Recurrence risk after withdrawal of antiepileptic drugs in children with epilepsy: a prospective study.

Aim: To study recurrence risk after withdrawal of antiepileptic drugs in children with epilepsy.

Methods: All children younger than 14 with two or more unprovoked seizures 24h apart who were seen at our Hospital between 1994 and 2004 were included consecutively and prospectively followed. Patients previously examined in other centres were excluded. All patients who entered a remission were proposed to stop medication and were followed.

Results: Three hundred and fifty three children with two or more unprovoked seizures were attended. A total of 238 entered a remission period and were proposed to stop medication, 216 accept. Mean seizure-free time before medication withdrawal was 2.2 years. Kaplan-Meier estimate of recurrence risk was 23% at 2 years (95% CI: 17-29) and 28% at 5 years (95% CI: 22-34). A remote symptomatic etiology, various seizure types and a history of prior febrile seizures or prior neonatal seizures were associated with a significant increase in recurrence risk in univariable and multivariable analyses using Cox proportional hazards model. Recurrence risk at 2 years was 17% (95% CI: 11-23) for idiopathic/cryptogenic epilepsies and 41% (85% CI: 28-54) for remote symptomatic epilepsies. Recurrence risks at 2 years by epileptic syndrome were West syndrome (0%), benign rolandic epilepsy (10%), epilepsy without unequivocal partial or generalized seizures (11%), benign infantile seizures (13%), absence epilepsy (16%), cryptogenic partial epilepsies (20%), symptomatic partial epilepsies (45%), symptomatic generalized epilepsies (54%).

Conclusions: Recurrence risk after withdrawal of antiepileptic treatment in children is low. Etiology and syndromic diagnosis are the main predictive factors.