Abstracts

If the prognosis of childhood epilepsy is known from the moment the patient is first seen, this may have important implications for the choice of the therapeutic strategy. The Dutch Study of Epilepsy in Childhood was designed to study the prognosis of childhood epilepsy in a prospective follow-up study of a large cohort; to compare terminal remission (TR) after five years with the course during the follow-up; to identify prognostically relevant variables; to develop prognostic models and study the reliability with which they indicate the individual prognosis; and to evaluate the concept of intractability in childhood epilepsy.
453 Children out of the cohort of the Dutch Study of Epilepsy in Childhood were followed for five years. TR after five years (TR5) was compared with the result after two years and with the longest remission during the follow-up. Univariate and multivariate analyses using variables defined at intake and after six months of follow-up were performed to determine their prognostic relevance. Based upon these, models were developed to determine the individual prognosis. Data about the use of anti-epileptic drugs and the outcome of various epileptic syndromes were also studied.
76% Had a TR5 of more than one year, 64% more than two years and 14% had not had any seizure during the entire follow-up. 34 Out of 108 children with TR5 less than one year (8% of the entire cohort) were considered really intractable according to predefined criteria. Significant variables in the multivariate analyses were non-idiopathic etiology, large number of seizures before intake, history of febrile convulsions and the combination of epileptiform EEG and age below four for intake variables only; and epileptiform EEG and age below four, postictal signs and course of the disease during the first six months of follow-up for intake and 6-month variables combined. The resulting models predicted the outcome correctly in 72 and 75% of cases. Comparison of the course during the follow-up or terminal remission after two years on the one hand and five-year outcome on the other demonstrates that improvement occurred more often than deterioration.
The prognosis of childhood epilepsy is generally good, and tends to improve after a longer follow-up. A poor outcome is associated mainly with symptomatic or cryptogenic etiology, the combination of young age at onset and epileptiform EEG abnormalities, and the early course of the disease. With the help of the variables used, a prognosis can be calculated in individual cases with a reasonable sensitivity and specificity. This study seems to make the way for trials of individualized treatment strategies for children with epilepsy with varying prognosis.
[Supported by: Dutch National Epilepsy Fund, Janivo Foundation]