Seizure (2006) 15, 420—427

www.elsevier.com/locate/yseiz

Seizure outcome after surgery for epilepsydue to focal cortical dysplastic lesions

Veriano Alexandre Jr.a, Roger Walz a, Marino M. Bianchin a,Tonicarlo R. Velasco a, Vera C. Terra-Bustamante a,Lauro Wichert-Ana a, David Araujo Jr.b, Helio R. Machado c,Joao A. Assirati Jr.c, Carlos G. Carlotti Jr.c, Antonio C. Santos b,Luciano N. Serafini d, Americo C. Sakamoto a,*

aDepartment of Neurology, Psychiatry and Psychology, Ribeirao Preto School of Medicine,University of Sao Paulo, BrazilbDepartment of Radiology, Ribeirao Preto School of Medicine, University of Sao Paulo, BrazilcDepartment of Neurosurgery, Ribeirao Preto School of Medicine, University of Sao Paulo, BrazildDepartment of Pathology, Ribeirao Preto School of Medicine, University of Sao Paulo, Brazil

Received 9 November 2004; received in revised form 25 November 2005; accepted 23 May 2006

KEYWORDSMalformation ofcortical development;Focal corticaldysplasia;Intractable epilepsy;Epilepsy surgery

Summary Neocortical development is a highly complex process encompassingcellular proliferation, neuronal migration and cortical organization. At any time thisprocess can be interrupted or modified by genetic or acquired factors causingmalformations of cortical development (MCD).

Epileptic seizures are the most common type of clinical manifestation, besidesdevelopmental delay and focal neurological deficits. Seizures due to MCD are fre-quently pharmacoresistant, especially those associated to focal cortical dysplasia(FCD).

Surgical therapy results have been reported since 1971, however, currently avail-able data from surgical series are still limited, mainly due to small number of patients,distinct selection of candidates and surgical strategies, variable pathological diag-nosis and inadequate follow-up.

This study addresses the possibilities of seizure relief following resection of focalcortical dysplasia, and the impact of presurgical evaluation, extent of resection andpathological findings on surgical outcome.

We included 41 patients, 22 adults and 19 children and adolescents, with medicallyintractable seizures operated on from 1996 to 2002. All were submitted to standar-dized presurgical evaluation including high-resolution MRI, Video-EEG monitoring andictal SPECT. Post-surgical seizure outcome was classified according to Engel’s schema.Univariate and multivariate analysis were performed.

* Corresponding author. Campus Universitario, Ribeirao Preto, SP, CEP 14048-900, Brazil. Tel.: +55 16 3633 9000; fax: +55 16 3633 0760.E-mail address: sakamoto@fmrp.usp.br (A.C. Sakamoto).

1059-1311/$ — see front matter # 2006 British Epilepsy Association. Published by Elsevier Ltd. All rights reserved.doi:10.1016/j.seizure.2006.05.005

Seizure outcome after surgery for epilepsy due to focal cortical dysplastic lesions 421

Fifteen patients had temporal and 26 extratemporal epilepsies. Of the total 26patients (63.4%) reached seizure-free status post-operatively. There was no correla-tion between outcome and age at surgery, duration of epilepsy, frequency of seizures,and pathological findings. There was, however, a clear correlation with topography ofFCD (temporal versus extratemporal) and regional ictal EEG onset, on univariate aswell as multivariate analysis.# 2006 British Epilepsy Association. Published by Elsevier Ltd. All rights reserved.

Introduction

Neocortical development is a highly complex inter-play of multiple processes involving cellular prolif-eration, neuronalmigration andcortical organizationduring which neurons and glia cells originate,migrate, differentiate and establish structural andfunctional connectivity. Atany timeduring this devel-opment, these processes can be interrupted or mod-ified by either genetic or acquired factors, or both,causing focal, multifocal or diffuse anatomicalderangements known as malformations of corticaldevelopment (MCD). Different classifications forthese structurally and functionally complex lesionshave been proposed over the years.1 However, con-sensual classification schemes are difficult to beachievedmainly because these lesions can take placein different stages of development, at various loca-tions, and associated to multiple etiologies and dif-ferent pathologic substrates.

