- Email: [email protected]
Temporal lobe surgery in medically refractory epilepsy: A comparison between populations based on MRI findings
Seizure 23 (2014) 20–24
Contents lists available at ScienceDirect
Seizure journal homepage: www.elsevier.com/locate/yseiz
Temporal lobe surgery in medically refractory epilepsy: A comparison between populations based on MRI findings Tsz Lau a, Timothy Miller a, Travis Klein a, Selim R. Benbadis b, Fernando L. Vale a,* a b
Department of Neurosurgery and Brain Repair, Morsani College of Medicine, University of South Florida, Tampa, United States Department of Neurology, Morsani College of Medicine, University of South Florida, Tampa, United States
A R T I C L E I N F O
A B S T R A C T
Article history: Received 14 May 2013 Received in revised form 12 August 2013 Accepted 6 September 2013
Introduction: High resolution MRI findings suggestive of mesial temporal sclerosis (MRI-MTS) correlate with good outcome after surgery. However, a large group of patients present with normal brain MRI (NMRI) and temporal lobe epilepsy (TLE). We aim to compare pre-operative ictal EEG patterns in predicting surgical outcomes in the population with MRI-MTS vs. N-MRI after selective anterior-mesial temporal lobe (AMTL) resection. Methods: 241 patients with unilateral anterior ictal EEG findings underwent selective AMTL resection. 143 MRI-MTS and 98 N-MRI patients were identified. Outcome was based on the modified Engel classification, ictal EEG pattern at seizure onset, demographics and MRI findings. Results: Seizure-free outcome was seen in the MRI-MTS in 79% of patients, compared to 59.1% (p < .005) of the N-MRI group. No significant difference was identified in ictal EEG patterns at presentation between groups. Class I outcome was achieved in 78.9% of patients that had theta rhythm and MRI-MTS compared to 57.9% of patients that had theta rhythm and N-MRI (p < 0.05). Discussion and conclusion: Surgical treatment for mesial TLE is effective. Positive MRI suggestive of mesial temporal sclerosis (MTS) predicts better seizure control after surgery. Theta rhythm is the most common ictal pattern and seems to carry the best prognosis for TLE. However, a well-selected group of patients with N-MRI will benefit from surgical intervention, and similar outcome to MRI-MTS patients can be achieved if delta ictal EEG pattern is presented at initial onset. Early referral to an epilepsy center cannot be emphasized enough, even in situations when high-resolution brain MRI is normal. ß 2013 British Epilepsy Association. Published by Elsevier Ltd. All rights reserved.
Keywords: Epilepsy surgery Temporal lobe Nonlesional Negative MRI Ictal EEG Outcomes
1. Introduction Mesial temporal sclerosis (MTS) is typically seen in medically refractory mesial temporal lobe epilepsy. About 75% of patients with mesial temporal lobe epilepsy (TLE) are resistant to medical management.1 Surgical resection usually provides excellent seizure control in patients with TLE,2 especially in the setting of positive magnetic resonance imaging (MRI) evidence of mesial temporal sclerosis (MRI-MTS).3–10 Surgery for patients with temporal lobe epilepsy and normal MRI findings (N-MRI) is less effective.11–17 Studies have shown 37– 76% seizure control in this group of patients, but they were either
Abbreviations: MTS, mesial temporal sclerosis; TLE, temporal lobe epilepsy; MRIMTS, mesial temporal sclerosis positive MRI; N-MRI, normal MRI; AMTL, anterior mesial temporal lobe; MTLE, mesial temporal lobe epilepsy. * Corresponding author at: Department of Neurosurgery and Brain Repair, University of South Florida, 2 Tampa General Circle, Tampa, FL 33606, United States. Tel.: +1 813 259 0605; fax: +1 813 259 0944. E-mail address: [email protected] (F.L. Vale).
retrospective in nature, included heterogeneous groups of patients with small a percentage of patients with normal MRI, or were based on small cohorts of patients. Still to be well elucidated is the contribution of ictal EEG patterns to the outcome of temporal lobe surgery in treating TLE. Some small studies have demonstrated a difference in anterior temporal ictal rhythms (slow (delta) vs. faster (theta) rhythms) in patients with MRI-MTS vs. N-MRI, but further research into this issue is likely warranted.18 The goal of this study is to compare the surgical outcomes in patients with mesial TLE with either N-MRI or ipsilateral MRI-MTS, and to identify factors that would predict better outcome based on pre-operative clinical evaluation. 2. Methods 2.1. Patient population Data were obtained from our institution’s epilepsy surgery database. This is a prospective database established in 1998 after
1059-1311/$ – see front matter ß 2013 British Epilepsy Association. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.seizure.2013.09.004
T. Lau et al. / Seizure 23 (2014) 20–24
receiving the institutional Internal Review Board approval. Informed consent was obtained from all patients. Patients with mesial temporal semiology and documented anterior temporal ictal recordings that underwent anterior-mesial temporal lobe (AMTL) resection for temporal lobe epilepsy were identified. Patients were divided into two groups based on results of preoperative high resolution MRI. The first group included patients with MRI evidence of ipsilateral mesial temporal sclerosis, the second included patients with normal MRI. Patient evaluation for surgical treatment as well as MRI protocol has been previously described.19 All patients were initially evaluated by board certified, fellowship trained epileptologists. Patients with semiology that suggested mesial temporal lobe epilepsy (MTLE) proceeded to phase 1 evaluation: long-term video-EEG monitoring. Sphenoidal or nasopharyngeal electrodes were not used on any patients, but bilateral basilar temporal placement, such as T1, T2 electrodes, were done routinely. Ictal EEG pattern was recorded at seizure onset. Anterior temporal rhythms were recorded as follow: (a) Slow (delta) rhythm (2– 4 Hz), (b) Faster rhythms (theta rhythm (5–7 Hz) or alpha (8– 12 Hz)) and (c) delta–theta (progression from slow to faster rhythms within 5 s). All EEG recordings were reviewed in the weekly comprehensive epilepsy surgery conference. At least two board-certified epileptologists were present on each conference and consensus was achieved between the groups. Inter-ictal positron emission tomography (PET) was performed if possible. A FDG-PET scan was considered positive when it showed an area of hypometabolism in the ipsilateral mesial-basal temporal lobe. Subtraction ictal single photon emission CT (SPECT) coregistered to MRI (SISCOM) was not available at our institution and only a few of our patients had SPECT imaging for statistical analysis. Following initial evaluation with video-EEG