Perfusion Imaging in Acute Ischemic Stroke: Let Us Improve the Science before Changing Clinical Practice

Article (PDF Available)inRadiology 266(1):16-21 · January 2013with12 Reads
DOI: 10.1148/radiol.12112134 · Source: PubMed
Reviews and CommentaRy
Radiology: Volume 266: Number 1—January 2013
Perfusion Imaging in Acute
Ischemic Stroke: Let Us Improve
the Science before Changing Clinical
Mayank Goyal, MD, FRCPC
Bijoy K. Menon, MD, DM
Colin P. Derdeyn, MD
Published online
10.1148/radiol.12112134 Content codes:
Radiology 2013; 266:16–21
From the Departments of Radiology and Clinical Neurosci-
ences and Hotchkiss Brain Institute, University of Calgary,
Calgary, Alberta, Canada (M.G., B.K.M.); and Mallinckrodt
Institute of Radiology, Washington University School of
Medicine, St Louis, Mo (C.P.D.). Received March 24, 2012;
revision requested April 16; final revision received July 12;
accepted August 3; final version accepted August 25. Ad-
dress correspondence to M.G., Department of Radiology,
Foothills Medical Centre, 1403 29th St NW, Calgary, AB,
Canada T2N 2T9 (e-mail:
Conflicts of interest are listed at the end of this article.
See also the article by Lev in this issue.
RSNA, 2013
eperfusion is the only proved ef-
fective therapy for patients with
acute ischemic stroke (1). There
is an urgent need for improvement in
acute stroke care, as only 15%–20%
of patients, at best, are eligible for in-
travenous tissue plasminogen activator
(tPA), and many patients so treated do
not achieve good clinical outcome (2,3).
This need has resulted in an enormous
amount of research about developing
better methods of revascularization
and neuroprotection and designing
better imaging paradigms for patient
selection. A key element of these ad-
vanced imaging paradigms is perfusion
We intend to bring focus on the
extent of our current knowledge on
the imaging triaging of acute ischemic
stroke patients by using perfusion imag-
ing and to make recommendations for
practice and clinical research. Our po-
sition is that perfusion imaging should
not be used outside of clinical studies
seeking to establish the utility of these
measurements. We do not intend to
focus on the utility of perfusion imag-
ing in other relevant clinical scenarios
where this imaging modality has de-
fined its role, and neither do we intend
to go into detail on the mathematic con-
structs behind these techniques.
Definitions of Core, Penumbra, and
Benign Oligemia
The core is tissue that is already dead
at the time of imaging. For our pur-
pose, we shall define penumbra as is-
chemic, nonfunctioning but living brain
tissue that will die unless blood flow is
restored. Benign oligemia is tissue that
is underperfused but functioning nor-
mally and that will survive irrespective
of improvement in blood supply (4–6).
The first indication of the presence
of penumbra in ischemic brain tissue
came from electrophysiologic studies
in animals. In a series of experiments
in cats, Hossman et al (7) reported
the restoration of electrical activity in
individual neurons with the restora-
tion of cerebral blood flow. They also
found that the longer the duration and
the greater the depth of ischemia, the
less likely that electrical function would
return. The threshold level for neuro-
nal dysfunction in these studies was a
cerebral blood flow (CBF) of approxi-
mately 20 mL/100 g/min. Tissue with
a value above this CBF threshold level
never infarcts; this region was defined
as benign oligemia. These animal ex-
periments also demonstrated the dy-
namic nature of the ischemic penum-
bra: Without reperfusion, the infarct
core expands into the penumbra over
time (4,8).
Imaging of Core, Penumbra, and Benign
Over the past 20 years, we have devel-
oped good imaging methods to delin-
eate core with some certainty. These in-
clude low attenuation on nonenhanced
head computed tomography (CT) and
on CT angiographic source images, low
cerebral blood volume or low CBF at
perfusion CT, and restricted diffusion
by using magnetic resonance (MR) im-
aging (9–15). These imaging modalities
are used to identify core by allowing a
comparison of baseline imaging param-
eters mentioned above with final infarct
in early reperfusers, unlike positron
emission tomographic (PET) studies
in which pathophysiological constructs
such as oxygen extraction fraction or
flumazenil binding are used (16,17).
