Gender and hemispheric differences in temporal lobe epilepsy: A VBM study

Gender

and

hemispheric

differences

in

temporal

lobe

epilepsy:

A

VBM

study

Maria

Teresa

Castilho

Garcia

Santana

*

,

Andrea

Parolin

Jackowski

,

Fernanda

dos

Santos

Britto

,

Gabriel

Barbosa

Sandim

,

Luı´s

Ota´vio

Sales

Ferreira

Caboclo

,

Ricardo

Silva

Centeno

_,

_Henrique

_Carrete

_Jr.

_,

_Elza

_Ma´rcia

_Targas

_Yacubian

a_{UnidadedePesquisaeTratamentodasEpilepsias(UNIPETE),DepartmentofNeurologyandNeurosurgery,UniversidadeFederaldeSa˜oPaulo,}

Sa˜oPaulo,Brazil

b_{Laborato´rioInterdisciplinardeNeurocieˆnciasClı´nicas(LiNC),DepartmentofPsychiatry,UniversidadeFederaldeSa˜oPaulo,Sa˜oPaulo,Brazil} c_{DepartamentodeDiagno´sticoporImagem,UniversidadeFederaldeSa˜oPaulo,Sa˜oPaulo,Brazil}

1. Introduction

Gender differences are recognized in the functional and anatomicalorganizationofthehumanbrain.1Differencesbetween gendersareprobablyexpressedearlyinlife,whendifferentialrates of cerebral maturation occur. In 1960s, Taylor suggested the biologicalbasisforahighervulnerabilityofthemalebrainandof the left hemisphere.2 _{According to this hypothesis, cerebral}

maturationwouldbemorerapidingirls,sothatboyswouldbe atagreaterriskforalongertimeinsuchawaythatapotential seizure-producinginsultwouldaffectthelessfunctionallyactive side,thelefthemisphere.

Sexual dimorphism has already beendescribed in temporal lobeepilepsywithmesialtemporalsclerosis(TLE-MTS)inrelation tothephenomenologyofseizures.3–5

Structural6_{andfunctionalneuroimaging,}7_aswellas

anatomo-pathologicalstudies,8_{haveshowngenderdifferencesinTLE-MTS.}

Men exhibit more volume deficits in brain areas other than ipsilateralhippocampusthanwomeninMRIstudies.6Thereisalso evidenceofgender-baseddifferencesinglucosehypometabolism measuredbyPET.9_{Finally,gendermaydifferentiallyinfluencethe}

degreeofgliosisinTLE-MTSpatients.8

Voxel-basedmorphometry(VBM),afullyautomated comput-erized quantitative MRI analysis technique, is a widely used method to identify gray and white matter abnormalities in epilepsy,10 _{enabling a comprehensive analysis of global brain}

structure.11 _{Several VBM studies have shown that TLE-MTS}

extends beyond mesialtemporal structures, and that there are differencesintheextentofanatomicaldamage,inboth,grayand whitematter,whencomparingleftandrighthemispheres.Someof themhaveinvestigatedtherelationshipbetweenTLEandclinical variables such as history of febrile seizures, age at onset and duration of epilepsy,12–14 _{presence of secondarily generalized}

seizures13 _{and side of MTS,}15,16 _{while none have approached}

genderdifferences.

ARTICLE INFO

Articlehistory:

Received16March2013

Receivedinrevisedform10December2013 Accepted12December2013

Keywords:

Gender

Mesialtemporalsclerosis Temporallobeepilepsy Voxel-basedmorphometry

ABSTRACT

Purpose:Genderdifferencesarerecognizedinthefunctionalandanatomicalorganizationofthehuman brain.Differencesbetween gendersareprobablyexpressedearlyinlife,whendifferential rates of cerebralmaturationoccur.Sexualdimorphismhasbeendescribedintemporallobeepilepsywithmesial temporalsclerosis(TLE-MTS).Severalvoxel-basedmorphometry(VBM)studieshaveshownthat TLE-MTSextends beyond mesial temporal structures, and that there are differences in the extent of anatomicaldamagebetweenhemispheres,althoughnonehaveapproachedgenderdifferences.Ouraim wastoinvestigategenderdifferencesandanatomicalabnormalitiesinTLE-MTS.

Methods:VBM5wasemployedtoanalyzegenderandhemisphericdifferencesin120patientswith TLE-MTSand50controls.

Results:VBMabnormalitiesweremorewidespreadinleft-TLE;whileinwomenchangesweremostly seenintemporalareas,frontalregionsweremoreaffectedinmen.

