NEUROLOGY/CLINICALPOLICY Clinical Policy: Use of Intravenous tPA for the Management of Acute Ischemic Stroke in the Emergency Department This clinical policy is the result of a collaborative project of the American College of Emergency Physicians and the American Academy of Neurology. Development Panel Robert L. Wears, MD, MS (Methodologist; Department Jonathan A. Edlow, MD (Department of Emergency of Emergency Medicine, University of Florida, Medicine, Beth Israel Deaconess Medical Center, Jacksonville, FL) Harvard Medical School, Boston, MA) Wyatt W. Decker, MD (Vice President and Trustee Mayo Eric E. Smith, MD, MPH (Department of Clinical Clinic, CEO Mayo Clinic Arizona, Scottsdale, AZ) Neurosciences, Hotchkiss Brain Institute [E.E.S.], Providing Project Support: University of Calgary, Foothills Medical Centre, Rhonda R. Whitson, RHIA, Clinical Practice Manager, Calgary, Canada) American College of Emergency Physicians LathaGantiStead,MD,MS,MBA(ProfessorofEmergency Thomas S. D. Getchius, Associate Director, Clinical MedicineandNeurologicalSurgery;Director,Centerfor Practice, American Academy of Neurology BrainInjuryResearchandEducation,Universityof Florida,Gainesville,FL) GaryGronseth,MD(DepartmentofNeurology,Universityof Approved by the ACEP Board of Directors, June 13, KansasMedicalCenter,KansasCity,KS) 2012 StevenR.Messé,MD(DepartmentofNeurology,Hospital Endorsed by the American Academy of Neurology, oftheUniversityofPennsylvania,Philadelphia,PA) December 6, 2012 Andy S. Jagoda, MD (Professor and Chair, Department Supported by the Emergency Nurses Association, of Emergency Medicine Mount Sinai School of December 11, 2012 Medicine; Medical Director, Emergency Department, Endorsed by the Neurocritical Care Society, January 4, Mount Sinai Hospital, New York, NY) 2013 Policy statements and clinical policies are the official policies of the American College of Emergency Physiciansand,assuch,arenotsubjecttothesamepeerreviewprocessasarticlesappearingintheprint journal.PolicystatementsandclinicalpoliciesofACEPdonotnecessarilyreflectthepoliciesandbeliefs of Annals of Emergency Medicine and its editors. 0196-0644/$-seefrontmatter Copyright©2012bytheAmericanCollegeofEmergencyPhysicians. http://dx.doi.org/10.1016/j.annemergmed.2012.11.005 [AnnEmergMed.2013;61:225-243.] treatedbetween3to4.5hoursaftersymptomonset? Evidencewasgradedandrecommendationsweregiven ABSTRACT basedonthestrengthoftheavailabledatainthemedical literature. Thispolicywasdevelopedbyajointwritingpanelofthe AmericanCollegeofEmergencyPhysiciansandtheAmerican AcademyofNeurology.Thepanelreviewedtheliteratureto INTRODUCTION deriveevidence-basedrecommendationstohelpclinicians Itisestimatedthatthereare795,000newstrokesinthe answerthefollowingcriticalquestions: UnitedStateseachyear.1Strokeisthethirdleadingcauseof (1)Isintravenoustissueplasminogenactivator(tPA) deathintheUnitedStates,causing1ofevery17deathsin safeandeffectiveforacuteischemicstrokepatientsif 2005.1 givenwithin3hoursofsymptomonset?(2)Isintravenous In1996,theFoodandDrugAdministration(FDA) tPAsafeandeffectiveforacuteischemicstrokepatients approvedintravenous(IV)tissueplasminogenactivator(tPA)as Volume,. : February Annalsof EmergencyMedicine 225 ClinicalPolicy atreatmentforacuteischemicstroke.Sincethen,theuseofIV formulatakingintoaccountdesignandqualityofstudy tPAforstrokehasbeenoneofthemostcontentiousmedical (AppendixB).Articleswithfatalflawsweregivenan“X”grade treatments. andnotusedinformulatingrecommendationsinthispolicy. Evidencegradingwasdonewithrespecttothespecificdata METHODOLOGY beingextractedandthespecificcriticalquestionbeingreviewed. AjointdevelopmentpanelwasappointedbytheAmerican Thus,thelevelofevidenceforanyonestudymayvary CollegeofEmergencyPhysicians(ACEP)andtheAmerican accordingtothequestion,anditispossibleforasinglearticleto AcademyofNeurology(AAN)toproduceaclinicalevidence– receivedifferentlevelsofgradingasdifferentcriticalquestions basedguidelineontheuseoftPAforacuteischemicstroke. areanswered.Question-specificlevelofevidencegradingmaybe Thisclinicalpolicywascreatedaftercarefulreviewand foundintheEvidentiaryTableincludedattheendofthis criticalanalysisofthemedicalliterature.Multiplesearchesof policy. MEDLINEandtheCochraneDatabaseforarticlespublished Clinicalfindingsandstrengthofrecommendationsregarding betweenJanuary1999andMay2011wereperformedusinga patientmanagementwerethenmadeaccordingtothefollowing combinationofkeywords,including“cerebrovascular criteria: accident,”“tissueplasminogenactivator,”“tPA,”“thrombolytic LevelArecommendations.Generallyacceptedprinciplesfor therapy,”“stroke,”“intracerebralhemorrhage,”“subarachnoid patientmanagementthatreflectahighdegreeofclinical hemorrhage,”“emergencydepartment,”“emergencyservice,” certainty(ie,basedonstrengthofevidenceClassIor “emergencyroom,”“therapyinemergencydepartment,”and overwhelmingevidencefromstrengthofevidenceClassII “treatmentinemergencydepartment.”Thesearcheswere studiesthatdirectlyaddressalloftheissues). limitedtotheEnglishlanguageandhumanstudies.Additional LevelBrecommendations.Recommendationsforpatient articleswerereviewedfromthebibliographiesofstudiescited. managementthatmayidentifyaparticularstrategyorrangeof Panelmemberssuppliedarticlesfromtheirownknowledgeand managementstrategiesthatreflectmoderateclinicalcertainty files,andmorerecentarticlesidentifiedduringtheprocesswere (ie,basedonstrengthofevidenceClassIIstudiesthatdirectly alsoincluded. addresstheissue,decisionanalysisthatdirectlyaddressesthe ThereasonsfordevelopingACEP’sclinicalpoliciesandthe issue,orstrongconsensusofstrengthofevidenceClassIII approachesusedintheirdevelopmenthavebeenenumerated.2 studies). Expertreviewcommentswerereceivedfromemergency