Major depressive disorder as a predictor of a worse seizure outcome one year after surgery in patients with temporal lobe epilepsy and mesial temporal sclerosis

Major

depressive

disorder

as

a

predictor

of

a

worse

seizure

outcome

one

year

after

surgery

in

patients

with

temporal

lobe

epilepsy

and

mesial

temporal

sclerosis

Gerardo

Maria

de

Arau´jo

Filho

*

,

Francinaldo

Lobato

Gomes

,

Lenon

Mazetto

,

Murilo

Martinez

Marinho

,

Igor

Melo

Tavares

,

Luı´s

Ota´vio

Sales

Ferreira

Caboclo

,

Elza

Ma´rcia

Targas

Yacubian

_,

_Ricardo

_Silva

_Centeno

a_{DepartmentofNeurologyandNeurosurgery,UniversidadeFederaldeSa˜oPaulo(UNIFESP),Sa˜oPaulo,Brazil}

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

1. Introduction

Epilepsysurgeryhasbecomeanimportanttreatmentoptionfor patientswithrefractorytemporallobeepilepsy(TLE),and30–40% of patients with TLE are surgical candidates. Current evidence suggests that epilepsy surgery is associated with a 60–70% remissionrateforlong-termepilepticsymptomsandsignificant improvements in quality of life (QOL).1,2 _{Data from previous} studieshavedemonstratedthat cortico-amygdalohippocampect-omy(CAH)isasafe,efficientsurgicalprocedureforpatientswith refractoryTLEandmesialtemporalsclerosis(TLE–MTS),oneofthe commonesttypes ofsurgically remediableepilepticsyndromes.

TLE–MTS compromises the primary structures of the limbic system,particularlythehippocampusandamygdala.3–7

Previous research has demonstrated a prevalence rate of psychiatric comorbidity of 20–40% in TLE–MTS patients, rising to70%inpatientswithrefractoryformsofthisseizuredisorder.8–13 Mood disorders are the most common (24–74%), followed by anxiety (10–25%), psychotic disorders (2–9%) and personality disorders(1–2%).6–11_{Giventhehighprevalenceofpre-surgicalPDs} inthispopulation,thepsychiatricoutcomesofsuchpatientsafter epilepsysurgeryhaveattractedtheparticularinterestof research-ers.6,7_{Moreover,theassociationbetweenpre-surgicalPDsanda} worse post-surgical seizure outcome in patients has been increasinglyrecognized.14–17

A number of risk factors for a poor post-surgical seizure outcomehavebeenrecognized.Theseincludealifetime psychiat-richistory,ahistory ofsecondarilytonic–clonicseizures, along duration of the seizure disorder, larger resections of mesial temporalstructuresandthepresenceofneuropathologic abnor-malities.14–17 _{However, there are a number of methodological} ARTICLE INFO

Articlehistory:

Received5January2012 Receivedinrevisedform4July2012 Accepted5July2012

Keywords:

Temporallobeepilepsy Mesialtemporalsclerosis Epilepsysurgery Psychiatricdisorders Seizureoutcome

ABSTRACT

Purpose:Theassociationbetweenpre-surgicalpsychiatricdisorders(PDs)andworseseizureoutcomein patientswithrefractoryepilepsysubmittedtosurgeryhasbeenincreasinglyrecognizedintheliterature. Thepresentstudyaimedtoverifytheimpactofpre-andpost-surgicalPDonseizureoutcomeinaseries ofpatientswithrefractorytemporallobeepilepsyandmesialtemporalsclerosis(TLE–MTS). Method: Datafrom115TLE–MTSpatients(65females;56.5%)whounderwent cortico-amygdalohip-pocampectomy(CAH)wereanalyzed.Pre-andpost-surgicalpsychiatricevaluationswereperformed usingDSM-IVandILAEcriteria.TheoutcomesubcategoryEngelIAwasconsideredascorrespondingtoa favorableprognosis.Amultivariatelogisticregressionmodelwasappliedtoidentifypossibleriskfactors associatedwithaworseseizureoutcome.

