FULL-LENGTH ORIGINAL RESEARCH

The significance of parahippocampal high gamma activity for memory preservation in surgical treatment of atypical temporal lobe epilepsy Naoto Kunii, 1Kensuke Kawai, 2Kyousuke Kamada, 3Takahiro Ota, and Nobuhito Saito Epilepsia, 55(10):1594–1601, 2014 doi: 10.1111/epi.12764

SUMMARY

Naoto Kunii is a Neurosurgeon at The University of Tokyo Hospital, Tokyo, Japan.

Objective: Resective surgery for mesial temporal lobe epilepsy (MTLE) with a correspondent lesion has been established as an effective and safe procedure. Surgery for temporal lobe epilepsies with bilateral hippocampal sclerosis or without correspondent lesions, however, carries a higher risk of devastating memory decline, underscoring the importance of establishing the memory-dominant side preoperatively and adopting the most appropriate procedure. In this study, we focused on high gamma activities (HGAs) in the parahippocampal gyri and investigated the relationship between memory-related HGAs and memory outcomes after hippocampal transection (HT), a hippocampal counterpart to neocortical multiple subpial transection. The transient nature of memory worsening after HT provided us with a rare opportunity to compare HGAs and clinical outcomes without risking permanent memory disorders. Methods: We recorded electrocorticography from parahippocampal gyri of 18 patients with temporal lobe epilepsy while they executed picture naming and recognition tasks. Memory-related HGA was quantified by calculating differences in power amplification of electrocorticography signals in a high gamma range (60–120 Hz) between the two tasks. We compared memory-related HGAs from correctly recognized and rejected trials (hit-HGA and reject-HGA). Using hit-HGA, we determined HGA-dominant sides and compared them with memory outcomes after HT performed on seven patients. Results: We observed memory-related HGA mainly between 500 and 600 msec poststimulus. Hit-HGA was significantly higher than reject-HGA. Three patients who had surgery on the HGA-dominant side experienced transient memory worsening postoperatively. The postoperative memory functions of the other four patients remained unchanged. Significance: Parahippocampal HGA was indicated to reflect different memory processes and be compatible with the outcomes of HT, suggesting that HGA could provide predictive information on whether the mesial temporal lobe can be resected without causing memory worsening. This preliminary study suggests a refined surgical strategy for atypical MTLE based on reliable memory lateralization. KEY WORDS: Electrocorticography, High gamma activity, Hippocampus, Human, Intracranial.

Accepted July 22, 2014; Early View publication September 2, 2014. Department of Neurosurgery, The University of Tokyo, Tokyo, Japan 1 Present address: Department of Neurosurgery, NTT Medical Center Tokyo, Tokyo, Japan. 2 Present address: Department of Neurosurgery, Asahikawa Medical University, Hokkaido, Japan. 3 Present address: Department of Neurosurgery, Tokyo Metropolitan Tama Medical Center, Tokyo,Japan. Address correspondence to Kensuke Kawai, Department of Neurosurgery, NTT Medical Center Tokyo, 5-9-22 Higashi-Gotanda, Shinagawa-ku, Tokyo 141-8625, Japan. E-mail: [email protected] Wiley Periodicals, Inc. © 2014 International League Against Epilepsy

