Extending Reperfusion Therapy for Acute Ischemic Stroke: Emerging Pharmacological, Mechanical, and...

Section Editors: Marc Fisher, MD, and Antoni Davalos, MD

Extending Reperfusion Therapy for Acute Ischemic StrokeEmerging Pharmacological, Mechanical, and Imaging Strategies

Carlos A. Molina, MD; Jeffrey L. Saver, MD

Background and Purpose—Reperfusion is the most beneficial of all therapeutic strategies for acute ischemic stroke.However, the standard cerebral reperfusion treatment of the first decade of the reperfusion era, noncontrast computedtomography (CT)–guided, �3 hours, intravenous tissue plasminogen activator, has many limitations. This reviewsurveys emerging strategies that have the potential to extend cerebral reperfusion therapy to larger numbers of patients.

Summary of Review—Innovative intravenous pharmacological reperfusion strategies include novel fibrinolytic agents(tenecteplase, reteplase, desmetolplase, plasmin, and microplasmin), glycoprotein (GP) IIb/IIIa antagonists with plateletdisaggregating effects (abciximab and tirofiban), combination therapies to improve efficacy of clot lysis (fibrinolyticsand GP IIb/IIIa agents, and fibrinolytics and direct thrombin inhibitors), increase the time window for clot lysis(fibrinolytics and neuroprotectants), and reduce the frequency of hemorrhagic complications (fibrinolytics andvasoprotectants), and externally applied ultrasound to enhance enzymatic fibrinolysis. Promising intra-arterialpharmacological reperfusion approaches include novel fibrinolytic agents, combined intravenous and intra-arterialfibrinolysis, and combined fibrinolytics and GP IIb/IIIa agents. Emerging endovascular mechanical reperfusionstrategies include intra-arterial thrombectomy (clot retrieval devices and suction thrombectomy devices), mechanicaldisruption (micro-guidewire passage, laser photoacoustic emulsification, and primary intracranial angioplasty), andaugmented fibrinolysis by endovascular ultrasound. Multimodal imaging, with magnetic resonance (MR) or CT, canrapidly assess infarct core, penumbra, site of vessel occlusion, and tissue hemorrhagic propensity, enabling improvedselection of patients for reperfusion therapy beyond any arbitrary fixed time window.

Conclusions—Therapeutic reperfusion is emerging as a treatment strategy of remarkable power and scope for rescuingpatients experiencing acute brain ischemia, applicable within and beyond the 3-hour time window. (Stroke. 2005;36:2311-2320.)

Key Words: endovascular therapy � reperfusion � stroke, acute � thrombolysis

Reperfusion of the ischemic brain is the most effectivetherapy for acute ischemic stroke ever known and

ever likely to be discovered. By restoring nutritional bloodflow to threatened tissues before they progress to infarc-tion, reperfusion therapies salvage penumbral tissue, re-duce final infarct size, and enable improved clinicaloutcomes. In the decade since the advent of reperfusion asa proven treatment strategy in acute ischemic stroke,accumulated research data and clinical experience haveconfirmed the dramatic benefit of early cerebral reperfu-sion.1 Intravenous fibrinolytic therapy within 3 hours ofonset yields a benefit at least an order of magnitude greaterthan aspirin, the only other widely available pharmacolog-ical agent of proven efficacy for ischemic stroke. Amongpatients matching the population in the pivotal National

Institute of Neurological Disorders and Stroke (NINDS)trials, the number needed to treat for benefit is �3.1.2 Forevery 1000 patients treated, �323 will attain a betteroutcome. A worldwide consensus recognizing the efficacyof reperfusion therapy for stroke now exists, with within–3-hour intravenous tissue plasminogen activator (tPA)approved by independent drug regulatory authorities in theUnited States, Canada, South America, Australia, and theEuropean Union.

However, the standard reperfusion strategy of the firstdecade of the reperfusion era, noncontrast computed tomog-raphy (CT)–guided, �3 hours, intravenous tPA, has manylimitations, including a short treatment time window,achieved recanalization rates of only �50%, and a substantialrisk of symptomatic hemorrhagic transformation. As a result,

Received November 2, 2004; final revision received February 23, 2005; accepted April 21, 2005.From the University of California (J.L.S.), Los Angeles; and Hospital Universitar Vall d’Hebron (C.M.), Barcelona, Spain.Correspondence to Jeffrey Saver, MD, Department of Neurology, University of California, UCLA Stroke Center, 710 Westwood Plaza, Los Angeles,

CA 90095-1769. E-mail [email protected]© 2005 American Heart Association, Inc.

Stroke is available at http://www.strokeaha.org DOI: 10.1161/01.STR.0000182100.65262.46

2311

Emerging Therapies

by guest on January 12, 2016http://stroke.ahajournals.org/Downloaded from

http://stroke.ahajournals.org/

few (typically 1% to 3%) patients currently receive reperfu-sion therapies in actual practice.

Several emerging strategies have the potential to extendcerebral reperfusion therapy to larger numbers of patients,including patients presenting beyond the current 3-hour timewindow. This review highlights recent advances shaping thecoming era of expanded reperfusion treatments for acuteischemic stroke.

Extending the Time Window for ConventionalIntravenous tPA

Pooled individual patient data analysis indicates that intrave-nous tPA may be of modest benefit beyond 3 hours inrelatively unselected ischemic stroke patients. The ongoingECASS-3 trial will confirm or disconfirm this finding in the3- to 4-hour time window by enrolling 800 patients in a fullydouble-blind, placebo-controlled study. The InternationalStroke Trial 3 (IST 3) will also provide some data of interest,although lack of blinding and selection criteria that will allowenrollment of patients at hospitals that have not clearly madean institutional commitment to the safe delivery ofthrombolytic therapy will make study interpretation difficult.

Novel Fibrinolytic AgentsThe failure of tPA to achieve rapid reperfusion in manypatients and its bleeding risk have prompted the developmentof more fibrinolytic agents with greater fibrin specificity andbetter risk/benefit profiles. Novel agents that achieve higherrecanalization rates, lower hemorrhage rates, or both, wouldextend the time window in which intravenous fibrinolytictherapy is beneficial.

Tenecteplase (TNK) is a genetically modified form of tPAthat has 14-fold greater fibrin specificity, longer half-life, and80-fold greater resistance to inhibition by plasminogen acti-vator inhibitor type 1.3 The long lifetime of TNK allows theuse of a single-bolus administration. High fibrin specificityshould confer the ability to induce faster and more completeclot lysis, with less bleeding complications. TNK administra-tion has been demonstrated to avoid the systemic plasmino-gen activation and plasmin generation commonly seen aftertPA therapy. Further, the lack of a procoagulant effectexhibited by TNK may reduce early reocclusion. In compar-ative trials in myocardial infarction (MI) patients, TNKshowed equivalent efficacy to tPA, a similar rate of intracra-nial hemorrhage (ICH), fewer noncerebral bleeding compli-cations, and less need for blood transfusion.3

In a pilot dose-escalating study, 75 stroke patients weretreated with intravenous TNK �3 hours after symptomonset.4 Patients were enrolled in 3 dose tiers of TNK: 0.1, 0.2,and 0.4 mg/kg. No case of symptomatic ICH was observedduring the first 72 hours after treatment. However, a dose-response relationship between TNK and neurological im-provement at 24 hours was not demonstrated. Currently, TNKfor the treatment of acute ischemic stroke is being investi-gated in a larger phase 2 trial.

