Abstract 12159: Metabolite Profiles Are Associated With Metabolically Unhealthy Obesity in Children and Adolescents

Abstract

Background: Metabolite profiling has been used to identify mechanisms of cardiometabolic disease (CMD). Whether metabolites can identify metabolic healthy obesity (MHO) versus unhealthy obesity (MUO) remains unclear. Hypothesis: We hypothesized that specific metabolite profiles would be associated with adverse cardiometabolic phenotypes in non-diabetic overweight/obese youth. Methods: We performed untargeted metabolomics (81 polar metabolites) in 90 overweight/obese youth (age 15.5±4.8 years, 54% female, BMI 33.8±10.0 kg/m 2 ), and used principal components analysis to identify metabolite clusters. Metabolically unhealthy obesity (MUO) was defined as one of the following: homeostatic model of insulin resistance (HOMA-IR) ≥ 3.16, hypertension (systolic BP ≥140 mmHg or diastolic BP ≥90 mmHg), triglycerides ≥ 150 mg/dl, HDL <40 mg/dl or glucose ≥100. We used logistic regression to estimate the association between metabolite clusters and MUO. Results: Five principal components (PC) accounted for 61.9% variance in polar metabolites. Polar PC2 contained aromatic amino acids (phenylalanine, tyrosine), branched-chain amino acids (e.g., leucine, valine, isoleucine), and their downstream metabolites (e.g., glutamine) and was associated with MUO (odds ratio OR=1.92, 95% confidence interval CI 1.06-3.48, P=0.03). Conclusions: We identified patterns of metabolites previously implicated in diabetes risk in adults (e.g., branched chain and aromatic amino acids) associated with CMD phenotypes in overweight/obese youth. Common metabolic pathways contribute to heterogeneity in cardiometabolic health in pediatric obesity as in adults.