Epileptic seizuresareby far themost commontypeof clinical manifestation, besides developmentaldelay and focal neurological deficits. Seizures dueto MCD are frequently pharmacoresistant2 andbetween all malformations the focal cortical dyspla-sia (FCD) emerges as themost important pathologicalsubstrate of surgically remediable epilepsies. Surgi-cal results have been reported since 1971,3 but espe-cially after the advent of MRI.4 However, currentlyavailable data from surgical series are still limited,due to the small numberof patients, heterogeneity ofFCD, distinct selection of surgical cases, variablepathological diagnosis, different surgical strategiesand inadequate follow-up.

In this study we aimed to assess the possibilitiesof seizure relief following resection of FCD, and toverify the impact of the presurgical evaluation,surgical resection and pathological findings withrespect to surgical outcome.

Patients and methods

We analyzed all patients with medically intractableepilepsy due to FCD who underwent resective sur-gery at the Epilepsy Surgery Center, Ribeirao PretoMedical School, from February 1996 to June 2002.

Forty-one patients were included according to thefollowing criteria: (1) patients submitted to stan-dardized presurgical evaluation including clinicaldata, non-invasive and invasive (whenever per-formed) electrophysiological evaluation, high-reso-lution MRI and single photon emission computedtomography (SPECT); (2) pathologically confirmedFCD; (3) post-surgical follow-up �1 year. Patientswith dual pathology and associated tumor,5 tuberoussclerosis,6 bilateral or diffuse MCD on MRI4 or insuf-ficient follow-up time were excluded (one patientdied days after surgery for unclear reasons).

Clinical data included age at onset of seizures,age at surgery, duration of epilepsy and frequency ofseizures. According to their age at the time ofsurgery the patients were divided into two sub-groups: children/adolescents (�18 years old) andadults (>18 years old). The frequency of the sei-zures was categorized as daily, weekly or monthlyseizures, when subjects had averages of at least oneseizure per day, per week or per month, respec-tively. All patients had comprehensive neuropsycho-logical testing whenever they were able tocooperate and perform it.

The electrophysiological evaluation during video-electroencephalography (VEEG) monitoring (Van-gard system) with scalp electrodes included inter-ictal and ictal recordings. Interictal findings werecategorized as ‘‘regional’’ when spikes were con-fined to one region (in different clusters maximumover a single lobe or in two contiguous regions), and‘‘non-regional’’, when spikes were multiregionaland independent (involving two or more regions inmore than one lobe), bilateral or generalized. IctalEEG patterns were similarly classified as ‘‘regional’’when the ictal onset was limited to only one region,and ‘‘non-regional’’, when it involved more thanone region from one hemisphere (lateralized), orwas non-lateralized or diffuse. The seizures typeswere classified according International LeagueAgainst Epilepsy.7

Invasive electrophysiological studies throughchronically implanted subdural electrodes were per-formed whenever the non-invasive methodologies,including high resolution MRI, resulted non-conver-gent, or when the localization of the presumed epi-leptogenic lesionwas close toor overlappedeloquent

422 V. Alexandre Jr. et al.

areas, thus requiring electrical stimulation for cor-tical mapping. Intra-operative electrocorticography(ECoG) was performed in selected cases, with theobjective of optimizing the resection of the lesionand the irritative zone.

The ictal SPECT (Tecnecio-99m-etilenodicisteine)was performed during the VEEG and categorized as‘‘convergent’’ whenever a single region of hyper-perfusion appeared and converged to the surgicallyresected region. Whenever more than one region ofhyperperfusion appeared, we classified theseSPECTs as partially convergent.

All patients had performed high-resolution MRI(1.5T) with special protocols for epilepsy using T1-and T2-weighted proton density, fluid-attenuatedinversion-recovery (FLAIR) sequences in axial, cor-onal, and sagital planes. MRI findings were indepen-dently analyzed by two neuroradiologists withexperience in the field of epileptology.

