Diagnostic
decision-making
after
a
first
and
recurrent
seizure
in
adults
Jessica
Askamp
a,*
,
Michel
J.A.M.
van
Putten
a,baDepartmentofClinicalNeurophysiologyatMIRAInstituteforBiomedicalEngineeringandTechnicalMedicine,UniversityofTwente,Enschede,TheNetherlands bDepartmentofClinicalNeurophysiology,MedischSpectrumTwente,Enschede,TheNetherlands
1. Introduction
When an adultpresentsto theemergency department(ED) afterafirstseizure,animportantquestioniswhetherornotthere isanincreasedrisk ofseizurerecurrence.1After anunprovoked
firstseizure,symptomaticetiologyandepileptiformEEGactivity are the two mostconsistent predictors of seizure recurrence.2 Therefore,MRIandroutine(20–30min)EEGincluding hyperven-tilation and photic stimulation, are both part of the standard diagnosticapproachin first-seizurepatients. Thisstudy specifi-callyaddressestheroleofEEGinfirst-seizurediagnosis.
The estimated probability of seizure recurrence aftera first seizure in adults withepileptiform EEGabnormalities is 49.5%, comparedtoonly27.4%inindividualswhoseEEGsarecompletely normal.3Still,thevalueofaroutineEEGafterafirstseizurehas been debated.4 A normal routine EEG does not exclude the
presenceofaseizuredisorder.Furthermore,presenceof epilepti-form activitywasreported inonly 8–50% offirst-seizure adult patients.3RoutineEEGsarethereforeoftenrepeated,orfollowed
byasleep-deprivedEEG,asthismayincreasesensitivity.5,6Still,
thereareepilepsypatientsinwhomrepeatedEEGsdonotshow anyepileptiformabnormality.Themainreasonmaybethatscalp electrodes sample only one-third of the cortex. Thislimits the sensitivity forIEDs arisingfrom withinsulciorwithtangential dipoles.IEDs mayalsobegeneratedbysuchasmallamountof cortex,thattheresultingextracellularcurrentsareinsufficientto allow reliable detection with scalp EEG.7 Another issue is the
limiteddurationofroutineEEG-registrations,whichwillnotshow anydischargesthatoccurinfrequently.Ontheotherhand,evenif
ARTICLE INFO
Articlehistory:
Received19December2012 Receivedinrevisedform25March2013 Accepted26March2013 Keywords: Firstseizure EEG Diagnosticdecision-making Survey ABSTRACT
Purpose:TheroleofEEGafterafirstseizurehasbeendebated.EpileptiformEEGactivityisagood predictorofseizurerecurrence,butisreportedinonly8–50%offirst-seizureadultpatients.Evenifthe EEGisabnormal,theopinionsabouttreatmentafterafirstseizurediffer.TheroleofEEGintreatment decisions after remission or recurrence is also unclear. This study aims to identify neurologists’ diagnosticstrategiescomparedtoguidelinesabouttheuseofEEG(i)afterafirstunprovokedgeneralized seizurein adults, (ii)after arecurrent seizureand (iii) intreatmentdecisions afterrecurrenceor remission.
Method: AllmembersoftheDutchNeurologicalSocietywereinvitedtoparticipateinouron-linesurvey abouttheuseofEEGafterafirstseizure,afterrecurrentseizuresandintreatmentdecisions.Tenpercent (N=110)ofinviteesparticipated,includingmainlyclinicalneurophysiologists,generalneurologistsand neurologists-in-training.
Results:Ninety-fivepercentoftherespondentswouldrequestaroutineEEGafterafirstseizure.After normalMRIandEEGfindings,4%wouldrecordasecondroutineEEG,48%asleep-deprivedEEGand45% wouldnotrepeattheEEG.Ifarecurrentseizureoccurswithinsix,orafter12or24months,87%,67%and 44%wouldrespectivelyconcludethatthepatienthasepilepsy,while57%,65%and72%wouldrequestan EEG.Whenapatientexperiencesarecurrencewhilebeingtreatedwithanti-epilepticdrugs,11%ofthe respondents would request an EEG. Twenty-five percent would request an EEG before stopping medicationaftertwoyearsofremission.
Conclusion:Thevariabilityinneurologists’reportedstrategiesabouttheuseofEEGinthediagnosisof seizuresisremarkably large.Consequencesfortheindividualpatientmay besignificant,including treatmentdecisionsand drivingrestrictions. Theavailabilityand useofmore sensitivediagnostic methodsmaybenecessarytoenhanceagreementbetweenneurologists.
