Fig2EventrelatedpotentialsprinciplesElectroencephalographicEEGactivityisrecordedbyscalpelectrodesandstimuliarepresentedseveraltimesaBymergingepochsofEEGaccordingtothebeginningofastimulusthea ID: 232041
Download Pdf The PPT/PDF document "PredictingValuesofERPsPredictingfunction..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
PredictingValuesofERPsPredictingfunctionaloutcomeaftercomaisanimportantaspectofintensivecare.PredictingvaluesofERPsareestimatedbycalculatingtherelationshipbetweenthepresence(ortheabsence)ofacomponentandtheoutcomeofpatients.Positivepredictivevalueforfavorableoutcome(PV+)estimatesthepercentageofpatientswhowillrecoverwhenacomponentisevoked(itisthenumberofpatientswithfavorableoutcomeandwiththecomponent/totalnumberofpatientswiththecomponent).Incontrast,positivepredictivevalueforunfavorableoutcome(PVestimatesthepercentageofpatientswhowontrecoverwhennocomponentisevoked(itisthenumberofpatientswithunfavorableoutcomeandwithoutthecomponent/totalnumberofpatientswithoutthecomponent).Therelation-shipbetweenERPsandoutcomecouldbealsoassessingfromtheoutcome.Sensitivityforunfavorableoutcome(orspecicityforfavorableoutcome;Sp+)estimatestheper-centageofpatientswhohadnotacomponentwhentheoutcomeisbad(itisthenumberofpatientswithunfa-vorableoutcomeandwithoutthecomponent/totalnumberofpatientswithunfavorableoutcome).Incontrast,sensi-tivityforfavorableoutcome(orspecicityforunfavorableoutcome;Sp)estimatesthepercentageofpatientswhohadacomponentwhentheoutcomeisgood(itisthenumberofpatientswithfavorableoutcomeandwiththecomponent/totalnumberofpatientswithfavorableoutcome).Short-LatencyERPsShort-latencyERPsareroutinelyusedbecausetheyareavalidatedmeanstopredictpooroutcome(forarecentreview,seeLaureysetal.[]).Theyshowaverylowrateoffalse-negativepredictions,i.e.,mostpatientsbilaterallymissingthesepotentialsshowanegativepredictionout-come.Ontheotherhand,therateoffalse-positivepredictionsishigh,i.e.,well-preservedpotentialsarerecordedinpatientswhohaveabadclinicalprognosis.Ifbothsomatosensory(i.e.,repeatedelectricalstimulationofthemediannerveatthewrists)andauditorystimulations(i.e.,presentationofclicksusingheadphones)havebeenused,theybringrelativelysimilarconclusions.VisualERPs(usinggoggleswithashingLEDs)areoflimiteduseattheintensivecaresettingbecausetheyfailtoshowsystematicactivationeveninhealthysubjects.Arecentmeta-analysisconrmedthatsomatosensoryevoked-potentials(SEPs),whichreecttheactivationofthesubcorticalsomatosensorypathwaysandoftheprimarysomatosensorycortex,aresuperior,withfewexceptions,topupillaryresponses,motorresponses,GlasgowComaScale,EEG,andcomputedtomographyforthepredictionoffavorableorunfavorableoutcomeafteracuteseverebraindamage[].BilateralabsenceofSEPsamongpatientsincomaisstronglyassociatedwithanon-awak-eningprognosis(PVnear100%),i.e.,deathorpermanentVS[].Atthesametime,allpatientswithfavorableoutcomehavedevelopednormalSEPs(Spnear100%).Incontrast,normalSEPsareassociatedtoagreatnumberoffalsepositives(forreviewssee[])suggestingthatSEPsarenotagoodpredictorofrecovery.Theabsenceofbrainstemauditory-evokedpotentials(BAEPs),whichareevokedintherst10msandrevealtheactivityfromtheauditorynervetotheinferiorcol-liculus[],canbeseenasareliableprognosticatorforpooroutcomewhenthereisnoevidenceofperipheralauditorydamage.Ontheotherhand,thepresenceofnor-malBAEPsdoesnotreliablyindicateagoodoutcome[].Similarly,theabsenceofmiddle-latencyauditory-evokedpotentials(MLAEPs;appearingbetween10and50msandprobablysigningtheactivationofthal-amusandprimaryauditorycortex)isstronglyassociatedwithbadoutcomeinpostanoxiccoma[CognitiveERPsIfshort-latencyERPsareusefultopredictunfavorableoutcomesincomasurvivors,theyarelesshelpfulinprog-nosticatingrecovery.Moreover,theyonlyestimatetheintegrityofascendingpathwaysandnotofpossibleresidualcognitivefunctioning.Somerecentstudiessuggestthatthe Fig.2Event-relatedpotentialsprinciples.Electroencephalographic(EEG)activityisrecordedbyscalpelectrodesandstimuliarepresentedseveraltimes(a).BymergingepochsofEEG,accordingtothebeginningofastimulus,theactivitytime-lockedtothestimulusisrevealedandthespontaneousbrainactivitycancels.Themorestimulipresentedthemorespontaneousbrainactivitycancelssimplyforstatisticalreasons(b)(adaptedfromGueritetal.