Insulin sensitivity as a predictor of longitudinal changes on body mass index in Brazilian adolescents.

Objectives: We aimed to investigate the effect of insulin sensitivity and insulin resistance status at baseline on longitudinal body mass index, and the possible effect modification by sex.

Methods: This is a secondary analysis of a randomized intervention community trial, in which a subgroup of 84 adolescents, aged between 10 and 12 years, were analyzed. Body weight, height, and body mass index (BMI) were determined before and after 8 months of follow-up. Glucose and serum insulin were examined at baseline and IR was defined based on the homeostasis model assessment-insulin resistance (HOMA-IR), with a cutoff >2.5 for both genders. Linear mixed-effects models were performed to evaluate the influence of HOMA-IR at baseline on BMI changes over time. Models were adjusted for age, pubertal stage, and stratified by sex.

Results: The sample comprised 65.4% of girls and the prevalence of overweight/obesity was 54.7% among girls and 50.0% among boys. The overall prevalence of IR was 75.3%, of which 60.7% for boys and 83.0% for girls. We found an interaction effect by sex (p = .004) for HOMA-IR as a continuous variable, with a decreased BMI rate of change among boys (β = -0.13; p = .03) but not for girls (β = +0.03; p = .36). Longitudinal BMI changes considering IR status at baseline (IR vs. non-IR) did not demonstrate any statistically significant difference for both boys (-0.1 vs. +0.4; p = .28) and girls (+0.7 vs. +1.0; p = .44).

Conclusion: Increased HOMA-IR values at baseline were associated with greater BMI reduction over time among boys but not girls, with no influence of IR status.