NEUROLOGICAL PROGRESS

Advanced Imaging to Extend theTherapeutic Time Window of Acute

Ischemic Stroke

Marc Fisher, MD,1 and Gregory W. Albers, MD2

Reperfusion therapy for acute stroke has evolved from the initial use of intravenous tissue plasminogen activator(tPA) within 3 hours of symptom onset to more recent guideline-recommended use up to 4.5 hours. In addition,endovascular therapy is increasingly utilized for stroke treatment and is typically initiated up to 8 hours after onset.Recent studies demonstrate that imaging of the ischemic penumbra with diffusion/perfusion magnetic resonanceimaging (MRI) can identify subgroups of patients who are likely to improve following successful reperfusion (TargetMismatch profile) and others who are at increased risk for hemorrhage and poor clinical outcomes (Malignantprofile). New data indicate that stent retriever devices provide better recanalization efficacy and clinical outcomesthan the previously available mechanical thrombectomy devices. Going forward, we believe that the use ofpenumbral imaging with validated MRI techniques, as well as the currently less well-validated computed tomography(CT) perfusion approach, will maximize benefit and reduce the risk of adverse events and poor outcomes when usedboth early after stroke onset and at later time points. New trials that feature diffusion/perfusion MRI or CTperfusion-based patient selection for treatment with intravenous tPA and or endovascular therapies versusnonreperfused control groups are planned or in progress. We predict that these trials will confirm the hypothesisthat penumbral imaging can enhance patient selection and extend the therapeutic time window for acute ischemicstroke.

ANN NEUROL 2013;73:4–9

The development of new, effective acute ischemic

stroke therapies is at a crossroads. The recently

reported negative trials of granulocyte colony stimulation

factor (AXIS II) and citicoline cast further pessimism on

the future of neuroprotection.1,2 In contrast, thrombo-

lytic therapies (intravenous tissue plasminogen activator

and intra-arterial prourokinase) have been established to

be effective, although prourokinase did not receive regu-

latory approval based on the results of a single trial.3

A key determinant for patient selection for throm-

bolytic therapy has been time from stroke onset, and the

efficacy for intravenous (i.v.) tissue plasminogen activator

(tPA) is established only up to 4.5 hours after symptom

onset.4 Some evidence of very modestly improved func-

tional outcomes following tPA administration up to 6

hours after symptom onset was recently demonstrated in

the IST 3 trial, but this benefit was largely driven by

patients treated within 3 hours after stroke onset.5 The

benefit of i.v. tPA appears to decline gradually with lon-

ger durations between symptom onset and tPA therapy,6

and recently the FDA did not approve Genentech’s appli-

cation for an extension of regulatory approval of i.v. tPA

beyond 3 hours.

In addition to pharmacological thrombolysis, the

other available options for reperfusion of acute ischemic

stroke are devices, as either alternatives or adjuvants to

i.v. tPA. Two thrombectomy devices, the Merci Retriever

and the Penumbra System, were cleared by the US Food

and Drug Administration (FDA) for intracranial clot re-

moval several years ago, but efficacy of these devices has

not been established in a randomized clinical trial.7,8

Recently, a third device, the SOLITAIRE stent-retriever,

was cleared by the FDA after the SWIFT trial not only

demonstrated significantly greater recanalization with this

device compared with the Merci device, but also a higher

rate of favorable clinical outcomes at day 90.9 Similar

advantages over the Merci device were recently reported

in a randomized comparison trial with the TREVO

View this article online at wileyonlinelibrary.com. DOI: 10.1002/ana.23744

Received Jun 25, 2012, and in revised form Jul 28, 2012. Accepted for publication Aug 15, 2012.

Address correspondence to Dr Fisher, UMASS/Memorial Healthcare, 119 Belmont Street, Worcester, MA 01605. E-mail: fisherm@ummhc.org

From the 1Department of Neurology, University of Massachusetts School of Medicine, Worcester, MA and 2Department of Neurology, Stanford University

School of Medicine, Palo Alto, CA.

