Clinical characteristics of headache in 1,264 children with newly diagnosed idiopathic epilepsy were studied from 2006 to 2009 in five Italian Child Neurology University Centers. Seizures were generalized in 61 (tonic-clonic in 45) and partial in 69 patients (benign rolandic in 15). Of 130 children who showed an associated peri-ictal and/or interictal headache, 62% had post-ictal headache, 30% pre-ictal, and 57.6% inter-ictal headache. Rare cases of ictal epileptic headache (“migralepsy”) were excluded. Pre-ictal, post-ictal, and inter-ictal headaches met criteria for migraine in 93%, 81.4% and 87%, respectively. Migraine without aura was most commonly associated with partial epilepsy, and occurred in 82% patients with peri-ictal headache and in 76.5% patients with post-ictal headache. Tension-type headache occurred in 19% patients. Focal epileptiform discharges in 39 children with partial epilepsy were frontal in 31%, temporal in 13%, and occipital in 41%. Peri-ictal headaches were post-ictal in 29 (74%) and pre-ictal in 10 (26%) patients with focal EEGs. Type of headache was not significantly correlated with localization of EEG abnormalities. 
COMMENT. This study confirms the prevalence of post-ictal headache associated with idiopathic epilepsy and an association of partial epilepsy with peri-ictal headache, most commonly migrainous. EEG is not routinely indicated in the evaluation of children with headache but may be appropriate in patients with atypical migrainous aura or episodic loss of consciousness. Several studies have addressed the utility of the EEG in children with headache, some concluding that a beneficial effect of anticonvulsant drugs in treatment of chronic migraine is not correlated with EEG epileptiform abnormalities. In 30 children with migraine studied in the 1970s and treated with the anticonvulsant, phenytoin, 77% had a decreased recurrence of headaches. The response rate was 61% in 13 patients with abnormal EEGs and 88% in 17 with normal EEGs, and the difference was not significant. Response to AED was independent of the EEG abnormality . In an analysis of records of 257 children with chronic headache, the EEG showed epileptiform activity in 12% and slowing in 8%. 
The Cochrane Central Register of Controlled Trials of anticonvulsants in the prophylaxis of migraine in adults recorded 14 reported trials comparing AED with placebo in a total of 2024 patients: 4 trials with divalproex sodium, 2 sodium valproate, 3 topiramate, 2 gabapentin, and one trial each of carbamazepine, clonazepam, and lamotrigine. Sodium valproate/divalproex sodium and topiramate are effective in reducing migraine frequency and are reasonably well tolerated. Neither clonazepam nor lamotrigine was superior to placebo.