Long-term epilepsy surgery outcomes in patients with PET-positive, MRI-negative temporal lobe epilepsy

Epilepsy & Behavior 41 (2014) 91–97

Contents lists available at ScienceDirect

Epilepsy & Behavior journal homepage: www.elsevier.com/locate/yebeh

Long-term epilepsy surgery outcomes in patients with PET-positive, MRI-negative temporal lobe epilepsy Peng-Fan Yang a,⁎, Jia-Sheng Pei a, Hui-Jian Zhang a, Qiao Lin b, Zhen Mei b, Zhong-Hui Zhong b, Jun Tian a, Yan-Zeng Jia b, Zi-Qian Chen c, Zhi-Yong Zheng d a

Department of Neurosurgery, Fuzhou General Hospital of Nanjing Command, PLA, Fuzhou 350025, China Department of Epileptology, Fuzhou General Hospital of Nanjing Command, PLA, Fuzhou 350025, China Department of Neuroradiology, Fuzhou General Hospital of Nanjing Command, PLA, Fuzhou 350025, China d Department of Pathology, Fuzhou General Hospital of Nanjing Command, PLA, Fuzhou 350025, China b c

a r t i c l e

i n f o

Article history: Received 1 August 2014 Revised 18 September 2014 Accepted 19 September 2014 Available online xxxx Keywords: Temporal lobe epilepsy Nonlesional MRI Presurgical evaluation Postoperative outcomes Positron emission tomography (PET) Anterior temporal lobectomy (ATL)

a b s t r a c t This study compared the long-term efficacy of anterior temporal lobectomy (ATL) for the treatment of medically refractory temporal lobe epilepsy (TLE) in patients who presented with ipsilateral temporal PET hypometabolism and nonlesional magnetic resonance imaging (PET+/MRI−) with that in patients who had mesial temporal sclerosis (MTS) on MRI. We described the electroclinical, MRI, PET, and pathological characteristics and seizure outcome of 28 PET+/MRI − patients without discordant ictal and interictal electroencephalography (EEG) who underwent ATL (2004–2007) for medically refractory partial epilepsy while avoiding intracranial monitoring. The primary outcome was the percentages of Engel Class I outcomes at 2 and 5 years of PET+/MRI− patients compared with those of patients with MTS on MRI; neuropsychological testing was used as the secondary outcome. At 2-year follow-up, 21 (75%) patients in the PET+/MRI − group were in Engel Class I compared with 66 (75.9%) patients with MTS, and at 5-year follow-up, 20 (71.4%) patients in the PET+/MRI− group were in Engel Class I compared with 64 (73.6%) patients in the group with MTS. There were no significant differences between the groups at either time period. We concluded that normal MRI results should not preclude presurgical evaluations in patients with medically refractory TLE, as favorable long-term postoperative seizure outcomes are possible, especially in patients with unilateral anterior interictal epileptiform discharges and ipsilateral temporal PET hypometabolism. © 2014 Elsevier Inc. All rights reserved.

1. Introduction Anterior temporal lobectomy (ATL) has been well established as an effective treatment for medically refractory temporal lobe epilepsy (TLE) [1]. To determine who would benefit from ATL, presurgical evaluation of TLE is usually performed using high-resolution magnetic resonance imaging (MRI) due to its high sensitivity (97%) and specificity (83%) for hippocampal sclerosis (HS), the most common pathologic basis of TLE [2]. This is supported by various studies showing that up to 60–80% of patients become seizure-free following an anterior temporal lobectomy (ATL) if an MRI-apparent structural abnormality, such as mesial temporal sclerosis (MTS), is concordant to the seizure-onset zone [2–5], with seizure-free rates for those with normal MRI results varying between 18% and 80% [2,6–16]. However, although it is

⁎ Corresponding author at: Department of Neurosurgery, Fuzhou General Hospital of Nanjing Command, PLA, North Road 156, West Second Ring, Fuzhou 350025, China. Tel.: +86 0591 22859387; fax: +86 0591 87640785. E-mail address: [email protected] (P.-F. Yang).

http://dx.doi.org/10.1016/j.yebeh.2014.09.054 1525-5050/© 2014 Elsevier Inc. All rights reserved.

established that MTS is the most reliable predictor of successful surgical outcomes [2,17], only 58–72% of patients with TLE show signs of MTS on MRI, with 16% of patients with medically refractory TLE demonstrating no MRI abnormality at all [1,2]. This shows a need to supplement MRI evaluation with other measures, including fluorodeoxyglucose-positron emission tomography (PET) and intracranial monitoring. As FDG-PET hypometabolism has been shown to correlate with successful surgical outcomes for TLE [18–22], interictal PET imaging has been routinely used for presurgical evaluation of patients and has been used in our center since 2003. However, unlike MRI, PET results do not show a relationship with HS or MTS. For example, studies correlating PET with pathology have found that the severity of cell loss in the sclerotic hippocampus does not correspond with the degree of PET hypometabolism [23,24], while the pattern of PET hypometabolism does not correspond to the severity of MTS as analyzed using MRI [25]. This suggests that FDG-PET hypometabolism is not necessarily detected using MRI and that PET could be a good supplemental diagnostic technique for MRI− patients. The purpose of our study was to compare surgical outcomes of PET +/MRI − patients with those showing evidence of MTS on MRI in

92

P.-F. Yang et al. / Epilepsy & Behavior 41 (2014) 91–97

order to investigate if PET could add value to MRI. Our primary outcome was seizure recurrence, while the secondary outcome was the comparison of neuropsychological testing before and after ATL surgery.