Desmetoplase is 1 of 4 distinct proteases found in thesaliva of the blood-feeding vampire bat Desmodus rotundus,collectively referred to as D rotundus salivary plasminogenactivators (DSPAs). Desmetoplase is the �-1 variant among

the DSPAs and exhibits �72% amino acid sequence identitywith human tPA. Unlike human tPA, DSPA�-1 exists as asingle-chain molecule, and its catalytic activity is exquisitelydependent on the presence of fibrin as cofactor. In models ofarterial thrombosis, DSPA�-1 induces faster and more sus-tained recanalization than tPA and produces less antiplasminconsumption and fibrinogenolysis. Moreover, unlike tPA,DSPA�-1 does not enhance N-methyl-D-aspartate–mediatedneurodegeneration.5 Desmoteplase has shown promise in 2phase 2 ischemic stroke trials enrolling patients 3 to 9 hoursafter onset when a MRI diffusion–perfusion mismatch patternis present.

Reteplase is a recombinant peptide that consists of kringle2 and protease domains of human tPA. The long half-life ofreteplase allows administration as a double-bolus injection.Reteplase produces rapid and effective coronary arterythrombolysis. Although easier to administer, reteplase did notprovide an additional survival benefit compared with anaccelerated infusion of alteplase in the treatment of acute MI.6

In a small prospective study in stroke patients, Qureshi et aldemonstrated that reteplase given intra-arterially up to 9hours after symptom onset, with or without angioplasty,resulted in a high rate of complete recanalization.7

Plasmin and microplasmin, a truncated form of plasmin,are emerging fibrinolytic agents. Standard plasminogen acti-vating drugs depend on the local availability of plasminogento generate active, fibrin-digesting, plasmin. In contrast,plasmin and microplasmin act directly on fibrin. Becausehuman plasmin is rapidly inactivated by circulating antiplas-min, it is potentially very useful as a local, intra-arteriallyapplied therapeutic agent8 but not suitable for use as anintravenous therapeutic agent. Microplasmin retains the pro-tease domain of plasminogen and is resistant to rapid inacti-vation by antiplasmin, rendering it suitable for considerationfor intravenous application. In rabbit small and large clotembolic stroke models, microplasmin infusion resulted in ahigh rate of clot lysis and, unlike tPA and TNK, did notincrease the rate of intracranial bleeding compared withcontrol animals. Moreover, microplasmin showed nonlytic-dependent neuroprotective effects improving behavioral rat-ing scores.9 Given its combined thrombolytic and neuropro-tective properties, microplasmin is an attractive stroketherapy candidate.

Glycoprotein IIb/IIIa AntagonistsGlycoprotein (GP) IIb/IIIa antagonists potently block theplatelet GP IIb/IIIa receptor, the final mediator of aggrega-tion. GP IIb/IIIa antagonists reduce thrombus growth andprevent reocclusion after mechanical or lytic-driven recana-lization. Moreover, GP IIb/IIIa antagonists have the ability todissolve platelet-rich clots and to improve flow in coronaryand cerebral microcirculation.

Abciximab is the Fab fragment of a chimeric human/mouseantibody directed against the platelet GP IIb/IIIa receptor.Abciximab administration at a bolus dose of 0.25 mg/kgfollowed by a continuous infusion for 12 hours, rapidlyproduces a profound hemostatic effect, with blockade of 80%of GP IIb/IIIa receptors, marked reduction of platelet aggre-gation, and prolongation of the bleeding time. The combina-

2312 Stroke October 2005



tion of abciximab, aspirin, and adjusted-dose heparin inducesa high rate (up to 50%) of coronary artery recanalization. Inthe AbESTT phase 2b trial, 400 patients were randomized toabciximab or control within 6 hours of observed stroke onsetor 3 hours of awakening with stroke; �50% were treated 3 to5 hours after onset. Abciximab showed a reasonable safetyprofile, with an ICH rate of 3.6%. A signal of potentialefficacy was identified, with favorable functional outcome(modified Rankin Scale [mRS] score 0 to 1) in 48% ofabciximab versus 40% of placebo patients (P�0.087).10

Abciximab is being investigated currently in a phase 3international trial (AbESTT II) enrolling 1800 patients.

Tirofiban is a tyrosine-derived nonpeptide molecule that ishighly specific for GP IIb/IIIa receptor. Tirofiban appearsparticularly suited for platelet disaggregation, given its hightargeted receptor specificity, and has a long, 1.6-hour half-life. Pilot data indicate that intravenous tirofiban can be safelyadministered in acute stroke patients. In an open-label pilotstudy, 18 patients with progressively deteriorating acuteischemic stroke were treated with body weight–adjustedintravenous tirofiban for a mean period of 46 hours.11 Nomajor ICH was observed, and the rate of asymptomatic ICHon CT was comparable to that observed in matched controls.Moreover, treatment with tirofiban was associated with asmaller 1-week MR infarct size compared with matchedcontrols.12 SaTIS (Safety of Tirofiban in Acute IschemicStroke) is an ongoing phase 2 multicenter, prospective,randomized, placebo-controlled safety trial of intravenoustirofiban in 240 stroke patients with National Institutes ofHealth Stroke Scale (NIHSS) score of 4 to 18 and treatmentinstituted in an extended time window up to 22 hours afteronset.

Combined Pharmacological ApproachesCombination pharmacotherapy strategies to expand the intra-venous fibrinolysis time window beyond 3 hours are underactive investigation. A rational combination of agents withadditive effects on clot lysis and clot formation may yieldhigher rates of arterial recanalization, lower rates of reocclu-sion, reductions in the dose of fibrinolytic agent required, andreduced frequency of hemorrhage transformation. Combiningneuroprotective therapies with fibrinolytics may potentiatetreatment benefit and extend the time window in whichsalvageable tissue persists to be rescued by reperfusion.Coadministering agents that block blood–brain barrier deg-radation may markedly reduce hemorrhagic complications offibrinolysis, permitting extension of therapy to a wider rangeof patients.

Lytics and AntithromboticsCombination therapy with fibrinolytic and GP IIb/IIIa agentsis under wide-ranging investigation.

In a series of studies, the Dusseldorf group treated up to 37patients within 3 hours of onset with reduced doses ofintravenous tPA (typically 20 mg) and a 24-hour infusionof tirofibran. Combined therapy resulted in a high rate (68%)of middle cerebral artery (MCA) recanalization on MRangiography, greater salvage of perfusion MR–defined tissue

at risk, and better clinical outcome than standard intravenoustPA.13,14 Low rates of symptomatic ICH were observed.

A pilot study in 27 patients found combining abciximabwith low-dose tPA (0.45 mg/kg) appeared safe and resulted inhigher rates of MCA recanalization compared with full-dosetPA alone.15

A combined thrombolytic regimen with reteplase withabciximab in MI and peripheral artery thrombosis patientsyields faster, more consistent, and sustained reperfusion, anda decreased rate of distal embolization.16 In stroke, prospec-tive trials under way include a 20-patient, dose-escalationsafety trial of intra-arterial (IA) reteplase and intravenousabciximab administered 3 to 6 hours after onset (A Qureshi,personal communication, 2004) and a 72-patient, dose-ranging safety trial of intravenous reteplase and intravenousabciximab 3 to 24 hours after onset ReoPro Retavase Reper-fusion of Stroke Safety Study—Imaging Evaluation (ROSIE/ROSIE-CT trials).

Eptifibatide is a highly selective GP IIb/IIIa antagonistcurrently being tested in combination with tPA within 3 hoursof onset in a multicenter phase 2 dose-escalation studyenrolling 100 patients (CLEAR). In addition, a randomizedopen-label, dose-escalation and safety trial of combinedadministration of tPA, eptifibatide, aspirin, and tinzaparin instroke patients �3 hours is under way (ROSIE-2).

