Between September 1994 and August 2001, 38 patients

Transcription

1 KISEP J Korean Neurosurg Soc 36 : , 2004 Clinical Article Parietal Lobe Epilepsy : Surgical Treatment and Outcome Chi Heon Kim, M.D., 1 Chun - Kee Chung, M.D., 1 Sang Kun Lee, M.D., 2 Yoon Kyung Lee, M.D., 3 Je G. Chi, M.D. 3 Departments of Neurosurgery, 1 Neurology, 2 Pathology, 3 Seoul National University College of Medicine; Clinical Research Institute, Seoul National University Hospital; Neuroscience Research Institute, Medical Research Center, Seoul National University, Seoul, Korea Objective : Parietal lobe epilepsy(ple) is neither common nor easily diagnosed because of its variable clinical features. To elucidate its characteristics and surgical outcome, the authors review their surgical experiences. Methods : Between September 1994 and August 2001, 38 patients with parietal lobe epilepsy received surgical treatment at our Hospital. All patients underwent resection, mainly involving the parietal lobe. In most patients, preoperative evaluation included interictal and ictal electroencephalography, magnetic resonance image, positron emission tomography, and interictal and ictal single photon emission computed tomography. The male to female ratio was 23:15. Age at surgery ranged from 4 to 38 years (median, 25). Results : Preoperatively over 60% were considered not to be PLE, even though PLE was the most common (15/38, 39.8 %). The most common seizure type was complex partial seizure (26/38, 68.4%) rather than simple partial seizure (3/38, 7.9%). Invasive study was performed in 37 of the 38 patients. Monitoring duration was from 4 to 18 days (median, 8 days). Awake operations under regional anesthesia were performed in 20 patients (52.6%). Follow-up periods ranged from 14 to 81 months (mean, 50.7). Seizure disappeared in 15 (Engel s classification I, 39.5%), and rare seizure remained in 5 (Engel II, 13.2%). Thirteen patients showed a worthwhile improvement (Engel III, 34.2%) and 5 no worthwhile improvement (Engel IV, 13.2%). Pathologies were diverse, the most common being cortical dysplasia (94.3%). Conclusion : Since PLE is difficult to diagnose preoperatively, an invasive study covering the parietal lobe is mandatory, if PLE is suspected. Cortical dysplasia was the most common etiology, thus awake operation under regional anesthesia and intraoperative brain mapping is helpful during extensive resection in order to spare the eloquent cortex. Using this protocol, PLE can be controlled surgically with a satisfactory result and a reasonably low level of complications. KEY WORD : Parietal lobe epilepsy. Introduction Surgery for intractable epilepsy provides many patients not only with freedom or substantial relief from seizures, but also functional improvement and an increased quality of life 23). Parietal seizures account for 6% of all seizures, but the frequency of parietal lobe epilepsy(ple) may be underestimated because of difficulties in its diagnosis, especially in cases with no structural lesion 3,9,11,23,27). To elucidate the characteristics and surgical outcome of intractable epilepsy, the authors reviewed their surgical experiences. Materials and Methods Between September 1994 and August 2001, 38 patients with parietal lobe epilepsy received surgical treatment at Seoul National University Hospital. All patients underwent resection involving the parietal lobe. We defined parietal lobe epilepsy not as seizure occurred only in the parietal lobe but seizure occurred mainly in the parietal lobe. Clinical data and the imaging findings of 38 patients were reviewed. Information about aura and semiology were obtained by reviewing patients charts and by self-descriptions obtained from patients. We also reviewed data upon clinical features, which were prospectively maintained. Age at surgery ranged from 4 to 38 years (median, 25), and age at onset of seizure ranged from 3 months to 25 years (median, 10 years). Postoperative follow-up duration ranged from 14 to 81 months (mean, 50.7), and the male to female ratio was 23:15.