ß2013BritishEpilepsyAssociation.PublishedbyElsevierLtd.Allrightsreserved.
*Corresponding author at: University of Twente, Department of Clinical Neurophysiology,BuildingCarre,P.O.Box217,7500AEEnschede,TheNetherlands. Tel.:+31534895310.
E-mailaddresses:J.Askamp@utwente.nl(J.Askamp),
M.J.A.M.vanPutten@utwente.nl(MichelJ.A.M.vanPutten).
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
1059-1311/$–seefrontmatterß2013BritishEpilepsyAssociation.PublishedbyElsevierLtd.Allrightsreserved.
theEEGisabnormal,theopinionsontreatmentafterafirstseizure differ.Immediatetreatmentreduces thenumberofrecurrences, butsomepatientsmaybetreatedunnecessarily.8,9
The majority of patients (92%) who present with a first unprovoked seizure and are treated with anti-epileptic drugs attainatwoyearremissionwithinfiveyearsafterthefirstevent, regardlessofimmediateordeferredtreatment.10Wheninforming
patientsabouttheconsequencesofdiscontinuingtreatmentafter such a period of seizure-freedom, accurate risk assessment is essential.Inthesecircumstances,theroutineEEGmayassist,asit allowsprognosticationaboutthelikelihoodofremissionormaybe usedtopredict seizurerecurrence in the eventof epileptiform activity.11Inchildren,persistentinterictalepileptiformactivityis
associatedwithanincreasedriskofseizurerecurrenceifAEDsare discontinuedafterremission.However,inadults,therelevanceof intericticalepileptiformactivityismuchlesscertain.12
Guidelinesmayassistphysiciansindecision-making concern-ingthediagnosisandtreatmentafterafirstseizureorrecurrent seizures.Theclarityandclinicalapplicabilityofaguidelinemaybe important attributes that contribute to the effects of practice guidelines.13Severalguidelinesorpracticeparametersregarding
theuseofEEGafterafirstseizureorrecurrentseizureshavebeen developed.
The American Academy of Neurology (AAN) and American Epilepsy Society (AES) developed a practice parameter for evaluation of adults presenting with an apparent unprovoked first seizure. They conclude that in these patients, the EEG is probablyhelpfulandthattheroutineEEGshouldbeusedaspartof the neurodiagnostic evaluation of the adult because it has substantial yield and value in determining the risk of seizure recurrence.3
TheNationalInstituteforHealthandClinicalExcellence(NICE) guidelines about the use of EEG state that an EEG should be performed only tosupporta diagnosisof epilepsy in adults in whomtheclinicalhistorysuggeststhattheseizureislikelytobe epilepticin origin.In those presentingwith a firstunprovoked seizure,unequivocalepileptiformactivityonEEGshouldbeusedto assesstherisk ofseizurerecurrence.Ifdiagnosisisstillunclear afterastandardEEG,repeatedstandardEEGsmaybehelpfulbut shouldnotbeusedinpreferencetosleeporsleep-deprivedEEGs. Further,whenaroutineEEGhasnotcontributedtodiagnosisor classification, a sleep EEG should be performed. There is no informationintheguidelinesabouttheroleofEEGinthedecision tostopmedicationafterremission.14
The International League Against Epilepsy (ILAE) has no international guidelines on the use and role of EEG in first-seizureandepilepsydiagnosis.However,accordingtotheItalian LeagueAgainstEpilepsy,anEEGshouldbeperformedwithin24h after a seizure, particularly in children. If the EEG is normal duringwakefulness,asleep EEGis recommended.15TheDutch
NeurologicalSocietyhas,togetherwiththeDutchLeagueAgainst Epilepsy, developed guidelines for diagnosisand treatmentof epilepsy(revised,2ndversion,2006).Theseguidelineshowever donotgiveinformationaboutwhichEEGsshouldbeusedaftera firstseizureorrecurrentseizuresinadults.