LevelCrecommendations.Otherstrategiesforpatient physicians,neurologists,andindividualmembersofthe managementthatarebasedonClassIIIstudies,orinthe AmericanAcademyofFamilyPhysicians,AmericanCollegeof absenceofanyadequate,publishedliterature,basedonpanel Physicians,EmergencyNursesAssociation,AmericanStroke consensus. Association,NationalStrokeAssociation,NeurocriticalCare Therearecertaincircumstancesinwhichthe Society,andtheSocietyforAcademicEmergencyMedicine. recommendationsstemmingfromabodyofevidenceshouldnot Theirresponseswereusedtofurtherrefineandenhancethis beratedashighlyastheindividualstudiesonwhichtheyarebased. policy;however,theirresponsesdonotimplyendorsementof Factorssuchasheterogeneityofresults,uncertaintyabouteffect thisclinicalpolicy.Commentswerealsoreceivedfrominternal magnitudeandconsequences,andpublicationbias,amongothers, ACEPandAANcommitteesandworkgroups.ACEPclinical mightleadtosuchadowngradingofrecommendations. policiesarescheduledforrevisionevery3years;however, Whenpossible,clinicallyorientedstatistics(eg,likelihood interimreviewsareconductedwhentechnologyorthepractice ratios,numberneededtotreat)willbepresentedtohelpthe environmentchangessignificantly.ACEPandAANarethe readerbetterunderstandhowtheresultscanbeappliedtothe fundingsourceforthisclinicalpolicy. Thesearchesresultedin1,140articles,ofwhich303were individualpatient.Forfurtherdefinitionofthesestatistical selectedforadditionalreviewandgrading.Allarticlesusedin concepts,seeAppendixC. theformulationofthisclinicalpolicywereindependently Thispolicyisnotintendedtobeacompletemanualonthe gradedbyatleast2panelmembersforstrengthofevidenceand evaluationandmanagementofadultpatientswithacute classifiedbythepanelmembersinto3classesofevidenceonthe ischemicstrokebutratherafocusedexaminationofcritical basisofthedesignofthestudy,withdesign1representingthe issuesthathaveparticularrelevancetothecurrentpracticeof strongestevidenceanddesign3representingtheweakest emergencymedicine.Itisthegoalofthispaneltoprovidean evidencefortherapeutic,diagnostic,andprognosticclinical evidence-basedrecommendationwhenthemedicalliterature reports,respectively(AppendixA).Articleswerethengradedon providesenoughqualityinformationtoansweracritical 6dimensionsthoughttobemostrelevanttothedevelopmentof question.Whenthemedicalliteraturedoesnotcontainenough aclinicalguideline:blindedversusnonblindedoutcome qualityinformationtoansweracriticalquestion,themembers assessment,allocation,directorindirectoutcomemeasures, ofthepanelbelievethatitisequallyimportanttoalert biases(eg,selection,detection,transfer),externalvalidity(ie, physicianstothisfact.Recommendationsofferedinthispolicy generalizability),andsufficientsamplesize.Articlesreceiveda arenotintendedtorepresenttheonlydiagnosticand finalgrade(ClassI,II,III)onthebasisofapredetermined managementoptionsthatthephysicianshouldconsider.ACEP 226 Annalsof EmergencyMedicine Volume,. : February ClinicalPolicy andAANclearlyrecognizetheimportanceoftheindividual symptomonsetwererandomized1:1toplaceboversusIV physician’sjudgment.Rather,thisguidelinedefinesforthe treatmentwith0.9mg/kgofthehumanrecombinanttPA physicianthosestrategiesforwhichmedicalliteratureexiststo alteplase,with10%ofthetotaldoseadministeredasabolus providesupportforanswerstothecriticalquestionsaddressed andtheremaining90%infusedover60minutes(maximum inthispolicy. dose90mg).Randomizationwasstratifiedbyclinicalcenter ScopeofApplication.Thisguidelineisintendedfor andbytimefromtheonsetofstroketotreatment(0to90 physiciansworkinginhospital-basedemergencydepartments minutesand91to180minutes).Theprespecifiedprimary (EDs). outcomeofNINDSpartI(n(cid:1)291)wasearlyclinical InclusionCriteria.Thisguidelineisintendedforadult improvement,definedascompleteresolutionofthestroke patientspresentingtotheEDwithacuteischemicstroke. symptomsoranimprovementintheNationalInstitutesof ExclusionCriteria.Thisguidelineisnotintendedtobe HealthStrokeScale(NIHSS)(Figure1)scoreby4ormore appliedtochildrenyoungerthan18years. pointsat24hours.Therewasnodifferenceinearlyclinical improvementinthetPAgroupcomparedwiththeplacebo CRITICALQUESTIONS group(relativerisk1.2;95%confidenceinterval[CI]0.9to 1. IsIVtPAsafeandeffectiveforacuteischemicstroke 1.6;P(cid:1).21).TheprespecifiedprimaryoutcomeofNINDS patientsifgivenwithin3hoursofsymptomonset? partII(n(cid:1)333)wasafavorableoutcomeat3months, 2. IsIVtPAsafeandeffectiveforacuteischemicstroke determinedusing4assessmentscales:theBarthelIndex patientstreatedbetween3to4.5hoursaftersymptom (Figure2),modifiedRankinScale(Table1),Glasgow onset? OutcomeScale(Table2),andNIHSS(Figure1).Totestthe primaryhypothesis,aglobalendpointwasderivedfromthe PatientManagementRecommendations individualscaleswiththeuseofscale-specificcutpoints.The LevelArecommendations.Inordertoimprovefunctional oddsratio(OR)forafavorableoutcomeinthetPAgroup, outcomes,IVtPAshouldbeofferedtoacuteischemicstroke definedasminimalornodisabilityat90days,was1.7(95% patientswhomeetNationalInstituteofNeurologicalDisorders CI1.2to2.6;P(cid:1).008).AfavorableoutcomeforthetPA andStroke(NINDS)inclusion/exclusioncriteriaandcanbe groupwasobservedoneachofthe4assessmentscales treatedwithin3hoursaftersymptomonset.* (P(cid:1).02to.03),withabsolutepercentagedifferencesbetween LevelBrecommendations.Inordertoimprovefunctional tPAandplaceborangingfrom11%to13%.Forexample,a outcomes,IVtPAshouldbeconsideredinacuteischemicstroke modifiedRankinScalescoreoutcomeof0or1,indicating patientswhomeetEuropeanCooperativeAcuteStrokeStudy noresidualdisablingstrokesymptoms,wasachievedin39% (ECASS)IIIinclusion/exclusioncriteriaandcanbetreated oftPA-treatedpatientsversus26%ofplacebo-treated between3to4.5hoursaftersymptomonset.