DXA-based Fat Mass With Risk of Worsening Insulin Resistance in Adolescents: A 9-Year Temporal and Mediation Study.

Context: Surrogate measures of childhood and adolescent obesity have impaired the understanding of the relationship of body composition with insulin resistance in the young population.

Objective: We aim to examine the longitudinal associations of directly measured total fat mass, trunk fat mass, and lean mass with the risk of hyperglycemia, hyperinsulinemia, and insulin resistance from ages 15 to 24 years, the mediation path through which lipids and inflammation influence insulin resistance, and whether increased fat mass temporally precede insulin resistance.

Methods: We studied 3160 adolescents from the Avon Longitudinal Study of Parents and Children (ALSPAC), UK birth cohort, who had complete dual-energy x-ray absorptiometry measure and fasting blood samples at age 15 years and repeated measures at ages 17- and 24-years clinic visit. Fasting glucose greater than 6.1 mmol/L, insulin greater than 11.78 mU/L, and homeostatic model assessment for insulin resistance (HOMA-IR) greater than or equal to the 75th percentile were categorized as hyperglycemia, hyperinsulinemia, and high insulin resistance, respectively. Longitudinal associations were examined with generalized logit-mixed-effect models, while mediation and temporal path analyses were examined using structural equation models, adjusting for cardiometabolic and lifestyle factors.

Results: Among 3160 participants (51% female), fat mass and lean mass increased linearly both in males and females, while glucose, insulin, and HOMA-IR had a U-shaped course from age 15 through 24 years. After full adjustment, each 1-kg cumulative increase in total fat mass (odds ratio 1.12 [95% CI, 1.11-1.13]) and trunk fat mass (1.21 [1.19-1.23]) from ages 15 through 24 years were associated with a progressively worsening risk of high insulin resistance as well as hyperglycemia and hyperinsulinemia. The association of increased total fat mass with increased insulin resistance was partly mediated by triglycerides (9% mediation). In the temporal path analysis, higher total fat mass at age 15 years was associated with higher insulin resistance at age 17 years, but not vice versa. Higher total fat mass at age 17 years was bidirectionally associated with higher insulin resistance at 24 years.

Conclusion: Mid-adolescence may be an optimal time for interrupting the worsening fat mass-insulin resistance pathologic cycle and attenuating the risk of progressively worsening metabolic dysfunction before young adulthood.