1594

1595 Memory Related High Gamma Activity Removal of the mesial temporal lobe (MTL) achieves seizure freedom in up to 70% of patients with intractable MTL epilepsy (MTLE),1 but carries the potential risk of damaging verbal episodic memory functions, particularly in cases with bilateral hippocampal sclerosis or without correspondent lesions, that is, atypical MTLE.2–4 To avoid postoperative memory decline, it is important to evaluate memory functions in the bilateral MTLs preoperatively, a protocol for which has yet to be established. The Wada test, the gold standard for testing language dominance, is also used to indicate the memory-dominant side. The results of the Wada test for memory function, however, have sometimes been incompatible with the clinical outcomes.5,6 Functional magnetic resonance imaging (fMRI) is a promising imaging modality for evaluating memory functions noninvasively. In recent studies that compared the reliability of fMRI with postoperative memory outcomes,7,8 fMRI alone appeared to be unable to predict memory outcome with acceptable accuracy. Recently, various kinds of brain oscillatory activities have gained attention as key measures to elucidate brain functions. Among them, power in the high gamma range, that is, high gamma activity (HGA), has been shown to be strongly correlated with important brain functions, including memory.9,10 Hippocampal gamma oscillations are reported to increase with memory load,11 which suggests they can be used as indices for memory functions. To date, the usefulness of HGA as a memory index has not been investigated because there is no established way of confirming that a given side is responsible for verbal episodic memory other than actually removing the MTL. Unfortunately, resection of the MTL carries a high risk of damaging episodic memory, particularly in surgeries on the language-dominant side without typical hippocampal sclerosis.12 This dilemma has prevented researchers from designing a study that can demonstrate the usefulness of any particular memory indices. We adopt hippocampal transection (HT) in cases with atypical MTLE of the language-dominant side, particularly in patients whose memory functions are maintained. HT is a novel memory-preserving procedure developed as a hippocampal counterpart of neocortical multiple subpial transections (MSTs).13–15 Similar to transient cortical dysfunction after MSTs,16 it is reported that memory functions decline transiently after HT before recovering to the preoperative level within 6 months.13,17 This transient nature of memory worsening after HT on the memory dominant side provided us with a unique opportunity to investigate, with little selection bias, whether the HGA distribution is compatible with the actual memory lateralization. In atypical MTLE cases, we have used subdural electrodes covering bilateral parahippocampal gyri. We utilized these parahippocampal electrodes in this study. In the first part, we elaborated a memory task paradigm that induces

HGA in the parahippocampal gyri, reflecting different memory processes. In the second part, we investigated the relationship between the laterality of memory-related HGA and transient memory worsening after HT.

Methods Subjects This study included patients with intractable epilepsy who underwent implantation of subdural electrodes for diagnostic purposes. We selected patients whose bilateral parahippocampal gyri were covered by subdural electrodes, since they were MRI-negative or phase I studies indicated epileptic foci on both sides. Patients with a low intelligence quotient (IQ; 0.3 or LI < 0.3; otherwise, memory lateralization was undetermined. We compared the results of memory evaluation by HGA with memory decline after HT with Fisher’s exact test (p < 0.05) and the kappa statistic.23 Hippocampal transection (HT) This procedure is developed based on the theory that transverse transections on the hippocampus disrupt longitudinal seizure propagation while preserving memory functions that are dependent on transverse connections. In a recent study, favorable seizure outcome and preservation of verbal memory with HT were reported, supporting the theory. Using a transsylvian approach, the interior horn of the lateral ventricle was opened. The hippocampus and parahippocampal gyrus were transected perpendicularly to the long axis of the hippocampus with a 4 mm interval, preserving the fimbria and lateral connection.13,14 We employed this procedure for language dominant–sided surgery of patients without obvious hippocampal sclerosis on MRI and with memory scores near normal to avoid severe postoperative memory decline. Despite the memory-preserving character of this procedure, a transient memory decline immediately after surgery seems inevitable when performed on the memory-dominant side.14,17 The transient memory worsening, therefore, indicated that the operated side was memory dominant.

1597 Memory Related High Gamma Activity

Figure 1. Task paradigm. The memory task paradigm used in this study included two sessions, that is, the picture naming task and picture recognition task. The pictures used in both sessions consisted of 10 target and 32 lure pictures of familiar objects, each of which was displayed two or three times, yielding 30 target and 70 lure trials. Each picture was presented for 1,000 msec, with an interstimulus interval ranging between 2,700 and 3,300 msec. In the picture naming task, subjects were shown pictures (targets and lures) in random order and instructed to name them silently. The 10 target pictures were then presented again and the patient was given 3 min to remember them (the encoding period). Fifteen minutes after the encoding period, subjects were shown the same pictures presented in the picture naming task in a different order (the picture recognition task). They were instructed to respond regarding whether the picture had been presented in the encoding period (targets) or not (lures) by pushing a button. Based on the answers, trials in the picture recognition task were labeled as correctly hit, correctly rejected, and others. Epilepsia ILAE