monitoring, patients received imaging with high-resolution MR using a 1.5 or 3.0-T magnet with special focus on the temporal lobes. Coronal and axial T2-weighted and FLAIR images were obtained and reviewed by a board-certified neuro-radiologist. Abnormal signal on FLAIR and T2weighted images, decreased volume and loss of anatomical configuration of the hippocampal formation are considered the hallmarks of radiographically confirmed MTS. In addition, most patients also underwent Wada testing to evaluate for memory asymmetry. Memory recall 2.5/8 points difference was considered asymmetric. Some of these pre-operative evaluations were not performed due to patient financial restraints, patient refusal, or patient non-compliance due to young age or low intelligence quotient. Decisions regarding surgical therapy were made after a review of all the clinical data for each patient in conjunction with the epilepsy neurologists, neurosurgeon and neuroradiologist of the Comprehensive Epilepsy Center. Invasive subdural contact electrode placement (phase II evaluation) along the basal and lateral neocortex was considered when: (i) there was an inadequate ictal recording due to extensive muscle artifact or poorly localized ictal onset in the setting of typical mesial temporal semiology or (ii) there was discordant pre-operative evaluation. All patients were followed at 3, 6, and 12 months, then yearly after. All patients with <1 year followup were excluded from analysis. Clinical outcome was based on the modified Engel classification: Class I, seizure free or residual aura (CL-I); Class II, rare disabling seizures (<3 complex partial seizures per year) (CL-II); Class III, worthwhile seizure reduction; and Class IV, no worthwhile improvement.20 2.2. Surgical technique Details of the surgical technique have been previously well described.21 In brief, a temporal trephine craniotomy is performed. The inferior temporal gyrus is identified and resected. Cortical resection is performed from lateral to medial structures. Landmarks
21
along the way from lateral to medial are the occipitotemporal sulcus, fusiform gyrus, and collateral sulcus. Next, the floor of the temporal horn is opened, followed by the resection of parahippocampal gyrus, amygdala, and uncus. Lastly, the hippocampus is resected en bloc, which can then be used for pathological analysis. Subpial fashion dissection of the mesial structure ensures protection of the underlying brain stem and perimesencephalic vasculature. 2.3. Statistical analysis Microsoft Excel (Microsoft, Redmond, WA, USA) software was used to store prospective clinical data. IBM Statistical Package for the Social Sciences 19.0 for Windows was used to perform statistical analysis. Results of pre-operative work-up (Wada test, ictal EEG pattern at seizure onset, PET) were evaluated to identify predictors of surgical outcome based on the modified Engel Classification. Fisher Exact test and Chi-squared were used to analyze categorical variables. The non-parametric Mann–Whitney U-test was used to test the distribution of age among the two groups. Statistically significant is defined as p value 0.05. 3. Results A total of 299 patients with TLE and unilateral ictal EEG findings that underwent selective AMTL resection were identified from our database. After excluding patients with less than 1-year follow-up, or dual pathology, 241 patients remained for statistical analysis. One hundred and forty three patients had evidence of mesial temporal sclerosis on preoperative high resolution MRI and 98 patients had normal high resolution MRI. 17 patients in the N-MRI group underwent phase II invasive monitoring as opposed to 5 patients in the MRI-MTS group. Demographics and background characteristics are summarized in Table 1. Mean follow-up was 2.9 years for the NMRI group and 3.4 years for the MRI-MTS group. The average age for both groups was 38-year-old. Sex distribution and the results of Wada testing between the two groups were also well matched. The only difference in pre-operative evaluation was a higher percentage of patients in the MRI-MTS group with positive PET study compared to the N-MRI group (94% vs. 81%, respectively). PET scan was performed in 94 (96%) patients in the N-MRI, whereas 76 (81%) demonstrated ipsilateral hypometabolism. In the MRI-MTS group, 99 (71%) studies were performed, while 93 (94%) studies were deemed positive. On the other hand, Wada test was performed in 91 patients in the N-MRI whereas 45 (49%) demonstrated significant asymmetry. In the MRI-MTS group, 130 tests were performed; however, only 69 (53%) patients demonstrated significant difference. Overall outcome based on modified Engel classification was statistically significant between the two groups (p < 0.05). In the MRI-MTS group, 79% had class I outcome (n = 113) compared to 59.2% (n = 58) in N-MRI group at last follow-up (Fig. 1). In patients with positive PET, similar percentages were observed (p < 0.05) Table 1 Patient characteristics.
Avg. age (range) Avg. F/U (range) Sex M F PET scan Positive Negative Wada test Asymmetry No Asymmetry
Negative MRI (n = 98)
MRI-MTS (n = 143)
p-Value
38 (10–65) 2.9 years (1–11.5)
38 (7–68) 3.4 years (1–11.5)
0.857
40% 60% n = 94 81% 19% n = 91 49% 50%
41% 59% n = 99 94% 6% n = 130 53% 47%
0.905
0.006
0.595
T. Lau et al. / Seizure 23 (2014) 20–24
22
Overall Outcome 90.0 80.0 Percentage
70.0 60.0 MTS-MRI (n=143)
50.0
nMRI (n=98)
40.0 30.0 20.0 10.0 0.0 IIIIIIIV
Engel Class
Percentage
Fig. 1. Overall outcome – negative MRI vs. MRI-MTS, p < 0.05.
90.00% 80.00% 70.00% 60.00% 50.00% 40.00% 30.00% 20.00% 10.00% 0.00%
MTS-MRI (n=93) nMRI (n=76)
I
II,III,IV Outcome
Fig. 2. Overall outcome in PET/SPECT positive patients, p < 0.05.
Theta
Percentage
100.00% 80.00% 60.00%
MTS-MRI (n=95)
40.00%
nMRI (n=57)
20.00% 0.00% I
Outcome
II,III,IV
Fig. 3. Outcome in patients with theta rhythm, p < 0.05.
(Fig. 2). Ictal EEG pattern at seizure onset is summarized in Table 2. Nine patients had limited or missing pre-operative ictal EEG documentation to allow adequate classification during the retrospective review process. Of the remaining 232 patients, theta rhythm was the most common ictal pattern identified for TLE. Ictal theta onset was identified in one hundred and fifty four (66.3%) out of 232 patients. Slow (delta) pattern was seen in 35 out of 232 (15%), while delta/theta was seen in 13.3% (31 out of 232) of the total population. Twelve patients had a poorly localizable ictal EEG pattern. Faster theta rhythm was seen in 58.2% in the N-MRI group and 70.9% in the MRI-MTS group. Delta rhythm was seen in 21.4% and 10.4% respectively. 14.3% of N-MRI patients had Delta–theta rhythm compared to 12.7% of MRI-MTS patients. Similar number of patients in both groups had nonTable 2 Ictal EEG pattern at seizure onset.
