Seizure outcome after AED failure in pediatric focal epilepsy: impact of underlying etiology

This study aimed to identify long-term seizure outcome in pediatric nonsyndromic focal epilepsy after failure of serial antiepileptic drugs (AEDs) due to lack of efficacy.

Children (1 month-17 years) with new-onset focal epilepsy not meeting the criteria for a defined electroclinical syndrome diagnosed between 1980 and 2009 while residing in Olmsted County, MN, were retrospectively identified. Medical records of those followed for ≥2 years were reviewed to assess etiology, the number of AEDs that failed due to lack of efficacy, and seizure outcome at final follow-up. Etiology was classified into structural/metabolic, genetic, or unknown. Favorable outcome was defined as seizure freedom ≥1 year, on or off AEDs, without prior epilepsy surgery. Poor outcome was defined as ongoing seizures in the preceding year or having undergone prior epilepsy surgery.

Nonsyndromic focal epilepsy accounted for 275/468 (59%) of all patients with newly diagnosed epilepsy--of these, 256 (93%) were followed for a minimum of two years and were included in the study. Median duration of follow-up was 10.0 years. At least one AED had failed due to lack of efficacy in 100 (39.1%) children. Favorable outcomes occurred in 149/156 (95.5%) children with no AED failure, 16/30 (53.3%) with one AED failure, 8/25 (32%) with two AED failures, and only 2/45 (4.4%) with three AED failures. After two AED failures, the seizures of nearly one-quarter of children who had epilepsy with an unknown cause responded favorably to the third AED compared with only 7.8% of the cohort that had epilepsy with a structural/metabolic cause. Children with a remote brain insult had a significantly higher likelihood of favorable outcome with serial AEDs than those with other structural abnormalities.

Etiology is an important determinant of pharmacoresistance in nonsyndromic focal epilepsy. Surgical evaluation should be considered after failure of 1-2 AEDs in those who have epilepsy with structural causes, excluding remote brain insults. Conversely, as surgical success is lower with normal MRI or more diffuse brain insults, it appears reasonable to hold off surgical evaluation until 2-3 AEDs have failed in such children.