2 Parietal Lobe Epilepsy Preoperative diagnostic studies included ictal/interictal scalp-electroencephalography(eeg) monitoring, magnetic resonance(mr) imaging, photon emission tomography(pet), and ictal/interictal single photon emission computed tomography(spect), functional MR imaging and the Wada test. Seizure was classified according to the ILAE classification 5). Video ictal EEG monitoring was performed in all patients and this data was reviewed. Mean duration of monitoring was eight days (4 to 18 days). Pre- and postoperative MR images were reviewed. Preoperative images were available in 27 patients, and postoperative MR images in 23. If images were not available, the neuroradiologist s reading results were reviewed instead. According to these data, abnormalities in the ipsilateral parietal lobe occurred in 11/38 (28.9 %). Positron emission tomograph (PET) and ictal/interictal single photon emission computed tomography (SPECT) was also performed. Aura, semiology, MRI findings, and the result of PET and ictal/interictal SPECT were all analyzed to localize the seizure onset zone. Invasive study (subdural grid, strip insertion, and monitoring) covering the suspicious region was performed to all patients except one, and if the result was not diagnostic then a second invasive study was performed (12 of 37, 32.4%). A third invasive study was performed in one patient. Invasive study was not performed in one patient because of a discrete cystic lesion clearly visible by MRI and PET, showing decreased metabolism in this area. Mean duration of monitoring was 8 days (4 to 18). After mapping the seizure onset zone, extraoperative functional mapping via stimulation of the grid and strip was performed. Awake operation under regional anesthesia of the scalp and intraoperative brain mapping by electrical cortical stimulation was also performed in 20 patients, if the desired resection site was close to the motor or sensory cortex, or in the left parietal lobe. Resection was limited to the seizure onset zone and irritative zone, and the functional cortex as confirmed by intra/extraoperative brain mapping was saved. After resection of the seizure onset zone, anti-epileptic medication was continued and tapered-out according to seizure outcome during the follow up. We utilized 0.3-msec pulses of alternating polarity with pulses of each polarity separated by 20 msec with 50-pulseper-second trains of 5 to 7 seconds duration. The starting amplitude was 1.5mA. We increased intensity in steps of 1mA, until functional changes were observed, or maximum amplitude of 17.5mA (isolated biphasic stimulator S12; Grass instrument; Massachusetts; USA, 1995). The physicians watched for any evidence of spontaneous changes such as muscle twitching, and the patient was asked to report if any somatosensory, visual, psychic, auditory, or other changes occurred. To define the language area, we presented the figures of common objects for naming and sentences for reading; these were used previously in the Wada test. After administering fentanyl 50~100mcg intravenously, a scalp nerve block was done to supraorbital, supratrochlear, auriculotemporal, and to the lesser and greater occipital nerves on the surgical side with 0.25% bupivacaine containing 1:200,000 epinephrine. The incision site was infiltrated with the same local anesthetics. Oxygen was given by nasal cannula. During operation fentanyl was infused. Propofol was not infused until mapping was successfully performed. Invasive blood pressure, end-tidal CO2 and the respiratory rate were monitored throughout the operation and arterial blood gas analysis was performed intermittently. Before the operation, information was offered regarding the reasons for an awake craniotomy and what would be done during that portion of period. The operating table was well padded. Both arms were positioned to be readily visualized by testing physicians. A support holding surgical drapes was prepared for the patient to see the test material or the physicians. Stimulation was delivered via grid electrodes or a stimulating probe. The electrical stimulation and testing methods were the same as those used for the extraoperative stimulation. Outcome was subdivided as over and as less than a 90% seizure reduction. Factors that had influence on outcome were analyzed for using Pearson Chi-square and Fischer s exact test. The Mann-Whitney test was used for analysis the influence of age at operation and the age at onset. Results Seven patients had a history of head trauma, three patients had encephalitis, and two patients had meningitis. One patient had a history of birth trauma. Febrile convulsion was seen in 10 patients. A history of pre-term birth or febrile disease of an J Korean Neurosurg Soc 36