Theta, n = 154 (66.4%) Delta, n = 35 (15.1%) Delta/theta, n = 31 (13.4%) Non-localized, n = 12 (5.2%)
Negative MRI, n = 98
MTS MRIa, n = 134
59 21 14 4
95 14 17 8
(60.2%) (21.4%) (14.3%) (4.1%)
(70.9%) (10.4%) (12.7%) (6.0%)
p = 0.11. a 9 patients with missing or limited pre-operative ictal recordings.
localized scalp rhythm. There was no statistically significant difference in the distribution of ictal EEG pattern between the groups (p = 0.11). Class I outcome was achieved in 75 out of 95 (78.9%) of patients that had theta rhythm and MRI-MTS compared to 33 out of 57 (57.9%) of patients that had theta rhythm and N-MRI (p < 0.05) (Fig. 3). For patients that had delta rhythm, class I outcome was achieved in 71% in both groups (Fig. 4). Class I outcome was achieved 64% of N-MRI patients (n = 14) and in 82% of MRI-MTS patients (n = 17) that had delta–theta rhythm (p > 0.05). Histopathology demonstrated hippocampal sclerosis (severe, long-standing) in all cases with pre-operative MRI-MTS. However, N-MRI population had evidence of neuronal loss and gliosis at the level of hippocampus compatible with mild to moderate MTS in 40 out of 98 patients (41%). No severe MTS was identified in the latter group of patients. Overall surgical complication rate was 2.9%. One patient had a superficial wound infection that required wound revision. One patient developed cerebral spinal fluid leak that required surgical repair. Four patients developed post-operative subdural hematoma (one subacute and three chronic) in which 2 cases required surgical evacuation in a delayed fashion without any long-term morbidity. One patient developed minor lacunar infarct postoperatively, remote from the operative site.
Percentage
T. Lau et al. / Seizure 23 (2014) 20–24
23
Delta
80.00% 70.00% 60.00% 50.00% 40.00% 30.00% 20.00% 10.00% 0.00%
MTS-MRI (n=14) nMRI (n=21)
I
II,III,IV Outcome
Fig. 4. Outcome in patients with delta rhythm, p = 1.
4. Discussion Previous reports have demonstrated that theta rhythm is the most common ictal anterior temporal pattern in the setting of TLE.22,23 Our results support the idea that theta rhythm at ictal onset seems to be most commonly associated with TLE and carries the best prognosis for seizure control in patients with mesial temporal semiology. However, using invasive monitoring, Ebersole and Pacia demonstrated that ictal EEG pattern at seizure onset often correlates with specific location of epileptogenic focus.24 Delta (slow) rhythm (Type II, 2–5 Hz) had 84% specificity of epileptogenic foci originating in the temporal neo-cortical regions and theta (faster) rhythm (Type I, 5–9 Hz) had 86% of predicting epileptogenic foci originating in the hippocampal (mesial) area. Nonetheless, it is still uncertain whether these correlations between ictal EEG rhythm and foci location can be equally applied to both N-MRI and MRI-MTS populations. In addition, theta ictal (Type I) EEG rhythm has been suggested to correlate with better surgical outcome,25 even in patients with normal MRI.17 Our data support the thought that surgical resection in patients with temporal lobe epilepsy (TLE) and negative MRI is less effective than in patients with MRI evidence of mesial temporal sclerosis.15–17,26,27 Our result of 59.1% seizure free in NMRI patients is similar to the results of smaller cohorts.11,16 Nonetheless, theta rhythm at ictal onset seems to be most commonly associated with TLE even in the N-MRI population. However, our data also suggest that better outcome can be achieved in a subgroup of the N-MRI population. Class 1 outcome in N-MRI patients increased from 57.8% to 71% if patients had delta rhythm on EEG at seizure onset as opposed to theta rhythm. This percentage of Engel class I outcome is the same as MRI-MTS patients that had similar ictal EEG rhythm. We hypothesize that delta EEG pattern at seizure onset may suggest similar physiology but not necessarily similar pathology of epileptogenic focus regardless of pre-operative MRI findings. In contrast, a similar assessment between pre-operative theta ictal EEG rhythms in the N-MRI versus MRI-MTS population cannot be made. Interestingly, we found a lack of statistical significance in the EEG ictal pattern (delta vs. theta) between the N-MRI and MRIMTS groups. In addition, previously published studies have touted the idea of theta rhythm as a good predictor in outcome for N-MRI patients.18 The most likely explanation for this dissimilarity is that a categorical classification between delta and theta is too simplistic, and there are several other criteria used in the analysis of ictal EEG patterns, such as semiology (video monitoring), time relationship of EEG onset vs. clinical onset, spatial distribution, and seizure evolution. We hypothesized that this group of patients share a common electrophysiology within the temporal lobe but not necessarily a common inciting pathology. Patients with TLE can present with structural
abnormalities, which may extend beyond hippocampal pathology. We can imagine that simply due to geographic proximity, a seizure focus may incite stimulation of the hippocampus (mesial structures) that could result in clinical presentation consistent with mesial TLE without MRI abnormalities. Temporal lobe surgery remains safe and effective for the treatment of medically refractory epilepsy. This study includes a fairly homogenous group of patients that presented with similar semiology and anterior-basal temporal ictal onset based on surface or invasive EEG that underwent a similar surgical approach despite not so similar radiographic findings. To the best of our knowledge, our surgical series in patients with temporal lobe epilepsy patients and normal MRI is the largest cohort in the literature. Our analysis of comparing results based on MRI findings and EEG rhythm at seizure onset is also unique. Limitations of our study include the retrospective study design, lack of inter-ictal EEG analysis, lack of invasive monitoring in all patients, lack of neo-cortical tissue for pathological analysis, number of pre-operative seizures was not a controlled factor and un-blinded evaluators. In addition, larger number of patients in each category could allow for multivariate analysis which may result in additional predictive factors for outcome. 5. Conclusion Surgical treatment for patients with intractable temporal lobe epilepsy is effective. Patients that had MRI evidence of mesial temporal sclerosis had better chance for seizure control after surgery when compared to patients that had negative MRI. Surgical outcome is improved in a subgroup of negative MRI population that had delta EEG rhythm at seizure onset. Although the overall rate of seizure control is not as high in patients with MTS-MRI, 59.1% of patients in this group can still be seizure free after surgical resection. This observation suggests that the functional anatomy of the anterior mesial temporal lobe may vary between individuals with mesial TLE. Therefore, early referral to an epilepsy center for further evaluation, even in situations when high-resolution brain MRI is normal needs to be emphasized. However, caution is recommended since TLE may include patients with different pathology but similar physiology during the pre-operative evaluation and larger group of patients may be needed for a more accurate correlation. Conflict of interest None of the authors has any conflict of interest to disclose. References 1. Spencer SS. When should temporal-lobe epilepsy be treated surgically? Lancet Neurol 2002;1:375–82. 2. Wiebe S, Blume WT, Girvin JP, Eliasziw M. A randomized, controlled trial of surgery for temporal-lobe epilepsy. N Engl J Med 2001;345:311–8.
