Porencephaly

To evaluate the usefulness and limitations of magneto-encephalography (MEG) for epilepsy surgery, we compared 'interictal' epileptic spike fields on MEG with ictal electrocorticography (ECoG) using invasive chronic subdural electrodes in a patient with intractable medial temporal lobe epilepsy (MTLE) associated with vitamin K deficiency intracerebral hemorrhage. A 19-year-old male with an 8-year history of refractory complex partial seizures, secondarily generalized, and right hemispheric atrophy and porencephaly in the right frontal lobe on MRI, was studied with MEG to define the interictal paroxysmal sources based on the single-dipole model. This was followed by invasive ECoG monitoring to delineate the epileptogenic zone. MEG demonstrated two paroxysmal foci, one each on the right lateral temporal and frontal lobes. Ictal ECoG recordings revealed an ictal onset zone on the right medial temporal lobe, which was different from that defined by MEG. Anterior temporal lobectomy with hippocampectomy was performed and the patient has been seizure free for two years. Our results indicate that interictal MEG does not always define the epileptogenic zone in patients with MTLE. [Neurol Res 2001; 23: 830-834]

Keywords: Magnetoencephalography; temporal lobe epilepsy; epilepsy surgery; epileptogenic zone

INTRODUCTION

The primary objective in a pre-operative evaluation for epilepsy surgery is the accurate localization of the epileptogenic area. While invasive electrophysiological monitoring such as subdural and depth recording of ictal activities remains the `gold standard' for seizure localization, a variety of modalities including electrophysiological monitoring, anatomical imaging, and physiological imaging techniques have also been shown to provide useful ancillary information. Among them, magnetoencephalography (MEG) is a particularly promising new tool as a noninvasive method of observing epileptiform activity1-3. Because brain magnetic activities do not undergo modifications and distortions such as voltage attenuation and smearing effects from volume conductors including skull and cerebrospinal fluid, MEG can provide accurate information on the spatial distribution and temporal patterns of epileptiform activity4,5. Magnetic source imaging (MSI) combines the measurement of MEG with such anatomical data as comes from magnetic resonance imaging (MRI) to portray the spatial distribution of sources of measured magnetic fields relative to anatomy.

MSI studies in epilepsy are routinely conducted concurrent with standard scalp EEG, using protocol designed to identify and map locations of 'interictal' epileptic activities onto anatomic images1,6-11. When evaluating data available from this technique, one must keep in mind that it is interictal in nature and therefore prone to all the problems associated with such interictal data11-13.

We investigated 'interictal' epileptic activity with a 37-channel biomagnetometer and mapped the data onto a three dimensional (3-D) image in a patient with medial temporal lobe epilepsy (MTLE) associated with vitamin K deficiency intracranial hemorrhage in the ipsilateral frontal lobe. The findings did not correlate with the results of invasive electrophysiological monitoring obtained from chronically placed subdural electrodes. The limitations of MSI as a pre-operative evaluation method for epilepsy surgery are thus discussed.

PATIENT PROFILE

A 19-year-old right-handed male had been identified as having vitamin K deficiency intracranial hemorrhage in the right frontal lobe at the age of one month. He had febrile convulsions at the ages of 6, 12 and 18 months. He developed onset of complex partial seizure (CPS) at age 11 years. The seizures began with an aura of a peculiar sensation and were followed by unconsciousness, lip smacking and hand automatism. The seizures lasted usually 1 min or longer with secondary generalization. His seizures occurred several times a week, and remained refractory to maximally tolerated combination therapy with anti-epileptic drugs such as phenobarbital, phenytoin, zonisamide, carbamazepine, valpronic acid and nitrazepam.