The location of the FCD was classified in temporaland extra-temporal, based on presurgical MRI in 38cases, and surgical resection in three other caseswith normal MRI. The completeness of the FCDresection was assessed through postsurgical MRIand classified as complete or incomplete.

Table 1 Univariate analysis for outcome correlations

Variable Seizure-free(Class I)

Non se(Class

Age at surgeryChildren/adolescents 13 6Adults 13 9

LocationTemporal 13 2Extratemporal 13 13

Interictal EEGRegional 15 5Non-regional 11 10

Ictal EEGRegional 22 7Non-regional 4 8

Invasive EEGPerformed 19 13Non-performed 7 2Intra-operative 10 8Extra-operative 9 5

Completeness of resectionComplete 12 2Incomplete 9 9

HistopathologicArchitectural 11 5Cytoarchitectural 15 10

OR: odds ratio; 95% CI: 95% confidence interval; *: variable with sig

The surgical specimens (hematoxylin/eosin-stained sections, 5 mm thicknesses) were reviewedand patterns of FCD classified according to previousreports8 in two grades: (I) disorganization of corticalarchitecture (architectural dysplastic lesion) and(II) disorganization of cortical architecture plusneuronal cytomegaly and balloon cells (cytoarchi-tectural dysplastic lesion).

The combination of clinical, non-invasive andinvasive electrophysiological data, and neuroima-ging results made it possible to define the bestsurgical strategy for each case.

The surgical outcomewas classified9 after at least1 year10 of follow-up: class I, seizure-free; class II,rare seizures (no more than two seizures a year);class III, worthwhile improvement (more than 75%seizure reduction); class IV, no worthwhile improve-ment (less than 75% seizure reduction). For thepurposes of this study, we divided the patients ina seizure-free group (Class I) and in a non-seizure-free group (Class II, III, and IV). Outcome was cor-related with: age at surgery, location of FCD, elec-trophysiological evaluation, completeness ofresection and pathological examination of surgicalspecimen.

izure-freeII, III, IV)

Significance

( p) OR 95% CI

0.74 1.5 0.41—5.43

0.02 * 6.5 1.21—34.73

0.19 2.7 0.72—10.27

0.01 * 6.2 1.44—27.37

0.44 0.4 0.07—2.33

0.72 0.6 0.16—2.91

0.06 6.0 1.03—34.85

0.74 1.4 0.38—5.52

nificant p-value included for multivariate analysis.

Seizure outcome after surgery for epilepsy due to focal cortical dysplastic lesions 423

Data were analyzed using commercially availablestatistical software SPSS 10.0 (Chicago, IL); statis-tical analysis consisted of application of two tailedFisher’s exact test to correlate different variableswith surgical outcome. A p-value of <0.05 wasconsidered significant and multivariate analysiswas calculated for variables with a significant p-value.

Results

Results of univariate and multivariate analyzesregarding post-surgical seizure outcome are pre-sented in Tables 1 and 2, respectively.

Overall post-surgical seizure outcome andcomplications

After a minimum follow-up of 1 year, 26 patients(63.4%) were seizure-free (class I), one (2.4%) hadno more than two seizures a year (class II), 12(29.3%) had a worthwhile improvement (class III),and only 2 (4.9%) had no improvement (class IV).Four patients had a second surgical intervention foruncontrolled seizures.

The most common complications of surgical pro-cedure were motor and visual deficits. The motordeficits occurred in six patients (14.6%) and includedincrease of hemiparesis in two patients (submittedto hemispherectomy), transitory hemiparesis in onepatient, mild definitive hemiparesis in two patientsandmoderate in other one. In one patient submittedto subdural electrodes implantation and MST asso-ciated to lesionectomy, an intracerebral hematomacaused the deficit. The visual deficits occurred in 11patients (26.8%) and included partial deficit tohemianopsia in one patient, hemianopsia in fivepatients with occipital resections, and quadranta-nopsia in five patients with temporal resection.Seven patients could not be investigated for visualdeficits because of poor cooperation with oftalmo-logic examination. Other complications includedrespiratory infection in two patients (4.8%), scalpbulging in four (9.6%), cerebro-spinal fluid fistulae in