Studies validating imaging measures of
core at nonenhanced CT (hypoattenu-
ation) or MR diffusion-weighted imag-
ing with PET have been impossible to
perform early enough and close enough
to PET such that the core infarct mea-
surements are accurate (18,19). With
Note: This copy is for your personal non-commercial use only. To order presentation-ready
copies for distribution to your colleagues or clients, contact us at
Radiology: Volume 266: Number 1—January 2013
n 17
CONTROVERSIES: Perfusion Imaging in Acute Ischemic Stroke Is Promising Goyal et al
make certain assumptions. One source
of error is with the choice and measure-
ment of the arterial input function (AIF)
(30). Perfusion techniques rely on the as-
sumption that the AIF is from an artery
that is the sole blood supply to the region
where blood flow is measured. This as-
sumption is not always true. Variability in
the choice of AIF can lead to variability
in blood flow measurements caused by
varying delay and/or dispersion (31,32).
Singular-value decomposition techniques
correct for these errors; however, not all
available perfusion software uses these
techniques (33,34). Partial volume av-
eraging of the AIF can lead to errors in
measuring blood flow (35). These errors
can be corrected by appropriate scaling
techniques by using the venous output
function. MR perfusion techniques are
affected because of low spatial resolution,
orientation of the vessel with respect to
the direction of the magnetic field, and
the mixing of tissue signal intensity with
signal from the artery. Additional as-
sumptions that are not always accurate
are in determining the proportionality
constant K, an integral part of decon-
volution algorithms and dependent on
capillary hematocrit levels that can vary
in ischemic tissue (36,37). To overcome
this drawback and to achieve a means
of comparison between estimated CBF
the penumbra into core. If leptomen-
ingeal collateral vessels fail (Figure)
or CBF varies with time because of as
yet unknown factors, defining penum-
bra by using perfusion imaging at a sin-
gle time will be error prone (27).
The dynamic nature of the penum-
bra is reflected in variable infarct growth
rates over time. The transformation
of penumbra into core in the infarct
“border zone” is dependent on many
complex and interrelated factors, in-
cluding leptomeningeal collateral supply
(Figure), ischemic tissue susceptibility to
necrosis, and peri-infarct depolarization
(6,25). These same factors lead to var-
iable growth of penumbra into regions
of oligemia and result in clinical wors-
ening (28). In addition, in vitro studies
hint at the possibility of selective neuro-
nal death in the penumbra and protein
dysfunction in the region with benign
oligemia (29). The use of blood flow
threshold levels to define irreversibility
and salvageability of brain tissue and to
predict infarct growth rates is affected
conceptually by these physiologic issues.
Technical Limitations of Current
Perfusion Techniques
Present perfusion techniques (either de-
convolution or nondeconvolution based)
current advances in endovascular treat-
ment, the imaging definition of what is
core at perfusion CT or MR imaging is
constantly changing as the benchmark
for “early successful reperfusion” re-
duces. It is, however, our opinion that
the current definition of core is reli-
able enough for clinical decision mak-
ing (20–22). The accurate separation
of penumbra from benign oligemia is
a challenge (23). In addition, there is
very likely tissue within imaging-defined
penumbra that is not dead yet, but may
be doomed to die even if reperfused
(because of apoptosis or other mech-
anisms) (24). A single measurement
of CBF or other related perfusion esti-
mates made at a single point in time is
not a reliable indicator of whether that
tissue will live if left alone or survive if
reperfused. Our ability to identify core,
penumbra, and benign oligemia by us-
ing blood flow measures is affected by
conceptual and technical flaws.
Conceptual Issues with Defining Tissue
State by Using Perfusion Measures
The concept of defining the size and
extent of penumbra and distinguishing
it from infarct core and benign olige-
mia by using perfusion threshold levels
(CBF, cerebral blood volume, time
to maximum, and mean transit time,
or summary parameters such as time
to peak) has several major problems.
First, these perfusion threshold levels
are time dependent. Tissue with a CBF
of less than 10 mL/100 g/min may sur-
vive for 30 minutes, but probably not
for 3 hours. Perfusion threshold levels
that separate core from penumbra
could therefore depend on when tissue
was imaged after ischemia onset (25).
Second, perfusion measurements do
not reflect metabolic activity of brain
tissue. Dead, dying, or doomed tissue
may have normal flow. Perfusion thresh-
old levels for infarction may be different
in various regions of the brain (26). It
is also important to note that the pen-
umbra is a dynamic state. Any attempt
to determine the status of ischemic and
oligemic tissue is a snapshot in time
(6). We do not know the factors that
determine the rate of transformation of
Schematic shows regions beyond an arterial occlusion with border zones among core, penumbra, and benign
oligemia. These regions may change over time owing to possible change in leptomeningeal collateral vessels
and other toxic metabolic or physiologic processes, including periinfarct depolarization.