3 CH Kim, et al. unknown cause was obtained in one patient each. In 18 patients (47.4%), there was no remarkable preceding event. Thirty patients (78.9%) patients showed aura. Sensorimotor auras were seen in only 12 (31.6 %) patients. Among these, 2 showed bilaterally and one showed ipsilateral to the seizure onset zone. Thus, true somatosensory aura was seen in only 9 patients (23.7%). Motor aura was seen in only one patient (sense of weakness). A tingling sensation was the most common sensory aura followed by pain, a thermal sense, and numbness. Visual aura was seen in 5 patients : visual darkness in 2, flashing in 2, and diplopia in one. Oroalimentary aura (OAA) was seen in 4 and auditory hallucination in 2. A sense of fear, anxiety, a light headache, and a strange feeling were also noted. Some patients showed more than one aura. The most common seizure type was complex partial seizure (CPS : 26/38, 68.4%) rather than simple partial seizure (SPS : 3/38, 7.9%), and this was followed by generalized tonic clonic seizure (GTC : 5/38, 13.2%). More than one seizure type was showed by 4 patients (CPS and GTC, CPS and SPS, SPS and GTC). Secondary generalization was seen in 32 patients (84.2%). The most common presentation was tonic and/or clonic movement (32/38, 84.2%) contralateral to the seizure onset zone in 14, followed by bilateral in 17, and ipsilateral in one. Interestingly, automatism of oral, oroalimentary or of an extremity was seen in 18 patients, and behavioral arrest or a motionless staring in 12, suggesting the temporal lobe as a symptomatogenic zone. Other symptoms were head and/or eye version in 12, eye blinking in 4, laughing in one, a silly smile in one, head hobbing in one and fencing posture in one. Many patients showed more than two presentations; these findings implied a variable propagation pathway of the epileptic discharge. Data on interictal scalp (EEG) was available in all patients except one. Lateralization was possible in 25 patients, localization in 3, false lateralization in 3, normal in 3, and both hemispheres were lateralized in 3. Video-EEG results were available in all patients. Lateralization was possible in 31 patients, localization in 4, and neither lateralization nor localization in 3. Magnetic resonance imaging(mri) revealed an abnormality in twenty-six 26 patients (68.4%). However, abnormalities involving the parietal lobe only were observed in 14 patients. Of these, 3 patients showed bilateral parietal cerebromalacia. Only 11 cases showed an ipsilateral parietal lesion. Other abnormalities were; hippocampal atrophy (HA) in 4, cortical atrophy in 3, hippocampal atrophy and cortical atrophy in 2, cortical atrophy and biventricular signal change in one, cortical dysplasia and cortical atrophy in one, and a tumorous lesion in one. Of these 11 cases, 4 patients had a dual lesion and in 3 patients, hippocampal atrophy was accompanied (cortical atrophy+biventricular signal change, 1; bilateral cortical atrophy+ha, 1; cortical atrophy+ha, 2). One patient showed bilateral hippocampal atrophy. In positron emission tomography(pet) localization of the seizure onset zone was possible in 11 patients (28.9%) and lateralization in 15 patients (39.5%). False lateralization was seen in one patient and bilateral hypometabolism also in one. Ictal SPECT was performed in 28 patients. Localization was possible in 7 (25%), lateralization in 15 (53.6%), false lateralization in 2, bilateral hyperfusion in 3, and normal in one. Interictal SPECT was performed in 21 patients. Lateralization was possible in 7 (52.4%), localization in 8 (38.1%), no lateralization/localization was possible in one and another was falsely lateralized. With these data, preoperative diagnosis was variable. Initially, parietal lobe epilepsy (PLE) was diagnosed in 13 (39.8%). Of these, multilobar involvement was suspected in 2 (FLE+PLE, TLE+PLE). In nearly 60% of patients, other diagnoses (lateral TLE, 9; FLE, 6; OLE, 5; etc.) were made preoperatively. Surgical resection was performed in all patients, and included the parietal lobe. The majority of patients underwent invasive study (subdural grid and strip insertion, and and ictal EEG monitoring) (37 of 38, 97.4%), because in most patients, the exact location of the epileptogenic zone was unsure by EEG, MRI, PET, and ictal/interictal SPECT. Moreover, the seizure onset zone is close to the eloquent sensory motor and language areas, so it is mandatory to verify functional brain mapping via electrical stimulation to the inserted grid and strip to prevent or minimize morbidity and neurological deficit. One patient showed a discrete cystic lesion on the right superior parietal