4 VC 2012 American Neurological Association

stent-retriever.10 Both of these trials enrolled patients up

to 8 hours from stroke onset. Therefore, endovascular

therapy with stent-retrievers has taken center stage as 1

of the most promising therapies for acute stroke.

Some of the enthusiasm for endovascular therapy

was recently dampened, however, by the news that the

IMS III trial was halted due to futility. IMS III assessed

the efficacy i.v. tPA alone (started within 3 hours of

stroke onset) versus the combination i.v. tPA plus endo-

vascular therapy (either mechanical thrombectomy devi-

ces and/or intra-arterial tPA). This result was surprising

to many, because in prior studies both intra-arterial

administration of thrombolytic agents and thrombectomy

devices have demonstrated greater recanalization rates

than i.v. tPA alone,11,12 and early recanalization is associ-

ated with improved clinical outcomes.13 Although the

full results of IMS III are not yet available, there are sev-

eral potential explications for the lack of clinical benefit.

One perspective is that reperfusion is only beneficial

under specific conditions; in particular, blood flow must

be restored to ischemic brain tissue prior to the develop-

ment of irreversible injury. Yet, the selection criteria for

the larger studies of reperfusion therapies, including IMS

III, have not included methods to identify potentially sal-

vageable tissue. Furthermore, there is evidence that reper-

fusion of irreversibly injured tissue may result in adverse

effects including brain edema and hemorrhage.5,14

A central consideration in the development and

optimization of novel acute stroke therapies is the con-

cept of the ischemic penumbra. Clearly, the target of

acute stroke therapies is salvage of some portion of the

ischemic penumbra, leading to a reduction is infarct size

and most importantly, improved functional outcome.15

The ischemic penumbra has multiple definitions, but the

1 most relevant for therapy development is ischemic tis-

sue that is potentially salvageable as distinguished from

the ischemic core that has already sustained irreversible

injury.16 Many factors affect the evolution of the ische-

mic penumbra into the ischemic core, and the rate of

progression of irreversible injury appears to be highly

variable between individuals (Fig 1). This variability is

likely mediated by the adequacy of collateral blood flow

as well as the metabolic milieu of individual stroke

patients.17

The individuality of penumbral evolution among

stroke patients implies that identifying the extent of the

ischemic core and penumbra could be useful in making

treatment decisions. Currently diffusion-weighted imag-

ing (DWI)/perfusion-weighted imaging (PWI) magnetic

resonance imaging (MRI) affords the best opportunity

for approximating the ischemic penumbra and core in

real time clinical practice.18 Despite a limited propensity

for permanent reversal following early reperfusion, the

DWI lesion provides a dependable estimation of the is-

chemic core.19,20 PWI identifies hypoperfused, ischemic

tissue and regions defined as abnormal on PWI that do

not demonstrate a DWI abnormality, often referred to as

the DWI/PWI mismatch, can approximate the ischemic

penumbra.21 The key issue related to PWI is to employ

an appropriately validated threshold parameter that

excludes ischemic tissue with modest blood flow reduc-

tion (ie, benign oligemia), because this tissue is unlikely

to become infarcted even if reperfusion does not occur.

Which PWI parameter is optimal, as well as what thresh-

old to use to define critical hypoperfusion, has been an

area of contentious debate.22 Recent studies support the

use of Tmax as a clinically useful PWI parameter, and

maps of tissue with a Tmax delay of >5 to 6 seconds, as

compared to normally perfused tissue, are reasonably pre-

dictive of ischemic tissue destined to become infarcted if

timely reperfusion does not occur.23 This Tmax threshold

correlates well with the penumbral range of cerebral

blood flow decline as determined by positron emission

tomography,24 and several ongoing studies are using Tmax

to identify the extent of the hypoperfused zone.