Inthis paper, we present neurologists’ reported diagnostic decisionsinadultswherewewillspecificallyemphasizeonthe useof EEG(i)after a firstunprovoked generalizedseizure,(ii) after recurrentseizures atdifferent time-intervals,and (iii)in thedecisiontostartorchangemedicationafterrecurrenceorto stopmedicationafterremission.Responsesreflectneurologists’ diagnosticdecisionsregarding patientsin theNetherlandsand willbecomparedtobothnationalandinternationalguidelines. 2. Methods
Approximately1100members(neurologistsandneurologists in training) of the Dutch Neurological Society wereinvited to participateinouronlinesurveyabouttheuseofEEGafterafirst seizureinadults,afterasecondseizureatdifferenttime-intervals from the first, and when making treatment decisions after recurrence orremission.The invitationincludedan informative letterwithalinktotheon-linesurveyatwww.epilepsydata.eu. Participationinthesurveywasanonymous.
Participantswerefirstaskedfortheireducationalbackground, typeof hospital they workedat andtheir number of yearsin practice.Acasewasthendescribedinwhichanadultpresentsto theEDafterafirstunprovokedgeneralizedseizure.Participants wereaskedwhattheirpolicy wouldbe,whatconclusions they would draw and which EEGs they would request for several scenarios, including normal and abnormal EEG findings and seizure recurrence. Corresponding questions and response options are listed in Table 1.Second, participantswere asked whetherthey wouldperform additionalEEG measurementsin treatedepilepsyifarecurrenceoccursafter18monthsofseizure freedomorwhenconsideringstoppingmedicationaftertwoyears ofremission.Thesurveyincludedmultiple-choice,yes–noaswell as open questions. Data were analyzed using PASW Statistics version18.0.0,SPSSInc.,bymeansofdescriptivestatistics. Non-responseswereexcluded.
3. Results
Tenpercent(N=110)oftheinvitedneurologists(intraining) respondedbeforethedeadline.Themajorityoftherespondents (56%)hadworkedformorethanfiveyearsasaneurologist(see
Table2). Primarily,general neurologists,clinical neurophysiolo-gistsandneurologistsintraining(87%intotal)respondedtothe survey(seeTable2).
Table1
Threesurveyquestionsforthecase:‘AnadultpresentstotheEDafterafirstunprovokedgeneralizedseizure–theneurologicalexaminationwasnormal’. Q1.Whatisyourpolicy?
Youcanchoosemultipleanswersfrom theoptionsbelow:
Q2.Whatwouldbeyourconclusionandpolicyafter eachofthefindingsbelow?
TheMRIwasnormal.
Q3.Whatwouldbeyourconclusionandpolicyafter eachofthefindingsbelow?
-MRI-scan -RoutineEEG -Sleep-deprivedEEG -Long-termEEG(2h) -Startmedication
-Hospitalizationforoneday/night -Routinebloodtests
-Noneoftheoptionsabove
(1)IftheroutineEEGisnormal
(2)Ifbothroutineandsleep-deprivedEEGsarenormal (3)Afterarecurrencewithin6monthsafterthefirstone (4)Afterarecurrenceafter12monthsafterthefirstone (5)Afterarecurrenceafter24monthsafterthefirstone
(1)IftheroutineEEGcontains2temporal spike-and-wavedischarges
(2)IftheroutineEEGcontainsgeneralized spike-and-wavedischarges
ResponseoptionsforQ2andQ3:conclusion:epilepsy/firstseizure,noepilepsy/possiblyepilepsybutIwillwait/noconclusion,Istillmissinformationpolicy:noEEG/routine EEG/sleep-deprivedEEG/long-termEEG(2h)/long-termEEG(>12h).
3.1. Firstseizure
Thelargemajorityoftherespondentswouldrequestan MRI-scan (96%) and routine EEG (95%) after a first unprovoked generalizedseizureinadults.Seventy-sixpercentwouldperform routinebloodtests(seeTable2).
Afteranormalneurologicalexamination,MRIandroutineEEG, 56%oftherespondentswoulddiagnosethepatientwithasingle seizure,notepilepsy, and23% wouldnot drawanyconclusions becauseofmissinginformation(seeTable3).Oftherespondents whoreportedthatinformationwasmissing,88%wouldrequesta sleep-deprived EEG. In general, 45% would not repeat thefirst normalEEG,whereas48%wouldrequestasleep-deprivedEEG(see
Table3).Only33%ofthegeneralneurologistswouldnotrepeata normalroutineEEG,whereas53%oftheclinicalneurophysiologists andresidentswouldnotrepeattheEEG.