[ NeurocritCare(2008)8:262270 useofauditorycognitiveERPsmightcomplementshort-latencyERPsevaluationattheintensivecareunit.AscomparedtoearlyERPs,cognitiveERPsareverydependentontheexperimentalconditions.Thus,itisveryimportanttoperformclinicalexplorationsinoptimalvigilance,attention,andhabituationsettings.ThisimpliestouseanumberofstimuliwhichisoptimizedtobothdigupERPsfromthebackgroundEEGactivityandtoavoidhabituationphe-nomena.Sinceallpublishedstudiesarenotconductedinsimilarconditions,theresultsarenothomogenous.WereviewedMedlinefromJanuary1,1980toJune1,2007(searchterms:coma,outcome,andevokedpotentials)andselectedallstudiesinvestigatingcognitiveERPs(N100,MMN,P300,orN400)ofcomatoseandpost-comatose(VSandMCS)patients,andinwhichprognosticvaluesweredescribed.TheN100ComponentTheN100waveisanegativedeection,elicitedaround100ms,inresponsetoanyauditorystimulus(seeFig.Fig.26].Itcorrespondstotheactivationoftheauditorycortex(andperhapsalsoofdorsolateralprefrontalareas)[IfthepresenceofaN100incomatoseandpost-coma-tosepatientssuggeststhattheprimaryauditorycortexisfunctionallypreserved,itdoesnotappeartobeagoodpredictorofbadorgoodoutcome(seeFig.).Someauthorssuggestthatitspresencepredicts(seepredictivevaluesofN100inTable)recovery[]whileothersdonot[].Ontheotherhand,twostudiessuggestthatitsabsenceappearstobeagoodpredictorofbadoutcomeandthat,amongpatientswhorecovered,ahighproportionofthemhadaN100[TheMismatchNegativity(MMN)TheMMNisanegativecomponentelicitedafter100200msbyanychangeormismatch,inasequenceofmonotonousauditorystimuli(i.e.,anoddballparadigm)ininattentivesubjectswhoforexamplewatchTV(seeb)[].Itisrelativelysmallinamplitude;thusitisgenerallydisplayedasadifferencewavecomputedbysubtractingtherepetitivefromthedeviantresponse.Fortheauditorymodality,theprimaryauditorycortexandpre-frontalareasparticipatetoitsgeneration[].MMNmayindexanautomaticstageofinformationprocessingafteracomparisonprocessbetweentheafferentinputandamemorytracedevelopedbytherepetitivestimulation[DataonMMNincomatosepatientsconvergetotheconclusionthatitisaverygoodpredictorofrecovery(seeforpredictivevaluesofMMN),notablyinanoxiccoma[].Inaprospectivecohortstudyon346comatosepatients,Fischeretal.[]reportedahighprognosticvalueoftheMMNforrecovery,alletiol-ogyconfounded(seeFig.).Inbothgroups,averyhighproportionofpatientswhodidnotrecoverdidnotevokeaMMNcomponentbutlowSpandPVwerealsoobserved.Similarresultshavebeenobtained,morerecently,byNaccacheetal.[Kotchoubeyetal.foundaMMNin65%ofpatientsinVSandin34%ofpatientsinMCS[].Interestingly,theyshowedthatcomplextoneselicitedaMMNsigni-cantlymorefrequently(in50%ofpatients)thanpuretones(24%).Veryrecently,acorrelationhasbeenfoundbetween Fig.3AuditorycognitiveERPsinhealthyparticipants.()Inoddballparadigms,monotonousstimuli(graytrace)elicittheN100,P200waves,anddeviantstimuli(darktrace)elicittheN100,P200,N200,andP300waves(notethatP200andN200waveswerenotsystematicallyinvestigatedinpatientswithdisordersofconsciousness).()Ininattentiveoddballparadigms,aMMNisevokedbythedeviantstimulus(darktrace),followingtheN100wave.()Incongruouswords(darktrace)elicitN100andN400componentswhereascongruouswordsdonotelicittheN400 Fig.4AuditorycognitiveERPsincomatosepatients.N100andMMNcomponentsareelicitedbothinhealthysubjects(leftpartofthegure)andincomatosepatients,withGCS8(rightpart)adaptedfromFischeretal.[ NeurocritCare(2008)8:262270