Evaluation of memory function The revised Miyake Paired-Associate Word Learning (MPAL) test and Wechsler Memory Scale-Revised (WMSR) were administered for memory evaluation. Patients performed MPAL and WMS-R preoperatively, only MPAL 2–4 weeks after surgery, and again MPAL and WMS-R 6 months to 1 year after surgery. Because three sets of MPAL batteries were available, we assessed memory soon after surgery only by MPAL so that we could rule out any learning effect.24 MPAL is a widely used verbal episodic memory examination in Japan for its convenience.25 In MPAL, patients learned 10 word pairs with a semantic relation and 10 word pairs without a semantic relation, and we counted the numbers of recalled pairs of nouns after three learning trials. The results were expressed as the percentage of correct answers for 10 unrelated pairs in the third trial. We compared the results between, before, and soon after surgery. In this study, memory decline was defined if there was decrease in MPAL scores by ≥30% after surgery.

Results Behavioral data were available for 16 of 18 patients who performed the memory tasks. The rate of correct responses, that is, the sum of correctly hit and correctly rejected responses, was 94.6  5.0% (mean  standard deviation [SD]). ECoG data were collected from 144 electrodes placed on the parahippocampal gyri of the 18 patients. A grand average of the memory-related spectrograms was obtained from the data of the 18 patients (Fig. 2A).

There was a prominent HGA between 500 and 600 msec poststimulus. Based on this finding, we focused the following quantitative analyses on the ECoG between 300 and 800 msec poststimulus to include this critical period. We calculated hit-PSCs and reject-PSCs using data for all electrodes and compared them across frequencies with the Wilcoxon signed rank test (Fig. 2B). The spectral estimates of hit-PSC were significantly higher than those of rejectPSC in the high gamma range. We performed HT on seven patients, for whom preoperative and postoperative evaluations of memory functions were available. An illustrative case is shown in Figure 3. Preoperative memory lateralization by HGA and postoperative memory outcomes are shown in Table 2. HGA was dominant on the left side in four patients, and only one patient showed right-dominant HGA. Two patients showed no lateralization of HGA, which was caused by low values of maximum hit-HGA on either side. Of the two patients, one patient underwent HT on the left side and the other on the right side, with significant memory decline in neither case. Whereas three patients had MST on the HGA-dominant side and experienced significant memory worsening (MPAL change >30%) in the early stage of the postoperative period, memory scores of the other four patients were maintained (p = 0.0286, kappa = 1). Focusing on the three patients who experienced early postoperative memory decline, we summarized the time coarse of their memory scores (Table 3). In all three patients, both MPAL and WMS-R recovered to the preoperative level at 6 month after surgery. Epilepsia, 55(10):1594–1601, 2014 doi: 10.1111/epi.12764

1598 N. Kunii et al. A

B

Figure 2. Gross inspection of memory-related spectrography and power spectral changes averaged among 18 patients. (A) Grand average of memory-related spectrography. Each time-frequency spectral power plot was standardized to baseline activity (600 to 100 msec), with time on the x-axis (600 to 1,500 msec, with stimulus presentation at 0) and frequency on the y-axis (0–160 Hz). The grand average was obtained by using recordings from 144 electrodes from bilateral parahippocampal gyri of 18 patients. Power increases in the high gamma range (60–120 Hz) were observed at around 500–600 msec. The solid black line in the plot represents the onset of task presentation. (B) Mean power spectral changes in the hit and reject trials. The mean power spectral changes were obtained by using recorded data from 144 electrodes from bilateral parahippocampal gyri of 18 patients. Red and blue lines indicate the grand averages of spectral density functions of hit- (red) and reject- (blue) trials, respectively. The shaded area around the solid line indicates standard errors of the mean values. Black bars above the lines indicate significant differences in spectral density functions between the two conditions (p < 0.05; corrected for multiple comparisons by the number of frequency bins). Spectral density estimates of hit-trials were significantly higher than those of reject-trials. The light gray zone indicates the high gamma frequency range (60–120 Hz). Epilepsia ILAE