Conclusions: Ourstudyconfirmedthatgenderandlateralityareimportant factorsdeterminingthe natureandseverityofbraindamageinTLE-MTS.Differentialratesofmaturationbetweengenderand hemispheresmayexplainthedistinctareasofanatomicaldamageinmenandwomen.

ß2013BritishEpilepsyAssociation.PublishedbyElsevierLtd.Allrightsreserved.

* Correspondingauthorat:UnidadedePesquisaeTratamentodasEpilepsias, DepartmentofNeurologyandNeurosurgery,715,Napolea˜odeBarros,13thFloor, 04024-002Sa˜oPaulo,SP,Brazil.Tel.:+551155764236.

E-mailaddress:teresacgsantana@hotmail.com(M.T.C.G.Santana).

ContentslistsavailableatScienceDirect

Seizure

j o urn a l hom e pa g e : ww w . e l se v i e r. c om / l oca t e / y se i z

1059-1311/$–seefrontmatterß2013BritishEpilepsyAssociation.PublishedbyElsevierLtd.Allrightsreserved.

Theaimofthisstudywastoinvestigategenderdifferencesand anatomicalabnormalitiesinTLE-MTSusingVBM.

2. Methods 2.1. Subjects

One-hundredand twenty patients with refractoryunilateral TLE-MTS (57 males) underwent a comprehensive pre-surgical evaluationattheEpilepsySectionoftheDepartmentofNeurology andNeurosurgery,UniversidadeFederaldeSa˜oPaulo,Sa˜oPaulo, Brazil, between February 2004 and July 2010. This evaluation consistedofadetailedclinicalhistory,neurologicalexamination, 1.5T brainMRI,neuropsychological and psychiatricevaluations, andpsychosocialassessments.Patientsalsounderwent3–6days of continuous video-EEG monitoring with 32-channel EEG recordings,withelectrodesplacedaccordingto10-10systemon the temporal lobes, plus bilateral sphenoidal electrodes. All patients had clear MRI findings consistent with unilateral MTS.17,18_{PatientswithunilateralMTSandadditional}

abnormali-ties besides brain atrophy detected by visual inspection were excluded.

Clinicalfeatureswereassessedretrospectivelyaccordingtoa specificprotocoldevelopedforthisstudy.Weregisteredageoffirst seizureaccordingtoTaylorinhisoriginalstudy.2_{Asinthisseries,}

ageattimeofthefirstepilepticeventbefore10yearsoflifewas recordedbygenderandside.

Natural history of TLE-MTS was also reviewed, including presenceandageofaninitialprecipitatinginjury(IPI),19_ageat

onsetanddurationofepilepsy.

2.2. Controls

Fifty age and gender-matched healthy control subjects (24 males),membersofthehospitalpersonnelwithnohistoryofhead injury or significant medical or psychiatric illnesses were submittedto1.5TbrainMRIunderconditionsidenticaltopatients. AllcontrolshadnormalMRIonvisualinspection.

TheEthicsCommitteeofourinstitution approvedthestudy, andinformedconsentwasobtained fromallparticipantsbefore theirinclusioninthisprotocol.

2.3. MRIdataacquisition

Patientswereretrospectivelyselectedduringa 6-yearperiod andasampleof120subjectsunderwentthesameMRIprotocol suitableforVBM.

Examinationofthebrainwasperformedinallsubjectswitha 1.5T MRI (Magnetom Sonata [Maestro Class] – Siemens AG, Medical Solutions, Erlangen, Germany)using aneight-channel headcoil. The followingsequences wereacquired: (1)sagittal images (T1-weighted spin echo); (2) coronal images [(T2-weightedfastspinecho;T2-weightedfluidattenuatedinversion recovery (FLAIR); T1-weighted inversion recovery)]; (3) axial images(T2-weightedFLAIR;T2-weightedgradientecho).

ForVBManalysistheconventionalsequenceperformedwas: sagittal T1-weighted gradient echo volumetric acquisition for multiplanarreconstruction (TR=2000ms,TE=3.42ms, flip an-gle=15, FOV=245mm, 1.0-mm slice thickness with no gaps, totaling160slices/slab,matrixsize=256256,NEX=1).