On admission, mild mental retardation and left homonymous hemianopsia were noted. MRI revealed a small sized calvarium and hemisphere on the right side, and a porencephaly in the medial part of the right frontal lobe (Figure IA). MRI with fluid-attenuated inversion recovery (FLAIR) sequence showed high signal intensity of the right atrophic hippocampus and occipital lobe (Figure 18). Interictal EEG showed paroxysmal activities in right frontal (Fp2-F4) and temporal regions (F8-T4). Ictal EEG demonstrated rhythmic slow waves predominantly in the bilateral fronto-temporal region, but failed to reveal the ictal onset zone. Interictal [^sup 18^F]fluoro-2-deoxy-D-glucose positron emission tomography (FDG-PET) and single photon emission computed tomography with ethyl-cysteinate dimer (ECDSPECT) showed decreased metabolic activity and perfusion over the entire right hemisphere. Wada testing showed left-hemispheric dominancy for language and memory.

MEG METHODS

Our MEG techniques have been described previously12,14-19. Spontaneous magnetic activities were recorded with a 37-channel large-array biomagnetometer (Magnes, 4-D Neuro Imaging Inc., San Diego, CA, USA). Simultaneous EEG recordings, which were made bilaterally with 21 silver disc electrodes, were done according to the International 10-20 system using ear referential derivations (Synafit 1000, NEC Medical Systems, Tokyo, Japan).

Two different types of stereotypic interictal paroxysms were recorded in the right frontal (Figure 2A) and temporal (Figure 2B) regions, respectively, on both EEG and MEG. The MEG and EEG paroxysmal activities did not completely coincide in waveform with each other, and the MEG paroxysmal activities showed a shorter duration (Figure 2). The difference between MEG and EEG manifestations was probably the result of MEG being performed with 37 channels placed over the right frontal and temporal regions while EEG was performed with a standard 10-20 electrodes placed over the entire scalp.

Source localization of the equivalent current dipole (ECD) was performed by least squares search20 on both MEG paroxysmal activities. A spherical model was fitted to the digitized head shape of the patient. The best fit single ECD was thus estimated for each time. The correlation between the theoretical field generated by the estimated ECD and the observed field was used to evaluate the estimation and thus to determine whether or not a good fit was achieved. Only the dipole sources that were estimated with correlation coefficients greater than 98% were analyzed. The estimated ECD sources of both MEG paroxysms were overlaid on the reconstructed 3-dimensional MRI which were made using Cemax VIP Station software (Cemax Inc., Fremont, CA, USA) (Figure 3). The estimated ECDs of the interictal MEG paroxysmal activities were located on the right lateral frontal lobe near the porencephalic cyst (Figure 3A) and on the right lateral temporal lobe (Figure 38).

INVASIVE ELECTROPHYSIOLOGICAL MONITORING

With the patient under general anesthesia, a right frontotemporo-parietal craniotomy was performed. Both frontal and temporal lobes were atrophic and arachnoid membrane was cloudy. There was a porencephalic cyst in the interhemispheric surface of the frontal lobe. The cyst wall was smooth and yellowish.

A trapezoid shape subdural electrode with eight contacts21 was placed on the medial and basal temporal lobe. Three matrices of subdural grid electrodes consisting of 15 or 20 contacts were placed over the lateral surface of the fronto-temporal lobes and the base and pole of the frontal lobe. In addition, a strip electrode consisting of five contacts was utilized to cover the interhemispheric surface of the frontal lobe.

Through these chronically placed subdural grid and strip electrodes, long-term electrocorticographic (ECoG) monitoring with a digital EEG recorder (EEG-DAE-2100, Nihon, Kohden, Tokyo, Japan) was performed. Interictal ECoG revealed multiple independent paroxysmal activities in the lateral aspect of the frontal and temporal lobes and in the medial part of the temporal lobe. Three stereotypic CPSs with secondary generalization were recorded. The seizure discharges began in the medial temporal lobe (Figure 4 arrow), were attenuated, and then spread to the entire temporal lobe and frontal base (Figure 4).

SURGERY AND POST-OPERATIVE COURSE

Following functional mapping by electrical cortical stimulation testing, the patient underwent a repeat craniotomy under general anesthesia. Anterior temporal lobectomy with hippocampectomy was performed. Histologically, pyramidal cells of the CAl hippocampus were absent and gliosis was noted. Gliary fibrillary acidic protein positive astrocytes were also prominent in the lateral temporal lobe.