Radiology: Volume 266: Number 1—January 2013
CONTROVERSIES: Perfusion Imaging in Acute Ischemic Stroke Is Promising Goyal et al
values from different patients, cross-
calibration is used by some researchers
in conjunction with deconvolution-based
approaches (36,37). Although absolute
quantification may indeed vary, we do
agree that the use of relative threshold
levels could help in unilateral ischemic
stroke evaluation. Nonetheless, white
matter disease, age-related changes and
previous infarcts can compromise the va-
lidity of cross-calibration. In addition, dis-
criminating between perfusion threshold
levels for ischemia in the gray and white
matter is prone to errors (38,39).
The use of parameters such as
mean transit time and time to max-
imum are confounded by reliability
and by issues of dispersion and delay
(40). There is a distinct lack of stan-
dardization of postprocessing tools
in perfusion imaging not only across
modalities (CT vs MR imaging) but
also across vendors and laboratories
(41,42). Lack of standardization of
postprocessing tools is, in our opin-
ion, a major limitation. Efforts toward
increasing the speed of postprocessing
have led to automated postprocessing
tools that could further worsen these
measurement errors, although this
idea is debatable (43,44).
The Issue of Time Taken for Perfusion
There seems to be a misconception
that perfusion imaging in real time only
takes a few minutes. This may be true
for image acquisition time (45,46) but
not for time from entry into the imaging
suite to making a treatment decision.
For CT perfusion, this time includes
setting up the imaging unit and injec-
tor, acquiring and transferring images,
postprocessing, and finally interpret-
ing the data. For MR perfusion, other
additional steps are required: making
sure the patient is MR safe, patient
positioning, motion, claustrophobia,
and obtaining localizing images. These
steps may easily extend the total imag-
ing time by 30 minutes (47). Perfusion
imaging–based paradigms may not be
practical in a real world scenario unless
ongoing efforts at reducing image post-
processing time succeed (48).
Other Limitations of Perfusion Imaging
Limited coverage of perfusion imaging
and the need for a different sequence
to image neck vessels and the arch of
aorta for planning endovascular treat-
ment are also disadvantages. This limi-
tation can be addressed by newer-gen-
eration whole-brain CT scanners or MR
imaging units; however, not all centers
have access to these tools. Additional
radiation and contrast agent dose given
to patients is also a concern and must
be justified by potential benefits ob-
tained when making clinical decisions.
Appropriate Imaging for Patient
Selection in Clinical Practice
In the next few paragraphs, we explore
the questions that clinicians need an-
swers for when making treatment
decisions in patients with acute ische-
mic stroke (49). Our effort is to ex-
plore whether these questions can be
answered satisfactorily without using
perfusion imaging. The questions are
as follows: (a) Is there an arterial oc-
clusion? (b) What is the extent of core
within or relative to the hypoperfused
arterial territory? (c) Can the core so
defined possibly explain the patient’s
clinical deficits? (d)What is the rate of
growth infarct?
CT angiography is a reliable, safe,
quick, and widely available tool to de-
termine the presence of an arterial oc-
clusion. Tools like diffusion-weighted
imaging and apparent diffusion coeffi-
cient in MR imaging define core as well
as or even better than perfusion mea-
surements (20). Hypoattenuation on a
good-quality nonenhanced CT image
unless very early is also fairly reliable
in estimating core (9,50,51). We agree
that perfusion imaging also identifies
core (15,52). At our centers, with rea-
sonable training, patients with large
early ischemic changes (correspond-
ing to an Alberta Stroke Program Early
CT Score of four or lower) are reliably
identified and excluded from throm-
bolytic treatment unless their disease
manifests very early (53,54). Propo-
nents of perfusion imaging use their
technique to do exactly the same.