2.4. MRIdataprocessingandanalysis:VBM

Patients and controls were analyzed using VBM5 toolbox

(http://dbm.neuro.uni-jena.de), implemented in Statistical

Parametric Mapping (SPM5)11,20 _{and executed in Matlab 7.0}

(Mathworks, Sherborn, MA). The sagittal T1 DICOM files were converted to NIFTI-1 (http://nifti.nimh.nih.gov) format. The convertedfileswerethensegmentedintograyandwhitematter and normalized. Voxel values weremodulated by the Jacobian determinants derived from the spatial normalization.The final voxelresolutionafternormalizationwas1mm3_{.Theobtainedgray}

matter(GM)imageswerefinallysmoothedwithaGaussianfilter atthefullwidthatheightmaximumequalto8mmandenteredin statistical analysis.Additionally, globalGM,whitematter(WM) and cerebral spinal fluid volumesas well as total intracranial volumeswerecomputed usingthenative-space tissuemaps of eachsubject.

2.5. Statistics

2.5.1. Clinicalanalysis

Descriptiveanalysesofquantitativevariableswerereportedby meanandstandarddeviation(SD).

Priortoconductinganalyses,measuresweretestedfornormal distributionusingKolmogorov–Smirnovtest.Allcategorical and quantitativevariableswereassessedaccordingtosideandgender using Chi-square and Mann–Whitney test, respectively. Age at onsetoffirstevent,varyingbygenderandsideoflesion,wasalso analyzed.

Thelevelofstatisticalsignificancewassetatp<0.05.

2.5.2. VBManalysis

Inordertoinvestigatepossiblegenderandhemisphericgray matter volume (GMV) differences between males and females, rightandleft-TLEpatients,andcontrolgroup,weemployedthe General Linear Model (GLM) with age and brain volume as covariatesofnointerestforallVBManalyses.

Resultingclusterswerereportedassignificantatap<0.001

level,two-tailed,uncorrectedformultiplecomparisons.Asmall volumecorrection(SVC)wasapplied,whentherewasastricta priori hypothesis which was already implicated in the patho-physiology of TLE (hippocampus, amygdala, entorhinal and perirhinal cortices, thalamus)21–23 _{emerging from the}

whole-brain analyses. Since abnormalities in lateral temporal and frontocentral areas have been previously reported13,24 _these

structureswereconsideredasaprioriregions.Asaconsequence, GMV findings were corrected for multiple comparisons using SVC. We performed SVC placing a sphere with 5mm radius centered atthe localmaxima, whichwas equivalentto a volumeof 500mm3_{, witha thresholdof p<0.01, corrected}

for multiple comparisons using False Discovery Rate (FDR). Unpredicted findings were considered as significant only if they survived FDR correction for multiple comparisons (p<0.05).

3. Results 3.1. Demographics

VBMdataof120patientsaresummarizedinTable1.

3.2. Hemisphericdifferences

Both in comparison to controls and between sides, left-TLEpatientsexhibitedgreaterextensionofGMVreductions (Fig. 1). Right-TLE showed decreased GMV mainly in temporal areas, mesial and lateral, while reductions in left-TLE were found in widespread regions: temporal and occipital lobes,aswell as thalamus,cingulumandcerebellum

Table2

ResultsfromVBM:analysisofhemisphericdifferences.

GroupComparison MNIcoordinates Brainregion Z-Score ke

GMreductionsinRTLE Comparisontocontrols

(45 1714) Rightinsula 3.73 261

(35 24 12) Righthippocampusa _{5.65 663}

(4 196) Rightthalamus 4.51 1316

(49 1039) Rightprecentralgyrus 5.48 234 ( 49 1344) Leftprecentralgyrus 4.35 477 (38 51 36) Rightcerebellarhemisphere 4.55 4018 (58 53 11) Rightinferiortemporalgyrus 4.34 264 (57 4512) Rightsuperiortemporalgyrus 3.44 85 (56 4031) Rightsupramarginalgyrus 3.63 143 ( 49 60 40) Leftcerebellarhemisphere 3.77 691 GMreductionsinLTLE

Comparisontocontrols

( 13688) Leftsuperiorfrontalgyrus 4.36 218 ( 11 2611) Leftthalamus 4.81 2149

(2 170) Rightthalamus 3.68 2149

( 32 16 19) Lefthippocampusa _{4.53 201}

( 44 1640) Leftprecentralgyrus 4.90 718 ( 63 18 22) Leftinferiortemporalgyrus 4.43 255 ( 39 53 35) Leftcerebellarhemisphere 4.60 8502 (33 67 33) Rightcerebellarhemisphere 4.01 1319 ( 51 6413) Leftmiddletemporalgyrus 5.16 1122 ( 55 5536) Leftsupramarginalgyrus 4.42 362 ( 3 6831) Leftprecuneus 4.17 1578