Discussion We analyzed memory-related parahippocampal HGAs of 18 patients with intractable MTLE. Of interest, correctly hit trials induced significantly higher HGAs than correctly rejected trials, which suggested a strong correlation between successful memory recognition and parahippocampal HGAs. Of seven patients who underwent HT, only three who showed dominant memory-related HGA on the operative side had postoperative memory decline. The other four patients experienced no memory worsening. The significance of parahippocampal HGA In healthy subjects and patients with late-onset epilepsy, language and verbal memory processes are strongly left-lateralized.26 This assumption is not applicable to cases with long-standing epilepsy. Hippocampal sclerosis usually indicates decreased memory function on the same side, making the surgical strategy straightforward. On the other hand, preoperative evaluations for atypical MTLE yield few clues regarding memory lateralization. In some cases, the epileptic focus and memory function may colocalize on the same side. In other cases, memory function may have already moved to the opposite side due to functional reorganization. Therefore, most cases with atypical MTLE of the language dominant side are excluded as candidates for hippocampal resection to avoid postoperative memory decline. This strategy may have overlooked a considerable number of resectable cases with left MTLE (i.e., Fig. 3). The most Epilepsia, 55(10):1594–1601, 2014 doi: 10.1111/epi.12764

significant contributions of this study are the clarification of such cases with dissociated lateralization of memory and language functions and the finding that the memory-related HGA colocalized with the clinically determined memorydominant side. Although the technique developed in this study is applicable to only cases that involve implantation of subdural electrodes bilaterally on the parahippocampal gyri, we consider such invasiveness does not reduce the significance of this study, since it mostly contributes to cases with atypical MTLE, which primarily require bilateral intracranial electrode implantation to localize seizure origin. It should be noted that HGAs calculated in this study derived from recordings of the parahippocampal gyri, leaving a concern that it might dissociate from hippocampal activities. We should be careful not to simply apply surgical strategy based on this study in all MTLE cases. Memory and hippocampal transection (HT) The memory status of all three patients who experienced transient memory worsening recovered to the preoperative level at 6 months after HT. This postoperative course agreed with those previously reported.14,17 This study premised that if HT is performed on the memory-dominant side, transient memory worsening is unavoidable. Considering that HT is usually performed in patients with near-normal memory functions, this assumption seems quite reasonable. In fact, the worsening of memory appeared to occur in an all-or-none manner; while all cases with transient memory worsening showed a >50% decrease in MPAL

1599 Memory Related High Gamma Activity

Figure 3. Illustrative case. A 35-year-old woman (case 5) with mild atrophy in the left hippocampus underwent bilateral electrode placement on the parahippocampal gyri, because less invasive examinations failed to determine the epileptic focus. Long-term ECoG monitoring captured habitual seizures originating from the left hippocampus. Her verbal memory function was intact, according to a preoperative Revised Miyake Paired-Associate Word Learning (MPAL) test. Both Wada language and memory tests indicated dominant language and memory functions on the left side. Spectral analyses of the ECoG data obtained during the memory task (hit-power spectral changes; see Methods) revealed marked high gamma activity in the right parahippocampal gyrus. We performed hippocampal transection on the left hippocampus to avoid permanent damage to her verbal memory function, even though transient memory decline seemed inevitable. Postoperative MPAL, however, showed her verbal memory function was intact even immediately after surgery. In this case, memory outcome agreed with parahippocampal high gamma activity, but not with the Wada test. According to the high gamma activity, we could have performed hippocampal resection without postoperative memory decline. Epilepsia ILAE

Table 2. Memory lateralization by HGA and postoperative memory outcomes Max hit-HGA No.