lobe, and PET localized this lesion, so invasive study was not performed. Invasive study was performed twice in 12 and three times in one. After extraoperative mapping of the seizure onset zone and functional brain mapping, an awake operation under general anesthesia was performed in 20 patients (52.6%). Motor and sensory, language, and right/left orientation were monitored intraoperatively. Resection included the whole extent of the visible lesion, if present, or the whole ictal onset zone and irritative zone, bounded by sulci or the functional cortex. Underlying white matter was resected if feasible. During and after resection, intraoperative mapping was performed intermittently to ensure that the functional cortex was preserved. The operation was performed twice in one patient due to recurred seizure (refer to the illustrative case

4 Parietal Lobe Epilepsy Table 1. The site of resection (n=38) Table 2. Pathologic classification (n=38) Table 3. Cortical dysplasia and outcome below). Right parietal resection was performed in 19 patients and left in 19. Sixteen patients received parietal resection only, and the remaining received resection of other lobes also. Accompanying resection sites were temporal (12), followed by frontal (8) and occipital (7). A double lesion was occurred in 8 patients (Table. 1). Complications occurred postoperatively in 20 patients (52.6%). The most common complication was visual field defect (10 of 38, 26.3%). If visual field defect was considered an unavoidable side effect rather than a complication, complications were found in 15 (39.5%). Sensory motor complications occurred in 7 patients and 3 patients showed a permanent motor complication. Left hand weakness (motor grade III/V) after right fronto-parieto-temporal resection was noted in one and permanent right hemiparesis (motor grade III/V) after left temporo-parietal resection in one. Unfortunately, although extraoperative functional mapping was performed, intraoperative functional mapping was not possible in these two children, because of poor cooperation. However, permanent slight right leg weakness (motor grade IV/V) occurred after left parietal resection by an awake operation in one child; she was able to walk independently, but complained of difficulty in running. Transient dysarthria developed in two patients after left fronto-parietal and left parietal resection, respectively. Transient Gerstman s syndrome was seen after left parietal resection under an awake operation in one. Postoperative epidural hemorrhage was detected in two patients and surgical infection in two. There was one case of delayed hydrocephalus development, which caused hemiparesis and was resolved with a shunt. There was no operative mortality, however, one patient expired 14 months after the operation due to status epilepticus. With antiepileptic medication, her seizures had improved (Engel III). However, she had not taken her meications for 2 months before the onset of status epilepticus. Pathology was available in 34 patients (Table 2). Cortical dysplasia was the most common pathologic finding (33 of 35, 94.3%). The pathological severity of cortical dysplasia was classified according to Mischel 15). Mild cortical dysplasia was present in 16, moderate in 4, and severe in 13. Among these 33, two cases showed mild cortical dysplasia around a tumorous lesion (pleomorphic xanthoastrocytoma and oligodendroglioma). One case showed oligodendroglioma without surrounding cortical dysplasia. Cortical dysplasia only occurred in 31 patients. Among these, only 17 patients (54.8%) showed an MRI abnormality. There was one case of chronic nonspecific inflammation, probably related to a previous head injury. Postoperative follow-up was possible in all patients and duration ranged from 14 to 81 months (mean, 50.7). The results of epilepsy surgery were grouped as seizure free, rare seizure, worthwhile improvement, and no worthwhile improvement, according to Engel s classification s I, II, III, and IV. At the last follow-up visit, seizures had disappeared in 15 patients (Engel s classification I, 39.5%) and seizures were rare in 5 (Engel s classification II, 13.5%). Thirteen patients showed worthwhile improvement (Engel s classification III, at least a 90% improvement in seizure frequency versus preoperative status; 34.2%) and 5 showed no worthwhile improvement (Engel s classification IV, 13.5%). Cortical dysplasia only was shown by 31 patients (Table. 3). In 3 patients with a tumor, one was Engel J Korean Neurosurg Soc 36