A lack of attention to improving the
quality of nonenhanced CT and to train-
ing in interpretation of early ischemic
changes on nonenhanced CT images
results in underutilization of this eas-
ily available and rapid tool (55). This
lack of attention to nonenhanced CT
has developed to such an extent that no
attention is paid to change in hypoat-
tenuation over time, a marker of time
from stroke onset (56). We would like
to remind the readers that perfusion
imaging and the mismatch paradigm
gives a sense of extent of “core” and
“salvageable brain tissue” but does not
provide a sense of “time from stroke
onset to imaging.” In our opinion, in-
formation similar to that obtained from
perfusion imaging can be gleaned from
nonenhanced CT and CT angiography
by assessing extent of early ischemic
changes and degree of leptomeningeal
collateral vessels. In addition, assess-
ment of hypoattenuation on nonen-
hanced CT images can provide us with
a sense of time from stroke symptom
onset to imaging. Diffusion-weighted
imaging fluid-attenuated inversion-re-
covery mismatch does provide a sense
of this “image time” conceptually, but
that paradigm is not an argument for
perfusion imaging.
Clinicians have a fairly good idea of
the presence or absence of clinical-im-
aging mismatch (57). This point merits
some discussion, as there is contra-
dictory literature on the accuracy and
relevance of clinical-imaging mismatch
(58,59). A careful look at this literature
reveals that the use of a volume thresh-
old level for small core rather than clin-
ical judgment on the basis of anatomy
and eloquence could be responsible for
the lower specificity of clinical-imaging
(CT or MR) mismatch when compared
with perfusion-diffusion mismatch in
predicting clinical outcome (60). Pre-
dicting infarct growth rate by using
perfusion estimates is possible on the
basis of an assessment of severity of is-
chemia within penumbra; nonetheless,
other factors such as change in lepto-
meningeal collateral vessel status, sus-
ceptibility of brain tissue to ischemia
and peri-infarct depolarization could
confound this assessment. In summary,
Radiology: Volume 266: Number 1—January 2013
n 19
CONTROVERSIES: Perfusion Imaging in Acute Ischemic Stroke Is Promising Goyal et al
these studies are accomplished, design-
ing randomized acute stroke trials by
using perfusion imaging for patient se-
lection should be attempted.
Only then should a clinician use this
technology in the real world to make
clinical decisions. We do believe that
the alternative CT-based paradigm that
we propose, a small-core–occlusion par-
adigm should also go through the same
hierarchy of steps mentioned above.
Finally, we point out that perfusion
imaging helps in detecting the presence
of distal occlusions that may be missed
at CT angiography, albeit not changing
a decision to use thrombolytic therapy.
It has the potential to identify hemor-
rhagic transformation better than non-
enhanced CT hypoattenuation or diffu-
sion-weighted imaging fluid-attenuated
inversion recovery, although that has to
be tested (69). It also identifies stroke
mimics. However, even when faced with
the possibility of stroke mimics such as
seizures, by identifying the presence or
absence of an occlusion, CT angiogra-
phy helps in identifying patients who are
candidates for thrombolysis (70). CT an-
giography is also as good, if not better,
at identifying worsening in patients with
a transient ischemic attack (71).
The patient’s clinical stroke severity at
presentation gives a fairly good esti-
mate of dysfunctional brain tissue. The
clinician is able to determine with rea-
sonable certainty whether the patient’s
clinical deficits are explained by tissue
that is defined as core on a good-qual-
ity nonenhanced CT scan. CT angiog-
raphy reveals the presence of an oc-
clusion and also allows for assessment
of collateral vessels. This small-core–
proximal occlusion paradigm with clin-
ical-imaging mismatch is a very good
surrogate for information that can be
obtained by the penumbra-core mis-
match paradigm at perfusion imaging
and takes less imaging time. Although
no one disputes the value of the con-
cept of penumbra and mismatch in
acute ischemic stroke treatment, cur-
rent perfusion techniques lack robust
evidence validating their use in reliably
of treatment effect with intravenous
tPA in the time window of greater
than 3 hours (20,63–65). In addition
to the many reasons mentioned for
the lack of substantial treatment ef-
fect of these trials (65), the lack of
a treatment endovascular arm result-
ing in faster reperfusion may also be
a reason. Faster reperfusion is im-
perative. Prolyse in Acute Cerebral
Thromboembolism II supports the use
of endovascular therapy for proximal
occlusions in this time window (66).
Post hoc analysis of data from Prolyse
in Acute Cerebral Thromboembolism
II and Echoplanar Imaging Thrombo-
lytic Evaluation Trial shows treatment
effects with small cores identified by
using either nonenhanced CT or dif-
fusion-weighted imaging (67,68). In
the absence of compelling evidence
for the use of intravenous tPA with
or without perfusion imaging in this
time window (20,21) and the need for
identifying a proximal occlusion be-
fore initiating endovascular therapy,
the use of nonenhanced CT and CT
angiography preferably or diffusion-
weighted imaging and MR angiog-
raphy and the small-core–occlusion
imaging paradigm could be faster and
more clinically relevant.