Table 2 Multivariate analysis for outcome correla-tions

Variable Significance

( p) OR 95% CI

Location 0.02 9.1 1.31—63.84Ictal EEG 0.01 8.7 1.5—50.60

OR: odds ratio; 95% CI: 95% confidence interval.

one (2.4%), status epilepticus after surgery in one(2.4%), acute subdural hematoma in one (2.4%),edema and increased intracranial pressure in one(2.4%), and scar infection in another (2.4%). Onepatient with intracranial electrodes developed anacute subdural hematoma (2.4%) but no descom-pression procedure was necessary. One patient diedhours after surgery for unclear reasons and was notincluded in this study due to insufficient follow-up.

Clinical data

Seizures began from 3 days to 36 years old (mean6.4, median 4 years) and were intractable to highdoses of multiple antiepileptic medications. Nine-teen children and adolescents (46.3%) and 22(53.7%) adults were included. There was no signifi-cant correlation between outcome and age at sur-gery ( p = 0.74). Duration of epilepsy prior to surgeryranged from 1 to 35 years (mean 13.3, median 12years). Twenty-five patients (61.0%) had daily sei-zures, 10 (24.4%) weekly seizures and six (14.6%)monthly seizures.

Location of FCD

The location of FCD was extra-temporal in 26patients (63.4%) and temporal in 15 (36.6%). Betterseizure outcome was observed following temporalresections (p = 0.02). In the temporal group (15cases), 13 patients were seizure-free and twopatients had a worthwhile improvement of seizurefrequencies.

Electrophysiological evaluation

Video-EEG monitoring data disclosed regional inter-ictal discharges in 20 patients (48.8%) and non-regional in 21 (51.2%). The non-regional interictaldischarges included multiregional spikes in 11patients, generalized spikes in four, and associationof them in six. Patients with regional discharges hada trend towards a better outcome, but the differ-ence was not significant when compared to patientswith non-regional discharges ( p = 0.19).

The ictal scalp EEG was classified as regional in 29patients (70.7%) and non-regional in 12 (29.3%). Thenon-regional ictal EEG included diffuse activity ineight patients and more than one pattern (diffuse,lateralized, non-lateralized) in the other four.Patients with regional ictal patterns had betteroutcome when compared to patients with non-regional ictal patterns ( p = 0.01).

The seizures types were simple partial seizures(SPS) in four (9.7%), complex partial seizures withautomatisms (CPS) in nine (21.9%), aura and CPS in

424 V. Alexandre Jr. et al.

six patients (14.6%), and tonic seizures with impair-ment of conscious in 22 (53.6%).

Thirty-two patients had invasive electrophysio-logical studies through intraoperative acute elec-trocorticography or extraoperative recordingsthrough chronically implanted subdural electrodes.Seizure outcome was not influenced by the use ofsubdural electrodes, nor there was any differencebetween acute (18 patients) and chronic ECoG (14patients) ( p = 0.72). Of those 14 subjects that hadsubdural evaluation, nine had in addition intrao-perative recordings in the vicinity of the resection.Out of the 26 patients who had extra-temporallocation of FCD, invasive electrophysiological stu-dies were performed in 24 cases (12 intraoperative,five extraoperative and seven combined record-ings). In the 15 patients with temporal FCD, invasiverecordings were performed in only eight cases (sixintraoperative and two combined).

SPECT

Thirty-one patients had performed ictal SPECT. In 15patients the location of the hyperperfusion area wasconvergent to the resected region. In 9 subjects theictal SPECTwas non-convergent, in the remaining 7patients there was only a partial convergencebetween the hyperperfusion area and the surgicallyremoved region.