5 CH Kim, et al. Table 4. Factors versus outcome (P<0.05) class I and the others class II. Hippocampal atrophy was shown in 7 patients, and hippocampectomy was performed in 3; Engel class I in 2 and class III in one. Hippocampectomy was not performed in the other four; Engle class I in two and class III in two. Overall most patients (33/38, 86.8%) showed more than 90% seizure reduction. Outcome was not dependent upon age at onset or at the time of the operation. Cortical dysplasia and outcome were regrouped ; mild versus moderate and severe cortical dysplasia, Engel classes I and II versus III and IV. The value of this comparison was 0.052, showing strong trend for outcome. The influence of other factors was not analyzed statistically due to the small number of patients (Table. 4). All patients with a febrile convulsion history showed more than a 90% seizure reduction versus 23 of 28 (82%) of patients without such a history. PET localization was possible in 11 patients and all showed over a 90% seizure reduction, versus 12 of 15 (80%) patients with PET lateralization and 9 of 11 (81%) without PET localization/ lateralization. Using these data, a presumptive trend could be suggested. If PLE was suspected in a patient who had a febrile convulsion history and localization of the seizure onset zone was possible with PET, a good surgical outcome (over 90% seizure reduction) may be expected, especially if the resected tissue showed more than moderate cortical dysplasia. The operation was performed twice in one patient, a 27-yearold male patient admitted with a chief complaint of complex partial seizure for ten years, due to dual pathologies. There was no aura. His typical seizure was brief behavior arrest followed by hand automatism. Brain MR images showed right hippocampal sclerosis (Fig. 1A, B). Video EEG and interictal EEG suggested the right temporal lobe as the seizure onset zone. PET showed right fronto-temporal hypometabolism, and ictal/interictal SPECT revealed the right temporal area as the seizure onset zone. Under the preoperative diagnosis of right TLE, a right anterior temporal lobectomy with amygdalohippocampectomy was performed. Pathology showed hippocampal sclerosis. Postoperatively, his seizures disappeared for 1 year with medication. However, after one year a complex partial seizure occurred without the motor component. Brain MR imaging showed no specific finding except the previously resected right temporal lobe and hippocampus (Fig. 2A, B). Video-EEG suggested the right posterior temporal area (posterior to the previous resection margin) as the seizure onset zone. PET study showed right fronto-temporal hypometabolism and ictal SPECT right temporal hyperfusion. The first invasive study indicated that the right posterior temporal and parietal area as the seizure onset zone (Fig. 3A, B). This finding suggested secondary propagation rather than a primary seizure focus. Thus, repositioning of the grid and strip was needed. A secondary invasive study covering the posterior temporal, parietal, and occipital lobe indicated that the parietal lobe as the seizure onset zone. Seizure onset zone was localized (Fig. 4A, B). Parietal lobe resection under an awake operation was performed and a pathologic examination showed severe cortical dysplasia. Postoperatively, the seizures disappeared (Engel I) for 28 months without neurological complication.

6 Parietal Lobe Epilepsy Discussion Aura is defined as that portion of the seizure which occurs before consciousness is lost and which is retained in memory afterwards 7). In a previous study, aura was found in most PLE patients (94%), and the most common aura type encountered was somatosensory (63%). In addition, the sidedness of all auras, except one, was contralateral to the seizure onset zone, suggesting that aura has a localization value 24). In the present study, aura was shown by 30 patients (78.9%). However, the presentation was diverse and sensorimotor aura was seen only in 12 patients (31.6%). Of these, only 9 were true somatosensory aura (23.7%). Although many authors reported that somatosensory aura is the most common in PLE and has a localization value 9,23,24,26), many patients in the present study showed different auras that were more commonly seen in other epilepsy 6,9,23,24,28,31). Moreover, somatosensory auras were not limited to patients with parietal lobe lesions, but could be also present in patients with extraparietal epilepsy 6,28,31). Although somatosensory aura suggests PLE, the relation might be inconsistent. Clinical presentations were also diverse in the present cases and this might be explained with the different pathways of seizure spread 2,24,25,28). The rostral parietal areas are interconnected to the primary motor, premotor, and supplementary motor areas by association and by commissural fibers, and the caudal/posterior parietal cortex is connected to the cingulate gyrus, the insula, and to the parahippocampal gyrus 1,4,5,7,9,14,18,23,24,28). With this explanation, one patient might have more than one aura and ictal semiology, and patients without PLE might have ictal semiology similar to PLE 23, 24). Moreover, seizure activity may spread rapidly and the perceived symptom may reflect J Korean Neurosurg Soc 36

