Memory tests are not good predictors of surgical outcome in patients with mesial temporal Lobe epilepsy associated with hippocampal sclerosis

Abstract

Introduction: One of the objectives of pre-surgical evaluation in mesial temporal epilepsy associated to hippoocampal sclerosis is the identification of patients with bad surgical prognosis for seizure control. At least theoretically, neuropsychological tests could be used in this venue. Objective: To evaluate whether verbal and visual memory tests can be used as isolate predictors of the post-surgical seizure outcome in patients with mesial temporal lobe epilepsy associated with hippocampal sclerosis refractory to pharmacological treatment. Methods: In a retrospective cohort study using the control of epileptic seizures as end-point, we evaluated 187 patients and calculated the correlation of clinical variables, cognitive evaluation, neuroimaging data, demographic data and electrophysiological findings with the result of seizure control after lobectomy in these patients. Results: An unfavorable prognosis during the postoperative period was observed only in association with low visual reproduction scores (visual memory). However, after Bonferroni correction, which was necessary to reduce the chance of type I error, this result was found to be spurious. Conclusion: We conclude that neuropsychological tests of verbal and visual memory such as those used in the routine presurgical evaluation of our patients with temporal lobe epilepsy are not good isolated predictors of surgical outcome.