OBJECTIVE:To establish the prevalence of insulin resistance and impaired glucose tolerance (IGT) and their determinants in a cohort of obese children and adolescents. METHODS: A retrospective design was used. The study group included 234 patients with a body mass index (BMI) greater than the 95th percentile for age and gender and 22 patients with a BMI between the 85th and 95th percentile for age and gender referred for evaluation to a major tertiary-care center in Israel. Ages ranged from 5 to 22 y. Estimates of insulin resistance (homeostatic model assessment (HOMA-IR)); insulin sensitivity (ratio of fasting glucose (GF) to fasting insulin (IF) (GF/IF), the quantitative insulin sensitivity check index (QUICKI)), and pancreatic b-cell function (HOMA-derived b-cell function (HOMA %B)) were derived from fasting measurements. An oral glucose tolerance test (OGTT) was performed in 192 patients to determine the presence of IGT. RESULTS: Insulin resistance was detected in 81.2% of the patients, IGT in 13.5%, and silent diabetes in one adolescent girl. Only two patients with IGT also had impaired fasting glucose (IFG). The prevalence of IGT was higher in adolescents than prepubertal children (14.7 vs 8.6%). GF/IF and QUICKI decreased significantly during puberty (Po0.005), whereas HOMA-IR and HOMA %B did not. Insulin resistance and insulin sensitivity indexes were not associated with ethnicity, presence of acanthosis nigricans or family history of type 2 diabetes. Patients with obesity complications had lower insulin sensitivity indexes than those without (P ¼ 0.05). Compared with subjects with normal glucose tolerance (NGT), patients with IGT had significantly higher fasting blood glucose (85.976.5 vs 89.2710.6 mg/dl, Po0.05), higher 2-h post-OGGT insulin levels (101.2774.0 vs 207.67129.7 mU/ml, Po0.001), a lower QUICKI (0.32370.031 vs 0.30970.022, Po0.05), and higher fasting triglyceride levels (117.4753.1 vs 156.9768.9, P ¼ 0.002). However, several of the fasting indexes except QUICKI failed to predict IGT. There was no difference between the group with IGT and the group with NGT in fasting insulin, HOMA-IR, HOMA %B or the male-to-female ratio, age, BMI-SDS, presence of acanthosis nigricans, ethnicity, and family history of type 2 diabetes. CONCLUSIONS: Insulin resistance is highly prevalent in obese children and adolescents. The onset of IGT is associated with the development of severe hyperinsulinemia as there are no predictive cutpoint values of insulin resistance or insulin sensitivity indexes for IGT, and neither fasting blood glucose nor insulin levels nor HOMA-IR or HOMA %B are effective screening tools; an OGTT is required in all subjects at high risk. Longitudinal studies are needed to identify the metabolic precursors and the natural history of the development of type 2 diabetes in these patients.