Objective: To examine links between obstructive sleep apnea (OSA), insulin resistance, and dyslipidemia. Study design: Obese (body mass index [BMI] >95th percentile for age and gender) children who snored (n = 62, 46 males, age 10.89 [5-16 years] underwent polysomnography and metabolic studies. Results: Respiratory disturbance index (RDI) was 9.23 (0-95), with 23 children (39%) recommended for treatment. Fasting insulin levels were 154.6 pmol/L ± 79.8 (52-486), and fasting glucose levels were high in 7 children (11%). Fasting insulin levels correlated with sleep variables, including log transformed RDI (log₁₀RDI) (P = .01), desaturation events (P = .05), arousal index (P = .01), and sleep-time with oxygen saturation in arterial blood <90% (P = .03) (adjusted r² = 0.21, F = 3.9, P = .005), but not with age, or BMI Z score. Log₁₀RDI correlated with fasting insulin (P = .001) and BMI Z score (P = .03) (adjusted r² = 0.12, F = 3.9, P = .005), but not age or other metabolic variables. The correlation between log₁₀RDI and fasting insulin persisted in models combining sleep and metabolic variables: log₁₀RDI, adjusted r² = 0.75, F = 35.2, P < .001, and for fasting insulin, adjusted r² = 0.42, F = 6.1, and P < .001. Conclusions: The severity of OSA (log₁₀RDI) correlated with fasting insulin levels, independent of BMI. Insulin levels may be further elevated as a consequence of OSA in obese children. (J Pediatr 2002;140:654-9)