Results:Pre-surgical PDs, particularly major depressive disorder (MDD), anxiety and psychotic disorders,werecommon,beingfoundin47patients(40.8%).Fifty-sixpatients(48.7%)wereclassified ashavingachievedanEngelIAoneyearafterCAH.Accordingtothelogisticregressionmodel,the presence of pre-surgical MDD (OR=5.23; p=0.003) appeared as the most important risk factor associatedwithanon-favorableseizureoutcome.

Conclusion:Althoughepilepsysurgerymaybethebesttreatmentoptionforpatientswithrefractory TLE–MTS,ourfindingsemphasizetheimportanceofperformingadetailedpsychiatricexaminationas partofthepre-surgicalevaluationprotocol.

ß_{2012BritishEpilepsyAssociation.PublishedbyElsevierLtd.Allrightsreserved.}

*Correspondingauthorat:GerardoMariadeAraujoFilhoRuaBotucatu,740– VilaClementino,CEP:04023-900,Sa˜oPaulo,SP,Brazil.Tel.:+551155391792; fax:+551155493819.

E-mailaddress:filho.gerardo@gmail.com(G.M.deArau´joFilho).

1_{GMAFandFLGcontributedequallytothiswork.}

ContentslistsavailableatSciVerseScienceDirect

Seizure

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

concerns about thesestudies: differenttypes and etiologies of epilepsy have been analyzed together, patients submitted to differentsurgical procedures were analyzed together, different pre-andpost-surgicalpsychiatricevaluationshavebeenused,the psychiatric assessments were solely based on dimensional instruments,analyses have only included a limited number of clinicalandsociodemographicvariablesandstudieshaveincluded differentfollow-upperiodsaftersurgery.18_{Thesemethodological} problemshave limitedourunderstanding oftheriskofsurgical procedureswithrespecttopost-surgicalPDsofar.18_Thepresent studyaimedtoverifytheriskof pre-andpost-surgical PDand otherclinicalandsociodemographiccharacteristicsin predispos-ingindividualstoaworseseizureoutcomeoneyearaftersurgery inahomogeneousseriesofpatientswithrefractoryTLE–MTSwho weresubmittedtoCAH.

2. Methods

2.1. Subjects

AllTLE–MTSpatientswerefollowedintheoutpatientclinicofa tertiarycenter(EpilepsySectionoftheUniversidadeFederaldeSa˜o Paulo,Brazil)fromJanuary2003toDecember2011andunderwent video-electroencephalographic(VEEG) monitoringand a psychi-atricevaluation.Inclusioncriteriawerepatientswhowereabove 18yearsold,hadanelectroclinicaldiagnosisofTLEbasedonILAE criteria,19_{hadbeensubmittedtoCAHandhadbeensubjectedto} follow-up for one year. All participants presented clear MRI findings of unilateral MTS and concordant interictal and ictal electroencephalographic(EEG)data. Exclusioncriteriawere the presenceofotherclinicalorneurologicaldiseasesbesidesepilepsy, cognitiveimpairmentsprecludingevaluations,bilateralMTSand/ or age younger than 18 years old. All patients gave written informedconsenttoreviewtheirmedicalrecords.

Data from 115 TLE–MTS patients (65 females; 56.5%) were includedinthestudy.The meanpatientage,meanduration of epilepsy and mean age at epilepsy onset were 36.910.77, 27.112.14and9.58.1years,respectively.MTSoccurredmore frequently on the left side (75 patients; 65.2%). Background asymmetry on pre-surgical EEG was observed in 31 patients (26.9%)andwasduetoaslowingoftheaffectedtemporalregions (19patients;61.3%),aslowingoftheaffectedcerebralhemispheres(9 patients;29.0%) or to absenceof physiological alpha and/or beta rhythms in the affected cerebral hemispheres (3 patients; 9.7%). Twenty-ninepatients(25.2%)hadahistoryofapotentiallycausative braininjury,withfebrileseizuresbeingthemostfrequent(25cases; 86.2%).Allpatientshadusedatleasttwoormoreantiepilepticdrugs (AEDs) prior to surgery. carbamazepine (CBZ) was the most frequently used AED, prescribed to 75 patients (65.3%), followed byphenytoin (PHT) (49 patients;42.6%), phenobarbital (PB) (34; 29.5%),topiramate(TPM)(24;21.2%)andoxcarbazepine(OXC)(12; 10.4%).Benzodiazepines(BZDs), particularlyclobazam(CLB),were the most common adjunctive drugs and were prescribed to 49 patients(42.6%).