Argatroban is a synthetic direct thrombin inhibitor. Block-ing thrombin inhibits fibrin formation in the thrombus andreduces platelet aggregation in the microcirculation. In con-junction with tPA, argatroban may enhance clot lysis, preventreocclusion, and limit the no-reflow phenomenon in themicrocirculation. Argatroban alone in human stroke appearedrelatively safe, although without a strong signal of potentialefficacy in the 171-patient ARGIS-1 trial. The NIH-sponsored tPA Argatroban Stroke Study (TARTS) is inves-tigating the combination of argatroban and tPA in a pilotdose-escalating safety trial in 40 patients with a documentedMCA occlusion on transcranial Doppler (TCD) within 3hours.17

Lytics and NeuroprotectantsNeuroprotective therapies have been shown to be moreeffective in animal models of ischemia when administrationis followed by reperfusion rather than persisting occlusion.Further, the effects of reperfusion injury may be limited orreversed by adding neuroprotectants to reperfusion strategies.Hypothermia probably represents the most potent neuropro-tectant currently under study. The COOL-AID (Cooling forAcute Ischemic brain Damage) phase 2 trial in stroke patientswithin 12 hours of onset demonstrated that hypothermia waswell tolerated in most patients,and a trend to attenuation ofdiffusion-weighted imaging (DWI) lesion growth was seen inhypothermic patients. Endovascular cooling to 33°C seems tobe feasible and safe in nonanesthetized stroke patients, evenin those treated with thrombolysis.18

By stabilizing threatened brain tissue, early neuroprotec-tive therapy may extend the time window for subsequenteffective administration of reperfusion agents. However, inmost human trials performed of combined neuroprotectionand thrombolytic therapy, neuroprotective interventions have

Molina and Saver Extending Reperfusion Therapy for Acute Stroke 2313



been initiated in hospital only after the start of intravenoustPA. In the FAST-MAG (Field Administration of StrokeTherapy—Magnesium) pilot trial, paramedic initiation ofmagnesium sulfate neuroprotective therapy in stroke patientsin the field was shown to be feasible and appeared safe.19

Among the 20 enrolled patients, 2 received subsequentin-hospital reperfusion interventions without hemorrhagiccomplication. In the NIH-funded FAST-MAG Phase 3 ran-domized trial, paramedics are initiating magnesium sulfate orplacebo in 1298 patients within 2 hours of stroke onset in alland within 1 hour of onset in approximately half. TheFAST-MAG Trialists anticipate that �20% of enrolled pa-tients will receive a Food and Drug Administration (FDA)–approved reperfusion intervention (intravenous tPA or MerciRetriever) on hospital arrival, providing substantial statisticalpower to explore whether hyperacute neuroprotection poten-tiates the benefits of subsequent reperfusion therapy.

Lytics and VasoprotectantsCerebral ischemia damages the cerebral vessels as well as theneuronal parenchyma, disrupting vascular integrity and pre-disposing to intracerebral hemorrhage. Fibrinolytic agentsexacerbate this hemorrhagic risk. Administering agents thatare vasoprotective along with reperfusion interventions mayreduce hemorrhagic transformation rates, improve the bene-fit/risk ratio, and increase the permissible time window forreperfusion therapy. In preclinical studies, the rate of tPA-induced hemorrhage was markedly reduced by administrationof the matrix metalloproteinase (MMP) inhibitor batimastat(BB-94) or the spin trap agent �-phenil-N-t-butylnitrine.20

The spin trap agent NXY-059 (cerovive) is currently in aphase 2 clinical trial in which the coadministration of intra-venous tPA is allowed. The clinical development of MMPinhibitors, free radical scavengers, and other vasoprotectivecompounds for combination therapy with fibrinolytic andmechanical reperfusion interventions may substantially ex-pand the time window in which reperfusion interventionsmay be undertaken safely.

SonothrombolysisExperimental and clinical studies have consistently demon-strated the capability of ultrasound (US) to enhance enzy-matic thrombolysis. US application increases the transport oftPA into the thrombus, promotes the opening and cleaving ofthe fibrin polymers, and improves the binding affinity of tPAto fibrin. In an observational pilot trial of combined therapywith 2-MHz continuous US monitoring and intravenous tPAin 55 patients with a documented MCA occlusion treated �3hours of stroke onset, complete recanalization at 2 hours oftPA bolus was achieved in 36% of patients. In a small studyusing transcranial color-coded sonography (TCCS), 32 pa-tients were randomly allocated to be treated with combinedTCCS and intravenous tPA or tPA alone �6 hours ofsymptom onset. Combined treatment was associated withhigher rates of recanalization but also with a higher rate ofICH.21 CLOTBUST, a phase 2 multicenter randomized trial,recently demonstrated that 2-hour continuous monitoringwith 2-MHz TCD, a commercially available device widelyused for diagnosis, in combination with standard tPA is safe

and may improve outcome.22 Among 126 patients random-ized to tPA plus 2-hour TCD monitoring (target group) ortPA alone (control group), symptomatic ICH occurred in4.8% of target and 4.8% of control patients. Completerecanalization or dramatic clinical recovery at 2 hours aftertPA bolus were observed in 49% of target and 29% of controlpatients (P�0.02). Moreover, trends toward better clinicaloutcomes at 24 hours and long term were noted insonothrombolysis patients. A phase 3 of the CLOTBUST trialis planned to begin in 2006. Enhancement of enzymaticthrombolysis by US may allow testing regimens with low-dose tPA to reduce the risk of ICH. The capability ofmicrobubbles to further accelerate US-enhanced lysis instroke patients is currently under investigation.

IA ApproachesEndovascular methods to achieve recanalization in acuteischemic stroke comprise a wide range of pharmacologicaland mechanical techniques. IA techniques expand the timewindow for reperfusion therapy by more frequently and morerapidly removing the offending thrombus than intravenousapproaches and by reducing or eliminating exposure tofibrinolytic agents and their attendant bleeding risks.

IA FibrinolysisIn local IA fibrinolysis, fibrinolytic agents are infused distalto, proximal to, or directly within thrombotic occlusionsusing a microcatheter delivery system. Compared with stan-dard intravenous administration, the IA route offers severaltheoretical advantages, including: higher concentrations offibrinolytic agent at the clot site; reduced systemic exposureto thrombolytics; an opportunity to carry out gentle mechan-ical disruption of the clot with the delivery catheter and wire;precise imaging of case-specific vascular anatomy, pathol-ogy, and collateral patterns; and exact knowledge of thetiming and degree of recanalization achieved. In open clinicalseries, IA cerebral thrombolysis has yielded higher earlyrecanalization rates than intravenous therapy (50% to 80% forIA and 30% to 50% for intravenous).23

IA fibrinolysis also has a number of potential disadvan-tages, including: manipulation of a catheter within cerebralvessels, potentially increasing vulnerability to hemorrhage;the requirement for heparin administration intraprocedurallyto deter catheter-induced thrombosis (potentially increasinghemorrhage risk); delay in initiation of fibrinolysis while thediagnostic angiogram is performed and the delivery micro-catheter positioned (start of IA lytic infusion typically occurs50 to 90 minutes later than start of intravenous lytic infusion);the procedure is labor- and capital-intensive; and the inter-vention can only be performed at tertiary and secondaryhospitals capable of acute endovascular therapy.

