This population-based study will evaluate the significance of the EEG as a predictor for recurrence of seizures in those children who have had a simple febrile convulsion. Outcomes reported are febrile seizure recurrence and afebrile seizure occurrence. The evolution of the EEG pattern will be described, and patterns will be correlated with the clinical outcome. The clinical study was carried out in Skopje, Macedonia (Yugoslavia), at the Pediatric Clinic of the University of Skopje. The study began in FY 1982. Patient accrual was completed in December, 1984, by which time approximately 400 patients with a febrile seizure, no prior complex or multiple seizures and with a normal or nonspecific abnormal EEG, were registered into the study and began the study protocol and follow-up. An additional 300 patients with a specific abnormal EEG were entered for baseline information and follow-up. Additional efforts by the clinical center were needed to collect data from those patients lacking a return visit and those who did not have long term follow-up. Final follow-up visits were completed in FY 1991. Initially 22% of the 676 children had an EEG classified as paroxysmally abnormal, which was associated by logistic regression analysis with older age, number of previous febrile seizures, preexisting motor abnormality, and focal index seizures. Statistical analysis of baseline EEG and its association with characteristics of the child and family, and the clinical characteristics of the seizure has been published. Analysis is currently being conducted to examine: the effectiveness of the initial EEG in predicting recurrent febrile seizures; the evolution of EEGs in children with febrile seizures; and the value of changes in EEGs in predicting febrile seizure recurrence. Among the more than 76% who were followed for an average of 29 months, one-fourth experienced at least one recurrent febrile seizure. The recurrence rate was 25%, 24%, and 23%, respectively, for those with normal, nonspecifically abnormal, and specifically abnormal initial EEGs. Initial EEG was also not predictive of multiple recurrences. The classification of EEG at presentation was not related to the likelihood of recurrence of febrile seizures. Additional publications from this project will be reported under ZO1 NS 02652-09 (Statistical Collaboration and Consultation).