Dominant hand

Language lateralitya

1 2 3 4 5 6 7

R R L R L R R

L L Bil L L L L

L

R

2.20 0.98 0.95 0.32 0.18 0.79 0.34

1.07 0.00 0.39 0.22 0.67 0.39 0.56

HGA-LI (indicated side)

Operative side

34.6 (L) 100 (L) 41.6 (L) 18.3 (N) 57.6 (R) 33.4 (L) 24.6 (N)

L L R L L L R

MPAL change (%) 50 60 10 10 0 80 20

Max hit-HGA, the maximum value of significant high gamma activity of hit trials in four electrodes; HGA-LI, laterality index of high gamma activity; MPAL, Revised Miyake Paired-Associate Word Learning Test; N, not determined. a Determined by Wada test.

scores, other cases experienced little memory worsening even soon after the surgery. Paradoxically, although these three patients were at high risk for postoperative memory decline, seizure outcomes of the three patients were more favorable. We confirmed that HT was performed in a

formulated manner among all the patients, ensuring minimum confounding biases in the treatment strategy. Two cases showed no obvious laterality of memoryrelated HGA. Two different interpretations are possible. First, we assumed their memory functions were distributed Epilepsia, 55(10):1594–1601, 2014 doi: 10.1111/epi.12764

1600 N. Kunii et al. Table 3. Changes in memory functions of patients with transient memory decline after surgery

No. 1 2 6

Pre-op MPAL (%)

Early post-op MPAL (%)

Late post-op MPAL (%)

70 100 90

20 40 10

100 100 80

Pre-op WMS-R

Late post-op WMS-R

Verbal

Visual

Verbal

Visual

91 93 94

121 113 113

76 96 95

105 115 111

MPAL, Revised Miyake Paired-Associate Word Learning Test; WMS-R, Wechsler Memory Scale-Revised.

bilaterally or partially moved to other brain areas as a result of functional reorganization. Many studies have suggested that reorganization of memory function can occur if the MTL is damaged extensively over a prolonged period. Such reorganization involves transference of processing to the contralateral MTL,27 the ipsilateral neocortical regions,28 or even the posterior hippocampus on the resection side.7 The durations of epilepsy in the two patients were 7 and 17 years, respectively, which were long enough to support this possibility. Second, we might have failed to detect dominant HGAs due to technical problems such as an electrode positions or a task design. Although the results of this present study might suggest less memory worsening after HT in patients who show no obvious laterality in memory-related HGAs, we should be careful in the interpretation of these cases. This study included only two cases that underwent HT on the HGA-nondominant side, which is too small number to ensure postoperative memory preservation. To support this point, we attached Table S1, presenting clinical data of four resection cases that were excluded from the present analyses. The right and left MTL were resected in two cases, respectively, and language function was left-dominant in all cases. In each case, resection was done in the HGA-nondominant side, resulting in no postoperative memory decline. Considering this supplementary data, it seems safe to resect HGA-nondominant sides. In addition, this study lacks cases with memory decline after right HT, raising a concern that HGA on the right side has less clinical impact. This is a possible limitation of this study that focused on HT, a memory-preserving procedure, which tends to be performed on the language-dominant side. Relationship with limbic event–related potentials Since Halgren et al. first described limbic event–related potentials (ERPs) in humans, several authors have reported on these electrical phenomena.29–33 Limbic ERPs include the MTL P300 elicited by an oddball task, the anterior MTL N400, which is a response to “new” stimuli, and the late negative component (LNC), which responds to correctly classified “old” items in a recognition task.29 Although lowering of limbic ERP amplitude has been reported to indicate Epilepsia, 55(10):1594–1601, 2014 doi: 10.1111/epi.12764

epileptogenicity, its relationship with memory functions has rarely been investigated. We used a recognition task in which patients are required to detect “old” stimuli, instead of a novelty detection task. The parahippocampal HGA, or hit-HGA observed in our study, therefore, appears to correspond to LNC in the hippocampus. The peak latency of LNC is reported to be around 800 msec. The relatively late onset of the parahippocampal HGA seems to be compatible with the latency of LNC. Reject-HGA, seemingly related to recognition of “old” stimuli, is expected to be useful as another memory index, which is beyond the scope of this study. Future direction To achieve better operative outcomes for atypical MTLE including seizure control rate and postoperative memory status, it is essential to establish accurate preoperative evaluations of detailed memory lateralization and select the best surgical option. Although preliminary, this study suggested a possible strategy that the hippocampus on the language dominant side can be resected safely, if HGA indicates memory lateralization to the non–language-dominant side. Although the number of cases in this study was too small to provide enough evidence to support this strategy, it suggests the possibility for improving surgery for atypical MTLE and deserves further validation with a larger sample size. Likewise, there have been a small number of reports on HT from a limited number of institutions. Clinical outcomes after HT need to be accumulated with long-term follow-up, as was accomplished for resective surgery.1

