Mediating role of inflammatory biomarkers in the causal effect of body composition on glycaemic traits and type 2 diabetes

Abstract Objective The aim was to investigate the mediating role of inflammatory biomarkers in the causal effect of body composition on glycaemic traits and type 2 diabetes. Methods A retrospective observational study and a Mendelian randomization (MR) study were used. Observational analyses were performed using data from 4717 Chinese children and adolescents aged 6–18 years who underwent dual‐energy X‐ray absorptiometry for body composition. MR analyses were based on summary statistics from UK Biobank, deCODE2021, Meta‐Analysis of Glucose and Insulin‐Related Traits Consortium (MAGIC) and other large consortiums. Inflammatory biomarkers included leptin, adiponectin, osteocalcin, fibroblast growth factor 23 (FGF23) and parathyroid hormone (PTH). Results In a retrospective observational study, increased fat mass had a positive effect on homeostasis model assessment of insulin resistance (HOMA‐IR) and homeostasis model assessment of pancreatic beta cell function (HOMA‐β) through FGF23, whereas fat‐free mass produced the opposite effects. PTH and osteocalcin played significant roles in the association of fat mass and fat‐free mass with fasting glucose, fasting insulin and HOMA‐IR (all p < 0.05). Mediation MR results indicated that childhood body mass index affected glycaemic traits through leptin and adiponectin. There existed a causal effect of fat‐free mass on type 2 diabetes via FGF23 (indirect effect: OR [odds ratio]: 1.14 [95% CI, confidence interval: 1.01–1.28]) and adiponectin (OR: 0.85 [95% CI: 0.77–0.93]). Leptin mediated the causal association of fat mass (indirect effect: β: −0.05 [95% CI: −0.07, −0.02]) and fat‐free mass (β: 0.03 [95% CI: 0.01, 0.04]) with fasting glucose. Conclusions Our findings suggest that different body compositions have differential influences on glycaemic traits and type 2 diabetes through distinct inflammatory biomarkers. The findings may be helpful in tailoring management of body composition based on inflammatory biomarkers with different glycaemic statuses.