24
T. Lau et al. / Seizure 23 (2014) 20–24
3. Spencer SS, Berg AT, Vickrey BG, Sperling MR, Bazil CW, Shinnar S, et al. Predicting long-term seizure outcome after resective epilepsy surgery: the multicenter study. Neurology 2005;65:912–8. 4. Smyth MD, Limbrick Jr DD, Ojemann JG, Zempel J, Robinson S, O’Brien DF, et al. Outcome following surgery for temporal lobe epilepsy with hippocampal involvement in preadolescent children: emphasis on mesial temporal sclerosis. J Neurosurg 2007;106:205–10. 5. Ramos E, Benbadis S, Vale FL. Failure of temporal lobe resection for epilepsy in patients with mesial temporal sclerosis: results and treatment options. J Neurosurg 2009;110:1127–34. 6. Kelemen A, Barsi P, Eross L, Vajda J, Czirjak S, Borbely C, et al. Long-term outcome after temporal lobe surgery – prediction of late worsening of seizure control. Seizure 2006;15:49–55. 7. Jeha LE, Najm IM, Bingaman WE, Khandwala F, Widdess-Walsh P, Morris HH, et al. Predictors of outcome after temporal lobectomy for the treatment of intractable epilepsy. Neurology 2006;66:1938–40. 8. Kumlien E, Doss RC, Gates JR. Treatment outcome in patients with mesial temporal sclerosis. Seizure 2002;11:413–7. 9. Jutila L, Immonen A, Mervaala E, Partanen J, Partanen K, Puranen M, et al. Long term outcome of temporal lobe epilepsy surgery: analyses of 140 consecutive patients. J Neurol Neurosurg Psychiatry 2002;73:486–94. 10. Cascino GD. Surgical treatment for epilepsy. Epilepsy Res 2004;60:179–86. 11. Bell ML, Rao S, So EL, Trenerry M, Kazemi N, Stead SM, et al. Epilepsy surgery outcomes in temporal lobe epilepsy with a normal MRI. Epilepsia 2009;50:2053–60. 12. Alarcon G, Valentin A, Watt C, Selway RP, Lacruz ME, Elwes RD, et al. Is it worth pursuing surgery for epilepsy in patients with normal neuroimaging? J Neurol Neurosurg Psychiatry 2006;77:474–80. 13. Immonen A, Jutila L, Muraja-Murro A, Mervaala E, Aikia M, Lamusuo S, et al. Long-term epilepsy surgery outcomes in patients with MRI-negative temporal lobe epilepsy. Epilepsia 2010;51:2260–9. 14. Cohen-Gadol AA, Wilhelmi BG, Collignon F, White JB, Britton JW, Cambier DM, et al. Long-term outcome of epilepsy surgery among 399 patients with nonlesional seizure foci including mesial temporal lobe sclerosis. J Neurosurg 2006;104:513–24.
15. Chapman K, Wyllie E, Najm I, Ruggieri P, Bingaman W, Luders J, et al. Seizure outcome after epilepsy surgery in patients with normal preoperative MRI. J Neurol Neurosurg Psychiatry 2005;76:710–3. 16. Holmes MD, Born DE, Kutsy RL, Wilensky AJ, Ojemann GA, Ojemann LM. Outcome after surgery in patients with refractory temporal lobe epilepsy and normal MRI. Seizure 2000;9:407–11. 17. Sylaja PN, Radhakrishnan K, Kesavadas C, Sarma PS. Seizure outcome after anterior temporal lobectomy and its predictors in patients with apparent temporal lobe epilepsy and normal MRI. Epilepsia 2004;45:803–8. 18. Tatum WO, Benbadis SR, Hussain A, Al-Saadi S, Kaminski B, Heriaud LS, et al. Ictal EEG remains the prominent predictor of seizure-free outcome after temporal lobectomy in epileptic patients with normal brain MRI. Seizure 2008;17: 631–6. 19. Vale FL, Effio E, Arredondo N, Bozorg A, Wong K, Martinez C, et al. Efficacy of temporal lobe surgery for epilepsy in patients with negative MRI for mesial temporal lobe sclerosis. J Clin Neurosci 2012;19:101–6. 20. Engel J. Surgical treatment of the epilepsies. New York: Raven Press; 1993. 21. Duckworth EA, Vale FL. Trephine epilepsy surgery: the inferior temporal gyrus approach. Neurosurgery 2008;63:ONS156–6 [discussion ONS160– 151]. 22. Vossler DG, Kraemer DL, Haltiner AM, Rostad SW, Kjos BO, Davis BJ, et al. Intracranial EEG in temporal lobe epilepsy: location of seizure onset relates to degree of hippocampal pathology. Epilepsia 2004;45:497–503. 23. Steinhoff BJ, So NK, Lim S, Luders HO. Ictal scalp EEG in temporal lobe epilepsy with unitemporal versus bitemporal interictal epileptiform discharges. Neurology 1995;45:889–96. 24. Ebersole JS, Pacia SV. Localization of temporal lobe foci by ictal EEG patterns. Epilepsia 1996;37:386–99. 25. Assaf BA, Ebersole JS. Visual and quantitative ictal EEG predictors of outcome after temporal lobectomy. Epilepsia 1999;40:52–61. 26. Kuzniecky R, Burgard S, Faught E, Morawetz R, Bartolucci A. Predictive value of magnetic resonance imaging in temporal lobe epilepsy surgery. Arch Neurol 1993;50:65–9. 27. Madhavan D, Kuzniecky R. Temporal lobe surgery in patients with normal MRI. Curr Opin Neurol 2007;20:203–7.
