Predicting long-term outcome of childhood epilepsy in Nova Scotia, Canada, and Turku, Finland. Validation of a simple scoring system

M. Sillanpaa, P. Camfield and C. Camfield
Department of Child Neurology, University Hospital, Turku, Finland.

BACKGROUND: Based on a follow-up study of 486 children, a scoring system for the long-term outcome for childhood epilepsy has been developed in Nova Scotia, Canada. The scoring system consists of four weighted predictor variables available at the time of diagnosis (ie, age at onset, intelligence, neonatal seizures, and number of seizures before treatment). OBJECTIVE: Validation of the scoring system using long-term follow-up data for childhood epilepsy from a regional cohort in Turku, Finland. DESIGN: Patients were selected from the Turku cohort with the same entry criteria as the Nova Scotia cohort. The scoring system was used to predict their epilepsy outcome. RESULTS: A total of 141 patients were selected from the Finland cohort. After 30 years of follow-up, 60% of the patients were in remission. The Nova Scotia scoring system correctly predicted outcome in 61% of Finnish cases (sensitivity, 43%; specificity, 88%; and positive predictive value, 84%). CONCLUSIONS: Long-term good outcome for childhood epilepsy is well predicted by the Nova Scotia scoring system.