Acknowledgments This work was supported in part by the Japan Epilepsy Research Foundation, a grant from the Suhara Memorial Foundation, Grant-in-Aid No. 21390405 and 24390337 for Scientific Research (B) from the Japan Society for the Promotion of Science, Grant-in-Aid No. 23659679 for Challenging Exploratory Research from the Ministry of Education, Culture, Sports, Science and Technology of Japan (MEXT), Grant-in-Aid No. 23119701 for Scientific Research on Innovative Areas, “Face perception and recognition” from MEXT, a Research Grant for “Decoding and controlling brain information” from the Japan Science and Technology Agency, and a Grant H23- Nervous and Muscular-General-003 for Comprehensive Research on Disability, Health and Welfare from the Ministry of Health, Labour and Welfare of Japan.

1601 Memory Related High Gamma Activity

Disclosure None of the authors has any conflict of interest to disclose. We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.

References 1. Wiebe S, Blume WT, Girvin JP, et al. A randomized, controlled trial of surgery for temporal-lobe epilepsy. N Engl J Med 2001;345:311– 318. 2. Binder JR. Preoperative prediction of verbal episodic memory outcome using FMRI. Neurosurg Clin N Am 2011;22:219–232, ix. 3. Engel J Jr, Wiebe S, French J, et al. Practice parameter: temporal lobe and localized neocortical resections for epilepsy: report of the Quality Standards Subcommittee of the American Academy of Neurology, in association with the American Epilepsy Society and the American Association of Neurological Surgeons. Neurology 2003;60:538–547. 4. Hermann BP, Wyler AR, Somes G, et al. Declarative memory following anterior temporal lobectomy in humans. Behav Neurosci 1994;108:3–10. 5. Dodrill CB, Ojemann GA. An exploratory comparison of three methods of memory assessment with the intracarotid amobarbital procedure. Brain Cogn 1997;33:210–223. 6. Loring DW, Lee GP, Meador KJ, et al. The intracarotid amobarbital procedure as a predictor of memory failure following unilateral temporal lobectomy. Neurology 1990;40:605–610. 7. Bonelli SB, Powell RH, Yogarajah M, et al. Imaging memory in temporal lobe epilepsy: predicting the effects of temporal lobe resection. Brain 2010;133:1186–1199. 8. Binder JR, Swanson SJ, Sabsevitz DS, et al. A comparison of two fMRI methods for predicting verbal memory decline after left temporal lobectomy: language lateralization versus hippocampal activation asymmetry. Epilepsia 2010;51:618–626. 9. Sederberg PB, Schulze-Bonhage A, Madsen JR, et al. Hippocampal and neocortical gamma oscillations predict memory formation in humans. Cereb Cortex 2007;17:1190–1196. 10. Axmacher N, Mormann F, Fernandez G, et al. Sustained neural activity patterns during working memory in the human medial temporal lobe. J Neurosci 2007;27:7807–7816. 11. van Vugt MK, Schulze-Bonhage A, Litt B, et al. Hippocampal gamma oscillations increase with memory load. J Neurosci 2010;30:2694– 2699. 12. Helmstaedter C, Petzold I, Bien CG. The cognitive consequence of resecting nonlesional tissues in epilepsy surgery–results from MRIand histopathology-negative patients with temporal lobe epilepsy. Epilepsia 2011;52:1402–1408. 13. Uda T, Morino M, Ito H, et al. Transsylvian hippocampal transection for mesial temporal lobe epilepsy: surgical indications, procedure, and postoperative seizure and memory outcomes. J Neurosurg 2013;119:1098–1104. 14. Shimizu H, Kawai K, Sunaga S, et al. Hippocampal transection for treatment of left temporal lobe epilepsy with preservation of verbal memory. J Clin Neurosci 2006;13:322–328. 15. Morrell F, Whisler WW, Bleck TP. Multiple subpial transection: a new approach to the surgical treatment of focal epilepsy. J Neurosurg 1989;70:231–239. 16. Blount JP, Langburt W, Otsubo H, et al. Multiple subpial transections in the treatment of pediatric epilepsy. J Neurosurg 2004;100:118–124.

