Assessing causality in the association between maternal pre-pregnancy obesity and child neurodevelopment: observational and mendelian randomization analyses

摘要

Background: Observational studies have shown that high maternal pre-pregnancy body mass index (BMI) may impair infant neuropsychological development; however, it is unclear whether reported associations are due to intrauterine mechanisms or are consequence of residual confounding by socio-economic factors. Objective: We aimed to determine whether maternal pre-pregnancy BMI causally influences child neurodevelopment by performing an observational analysis in 24 birth cohorts (> 100,000 mother/infant pairs) and a Mendelian randomization analysis in 10 of these cohorts (> 10,000 mother/infant pairs). Methods: We will use maternal BMI genetic score (30 established single nucleotide polymorphisms) as an instrumental variable for maternal pre-pregnancy BMI. We will also use paternal BMI as a negative control exposure and test the role of child BMI genetic score. We will assess infant cognition, psy-chomotor, and behaviour including attention-deficit hyper-activity disorder and autism symptoms from up to 11 years of age. Bias from pleiotropy will be tested using Egger regression. Results: In observational analysis, preliminary results using data from the INMA (Environment and Childhood) Spanish cohort (n = 1402) revealed that each kg/m increase of maternal pre-pregnancy BMI was associated with a reduction of 0.23 infant cognitive scores (95% CI: -0.42, -0.04) at the age of 1.5 years. In Mendelian randomization analysis, each kg/m increase of maternal BMI genetic score was associated with a reduction of 1.56 infant cognitive scores (95% CI: -3.06, -0.05) at the same age. Egger analysis showed no bias from pleiotropy. Conclusions: This study using different population settings and Mendelian randomization approach will serve to disentangle whether there is an intrauterine effect of maternal pre-pregnancy obesity on offspring neurodevelopment.