Using a difference between the volume of the base-

line PWI Tmax lesion and the DWI volume to identify

mismatch, the DEFUSE and EPITHET studies found

that most patients with a PWI/DWI mismatch appeared

to respond favorably if reperfusion occurred following i.v.

tPA treatment in the 3- to 6-hour time window. How-

ever, despite having a mismatch, patients with very large

baseline DWI lesions (large early core volumes) or

patients with very large volumes of severe Tmax delay had

highly unfavorable outcomes following reperfusion.

Patients with this MRI pattern, referred to as the

FIGURE 1: Initial diffusion-weighted imaging (DWI) volumesin patients with complete occlusion of the middle cerebralartery (n 5 16) or distal internal carotid artery (n 5 6) onmagnetic resonance angiography in the DEFUSE study.Note variability of DWI lesion volume and lack of relation-ship between lesion size and time from symptom onset tomagnetic resonance imaging (MRI). [Color figure can beviewed in the online issue, which is available atwww.annalsofneurology.org.]

Fisher and Albers: Penumbral Imaging of AIS

January 2013 5

Malignant profile, had a significantly higher rate of both

parenchymal hemorrhage and severe disability/death if

reperfusion occurred (Fig 2).14

Mismatch patients who do not have the Malignant

profile have been designated as having a Target Mis-

match, and these patents appear to respond very favor-

ably to reperfusion following i.v. tPA therapy. In a

pooled analysis of DEFUSE and EPITHET, Target Mis-

match profile patients who experienced reperfusion had a

5-fold increase in favorable clinical response at 90 days

and significantly less infarct growth when compared to

those who did not reperfuse.25 No association between

favorable outcomes or reduction in infarct growth was

apparent for patients who did not have a mismatch.

Identification of MRI profiles in the acute setting has

been challenging, because it previously required cumber-

some postprocessing. Currently, automated or semiauto-

mated programs are available that allow volumetric DWI

and PWI profiles to be generated rapidly.26

Recently, the results of the DEFUSE 2 study were

published. This multicenter study, which utilized an

automated mismatch analysis program, established MRI

profiles prospectively in a consecutive cohort of patients

who underwent endovascular therapy. The study con-

firmed the concepts demonstrated in DEFUSE and EPI-

THET; Target Mismatch patients who achieve early

reperfusion therapy have less infarct growth and more

favorable clinical outcomes.27 No association between

reperfusion and favorable outcomes or infarct growth was

present in patients without Target Mismatch. Further-

more, in DEFUSE 2 patients with Target Mismatch who

were treated relatively late (6 to 12 hours after symptom

onset), the positive association between reperfusion,

favorable clinical response, and attenuation of infarct

growth did not diminish (Fig 3).28 This finding contrasts

sharply with prior studies that did not use penumbral

imaging to select patients and suggests that imaging find-

ings may be of equal or greater importance than time

from symptom onset for identification of optimal

patients who might benefit from reperfusion therapy.

How could Target Mismatch patients who are

treated at later time points have outcomes that are as

favorable as those of earlier treated patients? One possi-

bility is that at later time points, the Target Mismatch

profile is identifying patients in whom the ischemic core

is evolving at a relatively slow rate. Despite a large vessel

occlusion, the finding that the DWI lesion is still consid-

erably smaller than the PWI lesion at a late time point

may reflect good collateral circulation. These collaterals

may allow prolonged, but not permanent survival of the

hypoperfused mismatch region. Evidence that the mis-

match region is still at considerable risk for infarct

expansion, even at later time points, was provided by the

DEFUSE 2 finding that Target Mismatch patients

treated between 6 and 12 hours from symptom onset

consistently demonstrated substantial infarct growth if

reperfusion was not achieved.28 Patients with slowly

evolving infarct cores are likely ideal candidates for later

time window reperfusion therapy, particularly endovascu-

lar therapies. One of the drawbacks of the endovascular

approach is that the time between hospital arrival and

achievement of endovascular reperfusion is typically at

least 90 to 120 minutes. For patients with rapidly grow-

ing infarct cores (such as patients with the Malignant

FIGURE 2: Example of the Malignant Profile. (A) This mag-netic resonance image obtained 3 hours after witnessedsymptom onset demonstrates very large diffusion-weightedimaging (DWI) lesion shown in pink very severe perfusionlesion. (B) In addition a very severe perfusion lesion is pres-ent (red) and corresponds to a Tmax delay >10 seconds.

