Predicting epilepsy surgery outcome

PURPOSE OF REVIEW: This review summarizes recent evidence on the seizure, safety, cognitive and psychosocial outcomes of epilepsy surgery and their predictors. RECENT FINDINGS: Risks of serious surgical complications have dramatically decreased over years to drop below 1% for temporal lobe resections. Although chances of postoperative seizure freedom largely vary between recent series, some data suggest that long-term seizure control might be achieved in over 80% of patients with mesial temporal lobe epilepsy or neocortical epilepsy associated with type 2 focal cortical dysplasia, and in up to two-thirds of patients with extratemporal lobe epilepsy. In the same conditions, some recent series challenge the classic view that a normal MRI is associated with worse outcome, an important finding given the greater proportion of MRI-negative patients now considered for epilepsy surgery. SUMMARY: These provocative findings appear to partly reflect the advances in the optimal use or postprocessing of neuroimaging data, as well as the identification of temporal plus epilepsy. This latter condition, which is characterized by normal MRI in half of patients and very poor outcome following anterior temporal lobectomy, was found to represent the main predictor of temporal lobe epilepsy surgery failures. Postoperative cognitive and quality-of-life outcomes, which partly depend on seizure control, are also influenced by antiepileptic drugs and psychiatric comorbidities.