IfthefirstroutineEEGcontainstwotemporalorgeneralized spike-and-wave discharges, 40% and 76% of the respondents respectivelywouldconcludethatthepatienthasepilepsy.Inthe presence of temporal discharges, 35% would request a sleep-deprivedEEG,whereasinthepresenceofgeneralizeddischarges, thelargemajority(93%)wouldnotrepeattheEEG(seeTable3).
3.2. Recurrentseizure
Afterarecurrencewithinsix,after12orafter24months,87%, 67%and44%oftherespondentsrespectivelywouldconcludethat thepatienthasepilepsy.Thepercentageofneurologists respond-ing‘possiblyepilepsy,butIwillwait’increasesfrom9%,to30%,to 46% (seeTable4).Residentsarelesslikely todiagnoseepilepsy after a recurrence after 24 months (35%) compared togeneral neurologists(53%),whilemoreresidentswouldconcludethatitis ‘possiblyepilepsy’(66%)comparedtogeneralneurologists(36%). 3.3. TheroleofEEGintreatmentdecisions
When apatientexperiencesarecurrence after18monthsof seizure-freedomwhilebeingtreatedwithanti-epilepticdrugsand intheabsenceofsleep-deprivation,69%would‘wait’,11%would requestanEEG,9%wouldprescribeanotherAED,and11%would addasecondAED.
Ifdiscontinuationofmedicationisconsideredaftertwoyearsof seizure-freedom, 25% would request an EEG before stopping. Seventy-five percent of the respondents would taper off anti-epilepticmedicationwithoutanadditionalEEG.
4. Discussion
Afirstseizuremayhavealargeimpactonapatients’life.16Ina
small proportionof first-seizure patients, epilepsyis diagnosed soon after the event.In various patients, however, uncertainty aboutthediagnosismaypersistquitesometime.17Unfortunately,
a reliablebiomarkerthatwouldenablephysicianstoaccurately predicttheriskofseizurerecurrenceisstillmissing.18TheEEGis generally used, but has limited sensitivity in epilepsy.19 This
causesdiagnosticuncertaintyafterafirstevent,inparticularifthe initial routine EEG is normal. Guidelines based on scientific knowledgefromlargeandwell-performedpatientstudiesshould guidetheneurologistindecision-makingregardingtheuseofEEG afterafirstseizure.However,guidelinesarerestrictedtoepilepsy diagnosis, providing information about the value of EEGs in epilepsy, while they lack specific information about which decisions tomakeafter a firstseizure.Another difficultyis the fact that there is not one single definition of epilepsy. The International League Against Epilepsy defines epilepsy as a disorderofthebraincharacterizedbyanenduringpredisposition to generate epileptic seizures and that epilepsy requires the occurrence of at least one epileptic seizure.20 The American
EpilepsySociety,however,statesthatepilepsyrequiresrecurrent
Table2
Participants’demographicdataandpolicyafterafirstunprovokedgeneralized seizureinadults. Response Total:N=110 Neurologist 36(33%) Neurologist/clinicalneurophysiologist 30(27%) Neurologist/epileptologist 4(4%) Neurologist/pediatricneurologist 10(9%) Neurologistintraining 30(27%) 5yearsorlonger 61(56%) <5years 19(17%) Intraining 30(27%)
Academicmedicalcenter 39(35%) Peripheralteachinghospital 37(34%) Peripheralnon-teachinghospital 24(22%) Epilepsycenter 10(9%) Whatisyourpolicy?(Multipleanswersallowed)
MRI-scan 105(96%)
RoutineEEG 104(95%)
Routinebloodtests 83(76%) Sleep-deprivedEEG 12(11%) Hospitalization 12(11%) Long-termEEG(>2h) 3(3%) Startmedication 1(1%)
Table3
Neurologists’reportedconclusionsandEEGrequestsfollowingafirstseizureinadultsand(i)anormalroutineEEG,(ii)normalroutineandsleep-deprivedEEGs,(iii)two temporaldischargesintheroutineEEG,and(iv)generalizeddischargesintheroutineEEG.
Conclusion Normalroutine EEG Normalroutine andSD-EEG Temporaldischarges inroutineEEG Generalizeddischarges inroutineEEG Focalepilepsy 0(0%) 0(0%) 40(38%) 0(0%) Generalizedepilepsy 0(0%) 0(0%) 2(2%) 81(76%) Singleseizure,notepilepsy 61(56%) 79(72%) 21(20%) 7(7%) Possiblyepilepsy,butIwillwait 23(21%) 26(24%) 37(35%) 15(14%) Noconclusion,Istillmissinformation 25(23%) 4(4%) 6(5%) 3(3%)
Total N=109 N=109 N=106 N=106
SecondEEG?