2. Patients and methods 2.1. Study setting We reviewed the medical records of all patients ≥13 years of age presenting with medically refractory temporal lobe epilepsy who underwent the Spencer-modified ATL without intracranial monitoring between April 2004 and December 2007 at the comprehensive epilepsy center of Fuzhou General Hospital. Our center covers an area in southeast China with a population of 200 million people. We perform noninvasive surgical evaluation to the greatest extent possible. For patients with TLE, we perform ATL directly without intracranial evaluation as long as ictal and interictal EEG and the PET hypometabolism area are consistent, regardless of whether there is ipsilateral HS, or the findings are negative on both sides. We selected 28 PET +/MRI − patients with temporal PET hypometabolism ipsilateral to the surgical site and an MRI that did not reveal an epileptogenic lesion for the study. Patients with any extratemporal MRI findings that were potentially epileptogenic, including hemiatrophy, stroke, encephalomalacia, or other cortical lesions, were excluded from the study. Patients with discordant ictal and interictal electroencephalography (EEG) were excluded. Ictal EEG and interictal EEG were considered discordant, for example, if in patients with unilateral seizure onset and bilateral epileptiform discharges more than 75% of interictal spikes were contralateral to the side of seizure onset. On the other hand, they were not considered discordant if there were equally obvious interictal spikes bilaterally or there were more than 75% of interictal spikes ipsilateral to the side of seizure onset. Patients with bilateral MTS and discordant noninvasive testing who had chronic intracranial EEG monitoring with bilateral temporal depth electrodes and subdural strip grid electrodes were also excluded. Patients with radiographically apparent MTS or other structural abnormalities or lesions in the temporal lobe were excluded from the MRI− group. For comparison, 87 patients with MTS were also included. They were selected prospectively based on MRI. In these patients, the epileptic region was strictly related with MTS, and contralateral temporal lobe MRI was normal. Patients who had MRI findings revealing hippocampal atrophy, with or without increased mesial temporal signal intensity, were included in the group with MTS. Fig. 1 shows PET images and MRI images from a patient who met the inclusion criteria. The patient was a 22-year-old female who had complex partial seizures for 8 years. Electroencephalography in the interictal phase suggested that the seizures were derived from the right temporal lobe. In the interictal phase, the interictal spike during interictal phase epileptic discharge was dominant in the right temporal area (N75%). Pathological examination demonstrated gliosis. The study was approved by the Institutional Review Board of Fuzhou General Hospital of Nanjing Command, PLA. Written informed consent was obtained from all patients.

2.2. MRI evaluation The seizure protocol MRIs were performed using 1.5-T magnets and included coronal T1-spoiled gradient, coronal fluid-attenuated inversion recovery (FLAIR), sagittal T1, and axial T2 spin-echo sequences. Thin slices through the temporal lobes were obtained, with 4-mm slice thickness for the coronal T1 and 2-mm slices for the coronal FLAIR. The neuroradiologist affiliated with the epilepsy program (Dr. Chen Ziqian) initially interpreted all MRIs and reviewed the findings for this study.

2.3. Interictal and ictal EEG All patients had routine interictal scalp EEG and continuous video scalp EEG monitoring to record seizures. Scalp EEG recordings were obtained with 27 electrodes (modified 10–20 montage with subtemporal electrodes [26]). Three aspects of video-EEG monitoring were evaluated: long-term interictal EEG, seizure semiology, and ictal EEG. The seizure semiology was coded dichotomously as the presence or absence of typical temporal semiology [27], defined as a seizure duration longer than 1 min including at least four of the following five characteristics: abdominal or experiential aura, impaired consciousness, occurrence of automatisms, unilateral arm dystonia, or pronounced postictal confusion [27–29]. Interictal spike data were characterized as bilateral equally obvious interictal spikes, with more than 75% of ipsilateral interictal spikes, or no obvious interictal spikes. Ictal EEG seizure patterns were categorized as previously described by Ebersole and Pacia [30]. Briefly, type I seizures were characterized by a progressive buildup of a regular 5- to 9-Hz EEG rhythm of uniform morphology persisting for greater than 5 s with the same morphology localized principally to the subtemporal and temporal electrodes. Type IIA seizures were characterized by irregular EEG rhythms in the 2- to 5-Hz range that were lateralized to one hemisphere but were less often localized only to the temporal electrodes. These rhythms exhibited little stability, and changes in morphology and frequency occurred every few seconds. Additionally, if the type IIA pattern was followed by a type I seizure pattern, it was termed type IIB. Type III seizures were defined as those without a distinct EEG ictal discharge. In these patients, clinical seizure activity was accompanied by EEG showing an interruption of normal background activity commonly combined with a diffuse irregular slowing. Type 1 seizure patterns have been reported to be associated with seizures originating in the hippocampus and type II and III patterns with temporal neocortical seizures [30]. 2.4. Fluorodeoxyglucose-positron emission computed tomography studies All patients underwent interictal 18F-FDG-PET for cerebral metabolic rate for glucose utilization measurement using a GE Discovery LS 16 PET scanner (GE Medical PET/CT System, USA). No patients had experienced seizures for at least 2 days before PET studies, and patients were observed carefully during scans to exclude ictal activity. The 18F-FDGPET results were classified into three categories: normal, those showing unilateral temporal hypometabolism, or other (mainly bitemporal or extratemporal hypometabolism). Combinations of unilateral temporal hypometabolism with confined ipsilateral frontobasal, ipsilateral thalamic, or contralateral cerebellar hypometabolism were considered compatible with unilateral temporal abnormalities [31]. Patients who had discordant PET findings showing hypometabolism in regions other than the temporal lobe, thalamus, or cerebellum, as the latter two regions are commonly hypometabolic in patients with TLE [32, 33], were excluded from the PET+/MRI− group. 2.5. Neuropsychological testing Neuropsychological testing of memory, language, and intelligence was obtained prior to temporal lobectomy and then repeated at 3 months following surgery. The neuropsychological testing performed was as follows: verbal comprehension quotients from a Wechsler intelligence scale, a test of lexical verbal fluency (Controlled Oral Word Association Test), Trail Making Test, a Semantic Fluency Test (animals, fruits, vegetables), Chinese naming test, and the Auditory Verbal Learning Test. 2.6. Surgery All patients had anterior temporal lobectomies under general anesthesia without the aid of intraoperative electrocorticography. The