ANNALS of Neurology

6 Volume 73, No. 1

profile), substantial growth of the infarct core has been

reported despite endovascular reperfusion.29

Challenges facing the acute stroke therapy field

going forward are how to incorporate the new information

about the predictive value of imaging into randomized

clinical trials to demonstrate significant improvements

in outcome over an extended time window and how to

optimize the percentage of patients who obtain these bene-

fits. Currently, there are several ongoing or planned clini-

cal trials that are evaluating reperfusion therapies (both i.v.

and endovascular) at late time windows or in patients who

are discovered to have a stroke after awakening. For exam-

ple, the novel thrombolytic desmoteplase is being com-

pared to placebo in the 4.5- to 9-hour time window

DIAS 3 and 4 trials.30 The prior desmoteplase studies

included patients studied either by computed tomography

(CT) or magnetic resonance (MR) angiography, but an

occluded vessel was not required for inclusion. A restricted

core volume was assumed based on either a lack of early

ischemic changes on routine CT or DWI and a mismatch

was required for indusion. A post hoc analysis of the prior

DIAS trials indicated an encouraging response to desmote-

plase in patients who presented with a high-grade stenosis

or occlusion on baseline imaging, whereas patients without

a visualized vascular lesion generally did well with or with-

out treatment.31 Furthermore, the combination of an

occluded intracranial vessel and a limited DWI lesion was

also correlated with favorable responses to reperfusion in

both the DEFUSE and EPITHET studies.32,33

EXTEND is an ongoing randomized double blind

trial of i.v. tPA compared to placebo in the 3- to 9-hour

time window (4.5 to 9 hours in countries that routinely

use tPA up to 4.5 hours).34 The design of EXTEND was

based on the results of EPITHET and DEFUSE and uses

the same automated software program employed in

DEFUSE 2 to randomize only patients with the Target

Mismatch profile on either MRI or perfusion CT (CTP).

For MRI patients to qualify for enrollment, the PWI

lesion identified by a Tmax delay of >6 seconds must be

at least 20% larger than the DWI lesion volume, and the

infarct core cannot be >70ml on DWI. For the patients

enrolled with CTP, the infarct core is estimated based on

regions with >70% reduction in cerebral blood flow

compared with normal contralateral tissue, and the hypo-

perfused volume must be at least 20% larger than the

estimated infarct core. Critically hypoperfused tissue is

determined using the same Tmax delay of >6 seconds on

CTP that is used with MRI. A very similar trial will be

conducted in Europe (ECASS 4) and is scheduled to

begin later this year. ECASS 4 will enroll Target

Mismatch patients in the 4.5- to 9-hour window using

the same MRI selection criteria and software as

EXTEND. These 2 trials are directly testing the hypothe-

sis that penumbral identification can select patients that

respond to i.v. tPA in a late time window. The success or

failure of these and other ongoing trials will provide

valuable information to guide future clinical practice and

investigations.

Other ongoing trials, such as the National Institute

of Neurological Disorders and Stroke-funded MR WIT-

NESS study,35 are using MRI to help determine whether

FIGURE 3: A 74-year-old female with witnessed onset ofleft-sided paralysis (National Institutes of Health StrokeScale [NIHSS] 5 12). (A) Despite a middle cerebral arteryocclusion (not shown), the patient had a small DWI lesion(pink) and large PWI/DWI mismatch 8.5 hours after symptononset. The PWI Tmax >6 seconds lesion is shown in green.(B) Following endovascular reperfusion therapy (reperfusionoccurred 10 hours after symptom onset), the patientimproved (NIHSS 5 3) and had a small infarct on 5-day fluidattenuated inversion recovery imaging.