2.2. Procedures

Patientsunderwent2–6daysofcontinuousVEEGmonitoring with 32-channel EEG recording, with electrodes, including sphenoidalelectrodes,placedaccordingtothe10–10systemon thetemporallobe.MTSwasdeterminedtobepresentifatrophy,an increasedT2-weightedsignal,adecreasedT1-weightedsignaland adisruptedinternalstructureofthehippocampuswerepresent, accompaniedbyatrophyoftheamygdalaand/ortemporal pole signal alteration on visual inspection of the MRI pictures. The epileptogeniczonewasdeterminedbypredominantlyipsilateral

interictalepileptiformdischarges (80%cutoff)and seizureonset recorded during prolonged VEEG monitoring. The association betweenMTSsideandthefrequencyofinterictalsharpwaveson pre-surgical EEGwas also analyzed,and a cutoff of >80% was considered for predominantly ipsilateral activity. Asymmetric backgroundactivitywasdefined bythepresence ofslowing or absenceofphysiologicalalphaand/orbetarhythmsintheaffected temporallobesorcerebralhemispheresofbackgroundactivityon pre-surgical EEG. Refractoriness to medical treatment was consideredpresentifseizurespersistedaftertheutilizationofat leasttwofirst-linemedicationsforpartialseizuresatthehighest tolerateddoses foratleast 6months.The potentiallycausative braininjuriesweredefinedastheoccurrenceofseverecerebral eventsinthefirstyearoflifebeforetheappearanceofepilepsythat required medical intervention and/or hospitalization. Febrile seizures, meningoencephalitis, headtrauma or severeperinatal hypoxiawereconsideredpotentiallycausesofbraininjury.

Thesurgicalprocedureconsistedofanenblocresectionofthe superior, middle, inferior temporal and fusiform gyri, with a posteriorlimitof4.5cmfromthetipofthetemporallobe.After opening the temporal horn, the mesial temporal structures (hippocampus,amygdalaandparahippocampalgyrus)werealso resected.3–5 _{Engel’s classification was utilized to measure the} patients’ seizure outcomes one year after CAH. Only the subcategoryEngel IA(completely seizure-free)wasdefinedasa favorableprognosis.20_{AlloftheEngelsubcategoriesaredescribed} inTable1.

2.3. Psychiatricevaluation

Asinglepsychiatrist(GMAF)conductedtheclinicalinterviews forthisstudyusingtheDiagnosticandStatisticalManualofMental Disorders(DSM-IV)Axis Icriteria.21_{Becauseeachpatientcould} havemorethanoneAxisIpsychiatricdiagnosis,thenumber of patients diagnosed and all comorbidPDs diagnosed wereboth considered in the analysis. The presence of other specific psychiatric diagnoses commonly associated with epilepsy, not coveredbytheDSM-IVbutwell-describedintheliterature,suchas interictaldysphoricdisorder(IDD)andthepsychosesofepilepsy, wereevaluatedusingtheILAE criteria.These criteriawerealso

Table1

Engel’sclassificationofpost-surgicaloutcome.

ClassI:Freeofdisablingseizuresa

A.Completelyseizure-freesincesurgery

B.Nondisablingsimplepartialseizuresonlysincesurgery

C.Somedisablingseizuresaftersurgery,butfreeofdisablingseizures foratleasttwoyears

D.GeneralizedconvulsionswithAED*_{discontinuationonly}

ClassII:Raredisablingseizures(‘‘almostseizure-free’’) A.Initiallyfreeofdisablingseizuresbuthasrareseizuresnow B.Raredisablingseizuressincesurgery

C.Morethanraredisablingseizuressincesurgery,butrareseizures forthelasttwoyears

D.Nocturnalseizuresonly

ClassIII:Worthwhileimprovementb

A.Worthwhileseizurereduction

B.Prolongedseizure-freeintervalsamountingtogreaterthanhalf thefollow-upperiod,butnot<2years