Post-operatively, the patient did not develop any further neurological deterioration. In the two years following the procedure, he has been free from seizures.

DISCUSSION

Extensive experience with MEG has shown that MEG provides an accurate localization of the cortex generating interictal spikes1,6-11,22-25. Most of these studies have employed indirect comparisons with intra-operative and/or interictal ECoG recordings6-11,16-19,22-26. On the other hand, paroxysm generated in the deep medial temporal structures may escape MEG detection23,24,27, probably as a result of the quick decay of magnetic fields from the deeper sources28,29. In the present study, we could not find ECD localization in the deep medial temporal lobe and interhemispheric surface of the frontal lobe. ECD localizaitons were estimated on the lateral aspects of the fronto-temporal lobes.

It has been generally accepted that the zones of the cortex that generate interictal spikes (irritative zones) are not always identical to the ictal onset zones and/or the epileptogenic zones. Irritative zones may be more extensive than epileptogenic zones, and furthermore, some epileptogenic zones may extend beyond the irritative zones or even exist with an absence of any detectable irritative zone30. It would therefore be ideal for a pre-operative evaluation to detect the ictal onset zone, which could then result in a good surgical outcome.

There are few published reports in which interictal MEG data was confirmed with ictal ECoG recordings2,3,12. Otsubo et al.3 reported a case of frontocentral epilepsy, in which two MEG epileptic foci were confirmed by invasive ECoG monitoring. Minassian et al.2 also demonstrated that, in 10 of 11 patients with extra-temporal lobe epilepsy, the anatomical location of the epileptiform discharges as determined by MEG corresponded to the ictal onset zone established by the ictal ECoG. In these particular cases, irritative zones were identical to the epileptogenic zones of the extratemporal neocortex. In the present study, the dipole localization of the two interictal paroxysms (MEG irritative zones) were in lateral aspects of the temporal and frontal lobes and the MEG irritative zones did not coincide with the ictal onset zone as detected using invasive ECoG monitoring.

One of the major differences between the MEG and ECoG findings is the fact that MSI depicts the epileptogenic focus as a point, while ECoG depicts it as a zone. Since the current MSI method analyzes the source localization using a single dipole model, the irritative zone was depicted as a point. Thus, most of the authors2,3,12 defined the MEG irritative zone as the cluster of ECD generated in the interictal paroxysm.

Ictal MEG studies are difficult to obtain, because prolonged recordings are not practical22,31. Because movement artifacts interfere with MEG recording and analysis of dipole localization, the patient must be absolutely still. In addition, as the seizure begins to spread and the volume of neural tissue involved increases, the dipole rapidly changes and the chance for accurate localization of the point of initiation is lost12. Stefan et al.11 recorded ictal MEG activities in three patients with partial epilepsy by using a multichannel magnetometer. In our previous report12, we recorded subclinical ictal MEG activities. In these reports, the dipole localizations were calculated with a good correlation at the onset of seizure activities. During seizures, however, the dipole localization of the propagation activities could not be calculated with a single dipole model. In this case, we could not obtain ictal MEG. Even though we had ictal MEG, ictal onset magnetic fields generated in the deep medial temporal structures could not be recorded.

In summary, our results indicate that interictal MEG cannot always define the epileptogenic zone in patients with MTLE.

ACKNOWLEDGEMENTS

We thank Kazumi Takahashi and Yuko Somehara, ELEKTA K.K. and Yoshie Hirosawa for their technical assistance.