Completeness of resection

Completeness of the excision of the lesions wasestimated through surgeons’ operative reportsand comparison of pre- and postsurgical MRI. Itwas assessed in 32 patients, but not possible toevaluate the remaining 9 subjects (three had normalpre-operative MRI, two had hemispherectomies,and four did not have post-operative MRI. Fourteenpatients had a complete resection (46.7%), and 12 ofthem were seizure-free. Eighteen patients had anincomplete resection (53.6%), of those only 9 wereseizure-free. A trend toward a better outcome(Class I) was observed when the resection was com-plete ( p = 0.06). In the temporal group (15patients), 13 patients were submitted to lobectomyand two to lesionectomy. In the extratemporalgroup (26 patients), 14 patients were submittedto lesionectomy, four to lobectomy, four to multi-lobar resection, two to lesionectomy plus multiplesubpial transection (MST) and two to hemispherect-omy. In the extratemporal cases that presented withunilobar multifocal spikes the lobectomy was indi-cated. So in overall 17 patients were submitted tolobectomy and 18 patients to lesionectomy (twowith MSTassociated). The overall post-surgical out-

come for lobectomies was 12 patients seizure-free,and for lesionectomies was 11 patients seizure-free(no significant difference p = 0.72).

Histopathological findings

The histopathological study revealed cytoarchitec-tural MCD in 25 cases (61%) and architectural MCD in16 cases (39%). No correlation was found betweenthese two groups regarding outcome ( p = 0.74).

Discussion

In the treatment of epilepsy, the ideal outcomemuststill be seizure-freedom.11 We obtained 63.4% ofseizure-free patients. This compares favorably withoutcomeofpatientswithhippocampal sclerosis12 andlow-grade neoplasms,13 situations in which the sur-gical outcome has been viewed optimistically.14

Reports of surgical outcome have been variableand some authors have concluded that overall,approximately 41% of patients with MCD may berendered seizure-free over a 2-year follow-up.15

Recent series suggest that 50—70% seizure-free out-come may be the best obtained result.14,16—18

The outcome was no better, when comparingsurgeries performed in childhood and adolescentsas opposed to adulthood. We believe that the con-vergence of preoperative evaluation data and notjust the age of onset of epilepsy is relevant to thepost-operative outcome. However, the epilepsiesassociated with MCD usually begin in children, andthe surgical treatment has been considered early forall patients with these lesions.19 Therefore, in chil-dren with MCD and many seizures per day, theobjectives of the surgeries are the relief of cata-strophic epilepsy, resumption of developmental pro-gress, and improvement in behavior. Some authorsreported that children and adolescents appeared tohave a better seizure outcome, though the differ-ence in comparison with adults was not significant.18

Otherwise, a favorable outcome of seizures inpatients with FCD Taylor type and adult-onset epi-lepsy also can be achieved.20 In our series there wasa similar number of children/adolescents and adultsat time of surgery.

Weobserveda significantdifference in theseizure-free rate in patients undergoing temporal comparedto extratemporal resection. In our study this latergroup included cases of multilobar as well as hemi-spheric surgeries. The two patients (one children andone adolescent) submitted to hemispherectomyachieved seizure-free state and experienced devel-opmental progress after relief of their catastrophic

Seizure outcome after surgery for epilepsy due to focal cortical dysplastic lesions 425

epilepsy. Some series also reportedbetter outcome inpatients with temporal compared to extratemporalresections.21 Other studies failed to confirm this viewbased on topography.15,16 When the lesion directlyinvolves the hippocampus, the decision to undertakea mesial resection is relatively clear. However, in thesetting of a lateral lesion without involvement ofmesial structures, careful consideration must begiven to the potential risks and benefits of a mesialresection. In our series, of the 15 patients in thetemporal group, thirteen were submitted to ‘‘anbloc’’ temporal resection, includingmesial structuresand the lesion. Of these, only one patient had theclassic findings of hippocampal sclerosis and 11patients had mild cell loss in CA1 and fascia dentata.Extrahippocampal lesions do not rule out the hippo-campusasthesiteofseizureorigin.22 In lesionssuchaslowgradeneoplasms in the temporal lobe, the resultsof lesionectomy alone are disappointing. Other con-siderations, includingtheunderstandingofthespatialand causal relationships between structural lesionsand epilepsy (extrinsic versus intrinsic epileptogeni-city) are essential to rationalize the therapeutic stra-tegies.23 Caution is also needed when interpretingtemporal lobe MRI findings. FCD may mimic the MRIappearance of tumors and the reverse may also betrue. Because of the possibility of dual pathology, andbecause MRI may not show the full extent of thelesion, a more aggressive approach including mesialtemporal lobe resectionmay be necessary to achievesatisfactory seizure control.24

