Maternal and neonatal outcome and children's development after medically assisted reproduction with in-vitro matured oocytes-a systematic review and meta-analysis

Abstract

Background: IVM was implemented in medically assisted reproduction 25 years ago. IVM does not involve controlled ovarian stimulation (COS) and is mainly indicated in patients with a high risk of ovarian hyperstimulation syndrome, in particular in patients with polycystic ovary syndrome (PCOS); it is also an acknowledged option in fertility protection. However, the in-vitro culture of immature oocytes raises concerns over their developmental potential and the putative impact on children's health. Although an increasing number of studies on obstetric and neonatal outcomes of IVM children and their development have been published in recent years, study designs are difficult to compare, since IVM is used in women with various indications and IVM protocols do not follow the same standards.

Objective and rationale: The aim of this systematic review was to evaluate the current evidence from IVM children of an impact of in-vitro culture of immature oocytes. Primary outcome parameters were birthweight and children's development up to the age of 2 years. We also compared pregnancy pathologies and the outcome of IVM children and COS children in relation to maternal indications, in particular PCOS, and to the type of IVM protocols with or without ovulation trigger as the secondary outcome parameters. IVM is an accepted clinical option for many centres; however, a comprehensive analysis of the available data is needed to establish whether the use of human oocytes that are fully matured in vitro is safe for both children and their mothers.

Search methods: Google Scholar and PubMed were used for identifying peer-reviewed original articles and reviews through January 2020. A total of 191 studies were screened and 16 studies were included in the qualitative synthesis. Studies were stratified according to indications, the use of an ovulation trigger and multiplicity.

Outcomes: Birthweights of IVM singletons and multiples were comparable to their respective COS controls: birthweights were also similar if the analysis was restricted to mothers with PCOS. IVM children had a comparable birthweight to COS children, irrespective of whether an ovulation trigger was used in IVM cycles or not. The frequency of gestational diabetes (GD) in singleton pregnancies was comparable between IVM and COS, regardless of infertility background. There was also no difference in GD frequency between IVM and COS, if an hCG ovulation trigger in IVM cycles was used or not. Hypertensive disorders in singleton pregnancies of women with PCOS were significantly more frequent after IVM compared to COS, in particular if IVM cycles were performed only with in-vitro matured oocytes. There was no difference in the preterm birth rate of singleton pregnancies between IVM and COS. Preterm birth rates were still similar if only women diagnosed with PCOS were compared and whether an ovulation trigger in IVM was used or not. The malformation rate in IVM children did not differ in COS children versus children after natural conception. At the age of 2 years, IVM singletons showed similar anthropometric and mental development compared to COS children or children from natural conception.

Wider implications: The higher incidence of hypertensive disorders in IVM pregnancies needs monitoring during pregnancy. Current data on the development of IVM children are encouraging, although the quality of many studies is limited and long-term data beyond 2 years are scarce. Further studies should be based on generally accepted IVM protocols. Studies on long-term outcomes beyond 2 years are needed to search for potential long-time sequelae of IVM.

Keywords: IVM; adverse child outcome; medically assisted reproduction; mental development; neonatal outcome; obstetric outcome.