RESUMO
Introduction Hyperandrogenism (HA), either clinical or biochemical, is associated with obesity in adolescent girls. Long chain polyunsaturated fatty acids ω3 (LCPUFA-ω3) play protective roles in some obesity-associated morbidities, but their contribution to preventing HA is unclear. Our aim was to examine the potential positive relationships between erythrocyte LCPUFA-ω3, with or without supplementation, and hyperandrogenemia. Methods Secondary analysis of a clinical trial that was conducted previously to analyze the effect of LCPUFA-ω3 on insulin resistance and body weight. Here, we present a cross-sectional analysis of 180 girls with obesity, and a longitudinal analysis of 117 girls who completed a 3-month supplementation period (57 LCPUFA-ω3 [DO3] and 60 placebo [DP)]). Dehydroepiandrosterone sulfate (DHEAS), total testosterone (TT) and steroid hormone binding globulin (SHBG) were measured with chemiluminescence; free testosterone (FT) was calculated. Erythrocyte fatty acids were determined by gas chromatography. Non-parametric statistics was used for analysis. Results In cross-sectional analysis, age (odds ratio [OR] = 1.35; 95% confidence interval [CI] = 1.03, 1.78; p = 0.027), insulin (OR = 1.05; 95% CI: 1.00, 1.10; p = 0.018), and erythrocytes eicosapentaenoic acid (EPA) (OR = 0.04; 95% CI: 0.01, 0.65; p = 0.012) were predictors of hyperandrogenemia (FT >0.63 ng/mL). In longitudinal analysis, EPA, adiponectin and SHBG increased, while FT decreased, in the DO3 group (p < 0.05). The risk of hyperandrogenemia at the end of follow-up was predicted by basal hyperandrogenemia (OR = 18.16, 95% CI: 5.37, 61.4; p < 0.001) and by increases in EPA (OR = 0.40; 95% CI: 0.01, 0.65; p = 0.06 marginal significance). Conclusions Our results suggest a preventive role of EPA on the risk for hyperandrogenemia in girls with obesity, but further studies are needed to demonstrate a benefit.