Contents lists available at ScienceDirect
Seizure journal homepage: www.elsevier.com/locate/yseiz
Temporal lobe surgery in medically refractory epilepsy: A comparison between populations based on MRI findings Tsz Lau a, Timothy Miller a, Travis Klein a, Selim R. Benbadis b, Fernando L. Vale a,* a b
Department of Neurosurgery and Brain Repair, Morsani College of Medicine, University of South Florida, Tampa, United States Department of Neurology, Morsani College of Medicine, University of South Florida, Tampa, United States
A R T I C L E I N F O
A B S T R A C T
Article history: Received 14 May 2013 Received in revised form 12 August 2013 Accepted 6 September 2013
Introduction: High resolution MRI findings suggestive of mesial temporal sclerosis (MRI-MTS) correlate with good outcome after surgery. However, a large group of patients present with normal brain MRI (NMRI) and temporal lobe epilepsy (TLE). We aim to compare pre-operative ictal EEG patterns in predicting surgical outcomes in the population with MRI-MTS vs. N-MRI after selective anterior-mesial temporal lobe (AMTL) resection. Methods: 241 patients with unilateral anterior ictal EEG findings underwent selective AMTL resection. 143 MRI-MTS and 98 N-MRI patients were identified. Outcome was based on the modified Engel classification, ictal EEG pattern at seizure onset, demographics and MRI findings. Results: Seizure-free outcome was seen in the MRI-MTS in 79% of patients, compared to 59.1% (p < .005) of the N-MRI group. No significant difference was identified in ictal EEG patterns at presentation between groups. Class I outcome was achieved in 78.9% of patients that had theta rhythm and MRI-MTS compared to 57.9% of patients that had theta rhythm and N-MRI (p < 0.05). Discussion and conclusion: Surgical treatment for mesial TLE is effective. Positive MRI suggestive of mesial temporal sclerosis (MTS) predicts better seizure control after surgery. Theta rhythm is the most common ictal pattern and seems to carry the best prognosis for TLE. However, a well-selected group of patients with N-MRI will benefit from surgical intervention, and similar outcome to MRI-MTS patients can be achieved if delta ictal EEG pattern is presented at initial onset. Early referral to an epilepsy center cannot be emphasized enough, even in situations when high-resolution brain MRI is normal. ß 2013 British Epilepsy Association. Published by Elsevier Ltd. All rights reserved.
Keywords: Epilepsy surgery Temporal lobe Nonlesional Negative MRI Ictal EEG Outcomes
1. Introduction Mesial temporal sclerosis (MTS) is typically seen in medically refractory mesial temporal lobe epilepsy. About 75% of patients with mesial temporal lobe epilepsy (TLE) are resistant to medical management.1 Surgical resection usually provides excellent seizure control in patients with TLE,2 especially in the setting of positive magnetic resonance imaging (MRI) evidence of mesial temporal sclerosis (MRI-MTS).3–10 Surgery for patients with temporal lobe epilepsy and normal MRI findings (N-MRI) is less effective.11–17 Studies have shown 37– 76% seizure control in this group of patients, but they were either
Abbreviations: MTS, mesial temporal sclerosis; TLE, temporal lobe epilepsy; MRIMTS, mesial temporal sclerosis positive MRI; N-MRI, normal MRI; AMTL, anterior mesial temporal lobe; MTLE, mesial temporal lobe epilepsy. * Corresponding author at: Department of Neurosurgery and Brain Repair, University of South Florida, 2 Tampa General Circle, Tampa, FL 33606, United States. Tel.: +1 813 259 0605; fax: +1 813 259 0944. E-mail address: [email protected] (F.L. Vale).
retrospective in nature, included heterogeneous groups of patients with small a percentage of patients with normal MRI, or were based on small cohorts of patients. Still to be well elucidated is the contribution of ictal EEG patterns to the outcome of temporal lobe surgery in treating TLE. Some small studies have demonstrated a difference in anterior temporal ictal rhythms (slow (delta) vs. faster (theta) rhythms) in patients with MRI-MTS vs. N-MRI, but further research into this issue is likely warranted.18 The goal of this study is to compare the surgical outcomes in patients with mesial TLE with either N-MRI or ipsilateral MRI-MTS, and to identify factors that would predict better outcome based on pre-operative clinical evaluation. 2. Methods 2.1. Patient population Data were obtained from our institution’s epilepsy surgery database. This is a prospective database established in 1998 after
1059-1311/$ – see front matter ß 2013 British Epilepsy Association. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.seizure.2013.09.004
T. Lau et al. / Seizure 23 (2014) 20–24
receiving the institutional Internal Review Board approval. Informed consent was obtained from all patients. Patients with mesial temporal semiology and documented anterior temporal ictal recordings that underwent anterior-mesial temporal lobe (AMTL) resection for temporal lobe epilepsy were identified. Patients were divided into two groups based on results of preoperative high resolution MRI. The first group included patients with MRI evidence of ipsilateral mesial temporal sclerosis, the second included patients with normal MRI. Patient evaluation for surgical treatment as well as MRI protocol has been previously described.19 All patients were initially evaluated by board certified, fellowship trained epileptologists. Patients with semiology that suggested mesial temporal lobe epilepsy (MTLE) proceeded to phase 1 evaluation: long-term video-EEG monitoring. Sphenoidal or nasopharyngeal electrodes were not used on any patients, but bilateral basilar temporal placement, such as T1, T2 electrodes, were done routinely. Ictal EEG pattern was recorded at seizure onset. Anterior temporal rhythms were recorded as follow: (a) Slow (delta) rhythm (2– 4 Hz), (b) Faster rhythms (theta rhythm (5–7 Hz) or alpha (8– 12 Hz)) and (c) delta–theta (progression from slow to faster rhythms within 5 s). All EEG recordings were reviewed in the weekly comprehensive epilepsy surgery conference. At least two board-certified epileptologists were present on each conference and consensus was achieved between the groups. Inter-ictal positron emission tomography (PET) was performed if possible. A FDG-PET scan was considered positive when it showed an area of hypometabolism in the ipsilateral mesial-basal temporal lobe. Subtraction ictal single photon emission CT (SPECT) coregistered to MRI (SISCOM) was not available at our institution and only a few of our patients had SPECT imaging for statistical analysis. Following initial evaluation with video-EEG monitoring, patients received imaging with high-resolution MR using a 1.5 or 3.0-T