* patients.Therewasa12%absoluteincreaseinthenumber *TheeffectivenessoftPAhasbeenlesswellestablishedin ofpatientswithminimalornodisabilityinthetPAgroup, institutionswithoutthesystemsinplacetosafelyadministerthe medication. definedbytheglobalstatistic.Thiscorrespondstoanumber Note:Withinanytimewindow,oncethedecisionismadeto neededtotreatof8.3,meaningthat8.3patientswouldneed administerIVtPA,thepatientshouldbetreatedasrapidlyas tobetreatedfor1additionalpatienttoachieveafavorable possible.Asofthiswriting,tPAforacuteischemicstrokeinthe outcomewithessentiallynostroke-relateddisability.A 3-to4.5-hourwindowisnotFDAapproved. subsequentreanalysisofthetrialdatasuggestedthatthe LevelCrecommendations.Nonespecified. numberneededtotreattoproducea1-pointshiftinthe RankinScale,includingfromstatesofseveredisabilityto Mostischemicstrokesinadultsarecausedbythrombotic moremoderatedisability,maybeaslowas3.6 orembolicocclusionsofanartery.WithtPA,inactive CombinedanalysisofpartsIandIIoftheNINDSstudy plasminogenisconvertedintotheactiveformplasmin, showedaconsistenteffectofIVtPAonfavorableoutcomeat90 whichpromotesthrombolysisbycleavingfibrin.In1995,the days.3Thisbeneficialeffectwasobservedinboththe0-to90- NINDStPAStrokeStudyGrouppublisheda2-part minuteandthe91-to180-minutetimestrata.Mortalitywas randomizedcontrolledtrialshowingthathuman recombinanttPAimprovedoutcomesafterischemicstroke.3 similarinbothgroups(17%fortPAversus21%forplacebo; P(cid:1).30).Therewas,however,anincreaseinsymptomatic ThispublicationledtoFDAapprovalin1996.Reactionto theavailabilityoftPAforacuteischemicstrokehasranged intracerebralhemorrhageinthetPA-treatedgroupduringthe fromskepticism4tounbridledenthusiasm.5 first36hours(6%versus0.6%intheplacebogroup;P(cid:2).001). TheClassININDStPAstudywasdividedinto2parts.3 ManyofthesetPA-relatedhemorrhageswerefatal(45%). Eachpartwasperformedinaunique,independentlyenrolled Therefore,theimproved90-dayoutcomesinthetPAgroup populationofpatientswithacuteischemicstrokebutwith (withoutanincreasedoverallmortality)occurreddespitethe differentprespecifiedprimaryoutcomes.Inbothparts,acute excessmortalityinpatientswhohadsymptomaticintracerebral ischemicstrokepatientspresentingwithin3hoursof hemorrhage. Volume,. : February Annalsof EmergencyMedicine 227 ClinicalPolicy National Institutes of Health Stroke Scale. Level of consciousness 1a–1c: 6. Motor Leg: Raise leg to 30 degrees and hold for 5 1a. Alertness seconds; test both sides. 0=alert and responsive 0=No drift x 5 seconds 1=arousable to minor stimulation 1=Drift but does not hit bed 2=arousable only to painful stimulation 2=Some antigravity effort but cannot sustain 3=reflex responses or unarousable 3=No antigravity effort, but even minimal movement counts 1b. Orientation: Ask the patient his or her age and the 4=No movement at all month; answers must be exact. X=Unable to assess because of amputation, fusion, 0=Both correct fracture, etc 1=One correct (or dysarthria, intubated, foreign language) Left or Right 2=Neither correct 7. Limb Ataxia: Check finger to nose and heel to shin 1c. Commands: Ask the patient to open/close eyes and to (only scoring + if out of proportion to weakness). grip/release the nonparetic hand (or other 1-step command). 0=No ataxia (or aphasic, hemiplegic) Grip and release nonparetic 1=Ataxia in 1 limb 0=Both correct (OK if impaired by weakness) 2=Ataxia in 2 limbs 1=One correct X=Unable to assess because of amputation, fusion, 2=Neither correct fracture, etc Left or Right 2. Best Gaze: Only horizontal eye movements are checked by voluntary movement or reflective movement (Doll’s 8. Sensory: Use safety pin. eyes, not by calorics). Check grimace or withdrawal if stuporous. Score only 0=Normal stroke-related losses. 1=Partial gaze palsy 0=Normal 2=Forced eye deviation or total paresis that cannot be 1=Mild to moderate unilateral loss but patient aware of overcome by Doll’s eyes touch (or aphasic, confused) 2=Total loss, patient unaware of touch, coma, bilateral loss 3. Visual Field: Test using confrontation (or visual threat if necessary). 9. Best Language: Describe cookie jar picture, name 0=No visual loss objects, and read sentences (these standard items can be 1=Partial hemianopia, quadrantanopia, extinction found on the Web and at the American Heart Association 2=Complete hemianopia Web site). 3=Bilateral hemianopia or blindness (including cortical 0=Normal blindness) 1=Mild to moderate aphasia (partly comprehensible) 2=Severe aphasia (almost no information exchanged) 4. Facial Palsy: If stuporous, check symmetry of grimace 3=Mute, global aphasia, coma. to pain. 0=Normal 10. Dysarthria: Read list of words. 1=Minor paralysis, flat nasolabial fold or asymmetric smile 0=Normal 2=Partial paralysis (lower face) 1=Mild to moderate, slurred but intelligible 3=Complete paralysis (upper and lower face) 2=Severe, unintelligible or mute X=Intubation or mechanical barrier 5. Motor Arm: arms outstretched 90 degrees (patient sitting) or 45 degrees (patient supine) for 10 seconds. 11. Extinction/Inattention: Simultaneously touch patient Encourage patient for best effort. Assess both sides. on both hands, show fingers in both visual fields, ask 0=No drift x 10 seconds whether patient recognizes own left hand. 1=Drift but does not hit bed 0=Normal, none detected (visual loss alone) 2=Some antigravity effort but cannot sustain 1=Neglects or extinguishes to double simultaneous 3=No antigravity effort, but even minimal movement stimulation in any modality counts (visual, auditory, sensory, special or body parts) 4=No movement at all 2=Profound neglect in more than 1 modality, does not X=Unable to assess because of amputation, fusion, recognize own left hand fracture, etc The NIHSS is an 11-part scale that measures the neurologic examination in a codified manner. The scale ranges from 0 to 42. A score of less than 5 indicates a small stroke, and greater than 20 indicates a large stroke. Physicians can learn to perform the NIHSS on a training module on the Internet. Standard pictures (eg, the cookie jar picture) and lists of words can also be downloaded from the Internet. Figure 1. National Institutes of Health Stroke Scale. 