Fisher and Albers: Penumbral Imaging of AIS

January 2013 7

it is safe to treat stroke patients with unwitnessed symp-

tom onset with i.v. tPA. The premise of this approach is

that if a DWI positive lesion has not yet developed sig-

nificant positivity on fluid attenuated inversion recovery

imaging, then penumbral tissue is likely to be present,

and the ischemic lesion is unlikely to hemorrhage follow-

ing reperfusion.

Another study design could be envisioned in which

consecutive patients who meet current i.v. tPA eligibility

criteria and are treated <4.5 hours from symptom onset

are enrolled. A baseline DWI/PWI and MR angiogram

or CTP and CT angiogram are obtained either prior to

or concurrent with treatment. In addition, an early

assessment of reperfusion would be obtained several

hours later. The hypothesis to be tested by such a study

is whether reperfusion of patients following i.v. tPA is

associated with better functional outcomes in Target Mis-

match patients compared with the No Mismatch and

Malignant profiles. Recent pilot data suggest that the

Malignant profile can be detected with CTP in about

10% of tPA-eligible patients who are imaged within 3

hours of symptom onset,36 and these patients had

extremely poor clinical outcomes following tPA treat-

ment. A recent study that treated 75 stroke patients at a

mean of 3 hours from symptom onset demonstrated that

in CTP-selected patients with a favorable imaging pro-

file, intravenous tenecteplase was associated with signifi-

cantly better reperfusion and clinical outcomes than in-

travenous tPA.37 That statistically significant differences

in clinical outcomes were demonstrated in this study, de-

spite the very small sample size, provides support for the

concept that imaging-based selection can reduce sample

size requirements.

Based on the failure of IMS III to demonstrate ben-

efits of endovascular therapy over i.v. tPA (discussed

above), additional randomized assessment of endovascular

therapies is essential. MR Rescue is an NIH-sponsored

trial in which patients eligible for endovascular therapy

were randomized to treatment with mechanical throm-

bectomy devices versus no endovascular treatment, using

an 8-hour time window.38 A multimodal MRI or CT

was performed prior to endovascular treatment. The

study stratified randomization based on the initial imag-

ing findings, to test the hypothesis that patients with a

penumbral pattern benefit from endovascular therapy,

and patients with a nonpenumbral pattern do not

(data anticipated in early 2013). With the recent SWIFT

and TREVO 2 trials demonstrating superiority of the

new stent-retrievers over the Merci device, as well as the

negative IMS III study (discussed above), randomized

data comparing the new generation of mechanical devices

to no endovascular therapy are greatly needed.

We believe the use of advanced imaging to select

penumbral patients for trials of reperfusion therapies is

sensible and will allow conclusive studies to be completed

rapidly and with manageable sample sizes. Current evi-

dence suggests that imaging findings have the potential

to replace time from symptom onset as the key determi-

nant of the clinical response to reperfusion. If this pre-

mise can be validated in prospective randomized trials, it

will lead to a paradigm shift in acute stroke therapy. At

this time, penumbral identification with DWI/PWI is

further advanced than with CTP; there are more exten-

sive data available regarding the utility of MRI for identi-

fication of patients who are more or less likely to

respond to treatment. As the clinical trial experience with

CTP expands and thresholds for distinguishing core and

penumbral tissue are better validated, it is anticipated

that this imaging modality will lead future efforts to

implement penumbral imaging in daily clinical practice

because of its more widespread availability in most cen-

ters. Imaging-based identification of eligible patients for

reperfusion therapy will not only reduce treatment-

related complications but also result in treatment of a

larger population of patients over an extended time win-

dow than the current time-based approaches.

Potential Conflicts of Interest

M.F.: employment, UMass/Memorial Healthcare; grants/

grants pending, NINDS; stock/stock options, Photothera,

Brainsgate; Editor in Chief of Stroke (AHA). G.W.A.:

consultancy, Lundbeck, Covidien, Stryker, Genetech,

Concentric; grants/grants pending, NIH, Lundbeck;

stock/stock options, iSchemaView.

References1. Ringelstein EB for the AXIS-2 study group. The AXIS-2 study: AX 200

for the treatment of acute ischemic stroke. Stroke 2012;43:A172.