The only large-scale, multicenter, randomized clinical trialof IA fibrinolytic therapy demonstrated substantial benefit oftherapy initiated up to 6 hours after onset of an M1 or M2MCA occlusion. In the Prolyse in Acute Cerebral Thrombo-embolism II (PROACT II) trial, the prespecified primaryoutcome, a good-to-excellent score on the modified RankinScale (mRS) of handicap (mRS �2), was achieved by 40% ofpro-urokinase (pro-UK) patients versus 25% of control pa-




tients (P�0.043).24 Partial or full recanalization(thrombolysis in myocardial infarction [TIMI] 2 or 3) rates 2hours after initiation of infusion were increased markedly inthe pro-UK group (66% versus 18%). However, full recana-lization (TIMI 3) was infrequent even in the pro-UK group(19% versus 2% in the control group). The recanalizationrates in PROACT II reflect the effects of pharmacologicallysis only. Passage of a microwire to disrupt the clot andaugment enzymatic lysis, although a common concomitanttherapy in endovascular practice, was not permitted by thestudy protocol. Intracerebral hemorrhage rates at 36 hourswere increased for the pro-UK group for all hemorrhages(46% versus 16%) and for symptomatic hemorrhages (10%versus 2%); however, no difference in overall mortality wasobserved.

Pro-UK is not available in regular practice because theresults of the single PROACT II trial were insufficient toobtain FDA approval. However, multiple large case-seriescohorts suggest similar efficacy and safety profiles for other,widely available fibrinolytic agents administered via the IAroute, including urokinase and tPA. Based on these findings,American Stroke Association guidelines recognize IA fibri-nolysis as a treatment option in select patients with largevessel occlusions, supported by evidence of intermediateweight.

Combined Intravenous/IAPharmacological Strategies

A treatment strategy of combined intravenous/IA lytic ther-apy may combine the advantages of speed (intravenous) anddefinitive endovascular attack (IA). Sequential intravenousand IA fibrinolytic therapy with tPA proved somewhatdisappointing in the NIH Interventional Management ofStroke (IMS) trial 1.25 Eighty patients were treated withreduced-dose intravenous tPA (0.6 mg/kg over 30 minutes)initiated within 3 hours of onset, followed by IA tPA,beginning within 5 hours of onset, if residual clot wasvisualized. Compared with historical controls treated withconventional intravenous tPA, the combined intravenous/IA-treated tPA patients showed only a modest trend to improvedclinical outcomes (odds ratio for global test, 1.35; CI, 0.78 to2.37). However, alternative, nonfibrinolytic intravenousagents may be more advantageous, serving to initiate treat-ment and also to provide a pharmacological complement thatmay enhance the effectiveness of IA fibrinolysis. Preliminarystudies are investigating combined intravenous G2P3 agentsand IA fibrinolytics up to 6 hours after symptom onset.26,27

Endovascular Mechanical TherapiesEndovascular mechanical therapies offer several distinct ad-vantages over endovascular delivery of pharmacological fi-brinolytics. Mechanical therapies typically: work more rap-idly, achieving recanalization within a few minutes, ratherthan the up to 120 minutes required with IA fibrinolyticadministration; are associated with lower intracerebral andsystemic hemorrhage risk because of the avoidance of phar-macological lysis; are more effective in disposing of largeclot burdens in proximal vessels, such as carotid T occlu-sions, where the sheer volume of clot to be digested retards

pharmacological lysis; and may in general be more effica-cious at achieving full recanalization.28

IA mechanical interventions may be classified into thecategories of endovascular thrombectomy, mechanical dis-ruption, and augmented fibrinolysis devices.29

Endovascular ThrombectomyEndovascular thrombectomy devices extract occludingthrombi from the target vessel through a catheter. Subcate-gories include: (1) clot retrieval devices that physically graspcerebral thrombi and pull them out of the cerebral circulation,and (2) suction thrombectomy devices that aspirate occlusivematerial from the vessel.

Clot retrieval devices were first developed to capture errantcoils and other foreign bodies that had embolized within thecerebral circulation during endovascular procedures. A natu-ral next step was to apply these devices to capture and removenaturally arising thromboemboli. These devices ensnare athrombus and then withdraw it out of the body, via the guidecatheter, or release it into a safer, extracerebral vascularterritory. At least 3 retriever device types have been appliedto cerebral thrombi in acute ischemic stroke patients, includ-ing the Microsnare (a 90° angled wire loop; Microvena),30 theNeuronet (self-expanding nitinol basket; Guidant),31 and theMerci Retriever X5/X6/LX (self-expanding nitinol helix;Concentric Medical).32 Additional devices currently FDAapproved for foreign body capture that could be appliedoff-label to cerebral thrombi include the In-Time Retriever (4to 6 concentric wire loops; Target) and the EnSnare (3 wireloops in tulip shape; Medical Device Technologies).

The Merci Retriever X5 and X6 devices have advancedfarthest in clinical trial development and regulatory approval.In the Merci Retriever procedure: (1) 2 to 3 loops of thenitinol helix are deployed beyond the thrombus; (2) thedevice is retracted into the thrombus and the remaining loopsdeployed within the clot; (3) the helix is twisted 3 to 5 timesto more fully capture the thrombus; (4) a balloon positionedproximally in the internal carotid artery is briefly inflated,blocking anterograde flow for a few seconds; and (5) whilethe balloon is up, the Merci Retriever and the ensnared clotare withdrawn, first into the positioning catheter and then outof the patient’s body. The Merci Retriever X5 and X6 deviceswere tested in the multicenter Mechanical Embolus Removalin Cerebral Ischemia (MERCI) trial, a 25-site, noncontrolled,technical efficacy trial. Patients with internal carotid arteryocclusion, M1 or M2 MCA occlusion, and vertebral andbasilar artery occlusions were treated within 8 hours ofonset.32 Among 121 patients enrolled, 114 underwent �1 (of6 permitted) passes with a clot retriever. Partial or completerevascularization was achieved by the device alone in 54%.Successful recanalization was associated with markedly im-proved clinical outcomes (90-day mRS, 0 to 2 in 53% ofrecanalizers versus 6% of nonrecanalizers; P�0.0001).Symptomatic hemorrhage occurred in 5% of patients treatedwith the device alone and 24% treated with the device plus anadditional rescue reperfusion intervention because of incom-plete recanalization response to the device (most commonlyIA fibrinolysis).




The encouraging results of the MERCI trial led the FDA inAugust 2004 to clear the Merci Retriever as the first devicereperfusion therapy labeled specifically for use in acuteischemic stroke. The FDA labeling reads, “The Merci Re-triever is intended to restore blood flow in the neurovascula-ture by removing thrombus in patients experiencing ischemicstroke. Patients who are ineligible for treatment with intrave-nous tPA (intravenous tPA) or who fail intravenous tPAtherapy are candidates for treatment.” It is important toemphasize that the device is labeled for a technical outcome(removing thrombi to restore blood flow), not a clinicaloutcome (eg, treatment of acute ischemic stroke). Only arandomized, controlled, clinical trial of the MERCI device(such as the recently launched NIH-funded MR Recanaliza-tion of Stroke Clots Using Embolectomy [MR RESCUE]trial) or another thrombus capture device can demonstratedefinitively that clot retriever therapy improves patient out-come. Vessel recanalization in acute ischemic stroke is apowerful determinant of clinical outcome and a promisingcandidate surrogate marker of treatment activity. In a recentmeta-analysis of 62 studies enrolling 2284 stroke patients,recanalization increased the odds ratio of good outcome5.4-fold.33 However, recanalization is not yet a fully validatedsurrogate that can replace clinical end points.