228 Annalsof EmergencyMedicine Volume,. : February ClinicalPolicy Barthel Index.* Activity Feeding Toilet Use 0=unable 0=dependent 5=needs help cutting, spreading butter, etc or requires modified diet 5=needs some help, but can do something alone 10=independent 10=independent (on and off, dressing, wiping) Bathing Transfers (bed to chair and back) 0=dependent 0=unable, no sitting balance 5=independent (or in shower) 5=major help (1 or 2 people, physical), can sit 10=minor help (verbal or physical) Grooming 15=independent 0=needs help with personal care 5=independent face/hair/teeth/shaving (implements provided) Mobility (on level surfaces) 0=immobile or <50 yards Dressing 5=wheelchair independent, including corners, >50 yards 0=dependent 10=walks with help of 1 person (verbal or physical) >50 yards 5=needs help but can do about half unaided 15=independent (but may use any aid; for example, stick) >50 yards 10=independent (including buttons, zips, laces, etc) Stairs Bowels 0=unable 0=incontinent (or needs to be given enemas) 5=needs help (verbal, physical, carrying aid) 5=occasional accident 10=independent 10=continent TOTAL (0-100): Bladder 0=incontinent, or catheterized and unable to manage alone 5=occasional accident 10=continent *Mahoney FI, Barthel D. Functional evaluation: the Barthel Index. Maryland State Med J. 1965;14:56-61. Used with permission. The Barthel ADL Index: Guidelines 1. The index should be used as a record of what a patient does, not as a record of what a patient could do. 2. The main aim is to establish degree of independence from any help, physical or verbal, however minor and for whatever reason. 3. The need for supervision renders the patient not independent. 4. A patient's performance should be established using the best available evidence. Asking the patient, friends/relatives, and nurses are the usual sources, but direct observation and common sense are also important. However, direct testing is not needed. 5. Usually the patient's performance over the preceding 24 to 48 hours is important, but occasionally longer periods will be relevant. 6. Middle categories imply that the patient supplies over 50 percent of the effort. 7. Use of aids to be independent is allowed. The Barthel Index measures a person's ability to function in terms of the activities of daily living and mobility. It consists of 10 items, and scores range from 0 to 100. The higher the score, the more independent a patient is. Figure 2. Barthel Index. SecondarysubgroupanalysesofthecombinedNINDS StrokeTrial–Italy,9MulticenterAcuteStrokeTrial–Europe,10 partIandpartIIstudiesfailedtofindevidenceofadifferent andECASSI.11Allofthesestudiesfailedtodemonstratea effectoftPAaccordingtoage,sex,strokeseverity,andstroke benefitofthrombolysisforstroke,andsomewerehalted type.7 earlybecauseofexcessivemortalityinthetreatmentarm.9,10 In1995and1996,severalotherlargerandomizedtrialsof AllofthesestudiesweredifferentfromtheNINDSstudyinthat thrombolyticagentsinacuteischemicstrokewerepublished, theyuseddifferentthrombolyticagents(streptokinase),8-10 includingtheAustralianStreptokinasetrial,8MulticenterAcute differenttimeperiodsfortreatment(upto6hours),higher Volume,. : February Annalsof EmergencyMedicine 229 ClinicalPolicy Table1.ModifiedRankinScale.*(Usedwithpermission). protocolwasmodifiedandanewtrial,enrollingpatients0to5 Score Description hoursafterstrokeonset,wasbegun(ATLANTISPartB).In ATLANTISPartB,613patientswererandomized1:1to0.9 0 Nosymptoms mg/kgtPAorplacebo.14After31patientswereenrolled,the 1 Nosignificantdisabilitydespitesymptoms;abletocarryoutall usualdutiesandactivities timewindowwaschangedto3to5hoursaftersymptomonset 2 Slightdisability;unabletocarryoutallpreviousactivitiesbut becauseofFDAapprovalforIVtPAin1996.Theprimary abletolookafterownaffairswithoutassistance outcomewastheproportionofpatientswithanexcellent 3 Moderatedisability;requiringsomehelpbutabletowalk recovery,definedasanNIHSSscoreof0or1at90days.There withoutassistance 4 Moderatelyseveredisability;unabletowalkwithout wasnodifferenceintheprimaryoutcomebetweentPA-treated assistanceandunabletoattendtoownbodilyneeds patientsandplacebocontrols(34%versus32%;P(cid:1).65).Inthe withoutassistance tPA-treatedgroup,therewasahigherrateofsymptomatic 5 Severedisability;bedridden,incontinent,andrequiring intracerebralhemorrhage(7%versus1%;P(cid:2).001)andatrend constantnursingcareandattention towardhighermortality(11%versus6.9%;P(cid:1).09).14Themean 6 Dead timetotreatmentinthisstudywas4hours28minutes.Amongthe RankinJ.Cerebralvascularaccidentsinpatientsovertheageof60.II.Progno- sis.ScottMedJ.1957;2:200(cid:3)215.©Copyright1957RoyalSocietyofMedi- 61patientsrandomizedwithin3hours,ofwhom23were cinePress,UK. randomizedtotPAand38wererandomizedtoplacebo,moretPA- *ThemodifiedRankinScaleisa6-pointclinicaloutcomescalethatmeasuresa treatedpatientsachievedtheprimaryoutcome(61%oftPAversus patient’sfunctionandindependence.Alowerscoreindicatesabetteroutcome. 26%ofplacebo;P(cid:1).01)andhadsymptomaticintracerebral hemorrhage(13%oftPAversus0%ofplacebo;P(cid:1).05).15 Table2.GlasgowOutcomeScore.