( 6 892) Leftcuneus 5.87 3558

(7 856) Rightcuneus 5.32 3558

( 5 885) Leftcuneus 5.63 2173

(42 81 3) Rightinferioroccipitalgyrus 4.75 131 GMreductionsinLTLE

ComparisontoRTLE

( 32 16 19) Lefthippocampusa _{9.21 26656}

( 36 81) Leftinsula 4.51 26656

( 38 9 40) Leftinferiortemporalgyrus 4.67 26656 ( 25 3 14) Leftamygdala 5.02 26656 ( 55 62 4) Leftposteriorinferiortemporalgyrus 4.31 2657 ( 58 5 20) Leftmiddletemporalgyrus 3.79 26656 ( 55 5814) Leftposteriorsuperiortemporalgyrus 4.13 2657 ( 547 8) Leftsuperiortemporalgyrus 3.82 26656 ( 15 2711) Leftthalamus 4.07 26656 ( 14 584) Leftposteriorcingulum 3.73 392 (6 5928) Rightposteriorcingulum 3.53 142 ( 35 8033) Leftprecuneus 4.01 724

( 5 856) Leftcuneus 3.82 564

( 43 26 27) Leftfusiformgyrus 5.27 26656 GMreductionsinRTLE

ComparisontoLTLE

(34 21 13) Righthippocampusa _{11.05 24281}

(20 4 14) Rightamygdala 5.13 24281 (30 12 35) Rightperirhinalcortex 4.33 24281 (59 16 6) Rightsuperiortemporalgyrus 4.22 24281 (45549) Rightmiddlefrontalgyrus 4.01 165 (5898) Rightprecentralgyrus 4.09 24281 (4 351) Rightmedialfrontalgyrus 4.33 176 (31 31 25) Rightfusiformgyrus 3.85 24281 (63 2618) Rightpostcentralgyrus 4.55 1131 (61 19 17) Rightinferiortemporalgyrus 4.85 24281

(38 8 2) Rightinsula 5.28 24281

Brainregion,MNIcoordinatesofpeakeffect,significanceextent(SPM5-Zscore)andvoxelextentareindicated.AllresultsarereportedatpFDR<0.05(correctedformultiple comparisons).

a_{Forthesebrainareas,asmallvolumecorrection(SVC),withathresholdofpFDRcorrected<0.01.C=controls;GM=graymatter;LTLE=lefttemporallobeepilepsy;}

RTLE=righttemporallobeepilepsy.

Table1

ClinicalanddemographicaldataonthesubjectssubmittedtoVBManalysis. Numberof

patients,n(%)

Historyof IPI,n(%)

Age, ySD

Brain volume,mm3

AgeofIPI, ySD

Ageatonset ofepilepsy,ySD

Durationof epilepsy,ySD Totalgroup 120(100.0) 66(55.0) 35.09.46 1140.56131.96 2.683.96 9.959.54 22.9812.40 Male 57(47.5) 34(59.6) 33.609.38 1216.69127.88 3.115.25 11.6410.70 18.9212.60 Female 63(52.5) 32(50.8) 36.289.40 1072.4991.89 2.221.77 8.418.14 26.6411.08

p-Value** _{– 0.388 0.067 <0.001 0.215 0.049 <0.001}

LeftTLE 66(55.0) 36(54.5) 35.709.24 1143.04129.62 2.654.54 9.819.36 22.4112.98 RightTLE 54(45.0) 30(55.5) 37.279.57 1138.08134.31 2.713.19 10.119.84 23.6711.73

p-Value* _{– 0.884 0.007 <0.001 0.530 0.729 0.789}

Abbreviations:IPI=initialprecipitatinginjury;TLE=temporallobeepilepsy;VBM=voxel-basedmorphometry.

3.3. Genderdifferences

In comparison to gender-matched controls, male patients presented GMVreductions in thalamus and frontal gyri, while female exhibited reductions in temporal areas, thalamus and cerebellum(Table3).

Whenmaleandfemalepatients,separatedbysideofMTS,were compared to controls, both presented different areas of GMV abnormalities. In men, GMV reductions persisted in thalamus, frontalgyriandhippocampusbeingmoreextensiveinleft-TLE.In women with right-TLE, significant reductions were found in hippocampusandtemporalgyri,whileinleft-TLEgroupreductions involvedbilateraltemporalgyriandcerebellum.

When investigating gender effectin patients withTLE-MTS, similarpatternofresultswasobserved.Menpresentedreductions in frontal gyri, cingulum and thalamus, whereas temporal structures, especially amygdala, and also hippocampus were involvedinwomen(Fig.2).