magnet with special focus on the temporal lobes. Coronal and axial T2-weighted and FLAIR images were obtained and reviewed by a board-certified neuro-radiologist. Abnormal signal on FLAIR and T2weighted images, decreased volume and loss of anatomical configuration of the hippocampal formation are considered the hallmarks of radiographically confirmed MTS. In addition, most patients also underwent Wada testing to evaluate for memory asymmetry. Memory recall 2.5/8 points difference was considered asymmetric. Some of these pre-operative evaluations were not performed due to patient financial restraints, patient refusal, or patient non-compliance due to young age or low intelligence quotient. Decisions regarding surgical therapy were made after a review of all the clinical data for each patient in conjunction with the epilepsy neurologists, neurosurgeon and neuroradiologist of the Comprehensive Epilepsy Center. Invasive subdural contact electrode placement (phase II evaluation) along the basal and lateral neocortex was considered when: (i) there was an inadequate ictal recording due to extensive muscle artifact or poorly localized ictal onset in the setting of typical mesial temporal semiology or (ii) there was discordant pre-operative evaluation. All patients were followed at 3, 6, and 12 months, then yearly after. All patients with <1 year followup were excluded from analysis. Clinical outcome was based on the modified Engel classification: Class I, seizure free or residual aura (CL-I); Class II, rare disabling seizures (<3 complex partial seizures per year) (CL-II); Class III, worthwhile seizure reduction; and Class IV, no worthwhile improvement.20 2.2. Surgical technique Details of the surgical technique have been previously well described.21 In brief, a temporal trephine craniotomy is performed. The inferior temporal gyrus is identified and resected. Cortical resection is performed from lateral to medial structures. Landmarks
21
along the way from lateral to medial are the occipitotemporal sulcus, fusiform gyrus, and collateral sulcus. Next, the floor of the temporal horn is opened, followed by the resection of parahippocampal gyrus, amygdala, and uncus. Lastly, the hippocampus is resected en bloc, which can then be used for pathological analysis. Subpial fashion dissection of the mesial structure ensures protection of the underlying brain stem and perimesencephalic vasculature. 2.3. Statistical analysis Microsoft Excel (Microsoft, Redmond, WA, USA) software was used to store prospective clinical data. IBM Statistical Package for the Social Sciences 19.0 for Windows was used to perform statistical analysis. Results of pre-operative work-up (Wada test, ictal EEG pattern at seizure onset, PET) were evaluated to identify predictors of surgical outcome based on the modified Engel Classification. Fisher Exact test and Chi-squared were used to analyze categorical variables. The non-parametric Mann–Whitney U-test was used to test the distribution of age among the two groups. Statistically significant is defined as p value 0.05. 3. Results A total of 299 patients with TLE and unilateral ictal EEG findings that underwent selective AMTL resection were identified from our database. After excluding patients with less than 1-year follow-up, or dual pathology, 241 patients remained for statistical analysis. One hundred and forty three patients had evidence of mesial temporal sclerosis on preoperative high resolution MRI and 98 patients had normal high resolution MRI. 17 patients in the N-MRI group underwent phase II invasive monitoring as opposed to 5 patients in the MRI-MTS group. Demographics and background characteristics are summarized in Table 1. Mean follow-up was 2.9 years for the NMRI group and 3.4 years for the MRI-MTS group. The average age for both groups was 38-year-old. Sex distribution and the results of Wada testing between the two groups were also well matched. The only difference in pre-operative evaluation was a higher percentage of patients in the MRI-MTS group with positive PET study compared to the N-MRI group (94% vs. 81%, respectively). PET scan was performed in 94 (96%) patients in the N-MRI, whereas 76 (81%) demonstrated ipsilateral hypometabolism. In the MRI-MTS group, 99 (71%) studies were performed, while 93 (94%) studies were deemed positive. On the other hand, Wada test was performed in 91 patients in the N-MRI whereas 45 (49%) demonstrated significant asymmetry. In the MRI-MTS group, 130 tests were performed; however, only 69 (53%) patients demonstrated significant difference. Overall outcome based on modified Engel classification was statistically significant between the two groups (p < 0.05). In the MRI-MTS group, 79% had class I outcome (n = 113) compared to 59.2% (n = 58) in N-MRI group at last follow-up (Fig. 1). In patients with positive PET, similar percentages were observed (p < 0.05) Table 1 Patient characteristics.
Avg. age (range) Avg. F/U (range) Sex M F PET scan Positive Negative Wada test Asymmetry No Asymmetry
Negative MRI (n = 98)
MRI-MTS (n = 143)
p-Value
38 (10–65) 2.9 years (1–11.5)
38 (7–68) 3.4 years (1–11.5)
0.857
40% 60% n = 94 81% 19% n = 91 49% 50%
41% 59% n = 99 94% 6% n = 130 53% 47%
0.905
0.006
0.595
T. Lau et al. / Seizure 23 (2014) 20–24
22
Overall Outcome 90.0 80.0 Percentage
70.0 60.0 MTS-MRI (n=143)
50.0
nMRI (n=98)
40.0 30.0 20.0 10.0 0.0 IIIIIIIV
Engel Class
Percentage
Fig. 1. Overall outcome – negative MRI vs. MRI-MTS, p < 0.05.
90.00% 80.00% 70.00% 60.00% 50.00% 40.00% 30.00% 20.00% 10.00% 0.00%
MTS-MRI (n=93) nMRI (n=76)
I
II,III,IV Outcome
Fig. 2. Overall outcome in PET/SPECT positive patients, p < 0.05.
Theta
Percentage
100.00% 80.00% 60.00%
MTS-MRI (n=95)
40.00%
nMRI (n=57)
20.00% 0.00% I
Outcome
II,III,IV
Fig. 3. Outcome in patients with theta rhythm, p < 0.05.
(Fig. 2). Ictal EEG pattern at seizure onset is summarized in Table 2. Nine patients had limited or missing pre-operative ictal EEG documentation to allow adequate classification during the retrospective review process. Of the remaining 232 patients, theta rhythm was the most common ictal pattern identified for TLE. Ictal theta onset was identified in one hundred and fifty four (66.3%) out of 232 patients. Slow (delta) pattern was seen in 35 out of 232 (15%), while delta/theta was seen in 13.3% (31 out of 232) of the total population. Twelve patients had a poorly localizable ictal EEG pattern. Faster theta rhythm was seen in 58.2% in the N-MRI group and 70.9% in the MRI-MTS group. Delta rhythm was seen in 21.4% and 10.4% respectively. 14.3% of N-MRI patients had Delta–theta rhythm compared to 12.7% of MRI-MTS patients. Similar number of patients in both groups had nonTable 2 Ictal EEG pattern at seizure onset.