*(Usedwithpermission). TheNINDSpartIIstudyisthereforeuniqueinshowinga 5 GoodRecovery Resumptionofnormallifedespiteminor benefitinthepreselectedprimaryoutcomefor0.9mg/kgtPA deficits. forpatientswithischemicstrokeoflessthan3hours’duration.3 4 ModerateDisability Disabledbutindependent.Canworkin shelteredsetting. Thereproducibilityofthefindingissupportedbythereanalysis 3 SevereDisability Consciousbutdisabled.Dependentfor oftheNINDSstudy,whichfoundthat90-dayoutcomeswere dailysupport. againsignificantlyimproved,withoutadifferenceinmortality 2 Persistentvegetative Minimalresponsiveness rates.16Furthermore,aClassIIpatient-levelmeta-analysisthat 1 Death includesdatafromtheNINDS,ECASS,ATLANTIS,and JennettB,BondM.Assessmentofoutcomeafterseverebraindamage.Lancet. EchoplanarImagingThrombolyticEvaluationTrial 1975;1:480-484.©Copyright1975,withpermissionfromElsevier. *TheGlasgowOutcomeScoreisanothersimplemeasureoffunctionaloutcome. (EPITHET)studiesofpatientstreatedwithin3hoursalso supportstheefficacyoftPA.17Theincreasednumberof patientsinthismeta-analysisprovidedamoreprecise dosesoftPA(1.1mg/kg),11orallowedotherconcomitant estimateofthepotentialeffectoftreatment,andthe antithrombotics(aspirin).9 calculated95%CIssuggestedthattPA’sbenefitdiminished overtimebutremainedsignificantupto4.5hoursafter OtherrandomizedtrialsofIVtPA,usingthesamedosebutwith onsetofsymptoms.17 longertimeperiods,generatedmixedoutcomes.TheClassI TwoindependentgroupshavereanalyzedtheNINDStrial ECASSIItestedtPA(0.9mg/kg)versusplaceboinacuteischemic strokeoflessthan6hours’duration.12Theprimaryendpointwas data.First,anindependentcommitteewascommissionedbythe NINDStoverifythevalidityoftheNINDStrialresultsandto theproportionofpatientswithafavorableoutcomeonthe addresstheconcernthatanimbalanceinstrokeseverityat modifiedRankinScale,definedasascoreof0or1.Therewasno differenceinthisoutcomebetweentPA-treatedandplacebo baselinemayhaveconfoundedtheanalysisoftherelationship controlsintheoverallcohort(40%versus37%;P(cid:1).28)andin betweenIVtPAandthelikelihoodofagoodoutcome.16 patientstreatedwithin3hours(42%versus38%;P(cid:1).63), AlthoughthemedianbaselineNIHSSscorewasnotdifferentin althoughlessthan20%ofpatientsweretreatedwithinthattime thetPAandplacebogroups(P(cid:1).10),thereweremorepatients period.Parenchymalhemorrhageonposttreatmentcomputed inthe91-to180-minutestratumwithbaselineNIHSS0to5 tomography(CT)wasobservedin12%oftPAand3%ofplacebo whowererandomizedtotPAratherthanplacebo(29patients patients(P(cid:2).001).The90-daymortalityratewasequal(11%)for totPAversus7patientstoplacebo).Thecommitteefoundthat boththetPAandplacebogroups(P(cid:1).99). therelationshipbetweentPAuseandgoodoutcomeremained TheAlteplaseThrombolysisforAcuteNoninterventional robust(OR2.1;95%CI1.5to2.9)afteradjustmentfor TherapyinIschemicStroke(ATLANTIS)trialalsotestedIV baselineNIHSSandotherfactorsrelatedtostrokeoutcome, tPA(0.9mg/kg)versusplaceboinpatientswithstroke usingdatafromNINDSpartIandpartII.16Second,an symptomsoffewerthan6hours’duration.13Thetrialwas independentauthorgroupreanalyzedthedatawithgraphic stoppedprematurelyafterenrolling142patientsbecauseof analysisbutwithoutstatisticaltesting.18Theyconcludedthat increasedsymptomaticintracerebralhemorrhageinpatients tPAhadonlyasmalleffectonthechangeinNIHSSscore enrolled5to6hoursafterstrokesymptomonset.Thetrial betweenbaselineandday90.TheNIHSSchangewasnota 230 Annalsof EmergencyMedicine Volume,. : February ClinicalPolicy primaryoutcomeoftheNINDSpartIItrial,however,andthe States,Canada,andtheEuropeanUnionmandatedphaseIV authorsdidnotdisputethattPAhadastatisticallysignificant studiestodeterminewhetheroutcomesinclinicalpractice effectontheprimarytrialoutcome. matchedthoseachievedinthetrials.Single-center(orin1case, DatahavebeenaccumulatingaddressingtheuseofIVtPA singlesystem21)studiesfromearlyadopterssuggestedcausefor within3to4.5hoursafteronsetofsymptoms.Asnotedabove, concern,withmajorprotocolviolationsoccurringin9%to theClassIImeta-analysisofstudiesusing0.9mg/kgoftPA 67%oftreatedpatients.21-26Mostviolationswererelatedto confirmedabenefitfortPAwithin3hoursofonsetof timecriteria,bloodpressuremonitoringandcontrol,or symptomsandsuggestedthatthebenefitremainedsignificant provisionofantithromboticsoranticoagulantswithin24hours upto4.5hoursfromsymptomonset.17Thebenefitof0.9mg/ oftPAadministration.Somestudiesfoundthatprotocol kgtPAbetween3to4.5hoursaftersymptomonsetwasdirectly violationswereassociatedwithahigherrateofsymptomatic testedintheClassIECASSIIIrandomizedcontrolledtrial.19 intracerebralhemorrhage22andmortality.24 Thetrialusedthesamedosingregimenandinclusion/exclusion Thefirstlargepostmarketingmulticenterstudy,mandatedby criteriaastheNINDSprotocol(Figure3),withadditional theFDA,wastheClassIIIStandardTreatmentwithAlteplase exclusions:agegreaterthan80years,baselineNIHSSscore toReverseStroke(STARS)study.27Mostoftheparticipating greaterthan25,anyoralanticoagulantuse(regardlessofthe centershadpreviouslyenrolledpatientsinclinicaltrialsoftPA internationalnormalizedratio),andthecombinationofa forstroke.TheadministrationoftPAfollowedtheNINDS previousstrokeanddiabetesmellitus.Inaddition,incontrastto protocol.3OutcomesweresimilartothoseinthetPAarmofthe theNINDSprotocol,patientswerepermittedtoreceive NINDStrial(seeEvidentiaryTable).Twolargerregistriesfrom parenteralanticoagulantsforprophylaxisofdeepvenous CanadaandEuropefoundthattPAadministeredinclinical thrombosiswithinthefirst24hoursaftertreatmentwithtPA. practicehadratesofsymptomaticintracerebralhemorrhageof ThefrequencyoftheprimaryefficacyoutcomeinECASSIII 4%to5%andratesofdisabilityandmortalitysimilartothat (definedasmodifiedRankinScalescore0to1at90daysafter observedintheNINDStrial.28,29TheClassIICanadian treatment)wassignificantlygreaterwithtPA(291/418;52.4%) AlteplaseforStrokeEffectivenessStudy(CASES)28tracked thanplacebo(182/403;45.2%)(OR1.34;95%CI1.02to