4. Discussion

Evidenceofextratemporalinvolvementandevenhemispheric damageinTLE-MTShasbeenconfirmedbypreviousquantitative MRI studies.25,26 VBM series have analyzed the hemispheric damage in TLE, indicating widespread extrahippocampal GM abnormalitiesinunilateralMTS,withinvolvementofotherlimbic structures, suchas cingulum, insula and thalamus13 _and

para-hippocampalareas,27_{aswellasextralimbicareasincludingfrontal}

lobe,parieto-occipitalregionsandcerebellum.13,27,28_Inarecent

reviewofVBMstudiesinTLE,26brainregionswerefoundtobe significantlyreducedinvolumerelativetocontrols,althoughthere wasastrongasymmetricaldistributionofabnormalities.10

In our study, we confirmed the extension of the lesions observingthatbothgroups,leftandrightMTS,hadseveralGMV abnormalities,incomparisontocontrolsandalsointheanalysisof eachhemisphereseparately.Inaddition,thesereductionsinGMV were more widespread in left-TLE patients, involving not only distinct areas ipsilateral to the side of MTS, but also bilateral regions. In contrast, in right-TLE, GMV reductions were more restricted and confined to medial temporal lobe. Evidence for distinctneuronal networkdamagemorewidespreadinpatients withleft-sidedseizurefocushasalsobeenfoundbyothers.15,16,27

Ourresultsconfirmedalargerextentofstructural abnormali-ties in left-TLE, in accordance with previous VBM10,15,16,29 _and

corticalthicknessanalyses.30_{AssuggestedbyMRIstudyinhealthy}

individuals, temporofrontal networks maybe more extensively connected in the dominant hemisphere, likely due to their involvement in languagefunction.31 _{Such increasedanatomical}

connectivitymayleadtomoreintenseseizurepropagationinthe lefthemisphereandmoremarkedneuronallossinleft-TLE.

Furthermore,braindamagewouldbemorelikelytoaffectthe left hemisphere still undergoing rapid maturation in the first yearsoflife.AccordingtoTaylor’shypothesisageoffirstseizure variesinmenandwomenandinleftandrighthemispheredueto differentratesofmaturationbetweensexesandlaterality.There wouldbebiologicalbasisforthehighervulnerabilityofthemale brainandofthelefthemisphere.Cerebralmaturationwouldbe morerapidingirls,sothatboyswouldbeatagreaterriskfora

Fig.2.VBMresults:threeandtwodimensionalrepresentationsofgenderdifferencesinTLE-MTS.Comparisonbetweenmaleandfemalepatients.(a–d)GMVreductionsin men;(e–h)GMVreductionsinwomen.

longertime. A potentialseizure-producinginsult would affect the less functionally active side, the left hemisphere. This hypothesis, that the right hemisphere develops earlier than the left, is also confirmed by changes in cerebral blood flow measuredbySPECT.32

Toourknowledge,untilthepresentmoment,noVBMstudyhas investigatedgenderdifferencesinTLE-MTS.However,arecentDTI study in unilateral TLE-MTS demonstrated WM differences between gender, showingmore extensive changes in right-TLE andinfemalepatients.33

Table3

ResultsfromVBM:analysisofgender.

Groupcomparison MNIcoordinates Brainregion Z-score ke

Genderdifferencesbetweenpatientsandcontrols GMreductionsinmen

Comparisontocontrols

( 31562) Leftmiddlefrontalgyrusa _{4.05 245}

( 3 1713) Leftthalamusa _{3.60 341}

( 59 932) Leftprecentralgyrusa _{3.64 197}

( 50 833) Rightprecentralgyrusa _{4.13 152}

GMreductionsinwomen Comparisontocontrols

(310 47) Rightinferiortemporalgyrus 4.01 90 ( 47 1441) Leftprecentralgyrus 5.14 316

(4 203) Rightthalamus 3.62 247

(54 4812) Rightposteriorsuperiortemporalgyrus 4.23 353 (46 68 36) Rightcerebellarhemisphere 4.59 10828 ( 37 64 53) Leftcerebellarhemisphere 4.15 4559 GMreductionsinRTLEmen

Comparisontocontrols

( 31562) Leftmiddlefrontalgyrusa _{3.85 270}

(36 22 14) Righthippocampusa _{4.59 338}

(1 1411) Rightthalamusa _{3.51 130}

(16 92 14) Rightlingualgyrus 4.23 279 GMreductionsinLTLEmen

Comparisontocontrols

( 176615) Leftsuperiorfrontalgyrusa _{3.61 59}

( 511220) Leftinferiorfrontalgyrusa _{3.46 60}

( 32 16 18) Lefthippocampusa _{3.51 39}

( 52 6513) Leftposteriormiddletemporalgyrus 4.51 590 ( 6 2013) Leftthalamus 3.97 445