NosecondEEG 49(45%) 100(93%) 64(61%) 96(93%) SecondroutineEEG 4(4%) 2(2%) 1(1%) 4(4%) Sleep-deprivedEEG 53(48%) 1(1%) 36(35%) 1(1%) Long-termEEG(2h) 1(1%) 1(1%) 1(1%) 0(0%) Long-termEEG(>12h) 2(2%) 3(%) 2(2%) 2(2%)
seizures,twoormore,whicharenotprovokedbysystemicoracute neurologic insults.21 A similar definition was proposed by
epidemiologists,whodefineepilepsyastwoormoreunprovoked seizuresoccurringatleast24hapart,mainlybecauseevidenceofa recurrencemaybetheonlyinformation availabletoidentifyan ‘enduringpredispositiontogenerateseizures’.22Thepresentstudy
was performed in order to identify neurologists’ diagnostic strategiesabouttheuseofEEGafterafirstseizureinadults,after recurrentseizuresandduringtreatmentdecisions.
Our surveyamongDutchneurologistsshowedthataroutine EEGisalmostalwaysperformedafterafirstseizureinadults,as practically every neurologist (95%) reported to request this examination.Thisisconsistentwithearlierfindingsinover400 first-seizurepatients,where95%ofthepatientshadanEEGafter their first seizure.23 Also, this is consonant with the practice
parameteroftheAmericanAcademyforNeurologyandAmerican EpilepsySociety.3However,theNICEguidelinesseemmuchmore
difficulttointerpretatthispoint.14Ingeneral,thereseemstobeno
doubtabouttheusefulnessofthefirstroutineEEGaftera first seizureinadults.
AfteranormalMRIandroutineEEGhowever,4%wouldrequest asecondroutineEEG,48%wouldrequestasleep-deprived EEG, and45%wouldnotrepeattheEEG.Thisindicatesthatthereisno common opinion on whether or not a second EEG should be performed.Thepreferenceforsleep-deprivedoverroutineEEGs agreeswithearlierstudies5,24,25andwithNICEguidelines.14There is,however,notasingleexplanationforthefindingthatnearlyhalf oftheneurologistswouldnotrepeattheEEGwhiletheotherhalf would. In fact, the guidelines too lack information about the usefulnessofasecondEEGafteraninitialnormalroutineEEGin first-seizure adult patients. Our survey showed that, of the neurologistswho would not repeat a normal routine EEG,80% concludedthatitwasafirstseizure,notepilepsy.Therest(20%) concludedto ‘wait’, but that the patienthas possibly epilepsy. Presumably,theseneurologistssharetheopinionthatafterasingle seizureandnormalroutineEEG,sufficientinformationhasbeen collectedtoconcludethatthereisinsufficientevidenceofepilepsy. Here,severalconsiderationsmayplayarole.Itwasestimatedthat about5%ofthepopulationexperiencesaseizureatsomepointof his or her life.26 Some of the first-seizure patients will never
experiencearecurrence,buttheexactproportionisunknown.It wasestimatedthat,overall,40–50%ofuntreatedindividualscan expecta recurrence withintwo years after theinitial seizure,2
whichwouldimplyaremissionrateof50–60%.Also,neurologists andpatientsmayaccepttheriskofhavingarecurrence.
If,ontheotherhand,thefirstroutineEEGcontainstemporalor generalizeddischarges,40%and 76%oftherespondents respec-tivelywouldconcludethatthepatienthasepilepsy.Thisindicates that in the presence of generalized epileptiform discharges, neurologistsaremoreinclinedtothediagnosisofepilepsythan inthepresenceoftemporaldischarges,evenafterasingleseizure only.Itwasindeedreportedthatinparticulargeneralizedspike andwavedischargesorfocalspikesareassociatedwithagreater riskforseizurerecurrence.3Thelimitedpercentageofneurologists
whoreportedtodiagnosethepatientwithepilepsyinthepresence oftemporalspike-and-wavedischargesmaybecausedbythefact thatweaddedinthedescriptionofthecaseinoursurveythatonly two temporaldischarges werefound. Apparently,some neurol-ogists consider the presence of two temporal epileptiform discharges notsufficientforthediagnosisofepilepsy,whilefor others, these abnormalities are significant. The exact relation betweenthenumberofinterictalepileptiformdischargesinthe EEGandriskofseizurerecurrenceis,however,unknown.