2. Davalos A, Alvarez-Sabin J, Castillo J, et al. Citicoline in the treat-ment of acute ischaemic stroke: an international, randomised,multicentre, placebo-controlled study (ICTUS trial). Lancet 2012;380:349–357.

3. Khala AM, Grotta JC. Established treatments for acute ischaemicstroke. Lancet 2007;369:319–330.

4. Hacke W, Kaste M, Bluhmki E, et al. Thrombolysis with alteplase 3to 4.5 hours after acute ischemic stroke. N Engl J Med 2008;359:1317–1329.

5. IST-3 Collaborative Group. The benefits and harms of intravenousthrombolysis with recombinant tissue plasminogen activator within6 h of acute ischaemic stroke (the third international stroke trial[IST-3]): a randomized trial. Lancet 2012;379:2352–2363.

6. Lees KR, Bluhmki E, von Kummer R, et al. Time to treatment withintravenous alteplase and outcome in stroke: an updated pooledanalysis of ECASS, ATLANTIS, NINDS, and EPITHET trials. Lancet2010;375:1695–1703.

7. Smith WS, Sung G, Saver J, et al. Mechanical thrombectomy foracute ischemic stroke: Final results of the Multi MERCI trial. Stroke2008;39:1205–1211.

ANNALS of Neurology

8 Volume 73, No. 1

8. Penumbra Pivotal Stroke Trial Investigators. The penumbra pivotalstroke trial: safety and effectiveness of a new generation of me-chanical devices for clot removal in intracranial large vessel occlu-sive disease. Stroke 2009;40:2761–2768.

9. Saver JL, Jahan R, Levy EI, et al. Solitaire flow restoration deviceversus the MERCI Retriever in patients with acute ischaemic stroke(SWIFT): a randomised, parallel-group, non-inferiority trial. Lancet2012;380:1241–1249.

10. Nogueira RG, Lutsep HL, Gupta R, et al. TREVO versus MERCIretrievers for thrombectomy revascularization of large vesselocclusions in acute ischaemic stroke (TREVO2): a randomized trial.Lancet 2012;380:1231–1240.

11. Mazighi M, Serfaty JM, Labreche J, et al. Comparison of intrave-nous alteplase with a combined intravenous-endovascularapproach in patients with stroke and confirmed arterial occlusion(RECANALISE study): a prospective cohort. Lancet Neurol 2009;8:802–809.

12. Jovin TG, Liebeskind DS, Gupta RA, et al. Imaging-based endo-vascular therapy for acute ischemic stroke due to proximal intra-cranial anterior circulation occlusion treated beyond 8 hours fromtime last seen well: retrospective multicenter analysis of 237 con-secutive patients. Stroke 2011;42:2206–2211.

13. Rha JH, Saver JL. The impact of recanalization on ischemic strokeoutcome: a meta-analysis. Stroke 2007;38:967–973.

14. Mlynash M, Lansberg MG, De Silva DA, et al. Refining the defini-tion of the malignant profile: insights from the DEFUSE-EPITHETpooled data set. Stroke 2011;42:1270–1275.

15. Donnan GA, Baron JC, Ma H, Davis SM. Penumbral selection ofpatients for acute stroke therapy. Lancet Neurol 2009;8:261–269.

16. Fisher M. Characterizing the target of acute stroke treatment.Stroke 1997;28:866–872.

17. Martini SR, Kent TA. Hyperglycemia in acute ischemic stroke: avascular perspective. J Cereb Blood Flow Metab 2007;27:435–451.

18. Muir KW, Buchan A, von Kummer R, et al. Imaging of acutestroke. Lancet Neurol 2006;5:755–768.

19. Chemmanam T, Campbell BC, Christensen S, et al. Ischemic diffu-sion lesion reversal is uncommon and rarely alters perfusion-diffu-sion mismatch. Neurology 2010;75:1040–1047.

