High HOMA-IR, adjusted for puberty, relates to the metabolic syndrome in overweight and obese Chilean youths

Abstract
To determine how the homeostasis model assessment of insulin resistance (HOMA-IR) is related to metabolic risk in a sample of overweight and obese Chilean youths accounting for Tanner stage. A cross-sectional study assessing 486 overweight and obese youths (aged 5-15 years) recruited from the University of Chile, Pediatric Obesity Clinic. We measured anthropometry, Tanner stage, HOMA-IR, and laboratory tests related to metabolic risk. HOMA-IR was categorized by quartile for children (Tanner stages I and II) and adolescents (Tanner stage III and above) from a normative Chilean sample. Children and adolescents with HOMA-IR in the highest quartile were likely to have higher body mass index (BMI) Z-scores, elevated waist circumference, systolic and diastolic blood pressure, and triglycerides and low high-density lipoprotein. HOMA-IR had good negative predictive value for characteristics of the metabolic syndrome (MetS; 0.82). In a multivariate regression model, BMI Z-score [odds ratio (OR) 1.5] and HOMA-IR (OR 3.3) predicted 22% of the variance for the MetS, with 36% of the explained variance attributed to HOMA-IR. In a large clinical sample of overweight and obese Chilean youths, HOMA-IR ≥ 75th percentile was significantly associated with the cluster of factors referred to as the MetS. We emphasize the importance of establishing percentiles for HOMA-IR based on a normative sample and taking Tanner stage into account. Although BMI is easy to assess and interpret with minimal costs in a clinical setting, adding HOMA-IR explains more of the variance in the MetS than BMI Z-score alone.
Keywords