Future Directions
We advocate a hierarchy of steps to-
ward the use of perfusion imaging in
clinical decision making that is, in our
opinion, appropriate: (a) First, stan-
dardization of imaging protocols and
image processing tools (CT vs MR im-
aging, what algorithm to use, differ-
ences across vendors) and data acqui-
sition (rate of contrast agent injection,
amount of contrast agent, toggling table
vs static table) should be performed.
These issues are easier to solve and
require a cooperative effort between
the scientific community and industry.
In addition, rapid postprocessing soft-
ware (possibly automated) that gives
good-quality data to treating physicians
in real time needs to be designed. (b)
Next, carefully designed studies ad-
dressing some of the conceptual issues
highlighted above are needed. (c) Once
if a clinician has knowledge of where
the occlusion is and of how complex the
arterial access is and is able to glean
from imaging information on whether
the core is relatively small or large
when compared with a potentially hy-
poperfused brain region, we believe
he or she is able to make a decision
in favor or against thrombolytic ther-
apy either intravenous or endovascular.
We request our readers to question
whether perfusion imaging answers any
of the above questions better than a
small core–occlusion–based paradigm
by using good quality nonenhanced CT
or CT angiography. We also request
them to note the fact that clinicians in
many centers where perfusion imaging
is used make the decision to use throm-
bolytic therapy even before perfusion
imaging is available. On the basis of this
argument, we therefore suggest imag-
ing strategies classified in two groups
as follows: 0–4.5-hour time window or
longer than 4.5 hours from onset or un-
known time of onset.
0–4.5-Hour Time Window
In this time window, there is level 1 ev-
idence for the use of intravenous tPA
(61). Perfusion imaging has a limited
role in determination of appropriate
patients for intravenous thrombolysis
(20,21). The rationale for endovascular
treatment in this time window is based
on the low revascularization rates and
poor outcomes with intravenous tPA
in proximal occlusions. A patient with
moderate to severe stroke symptoms
with a small core identified at CT and
large-vessel occlusion identified at CT
angiography is most likely to benefit
from endovascular treatment (62). In
our opinion, a CT and CT angiography–
based imaging paradigm is quick and
easy within this time window without
prolonging door-to-needle time.
Longer than 4.5 Hours from Onset or
Unknown Time of Onset
Major clinical trials in which perfusion
imaging and a mismatch paradigm
(Desmoteplase in Acute Ischemic
Stroke 2 trial, Echoplanar Imaging
Thrombolytic Evaluation Trial) were
used have not shown level 1 evidence
Radiology: Volume 266: Number 1—January 2013
CONTROVERSIES: Perfusion Imaging in Acute Ischemic Stroke Is Promising Goyal et al
chemic stroke: a scientific statement from
the American Heart Association. Stroke
21. Schellinger PD, Bryan RN, Caplan LR, et
al. Evidence-based guideline: the role of
diffusion and perfusion MRI for the diag-
nosis of acute ischemic stroke—report of
the Therapeutics and Technology Assess-
ment Subcommittee of the American Acad-
emy of Neurology. Neurology 2010;75(2):
22. Schaefer PW, Mui K, Kamalian S, Nogueira
RG, Gonzalez RG, Lev MH. Avoiding “pseudo-
reversibility” of CT-CBV infarct core lesions
in acute stroke patients after thrombolytic
therapy: the need for algorithmically “delay-
corrected” CT perfusion map postprocessing
software. Stroke 2009;40(8):2875–2878.
23. Alawneh JA, Jones PS, Mikkelsen IK, et
al. Infarction of ‘non-core-non-penumbral’
tissue after stroke: multivariate modelling
of clinical impact. Brain 2011;134(pt 6):
24. Kametsu Y, Osuga S, Hakim AM. Apoptosis
occurs in the penumbra zone during short-
duration focal ischemia in the rat. J Cereb
Blood Flow Metab 2003;23(4):416–422.
25. Back T. Pathophysiology of the ischemic pen-
umbra: revision of a concept. Cell Mol Neu-
robiol 1998;18(6):621–638.