outcomesof1,135tPA-treatedpatients,whichtheauthors 1.76;riskratio1.16;95%CI1.01to1.34;P(cid:1).04).Mortality estimatedtorepresent84%ofalltreatedpatientsinCanada rateswereequivalent(7.7%fortPA-treatedpatientsversus duringthestudyperiod.Usingmultivariable-adjustedpredictive 8.4%forplacebo-treatedpatients).Symptomaticintracranial modeling,theauthorsfoundnodifferencebetweentheobserved hemorrhage,asdefinedbythecriteriausedintheNINDS rateofagoodoutcomeandtheexpectedratebasedonamodel study,wasreportedin33subjectstreatedwitht-PA(7.9%)and derivedfromtheNINDSdataset.TheClassIISafe in14subjectsgivenplacebo(3.5%)(OR2.38;95%CI1.25to 4.52;P(cid:1).006).Thehemorrhagerateswereslightlyhigherfor ImplementationofThrombolysisinStroke-MonitoringStudy (SITS-MOST)trackedoutcomesof6,442tPA-treatedpatients bothplaceboandtPA-treatedpatientscomparedwiththatin from285centersinEurope.29Theproportionwithgood theNINDSstudy,whichmaybeattributabletotheearlyuseof outcomewas38.9%,andsymptomaticintracerebral parenteraldeepvenousthrombosisprophylaxisallowedinthis hemorrhage,definedaccordingtocriteriausedinECASSII,12 study.ThebenefitinECASSIIIwasmoremodestthanthat was4.6%. observedintheNINDStrials,andthenumberneededtotreat TherearefewerdataontheuseoftPAinclinicalpracticein toachieve1excellentoutcomewas14inthisstudy.Thisis the3-to4.5-hourtimeperiod.TheClassIIISafe consistentwiththetPAmeta-analysiswithinthistimeframeand ImplementationofTreatmentsinStroke–InternationalStroke reinforcestheconceptthatearliertimetotreatmenthasalarge ThrombolysisRegistry(SITS-ISTR)3-to4.5-hourstudywasa impactonlikelihoodofgoodoutcomewithinanydefined timeframe.Therefore,althoughthetimewindowfortPA posthocassessmentofdataacquiredbetweenDecember2002 treatmentmayhavebeenlengthenedbasedontheECASSIII andFebruary2010fromanongoinginternationalregistry.30 results,theaggregatedatastronglysuggestthatpatientoutcomes Thisstudyreportedoutcomesin2,317patientstreatedwith willbeoptimizedbytheearliestpossibleadministrationoftPA tPAbetween3to4.5hoursafteronset.Mostpatientswere afterasafeandthoroughclinicalandbrainimagingevaluation. treatedafterpublicationoftheECASSIIItrialinOctober Thenotionthatthereis“plentyoftime”toevaluatepatients 2008.Therewere44.5%withgoodoutcome(modifiedRankin andadministertPAcouldleadtodelaysthatreducethe Scalescore0or1),whereas7.4%hadsymptomaticintracranial effectivenessofthedrug.20 hemorrhagebytheNINDStrialdefinitionand12.0%diedby Thesubstantialincreasedrateofsymptomaticintracerebral 3months.ComparedwiththeECASSIIItPA-treatedarm,the hemorrhageamongtPA-treatedpatientshastempered proportionwithgoodoutcomewassomewhatlowerandthe enthusiasmfortherapidadoptionoftPAasroutinecare,inpart proportionwithmortalitywassomewhathigher,probably becauseoftheconcernthattreatmentmaybelesssafeinroutine becausepatientsintheSITS-ISTRregistryhadhigherinitial clinicalpracticethaninthehighlymonitoredsettingofa strokeseverityandmoremedicalcomorbiditiesthanthe clinicaltrial.Asaresult,regulatoryagenciesintheUnited patientsenrolledintheECASSIIItrial. Volume,. : February Annalsof EmergencyMedicine 231 ClinicalPolicy NINDS and ECASS III inclusion and exclusion criteria for intravenous tPA for acute ischemic stroke. NINDS Criteria3 ECASS III Criteria19 Inclusion: Inclusion: Acute ischemic stroke with clearly defined time Acute ischemic stroke with a clearly defined of onset (who could be treated <3 hours of time of onset (who could be treated between 3- symptom onset) 4.5 hours from symptom onset) Measurable deficit on the NIH stroke scale Age 18-80 years Baseline brain CT scan that showed no evidence Stroke symptoms present for at least 30 minutes of hemorrhage. without significant improvement prior to treatment. Exclusion:* Baseline brain imaging that showed no evidence Another stroke or serious head injury within the of hemorrhage. preceding 3 months Major surgery within prior 14 days Exclusion:* History of intracranial hemorrhage Same as NINDS plus the following additional Systolic BP >185 mm Hg or diastolic BP >100 criteria: mm Hg Age >80 years Rapidly improving or minor symptoms Severe stroke (NIHSS >25) or by appropriate Symptoms suggestive of subarachnoid imaging techniques (defined as >1/3 of the hemorrhage middle cerebral artery territory) Gastrointestinal or genitourinary hemorrhage Combination of previous stroke and diabetes within the previous 21 days mellitus Arterial puncture at a noncompressible site Any oral anticoagulant use (regardless of INR within the previous 7 days or PT). Seizure at onset of stroke Use of anticoagulation: *Exclusions (or cautions) to tPA use that patients receiving heparin within the 48 were not specifically mentioned in either hours preceding the onset of stroke who have an study but are generally used: elevated PTT, Myocardial infarction within previous 3 months patients with a PT >15 seconds (or INR (AHA 2007 guidelines) >1.6), Pregnancy and early postpartum period patients with a platelet count <100,000 Known bleeding diathesis, recent pericarditis, Glucose level of <50 mg/dL or >400 mg/dL. recent lumbar puncture (Brain Attack Coalition http://www.stroke- site.org/guidelines/tpa_guidelines.html, accessed March 1, 2012). AHA, American Heart Association; BP, blood pressure; CT, computed tomography; ECASS, European Cooperative Acute Stroke Study; INR, International Normalized Ratio; NIH, National Institutes of Health; NIHSS, National Institutes of Health stroke scale; NINDS, National Institute of Neurological Disorders and Stroke; PT, prothrombin time; PTT, partial thromboplastin time; tPA, tissue plasminogen activator. Figure 3. NINDS and ECASS III inclusion and exclusion criteria for intravenous tPA for acute ischemic