REFERENCES

1 Aung M, Sobel DF, Gallen CC, Hirschkoff EC. Potential contribution of bilateral magnetic source imaging to the evaluation of epilepsy surgery candidates. Neurosurgery 1995; 37: 1113-1121

2 Minassian BA, Otsubo H, Weiss S, Elliott I, Rutka JT, Snead OC Ill. Magnetoencephalographic localization in pediatric epilepsy surgery: Comparison with invasive intracranial electroencephalography. Ann Neurol 1999; 46: 627-633

3 Otsubo H, Sharma R, Elliott I, Holowka S, Rutka JT, Snead OC III. Confirmation of two magnetoencephalographic epileptic foci by invasive monitoring from subdural electrodes in an adolescent with right frontocentral epilepsy. Epilepsia 1999; 40: 608-613

4 Barth DS, Sutherling WW, Engel J Jr, Beatty J. Neuromagnetic localization of epileptiform spike activity in the human brain. Science 1982; 218: 891-894

5 Barth DS, Sutherling WW, Engel J Jr, Beatty J. Neuromagnetic evidence of spatially distributed sources underlying epileptiform spikes in human brain. Science 1984; 223: 293-296

6 Eisenberg HM, Papanicolaou AC, Baumann SB, Rogers RL, Brown LM. Magnetoencephalographic localization of interictal spike sources. Case report. J Neurosurg 1991; 74: 660-664

7 Paetau R, Kajola M, Karhu J, Nousiainen U, Partanen J, Tiihonen J, Valalahti M, Hari R. Magnetoencephalographic localization of epileptic cortex: Impact on surgical treatment. Ann Neurol 1992; 32: 106-109

8 Paetau R, Hamalainen M, Hari R, Kajola M, Karhu J, Larsen TA, Lindahl E, Salonen 0. Magnetoencephalographic evaluation of children and adolescents with intractable epilepsy. Epilepsia 1994; 35:275-284

9 Paetau R, Granstrom M-L, Blomstedt G, Jousmaki V, Korkman M, Liukkonen E. Magnetoencephalography in presurgical evaluation of children with the Landau-Kleffner syndrome. Epilepsia 1999; 40:326-335

10 Stefan H, Schneider S, Abraham-Fuchs K, Bauer J, Feistel H, Pawlik G, Neubauer U, Rohrlein G, Huk WJ. Magnetic source localization in focal epilepsy. Multichannel magnetoencephalography correlated with magnetic resonance brain imaging. Brain 1990; 113: 1347-1359

11 Stefan H, Schneider S, Feistel H, Pawlik G, SchUler P, AbrahamFuchs K, Schlegel T, Neubauer U, Huk WJ. Ictal and interictal activity in partial epilepsy recorded with multichannel magnetoencephalography: Correlation of electroencephalography/electrocorticography, magnetic resonance imaging, single photon emission computed tomography, and positron emission tomography findings. Epilepsia 1992; 33: 874-887

12 Ishibashi H, Morioka T, Shigeto H, Nishio S, Yamamoto T, Fukui M. Three-dimensional localization of subclinical ictal activity by magnetoencephalography: Correlation with invasive monitoring; Surg Neurol 1998; 50: 157-163

13 Suthering WW, Crandall PH, Engel J Jr, Darcey TM, Cahan LD, Barth DS. The magnetic field of complex partial seizures agrees with intracranial localizations. Ann Neurol 1987; 21: 548-558

14 Hisada K, Morioka T, Nishio S, Muraishi M, Yamamoto T, Yoshida T, Fukui M. Magnetoencephalographic analysis of periodic lateralized epileptiform discharges (PLEDS). Clin Neurophysiol 2000; 111:122-127

15 Ishibashi H, Morioka T, Shigeto H, Yamamoto T, Nishio S, Fukui M. Magnetoencephalographic evaluation of anterior corpus

callostomy for intractable epilepsy in a patient with LennoxGastaut syndrome. J Epilepsy 1998; 11: 202-207

16 Morioka T, Yamamoto T, Nishio S, Takeshita I, Imamoto N, Fukui M. Magnetoencephalographic features in neurocysticercosis. Surg Neurol 1996; 45: 176-182

17 Morioka T, Nishio S, Ishibashi H, Muraishi M, Hisada K, Shigeto H, Yamamoto T, Fukui M. Intrinsic epileptogenicity of focal cortical dysplasia as revealed by magnetoencephalography and electrocorticography. Epilepsy Res 1999; 33: 177-187