P.-F. Yang et al. / Epilepsy & Behavior 41 (2014) 91–97

93

Fig. 1. Left panel: FDG-PET (coronal) images with hypometabolism in the right mesial temporal region (A. hippocampal head, B. hippocampal body). Right panel: MRI FLAIR (coronal) images without hippocampal atrophy and signal abnormality bilaterally (A. hippocampal head, B. hippocampal body).

temporal neocortex was removed between 3 and 3.5 cm (dominant temporal lobe) and 3.5 and 4.5 cm (nondominant temporal lobe) from the anterior temporal fossa along the middle temporal gyrus, leaving the upper temporal gyrus intact. The entire hippocampal head – from the choroidal point to the anterior end of the hippocampal head – was exposed and removed en bloc. Including the entorhinal area, the en bloc mass measured more than 2.5 cm in length. The hippocampus was is removed as far back as the level of the superior colliculus. The amygdala and the innermost part of the uncus were removed subpially to expose the anterior choroidal artery and the optic tract. 2.7. Pathological examination The surgical specimens were taken from two defined neuroanatomic regions (i.e., hippocampal formation and temporal pole). Each tissue sample was grossly inspected, measured, and then cut in coronal slices.

A selection of these slices was fixed in 10% buffered formalin, embedded in paraffin, and then stained with hematoxylin and eosin (H&E) stain. Each section was evaluated under light microscopy. All surgical pathologic tissue sections were reviewed by a board-certified neuropathologist (Dr. Zheng Zhiyong) and were classified into HS, malformation of cortical development (MCD), hamartia (atypical collections of small neuronal cells), nonspecific subpial gliosis, or normal histology.

2.8. Outcome assessment Patients' follow-up examinations were performed at 3 and 12 months after ATL and at yearly intervals, thereafter, by chart review, mail-out questionnaire, and telephone contact. Seizure outcomes were assessed using a modified Engel's classification [34]: Class I, seizurefree with or without auras; Class II, greater than 90% reduction in seizure

94

P.-F. Yang et al. / Epilepsy & Behavior 41 (2014) 91–97

with right TLE and those with left TLE, and the adjusted changes from the reference group and 95% confidence interval were calculated. To determine the difference between before and after surgery within the study groups or within patients with right TLE and those with left TLE, paired t-tests were used to examine continuous variables. Statistics Analysis System (SAS) software package, version 9.2 (SAS Institute Inc., Cary, NC, USA) was used for the statistical analysis. All statistical assessments were evaluated at a two-sided P-value of 0.05.

Table 1 Summary of demographic characteristics, interictal spike data, and ictal EEG pattern.

Age at seizure onseta (year) Genderb Women Men Age at surgerya (year) Epilepsy durationa (year) Preoperative seizure frequencya (frequency/month) History of GTCSb Preoperative aurasb Seizure lateralizationb Right Left Handednessb Right Left Ambidextrous Family history of epilepsyb History of febrile convulsionsb Negative Positive Nonresponders History of depressionb Interictal spike datab Bilateral equally obvious interictal spikes With more than 75% of ipsilateral interictal spikes No obvious interictal spikes Ictal EEGb Type 1 pattern Type 2 pattern Type 3 pattern

PET+/MRI− (n = 28)

MTS (n = 87)

P-value

16.0 ± 7.6

9.0 ± 3.4

b0.001⁎ 0.886

13 (46.4) 15 (53.6) 26.0 ± 7.5 10.1 ± 3.3 7.0 ± 7.8

37 (42.5) 50 (57.5) 28.4 ± 5.4 19.3 ± 4.7 8.0 ± 5.1

14 (50.0) 22 (78.6)

53 (60.9) 61 (70.1)

17 (60.7) 11 (39.3)

43 (49.4) 44 (50.6)

24 (85.7) 4 (14.3) 0 (0.0) 8 (28.6)

80 (92.0) 6 (6.9) 1 (1.1) 31 (35.6)

16 (57.1) 9 (32.1) 3 (10.7) 10 (35.7)

38 (43.7) 45 (51.7) 4 (4.6) 35 (40.2)

1 (3.6)

2 (2.3)

25 (89.3)

77 (88.5)

2 (7.1)

8 (9.2)

25 (89.3) 3 (10.7) 0 (0.0)

75 (86.2) 9 (10.3) 3 (3.4)

0.137 b0.001⁎ 0.533

3. Results

0.424 0.531 0.411

3.1. Baseline characteristics The 115 patients included in the analysis were divided into two groups: PET+/MRI − group (n = 28) and group with MTS (n = 87). The baseline characteristics of the two groups are summarized in Table 1. The PET +/MRI − group had a higher age at seizure onset (16.0 ± 7.6 years) and shorter epilepsy duration (10.1 ± 3.3 years) than the group with MTS (age at seizure onset: 9.0 ± 3.4 years, P b 0.001; epilepsy duration: 19.3 ± 4.7 years, P b 0.001), while all other characteristics showed no significant difference between groups.