ClassIV:Noworthwhileimprovement A.Significantseizurereduction B.Noappreciablechange C.Seizuresworse

a_{Excludesearlypost-surgicalseizures(firstfewweeks).}

b _{determinationof‘‘worthwhileimprovement’’requiresquantitativeanalysisof}

additional data,suchasthepercentseizurereduction, cognitivefunction and qualityoflife.

utilized todifferentiate postictal psychosis(PIP) frominterictal psychosis(IIP).22_{Dataconcerningthelifetimehistoryof} psychiat-ric treatment, defined as any past treatment with psychiatric drugs, were collected from patients during the first clinical interview.Information about familyhistory ofepilepsy andPD was also obtained from patients through broad questions regardingwhetheranyfirst-degreerelativewasintreatmentfor epilepsyand/oranyPDatthemomentoftheclinicalinterview.All patientsweresubmittedtoapre-surgicalandatleastone post-surgicalpsychiatricevaluationwithinthefirstyearaftersurgery, andthelatestpsychiatricevaluationswithinthefirstyearafter surgery were considered toevaluate the possible remission of psychiatric symptoms after surgery. Due to ethical issues, all patients with pre-surgical, post-surgical and/or de novo PD receivedpsychiatricfollow-upaftersurgery.

2.4. Statistics

Statisticalanalyseswereperformedusingversion10.0ofthe StatisticalPackageforSocialSciencessoftware(SPSS10.0,Chicago, Illinois).Patientsweredividedintotwogroups,namelythosewith or without a favorable post-surgical outcome one year after surgery.Bivariatestatistical analyseswereperformedusing the mostappropriatestatisticaltest(chi-square(

x

Student’sttestforunequalvariances).Amultivariateanalysiswas performedusinga logisticregressionmodeltoidentifypossible clinicalandsocio-demographicpredictorsofpost-surgicalPD.To preventmulticollinearity,apreviousprocedurefortheanalysisof thevariablecorrelationmatrixwasfollowedinwhichthevariable withthelower correlationvaluewas not utilized in theinitial model in the case where the correlationof two variables was higherthan0.6.Afterthispreviousanalysis,thevariablesincluded in the initial model were gender, age, presence and type of potentiallycausesofbraininjury, lifetimehistoryofpsychiatric treatment,presenceofpre-surgicalPD,familyhistoryofepilepsy andPD,epilepsyduration,ageatepilepsyonset,numberandtypes ofAED,lateralityofMTS,pre-surgicalEEGdataandthepresenceof afavorableseizureoutcome(EngelIA).Thevariable‘‘pre-surgical PD’’presented the higher variance inflation factor (VIF) (2.81), demonstratinganadequateadjustmentoftheinitialmodel.The oddsratio(OR)wascalculatedforsignificantriskfactors.Apvalue <0.05wasconsideredsignificant.

