Seizure outcomes of temporal lobe epilepsy surgery in patients with normal MRI and without specific histopathology

Abstract

Background: Seizure outcome following surgery in pharmacoresistant temporal lobe epilepsy patients with normal magnetic resonance imaging and normal or non-specific histopathology is not sufficiently presented in the literature.

Methods: In a retrospective design, we reviewed data of 263 patients who had undergone temporal lobe epilepsy surgery and identified 26 (9.9%) who met the inclusion criteria. Seizure outcomes were determined at 2-year follow-up. Potential predictors of Engel class I (satisfactory outcome) were identified by logistic regression analyses.

Results: Engel class I outcome was achieved in 61.5% of patients, 50% being completely seizure free (Engel class IA outcome). The strongest predictors of satisfactory outcome were typical ictal seizure semiology (p = 0.048) and localised ictal discharges on scalp EEG (p = 0.036).

Conclusion: Surgery might be an effective treatment choice for the majority of these patients, although outcomes are less favourable than in patients with magnetic resonance imaging-defined lesional temporal lobe epilepsy. Typical ictal seizure semiology and localised ictal discharges on scalp EEG were predictors of Engel class I outcome.

Keywords: Epilepsy surgery; Normal MRI; Normal or non-specific histopathology; Postsurgical seizure outcomes; Prognostic factors; Temporal lobe epilepsy.

Fig. 1

Location of the electrode contacts on superficial cortical anatomy of the temporal lobes, obtained by reconstruction of presurgical brain magnetic resonance imaging co-registered with computerised tomography scan after bilateral implantation of subdural strip electrodes, and localisation of the seizure onset zone on the laterobasal neocortex of the left temporal lobe, determined by the minimum norm solution of early ictal electroencephalographic activity, using methods as implemented in Curry software