20. Campbell BC, Purushotham A, Christensen S, et al. The infarctcore is well represented by the acute diffusion lesion: sustainedreversal is infrequent. J Cereb Blood Flow Metab 2012;32:50–56.

21. Baird AE, Warah S. Magnetic resonance imaging of acute stroke.J Cereb Blood Flow Metab 1998;18:583–609.

22. Dani KA, Thomas RG, Chappell FM, et al. Computed tomographyand magnetic resonance perfusion imaging in ischemic stroke:definitions and thresholds. Ann Neurol 2011;70:384–401.

23. Olivot J-M, Mlynash M, Thijs VN, et al. Optimal Tmax thresholdfor predicting penumbral tissue in acute stroke. Stroke 2009;40:469–475.

24. Zaro-Weber O, Moeller-Hartmann W, Heiss W-D, Sobesky J.Maps of maximum and time to peak for mismatch definition inclinical stroke studies validated with positron emission tomogra-phy. Stroke 2010;41:2817–2821.

25. Lansberg MG, Lee J, Christensen S, et al. RAPID automatedpatient selection for reperfusion therapy: a pooled analysis of theEchoplanar Imaging Thrombolytic Evaluation Trial (EPITHET) andthe Diffusion and Perfusion Imaging Evaluation for UnderstandingStroke Evolution (DEFUSE) Study. Stroke 2011;42:1608–1614.

26. Straka M, Albers GW, Bammer R. Real-time diffusion-perfusionmismatch analysis in acute stroke. J Magn Reson Imaging 2010;32:1024–1037.

27. Lansberg MG, Straka M, Kemp S, et al. MRI profile and responseto endovascular reperfusion after stroke (DEFUSE 2): a prospectivecohort study. Lancet Neurol 2012;11:860–867.

28. Albers GW, Mylnash M, Hamilton S, et al. Benefits of endovascu-lar reperfusion are maintained in Target Mismatch patients in latetime windows. European Stroke Conference abstract e-book.Basel, Switzerland: Karger, 2012:80.

29. Mylnash M, Lansberg M, Straka, M, et al. The Malignant MRI pro-file: implications for endovascular therapy. Stroke 2012;43:A53.

30. von Kummer R, Albers GW, Mori E on behalf of the DIAS SteeringCommittee. The desmoteplase in acute ischemic stroke (DIAS)clinical trial programme. Int J Stroke (in press).

31. Fiebach JB, Al-Rawi Y, Wintermark M, et al. Vascular occlusionenables selecting acute ischemic stroke patients for treatmentwith desmoteplase. Stroke 2012;43:1561–1566.

32. Lansberg MG, Thijs VN, Bammer R, et al. The MRA-DWI mismatchidentifies patients with stroke who are likely to benefit from reper-fusion. Stroke 2008;39:2491–2496.

33. De Silva DA, Churilov L, Olivot JM, et al. Greater effect of strokethrombolysis in the presence of arterial obstruction. Ann Neurol2011;70:601–605.

34. Ma H, Parsons MW, Christensen S, et al on behalf of the EXTENDinvestigators. Extend A multicentre, randomized, double-blinded,placebo-controlled phase III study to investigate EXtending thetime for Thrombolysis in Emergency Neurological Deficits(EXTEND). Int J Stroke 2012;7:74–80.

35. MR WITNESS. Available at: http://clinicaltrials.gov/ct2/show/NCT01282242?term¼MRþWITNESS

36. Inoue M, Mlynash M, Straka M, et al. Patients with the MalignantProfile within 3 hours of symptom onset have very poor outcomesfollowing IV tPA therapy. Stroke 2012;43:2494–2496.

37. Parsons M, Spratt N, Bivard A, et al. A randomized trial of tenec-teplase versus alteplase for acute ischemic stroke. N Engl J Med2012;366:1099–1107.

38. US National Institutes of Health Website, MR and recanalization ofstroke clots using embolectomy (MR RESCUE) trial. Available at:clinicaltrials.gov/ct/show/NCT01282242. Accessed July 24, 2012.

January 2013 9