7 CH Kim, et al. secondary spread rather than the primary seizure site 24,27). With these theories, it is difficult to envision which objective clinical seizure manifestations might be associated with seizure activity confined to the parietal lobe 31). In the present study, with ictal semiologies, tonic and/or clonic movement contralateral to the seizure onset zone, was shown only by 14 patients (36.8%). Preoperative work up composed of many kinds of study. Interictal/ictal EEG, inctal/interictal SPECT and PET. Scalp/intracranial interictal EEG is frequently negative or maybe misleading, furthermore, the spread of epileptic discharges from the parietal and occipital lobes to the frontal and temporal regions may obscure the seizure origin 8,9,23,24, 28,31). Previously, we reported that ictal scalp EEG was the most useful diagnostic tool in the localization of the epileptogenic zone in non-lesional neocortical epilepsy 8). However, in the present study, localization was possible only in 4 (10.5%). Although ictal scalp EEG is a very useful method for inferring the location of the seizure onset zone, other diagnostic methods such as PET and/or interictal/ictal SPECT may be used in a complementary fashion. It has been usually known that extratemporal epilepsies are infrequently associated with focal hypometabolism in PET in the absence of a structural lesion, unlike temporal lobe epilepsy, in which interictal temporal hypometabolism occurs in up to 85% of cases 9). However, in the present study, localization was possible only in 28.9% of patients with PET, which showed that interictal PET might not be as valuable a tool as patients with TLE. Ictal SPECT has shown a high yield in defining the source of epileptic activity in PLE 9,13). However, in the present study, ictal SPECT gave little help to verity ictal onset zone. Although theoretically, ictal SPECT is very good diagnostic method, changeover to a postictal hypoperfusion pattern occurs very rapidly in PLE, with a time span of about 12 seconds following seizure termination, and the actually preparation and injection of isotope within this time frame is practically very difficult 8, 10). Moreover, the magnitude of focal hyperperfusion during parietal lobe seizures is modest compared with the global temporal lobe hyperperfusion changes observed during TLE 8). The second International Conference on epilepsy surgery reported that extra-temporal resection, including lesional and non-lesional resection, showed, seizure free in 45.1%, improvement in 35.2%, and unchanged in 19.8%, which compared with seizure free in 67.9%, and improvement in 24% of patients receiving temporal resection 6,28). Results after parietal lesional resections were more favorable and comparable to those of the temporal lobe group : 66.6% were seizure free, 21.5% improved, and 11.9% unchanged 3,6,27). Limited resection in nontumoral posterior onset medically refractory epilepsy, even with a structural abnormality, is not often successful, presumably, because it is a manifestation of an extensive syndrome of developmental disturbance 25). So, extended resection may be successful in patients with posterior epilepsy, as long as interictal and ictal discharges are unilateral. Even if an MR image reveals an abnormal lesion, resection should not be limited to this abnormality but extended to the seizure onset zone and the irritative zone mapped by extraoperative and intraoperative EEG. However, the threat of visual field defect, sensory motor deficit, and language compromise may militate against extensive resection 4). Nearly 90% of patients realized more than a 90% seizure reduction in the present study, and postoperative neurological complications were avoided by extraoperative and intraoperative functional brain mapping. Extraoperative brain mapping using an inserted grid and strip, and intraoperative brain mapping during an awake operation may prevent major complication due to excessive resection of the seizure onset zone. With extensive resection, even without a focal abnormality in MRI, pathologic examinations have shown an abnormal lesion in all biopsy available patients. This finding suggests that structural abnormality, even if it is seen focally in MRI, is not always a refection of whole extent of the pathologic abnormality especially for cortical dysplasia; this