0.438

0.648 0.146

3.2. Engel epilepsy surgery outcome scale

0.839 1.000

The mean postsurgical follow-up duration was 7.0 ± 0.9 years for the PET +/MRI − group and 6.5 ± 0.8 years for the group with MTS. Class I surgical outcomes at 2 and 5 years were 21 (75.0%) and 20 (71.4%) for the PET+/MRI− group and 66 (75.9%) and 64 (73.6%) for the group with MTS, with no significant difference detected between the two groups at either time period (Table 2).

1.000

PET, positron emission tomography; MRI, magnetic resonance imaging; MTS, mesial temporal sclerosis; EEG, electroencephalography; GTCS, generalized tonic–clonic seizure. a Continuous data were presented as mean ± standard deviation and compared between different groups by Student's t-test. b Categorical variables were expressed by counts (percentages) and compared between different groups by the chi-square test or Fisher's exact test. ⁎ Indicates a significant difference between the two groups.

frequency, with rare CPS; Class III, 50%–90% reduction in seizure frequency; and Class IV, less than 50% reduction in seizure frequency. 2.9. Statistical analysis Continuous variables regarding symptoms both before and after surgery were presented as mean and standard deviation, which were compared between two study groups or patients with right TLE and those with left TLE by Student's t-test. Categorical variables were expressed by counts and percentages and were compared between two study groups by the chi-square test or Fisher's exact test. Analysis of covariance (ANCOVA) was used to adjust for the imbalanced neuropsychological testing performed prior to surgery in two study groups or in patients

3.3. Comparison of neuropsychological testing between patients with right TLE and those with left TLE Neuropsychological testing revealed significantly different scores between patients with right TLE and those with left TLE prior to surgery. Patients with right TLE had higher VC, COWAT, SFT, CNT, and AVLT scores and lower TMT score than patients with left TLE (all P ≤ 0.001). Because of imbalanced neuropsychological testing before surgery, an ANCOVA was used to compare changes in test scores following surgery. The tests revealed significantly higher changes in VC, COWAT, SFT, CNT, and AVLT scores for those with right TLE compared with those with left TLE (all P b 0.001). Specifically, in patients with right TLE, the VC and SFT scores decreased following surgery, but the TMT and AVLT scores improved (all P ≤ 0.007). In patients with left TLE, all neuropsychological tests showed improvement following surgery (all P b 0.001; Table 3). 3.4. Comparison of neuropsychological testing between PET+ and MRI− patients In the neuropsychological testing, only VC scores were significantly different between the PET +/MRI − group and group with MTS prior

Table 2 Seizure reoccurrence as measured by the Engel epilepsy surgery outcome scale. Engel Class

2 years

5 years

PET+/MRI− (n = 28) Counts (%)

MTS (n = 87) Counts (%)

21 (75.0) 3 (10.7) 3 (10.7) 1 (3.6)

66 (75.9) 10 (11.5) 9 (10.3) 2 (2.3)

P-valuea

PET+/MRI− (n = 28) Counts (%)

MTS (n = 87) Counts (%)

20 (71.4) 4 (14.3) 3 (10.7) 1 (3.6)

64 (73.6) 11 (12.6) 10 (11.5) 2 (2.3)

1.000 Class I Class II Class III Class IV

PET, positron emission tomography; MRI, magnetic resonance imaging; MTS, mesial temporal sclerosis. a Compared between groups by Fisher's exact test.

P-valuea

0.944

P.-F. Yang et al. / Epilepsy & Behavior 41 (2014) 91–97 Table 3 Comparison of neuropsychological testing between patients with right TLE and those with left TLE. Test

VC Before surgery After surgery Changea COWAT Before surgery After surgery Changea TMT Before surgery After surgery Changea SFT Before surgery After surgery Changea CNT Before surgery After surgery Changea AVLT Before surgery After surgery Changea

Right TLE (n = 60) Mean ± SD

Left TLE (n = 55) Mean ± SD

P-value

94.7 ± 5.7 96.7 ± 5.8† 6.02 (4.43, 7.62)

83.6 ± 4.0 81.3 ± 4.0† Reference

b0.001⁎ b0.001⁎ b0.001⁎

28.8 ± 1.4 28.6 ± 1.6 1.45 (0.83, 2.06)

26.6 ± 2.0 25.7 ± 2.0† Reference

b0.001⁎ b0.001⁎ b0.001⁎

80.8 ± 5.0 79.9 ± 4.7† 0.74 (−0.01, 1.49)

83.5 ± 3.9 81.5 ± 3.8† Reference

0.001⁎ 0.053 0.052

40.4 ± 2.2 43.4 ± 2.4† 6.71 (4.69, 8.74)

30.3 ± 1.5 28.6 ± 2.4† Reference

b0.001⁎ b0.001⁎ b0.001⁎

51.5 ± 2.3 50.9 ± 2.8 10.28 (8.18, 12.39)

43.3 ± 2.0 36.7 ± 2.8† Reference

b0.001⁎ b0.001⁎ b0.001⁎

12.5 ± 2.2 11.9 ± 1.9† 5.52 (4.5, 6.54)

