P-760 Monozygotic twin rate among IVF centers: A multicenter analysis of data from 18 Italian units

Abstract Study question Could factors inherent to individual centers, related to local clinical or laboratory variables, play a role in the increased incidence of monozygotic twins in IVF? Summary answer The rate of MZT did not significantly vary among IVF centers What is known already The risk of monozygotic twins (MZT) is increased in couples undergoing in vitro fertilization (IVF). Several systematic reviews have investigated the possible determinants linked to IVF, but results obtained have been inconclusive. Associations have been identified with the prolonged culture at blastocyst stage and young female age. Conventional IVF compared to ICSI, and assisted hatching emerged as possible additional risk factors, but evidence is controversial. Meta-analyses, however, cannot provide robust evidence because of the difficulty in performing multivariate analyses. Moreover, available investigations were retrospective and involved relatively small sample sizes hampering the quality of the collected data. Study design, size, duration This is a multicenter retrospective cohort study using the Italian ART National Registry database and involving the Centers reporting data from individual IVF cycles from 2015 to 2019. Eighteen IVF centers were included with a total of 87,076 IVF cycles reported during the study period. The primary aim of the study was to investigate whether the incidence of MZT differed among centers. Participants/materials, setting, methods Only single embryo transfer cycles were considered. Women who had sex-discordant twin deliveries were excluded. MZT rate was calculated as the number of multiple pregnancies out of the total number of clinical pregnancies. A binomial distribution model was used to determine the 95% CI of the frequency of MZT. The rates among centers were deemed heterogeneous if the 95% CI in at least one center did not include the mean MZT rate. Main results and the role of chance Ten centers reported data from 2015 to 2019, eight from 2017 to 2019. An analysis of the main characteristics of the centers revealed considerable variability. The total number of clinical pregnancies fulfilling our inclusion criteria were 10,440. Of these, 7 were found to be twin pregnancies with discordant gender and were excluded, leaving 10,433 for data analyses. The number of cases included per center varied between 11 and 2,823. Overall, 162 pregnancies were found to be multiple: 48 ended in a miscarriage, 28 spontaneously reduced to singletons and 86 women delivered two or more newborns. The vast majority were twins (n = 160), one was triplet, and one was quadruplet. Considering the denominator of 10,433 pregnancies, the rate of MZT was thus 1.5% (95%CI: 1.3 - 1.8%). The MZT rate among centers varied between 0% (95%CI: 0.0 - 25.9%) and 3.2% (95%CI: 1.3 - 8.1%). All the 95%CIs of the rates of MZT include the 1.5% common rate, thus rejecting the hypothesis that this incidence significantly differed among centers. Spearman correlations between the rate of MZT and variables associated with individual center policies could not identify any significant association. Even when restricting the analyses only to the largest centers, no association emerged. Limitations, reasons for caution Our study was limited by its retrospective nature. Furthermore, some centers provided less than 100 cases for data analysis and were therefore poorly informative. External confirmation from larger registries is therefore required. Wider implications of the findings This study provides evidences that the rate of MZT did not differ among IVF centers. Furthermore, no clinical and laboratory variables inherent to individual centers were identified in the determinism of IVF-associated MZT. Further studies are needed to identify the causes of increased risk of MZT in IVF practice. Trial registration number not applicable