Some patients may experience a recurrence after the first unprovokedgeneralizedseizure.Theroleofthetimebetweenthe firstandsecondseizureintheactualdiagnosisisunknown.The majorityofourparticipants(87%)woulddiagnoseepilepsyaftera recurrence within six months, but only 44% would diagnose epilepsyiftherecurrenceoccursafter24months.Simultaneously, thepercentageofrespondentsreportingthatitispossiblyepilepsy, buttowait,increasesfrom9%to46%.Thisindicatesthatthereisno common opinion on what thediagnosisepilepsyrequires with respect toseizure interval. Italso indicates that the confusion increaseswithincreasingtime-to-recurrence.Noneofthe guide-linesincludesthetimebetweenthefirstandrecurrentseizurein thedefinitionofepilepsy,whileitseemstobeaveryimportant issue,stronglyinfluencingthediagnosticdecisionsandverylikely alsotreatmentdecisionsinpatients.
In patients diagnosed with epilepsy and treated with anti-epilepticdrugs,remissionorrecurrencemayoccur.Ourfindings suggestthat,accordingtotherespondingneurologists,theroleof EEGafterseizurerecurrenceinapatientalreadytreatedwith anti-epilepticdrugsislimited.Only11%wouldrequestanEEGinthese circumstances, while the majority (69%) would ‘wait’. When consideringAEDdiscontinuanceafterremission,theroleofEEG seemstobealittlelarger,as25%ofourrespondentswouldrequest anEEGbeforestopping.SomeauthorssuggestedthattheEEGmay haveanimportantroleinAEDdiscontinuationdecisionmaking.11
However, our survey participants think that this role is quite limited.Forcomparison,asimilarpercentage(29%)indicatedto followthepatients’wishdirectlywithoutadditionalexaminations, ifthepatientwantstostopmedication.Thereasonmaybethat both physicians and patients are willing to accept the risk of seizurerecurrence.Indeed,it wasshownthat20%offamilies of childrenhavingepilepsywereevenwillingtodiscontinueAEDs withariskofseizurerecurrenceof75%.27
Although the results from our study are illustrative for neurologists’ diagnosticdecisions,conclusionsshould bedrawn with caution, as our survey has several limitations. The cases presentedwereverygeneral,which maymakedecision-making difficultbecausesomeinformationmightbemissingthatwould have been available in real practice. However, for each of the questionsaboutthefirst-seizureadultpatient,participantscould choose fortheoption‘No conclusion,Istillmiss information’and indeed,somerespondentsusedthisoption.Further,guidelinesare neverbasedonindividualcasesandtrytodescribethebestand generalwayofdiagnosingfirst-seizureorepilepsypatients.
Second,thevariabilityinoursurveyresultsmaybecausedby thelargevarietyinthebackgroundofourparticipants,including differencesineducationandyearsinpractice.Wedo notknow exactly how well this group represents the population of
Table4
Neurologists’reportedconclusionsand EEGrequestsafter arecurrentseizure withinsix,andafter12and24monthsfollowingthefirstone.
Conclusion Withinsix months After12 months After24 months Epilepsy 95(87%) 73(67%) 47(44%) Singleseizure,notepilepsy 0(0%) 1(1%) 9(8%) Possiblyepilepsy,butI
willwait
10(9%) 32(30%) 50(46%) Noconclusion,Istillmiss
information 4(4%) 2(2%) 2(2%) Total N=109 N=108 N=108 Policy:EEG? NoEEG 46(43%) 37(35%) 30(28%) RoutineEEG 25(23%) 32(30%) 44(41%) Sleep-deprivedEEG 25(23%) 24(22%) 23(21%) Long-termEEG(2h) 4(4%) 5(5%) 4(4%) Long-termEEG(>12h) 7(7%) 9(8%) 6(6%) Total N=107 N=107 N=107
neurologistsevaluatingfirst-seizureadultpatients.Furthermore, onlytenpercentoftheinviteesrespondedtothesurvey.Weknow the background of our participants, but we have no specific information about the background of non-respondents. The percentage of neurologists-in-training within the respondents (27%)wassimilartothepercentageofneurologists-in-trainingthat are members of the Dutch Neurological Society(approximately 29%),whichindicatesthatoursampleisatleastpartly representa-tiveforallthemembersoftheDutchNeurologicalSociety.