26. Payabvash S, Souza LC, Wang Y, et al. Re-
gional ischemic vulnerability of the brain
to hypoperfusion: the need for location
specific computed tomography perfusion
thresholds in acute stroke patients. Stroke
27. Menon BK, Smith EE, Modi J, et al. Re-
gional leptomeningeal score on CT angiogra-
phy predicts clinical and imaging outcomes
in patients with acute anterior circula-
tion occlusions. AJNR Am J Neuroradiol
28. Alawneh JA, Moustafa RR, Baron JC. Hemo-
dynamic factors and perfusion abnormalities
in early neurological deterioration. Stroke
29. Sharp FR, Lu A, Tang Y, Millhorn DE. Mul-
tiple molecular penumbras after focal cere-
bral ischemia. J Cereb Blood Flow Metab
30. Thijs VN, Somford DM, Bammer R, Rob-
berecht W, Moseley ME, Albers GW. Influence
of arterial input function on hypoperfusion
volumes measured with perfusion-weighted
imaging. Stroke 2004;35(1):94–98.
31. Sanelli PC, Lev MH, Eastwood JD, Gonza-
lez RG, Lee TY. The effect of varying user-
selected input parameters on quantitative
values in CT perfusion maps. Acad Radiol
32. Ferreira RM, Lev MH, Goldmakher GV,
et al. Arterial input function placement
for accurate CT perfusion map construc-
tion in acute stroke. AJR Am J Roentgenol
33. Konstas AA, Lev MH. CT perfusion imaging
of acute stroke: the need for arrival time, de-
5. Baron JC, Marchal G. Ischemic core
and penumbra in human stroke. Stroke
6. Heiss WD. Ischemic penumbra: evidence
from functional imaging in man. J Cereb
Blood Flow Metab 2000;20(9):1276–1293.
7. Hossmann KA, Heiss WD, Bewermeyer H,
Mies G. EEG frequency analysis in the course
of acute ischemic stroke. Neurosurg Rev
8. Jones TH, Morawetz RB, Crowell RM, et al.
Thresholds of focal cerebral ischemia in awake
monkeys. J Neurosurg 1981;54(6):773–782.
9. Barber PA, Demchuk AM, Zhang J, Buchan
AM. Validity and reliability of a quantitative
computed tomography score in predicting
outcome of hyperacute stroke before throm-
bolytic therapy: ASPECTS study group—Al-
berta Stroke Programme Early CT Score.
Lancet 2000;355(9216):1670–1674.
10. Wintermark M, Flanders AE, Velthuis B, et
al. Perfusion-CT assessment of infarct core
and penumbra: receiver operating charac-
teristic curve analysis in 130 patients sus-
pected of acute hemispheric stroke. Stroke
11. Parsons MW, Pepper EM, Chan V, et al. Per-
fusion computed tomography: prediction of
final infarct extent and stroke outcome. Ann
Neurol 2005;58(5):672–679.
12. Barber PA, Darby DG, Desmond PM, et al.
Prediction of stroke outcome with echopla-
nar perfusion- and diffusion-weighted MRI.
Neurology 1998;51(2):418–426.
13. Schäbitz WR, Fisher M. Diffusion weighted
imaging for acute cerebral infarction. Neurol
Res 1995;17(4):270–274.
14. Moseley ME, Cohen Y, Mintorovitch J, et al.
Early detection of regional cerebral ischemia
in cats: comparison of diffusion- and T2-
weighted MRI and spectroscopy. Magn Reson
Med 1990;14(2):330–346.
15. Campbell BC, Christensen S, Levi CR, et al.
Cerebral blood flow is the optimal CT per-
fusion parameter for assessing infarct core.
Stroke 2011;42(12):3435–3440.
16. Derdeyn CP, Videen TO, Yundt KD, et al.
Variability of cerebral blood volume and
oxygen extraction: stages of cerebral hae-
modynamic impairment revisited. Brain
2002;125(pt 3):595–607.
17. Heiss WD, Sobesky J. Comparison of PET
and DW/PW-MRI in acute ischemic stroke.
Keio J Med 2008;57(3):125–131.
18. Heiss WD, Sobesky J, Smekal U, et al. Prob-
ability of cortical infarction predicted by
flumazenil binding and diffusion-weighted
imaging signal intensity: a comparative pos-
itron emission tomography/magnetic reso-
nance imaging study in early ischemic stroke.
Stroke 2004;35(8):1892–1898.