18 Morioka T, Nishio S, Shigeto H, Goto Y, Yamamoto T, Minami T, Gondo K, Fujii K, Fukui M. Surgical management of intractable epilepsy associated with cerebral neurocytoma. Neurol Res 2000; 22:449-456

19 Shigeto H, Tobimatsu S, Morioka T, Yamamoto T, Kobayashi T, Kato M.Jerk-locked back averaging and dipole source localization of magnetoencephalographic transients in a patient with epilepsia partial is continua. Electroenceph Clin Neurophysiol 1977; 103: 440-444

20 Sarvas J. Basic mathematical and electromagnetic concepts of the biomagnetic inverse problem. Phys Med Biol 1987; 37: 11-22

21 Shimizu H, Suzuki I, Ohta Y, Ishijima B. Mesial temporal subdural electrode as a substitute for depth electrode. Surg Neurol 1992; 38: 186-191

22 Knowlton RC, Laxer KD, Aminoff MK, Roberts TPL, Wong STC, Rowley HA. Magnetoencephalography in partial epilepsy: Clinical yield and localization accuracy. Ann Neurol 1997; 42: 622-631

23 Smith JR, Schwartz BJ, Gallen C, Orrison W, Lewine J, Murro AM, King DW, Park YD. Utilization of multichannel magnetoencephalography in the guidance of ablative seizure surgery. J Epilepsy 1995; 8: 119-130

24 Sutherling WW, Crandall PH, Cahan LD, Barth DS. The magnetic field of epileptic spikes agrees with intracranial localizations in complex partial epilepsy. Neurology 1988; 38: 778-786

25 Sutherling WW, Levesque MF, Crandall PH, Barth DS. Localization of partial epilepsy using magnetic and electric measurements. Epilepsia 1991; 32 (Suppl. 5): S29-S40

26 Tiihonen J, Hari R, Kajola M, Nousiainen U, Vapalahti M. Localization of epileptic fosi using a large-area magnetometer and functional brain anatomy. Ann Neurol 1990; 27: 283-290

27 Brockhaus A, Lehnertz K, Wienbruch C, Kowalik A, Burr W, Elbert T, Hoke M, Elger CE. Possibilities and limitations of magnetic source imaging of methohexital-induced epileptiform patterns in temporal lobe epilepsy patients. Electroenceph Clin Neurophysiol 1997;102:423-436

28 Mikuni N, Nagamine T, Ikeda A, Terada K, Taki W, Kimura J, Kikuchi H, Shibasaki H. Simultaneous recording of epileptiform discharges by MEG and subdural electrodes in temporal lobe epilepsy. Neuroimage 1997; 5: 298-230

29 Shigeto H, Morioka T, Ishibashi H, Yamamoto T, Tobimatsu S, Kato M, Kobayashi T, Fukui M. Feasibility of magnetoencephalography in detecting epileptic discharges: Simultaneous recording of magnetic field and electrocorticography. Electroenceph Clin Neurophysiol 1997; 103 (Suppl.): 51 60

30 I-Ciders HO, Engel J Jr, Munari C. General principles. In: Engel J Jr, ed. Surgical Treatment of the Epilepsies, New York: Raven Press, 1993: pp. 137-153

31 Ebersole JS. Magnetoencephalography/magnetic source imaging in the assessment of patients with epilepsy. Epilepsia 1997; 38 (Suppl. 4): S1-S5

Kei Hisada, Takato Morioka, Shunji Nishio, Tomoya Yamamoto* and Masashi Fukui

Department of Neurosurgery, *Department of Otorhinolaryngology Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan

Correspondence and reprint requests to: Takato Morioka, MD, PhD, Department of Neurosurgery, Graduate School of Medical Sciences, Kyushu University, 3-1-1 Maidashi, Higashi-ku, Fukuoka 812-8582, Japan. [takato@ns.med.kyushu-u.ac.jp] Accepted for publication June 2001.

Copyright Forefront Publishing Group Dec 2001
Provided by ProQuest Information and Learning Company. All rights Reserved