Opinionsandattitudesmayvaryacrossspecialists.Forexample, residentsseemlesslikelytodiagnoseepilepsyafterarecurrence after24monthsthangeneralneurologists.Conclusionsaboutthe significanceofthesedifferencesshouldbedrawnwithcaution,as groups were relatively small and variable in size (e.g., four epileptologistscomparedtothirtyclinicalneurophysiologists).
Further, the results of the study reflect the opinions and attitudesofDutchneurologistsandarenotnecessarilyapplicable to physicians working in other countries. For example, the incidence of epilepsy differs between countries,28 which may
causedifferencesin(diagnostic)decisionsmadebyneurologistsin different countries. Also, differences in opinions and attitudes towardtheuseofEEGafterafirstseizureorinepilepsymaybe causedbysocio-culturalandeducationalbackgrounddifferences varyingovercountries.Besides,wedonotknowtowhatextent neurologistsuse(international)guidelinesduringdiagnosticand treatmentdecisions.
Insum,thepresentstudyhasshownthatthereislargevariability betweenneurologists’reporteddiagnosticdecisionsabouttheuseof EEGafterafirstseizure,afterrecurrentseizures,orintreatment decisions.TheroleofthefirstroutineEEGafterafirstseizurein adultsisnotinquestion,however,thereisnoconsensusaboutthe roleofasecondEEGafteraninitialnormalEEGandtheroleforEEGin treatmentdecisions.Thelargevariabilityindiagnosticstrategies reflectsthelimitedsensitivityoftheEEG.Furthermore,neurologists mayfollowdifferentdefinitionsofepilepsy.
Differencesindiagnosticdecisionsbetweenneurologistsmay havesignificantconsequences for theindividualpatient aftera visittooneoftheneurologists,includingtreatmentand driving restrictions.Agreementbetweenneurologistswould,therefore,be highly desirable. This may be enhanced by the use of more sensitivediagnosticmethodsandunambiguousguidelines. Acknowledgements
Wewouldliketothankallneurologistswhoparticipatedinour onlinesurveyfortheircontributiontothisstudy.
ThisstudywasfundedbytheHighTechHealthFarmandthe provinceofOverijssel.
References
1.Pohlmann-EdenB,NewtonM.Firstseizure:EEGandneuroimagingfollowingan epilepticseizure.Epilepsia2008;49(Suppl.1):19–25.
2.Berg AT. Risk of recurrence after a first unprovoked seizure. Epilepsia 2008;49(Suppl.1):13–8.
3.KrumholzA,WiebeS,GronsethG,ShinnarS,LevisohnP,TingT,etal.Practice parameter:evaluating anapparent unprovokedfirst seizurein adults(an
evidence-based review):report ofthe QualityStandardsSubcommitteeof the American Academy ofNeurologyand the AmericanEpilepsy Society. Neurology2007;69:1996–2007.
4.NoachtarS,Re´miJ.TheroleofEEGinepilepsy:acriticalreview.Epilepsyand Behavior2009;15:22–33.
5.LiporaceJ,TatumW,MorrisGL,FrenchJ.Clinicalutilityofsleep-deprivedversus computer-assistedambulatory16-channelEEGinepilepsypatients:a multi-centerstudy.EpilepsyResearch1998;32:357–62.
6.DoppelbauerA,ZeitlhoferJ,ZifkoU,BaumgartnerC,MayrN,DeeckeL. Occur-renceofepileptiformactivityintheroutineEEGofepilepticpatients.Acta NeurologicaScandinavica1993;87:345–52.
7.PillaiJ,SperlingMR.InterictalEEGandthediagnosisofepilepsy.Epilepsia 2006;47(Suppl.1):14–22[Table1].
8.FountainNB,FreemanJM.EEGisanessentialclinicaltool:proandcon.Epilepsia 2006;47(Suppl.1):23–5.
9.vanDonselaarC,GeertsA,SchimsheimerRJ.Idiopathicfirstseizureinadultlife: whoshouldbetreated?BritishMedicalJournal(ClinicalResearched)1991;302: 620–3.