Theepileptogenic zonewasdefinedusing informa-tion from clinical findings, EEG (most commonlyscalp, possibly supplemented by acute or chronicintracranial studies) and in some cases additionalfunctional methods (SPECT). The scalp EEG findingsof patients with MCD are variable, frequently disclos-ing multifocal or widespread interictal spikes, andpoorly localized ictal EEG.25 The objectives of pre-surgical evaluation are to identify the area of brainmost responsible for generating habitual seizures andto demonstrate that it can be removed withoutcausing additional and unacceptable deficits.9 Wenoticed a trend to a better outcome in patients withregional interictal EEG while in those with regionalictal EEG the surgical outcome was significantly bet-ter. Others studies concluded that scalp EEGwere notsignificantly correlated with the post-operativeresults.15,21,26 In our series of isolated MCD, tonicseizures were more common and SPS were less com-mon, in contrast with Bautista et al.14 who reportedSPS in 42% of their cases with isolated FCD.

The role of intracranial recordings remains uncer-tain. Although authors suggested that chronic sub-dural electrographic findings do no correlate withoutcome, there are insufficient published data to

confirm it.15,16,21 We were also unable to confirmwhether subdural evaluation improved seizure out-come after surgery for MCD. We understand thatfuture studies are necessary to assess it. Even whenMRI reveals the lesion, the boundaries were notalways well defined and in most cases, as revealedby stereo-EEG (SEEG), the epileptogenic zoneextendedbeyondthevisibleMRI lesion.27Ourfindingssuggest that ictal SPECTmay be a useful complemen-tary method during the presurgical evaluation andmayprovideaguide to invasiveEEG, helping todefinethe epileptogenic zone. In the three cases with noabnormality on MRI, SPECTwas convergent with thearea of surgical resection. These findings are inagreement with other authors.28

The surgical strategy includes the completeremoval of the MRI visible lesion coupled with resec-tion of the region of abnormal electrographic activitybut preserving the functional cortex.29 When theremoval of the lesion was complete, we observed atrend toward a better postoperative outcome. Thisdifferencewasnot significantprobablybecause therewere few patients in either group. Conversely, somecases achieved seizure-free state despite incompleteresection while others did not become seizure-freedespite of the completeness of the resection by MRIanalysis. The complete resection of FCD resulted ingood seizure outcome in children.30,31

There are reports of complete excision accordingto MRI not being associated with seizure-free out-come, attributed variously to ‘‘inaccuracy inmethod of quantitating lesion resection’’ or thepresence of MRI-occult pathology. Remains possiblethat even standard high-resolution MRI may fail toreveal the entire extent of MCD, so that perceivedcomplete resection is, in fact, not necessarily com-plete resection of all the pathology present.15 It ispossible that residual microscopic lesion couldaccount for persistent seizures arising adjacent toan apparently complete lesion removal as judged byMRI. The emerging MRI techniques and functionalneuroimaging may contribute to disclose theseoccult MCD. Otherwise, even if incomplete, resec-tion may remove enough of the epileptogenic zoneto render a patient seizure-free. It would be causedby a network disruption or functional disconnection,just like occur in cases of functional hemispherect-omy.32 So that subtotal resection did not preclude aseizure-free outcome.33 Attempts have been madeto grade the severity of dysplastic lesions on patho-logic findings. The least abnormality consists ofvertical and horizontal dyslamination translating adisorder of migration. Abnormal neurons and bal-loon cells can also be present.