The next several years will undoubtedly witness rapidtechnologic advance in clot retrieval devices as embolectomyinstruments proliferate that improve on or complement theMERCI Retriever X5/X6. Most likely, as with the MERCIRetriever, FDA will permit new clot retrieval devices tofollow a rapid 510K pathway to approval, requiring onlydemonstration of technical efficacy in clot removal in uncon-trolled trials, not clinical efficacy in improving patient out-come in controlled, randomized trials. One promising second-generation device, already being tested in humans in theMulti-MERCI clinical trial, is the Merci Retriever LX (Con-centric Medical). The Merci Retriever LX has concentrichelical loops with polymer filaments attached, increasing clottraction, and achieved higher recanalization rates than theX5/X6 Retrievers in preclinical studies. If technologicaladvances in clot retrievers proceed at a pace typical of othermedical devices after first in class approval, with new devicedesigns appearing every 18 months on average, a markedexpansion in the endovascular armamentarium for acuteischemic stroke will take place over the 5 years.

Suction thrombectomy devices use vacuum aspiration toremove occlusive clot in acute ischemic stroke. Comparedwith mechanical disruption devices, suction thrombectomyhas reduced risk of causing uncontrolled thrombus fragmen-tation and distal embolization. Simple syringe suction appliedto an endovascular catheter was successful in treating large,internal carotid artery thrombi in small case series.29 Moresophisticated, vortex aspiration devices have been developedfor the extracerebral circulation, using high-pressure streamsto generate Venturi forces that physically fragment, draw in,and aspirate thrombi, including the AngioJet (Possis Medi-cal), the Oasis (Boston Scientific), the Amplatz Thrombec-tomy Device (Microvena Corp.), and the Hydrolyzer (Cor-dis). The initial generation Angiojet successfully treatedinternal carotid and vertebrobasilar thromboses in case re-

ports, although lack of flexibility made navigation in theintracranial circulation difficult. The NeuroJet (Possis Medi-cal), a smaller, single-channel device, was developed specif-ically for the intracranial circulation, sized to enter the MCAtrunk. However, in the initial feasibility and safety study inacute arterial ischemic stroke, vessel dissection was noted,and the trial was interrupted after the first 5 patients.Although modifications to device and protocol were under-taken for a successor safety trial, further development of thisdevice for ischemic stroke has apparently now been halted.

Mechanical Disruption of Occlusive MaterialA wide range of endovascular devices are designed tomechanically fragment or completely obliterate thrombi,atherosclerotic plaque, and other vascular occlusions. Re-peated passage of a micro-guidewire through a thrombus is asimple form of mechanical disruption frequently undertakenduring IA fibrinolytic procedures. Laser-tipped endovascularcatheters rapidly disrupt clots through conversion of photoenergy into acoustic energy, resulting in clot emulsification.At least 2 systems have entered human clinical trials for acuteischemic stroke: the EPAR (Endovasix) and LaTIS (LaTISInc) systems. The more extensively studied EPAR systemwas applied to 34 patients in a multicenter safety andfeasibility trial. The EPAR system alone, before patientexposure to any adjunctive lytics and stent therapies,achieved recanalization in 35% (8 of 23) of patients receivingany firing of the laser and 57% (8 of 14) of patients receivingcomplete lasing per protocol34(Angsar Berlis, personal com-munication, 2004). No adverse effects directly attributable tolasing were noted, but 3 patients had symptomatic ICH.These results suggest that endovascular photoacoustic clotdisruption holds promise as a mechanical recanalizationstrategy in acute stroke.

Primary intracranial angioplasty is a promising endovas-cular reperfusion strategy in select clinical circumstances. Inacute MI, primary angioplasty and stenting are superior tofibrinolytic therapy, yielding higher recanalization rates andbetter long-term outcomes. Several case series have reportedsuccess with acute percutaneous balloon angioplasty forischemic stroke.35 Angioplasty appears particularly useful inpatients with intracranial atherosclerotic lesions and super-vening in situ thrombi. In these lesions, as in the coronarybed, angioplasty in part achieves recanalization throughcontrolled cracking and dissection of underlying atheroscle-rotic lesions on which supervening thrombus has developed.However, many cerebral occlusions are attributable tothrombi of proximal origin that embolize to lodge in recipientcerebral vessels without extensive underlying calcified ath-erosclerosis. These spongy cerebral clots often bounce backinto an occlusive position after balloon angioplasty. As aresult, primary cerebral angioplasty has tended to be lesssuccessful when applied in white populations (among whomthromboembolism to intracranial vessels is a frequent strokemechanism) than in Asian populations (among whom in situintracranial atherothrombosis is a frequent stroke mecha-nism).28,29 It may be speculated that primary stenting willbetter maintain patency than angioplasty without stentingwhen the target cerebrovascular lesion is an embolized




thrombus. If so, continued advances in the development ofintracranial stenting technology may expand the applicabilityof acute cerebral angioplasty to a broader range of patients.

Augmented FibrinolysisSeveral mechanical techniques may enhance pharmacologicalfibrinolysis. Passage of a micro-wire through an occlusionduring IA fibrinolytic procedures is a form of augmentedfibrinolysis, not only directly disrupting the clot but alsoincreasing penetration of fibrinolytic agent throughout thetarget thrombus.36 Endovascular US techniques to enhanceenzymatic, intra-arterially delivered fibrinolytic agents arebeing developed in a manner complementary to external UStechniques to enhance intravenously administered fibrinolyt-ics. The EKOS MicroLysUS infusion catheter (EKOS Corp)system for augmented thrombolysis was tested in a small,multicenter safety and feasibility trial within 6 hours afteronset of anterior and 13 hours of posterior circulationischemia. Partial or complete recanalization was achievedwithin 1 hour of therapy start in 8 of 14 (57%), andsymptomatic hemorrhagic transformation occurred in 2 of 14(14%).37 The Interventional Management of Stroke Trialistsare currently investigating a strategy of upfront intravenoustPA followed by IA tPA administered via the EKOS catheterfor lytic augmentation.

Combined Pharmacological—EndovascularMechanical Strategies

Early experience with this wide range of emerging endovas-cular interventions suggests that combined pharmacologicaland endovascular mechanical therapies will often be requiredto achieve optimum reperfusion.28 Mechanical devices arecurrently too bulky to pass into distal vessels and oftenfragment proximal clots, causing pieces to embolize distally.Cleanup IA fibrinolysis directed at distal residua will often bea consideration in patients treated with mechanical devices ifit can be pursued with low additional risk. Complementarytreatment approaches may be needed to address occlusivelesions of mixed composition. Initial application of fibrino-

lytics may lyse a small supervening thrombus superimposedon a near-occlusive atherosclerotic lesion, but follow-upangioplasty will be needed to maximize patency and avertearly reocclusion. “Rescue” therapy with �1 modalities afterinitial therapy has failed will often be desirable. An occlusioninitially thought to be an embolic thrombus but unresponsiveto thrombus treatments (eg, clot retrievers, aspiration, andlaser) should suggest underlying atherosclerosis and the needfor “rescue” angioplasty. Conversely, a vessel repeatedlyreoccluding after angioplasty may suggest spongy thrombusin a near-normal underlying vessel and the need for “rescue”fibrinolytics, clot retrieval, or other appropriate intervention.Tandem lesions may require tandem treatments (eg, primarystenting of an extracranial internal carotid stenosis or occlu-sion to permit access of a thrombus capture device or IAfibrinolytic delivery catheter to an artery-to-artery emboluslodged in the MCA). Tailored approaches chosen from arange of mechanical and pharmacological options likely willbe required to achieve optimum recanalization rates, alwaysbearing in mind that the cerebral vasculature is fragile and theamplitude of mechanical energies and intensity of pharmaco-logical therapies delivered to break up thrombi will be limitedby the need to protect vessel wall integrity.

