Leptin is secreted by the white adipose tissue and modulates energy homeostasis. Nutritional, neural, neuroendocrine, paracrine, and autocrine factors, including the sympathetic nervous system and the adrenal medulla, have been implicated in the regulation of leptin secretion. Classic congenital adrenal hyperplasia (CAH) is characterized by a defect in cortisol and aldosterone secretion, impaired development and function of the adrenal medulla, and adrenal hyperandrogenism. To examine leptin secretion in patients with classic CAH in relation to their adrenomedullary function and insulin and androgen secretion, we studied 18 children with classic CAH (12 boys and 6 girls; age range 2-12 yr) and 28 normal children (16 boys and 12 girls; age range 5-12 yr) matched for body mass index (BMI). Serum leptin concentrations were significantly higher in patients with CAH than in control subjects (8.1 +/- 2.0 vs. 2.5 +/- 0.6 ng/ml, P = 0.01), and this difference persisted when leptin values were corrected for BMI. When compared with their normal counterparts, children with CAH had significantly lower plasma epinephrine (7.1 +/- 1.3 vs. 50.0 +/- 4.2, P < 0.001) and free metanephrine concentrations (18.4 +/- 2.4 vs. 46.5 +/- 4.0, P < 0.001) and higher fasting serum insulin (10.6 +/- 1.4 vs. 3.2 +/- 0.2 microU/ml, P < 0.001) and testosterone (23.7 +/- 5.3 vs. 4.6 +/- 0.5 ng/dl, P = 0.003) concentrations. Insulin resistance determined by the homeostasis model assessment method was significantly greater in children with classic CAH than in normal children (2.2 +/- 0.3 vs. 0.7 +/- 0.04, P < 0.001). Leptin concentrations were significantly and negatively correlated with epinephrine (r = -0.50, P = 0.001) and free metanephrine (r = -0.48, P = 0.002) concentrations. Stepwise multiple linear regression analysis indicated that serum leptin concentrations were best predicted by BMI in both patients and controls. Gender predicted serum leptin concentrations in controls but not in patients with classic CAH. No association was found between the dose of hydrocortisone and serum leptin (r = -0.17, P = 0.5) or insulin (r = 0.24, P = 0.3) concentrations in children with CAH. Our findings indicate that children with classic CAH have elevated fasting serum leptin and insulin concentrations, and insulin resistance. These most likely reflect differences in long-term adrenomedullary hypofunction and glucocorticoid therapy. Elevated leptin and insulin concentrations in patients with CAH may further enhance adrenal and ovarian androgen production, decrease the therapeutic efficacy of glucocorticoids, and contribute to later development of polycystic ovary syndrome and/or the metabolic syndrome and their complications.