The presence of balloon cells and cytomegalicneurons has been linked to the occurrence of the

426 V. Alexandre Jr. et al.

most severe epileptic abnormalities34 and to poorerpostoperative outcome for patients with ballooncells in surgical specimen.35 Other authors founda relatively good postoperative prognosis in thepresence of balloon cells36 while others did not findsignificant difference in seizure outcome in patientswith or without balloon cells and cytomegalic neu-rons.31 The histological detection of abnormal cel-lular elements (balloon cells and cytomegalicneurons) in our study was not associated with adifference in postoperative seizure control. Wesuppose that the complete resection of the lesionand epileptogenic zone is more important for theseizure outcome than the presence of the ballooncells or cytomegalic neurons in the surgical speci-men, as the complete removal of the lesion in ourstudy had a trend toward a better postoperativeoutcome.

The multivariate analysis was significant for loca-tion of lesion in the temporal lobe and the presenceof regional ictal EEG. From our study we concludedthat patients with lesion in the temporal lobe andpatients with regional ictal EEG presented the bestpostoperative outcome. However, patients withextratemporal lesions and more extensive neuro-physiological changes may also become seizure-free, demonstrating that such findings should notpreclude further presurgical evaluation.

Acknowledgements

We express our gratitude to Mrs. Elettra Greene forher professional assistance in the English version.We thank the support of CNPq and FAEPA.

References

1. Barkovich AJ, Kuzniecky RI, Jackson GD, Guerrini R, DobynsWB. Classification system formalformations of cortical devel-opment: update 2001. Neurology 2001;57:2168—78.

2. Semah F, Picot MC, Adam C, et al. Is the underlying cause ofepilepsy a major prognostic factor for recurrence? Neurology1998;51:1256—62.

3. Taylor DC, Falconer MA, Bruton CJ, Corsellis JA. Focal dys-plasia of the cerebral cortex in epilepsy. J Neurol NeurosurgPsychiatry 1971;34:369—87.

4. Kuzniecky RI. Magnetic resonance imaging in developmentaldisorders of the cerebral cortex. Epilepsia 1994;35(Suppl.6):S44—56.

5. Degen R, Ebner A, Lahl R, Leonhardt S, Pannek HW, Tuxhorn I.Various findings in surgically treated epilepsy patients withdysembryoplastic neuroepithelial tumors in comparison withthose of patients with other low-grade brain tumors andother neuronal migration disorders. Epilepsia 2002;43:1379—84.

6. Sparagana SP, Roach ES. Tuberous sclerosis complex. CurrOpin Neurol 2000;13:115—9.

7. Commission on classification and terminology of the Interna-tional League Against Epilepsy. Proposal for revised clinicaland electroencephalografic classification of epileptic sei-zures. Epilepsia 1991;22:489—501.

8. Mischel PS, Nguyen LP, Vinters HV. Cerebral cortical dysplasiaassociated with pediatric epilepsy. Review of neuropatholo-gic features and proposal for a grading system. J NeuropatholExp Neurol 1995;54:137—53.

9. Engel Jr J. Surgery for seizures. N Engl J Med 1996;334:647—52.

10. Luders H, Murphy D, Awad I, et al. Quantitative analysis ofseizure frequency 1 week and 6, 12, and 24 months aftersurgery of epilepsy. Epilepsia 1994;35:1174—8.

11. Walker MC, Sander JW. The impact of new antiepileptic drugson the prognosis of epilepsy: seizure freedom should be theultimate goal. Neurology 1996;46:912—4.

12. Sperling MR, O’Connor MJ, Saykin AJ, Plummer C. Temporallobectomy for refractory epilepsy. JAMA 1996;276:470—5.

13. Britton JW, Cascino GD, Sharbrough FW, Kelly PJ. Low-gradeglial neoplasms and intractable partial epilepsy: efficacy ofsurgical treatment. Epilepsia 1994;35:1130—5.

14. Bautista JF, Foldvary-Schaefer N, Bingaman WE, Luders HO.Focal cortical dysplasia and intractable epilepsy in adults:clinical, EEG, imaging, and surgical features. Epilepsy Res2003;55:131—6.

15. Sisodiya SM. Surgery for malformations of cortical develop-ment causing epilepsy. Brain 2000;123(Pt 6):1075—91.

16. Edwards JC, Wyllie E, Ruggeri PM, et al. Seizure outcomeafter surgery for epilepsy due to malformation of corticaldevelopment. Neurology 2000;55:1110—4.