Using Multimodal Imaging to Extend theReperfusion Treatment Time Window

The duration of the ischemic penumbra varies widely frompatient to patient. Late reperfusion therapy is likely to benefitindividuals in whom substantial salvageable tissue still per-sists beyond the first few hours after symptom onset but notbenefit, and possibly harm, patients who have completed theirinfarction. Multimodal MR and CT imaging protocols rendera multidimensional depiction of the cerebral ischemic pro-cess: distinguishing infarct from hemorrhage; delineatingirreversibly injured infarct core, still salvageable penumbra,and unthreatened regions of benign oligemia; identifyingpropensities to hemorrhagic transformation; and ascertaininglarge vessel stenoses and occlusions, all in just 5 to 20minutes of table time (Figure). 38,39 These protocols poten-

The multimodal CT and MR strategies inacute stroke neuroimaging. MRA indi-cates MR angiography (copyright UCLAStroke Program).




tially expand the time window for reperfusion therapy,enabling patient selection based on an individualized tissueclock rather than a fixed time clock38,40 (Table).

It is important to emphasize that in any conceivablecircumstance, it will always be crucial to institute therapy assoon as possible. For early and late treatment windowpatients, there will always be a tradeoff between moreinformation and more dead brain. Moreover, within the first1 to 2 hours of onset, virtually all patients harbor substantialpenumbra, whereas among late-presenting patients, a steadilydecreasing proportion evidences persisting penumbra. Open-ing up late treatment time windows for select patients throughmultimodal imaging is a complementary strategy to achievingearly treatment times for all early presenting patients.

Multiple current clinical trials are refining or incorporatingMR strategies to expand patient eligibility for reperfusiontherapy, including studies: (1) identifying candidate MRmeasures for patient selection (no internal control group, allpatients treated irrespective of entry MR pattern); (2) validat-ing prespecified MR measures for patient selection (random-ized, controlled design, all patients enrolled irrespective ofentry MR pattern); and (3) already using MR measures forpatient selection, although these measures have not yet beenfully validated (randomized, controlled design; only patientsexhibiting MR pattern felt predictive of good treatmentresponse enrolled).

Identifying Candidate MRI Algorithms toSelect Patients for Late Reperfusion Therapy

DEFUSE (Diffusion-weighted imaging Evaluation For Un-derstanding Stroke Evolution) is an NIH-funded, multicenter,uncontrolled pilot study investigating whether specific DWI–

perfusion-weighted imaging (PWI) profiles predict a favor-able clinical response to intravenous tPA administered 3 to 6hours after stroke onset. Patients with a clinical diagnosis ofischemic stroke causing measurable moderate-to-severe neu-rological deficit (NIHSS score �5) are included, regardlessof MR PWI-DWI pattern.

Validating MRI AlgorithmsThe randomized EPITHET (Echoplanar Imaging Thrombol-ysis Evaluation Trial) is an ongoing double-blind, random-ized, multicenter study with a planned enrollment of 100patients. EPITHET aims to determine whether the extent ofthe ischemic penumbra apparent on perfusion–diffusion MRIidentifies patients who will benefit from intravenous tPA 3 to6 hours after stroke.

MR RESCUE is an ongoing, NIH-funded, multicenter,randomized clinical trial with a planned sample size of 120patients. The trial tests the hypothesis that the presence ofsubstantial ischemic penumbral tissue visualized on diffu-sion–perfusion MRI identifies patients most likely to respondto Merci Retriever mechanical embolectomy up to 8 hoursfrom symptom onset.

Using Multimodal MRI Algorithms to SelectPatients for Late Reperfusion Therapy

The DIAS (Europe/Australia) and DEDAS (United States/Canada) Trials are randomized, multicenter, placebo-controlled, dose-escalating trials assessing the safety andthrombolytic efficacy of intravenous desmoteplase in MRI-selected patients with acute ischemic stroke between 3 to 9hours after stroke onset. Patients are eligible if MRI within 8hours shows ongoing hypoperfusion (PWI abnormality �2cm in diameter in the hemispheric gray matter) and stillsalvageable penumbra (PWI/DWI mismatch �20%). Thepreliminary results of DIAS appear to validate the strategy oftreating late-presenting, imaging-selected patients with intra-venous fibrinolysis. A dose-response relationship was dem-onstrated between desmoteplase and reperfusion. At theapparent optimal dose of 125 �g/kg, reperfusion (PWIreduction �30% or TIMI change �2 post-thrombolysis)occurred in 71% of patients (versus 22% in placebo) andexcellent clinical outcome in 60% (versus 18% in placebo; S.Warach, International Stroke Conference San Diego, Calif,2004). The rate of symptomatic intracerebral hemorrhage waslow (3.3%) in this dose range (90 and 125 �g/kg). Safety andefficacy appeared independent of the time window. A pivotaltrial with 125 �g/kg IV desmoteplase in imaging-selectedpatients in the 3- to 9-hour window is planned.

The ROSIE trial is a phase 2 safety and dose-ranging studyof combined reteplase and abciximab initiated 3 to 24 hoursafter stroke onset. Leading entry criteria are NIHSS score�16, a perfusion MR deficit, and absence of a DWI abnor-mality more than one third of the MCA territory. Patientsreceive abciximab alone or abciximab plus 1 of 4 tiers ofreteplase.

Clinical-Diffusion MismatchClinical-DWI mismatch (CDM) represents a new diagnosticapproach that can extend MR identification of persisting

Provisional Multimodal MR/CT Algorithm for Selecting Patientsfor Late (>3 hours) Reperfusion Therapies

Very favorable candidate for intravenous or IA therapy

Distal M1 or proximal M2 MCA occlusion, beyond takeoff oflenticulostriate branches�substantial visualized penumbra*

Favorable candidate for intravenous or IA therapy

Proximal M1 MCA occlusion�substantial visualized penumbra

Somewhat favorable candidates for therapy

Distal internal carotid artery occlusion�substantial visualized penumbra(IA�IV)

Distal MCA/ACA branch occlusion�severe functional deficit�substantialvisualized penumbra (IV)

Penetrator occlusion�severe functional deficit�substantial visualizedpenumbra (IV)

Avoid fibrinolytic therapy; IA mechanical therapy may be considered butbenefit/risk ratio reduced

Large infarct core, more than one third of MCA territory (DWI change onMR, collapsed CBV on CT), but substantial visualized penumbra

Avoid therapy

No substantial visualized penumbra

*Methods for identifying when substantial penumbral tissue is present arerapidly evolving. Simple current approaches are MR PWI-DWI mismatch or CTCTP-collapsed CBV mismatch �20% in diameter on slice with largest lesion.Multivariate predictive equations offer greater accuracy but are less widelyaccessible.




penumbra to hospitals that have diffusion but not perfusionMR imaging. Because most ischemic brain tissue is clinicallysymptomatic, stroke severity as measured by the NIHSSscore correlates with the extent of hypoperfused tissue (PWIabnormality). CDM is defined as an NIHSS score �8 andischemic volume on DWI �25 mL. The NIHSS score of �8has been associated with cortical perfusion deficits and highrate of neurological deterioration. In 166 patients imagedwithin 12 hours of hemispheric ischemic stroke onset, CDMwas found in 87 (52%).41 The frequency of CDM decreasedas time from onset increased, being 74% at �3 hours, 48%from 3 to 6 hours, and 46% from 6 to 12 hours. The presenceof CDM was associated with a higher rate of early neurolog-ical deterioration, greater DWI lesion growth at 72 hours, andlarger infarct volume on T2-weighted MRI at day 30. How-ever, because the NIHSS score underestimates infarct volumein the right hemisphere, the CDM definition may be lesssensitive to estimate penumbra tissue in right-sided lesions.Prospective validation of the CDM definition is needed todetermine its reliability to rapidly identify patients with tissueat risk as candidates for reperfusion strategies.