Theta, n = 154 (66.4%) Delta, n = 35 (15.1%) Delta/theta, n = 31 (13.4%) Non-localized, n = 12 (5.2%)
Negative MRI, n = 98
MTS MRIa, n = 134
59 21 14 4
95 14 17 8
(60.2%) (21.4%) (14.3%) (4.1%)
(70.9%) (10.4%) (12.7%) (6.0%)
p = 0.11. a 9 patients with missing or limited pre-operative ictal recordings.
localized scalp rhythm. There was no statistically significant difference in the distribution of ictal EEG pattern between the groups (p = 0.11). Class I outcome was achieved in 75 out of 95 (78.9%) of patients that had theta rhythm and MRI-MTS compared to 33 out of 57 (57.9%) of patients that had theta rhythm and N-MRI (p < 0.05) (Fig. 3). For patients that had delta rhythm, class I outcome was achieved in 71% in both groups (Fig. 4). Class I outcome was achieved 64% of N-MRI patients (n = 14) and in 82% of MRI-MTS patients (n = 17) that had delta–theta rhythm (p > 0.05). Histopathology demonstrated hippocampal sclerosis (severe, long-standing) in all cases with pre-operative MRI-MTS. However, N-MRI population had evidence of neuronal loss and gliosis at the level of hippocampus compatible with mild to moderate MTS in 40 out of 98 patients (41%). No severe MTS was identified in the latter group of patients. Overall surgical complication rate was 2.9%. One patient had a superficial wound infection that required wound revision. One patient developed cerebral spinal fluid leak that required surgical repair. Four patients developed post-operative subdural hematoma (one subacute and three chronic) in which 2 cases required surgical evacuation in a delayed fashion without any long-term morbidity. One patient developed minor lacunar infarct postoperatively, remote from the operative site.
Percentage
T. Lau et al. / Seizure 23 (2014) 20–24
23
Delta
80.00% 70.00% 60.00% 50.00% 40.00% 30.00% 20.00% 10.00% 0.00%
MTS-MRI (n=14) nMRI (n=21)
I
II,III,IV Outcome
Fig. 4. Outcome in patients with delta rhythm, p = 1.
4. Discussion Previous reports have demonstrated that theta rhythm is the most common ictal anterior temporal pattern in the setting of TLE.22,23 Our results support the idea that theta rhythm at ictal onset seems to be most commonly associated with TLE and carries the best prognosis for seizure control in patients with mesial temporal semiology. However, using invasive monitoring, Ebersole and Pacia demonstrated that ictal EEG pattern at seizure onset often correlates with specific location of epileptogenic focus.24 Delta (slow) rhythm (Type II, 2–5 Hz) had 84% specificity of epileptogenic foci originating in the temporal neo-cortical regions and theta (faster) rhythm (Type I, 5–9 Hz) had 86% of predicting epileptogenic foci originating in the hippocampal (mesial) area. Nonetheless, it is still uncertain whether these correlations between ictal EEG rhythm and foci location can be equally applied to both N-MRI and MRI-MTS populations. In addition, theta ictal (Type I) EEG rhythm has been suggested to correlate with better surgical outcome,25 even in patients with normal MRI.17 Our data support the thought that surgical resection in patients with temporal lobe epilepsy (TLE) and negative MRI is less effective than in patients with MRI evidence of mesial temporal sclerosis.15–17,26,27 Our result of 59.1% seizure free in NMRI patients is similar to the results of smaller cohorts.11,16 Nonetheless, theta rhythm at ictal onset seems to be most commonly associated with TLE even in the N-MRI population. However, our data also suggest that better outcome can be achieved in a subgroup of the N-MRI population. Class 1 outcome in N-MRI patients increased from 57.8% to 71% if patients had delta rhythm on EEG at seizure onset as opposed to theta rhythm. This percentage of Engel class I outcome is the same as MRI-MTS patients that had similar ictal EEG rhythm. We hypothesize that delta EEG pattern at seizure onset may suggest similar physiology but not necessarily similar pathology of epileptogenic focus regardless of pre-operative MRI findings. In contrast, a similar assessment between pre-operative theta ictal EEG rhythms in the N-MRI versus MRI-MTS population cannot be made. Interestingly, we found a lack of statistical significance in the EEG ictal pattern (delta vs. theta) between the N-MRI and MRIMTS groups. In addition, previously published studies have touted the idea of theta rhythm as a good predictor in outcome for N-MRI patients.18 The most likely explanation for this dissimilarity is that a categorical classification between delta and theta is too simplistic, and there are several other criteria used in the analysis of ictal EEG patterns, such as semiology (video monitoring), time relationship of EEG onset vs. clinical onset, spatial distribution, and seizure evolution. We hypothesized that this group of patients share a common electrophysiology within the temporal lobe but not necessarily a common inciting pathology. Patients with TLE can present with structural
abnormalities, which may extend beyond hippocampal pathology. We can imagine that simply due to geographic proximity, a seizure focus may incite stimulation of the hippocampus (mesial structures) that could result in clinical presentation consistent with mesial TLE without MRI abnormalities. Temporal lobe surgery remains safe and effective for the treatment of medically refractory epilepsy. This study includes a fairly homogenous group of patients that presented with similar semiology and anterior-basal temporal ictal onset based on surface or invasive EEG that underwent a similar surgical approach despite not so similar radiographic findings. To the best of our knowledge, our surgical series in patients with temporal lobe epilepsy patients and normal MRI is the largest cohort in the literature. Our analysis of comparing results based on MRI findings and EEG rhythm at seizure onset is also unique. Limitations of our study include the retrospective study design, lack of inter-ictal EEG analysis, lack of invasive monitoring in all patients, lack of neo-cortical tissue for pathological analysis, number of pre-operative seizures was not a controlled factor and un-blinded evaluators. In addition, larger number of patients in each category could allow for multivariate analysis which may result in additional predictive factors for outcome. 5. Conclusion Surgical treatment for patients with intractable temporal lobe epilepsy is effective. Patients that had MRI evidence of mesial temporal sclerosis had better chance for seizure control after surgery when compared to patients that had negative MRI. Surgical outcome is improved in a subgroup of negative MRI population that had delta EEG rhythm at seizure onset. Although the overall rate of seizure control is not as high in patients with MTS-MRI, 59.1% of patients in this group can still be seizure free after surgical resection. This observation suggests that the functional anatomy of the anterior mesial temporal lobe may vary between individuals with mesial TLE. Therefore, early referral to an epilepsy center for further evaluation, even in situations when high-resolution brain MRI is normal needs to be emphasized. However, caution is recommended since TLE may include patients with different pathology but similar physiology during the pre-operative evaluation and larger group of patients may be needed for a more accurate correlation. Conflict of interest None of the authors has any conflict of interest to disclose. References 1. Spencer SS. When should temporal-lobe epilepsy be treated surgically? Lancet Neurol 2002;1:375–82. 2. Wiebe S, Blume WT, Girvin JP, Eliasziw M. A randomized, controlled trial of surgery for temporal-lobe epilepsy. N Engl J Med 2001;345:311–8.