5.3 ± 1.5 3.1 ± 1.2† Reference

b0.001⁎ b0.001⁎ b0.001⁎

TLE, temporal lobe epilepsy; VC, Verbal Comprehension Test; COWAT, Controlled Oral Word Association Test; TMT, Trail Making Test (Trail B); SFT, Semantic Fluency Test (animals, fruits, vegetables); CNT, Chinese naming test; AVLT, Auditory Verbal Learning Test. a Difference in test scores from before to after surgery compared between groups by ANCOVA, the mean change and 95% confidence intervals (CIs) were calculated. ⁎ P b 0.05 indicates a significant difference between patients with right TLE and those with left TLE. † P b 0.05 indicates a significant difference between before and after surgery.

to surgery (92.6 ± 7.7 vs. 88.3 ± 7.0, P = 0.007). Using an ANCOVA, we observed significant differences between the PET +/MRI − group and the group with MTS in the changes following surgery of VC, COWAT, and TMT scores. The PET +/MRI − group had a lower VC score but higher COWAT and TMT scores than the group with MTS (all P ≤ 0.002). In the PET+/MRI− group, the AVLT score decreased following surgery (P b 0.001). In the group with MTS, the COWAT, TMT, CNT, and AVLT scores improved following surgery (all P b 0.001; Table 4). 3.5. Histopathology In all cases, samples from the temporal lobe and hippocampus were investigated, although in some cases, the cornu ammonis region was not fully available for investigation. Hippocampal formation neuronal loss together with gliosis within the CA1 region was diagnosed in 85 of 87 patients with MTS and 5 of 28 PET +/MRI − patients. In addition, prominent gliosis without neuronal loss was observed in 2 patients with MTS and 8 PET +/MRI − patients. In the remainder of the PET+/MRI − group, 4 patients were shown to have cortical dysplasia (1 Taylor type IA architectural, 2 type IB cytoarchitectural, and 1 type 2 with balloon cells), while 12 subjects did not display any pathologic alteration in the examined surgical specimens. 4. Discussion Because MRI-identified structural abnormalities have been documented to be among the most reliable predictors of a favorable outcome for ATL [2,4,35,36], it is controversial as to whether patients with normal MRI findings should ever be considered for an ATL [13]. Our results demonstrate that ATL can be an effective treatment for temporal lobe

95

Table 4 Neuropsychological testing between patients with PET+/MRI− and patients showing MTS with MRI. Test

VC Before surgery After surgery Changea COWAT Before surgery After surgery Changea TMT Before surgery After surgery Changea SFT Before surgery After surgery Changea CNT Before surgery After surgery Changea AVLT Before surgery After surgery Changea

PET+/MRI− (n = 28) Mean ± SD

MTS (n = 87) Mean ± SD

92.6 ± 7.7 91.3 ± 10.9 −2.52 (−4.00, −1.03)

88.3 ± 7.0 88.7 ± 8.5 Reference

0.007⁎ 0.198 0.001⁎

27.7 ± 1.4 27.8 ± 2.2 0.99 (0.36, 1.62)

27.8 ± 2.2 27.0 ± 2.3† Reference

0.673 0.090 0.002⁎

82.8 ± 2.6 82.5 ± 2.5 1.62 (0.82, 2.41)

81.8 ± 5.2 80.1 ± 4.6† Reference

0.214 0.001⁎ b0.001⁎

35.8 ± 5.7 36.6 ± 7.0 −0.04 (−1.00, 0.92)

35.5 ± 5.3 36.2 ± 8.0 Reference

0.781 0.808 0.935

48.7 ± 4.8 46.1 ± 9.0 0.53 (−1.03, 2.09)

47.2 ± 4.6 43.4 ± 7.1† Reference

0.147 0.111 0.504

9.9 ± 4.5 8.1 ± 5.8† −0.64 (−1.44, 0.16)

8.8 ± 3.9 7.5 ± 4.3† Reference

0.207 0.643 0.114

P-value

PET, positron emission tomography; MRI, magnetic resonance imaging; MTS, mesial temporal sclerosis; TLE, temporal lobe epilepsy; VC, Verbal Comprehension Test; COWAT, Controlled Oral Word Association Test; TMT, Trail Making Test (Trail B); SFT, Semantic Fluency Test (animals, fruits, vegetables); CNT, Chinese naming test; AVLT, Auditory Verbal Learning Test. a Difference in test scores from before to after surgery compared between groups by ANCOVA, the mean change and 95% confidence intervals (CIs) were calculated. ⁎ P b 0.05 indicates a significant difference between patients with right TLE and those with left TLE. † P b 0.05 indicates a significant difference between before and after surgery.

epilepsy in patients with nonlesional MRI, provided they are carefully selected. In our cohort, the PET+/MRI− patient group showed excellent postsurgical outcomes, with 71% categorized as Class I at 5 years. In particular, the 5-year surgical outcomes for these patients were comparable with those of patients with MRI findings revealing MTS. These findings corroborate those found in previous studies [9,19,49] and provide further evidence that PET+/MRI− patients can be very good surgical candidates. It is common in some epilepsy centers for intracranial implantation to be performed when patients are PET+/MRI− and evidence of structural abnormality is lacking [12]. For example, one case–control study matched 30 PET+/MRI− patients who underwent standard or hippocampal-sparing resections with 30 age- and gender-matched patients showing MTS on MRI and concluded that surgical outcome was equivalent between groups [9]. At our center, the PET+/MRI− patients who underwent resection directly without undergoing surgical implantation still had excellent outcomes. These results suggest that in many patients who have TLE with PET and EEG that is concordant and whose cases are otherwise uncomplicated, and the MRI findings are negative, it is not necessary to carry out surgical implantation, which is both invasive and expensive [9,20]. This is a particularly important advantage in developing countries such as China. Because of economic reasons, we do as few invasive examinations as possible, and we developed strict EEG inclusion criteria. Unlike in the study by LoPinto-Khoury et al. [20] conducted in the US, in our study, the cases with significant epileptic discharges on both sides were excluded, and therefore, surgery had good efficacy. Our findings, similar to the findings of LoPinto-Khoury et al. [20], suggest that a PET study should be considered for patients with TLE who are MRI−. Although a meta-analysis by Tellez-Zenteno et al. [37]

