[Epilepsies and time to diagnosis. Descriptive results of the CAROLE survey].

CAROLE is a prospective survey of children and adults who experienced epileptic unprovoked seizure(s) diagnosed for the first time between May 1 1995 and June 30 1996 by 243 French neurologists and neuropediatricians. Case records forms at entry allowed to compare patients who had a single seizure or several seizures prior to diagnosis. In patients with recurrent seizures, the time elapsed between the onset of attacks and the diagnosis (diagnostic delay) was looked for. Half of the 1942 included individuals already experienced more than one seizure when diagnosed. Due to natural history of epilepsies, 13 p.100 of the patients did not come to medical attention after a single seizure. Time to diagnosis ranged from 0 day to 52 years. While the overall median delay was 6 days, it ranged from 0 day to 7-8 months according to the type of seizure and the epilepsy syndrome. Two thirds of generalized convulsive seizures were immediately diagnosed versus one third of partial seizures. One half of infantile spasms were identified within 2 weeks versus 6 weeks in complex partial seizures, 4 months in absence seizures, 6 months in simple partial seizures, and 7 months in myoclonic seizures. Three quarters of idiopathic partial epilepsies were diagnosed within 4 weeks versus 3 months in symptomatic generalized epilepsies, 8 months in symptomatic partial epilepsies, 15 months in idiopathic generalized epilepsies, 30 months in cryptogenic partial epilepsies, and 33 months in undetermined epilepsies. So, the time elapsed between a first epileptic event and its diagnosis is epilepsy-dependent: seizure type and cause. Other reasons of diagnostic delay do exist: doctor and patient. They will be addressed in another study.