( 5 892) Leftcuneus 4.84 1886

GMreductionsinRLTEwomen Comparisontocontrols

(302 46) Rightinferiortemporalgyrusa _{5.24 640}

(23 20 16) Righthippocampusa _{4.16 553}

(11 285) Rightthalamus 4.01 318 (56 4612) Rightposteriorsuperiortemporalgyrus 3.68 202 (37 54 34) Rightcerebellarhemisphere 4.30 6617 GMreductionsinLTLEwomen

Comparisontocontrols

( 46 1542) Leftprecentralgyrus 5.03 313 ( 63 18 21) Leftmiddletemporalgyrus 3.71 118 ( 12 2810) Leftthalamus 3.68 144 (52 4911) Rightposteriorsuperiortemporalgyrus 4.00 183 ( 38 65 52) Leftcerebellarhemisphere 4.05 7528 (46 69 36) Rightcerebellarhemisphere 4.10 7156 Genderdifferencesbetweenpatients

GMreductionsinwomen Comparisontomen

(290 27) Rightamygdalaa _{4.37 835}

(310 46) Rightinferiortemporalgyrusa _{4.62 735}

(24 9 28) Righthippocampusa _{3.65 835}

( 31 3 26) Leftamygdalaa _{3.95 237}

( 301 44) Leftinferiortemporalgyrusa _{3.34 63}

(35 49 34) Rightcerebellarhemisphere 4.03 511 GMreductionsinmen

Comparisontowomen

( 185035) Leftsuperiorfrontalgyrusa _{4.05 123}

( 43239) Leftmedialfrontalgyrusa _{3.76 116}

(521522) Rightinferiorfrontalgyrusa _{4.32 380}

(0 169) Rightthalamusa _{3.64 177}

GMreductionsinRTLEmenb _{(224733) Rightsuperiorfrontalgyrus}a _{3.59 39}

(42727) Rightcingulategyrusa _{3.59 1027}

( 272847) Leftmiddlefrontalgyrusa _{3.81 182}

( 37 4246) Leftinferiorparietallobule 4.71 357 GMreductionsinLTLEmenb _{(531025) Rightinferiorfrontalgyrus}a _{3.73 150}

( 1 188) Leftthalamusa _{3.45 108}

GMreductionsinRTLEwomenb _{(281 26) Rightamygdala}a _{3.52 198}

(301 45) Rightinferiortemporalgyrusa _{5.36 1201}

( 285 43) Leftinferiortemporalgyrusa _{4.05 464}

GMreductionsinLTLEwomenb _{(300 26) Rightamygdala}a _{3.29 41}

(60 49 15) Rightposteriorinferiortemporalgyrus 3.61 129 ( 24 47 24) Leftcerebellarhemisphere 3.47 519 ( 33 1 27) Leftamygdalaa _{3.81 89}

(28 53 22) Rightcerebellarhemisphere 3.16 537 Brainregion,voxelextent,significanceextent(SPM5-Zscore)andMNIcoordinatesofpeakeffectareindicated.AllresultsarereportedatpFDR<0.05(correctedformultiple comparisons).

a_{Forthesebrainareas,asmallvolumecorrection(SVC),withathresholdofpFDRcorrected<0.01.C=controls;GM=graymatter;LTLE=lefttemporallobeepilepsy;}

RTLE=righttemporallobeepilepsy;W=women.

One marked gender difference found in our data was the distinctregionsofabnormalitiesinmenandwomen,whichwere concentratedinfrontalregionsinmen,andintemporalareasin women.Similarfindingswerefoundinapreviousinvestigationof sexualdimorphismin TLE-MTS usingPET.9 _{Male patientsmore}

oftenhadafrontallobehypometabolismipsilateraltotheseizure onset and a spread of epileptiform activity to this region. By contrast,femaleexhibitedhypometabolismandictalspreadtothe contralateral temporallobe. Data fromanotherPET studyhave demonstrated gender-specific predominance of extramesiotem-poralhypometabolisminmalepatientswithTLE-MTSrelatedto abnormalitiesoftemporalandfrontallobefunctions.7

Sexualdimorphisminthephenomenologyofseizureshasbeen alsohighlightedinTLE.Womenhavemoreisolatedandspecific typesofaurasthanmen,includingsexualmanifestations,ictalfear and affective auras.3–5,34 _{On the other hand, men present}

secondarily generalized tonic-clonic seizures more often than women, suggesting a greater seizure spread.4 _{Involvement of}

mesiotemporalstructuresinwomen,andfrontalpredominanceof anatomicaldamageinmen,asdemonstratedinthis VBMstudy, mightbethestructuralbasisofthesefindings.

