Abstract

This study is aimed at exploring the associations of lipid metabolites with insulin resistance (IR) and hypertriglyceridemia among children and adolescents. We recruited 630 young individuals aged 4 to 17 years and divided them into two groups: low-IR and high-IR. MS metabolomic analysis was used to measure 75 lipid metabolites in plasma. IR-specific lipid metabolites are illustrated via volcano plot. We used regression models to evaluate the relationships between IR-specific lipid metabolites and IR and hypertriglyceridemia. Furthermore, we applied bootstrapping to test the mediating effect of triglyceride (TG) on the relationship between IR-specific lipid metabolites and IR. In our study, there were 379 individuals in the low-IR group and 251 individuals in the high-IR group. We identified 9 children-specific IR-associated lipid metabolites, of which 16:0–18:1 PE, 18:0–18:2 PE, and 18:0 PE were positively associated with the risk of IR (OR = 1.909, 95% CI = 1.084–3.361, p = 0.025; OR = 1.890, 95% CI = 1.065–3.355, p = 0.030; respectively). Additionally, correlation heat maps revealed associations between TG and 16:0–18:1 PE, 26:0 Lyso PC, and 22:1 (Cis) PC. Further multivariate binary logistics regression analysis revealed that 16:0–18:1 PE was significantly associated with hypertriglyceridemia (OR = 1.949, 95% CI = 1.212–3.133, p = 0.006). TG plays a partial mediating role in the relationship between 16:0–18:1 PE and HOMA-IR, with the proportion of the mediation effect relative to the total effect being 18.29% (95% CI of the mediating effect for TG was 0.004–0.082). Increased levels of 16:0–18:1 PE, 18:0–18:2 PE, and 18:0 PE are associated with increased risks of IR. Increased levels of 16:0–18:1 PE are associated with increased risks of hypertriglyceridemia. TG may mediate the impact of IR driven by 16:0–18:1 PE.