19. Heiss WD, Sobesky J, Hesselmann V. Identi-
fying thresholds for penumbra and irrevers-
ible tissue damage. Stroke 2004;35(11 suppl
20. Latchaw RE, Alberts MJ, Lev MH, et al.
Recommendations for imaging of acute is-
identifying the penumbra and, in addi-
tion, are affected by nonstandardized
nomenclature, conceptual issues, and
measurement errors. In our opinion,
therefore, the use of perfusion imaging
in clinical practice is premature when
seen in the context of other imaging
paradigms currently available that do
not use this tool.
Perfusion imaging in acute ischemic
stroke is a promising technique; we
have a responsibility to design research
studies that try to increase the under-
standing of the role of this tool in acute
ischemic stroke treatment by enroll-
ing more patients in research studies
rather than jumping the gun and using
the tool as a clinical aid. We believe
that a tool such as perfusion imaging
that can potentially give rich informa-
tion about cerebral hemodynamics de-
serves this approach to realize its true
potential. Seen this way, we are propo-
nents of this technique.
Disclosures of Conflicts of Interest: M.G. Fi-
nancial activities related to the present article:
none to disclose. Financial activities not related
to the present article: received less than $10 000
as consultant to help design a stroke trial from
ev3, institution received $16 900 as primary in-
vestigator for study to study carotid plaque us-
ing MR imaging from Bayer, received less than
$5000 for speaking engagements from Penum-
bra and ev3, holds stock from Calgary Scientific
and NONo. Other relationships: none to dis-
close. B.K.M. No relevant conflicts of interest
to disclose. C.P.D. Financial activities related to
the present article: none to disclose. Financial
activities not related to the present article: re-
ceived consultancy fees from W. L. Gore and As-
sociates (medical device company), holds stock
in nFocus for aneurysm treatment device and
Pulse Therapeutics for acute stroke treatment
device. Other relationships: none to disclose.
1. Rha JH, Saver JL. The impact of recanali-
zation on ischemic stroke outcome: a meta-
analysis. Stroke 2007;38(3):967–973.
2. Zivin JA. Acute stroke therapy with tissue
plasminogen activator (tPA) since it was ap-
proved by the U.S. Food and Drug Admin-
istration (FDA). Ann Neurol 2009;66(1):
3. Kleindorfer D, Lindsell CJ, Brass L, Koro-
shetz W, Broderick JP. National US estimates
of recombinant tissue plasminogen activator
use: ICD-9 codes substantially underesti-
mate. Stroke 2008;39(3):924–928.
4. Astrup J, Siesjö BK, Symon L. Thresholds in
cerebral ischemia - the ischemic penumbra.
Stroke 1981;12(6):723–725.
Radiology: Volume 266: Number 1—January 2013
n 21
CONTROVERSIES: Perfusion Imaging in Acute Ischemic Stroke Is Promising Goyal et al
61. Adams HP Jr, del Zoppo G, Alberts MJ, et
al. Guidelines for the early management of
adults with ischemic stroke: a guideline from
the American Heart Association/American
Stroke Association Stroke Council, Clinical
Cardiology Council, Cardiovascular Radiol-
ogy and Intervention Council, and the Ath-
erosclerotic Peripheral Vascular Disease
and Quality of Care Outcomes in Research
Interdisciplinary Working Groups—the
American Academy of Neurology affirms the
value of this guideline as an educational tool
for neurologists. Circulation 2007;115(20):
62. Goyal M. Poor clinical outcome despite suc-
cessful arterial recanalization: what went
wrong? how can we do better? Neuroradiol-
ogy 2010;52(5):341–343.
63. Hacke W, Furlan AJ, Al-Rawi Y, et al. Intra-
venous desmoteplase in patients with acute
ischaemic stroke selected by MRI perfusion-
diffusion weighted imaging or perfusion CT
(DIAS-2): a prospective, randomised, dou-
ble-blind, placebo-controlled study. Lancet
Neurol 2009;8(2):141–150.
64. Davis SM, Donnan GA, Parsons MW, et al.
Effects of alteplase beyond 3 h after stroke
in the Echoplanar Imaging Thrombolytic
Evaluation Trial (EPITHET): a placebo-
controlled randomised trial. Lancet Neurol
65. Nagakane Y, Christensen S, Brekenfeld
C, et al. EPITHET: positive result af-
ter reanalysis using baseline diffusion-
weighted imaging/perfusion-weighted im-
aging co-registration. Stroke 2011;42(1):
66. Furlan A, Higashida R, Wechsler L, et al.
Intra-arterial prourokinase for acute is-
chemic stroke. The PROACT II study: a
randomized controlled trial. Prolyse in
Acute Cerebral Thromboembolism. JAMA
67. Hill MD, Rowley HA, Adler F, et al. Selection of
acute ischemic stroke patients for intra-arterial
thrombolysis with pro-urokinase by using AS-
PECTS. Stroke 2003;34(8):1925–1931.