Selection Based on Multimodal CT CriteriaNovel CT techniques also show great promise as tools tostratify later-presenting patients into groups likely and notlikely to benefit from reperfusion. CT angiography (CTA) isa well-established technique to identify acute vascular occlu-sions. Several CT techniques are now available to imagetissue perfusion, including perfusion CT (PCT), CTA sourceimage analysis, and xenon-CT. Of these, particularly greatpotential is shown by dynamic PCT, in which images areacquired during first pass of a standard iodinated contrastbolus.

PCT cerebral blood flow (CBF) maps distinguish penum-bra from benign oligemia by differentiating regions withmoderate versus mild reductions in blood flow. PCT cerebralblood volume (CBV) maps distinguish infarct core frompenumbra by delineating regions with advanced tissue injury,loss of autoregulation, and vascular collapse, evident asmarkedly decreased CBV. Accordingly, PCT offers an ana-logue to the MR mismatch model of core and penumbra, withregions of collapsed CBV representing core and regions withreduced CBF but preserved CBV (CBF-CBV mismatch)representing penumbra.39 Penumbral regions identified by CTCBF-CBV mismatch correlate well with penumbral regionsidentified by MR DWI-PWI mismatch when both studies areobtained in the same acute stroke patients.42

Multimodal CT techniques are just beginning to be appliedin formal clinical trials and advanced clinical practice toextend treatment time windows by selecting patients for latereperfusion. Compared with MR, multimodal CT has thedisadvantages of less coverage of brain tissue (interrogatingonly 2 to 4 slices at present), use of iodinated contrast withallergic and renal toxicity, and poor visualization of infraten-torial tissues attributable to bone artifact. However, CTA/PCT has the advantages of more rapid patient positioning andscan acquisition, wider availability of hardware and staffing,and an easier upgrade path to implementation for many

hospitals, building on the existing infrastructure of emer-gency CT scanners.

ConclusionsThe ideal toward which reperfusion therapies for stroke striveis to achieve complete and lasting vessel patency, as rapidlyas possible, in all patients harboring salvageable tissue, withno risk of hemorrhagic transformation. This review hassurveyed �45 distinct, promising approaches to extendingreperfusion in cerebral ischemia currently under investigationin human clinical studies. Areas of advance include novelpharmacological classes, novel agents within classes, novelmechanical devices, novel imaging selection paradigms, andnovel combinations of these techniques. Further developmentof these therapies will require innovations in clinical trialdesign to meet the emerging challenges of testing combina-tion therapies, device therapies, and therapies tailored toindividual pathophysiology while retaining definitive phase 3trials as the gold standard for assessing treatment benefit.43 Ifstudied correctly, therapeutic reperfusion promises to emergeas a treatment strategy of remarkable power and scope forrescuing patients experiencing acute brain ischemia, applica-ble within and beyond the 3-hour time window.

References1. Hacke W, Donnan G, Fieschi C, Kaste M, von Kummer R, Broderick JP,

Brott T, Frankel M, Grotta JC, Haley EC Jr, Kwiatkowski T, Levine SR,Lewandowski C, Lu M, Lyden P, Marler JR, Patel S, Tilley BC, AlbersG, Bluhmki E, Wilhelm M, Hamilton S; ATLANTIS Trials Investigators;ECASS Trials Investigators; NINDS rt-PA Study Group Investigators.Association of outcome with early stroke treatment: pooled analysis ofATLANTIS, ECASS, and NINDS rt-PA stroke trials. Lancet. 2004;363:768–774.

2. Saver JL. Number needed to treat estimates incorporating effects over theentire range of clinical outcomes. Arch Neurol. 2004;61.

3. Davydov L, Cheng JW. Tenecteplase: a review. Clin Ther. 2001;23:982–997.

4. Lyden P; TNK for Stroke Investigators. Pilot study of tenecteplase (TNK)in acute ischemic stroke: preliminary report. Stroke. 2003;34:246A.

5. Liberatore GT, Samson A, Bladin C, Schleuning WD, Medcalf RL.Vampire bat salivary plasminogen activator (desmoteplase): a uniquefibrinolytic enzyme that does not promote neurodegeneration. Stroke.2003;34:537–543.

6. Noble S, McTavish D. Reteplase. A review of its pharmacological prop-erties and clinical efficacy in the management of acute myocardialinfarction. Drugs. 1996;52:589–605.

7. Qureshi AI, Ali Z, Suri MF, Kim SH, Shatla AA, Ringer AJ, Lopes DK,Guterman LR, Hopkins LN. Intra-arterial third-generation recombinanttissue plasminogen activator (reteplase) for acute ischemic stroke. Neu-rosurgery. 2001;49:41–48.

8. Marder VJ, Stewart D. Towards safer thrombolytic therapy. SeminHematol. 2002;39:206–216.

9. Lapchak PA, Araujo DM, Pakola S, Song D, Wei J, Zivin JA. Micro-plasmin: a novel thrombolytic that improves behavioral outcome afterembolic strokes in rabbits. Stroke. 2002;33:2279–2284.

10. Adams HP Jr, Leclerc JR, Bluhmki E, Clarke W, Hansen MD, Hacke W.Measuring outcomes as a function of baseline severity of ischemic stroke.Cerebrovasc Dis. 2004;18:124–129.

11. Junghans U, Seitz RJ, Aulich A, Freund HJ, Siebler M. Bleeding risk oftirofiban, a nonpeptide GPIIb/IIIa platelet receptor antagonist in pro-gressive stroke: an open pilot study. Cerebrovasc Dis. 2001;12:308–312.

12. Junghans U, Seitz RJ, Ritzl A, Wittsack HJ, Fink GR, Freund HJ, SieblerM. Ischemic brain tissue salvaged from infarction by the GP IIb/IIIaplatelet antagonist tirofiban. Neurology. 2002;58:474–476.

13. Straub S, Junghans U, Jovanovic V, Wittsack HJ, Seitz RJ, Siebler M.Systemic thrombolysis with recombinant tissue plasminogen activatorand tirofiban in acute middle cerebral artery occlusion. Stroke. 2004;35:705–709.




14. Seitz RJ, Meisel S, Moll M, Wittsack H-J, Junghans U, Siebler M. Theeffect of combined thrombolysis with rtPA and tirofiban on ischemicbrain lesions. Neurology. 2004;62:2110–2112.

15. Gahn G, Kunz A, Putz V, Becker U, Goldhagen T, Hahn G, et al.Recanalization of middle cerebral artery occlusion after either t-PA ort-PA combined with abciximab. Stroke. 2004;35:287 (abstract).

16. Ouriel K, Castaneda F, McNamara T, Swischuk J, Tepe G, Smith JJ,Clark J, Duda S. Reteplase monotherapy and reteplase/abciximab com-bination therapy in peripheral arterial occlusive disease: results from theRELAX trial. J Vasc Interv Radiol. 2004;15:229–238.

17. Ireland JK, Uchino K, Alexandrov AV, Ford SR, Shaw SG, MatherneDE, et al. TPA Argatroban Stroke Study (TARTS). Stroke. 2004;35:301(abstract).

18. De Georgia MA, Krieger DW, Abou-Chebl A, Devlin TG, Jauss M, DavisSM, Koroshetz WJ, Rordorf G, Warach S. Cooling for Acute IschemicBrain Damage (COOL AID): a feasibility trial of endovascular cooling.Neurology. 2004;63:312–317.