17. Hong SC, Kang KS, Seo DW, et al. Surgical treatment ofintractable epilepsy accompanying cortical dysplasia. J Neu-rosurg 2000;93:766—73.

18. Kral T, Clusmann H, Blumcke I, et al. Outcome of epilepsysurgery in focal cortical dysplasia. J Neurol Neurosurg Psy-chiatry 2003;74:183—8.

19. Hilbig A, Babb TL, Najm I, Ying Z, Wyllie E, BingamanW. Focalcortical dysplasia in children. Dev Neurosci 1999;21:271—80.

20. Siegel AM, Cascino GD, Elger CE, et al. Adult-onset epilepsy infocal cortical dysplasia of Taylor type. Neurology 2005;64:1771—4.

21. Hirabayashi S, Binnie CD, Janota I, Polkey CE. Surgicaltreatment of epilepsy due to cortical dysplasia: clinicaland EEG findings. J Neurol Neurosurg Psychiatry 1993;56:765—70.

22. Levesque MF, Nakasato N, Vinters HV, Babb TL. Surgicaltreatment of limbic epilepsy associated with extrahippocam-pal lesions: the problem of dual pathology. J Neurosurg1991;75:364—70.

23. Zentner J, Hufnagel A, Wolf HK, et al. Surgical treatment ofneoplasms associated with medically intractable epilepsy.Neurosurgery 1997;41:378—86.

24. Modha A, Vassilyadi M, Keene D, et al. Temporal lobe focalcortical dysplasia: MRI imaging using FLAIR shows lesionsconsistent with neoplasia. Childs Nerv Syst 2000;16:269—77.

25. Raymond AA, Fish DR. EEG features of focal malformationsof cortical development. J Clin Neurophysiol 1996;13:495—506.

26. Palmini A, Andermann F, Olivier A, et al. Focal neuronalmigration disorders and intractable partial epilepsy: a studyof 30 patients. Ann Neurol 1991;30:741—9.

27. Tassi L, Colombo N, Garbelli R, et al. Focal cortical dysplasia:neuropathological subtypes, EEG, neuroimaging and surgicaloutcome. Brain 2002;125:1719—32.

28. Kuzniecky R, Mountz JM, Wheatley G, Morawetz R. Ictalsingle-photon emission computed tomography demonstrates

Seizure outcome after surgery for epilepsy due to focal cortical dysplastic lesions 427

localized epileptogenesis in cortical dysplasia. Ann Neurol1993;34:627—31.

29. Palmini A, Gambardella A, Andermann F, et al. Intrinsicepileptogenicity of human dysplastic cortex as suggestedby corticography and surgical results. Ann Neurol 1995;37:476—87.

30. Hader WJ, Mackay M, Otsubo H, et al. Cortical dysplasticlesions in childrenwith intractable epilepsy: role of completeresection. J Neurosurg 2004;100(2 Suppl. Pediatrics):110—7.

31. Kloss S, Pieper T, Pannek H, et al. Epilepsy surgery in childrenwith focal cortical dysplasia (FCD): Results of long-termseizure outcome. Neuropediatrics 2002;33:21—6.

32. Rasmussen T. Hemispherectomy for seizures revisited. Can JNeurol Sci 1983;10:71—8.

33. Hudgins RJ, Flamini JR, Palasis S, et al. Surgical treatment ofepilepsy in children caused by focal cortical dysplasia.Pediatr Neurosurg 2005;41(2):70—6.

34. Cepeda C, Hurst RS, Flores-Hernandez J, et al. Morphologicaland electrophysiological characterization of abnormal celltypes in pediatric cortical dysplasia. J Neurosci Res 2003;72:472—86.

35. Palmini A, Gambardella A, Andermann F, et al. Operativestrategies for patients with cortical dysplastic lesions andintractable epilepsy. Epilepsia 1994;35(Suppl. 6):S57—71.

36. Rosenow F, Luders HO, Dinner DS, et al. Histopathologicalcorrelates of epileptogenicity as expressed by electrocortico-graphic spiking and seizure frequency. Epilepsia 1998;39:850—6.