Obstetrics and Gynaecology - Theses
Permanent URI for this collection
Search Results
1 - 1 of 1
-
ItemLife after the lab - the short-medium- and long-term outcomes after in vitro fertilisation (IVF) in Victoria( 2022-12)BACKGROUND In-vitro fertilisation (IVF) is an artificial mode of conception. Currently, IVF accounts for approximately 5% of births in Australia and has resulted in more than 8 million births world-wide. From an epidemiology perspective, IVF, like many other perinatal events, is a significant perinatal exposure which may have important long-term sequalae. AIMS This thesis aims to examine specific short-, medium- and long-term outcomes for children conceived by IVF. In doing so, this project was intended to contribute to the growing body of literature looking at outcomes among these children and help fill an existing knowledge gap. RESEARCH QUESTIONS To achieve the project aim, two research questions with were explored. First (project 1), what is the association between IVF using donor gametes and specific adverse outcomes of pregnancy, including hypertensive disorders and fetal growth restriction? Second (project 2), what is the estimated causal effect of IVF conception on school-age developmental and educational outcomes for children, compared with unassisted conception? This second question represents the major project of this thesis. METHODS These research questions were explored separately using contrasting methodologies. In project 1, a retrospective observational study was undertaken to examine the association between IVF using different categories of donated gametes (donated sperm, oocyte, or embryo) and 1) hypertensive disorders of pregnancy and 2) fetal growth restriction (FGR)), compared with IVF-conceived pregnancies using autologous gametes (own egg and partner sperm). The second and major project of this thesis was a state-wide, population-based record linkage study aimed at examining the school-age educational and developmental outcomes for children after IVF conception. IVF cycle data was obtained from the three major IVF providers in Victoria from 2005-2014 to identify a cohort of children conceived via IVF. This dataset was then combined by administrative data linkage with four other separate databases. These datasets, included the 1) Victorian Perinatal Data Collection – to enable identification of spontaneously conceived controls from the same time period and to add birth outcome data for the population, 2) The Australian Early Development Census (AEDC) – a global assessment of childhood development at school entry, 3) the National Assessment Program for Literacy and Numeracy (NAPLAN) – the universal Australian psychometric testing metric, and 4) the birth deaths and marriages registry – which was required to link maternal and child data. Project 2 was approached under the framework of causal inference. This approach stipulates that throughout the study design process, data collection, analysis and interpretation, researchers attempt to curtail bias arising from observational data such that under a set of assumptions, causal effects can be estimated. These methods have been developed since the 1970s but are not commonly used in health research. The application of these methods to the research question of interest in project 2 is a significant focus of this thesis. RESULTS Our findings from project 1 suggest a 2.3-fold increased risk of hypertensive disorders of pregnancy, after adjustments with multivariate regression, in pregnancies conceived via donated oocytes. Sperm donation did not appear to be associated with an increased risk (adjusted odds ratio: 0.94, 95% confidence interval: 0.73 to 1.21). The results from project 2 suggest that there is no causal effect of IVF conception on the risk of being developmentally vulnerable at school entry, as assessed by the AEDC outcome metric (average treatment effect (ATE) risk ratio: 0.97, 95% CI 0.77 to 1.25). Similarly, the findings suggest no causal effect of IVF conception on educational outcomes at age 7-9, with similar overall NAPLAN z-score seen in both IVF conceived children and controls (ATE mean difference: 0.003, 95% CI -0.018 to 0.077). The causal model included adjustment of important covariates including maternal age, education, socioeconomics and parity. DISCUSSION The findings of project 1 confirm that pregnancies with a history of egg donation should be managed as high risk. Previous evidence has suggested that “primi-paternity” (the first pregnancy conceived with a specific partner) increases the risk of hypertensive disorders of pregnancy. Our findings suggest that the concept of primi-paternity is only valid during natural conception. IVF is known to be associated with an increased risk of hypertensive disorders of pregnancy – and our findings did not suggest additional risk using donated sperm. As such, we propose that the products within seminal fluid may contribute to development of maternal immunotolerance and that IVF may subvert this natural process of immunotolerance. The findings in this project provide important reassurance for current and future families of IVF-conceived children. Additionally, this thesis demonstrates that via creation of a population-wide, covariate rich dataset and with considered application of these methods, causal effect can be estimated from observational data to determine the school-age educational and developmental outcomes for children after IVF-assisted conception compared to their peers.