Associations of birth weight, plasma metabolome in adulthood and risk of type 2 diabetes

Abstract Aims We aimed to examine the longitudinal associations of birth weight with plasma metabolites in adulthood, and further quantify the proportions of the links between birth weight and incident adult type 2 diabetes (T2D) that were mediated by plasma metabolites. Materials and Methods A total of 62,033 participants with complete nuclear magnetic resonance metabolomics and birth weight data from the UK Biobank were included in this study. Linear regression was used to assess the associations between birth weight and metabolites. Cox regression was used to estimate hazard ratios for T2D associated with metabolites. We further performed mediation analyses to estimate the extent to which metabolites might mediate the association between birth weight and T2D risk. Results Low birth weight was associated with the adverse metabolic responses across multiple metabolic pathways, including lipoprotein subclasses, amino acids, fatty acids (FA), and inflammation. Metabolites associated with higher birth weight tended to be associated with a lower risk of T2D (Pearson correlation coefficient: −0.85). A total of 62 metabolites showed statistically significant mediation effects in the protective association of higher birth weight and T2D risk, including large‐sized very low‐density lipoprotein particles and triglyceride concentrations as well as saturated, and monounsaturated FA and glycoprotein acetyls. Conclusions We identified a range of metabolites that reflect the adult metabolic response to birth weight, some of which might lie on the pathway between birth weight and adult T2D risk.