17. Kawai K, Kamada K, Ohta T, et al. Multiple hippocampal transection: seizure outcome and postoperative neuropsychometry. Epilepsia 2006;47 (Suppl. 4):12–13. 18. Kamada K, Sawamura Y, Takeuchi F, et al. Expressive and receptive language areas determined by a non-invasive reliable method using functional magnetic resonance imaging and magnetoencephalography. Neurosurgery 2007;60:296–305; discussion 305–296. 19. Koizumi S, Kawai K, Asano S, et al. Familial lateral temporal lobe epilepsy confirmed with intracranial electroencephalography and successfully treated by surgery. Neurol Med Chir (Tokyo) 2011;51:604–610. 20. Shimizu H, Suzuki I, Ohta Y, et al. Mesial temporal subdural electrode as a substitute for depth electrode. Surg Neurol 1992;38:186–191. 21. Kunii N, Kamada K, Ota T, et al. Characteristic profiles of high gamma activity and blood oxygenation level-dependent responses in various language areas. Neuroimage 2013;65:242–249. 22. Kunii N, Kamada K, Ota T, et al. The dynamics of language-related high-gamma activity assessed on a spatially-normalized brain. Clin Neurophysiol 2013;124:91–100. 23. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977;33:159–174. 24. Theisen ME, Rapport LJ, Axelrod BN, et al. Effects of practice in repeated administrations of the Wechsler Memory Scale Revised in normal adults. Assessment 1998;5:85–92. 25. Ishiai S, Koyama Y, Seki K, et al. Unilateral spatial neglect in AD: significance of line bisection performance. Neurology 2000;55:364– 370. 26. Binder JR, Sabsevitz DS, Swanson SJ, et al. Use of preoperative functional MRI to predict verbal memory decline after temporal lobe epilepsy surgery. Epilepsia 2008;49:1377–1394. 27. Golby AJ, Poldrack RA, Illes J, et al. Memory lateralization in medial temporal lobe epilepsy assessed by functional MRI. Epilepsia 2002;43:855–863. 28. Detre JA, Maccotta L, King D, et al. Functional MRI lateralization of memory in temporal lobe epilepsy. Neurology 1998;50:926–932. 29. Dietl T, Kurthen M, Kirch D, et al. Limbic event-related potentials to words and pictures in the presurgical evaluation of temporal lobe epilepsy. Epilepsy Res 2008;78:207–215. 30. Grunwald T, Lehnertz K, Pezer N, et al. Prediction of postoperative seizure control by hippocampal event-related potentials. Epilepsia 1999;40:303–306. 31. Puce A, Kalnins RM, Berkovic SF, et al. Limbic P3 potentials, seizure localization, and surgical pathology in temporal lobe epilepsy. Ann Neurol 1989;26:377–385. 32. Meador KJ, Loring DW, King DW, et al. Spectral power of human limbic evoked potentials: relationship to seizure onset. Ann Neurol 1988;23:145–151. 33. Halgren E, Squires NK, Wilson CL, et al. Endogenous potentials generated in the human hippocampal formation and amygdala by infrequent events. Science 1980;210:803–805.

Supporting Information Additional Supporting Information may be found in the online version of this article: Table S1. Clinical characteristics of patients who underwent resection of mesial temporal lobe.

Epilepsia, 55(10):1594–1601, 2014 doi: 10.1111/epi.12764

The significance of parahippocampal high gamma activity for memory preservation in surgical treatment of atypical temporal lobe epilepsy.

Resective surgery for mesial temporal lobe epilepsy (MTLE) with a correspondent lesion has been established as an effective and safe procedure. Surger...
472KB Sizes 2 Downloads 5 Views