3. Results

Pre-surgicalPDswerediagnosedin47patients(40.8%).Major depressivedisorder(MDD),whichrepresentedallofthediagnoses ofdepressioninthepresentseries,wasthemostcommonPDand was diagnosed in 27 patients (23.4%). Eleven patients (9.5%) presentedwithanxietydisorders;amongthem,ninepatientshad generalizedanxietydisorder(GAD),andtwopatientshadsocial phobia. Psychotic disorders were observed in seven patients (6.0%):PIPinfourpatients(3.4%)andIIPinthreepatients(2.6%). FourpatientsfulfilledthecriteriafortwoAxisIdisorders. Post-surgicalPDswereobservedin31patients(26.9%).Again,MDDwas themostcommonPDandwasdiagnosedin14patients(12.1%). Tenpatients (8.6%)presentedwithgeneralizedanxietydisorder (GAD),andIIPwasdiagnosedineightpatients(6.9%).Twopatients fulfilled the criteria for two Axis I disorders. Twenty-seven patients, whorepresented54%of thepatients withpre-surgical PD,reportedacompleteremissionofpsychiatricsymptomsafter surgery.Symptom remissionoccurredamongpatients withIDD (100%), PIP(100%), MDD(59.2%)and anxietydisorders(36.3%). However,11patients(9.6%)developeddenovoPDaftersurgery. Fivepatientspresentedwithpsychosis(45.4%),threepatientswith MDD(27.3%)andthreepatientswithGAD(27.3%).Fluoxetinewas the most commonly prescribed antidepressant drug prior to epilepsysurgery(prescribedto41patients),followedbysertraline (17).Risperidonewasthemostcommonlyusedantipsychoticdrug (6patients),followedbyhaloperidol(4patients)andolanzapine(2 patients).Psychiatricdrugswerediscontinuedinallpatientswho hadexperiencedcompleteremissionofsymptomsoneyearafter surgery.Fifty-sixpatients(48.7%)wereclassifiedasEngelIAone yearaftersurgery,whereastherewere27(23.4%)EngelIB/IC/ID,21 (18.2%)EngelII,9(7.9%)EngelIII,and2(1.8%)EngelIVpatients. Pre-surgical PDoccurred in 47 patients (40.8%), whereas post-surgicalPDoccurredin31patients(26.9%).Thenumberandtypes ofpre-surgicalPDobservedinbothgroups(EngelIAversus non-EngelIA)aredescribedinFig.1.

Statistical analyses regarding the differences in clinical and sociodemographic characteristics between groups were per-formed. The group with a non-favorable seizure outcome presented a higherincidenceof pre-surgicalPD (p=0.002) and pre-surgical MDD (p=0.001), as well as more asymmetric backgroundactivityonEEG(p=0.003).Nodifferenceswerefound

betweengroupsregardingthenumberortypeofAEDsused,orany other clinical or sociodemographic variables. The clinical and demographicdataofbothgroupsaresummarizedinTable2.

Inamultivariatelogisticregressionmodel(sensitivity:76.5%; specificity: 69.6%; positive predictive value: 71.2%; negative predictivevalue:66.5%;areaunderthecurve0.724),thepresence ofanypre-surgicalPDwasassociatedwithahigherriskforaworse outcome(OR=3.49;p=0.002)intheinitialmodel.However,when psychiatricdiagnoseswereanalyzedseparately,onlythediagnosis ofMDDpersistedasstatisticallysignificant(OR=5.23;p=0.003),

whereas the other PDs combined (except MDD) were not

significant (OR=1.57; p=0.42). The presence of asymmetric activity on pre-surgical EEG also appeared as a risk factor (OR=3.81;p=0.001).Thepresenceofpost-surgical PDwasnot associated with a worse seizure outcome (OR=1.67; p=0.32).

Table3showstheresultsofthefinaladjustedmodel.

4. Discussion

Theaimofthisstudywastoverifytheriskofpre-and post-surgicalPD and other clinical and sociodemographic factors in predisposingindividualstoaworseseizureoutcomeoneyearafter surgery.Weexamineda homogeneousseriesofpatientswitha specificandcommonepilepsysyndrome(TLE–MTS)whowereall submittedtothesamesurgicalprocedure(CAH)andfollowed-up oneyearaftersurgery.Pre-andpost-surgicalpsychiatric evalua-tions used the same diagnostic criteria based on the modern psychiatricnosography.21,22_{Accordingtothepresentfindings,the} pre-surgicaldiagnosisofMDDandthepresenceofbackgroundEEG asymmetry appeared as risk factors associated with a non-favorableseizureoutcome.

Inrecentyears,surgeryhasbecomeaveryimportanttreatment optionforpatientswithrefractoryTLE–MTS, andCAHhasbeen showntobeasafeandefficientsurgicalprocedure.1–5_However, giventhiselevated prevalenceofpre-surgicalPD,someauthors have highlightedtherelatively highrisk of post-surgical PD in patientssubmittedtoepilepsysurgery,whereasotherauthorsdo not share this concern.18,23–26 _{In addition,recent studies have} shown a correlation between the risk of a worse post-surgical seizureoutcomeandsomepre-surgicalclinicaland sociodemo-graphiccharacteristics.14–18 _{Previousstudieshavereported that} pre-surgicalPD,aswellasalifetimehistoryofdepression,predicts a worse post-surgical seizure outcome among patients with refractory TLE. The most important hypothesis explains this observation by positing that pre-surgical PD, and particularly MDD,maybeamarkerofmorediffusecerebraldisease.15–17_This hypothesisstressesthebidirectionalityoftheassociationbetween

depression and postoperative seizure status.15–17,26 _{In keeping} withthis,themajorityofcaseserieshavereportedanassociation between the absence of post-surgical PD and better surgical outcome.23–27