96

P.-F. Yang et al. / Epilepsy & Behavior 41 (2014) 91–97

demonstrated a significantly lower likelihood of seizure-free outcome for patients with MRI− epilepsy compared with those with MRI+ epilepsy, there still might be subgroups of MRI− patients with a more favorable surgical prognosis. LoPinto-Khoury et al. [20] proposed that their “data suggest that PET+/MRI− patients without discordant findings should have much higher rates of seizure freedom, likely equivalent to the lesional group.” We believe that our data also suggest this. Immonen et al. [12] found that in a series of MRI− patients, of the predictive factors for seizure outcome, noncongruent PET was the strongest for Engel Class III or IV outcomes. As pointed out by LoPinto-Khoury et al. [19], this suggests that PET scans may be of value for ruling out patients as candidates for surgery. Furthermore, we found that 18FFDG-PET was especially valuable when MRI findings were normal or when ictal EEG and MRI findings were not concordant, which is in line with indications for 18F-FDG-PET in the literature [9,24]. Our 18F-FDGPET results are also similar to those of previous reports that described hypometabolism ipsilateral to the EEG foci in 30–90% of patients with MRI-negative TLE [8,9,38–40]. Interestingly, a recent series [39] showed that both F-trans-4-fluoro-N-2-[4-(2-methoxyphenyl) piperazin-1-yl] ethyl-N-(2-pyridyl) cyclohexane carboxamide (18F-FCWAY-PET) and 18 F-FDG-PET may be even more helpful for locating epileptogenic zones in nonlesional TLE, as 18F-FDG-PET cannot distinguish mesial from lateral temporal foci, whereas 18F-FCWAY-PET showed less lateral temporal binding reduction in patients with mesial than lateral foci and may be more accurate than 18F-FDG-PET. The pre- and postoperative neuropsychological tests confirm previous findings that patients with right temporal lobe epilepsy and those with left temporal lobe epilepsy are neuropsychologically different, with patients with left temporal lobe epilepsy tending to have lower verbal abilities compared with those with right temporal lobe epilepsy at presentation, and which decline after surgery [41]. In patients whose mesial structures were resected, rates of memory decline are comparable with previously reported rates of verbal memory declines in dominant lesional TLE [42,43], whereas naming declines were higher in the PET +/MRI − group (61.5%, in 8 out of 13 with dominant TLE) compared with the group with MTS (37.8%, in 17 out of 45 with dominant TLE), suggesting higher risks for decline in confrontation naming among patients with nonlesional TLE. However, this will require further, well-designed, and controlled prospective research with a larger sample size and lesional comparison group. The underlying etiology of PET+/MRI − TLE continues to be enigmatic. The only significant clinical differences we found between these patients and those with MTS are that the PET+/MRI− patients developed epilepsy at a later age, had shorter disease durations at the time of surgery, and showed a trend toward more cognitive auras. Although our results suggest that these patients are largely surgically remediable, their distinction as a homogeneous clinical entity is not supported by our findings. This supports recent reports of heterogeneous causes of PET +/MRI − epilepsy such as small temporal pole encephaloceles [44] or Taylor-type focal cortical dysplasias [45,46]. Furthermore, Tassi et al. [47] reported that 32% of patients with type IA and 50% of patients with type IB histopathologically verified temporal lobe cortical dysplasias had normal MRIs, something seen in 4 of our patients. This is consistent with previous findings that neuronal loss and synaptic reorganization involving the hippocampal formation that is too trivial to be identified on MRI, which was seen in five of our patients, form a favorable post-ATL outcome group, suggesting that currently unrecognized cellular and subcellular alterations, not evident on routine pathological examinations, might be responsible for the remainder [48]. One might speculate that subtle lesions not appreciated by the pathologist could explain the surgical success seen in the cohort. If so, in the future, better molecular imaging, neuroimaging techniques, and volumetric analysis of the mesial temporal lobes could potentially distinguish patients with a higher chance of favorable surgical outcome. Our study had several limitations. One limitation was that the Wada test was not performed prior to surgery to assess language and memory