stroke. PuttingtheEvidenceIntoClinicalContext stroke,includingrapidaccesstolaboratorytestresults,brain SafeandeffectiveadministrationoftPAreliesonahospital imaging,andaccurateimageinterpretation.Protocolsmustbe havingasysteminplacefortreatingstrokepatients.Patients inplacefordrugadministration,closeclinicalmonitoring, mustundergorapidandaccuratediagnosisofacuteischemic activebloodpressuremanagement,andtreatmentof 232 Annalsof EmergencyMedicine Volume,. : February ClinicalPolicy hemorrhagiccomplications(systemicorintracerebral)ifthey “hypodensitygreaterthanonethirdofthecerebral occur.Ifagivenhospitalisunabletoprovidethisinfrastructure, hemisphere”anda“caution”forthepresenceofmajor protocolsshouldbeinplacefortransferringpatientstoafacility deficits,40andtheAmericanCollegeofChestPhysicians thatcan.Whateverahospital’sapproachis,anongoingquality guidelinesrecommendagainsttreatmentwhenclearly assuranceprogramoughttobeinplace.Physicianexpertiseand identifiablehypodensityispresentingreaterthanonethird writtenprotocolsarethereforehypothesizedtobeimportantfor ofthemiddlecerebralarteryterritorywhilenotdisallowing useoftPA31butmaybeinshortsupplyinsmallercenters treatmentinthepresenceofearlyischemicchangessuchas withoutanabundanceofstrokespecialists.TheCASESand subtlelossofgray-whitedifferentiationorsulcaleffacement SITS-MOSTstudies,whichincludedabroadselectionof withouthypodensity.41Patientswiththesecharacteristics academicandcommunityhospitals,showedresultssimilarto havebeenexcluded,underrepresented,ornotreportedonin thoseobservedintheNINDStrial.Additionally,bothstudies themajorobservationalstudies;therefore,dataonoutcomes failedtofindadifferenceinoutcomesinpatientstreatedat inthesepatientsubgroupsinclinicalpracticearelacking. moreexperiencedcenters,definedbytPAcasevolume, Theexceptionisadvancedage,forwhichseveralstudies comparedwithlessexperiencedcenters.TheSITS-MOST reportgenerallyworseoutcomescomparedwithyounger subjectsbutnoincreasedriskofsymptomaticintracerebral findingsmustbetreatedwithsomecaution,however,because hemorrhage.42-44Thisisnotasurprisingfinding,giventhatage allcenterswererequiredtohaveaneurologistorotherphysician with“considerableexperienceinstrokecare.”29Adequate isawell-establishedriskfactorforpooroutcomeregardlessof intervention.Althoughitisappropriatetoexercisecaution physicianacutestrokecareexpertisehasnotbeenrigorously whenconsideringtreatmentforthesesubgroupswithpoor definedintheliterature,basedoneithercredentialordegreeof prognosis,aposthocanalysisofthe1995NINDStrialfailedto experience,orstudiedinclinicaltrials.Thedefinitionshould showevidenceofadifferentialeffectoftPAaccordingtopatient notberestrictedtoneurologistsandshouldincludeemergency subgroups,includingthosewithadvancedage,severeclinical physiciansorotherphysicianswithexpertiseandexperiencein deficits,andmoreextensiveCTchanges.45 strokecare,accordingtorecommendationsfromtheBrain AttackCoalition31andtheCanadianStrokeConsortium.32 Forcenterswithouton-siteacutestrokespecialists,telestroke Addendum technologyoffersameanstoobtainremoteconsultationabout Afterthisdocumentwascompleted,theInternational theadministrationofIVtPA.InastudybyFisher,33the StrokeTrial3(IST-3)waselectronicallypublishedin formationof“telestroke”networksallowedinexperienced Lancet.46IST-3wasdesignedtoevaluatetheeffectsoftPA centerstoobtainexpertmedicalandradiologicconsultationby onpatientswithischemicstrokeupto6hoursfrom remotevideolinkage.Accumulatingdatashowthatthismodel symptomonsetinwhombenefitwasdeemedtobeuncertain ofstrokecareproducesresultssimilartothoseobtainedbyon- (thevastmajorityofwhomhadcontraindicationstotPA siteconsultationwithstrokeexperts.34-36AClassIIIstudyfrom definedbyNINDScriteriainthe0-to3-hourwindowor anetworkofhospitalsinBavaria,Germany,foundthat115 ECASS-3criteriainthe3-to4.5-hourwindow).IST-3 patientstreatedwithtPAatremotesitesusingtelestrokehad lookedatadifferentcohortofpatientsthanthoseonwhich similarinhospitalratesofsymptomatichemorrhage(7.8%)and thispolicyfocuses.Thepublishedtrialdatawerecarefully mortality(3.5%)comparedwithlocallytreatedpatientsatthe reviewedbythewritingpanel,anditwasdeterminedthatthe academicstrokecenters.37Arandomizedcontrolledtrialshowed study’smethodologywassuchthatthefindingsdidnotaffect thatmoreaccuratedecisionsaremadewhenvideoconsultation, therecommendationsmadeinthispracticeguideline. ratherthantelephoneconsultation,isused.38TheAmerican HeartAssociationpublishedrecommendationsontheuseof Disclosures telemedicineforacutestrokecare.39 Dr.EdlowservesontheExecutiveCommitteeofthe Therehasbeenclinicalconcernabouttreatmentof FoundationforEducationandResearchinNeurologic patientgroupswhowouldmeetNINDScriteriabuthavea Emergencies,coeditedthetextbookofNeurological poorprognosisforgoodoutcome,irrespectiveoftPAuse, EmergenciesforOxfordUniversityPress,andservesonthe includingthosewithadvancedage,severeclinicaldeficits, editorialboardoftheJournalofInternalandEmergency andCThypodensityinalargeportionofthemiddlecerebral MedicineandtheInternationalJournalofEmergency arteryterritoryorhemisphere.TheSITS-MOSTandSITS- Medicine.Healsoreviewsmedicalmalpracticecasesforboth ISTRtreatmentprotocolexcludedpatientsolderthan80 plaintiffanddefense. years,withNIHSSscoregreaterthanorequalto25,orwith Dr.Smithservedonascientificadvisoryboardfor “severestroke”onCT.29,30TheCanadianguidelineslistCT Genentechin2010,receivedspeakerhonorariafromthe evidenceofinfarctioninvolvingmorethanonethirdofthe CanadianConferenceonDementia,servesasanassistanteditor middlecerebralarteryterritoryasanexclusioncriterion.32 forStroke,hasservedonspeakers’bureausforQuantiaMDand TheAmericanHeartAssociation/AmericanStroke BMJBestPractice,isontheDataandSafetyMonitoringBoard AssociationguidelinesincludethatCTdoesnotshowa fortheMRWitnesstrialfundedbytheNationalInstitutesof Volume,. : February Annalsof EmergencyMedicine 233 ClinicalPolicy Health(NIH)/NINDS,andreceivesresearchsupportfromthe methodandapplicationtothrombolytictherapyforacutestroke. NIH/NINDS,CanadianInstitutesforHealthResearch, ArchNeurol.2004;61:1066-1070. 