10.MarsonA,JacobyA,JohnsonA,KimL,GambleC,ChadwickD.Immediateversus deferredantiepilepticdrugtreatmentforearlyepilepsyandsingleseizures:a randomisedcontrolledtrial.TheLancet2007;365:2007–13.
11.BrittonJW.SignificanceoftheEEGandepileptiformabnormalitiesin antiepi-lepticdrugdiscontinuance.JournalofClinicalNeurophysiology2010;27:249–54.
12.DuncanJS.Antiepilepticdrugsandtheelectroencephalogram.Epilepsia1987;28: 259–66.
13.ShekellePG,KravitzRL,BeartJ,MargerM,WangM,LeeM.Arenonspecific practiceguidelinespotentiallyharmful?Arandomizedcomparisonoftheeffect ofnonspecificversusspecificguidelinesonphysiciandecisionmaking.Health ServicesResearch2000;34:1429–48.
14.NationalInstituteforHealthandClinicalExcellence.Theepilepsies:the diag-nosisandmanagementoftheepilepsiesinadultsandchildreninprimaryand secondary care. CG137. London:NationalInstitute forHealthand Clinical Excellence;2010.
15.BeghiE,DeMariaG,GobbiG,VeneselliE.Diagnosisandtreatmentofthefirst epilepticseizure:guidelinesoftheItalianLeagueagainstEpilepsy.Epilepsia 2006;47(Suppl.5):2–8.
16.VelissarisSL,WilsonSJ,SalingMM,NewtonMR,BerkovicSF.Thepsychological impactofanewlydiagnosedseizure:losingandrestoringperceivedcontrol. EpilepsyandBehavior2007;10:223–33.
17.vanDonselaarC,StroinkH. Epilepsia2006;47(Suppl.1):9–13.
18.EngelJ.Progressinepilepsy:reducingthetreatmentgapandthepromiseof biomarkers.CurrentOpinioninNeurology2008;21:150–4.
19.SmithSJM.EEGinthediagnosis,classification,andmanagementofpatients withepilepsy.JournalofNeurologyNeurosurgeryandPsychiatry2005;76(Suppl. 2):ii2–7.
20.FisherRS,VanEmdeBoasW,BlumeW,ElgerC,GentonP,LeeP,etal.Epileptic seizuresandepilepsy:definitionsproposedbytheInternationalLeagueAgainst Epilepsy(ILAE) and the InternationalBureau forEpilepsy (IBE). Epilepsia 2005;46:470–2.
21.AmericanEpilepsySociety.Clinical epilepsyslides:definitions;2010. http:// www.aesnet.org/go/professional-development/educational-opportunities/epi-lepsy-education-program/clinical-epilepsy/nocase-slides[accessed01.10.12]. 22.ThurmanDJ,BeghiE,BegleyCE,BergAT,BuchhalterJR,DingD,etal.Standards
forepidemiologicstudiesandsurveillanceofepilepsy.Epilepsia2011;52(Suppl. 7):2–26.
23.HopkinsA,GarmanA,ClarkeC.Thefirstseizureinadultlife.Valueofclinical features,electroencephalography,andcomputerizedtomographicscanningin predictionofseizurerecurrence.TheLancet1988;1:721–6.
24.RoupakiotisSC,GatzonisSD,TriantafyllouN,MantouvalosV,ChioniA,Zournas C,etal.Theusefulnessofsleepandsleepdeprivationasactivatingmethodsin electroencephalographicrecording:contributiontoalong-standingdiscussion. Seizure2000;9:580–4.
25.El-AdB,NeufeldMY,KorczynAD.ShouldsleepEEGrecordalwaysbeperformed aftersleepdeprivation?ElectroencephalographyandClinicalNeurophysiology 1994;90:313–5.
26.GoodridgeDM,ShorvonSD.Epilepticseizures ina populationof6000.I. Demography,diagnosisandclassification,androleofthehospitalservices. BritishMedicalJournal(ClinicalResearched)1983;287:641–4.
27.GordonK,MacSweenJ,DooleyJ,CamfieldC,CamfieldP,SmithB.Familiesare contenttodiscontinueantiepilepticdrugsatdifferentrisksthantheir physi-cians.Epilepsia1996;37:557–62.
28.SanderJW.Theepidemiologyofepilepsyrevisited.CurrentOpinioninNeurology 2003;16:165–70.
