P-776 School-age outcomes among IVF and ICSI -conceived children: a causal inference analysis using linked population-wide data

Abstract

Abstract Study question What is the causal effect of method of oocyte insemination; standard in-vitro fertilisation (IVF) or intracytoplasmic sperm injection (ICSI), on childhood development at school entry? Summary answer The developmental outcomes at school entry for children conceived by ICSI, without severe male factor infertility, are equivalent to those conceived via standard IVF. What is known already Over 12 million children have been conceived globally with the assistance of assisted reproductive technology (ART). Within ART, ICSI continues to increase as the most common mode of oocyte insemination, even when not clearly indicated (i.e. without male factor infertility). Large cohort studies have suggested that ART, and ICSI in particular, may be associated with an increase in congenital abnormalities, autism spectrum disorder, developmental delay and intellectual disability. However, the studies are heterogenous and the evidence conflicting. The association between the ICSI technique itself and long-term childhood development remains unclear. Study design, size, duration Causal inference methods were used to analyse observational data in a way that emulates a target randomised clinical trial. For example, patients where standard IVF was not an option, (e.g., due to severe male factor infertility), were excluded. The state-wide cohort included linked maternal and childhood administrative data from Victoria, Australia. To eliminate potential confounding associated with IVF, comparison was made with children conceived via standard IVF (rather than conception without assistance). Participants/materials, setting, methods The study included singleton infants conceived via ART and born between 2005-2013. The primary exposure was conception via ICSI, with those born following conventional IVF insemination as controls. The outcome examined was the Australian Early Development Census (AEDC), a broad assessment of development and well-being performed in Australian schools, every triennium, at school entry (age 4-6). We combined multiple imputation and inverse probability weighting with regression adjustment to estimate population average causal effects. Main results and the role of chance The final cohort included 4,035 IVF and ICSI conceived children across the two outcome cohorts. Linked records were available for 2,468 ICSI children and 1,567 IVF-conceived controls. Maternal characteristics between the two treatment cohorts were similar for age, parity, education, language background, and number of eggs collected during treatment cycle. The ICSI conceived children were generally from residential areas of lower socio-economic status and conception was more likely to involve donated gametes. The unadjusted rate of the primary outcome; AEDC-defined “developmental vulnerability”, was 10.83% among the IVF group and 10.14% among the ICSI group. Given the use of observational data, there were missing data and inherent differences in the covariate profile between exposure cohorts. Multiple imputation, bootstrapping and doubly robust inverse probability weighted regression adjustment modelling was utilised to allow a causal interpretation of results. There was no evidence of a causal effect of ICSI-conception on the risk of AEDC-defined developmental vulnerability at school entry compared with spontaneously conceived children; adjusted risk difference -0.8% (95% CI -2.7% to 1.1%) and adjusted risk ratio 0.92 (95% CI 0.76 to 1.12). The adjustment variable set was determined a priori via a modified Delphi procedure. Limitations, reasons for caution Children who did not attend school due to severe disability were excluded, leading to potential selection bias. Within our study cohort of 2,468 ICSI children, no increased risk of developmental vulnerability was identified. It is possible that unmeasured confounders could have led to bias in estimating the treatment effects. Wider implications of the findings This study, in contrast to previous evidence, suggests that empiric use of ICSI compared with conventional IVF insemination does not affect early childhood developmental outcomes. These findings provide important reassurance for current and prospective parents, and clinicians alike. Trial registration number not applicable