24
T. Lau et al. / Seizure 23 (2014) 20–24
3. Spencer SS, Berg AT, Vickrey BG, Sperling MR, Bazil CW, Shinnar S, et al. Predicting long-term seizure outcome after resective epilepsy surgery: the multicenter study. Neurology 2005;65:912–8. 4. Smyth MD, Limbrick Jr DD, Ojemann JG, Zempel J, Robinson S, O’Brien DF, et al. Outcome following surgery for temporal lobe epilepsy with hippocampal involvement in preadolescent children: emphasis on mesial temporal sclerosis. J Neurosurg 2007;106:205–10. 5. Ramos E, Benbadis S, Vale FL. Failure of temporal lobe resection for epilepsy in patients with mesial temporal sclerosis: results and treatment options. J Neurosurg 2009;110:1127–34. 6. Kelemen A, Barsi P, Eross L, Vajda J, Czirjak S, Borbely C, et al. Long-term outcome after temporal lobe surgery – prediction of late worsening of seizure control. Seizure 2006;15:49–55. 7. Jeha LE, Najm IM, Bingaman WE, Khandwala F, Widdess-Walsh P, Morris HH, et al. Predictors of outcome after temporal lobectomy for the treatment of intractable epilepsy. Neurology 2006;66:1938–40. 8. Kumlien E, Doss RC, Gates JR. Treatment outcome in patients with mesial temporal sclerosis. Seizure 2002;11:413–7. 9. Jutila L, Immonen A, Mervaala E, Partanen J, Partanen K, Puranen M, et al. Long term outcome of temporal lobe epilepsy surgery: analyses of 140 consecutive patients. J Neurol Neurosurg Psychiatry 2002;73:486–94. 10. Cascino GD. Surgical treatment for epilepsy. Epilepsy Res 2004;60:179–86. 11. Bell ML, Rao S, So EL, Trenerry M, Kazemi N, Stead SM, et al. Epilepsy surgery outcomes in temporal lobe epilepsy with a normal MRI. Epilepsia 2009;50:2053–60. 12. Alarcon G, Valentin A, Watt C, Selway RP, Lacruz ME, Elwes RD, et al. Is it worth pursuing surgery for epilepsy in patients with normal neuroimaging? J Neurol Neurosurg Psychiatry 2006;77:474–80. 13. Immonen A, Jutila L, Muraja-Murro A, Mervaala E, Aikia M, Lamusuo S, et al. Long-term epilepsy surgery outcomes in patients with MRI-negative temporal lobe epilepsy. Epilepsia 2010;51:2260–9. 14. Cohen-Gadol AA, Wilhelmi BG, Collignon F, White JB, Britton JW, Cambier DM, et al. Long-term outcome of epilepsy surgery among 399 patients with nonlesional seizure foci including mesial temporal lobe sclerosis. J Neurosurg 2006;104:513–24.
15. Chapman K, Wyllie E, Najm I, Ruggieri P, Bingaman W, Luders J, et al. Seizure outcome after epilepsy surgery in patients with normal preoperative MRI. J Neurol Neurosurg Psychiatry 2005;76:710–3. 16. Holmes MD, Born DE, Kutsy RL, Wilensky AJ, Ojemann GA, Ojemann LM. Outcome after surgery in patients with refractory temporal lobe epilepsy and normal MRI. Seizure 2000;9:407–11. 17. Sylaja PN, Radhakrishnan K, Kesavadas C, Sarma PS. Seizure outcome after anterior temporal lobectomy and its predictors in patients with apparent temporal lobe epilepsy and normal MRI. Epilepsia 2004;45:803–8. 18. Tatum WO, Benbadis SR, Hussain A, Al-Saadi S, Kaminski B, Heriaud LS, et al. Ictal EEG remains the prominent predictor of seizure-free outcome after temporal lobectomy in epileptic patients with normal brain MRI. Seizure 2008;17: 631–6. 19. Vale FL, Effio E, Arredondo N, Bozorg A, Wong K, Martinez C, et al. Efficacy of temporal lobe surgery for epilepsy in patients with negative MRI for mesial temporal lobe sclerosis. J Clin Neurosci 2012;19:101–6. 20. Engel J. Surgical treatment of the epilepsies. New York: Raven Press; 1993. 21. Duckworth EA, Vale FL. Trephine epilepsy surgery: the inferior temporal gyrus approach. Neurosurgery 2008;63:ONS156–6 [discussion ONS160– 151]. 22. Vossler DG, Kraemer DL, Haltiner AM, Rostad SW, Kjos BO, Davis BJ, et al. Intracranial EEG in temporal lobe epilepsy: location of seizure onset relates to degree of hippocampal pathology. Epilepsia 2004;45:497–503. 23. Steinhoff BJ, So NK, Lim S, Luders HO. Ictal scalp EEG in temporal lobe epilepsy with unitemporal versus bitemporal interictal epileptiform discharges. Neurology 1995;45:889–96. 24. Ebersole JS, Pacia SV. Localization of temporal lobe foci by ictal EEG patterns. Epilepsia 1996;37:386–99. 25. Assaf BA, Ebersole JS. Visual and quantitative ictal EEG predictors of outcome after temporal lobectomy. Epilepsia 1999;40:52–61. 26. Kuzniecky R, Burgard S, Faught E, Morawetz R, Bartolucci A. Predictive value of magnetic resonance imaging in temporal lobe epilepsy surgery. Arch Neurol 1993;50:65–9. 27. Madhavan D, Kuzniecky R. Temporal lobe surgery in patients with normal MRI. Curr Opin Neurol 2007;20:203–7.