Our study was focused on the extension of damage and anatomicalchangesaccordingtogenderandlateralizationin TLE-MTS, based initially on Taylor’s hypothesis. However, some limitationsshouldbepointedout.Variablessuchashandedness, lateralizationofspeechanddifferencesofIQbetweenpatientsand controlswerenotcontrolledinVBManalyses.

Thus,wemustconsiderthatthesevariablesmayinfluencethe degree of anatomical differences between genders and hemi-spheres, therefore configuring some limitations of the present study.

Further studies should include these important aspects to confirmwhethertheymayormaynotinfluencetheextensionof braindamageandgenderdifferencesinTLE-MTS.

5. Conclusion

Our study confirmedgender and hemisphericdifferences in TLE-MTS.Genderandlateralityareimportantfactorstodetermine nature and severity of brain damage. Lesions were more widespreadinlefthemisphere.Menandwomenshowdifferent areas of anatomical involvement. While structural damage in malesextendedmainlytofrontallobes,infemalesdamagewere mostly seen in temporal areas.Differential rates of maturation between genderand hemispheres maybethebasisfordistinct areasofanatomicaldamageinmenandwomen.

Acknowledgement

ThisstudywassupportedbyFundac¸a˜odeAmparoa` Pesquisado EstadodeSa˜oPaulo(FAPESP).

References

1.GeschwindN,GalaburdaAM.Cerebrallateralization,biologicalmechanisms, associations,andpathology:I.Ahypothesisandaprogramforresearch.Arch Neurol1985;42:428–59.

2.TaylorDC.Differentialratesofcerebralmaturationbetweensexesandbetween hemispheres.Lancet1969;2:140–2.

3.Re´millardGM,Andermann F, Testa GF, et al.Sexual ictalmanifestations predominateinwomenwith temporallobeepilepsy:afindingsuggesting sexualdimorphisminthehumanbrain.Neurology1983;33:323–30.

4.JanszkyJ,SchulzR,JanszkyI,EbnerA.Medialtemporallobeepilepsy:gender differences.JNeurolNeurosurgPsychiatry2004;75:773–5.

5.ChiesaV,GardellaE,TassiL,etal.Age-relatedgenderdifferencesinreporting ictalfear:analysisofcasehistoriesand reviewoftheliterature. Epilepsia

2007;48:2361–4.

6.BriellmannRS,Berkovic SF,JacksonGD.Menmaybemorevulnerableto seizure-associatedbraindamage.Neurology2000;55:1479–85.

7.NickelJ,JokeitH,WunderlichG,EbnerA,WitteOW,SeitzRJ.Gender-specific differencesofhypometabolisminmTLE:implicationforcognitiveimpairments.

Epilepsia2003;44_:1551–61.

8.DohertyMJ,RostadSW,KraemerDL,VosslerDG, HaltinerAM.Neocortical gliosis in temporal lobe epilepsy: gender-based differences. Epilepsia

2007;48:1455–9.

9.SavicI,EngelJrJ.Sexdifferencesinpatientswithmesialtemporallobeepilepsy.

JNeurolNeurosurgPsychiatry1998;65:910–2.

10.KellerSS,RobertsN.Voxel-basedmorphometryoftemporallobeepilepsy:an introductionandreviewoftheliterature.Epilepsia2008;49:741–57.

11.AshburnerJ,FristonKJ.Voxel-basedmorphometry–themethods.Neuroimage

2000;11:805–21.

12.KellerSS,WieshmannUC,MackayCE,DenbyCE,WebbJ,RobertsN.Voxelbased morphometryofgreymatterabnormalitiesinpatientswithmedically intrac-tabletemporallobeepilepsy:effectsofsideofseizureonsetandepilepsy duration.JNeurolNeurosurgPsychiatry2002;73:648–56.

13.BernasconiN,DuchesneS,JankeA,LerchJ,CollinsDL,BernasconiA. Whole-brain voxel-basedstatistical analysisofgray matterand whitematter in temporallobeepilepsy.Neuroimage2004;23:717–23.

14.BonilhaL,RordenC,AppenzellerS,CoanAC,CendesF,LiLM.Graymatter atrophy associated with duration of temporal lobe epilepsy. Neuroimage

2006;32:1070–9.