finding was reported previously 19,20,24,26). Although 7 patients in the present study showed hippocampal atrophy in MRI (18.4%), this lesion was not routinely resected, and results were similar in the resected and non-resected groups. Moreover, nearly 10% percent of patients with an extra-temporal epileptogenic zone had hippocampal atrophy. There are two hypotheses for this observation; 1) limbic structures are vulnerable to many kinds of insult and these do not necessarily produce medial TLE; and 2) repeated seizure can contribute to hippocampal atrophy 11,14-16,32). Moreover, hippocampal atrophy was not always found in the seizure onset zone, but might be accompanied lesion. And, dual lesions warrant a precise definition of the true epileptogenic area prior to surgical treatment 15,28). Reversible postoperative neurological disturbances are common for surgery of eloquent brain areas, and do not affect the longer-term outcome 2). Moreover, these transient postoperative neurological disturbances were not considered as complications, but as acceptable side effects, if they resolved completely within 2 weeks 2). Sometimes, neurological sequelae occur due to epilepsy surgery, which are irreversible but cannot be avoided,

8 Parietal Lobe Epilepsy such as homonymous hemianopsia in occipital lobe resection 2) So, it is legitimate to consider these as side effects rather than complications 2). In previous studies permanent hemiparesis after parietal resection, occurred in 2% of patients 2,21). Although complications occurred in twenty patients (52.6%), visual field defect, which was unavoidable occurred in 10. Most of the other complications were transient and recovered fully within three months, but permanent complications occurred in three (7.9%), if visual field defects are considered as side effects rather than complications. Because two patients were children, intraoperative monitoring during an awake operation was not possible and the posto-perative neurologic deficits were severe and permanent. However, one patient who underwent an awake operation showed slight leg weakness without difficulty in daily life. During operations on the eloquent brain, awake operations and intraoperative monitoring are valuable tool for the prevention of major neurological sequelae. Cortical dysplasia has been increasingly recognized as a causative pathological substrate in epilepsy and may be an underlying factor in some apparently cryptogenic cases 28). Cortical dysplasia could be defined as a spectrum of derangements in the development of the neocortex, and to be associated with a range of morphologic features and multiple putative etiologic factors, including genetic and environmental influences 15,18). In addition, tumors may co-exist in cortical dysplasia, but no particular pattern or distribution of dysplasia appears to be associated with tumor type 18). Although the surgical outcome of cortical dysplasia is not as good as in hippocampal sclerosis, many authors have suggested that the severity of cortical dysplasia is correlated with surgical outcome 9,29,30), and the present study supports this trend. Conclusion Parietal lobe epilepsy is difficult to diagnosis due to its variable clinical presentation. Careful history taking about aura and semiology, and careful analysis of the results of many diagnostic methods, such as MRI, interictal/ictal EEG, PET, and inter/ictal SPECT are mandatory, although no single method can directly localize the seizure onset zone. Most of all, suspicion of PLE is important and if suspected, then an invasive study covering this region is important to localize the seizure onset zone. Because in the present study, cortical dysplasia was found to be the most common seizure onset zone, and extensive involvement over MRI abnormalities common, extensive resection sparing the functional area with supportive extra/ intra-operative functional brain mapping was necessary. Using this protocol, PLE can be controlled and results are satisfactory with acceptable complication rates. Acknowledgement This work was partially supported by a grant from Seoul National Univertisy Hospital and Seoul National University Medical Research Center. References 1. Aykut-Bingol C, Spencer SS : Nontumoral occipitotemporal epilepsy : localizing findings and surgical outcome. Ann Neurol 46 : , Behrens E, Schramm J, Zentner J, Konig R : Surgical and neurological complications in a series of 708 epilepsy surgery procedures. Neurosurgery 41 : 1-10, Blume WT, Whiting SE, Girvin JP : Epilepsy surgery in the posterior cortex. Ann Neurol 29 : , Commission on Classification and Terminology of the International League Against : Proposal for revised classification of epilepsies