19. Saver JL, Kidwell C, Eckstein M, Starkman S. Prehospital neuropro-tective therapy for acute stroke: results of the Field Administration ofStroke Therapy-Magnesium (FAST-MAG) pilot trial. Stroke. 2004;35:e106–e108.

20. Lapchak PA, Araujo DM. Reducing bleeding complications afterthrombolytic therapy for stroke: clinical potential of metalloproteinaseinhibitors and spin trap agents. CNS Drugs. 2001;15:819–829.

21. Eggers J, Koch B, Meyer K, Konig I, Seidel G. Effect of ultrasound onthrombolysis of middle cerebral artery occlusion. Ann Neurol. 2003;53:797–800.

22. Alexandrov AV, Molina CA, Grotta JC, Garami Z, Ford SR,Alvarez-Sabin J, Montaner J, Saqqur M, Demchuk AM, Moye LA, HillMD, Wojner AW; CLOTBUST Investigators. Ultrasound-enhancedsystemic thrombolysis for acute ischemic stroke. N Engl J Med. 2004;351:2170–2178.

23. Alexandrov AV, Demchuk AM, Burgin WS, Robinson DJ, Grotta JC.Ultrasound-enhanced thrombolysis for acute ischemic stroke: phase I.Findings of the CLOTBUST trial. J Neuroimaging. 2004;14:113–117.

24. Furlan A, Higashida R, Wechsler L, Gent M, Rowley H, Kase C, PessinM, Ahuja A, Callahan F, Clark WM, Silver F, Rivera F. Intra-arterialprourokinase for acute ischemic stroke. The PROACT II study: a ran-domized controlled trial. Prolyse in Acute Cerebral Thromboembolism.J Am Med Assoc. 1999;282:2003–2011.

25. Combined intravenous and intra-arterial recanalization for acute ischemicstroke: the Interventional Management of Stroke Study. Stroke. 2004;35:904–911.

26. Eckert B, Koch C, Thomalla G, Roether J, Zeumer H. Acute basilar arteryocclusion treated with combined intravenous abciximab and intra-arterialtissue plasminogen activator: report of 3 cases. Stroke. 2002;33:1424–1427.

27. Qureshi AI. Safety and efficacy of intra-arterial reteplase and intravenousabciximab in patients with acute ischemic stroke. Ongoing Clinical TrialsSession, 27th International Stroke Conference; San Antonio, Texas, 2002.

28. Leary MC, Saver JL, Gobin YP, Jahan R, Duckwiler GR, Vinuela F,Kidwell CS, Frazee J, Starkman S. Beyond tissue plasminogen activator:mechanical intervention in acute stroke. Ann Emerg Med. 2003;41:838–846.

29. Nesbit GM, Luh G, Tien R, Barnwell SL. New and future endovasculartreatment strategies for acute ischemic stroke. J Vasc Interv Radiol.2004;15:S103–S110.

30. Wikholm G. Transarterial embolectomy in acute stroke. AJNR Am JNeuroradiol. 2003;24:892–894.

31. Mayer TE, Hamann GF, Brueckmann HJ. Treatment of basilar arteryembolism with a mechanical extraction device: necessity of flow reversal.Stroke. 2002;33:2232–2235.

32. Starkman S; MERCI Investigators. Results of the combined MERCI I-II(Mechanical Embolus Removal in Cerebral Ischemia) Trials. Stroke.2004;35:A240.

33. Rha J-H, Saver JL. Recanalization as a surrogate outcome measure inthrombolytic clinical trials: a meta-analysis. Stroke. 2003;34:317(abstract).

34. Berlis A, Lutsep H, Barnwell S, Norbash A, Wechsler L, Jungreis CA,Woolfenden A, Redekop G, Hartmann M, Schumacher M. Mechanicalthrombolysis in acute ischemic stroke with endovascular photoacousticrecanalization. Stroke. 2004;35:1112–1116.

35. Ringer AJ, Qureshi AI, Fessler RD, Guterman LR, Hopkins LN. Angio-plasty of intracranial occlusion resistant to thrombolysis in acute ischemicstroke. Neurosurgery. 2001;48:1282–1288.

36. Qureshi AI, Siddiqui AM, Suri MF, Kim SH, Ali Z, Yahia AM, LopesDK, Boulos AS, Ringer AJ, Saad M, Guterman LR, Hopkins LN.Aggressive mechanical clot disruption and low-dose intra-arterial third-generation thrombolytic agent for ischemic stroke: a prospective study.Neurosurgery. 2002;51:1319–1327.

37. Mahon BR, Nesbit GM, Barnwell SL, Clark W, Marotta TR, Weill A,Teal PA, Qureshi AI. North American clinical experience with the EKOSMicroLysUS infusion catheter for the treatment of embolic stroke. AJNRAm J Neuroradiol. 2003;24:534–538.

38. Kidwell CS, Alger JR, Saver JL. Beyond mismatch: evolving paradigmsin imaging the ischemic penumbra with multimodal magnetic resonanceimaging. Stroke. 2003;34:2729–2735.

39. Wintermark M, Reichhart M, Thiran JP, Maeder P, Chalaron M,Schnyder P, Bogousslavsky J, Meuli R. Prognostic accuracy of cerebralblood flow measurement by perfusion computed tomography, at the timeof emergency room admission, in acute stroke patients. Ann Neurol.2002;51:417–432.

40. Schellinger PD, Fiebach JB, Hacke W. Imaging-based decision making inthrombolytic therapy for ischemic stroke: present status. Stroke. 2003;34:575–583.

41. Davalos A, Leira R, Pedraza S, Castellanos M, Silva Y, Montserrat P, etal. The clinical-DWI mismatch: a new diagnostic clue in the treatment ofacute ischemic stroke. Stroke. 2003;34:254A.

42. Schramm P, Schellinger PD, Klotz E, Kallenberg K, Fiebach JB, KulkensS, Heiland S, Knauth M, Sartor K. Comparison of perfusion computedtomography and computed tomography angiography source images withperfusion-weighted imaging and diffusion-weighted imaging in patientswith acute stroke of �6 hours’ duration. Stroke. 2004;35:1652–1658.

43. Fisher M; Stroke Therapy Academic Industry Roundtable. Recom-mendations for advancing development of acute stroke therapies:Stroke Therapy Academic Industry Roundtable 3. Stroke. 2003;34:1539 –1546.




Carlos A. Molina and Jeffrey L. SaverMechanical, and Imaging Strategies

Extending Reperfusion Therapy for Acute Ischemic Stroke: Emerging Pharmacological,

Print ISSN: 0039-2499. Online ISSN: 1524-4628 Copyright © 2005 American Heart Association, Inc. All rights reserved.

is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Stroke doi: 10.1161/01.STR.0000182100.65262.46

2005;36:2311-2320; originally published online September 22, 2005;Stroke.

http://stroke.ahajournals.org/content/36/10/2311World Wide Web at:

The online version of this article, along with updated information and services, is located on the

http://stroke.ahajournals.org//subscriptions/

is online at: Stroke Information about subscribing to Subscriptions:

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

document. Permissions and Rights Question and Answer process is available in the

Request Permissions in the middle column of the Web page under Services. Further information about thisOnce the online version of the published article for which permission is being requested is located, click

can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office.Strokein Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:


http://stroke.ahajournals.org/content/36/10/2311

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

http://stroke.ahajournals.org//subscriptions/


Extending Reperfusion Therapy for Acute Ischemic Stroke: Emerging Pharmacological, Mechanical, and...

Documents

Transcript of Extending Reperfusion Therapy for Acute Ischemic Stroke: Emerging Pharmacological, Mechanical, and...