutions to legal and ethical problems in routine infertility programs and may also be used for fertility preservation for medical and social reasons. Methods: We conducted an observational longitudinal cohort multicentric study to investigate the efficacy and reproducibility of oocyte cryopreservation outcomes in IVF/ICSI cycles. Moreover, the effects of patient and cycle characteristics on the delivery rate (DR) were analyzed. Results: In 486 cycles performed in 450 couples, 2721 oocytes were warmed and 2304 of them survived cryopreservation (84.7%). Of the 2182 oocytes subjected to ICSI, the rates of fertilization and development to top-quality embryos were 75.2 and 48.1%, respectively. A total of 128 deliveries were obtained (26.3% per cycle and 29.4% per transfer) for 450 patients (28.4%) and 147 babies were live born from 929 embryos transferred (15.8%). The forward logistic regression analysis on a per patient basis showed that female age [odds ratio (OR): 0.93, 95% confidence interval (CI): 0.88–0.98], number of vitrified oocytes (OR: 1.08, 95% CI: 1.01–1.17) and the day of transfer (OR: 1.97, 95% CI: 1.14–3.42) influenced DR. By recursive partitioning analysis, it can be estimated that more than eight oocytes vitrified are required to improve the outcome (22.6 versus 46.4% DR, respectively). When fewer oocytes are available in women aged .38 years, results are dramatically reduced (12.6 versus 27.5% DR, respectively). Conversely, when .8 oocytes are available, blastocyst culture represents the most efficient policy (62.1% DR; data from one center only). Conclusions: Oocyte vitrification is an efficient and reliable approach, with consistent results between centers and predictable DRs. It should be applied routinely for various indications. A predictive model is proposed to help patient counselling and selection.