In the present study, a statistically significant association between the absence of pre-surgical PD and a favorable post-surgicalseizureoutcomewasfound.Inaddition,thepre-surgical diagnosisofMDDwasassociatedwithahigher riskofa worse seizureoutcomeinthelogisticregressionmodel.Thesefindings areconcordantwithrecentdatafromotherstudiesandsupportthe currenthypothesisregardingtheassociationbetweenpre-surgical PD and seizure outcome.15–17 _{In accordance with the current} literature,thepresenceofepileptiformdischargescontralateralto theepileptogeniczone,aswellasasymmetricbackgroundactivity onpre-surgicalEEG,havebeenconsideredriskfactorsforaworse seizureandpsychiatricpost-surgicaloutcome,whilethepresence ofinterictaldischargesexclusivelyipsilateraltotheepileptogenic zone hasbeenconsidered tobea protectivefactor against the appearanceofpost-surgicalPD.14,16,26_{Inthisstudy,thepresenceof} abackgroundEEGasymmetry,whichhasbeenconsideredtobea possiblepredictorofamorediffusecerebraldisease,appearedasa riskfactorassociatedwithanunfavorableseizureoutcome.14,16,26 Thereareimportantlimitationsofthepresentstudy.Wecould not evaluate the prevalence of other PDs not covered by the diagnosticcriteriautilizedinthisstudy(DSM-IVandILAE),suchas attention-deficit hyperactivity disorder (ADHD), which is often observedamongpatientswithepilepsy.28_{Moreover,becauseonly} categoricalcriteriawereutilizedandnodimensionalinstruments were applied, we could not assess the impact of CAH on the psychiatricsymptomsofthosepatientswithapre-surgicalPDthat persistedaftersurgery.Furthermore,becausethestudyfocusedon AxisIPDs,wedidnotevaluatetheimpactofCAHonpersonality disorders(AxisIIPDs).Duetoincompleteand/orunreliabledata,

Table2

Clinicalanddemographicdataofpatientswithtemporallobeepilepsyandmesialtemporalsclerosissubmittedtoanteriortemporallobectomy.

Clinical/demographicdata EngelIA Non-EngelIA p

Numberofpatients(%) 56(48.7) 59(51.3) –

Ageatsurgery(meanSD) 35.89.6 36.710.1 0.82

Femalegender(%) 28(50.0) 37(62.7) 0.43

Ageatepilepsyonset(meanSD) 10.89.2 9.17.4 0.38

Yearsofepilepsyatsurgery(meanSD) 23.912.3 24.111.8 0.62

Lifetimepsychiatrictreatment(%) 13(23.2) 18(30.5) 0.71

Familyhistoryofepilepsy(%) 12(24.5) 24(36.4) 0.22

Familyhistoryofpsychiatricdisorders(%) 6(10.7) 12(20.3) 0.12

Presenceoffebrileseizures(%) 10(17.8) 15(25.4) 0.26

Presenceofleft-sidedMTS(%) 31(55.3) 44(74.5) 0.32

Presenceofpre-surgicalPD(%) 13(23.2) 37(62.7) 0.002*

Presenceofpost-surgicalPD(%) 11(22.4) 20(30.3) 0.40

AsymmetricbackgroundEEGactivity(%) 7(12.5) 24(40.6) 0.003*

ContralateralslowwavesonEEG(%) 13(23.2) 17(28.8) 0.89

ContralateralsharpwavesonEEG(%) 15(26.7) 19(32.2) 0.48

MTS:mesialtemporalsclerosis;SD:standarddeviation;PD:psychiatricdisorders;EEG:electroencephalogram.

