Descriptive study of symptomatic epilepsy by age of onset in patients with a 3-year follow-up at the Neuropaediatric Department of a reference centre

Objective: We conducted a descriptive study of symptomatic epilepsy by age at onset in a cohort of patients who were followed up at a neuropaediatric department of a reference hospital over a 3-year period PATIENTS AND METHODS: We included all children with epilepsy who were followed up from January 1, 2008 to December 31, 2010 RESULTS: Of the 4595 children seen during the study period, 605 (13.17%) were diagnosed with epilepsy; 277 (45.79%) of these had symptomatic epilepsy. Symptomatic epilepsy accounted for 67.72% and 61.39% of all epilepsies starting before one year of age, or between the ages of one and 3, respectively. The aetiologies of symptomatic epilepsy in our sample were: prenatal encephalopathies (24.46% of all epileptic patients), perinatal encephalopathies (9.26%), post-natal encephalopathies (3.14%), metabolic and degenerative encephalopathies (1.98%), mesial temporal sclerosis (1.32%), neurocutaneous syndromes (2.64%), vascular malformations (0.17%), cavernomas (0.17%), and intracranial tumours (2.48%). In some aetiologies, seizures begin before the age of one; these include Down syndrome, genetic lissencephaly, congenital cytomegalovirus infection, hypoxic-ischaemic encephalopathy, metabolic encephalopathies, and tuberous sclerosis.

Conclusions: The lack of a universally accepted classification of epileptic syndromes makes it difficult to compare series from different studies. We suggest that all epilepsies are symptomatic because they have a cause, whether genetic or acquired. The age of onset may point to specific aetiologies. Classifying epilepsy by aetiology might be a useful approach. We could establish 2 groups: a large group including epileptic syndromes with known aetiologies or associated with genetic syndromes which are very likely to cause epilepsy, and another group including epileptic syndromes with no known cause. Thanks to the advances in neuroimaging and genetics, the latter group is expected to become increasingly smaller.