and epileptic syndromes. Epilepsia 30 : , Fried I, Spencer DD, Spencer SS : The anatomy of epileptic auras: focal pathology and surgical outcome. J Neurosurg 83 : 60-66, Guldvog B, Loyning Y, Hauglie-Hanssen E, Flood S, Bjornaes H : Predictive factors for success in surgical treatment for partial epilepsy : a multivariate analysis. Epilepsia 35 : , Hong KS, Lee SK, Kim JY, Lee DS, Chung CK : Pre-surgical evaluation and surgical outcome of 41 patients with non-lesional neocortical epilepsy. Seizure 11 : , Ho SS, Berkovic SF, Newton MR, Austin MC, McKay WJ, Bladin PF: Parietal lobe epilepsy : clinical features and seizure localization by ictal SPECT. Neurology 44 : , Keene DL, Jimenez CC, Ventureyra E : Cortical microdysplasia and surgical outcome in refractory epilepsy of childhood. Pediatr Neurosurg 29 : 69-72, Laich E, Kuzniecky R, Mountz J, Liu HG, Gilliam F, Bebin M, et al : Supplementary sensorimotor area epilepsy. Seizure localization, cortical propagation and subcortical activation pathways using ictal SPECT. Brain 120 : , Lawn N, Londono A, Sawrie S, Morawetz R, Martin R, Gilliam F, et al : Occipitoparietal epilepsy, hippocampal atrophy, and congenital developmental abnormalities. Epilepsia 41 : , Lee SK, Lee SH, Kim SK, Lee DS, Kim H : The clinical usefulness of ictal SPECT in temporal lobe epilepsy : the lateralization of seizure focus and correlation with EEG. Epilepsia 41 : , Lehman R, Andermann F, Olivier A, Tandon PN, Quesney LF, Rasmussen TB : Seizures with onset in the sensorimotor face area: clinical patterns and results of surgical treatment in 20 patients. Epilepsia 35 : , Levesque MF, Nakasato N, Vinters HV, Babb TL : Surgical treatment of limbic epilepsy associated with extrahippocampal lesions : the problem of dual pathology. J Neurosurg 75 : , Mischel PS, Nguyen LP, Vinters HV : Cerebral cortical dysplasia associated with pediatric epilepsy. Review of neuropatholgic features and proposal for a grading system. J Neuropathol Exp Neurol 54 : , Nam H, Lee SK, Chung CK, Hong KS, Chang KH, Lee DS : Incidence and clinical profile of extra-medial-temporal epilepsy with hippocampal atrophy. J Korean Med Sci 16 : , Oka A, Kubota M, Sakakihara Y, Yanagisawa M : A case of parietal J Korean Neurosurg Soc 36

9 CH Kim, et al. lobe epilepsy with distinctive clinical and neuroradiological features. Brain Dev 20 : , Prayson RA, Estes ML : Cortical dysplasia : a histopathologic study of 52 cases of partial lobectomy in patients with epilepsy. Hum Pathol 26 : , Raymond AA, Cook MJC, Fish DR, Shoron SD : Magnetic Resonance Scanning and Epilepsy, ed 1. New York : Plenum Press, 1994, p Raymond AA, Fish DR, Sisodiya SM, Alsanjari N, Stevens JM, Shorvon SD : Abnormalities of gyration, heterotopias, tuberous sclerosis, focal cortical dysplasia, microdysgenesis, dysembryoplastic neuroepithelial tumour and dysgenesis of the archicortex in epilepsy. Clinical, EEG and neuroimaging features in 100 adult patients. Brain 118 : , Rydenhag B, Silander HC : Complications of epilepsy surgery after 654 procedures in Sweden, September : a multicenter study based on the Swedish National Epilepsy Surgery Register. Neurosurgery 49 : 51-57, Salanova V, Andermann F, Rasmussen T, Olivier A, Quesney LF : Parietal lobe epilepsy. Clinical manifestations and outcome in 82 patients treated surgically between 1929 and Brain 118 : , Salanova V, Andermann F, Rasmussen T, Olivier A, Quesney LF : Tumoural parietal lobe epilepsy. Clinical manifestations and outcome in 34 patients treated between 1934 and Brain 118 : , Shrvon SD, Fish DR, Andermann F, Bydder GM, Stefan H : Brain structure in epilepsy, ed 2. New York : Plenum Press, 1994, p Sim BS, Choi HY : Surgical Strategy of Epilepsy Arising from Parietal and Occipital Lobes, J Korean Neurosurg Soc 29 : , Sisodiya SM : Surgery for malformations of cortical development causing epilepsy. Brain 123 : , Spencer SS : Long-term outcome after epilepsy surgery. Epilepsia 37 : , Sveinbjornsdottir S, Duncan JS : Parietal and occipital lobe epilepsy : a review. Epilepsia 34 : , 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 125, , Urbach H, Scheffler B, Heinrichsmeier T, von Oertzen J, Kral T, Wellmer J, et al : Focal cortical dysplasia of Taylor s balloon cell type : a clinicopathological entity with characteristic neuroimaging and histopathological features, and favorable postsurgical outcome. Epilepsia 43 : 33-40, Williamson PD, Boon PA, Thadani VM, Darcey TM, Spencer DD, Spencer SS, et al : Parietal lobe epilepsy : diagnostic considerations and results of surgery. Ann Neurol 31 : , Wyler AR, Wilkus RJ, Rostad SW, Vossler DG : Multiple subpial transections for partial seizures in sensorimotor cortex. Neurosurgery 37 : , 1995