ESPE Abstracts

Background: Maternal triglycerides, fatty acids, and their metabolites together with sugars and metabolic intermediates may play a role in newborn outcomes. Furthermore, maternal amino acids, acylcarnitines, lipids and fatty acids and their metabolites are associated with cord C-peptide independent of maternal BMI and glycemia, highlighting the potential contribution of maternal metabolites to offspring outcomes. This study examined the associations of maternal metabolites during pregnancy with childhood glucose outcomes. Associations between cord blood metabolites and childhood glucose outcomes were also assessed in a subset of offspring.

Methods: Phenotype data from the multi-ethnic Hyperglycemia and Adverse Pregnancy Outcome (HAPO) Study and HAPO Follow Up Study (FUS) were used together with targeted and non-targeted metabolomics data obtained fasting and 1-hr after a glucose load at ~28 weeks gestation. Maternal metabolite association analyses included 2,145 mother offspring pairs while cord blood association analyses included 808 offspring. Multiple logistic and linear regression were used to examine associations between primary predictors, maternal metabolites and cord blood metabolites, and child glucose outcomes. Model 1 included field center, child age, sex, and pubertal status with sex by Tanner stage interaction term, maternal variables at pregnancy OGTT (age, height, mean arterial pressure, parity, smoking, alcohol, and gestational age), and child’s family history of diabetes in first-degree relatives. Model 2 included Model 1 + maternal BMI at pregnancy OGTT, Model 3 included Model 1 + maternal glucose z score sum at pregnancy OGTT, and Model 4 included Model 1 + maternal BMI and glucose z score sum at pregnancy OGTT.

Results: In the fully adjusted model (Model 4), no associations of fasting maternal metabolites and childhood glucose outcomes were identified. 1-hr maternal methionine, acetylcarnitine (C2), and hypoxanthine were positively associated with child HbA1c (b =0.02, 95%CI: 0.01-0.04, P<0.001), 2-hr glucose (b=1.69, 95%CI: 0-8-2.58, P<0.001), and insulinogenic index (b=0.02, 95%CI: 0.01-0.03, P<0.001) respectively. Multiple maternal fasting and 1-hr metabolites were significantly associated with childhood adiposity outcomes in Model 1 but were not significant after adjusting for maternal BMI and/or maternal glycemia. There were no significant associations between cord blood metabolites and childhood glucose outcomes.

Conclusions: Only maternal methionine, acetylcarnitine (C2), and hypoxanthine were associated with childhood glucose outcomes after adjusting for maternal BMI and glucose. Additional metabolites were significantly associated with child glucose outcomes after adjusting for maternal BMI but not maternal glucose, suggesting that maternal glucose accounts for these associations.