Prenatal exposome and childhood general, visceral and liver fat assessed by magnetic resonance imaging

Abstract

TPS 711: The exposome, Exhibition Hall, Ground floor, August 26, 2019, 3:00 PM - 4:30 PM Background/Aim: Studies on childhood adiposity have mostly assessed the risks of single or few exposures. We used an exposome approach to assess the associations of several lifestyle and environmental prenatal exposures with child general, visceral and liver fat. Methods: In a population-based prospective cohort study among 3,223 mother-child pairs, we assessed 46 lifestyle and environmental prenatal exposures and fat mass index by dual-energy X-ray absorptiometry, and visceral fat index and liver fat fraction by magnetic resonance imaging at 10 years. We used an exposure-wide association study (ExWAS) considering exposures individually in linear regression models, and a variable selection method, the deletion-substitution-addition (DSA) algorithm, considering all exposures simultaneously to build a final multi-exposure model. Results: In ExWAS, 21 exposures (including maternal body mass index (BMI), smoking habits, alcohol consumption, dietary intake and air pollutants) were associated with at least one of the child fat outcomes, but only the associations of maternal BMI, alcohol consumption and vitamin D levels remained significant after multiple testing correction (p<0.002). In the final DSA multi-exposure model, maternal BMI was positively associated whereas alcohol consumption was negatively associated with all child fat outcomes. Positive associations were observed for linoleic acid levels with fat mass index, for smoking habits and γ-linolenic acid levels with fat mass and visceral fat indices, and for caffeine intake and overall psychological distress with visceral fat index and liver fat fraction. Negative associations were observed for overall diet quality and elemental potassium concentration with fat mass index, for vitamin D levels with fat mass and visceral fat indices and for nitrogen dioxide concentration with liver fat fraction. Conclusions: Our exposome approach identified mainly lifestyle prenatal exposures that are associated with child general, visceral and liver fat.