*_p<0.05.

Table3

Logisticregressionresultsofthefinaladjustedmodel.

Riskfactors Oddsratio p>z

FamilyhistoryofPDs 2.36 0.12

Presenceoffebrileseizures 4.21 0.07

Lifetimepsychiatrictreatment 1.62 0.23 Pre-surgicalPDs(exceptdepression) 1.56 0.38

Post-surgicalPDs 1.67 0.32

Pre-surgicalMDD 5.23 0.003*

AsymmetricbackgroundEEGactivity 3.81 0.001*

MDD: major depressive disorder; PDs: psychiatric disorders; EEG: video-electroencephalogram.

* _p

we could not evaluate the possible associations between pre-surgical seizure frequency and seizure outcomesafter surgery. Finally, ourfindings arebased on a relatively smallnumber of patients and a relatively short follow-up period. However, our findingsareconcordantwiththepreviousliteratureandarebased onarelativelyhomogeneouspopulationofTLE–MTSpatients.

Toconclude,thepsychiatricissuesthatneedtobeconsideredin surgicalcandidatesarecomplexandincludetheriskofdeveloping post-surgicalPDsandpoorerpost-surgicalseizureoutcome.15–17 Inviewoftherelativelyhighpsychiatriccomorbidityobservedin surgical candidates and its potential negative impact on post-surgicalseizureoutcomes,patientswithTLE–MTSrequirecareful pre- and post-surgical psychiatric evaluation.15–17 _{As many} questions remain, further prospective studies with longer fol-low-upperiodsarerequired.18

Acknowledgments

ThisworkwassupportedbyCAPESandFAPESPfromBrazil.

References

1.HildebrandtM,SchulzR,HoppeM,MayT,EbnerA.PostoperativeroutineEEG correlates with long-termseizure outcome after epilepsysurgery. Seizure 2005;14:446–51.

2.JanskyJ,JanskyI,SchulzR,HoppeM,BehneF,PannekHW,et al. Temporallobe epilepsywithhippocampalsclerosis:predictorsforlong-termsurgical out-come.Brain2005;128:395–404.

3.FalconerMA,SerafetinidesEA.Afollow-upstudyofsurgeryintemporallobe epilepsy.JournalofNeurologyNeurosurgeryandPsychiatry1963;26:154–65. 4.WiebeS,BlumeWT,GirvinJP,EliasziwM.Fortheeffectivenessandefficacyof

surgeryfortemporallobeepilepsystudygroup.Arandomized,controlledtrial of surgery for temporal lobe epilepsy. New England Journal of Medicine 2001;345:311–8.

5.GuarnieriR,WalzR,HallakJEC,CoimbraE,AlmeidaE,CescatoMP,et al. Do psychiatriccomorbiditiespredictpostoperativeseizureoutcomeintemporal lobeepilepsysurgery?Epilepsy&Behavior2009;14:529–34.

6.FoongJ,FlugelD.Psychiatricoutcomeofsurgeryfortemporallobeepilepsyand presurgicalconsiderations.EpilepsyResearch2007;75:84–96.

7.PintorL,BaillesE,Ferna´ndez-EgeaE,Sa´nchez-GistauV,TorresX,Carren˜oM, et al. _{Psychiatricdisordersintemporallobeepilepsypatientsoverthefirstyear} aftersurgicaltreatment.Seizure2007;16_:218–25.

8.Tellez-ZentenoJF,WiebeS.Prevalenceofpsychiatricdisordersinpatientswith epilepsy:whatwethinkweknowandwhatweknow.In:KannerAM,Schachter S,editors.Psychiatriccontroversiesinepilepsy.SanDiego:AcademicPress;2008 .p.1–18.

9.GaitatzisA,TrimbleMR,SanderJW.Thepsychiatriccomorbidityofepilepsy. ActaNeurologicaScandinavica2004;110:207–20.

10.DevinskyO.Psychiatriccomorbidityinpatientswithepilepsy:implicationsfor diagnosisandtreatment.Epilepsy&Behavior2003;4:2–10.