functions. We based the determination on the experience of left handedness to ensure that in most patients the left hemisphere was the dominant hemisphere as is the case with most Chinese. The surgical excision of the left neocortex is smaller, and the Spencer type of resection was used to retain the superior temporal gyrus in order to keep the language areas intact. Another limitation was that we used a 1.5-T MRI scanner during the study period as the use of a 3.0-T MRI scanner did not begin until after 2007. Because using a 1.5-T MRI scanner cannot show the internal structure of the hippocampus, there may be false-negative diagnoses in patients with MTS. An additional limitation is that this study had inadequate statistical power. Based on the results in Table 3, a total of 6555 patients per group would be needed to detect any difference between two groups using Fisher's exact test with a power of 0.8. It is not feasible to conduct this type of study with such a large number of patients. 5. Conclusions This study demonstrates that using noninvasive screening to eliminate those with discordant interictal or ictal EEGs and nonconcordant PET significantly improves the seizure-free success rate of patients presenting with PET +/MRI − scan results to be comparable with those presenting with MTS. Notably, this was done without using invasive intracranial electrodes for monitoring. Unfortunately, this limits the level of postsurgical monitoring, a potential limitation of this study. Our results give weight to the still controversial ability of ATL to treat those with MRI − TLE by suggesting better screening methods to identify those who would benefit from ATL and by providing a relatively large sample size. Acknowledgments None. Conflict of interest The authors declare that there are no conflicts of interest. References [1] Wiebe S, Blume WT, Girvin JP, Eliasziw M. Effectiveness, Efficiency of Surgery for Temporal Lobe Epilepsy Study G. A randomized, controlled trial of surgery for temporal-lobe epilepsy. N Engl J Med 2001;345:311–8. [2] Berkovic SF, McIntosh AM, Kalnins RM, Jackson GD, Fabinyi GC, Brazenor GA, et al. Preoperative MRI predicts outcome of temporal lobectomy: an actuarial analysis. Neurology 1995;45:1358–63. [3] 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. [4] Radhakrishnan K, So EL, Silbert PL, Jack Jr CR, Cascino GD, Sharbrough FW, et al. Predictors of outcome of anterior temporal lobectomy for intractable epilepsy: a multivariate study. Neurology 1998;51:465–71. [5] Tellez-Zenteno JF, Dhar R, Wiebe S. Long-term seizure outcomes following epilepsy surgery: a systematic review and meta-analysis. Brain 2005;128:1188–98. [6] 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. [7] 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. [8] Bien CG, Szinay M, Wagner J, Clusmann H, Becker AJ, Urbach H. Characteristics and surgical outcomes of patients with refractory magnetic resonance imagingnegative epilepsies. Arch Neurol 2009;66:1491–9. [9] Carne RP, O'Brien TJ, Kilpatrick CJ, MacGregor LR, Hicks RJ, Murphy MA, et al. MRInegative PET-positive temporal lobe epilepsy: a distinct surgically remediable syndrome. Brain 2004;127:2276–85. [10] Chapman K, Wyllie E, Najm I, Ruggieri P, Bingaman W, Lüders J, et al. Seizure outcome after epilepsy surgery in patients with normal preoperative MRI. J Neurol Neurosurg Psychiatry 2005;76:710–3. [11] 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. [12] Immonen A, Jutila L, Muraja-Murro A, Mervaala E, Aikia M, Lamusuo S, et al. Longterm epilepsy surgery outcomes in patients with MRI-negative temporal lobe epilepsy. Epilepsia 2010;51:2260–9.

P.-F. Yang et al. / Epilepsy & Behavior 41 (2014) 91–97 [13] Scott CA, Fish DR, Smith SJ, Free SL, Stevens JM, Thompson PJ, et al. Presurgical evaluation of patients with epilepsy and normal MRI: role of scalp video-EEG telemetry. J Neurol Neurosurg Psychiatry 1999;66:69–71. [14] Siegel AM, Jobst BC, Thadani VM, Rhodes CH, Lewis PJ, Roberts DW, et al. Medically intractable, localization-related epilepsy with normal MRI: presurgical evaluation and surgical outcome in 43 patients. Epilepsia 2001;42:883–8. [15] 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. [16] Kumar A, Valentín A, Humayon D, Longbottom AL, Jimenez-Jimenez D, Mullatti N, et al. Preoperative estimation of seizure control after resective surgery for the treatment of epilepsy. Seizure 2013;22:818–26. [17] 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. [18] Delbeke D, Lawrence SK, Abou-Khalil BW, Blumenkopf B, Kessler RM. Postsurgical outcome of patients with uncontrolled complex partial seizures and temporal lobe hypometabolism on 18FDG-positron emission tomography. Invest Radiol 1996;31: 261–6. [19] Dupont S, Semah F, Clémenceau S, Adam C, Baulac M, Samson Y. Accurate prediction of postoperative outcome in mesial temporal lobe epilepsy: a study using positron emission tomography with 18fluorodeoxyglucose. Arch Neurol 2000;57:1331–6. [20] LoPinto-Khoury C, Sperling MR, Skidmore C, Nei M, Evans J, Sharan A, et al. Surgical outcome in PET-positive, MRI-negative patients with temporal lobe epilepsy. Epilepsia 2012;53:342–8. [21] Willmann O, Wennberg R, May T, Woermann FG, Pohlmann-Eden B. The contribution of 18F-FDG PET in preoperative epilepsy surgery evaluation for patients with temporal lobe epilepsy. A meta-analysis. Seizure 2007;16:509–20. [22] Valentín A, Alarcón G, Barrington SF, García Seoane JJ, Martín-Miguel MC, Selway RP, et al. Interictal estimation of intracranial seizure onset in temporal lobe epilepsy. Clin Neurophysiol 2014;125:231–8. [23] Foldvary N, Lee N, Hanson MW, Coleman RE, Hulette CM, Friedman AH, et al. Correlation of hippocampal neuronal density and FDG-PET in mesial temporal lobe epilepsy. Epilepsia 1999;40:26–9. [24] O'Brien TJ, Newton MR, Cook MJ, Berlangieri SU, Kilpatrick C, Morris K, et al. Hippocampal atrophy is not a major determinant of regional hypometabolism in temporal lobe epilepsy. Epilepsia 1997;38:74–80. [25] Chassoux F, Semah F, Bouilleret V, Landre E, Devaux B, Turak B, et al. Metabolic changes and electro-clinical patterns in mesio-temporal lobe epilepsy: a correlative study. Brain 2004;127:164–74. [26] American Electroencephalographic Society guidelines for standard electrode position nomenclature. J Clin Neurophysiol 1991;8:200–2. [27] Marks Jr WJ, Laxer KD. Semiology of temporal lobe seizures: value in lateralizing the seizure focus. Epilepsia 1998;39:721–6. [28] Lüders H, Acharya J, Baumgartner C, Benbadis S, Bleasel A, Burgess R, et al. Semiological seizure classification. Epilepsia 1998;39:1006–13. [29] Kotagal P. Automotor seizures. In: Lüders HO, Noachtar S, editors. Epileptic seizures: pathophysiology and clinical semiology. Philadelphia: Churchill Livingstone; 2000. p. 449–57. [30] Ebersole JS, Pacia SV. Localization of temporal lobe foci by ictal EEG patterns. Epilepsia 1996;37:386–99.

