The aim of the study was to compare the outcome with respect to age of implant, aetiology and duration of epilepsy.One hundred thirty-five drug-resistant epileptic patients, excluded from ablative surgery, were submitted to vagal nerve stimulation (1995-2007). Aetiology was cryptogenic in 57 and symptomatic in 78 patients. Ages of implant were 0.5-6 years (18 patients), 7-12 years (32 patients), 13-18 years (31 patients) and more than 18 years (54 patients). Epilepsy types were Lennox-Gastaut (18 patients), severe multifocal epilepsy (33 patients) and partial (84 patients). Duration of epilepsy is 3 months to 57 years. Clinical outcome was determined by comparing the seizure frequency after stimulation at 3-6-12-18-24-36 months with the previous 3 months. 'Responders' were the patients experiencing a seizure frequency reduction of 50\% or more during follow-up. In statistical analysis, Wilcoxon and McNemar tests, general linear model for repeated measures, logistic regression and survival analysis were used.The seizure frequency reduction was significant in the group as a whole between baseline and the first follow-up (Wilcoxon test). The percentage of responder increases with time (McNemar test p = 0.04). Univariate analysis showed a significant effect of the age of implant on seizure frequency reduction: Adult patient had worst clinical outcome than children (p < 0.001) and adolescents (p = 0.08). Patients with severe multifocal epilepsy had better percentage seizure reduction compared with Lennox-Gastaut and partial (p = 0.03). Lesser duration of epilepsy had positive influence on outcome. Multivariate analysis confirmed age of implant to be the strongest factor influencing prognosis. Furthermore, positive is the association between lesional aetiology and young age.The best responder could be a young lesional epileptic patient; after 3 years of follow-up, the percentage of responders is still in progress.
