OBJECTIVE To describe the clinical effects of carbamazepine ingestion in a pediatric population. METHODS Case series of prospectively evaluated patients and a historical retrospective group. METHODS All patients younger than 18 years who presented to an urban emergency department with history of carbamazepine ingestion and positive laboratory confirmation. METHODS Patient demographics, findings on physical examination, serum carbamazepine levels, analysis of 12-lead ECGs, and time and doses of activated charcoal were recorded. RESULTS Seventy-seven patients were enrolled, of whom 17 were evaluated prospectively. Serum carbamazepine levels were greater than 12 micrograms/mL (50 mumol/L) in 61 patients. In those 61 patients, mean peak serum level was significantly higher in patients with dystonic reactions (P = .009), coma (P = .002), and apnea (P = .008) than in patients without these symptoms. There was no significant difference in mean peak serum levels between patients with and without seizures. Serum carbamazepine half-life was significantly shorter (P = .022) in patients who received multiple doses of activated charcoal (8.2 +/- 1.6 hours) than in those who received a single dose (12.1 +/- hours). CONCLUSIONS Pediatric patients with suspected carbamazepine ingestion are at higher risk for dystonic reactions, coma, and apnea if the peak serum carbamazepine level exceeds 28 micrograms/mL (117 mumol/L). The development of seizures is not related to peak serum level. Multiple doses of activated charcoal can significantly shorten serum carbamazepine half-life.