Efficacy and tolerability of topiramate in the treatment of pediatric migraine is studied by retrospective analysis of records of 37 children treated at St Christopher's Hospital for Children, Philadelphia, PA. The mean age was 14 years; the range, 7.3-20.5 years. The majority (30 [81%]) had migraine without aura, 4 (11%) had migraine with aura, and the remaining 3 had abdominal, ophthalmoplegic, and catamenial migraine in one each. Mean follow-up was 12 +/- 5 months. The mean dose of topiramate was 1.7 +/- 1 mg/kg/day (range, 0.5-5.5 mg/kg/day), or 50-200 mg/day. Headache frequency per month was 15 +/- 7 before treatment and 3 +/- 3.4 after treatment. Response was excellent or good, with >50% migraine reduction, in 28 (76%) patients. Adverse effects occurred in 10 (27%) patients; 5 had cognitive deficits, 3 drowsiness, 1 paresthesias, and 1 anhidrosis. No patient had significant weight loss. Side effects were directly related to dosage, and occurred especially in patients taking doses >2 mg/kg/day (mean toxic dose 2.8 +/- 1.5 mg/kg/day). The mean dose not associated with adverse events was 1.27 +/- 0.7 mg/kg/day. Seven (19%) patients discontinued treatment because of side effects, 5 (14%) with cognitive issues. The authors conclude that topiramate is an effective, safe prophylactic therapy for pediatric migraine. The acceptable risk/benefit maintenance dose is <2 mg/kg/day. 
COMMENT. In this retrospective, uncontrolled study, topiramate at one-year follow-up appeared to be an effective prophylactic therapy for pediatric migraine. Cognitive deficit was a significant adverse event, however, leading to withdrawal of therapy in 14% patients. Since headache disorders in children and adolescents tend to resolve spontaneously in a large proportion of patients, as shown in the previous study (Wang S-J et al, 2009), double-blind, placebo-controlled studies of migraine prophylaxis are essential.
Other anticonvulsants, including phenytoin and valproate, are effective in the prophylaxis of migraine, but the side-effects tend to outweigh the benefits. In an early study of the EEG and response to phenytoin in 30 children with migraine, 77% had headaches controlled . Response to phenytoin was not correlated with an abnormal EEG. In 13 patients with abnormal and 17 with normal EEGs, the beneficial response rates were 61% and 88%, respectively. Epileptiform EEGs were found in 18% of a total 100 consecutive children with recurrent headache, and with the same frequency in those with migraine. Kramer U, Harel S and associates found an 11% incidence of epileptiform EEGs in children with migraine or tension headaches; the incidence was 26% and significantly higher in children with chronic “very brief” headaches .