To propose a new criterion for the diagnosis of metabolic syndrome (MS) in adolescents and to check its consistency with those proposed by Jolliffe and Janssen and by the International Diabetes Federation (IDF).This is a cross-sectional study of 80 obese adolescents aged 14 to 19 years. Anthropometric (weight, height, and waist circumference) and laboratory (fasting triglycerides, HDLc, glucose, and insulin) parameters, as well as blood pressure were evaluated. The HOMA-IR index was used to characterize insulin resistance, and the presence of steatosis was assessed by hepatic ultrasound. Agreement analyses across the three criteria were made using the kappa coefficient.The prevalence of MS was 13.5, 15, and 25% for IDF and Jolliffe and Janssen's criteria and the proposed method, respectively. A nearly perfect agreement between Jolliffe and Janssen's and IDF (kappa = 0.94) criteria and a moderate agreement between the new criteria and the previous two (kappa = 0.46 and 0.41, respectively) were observed.The highest prevalence of MS was observed with the criterion proposed in this study, which included steatosis and insulin resistance as parameters, thus being able to diagnose a larger number of adolescents at metabolic risk.