7. TheNINDSt-PAStrokeStudyGroup.Generalizedefficacyoft-PA CanadianStrokeNetwork,theAlbertaHeritageFundfor foracutestroke.SubgroupanalysisoftheNINDSt-PAstroketrial. MedicalResearch,andtheHeartandStrokeFoundationof Stroke.1997;28:2119-2125. Canada. 8. DonnanGA,DavisSM,ChambersBR,etal.Streptokinasefor Dr.Steadiseditor-in-chiefoftheInternationalJournalof acuteischemicstrokewithrelationshiptotimeofadministration: EmergencyMedicine. AustralianStreptokinase(ASK)TrialStudyGroup.JAMA.1996; Dr.Gronsethservesasaneditorialadvisoryboardmemberof 276:961-966. 9. MulticentreAcuteStrokeTrial–Italy(MAST-I)Group.Randomised NeurologyNow,servesonaspeakers’bureauforBoehringer controlledtrialofstreptokinase,aspirin,andcombinationofboth Ingelheim,andreceiveshonorariafromBoehringerIngelheim intreatmentofacuteischaemicstroke.Lancet.1995;346:1509- andtheAmericanAcademyofNeurology. 1514. Dr.MesséreceivespublishingroyaltiesfromUp-To-Date, 10. TheMulticenterAcuteStrokeTrial–EuropeStudyGroup. formerlyservedonthespeakers’bureauforBoehringer Thrombolytictherapywithstreptokinaseinacuteischemicstroke. Ingelheim(ended4/2011),andreceivesresearchsupportfrom NEnglJMed.1996;335:145-150. Gore,theNIH(NIDDK,U01-DK060990,Endpoint 11. HackeW,KasteM,FieschiC,etal.Intravenousthrombolysis withrecombinanttissueplasminogenactivatorforacute AdjudicationCommittee),NationalHeart,Lung,andBlood hemisphericstroke.TheEuropeanCooperativeAcuteStroke Institute(NHLBI)(1R01HL084375-01A2,subinvestigator, Study(ECASS).JAMA.1995;274:1017-1025. neurologicassessments),NINDS(U01NS40406-04,local 12. HackeW,KasteM,FieschiC,etal.Randomiseddouble-blind principalinvestigator),andNIH(HHSN268200800003C, placebo-controlledtrialofthrombolytictherapywithintravenous backupmedicalmonitor). alteplaseinacuteischaemicstroke(ECASSII).SecondEuropean- Dr.JagodaservesontheexecutiveboardfortheBrainAttack AustralasianAcuteStrokeStudyInvestigators.Lancet.1998;352: 1245-1251. CoalitionandfortheFoundationforEducationandResearch 13. ClarkWM,AlbersGW,MaddenKP,etal.ThrombolyticTherapyin inNeurologicEmergencies,servesontheadvisoryboardforthe AcuteIschemicStrokeStudyInvestigators.ThertPA(alteplase)0- BrainTraumaFoundation,andisaconsultantforBanyan to6-houracutestroketrial,partA(A0276g):resultsofadouble- Biomarkers,Cyvek,Pfizer,andGORE.Heisalsoeditor-in- blind,placebo-controlled,multicenterstudy.Stroke.2000;31: chiefofEmergencyMedicinePracticeandservesontheeditorial 811-816. boardsforPediatricEmergencyMedicinePractice,Emergency 14. ClarkWM,WissmanS,AlbersGW,etal.AlteplaseThrombolysis MedicinePracticeGuidelines,EMCriticalCare,Annalsof forAcuteNoninterventionalTherapyinIschemicStrokeStudy Investigators.Recombinanttissue-typeplasminogenactivator EmergencyMedicine,ACEPNews,andAustralasianJournalof (alteplase)forischemicstroke3to5hoursaftersymptomonset. EmergencyMedicine. TheATLANTISStudy:arandomizedcontrolledtrial.JAMA.1999; Dr.WearsservesontheboardofdirectorsoftheEmergency 282:2019-2026. MedicinePatientSafetyFoundation,ontheeditorialboardfor 15. AlbersGW,ClarkWM,MaddenKP,etal.ATLANTIStrial:results AnnalsofEmergencyMedicine,andontheeditorialboardfor forpatientstreatedwithin3hoursofstrokeonset.Stroke.2002; HumanFactorsandInternationalJournalofRiskandSafetyin 33:493-496. 16. IngallTJ,O’FallonWM,AsplundK,etal.Findingsfromthe Medicine. reanalysisoftheNINDStissueplasminogenactivatorforacute Dr.DeckerservesastrusteeandVicePresident,Mayo ischemicstroketreatmenttrial.Stroke.2004;35:2418-2424. Clinic,CEOforMayoClinicinScottsdale,AZ. 17. LeesKR,BluhmkiE,vonKummerR,etal.Timetotreatmentwith intravenousalteplaseandoutcomeinstroke:anupdatedpooled analysisofECASS,ATLANTIS,NINDS,andEPITHETtrials.Lancet. 2010;375:1695-1703. REFERENCES 18. HoffmanJR,SchrigerDL.AgraphicreanalysisoftheNINDStrial. 1. Lloyd-JonesD,AdamsRJ,BrownTM,etal.AmericanHeart AnnEmergMed.2009;54:329-336. AssociationStatisticsCommitteeandStrokeStatistics 19. HackeW,KasteM,BluhmkiE,etal.ECASSInvestigators. Subcommittee.Heartdiseaseandstrokestatistics—2010 Thrombolysiswithalteplase3to4.5hoursafteracuteischemic update:areportfromtheAmericanHeartAssociation.Circulation. stroke.NEnglJMed.2008;359:1317-1329. 2010;121:e46-e215. 20. DelZoppoGJ,SaverJL,JauchEC,etal.AmericanHeart 2. SchrigerDL,CantrillSV,GreeneCS.Theorigins,benefits,harms AssociationStrokeCouncil.Expansionofthetimewindowfor andimplicationsofemergencymedicineclinicalpolicies.Ann treatmentofacuteischemicstrokewithintravenoustissue EmergMed.1993;22:597-602. plasminogenactivator:ascienceadvisoryfromtheAmerican 3. NationalInstituteofNeurologicalDisordersandStrokert-PA StrokeStudyGroup.Tissueplasminogenactivatorforacute HeartAssociation/AmericanStrokeAssociation.Stroke.2009; ischemicstroke.NEnglJMed.1995;333:1581-1587. 40:2945-2948. 4. MitkaM.TensionsremainovertPAforstroke.JAMA.2003;289: 21. WangDZ,RoseJA,HoningsDS,etal.Treatingacutestroke 1363-1364. patientswithintravenoustPA.TheOSFStrokeNetwork 5. LydenPD,LeesKR,DavisSM.Alteplaseforacutestroke experience.Stroke.2000;31:77-81. revisited:thefirst10years.LancetNeurol.2006;5:722-724. 22. Lopez-YunezAM,BrunoA,WilliamsLS,etal.Protocolviolations 6. SaverJL.Numberneededtotreatestimatesincorporatingeffects incommunity-basedrTPAstroketreatmentareassociatedwith overtheentirerangeofclinicaloutcomes:novelderivation symptomaticintracerebralhemorrhage.Stroke.2001;32:12-16. 234 Annalsof EmergencyMedicine Volume,. : February