15.BonilhaL,RordenC,HalfordJJ,etal._{Asymmetricalextra-hippocampalgrey} matterlossrelatedtohippocampalatrophyinpatientswithmedialtemporal lobeepilepsy.JNeurolNeurosurgPsychiatry2007;78_:286–94.

16.RiedererF, Lanzenberger R, KayaM,PrayerD, Serles W,BaumgartnerC. Networkatrophyintemporallobeepilepsy.Avoxel-basedmorphometrystudy.

Neurology2008;71:419–25.

17.BerkovicSF,AndermannF,OlivierA,etal.Hippocampalsclerosisintemporal lobe epilepsy demonstrated by magnetic resonance imaging. Ann Neurol

1991;29:175–82.

18.JacksonGD,BerkovicSF,DuncanJS,ConnellyA.Optimizingthediagnosisof hippocampalsclerosisusingMRimaging.AmJNeuroradiol1993;14:753–62.

19.MathernGW,BabbTL,VickreyBG,MelendezM,PretoriusJK.The clinical-pathogenicmechanismsofhippocampalneuronlossandsurgicaloutcomesin temporallobeepilepsy.Brain1995;118:105–18.

20.AshburnerJ,FristonKJ.Unifiedsegmentation.Neuroimage2005;26:839–51.

21.BernasconiN,BernasconiA,CaramanosZ,AntelSB,AndermannF,ArnoldDL. Mesialtemporaldamageintemporallobeepilepsy:avolumetricMRIstudyof the hippocampus, amygdala and parahippocampal region. Brain

2003;126:462–9.

22.BonilhaL,RordenC,CastellanoG,CendesF,LiLM.Voxel-basedmorphometryof thethalamusinpatientswithrefractorymedialtemporallobeepilepsy. Neuro-image2005;25:1016–21.

23.LabateA,CerasaA,AgugliaU,MumoliL,QuattroneA,GambardellaA. Voxel-basedmorphometryofsporadicepilepticpatientswithmesiotemporal sclero-sis.Epilepsia2010;51_:506–10.

24.BernhardtBC,WorsleyKJ,BessonP,etal._{Mappinglimbicnetworkorganization} intemporallobeepilepsyusingmorphometriccorrelations:insightsonthe relationbetweenmesiotemporalconnectivityandcorticalatrophy.Neuroimage

2008;42:515–24.

25.MarshL,MorrellMJ,ShearPK,etal.Corticalandhippocampalvolumedeficits intemporallobeepilepsy.Epilepsia1997;38:576–87.

26.BriellmannRS,JacksonGD,KalninsR,BerkovicSF.Hemicranialvolumedeficits inpatientswithtemporallobeepilepsywithandwithouthippocampal sclero-sis.Epilepsia1998;39(11):1174–81.

27.BonilhaL,RordenC,CastellanoG,etal.Voxel-basedmorphometryrevealsgray matternetworkatrophyinrefractory medialtemporallobeepilepsy. Arch Neurol2004;61:1379–84.

28.KellerSS,MackayCE,BarrickTR,WieshmannUC,HowardMA,RobertsN. Voxel-basedmorphometriccomparisonofhippocampalandextrahippocampal ab-normalitiesinpatientswithleftandrighthippocampalatrophy.Neuroimage

2002;16:23–31.

29.CoanAC,AppenzellerS,BonilhaL,LiLM,CendesF.Seizurefrequencyand lateralizationaffectprogressionofatrophyintemporallobeepilepsy.Neurology

2009;73:834–42.

30.BernhardtBC,BernasconiN,ConchaL,BernasconiA.Corticalthicknessanalysis intemporallobeepilepsy.Reproducibilityandrelationtooutcome.Neurology

2010;74_:1776–84.

31.PowellHWR,ParkerGJM,AlexanderDC,etal._{Hemisphericasymmetriesin} language-related pathways:a combined functional MRIand tractography study.Neuroimage2006;32:388–99.

32.ChironC,JambaqueI,NabboutR,LounesR,SyrotaA,DulacO.Therightbrain hemisphereisdominantinhumaninfants.Brain1997;120:1057–65.

33.OguzKK,TezerI, SanverdiE,etal.Effectofpatientsexonwhitematter alterations inunilateral medial temporallobeepilepsy withhippocampal sclerosis assessed by diffusion tensor imaging. Am J Neuroradiol

2013;34:1010–5.

34.TothV,FogarasiA,KaradiK,KovacsN,EbnerA,JanszkyJ. Ictalaffective symptomsintemporallobeepilepsyarerelatedtogenderandage.Epilepsia