Objectives: To examine the association and predictive capacity of adiponectin levels, HOMA-AD and HOMA-IR indexes with metabolic risk markers in children and adolescents.
Methods: A cross-sectional study was conducted with 691 children and adolescents (7-14 y), of both sexes. Demographic (sex, age), anthropometric (weight, height, body mass index, waist circumference, body fat), biochemical [total cholesterol, high density lipoprotein (HDL), low density lipoprotein (LDL), triglycerides, fasting glycemia, insulin and adiponectin] and clinical parameters (arterial blood pressure) were analyzed.
Results: In multiple linear regression models, metabolic risk were analyzed in relation to adiponectin levels, HOMA-AD and HOMA-IR. ROC curve analysis was used to define the cut-off for metabolic syndrome for each method studied. Adiponectin level was inversely correlated with weight (r = -0.12; p = 0.01), waist circumference (WC) (r = -0.12; p = 0.01), and triglycerides (r = -0.11; p = 0.02); it was directly correlated with HDL (r = 0.10; p = 0.03) only in the adolescents. In the final linear regression model, after adjustment, only triglycerides (p = 0.03) and HDL (p = 0.04) remained significant. However, HOMA-AD and HOMA-IR were associated with metabolic risk and were the most suitable methods for metabolic syndrome screening in both age groups. For children, independent variables explained 16.0% and 14.5% of HOMA-AD and HOMA-IR, respectively. For adolescents, R2 was higher in HOMA-AD and HOMA-IR models (R2adjusted = 31.9% and R2adjusted = 29.6%, respectively).
Conclusions: HOMA-AD and HOMA-IR are better explained by metabolic markers than adiponectin levels.
Keywords: Adiponectin; Adolescents; Children; HOMA-AD; HOMA-IR.