97

[31] Uijl SG, Leijten FS, Arends JB, Parra J, van Huffelen AC, Moons KG. The added value of [18F]-fluoro-D-deoxyglucose positron emission tomography in screening for temporal lobe epilepsy surgery. Epilepsia 2007;48:2121–9. [32] Henry TR, Mazziotta JC, Engel Jr J. Interictal metabolic anatomy of mesial temporal lobe epilepsy. Arch Neurol 1993;50:582–9. [33] Theodore WH, Fishbein D, Dietz M, Baldwin P. Complex partial seizures: cerebellar metabolism. Epilepsia 1987;28:319–23. [34] Engel JJ, Van Ness PC, Rasmussen TB, Ojemann LM. Outcome with respect to epileptic seizures. In: Engel J, editor. Surgical treatment of the epilepsies. New York: Raven Press; 1993. p. 609–21. [35] McIntosh AM, Wilson SJ, Berkovic SF. Seizure outcome after temporal lobectomy: current research practice and findings. Epilepsia 2001;42:1288–307. [36] Thadani VM, Williamson PD, Berger R, Spencer SS, Spencer DD, Novelly RA, et al. Successful epilepsy surgery without intracranial EEG recording: criteria for patient selection. Epilepsia 1995;36:7–15. [37] Tellez-Zenteno JF, Hernandez Ronquillo L, Moien-Afshari F, Wiebe S. Surgical outcomes in lesional and non-lesional epilepsy: a systematic review and metaanalysis. Epilepsy Res 2010;89:310–8. [38] Carne RP, O'Brien TJ, Kilpatrick CJ, Macgregor LR, Litewka L, Hicks RJ, et al. ‘MRInegative PET-positive’ temporal lobe epilepsy (TLE) and mesial TLE differ with quantitative MRI and PET: a case control study. BMC Neurol 2007;7:16. [39] Liew CJ, Lim YM, Bonwetsch R, Shamim S, Sato S, Reeves-Tyer P, et al. 18F-FCWAY and 18F-FDG PET in MRI-negative temporal lobe epilepsy. Epilepsia 2009;50:234–9. [40] Lamusuo S, Jutila L, Ylinen A, Kälviäinen R, Mervaala E, Haaparanta M, et al. [18F] FDG-PET reveals temporal hypometabolism in patients with temporal lobe epilepsy even when quantitative MRI and histopathological analysis show only mild hippocampal damage. Arch Neurol 2001;58:933–9. [41] Rausch R, Kraemer S, Pietras CJ, Le M, Vickrey BG, Passaro EA. Early and late cognitive changes following temporal lobe surgery for epilepsy. Neurology 2003;60:951–9. [42] Fong JS, Jehi L, Najm I, Prayson RA, Busch R, Bingaman W. Seizure outcome and its predictors after temporal lobe epilepsy surgery in patients with normal MRI. Epilepsia 2011;52:1393–401. [43] Lineweaver TT, Morris HH, Naugle RI, Najm IM, Diehl B, Bingaman W. Evaluating the contributions of state-of-the-art assessment techniques to predicting memory outcome after unilateral anterior temporal lobectomy. Epilepsia 2006;47:1895–903. [44] Abou-Hamden A, Lau M, Fabinyi G, Berkovic SF, Jackson GD, Mitchell LA, et al. Small temporal pole encephaloceles: a treatable cause of “lesion negative” temporal lobe epilepsy. Epilepsia 2010;51:2199–202. [45] Chassoux F, Rodrigo S, Semah F, Beuvon F, Landre E, Devaux B, et al. FDG-PET improves surgical outcome in negative MRI Taylor-type focal cortical dysplasias. Neurology 2010;75:2168–75. [46] Srikijvilaikul T, Najm IM, Hovinga CA, Prayson RA, Gonzalez-Martinez J, Bingaman WE. Seizure outcome after temporal lobectomy in temporal lobe cortical dysplasia. Epilepsia 2003;44:1420–4. [47] Tassi L, Colombo N, Garbelli R, Francione S, Lo Russo G, Mai R, et al. Focal cortical dysplasia: neuropathological subtypes, EEG, neuroimaging and surgical outcome. Brain 2002;125:1719–32. [48] Engel Jr J. Introduction to temporal lobe epilepsy. Epilepsy Res 1996;26:141–50. [49] Struck AF, Hall LT, Floberg JM, Perlman SB, Dulli DA. Surgical decision making in temporal lobe epilepsy: a comparison of [(18)F]FDG-PET, MRI, and EEG. Epilepsy Behav 2011;22:293–7.