Outcome of intracranial electroencephalography monitoring and surgery in magnetic resonance imaging-negative temporal lobe epilepsy

Highlights

• iEEG-guided anterior temporal lobectomy has excellent surgical outcomes.

• 25% of MRI-negative TLE patients have temporal neocortical seizure focus on iEEG.

• Verbal learning and memory outcomes remain stable after temporal neocorticectomy.

• iEEG-guided temporal neocorticectomy has a low surgical success rate.

Summary

We evaluated the outcomes of intracranial electroencephalography (iEEG) recording and subsequent resective surgery in patients with magnetic resonance imaging (MRI)-negative temporal lobe epilepsy (TLE). Thirty-two patients were identified from the Mayo Clinic Epilepsy Surgery Database (Arizona, Florida, and Minnesota). Eight (25.0%) had chronic iEEG monitoring that recorded neocortical temporal seizure onsets; 12 (37.5%) had mesial temporal seizure onsets; 5 (15.6%) had independent neocortical and mesial temporal seizure onsets; and 7 (21.9%) had simultaneous neocortical and mesial seizure onsets. Neocortical temporal lobe seizure semiology was the only factor significantly associated with neocortical temporal seizure onsets on iEEG. Only 33.3% of patients who underwent lateral temporal neocorticectomy had an Engel class 1 outcome, whereas 76.5% of patients with iEEG-guided anterior temporal lobectomy that included the amygdala and the hippocampus had an Engel class 1 outcome. Limitations in cohort size precluded statistical analysis of neuropsychological test data.

Introduction

The role of surgical resection has been well established in medically intractable temporal lobe epilepsy (TLE) (Wiebe et al., 2001, Yoon et al., 2003, Cohen-Gadol et al., 2006, Schmidt and Stavem, 2009). Seizure-free outcome rates after resective surgery average around 70–80% in patients with hippocampal atrophy and concordant interictal and ictal electroencephalographic (EEG) discharges (Berkovic et al., 1995, Cascino, 2004). Unfortunately, the success rate is much lower in patients with non-lesional TLE (Berkovic et al., 1995, Holmes et al., 2000, Chapman et al., 2005, Tatum et al., 2008, Bell et al., 2009). Given the fact that mesial temporal structures are important for memory, physicians may be more reluctant to recommend anterior temporal lobectomy, which includes the mesial temporal structures, when no structural abnormality is identified on magnetic resonance imaging (MRI). In these complicated patients, intracranial EEG (iEEG) monitoring is often required to guide temporal lobe resection.

Because increased complication rates are related to longer monitoring and a greater number of electrodes being implanted, the extent of electrode implantation for iEEG should be determined judiciously and should be limited to reduce the risk of complications (Hamer et al., 2002). The prognostic value of interictal spiking in intraoperative electrocorticography (ECoG) remains controversial. Luther et al. (2011) reported that intraoperative ECoG can be useful in a subset of patients with TLE and non-lesional MRI. However, other authors could not find correlation between seizure-free outcomes and complete resection of irritative zone on ECoG (Schwartz et al., 1997, San-juan et al., 2011, Wray et al., 2012). Therefore, the role of chronic extraoperative iEEG monitoring in patients with MRI-negative TLE remains important and deserves further study. The purpose of this study was to determine the yield of extraoperative iEEG monitoring in patients with MRI-negative TLE. In addition, we sought to determine the outcome of subsequent resective surgery in these patients.