11.DeAraujoFilhoGM,RosaVP,LinK,CabocloLO,SakamotoAC,YacubianEMT. Psychiatriccomorbidityinepilepsy:astudycomparingpatientswithmesial temporal sclerosis and juvenile myoclonic epilepsy. Epilepsy & Behavior 2008;13_:196–201.

12.DeAraujoFilhoGM,MazettoL,MacedoJS,CabocloLO,YacubianEMT. Psychi-atriccomorbidityinpatientswithtwoprototypesoffocalversusgeneralized epilepsysyndromes.Seizure2011;20:383–6.

13.SwinkelsWAM,BoasWE,KuykJ,vanDyckR,SpinhovenP.Interictaldepression, personalitytraitsandpsychologicaldissociationinpatientswithtemporallobe epilepsy(TLE)andextra-TLE.Epilepsia2006;47:2092–103.

14.AnhouryS,BrownRJ,KrishnamoorthyES,TrimbleMR.Psychiatricoutcome followingtemporallobectomy:apredictivestudy.Epilepsia2000;41:1608–15. 15.KannerAM.Depressioninepilepsy:prevalence,clinicalsemiology,pathogenic

mechanismsandtreatment.BiologicalPsychiatry2003;54:388–98.

16.KannerAM,ByrneR,SmithMC,BalabanovAJ,FreyM.Doesalifetimehistoryof depressionpredictaworsepostsurgicalseizureoutcomefollowingatemporal lobectomy?AnnalsofNeurology2006;10:19.

17.KannerAM.Shouldapsychiatricevaluationbeincludedineverypre-surgical work-up?In:KannerAM,SchachterS, editors.Psychiatriccontroversiesin epilepsy.SanDiego:AcademicPress;2008.p.239–54.

18.MacrodimitrisS,ShermanEMS,FordeS,Tellez-ZentenoJF,MetcalfeA, Her-nandez-RonquilhoL,et al. Psychiatricoutcomesofepilepsysurgery:a system-aticreview.Epilepsia2011;52:880–90.

19.CommissiononClassificationandTerminologyoftheInternationalLeague AgainstEpilepsy.Proposalforrevisedclassificationofepilepsiesandepileptic syndromes.Epilepsia1989;30_:389–99.

20.EngelJrJ,VanNessPC,RasmussenTB,OjemannLM.Outcomewithrespectto epilepticseizures.In:EngelJrJ,editor.Surgicaltreatmentoftheepilepsies.New York:RavenPress;1993.

21.AmericanPsychiatricAssociationDiagnosticandstatisticalmanualformental disordersDSM–IV(TextRevision).4thed.,Washington;2004.

22.KrishnamoorthyES,TrimbleMR,BlumerD.Theclassificationof neuropsychi-atricdisordersinepilepsy:aproposalbytheILAEcommissionon psychobiol-ogyofepilepsy.Epilepsy&Behavior2007;10:349–53.

23.BlumerD,WakhluS,DaviesK,HermannB.Psychiatricoutcomeoftemporal lobectomyforepilepsy:incidenceandtreatmentofpsychiatriccomplications. Epilepsia1998;39:478–86.

24.AltshulerL,RauschR,DeIrahimS,KayJ,CrandallP.Temporallobeepilepsy, temporallobectomy and majordepression.Journalof Neuropsychiatryand ClinicalNeurosciences1999;11:436–43.

25.ReuberM,AndersenB,ElgerCE,HelmstaedterC.Depressionandanxietybefore andaftertemporallobeepilepsysurgery.Seizure2004;13:129–35. 26.DevinskyO,BarrWB,VicreyBG,BergAT,BazilCW,PaciaSV,et al. Changesin

depression and anxiety after resective surgery for epilepsy. Neurology 2005;65:1744–52.

27.MetternichB,WagnerK,BrandtA,KraemerR,BuschmannF,ZentnerJ,et al. Preoperativedepressivesymptomspredictpostoperativeseizureoutcomein temporalandfrontallobeepilepsy.Epilepsy&Behavior2009;16_:622–8. 28.Gonzales-HeidrichJ,DoddsA,WhitneyJ,MacMillanC,WaberD,FaraoneSV,