68. Parsons MW, Christensen S, McElduff P, et al.
Pretreatment diffusion- and perfusion-MR le-
sion volumes have a crucial influence on clinical
response to stroke thrombolysis. J Cereb Blood
Flow Metab 2010;30(6):1214–1225.
69. Hom J, Dankbaar JW, Soares BP, et al.
Blood-brain barrier permeability assessed by
perfusion CT predicts symptomatic hemor-
rhagic transformation and malignant edema
in acute ischemic stroke. AJNR Am J Neuro-
radiol 2011;32(1):41–48.
70. Sylaja PN, Dzialowski I, Krol A, et al. Role
of CT angiography in thrombolysis decision-
making for patients with presumed sei-
zure at stroke onset. Stroke 2006;37(3):
71. Coutts SB, Modi J, Patel SK, et al. CT/CT
angiography and MRI findings predict re-
current stroke after transient ischemic at-
tack and minor stroke: results of the pro-
spective CATCH study. Stroke 2012;43(4):
46. Gentile NT, Cernetich J, Kanamalla US, et
al. Expedited computed tomography per-
fusion and angiography in acute ischemic
stroke: a feasibility study. J Emerg Med
47. Gupta R, Horev A, Wisco D, et al. Abstract
207: perfusion increases time to reperfusion
and may not enhance patient selection for
endovascular reperfusion therapies in acute
ischemic stroke. Stroke 2012;43:A207.
48. Straka M, Albers GW, Bammer R. Real-
time diffusion-perfusion mismatch analysis
in acute stroke. J Magn Reson Imaging
49. Goyal M, Almekhlafi MA. Dramatically re-
ducing imaging-to-recanalization time in acute
ischemic stroke: making choices. AJNR Am J
Neuroradiol 2012;33(7):1201–1203.
50. Demchuk AM, Hill MD, Barber PA, et al.
Importance of early ischemic computed to-
mography changes using ASPECTS in NINDS
rtPA stroke study. Stroke 2005;36(10):
51. von Kummer R, Bourquain H, Bastianello S,
et al. Early prediction of irreversible brain
damage after ischemic stroke at CT. Radiol-
ogy 2001;219(1):95–100.
52. Bivard A, McElduff P, Spratt N, Levi C, Par-
sons M. Defining the extent of irreversible
brain ischemia using perfusion computed
tomography. Cerebrovasc Dis 2011;31(3):
53. Coutts SB, Demchuk AM, Barber PA, et al.
Interobserver variation of ASPECTS in real
time. Stroke 2004;35(5):e103–e105.
54. Hill MD, Demchuk AM, Tomsick TA,
Palesch YY, Broderick JP. Using the base-
line CT scan to select acute stroke patients
for IV-IA therapy. AJNR Am J Neuroradiol
55. Modi J, Bai HD, Menon BK, Goyal M. En-
hancing acute ischemic stroke interpretation
with online aspects training. Can J Neurol Sci
56. Dzialowski I, Weber J, Doerfler A, Forst-
ing M, von Kummer R. Brain tissue water
uptake after middle cerebral artery oc-
clusion assessed with CT. J Neuroimaging
57. Dávalos A, Blanco M, Pedraza S, et al.
The clinical-DWI mismatch: a new di-
agnostic approach to the brain tissue at
risk of infarction. Neurology 2004;62(12):
58. Prosser J, Butcher K, Allport L, et al. Clin-
ical-diffusion mismatch predicts the puta-
tive penumbra with high specificity. Stroke
59. Ebinger M, Iwanaga T, Prosser JF, et al.
Clinical-diffusion mismatch and benefit from
thrombolysis 3 to 6 hours after acute stroke.
Stroke 2009;40(7):2572–2574.
60. Lansberg MG, Thijs VN, Hamilton S, et al.
Evaluation of the clinical-diffusion and perfu-
sion-diffusion mismatch models in DEFUSE.
Stroke 2007;38(6):1826–1830.
lay insensitive, and standardized postprocess-
ing algorithms? Radiology 2010;254(1):22–
34. Calamante F. Bolus dispersion issues re-
lated to the quantification of perfusion MRI
data. J Magn Reson Imaging 2005;22(6):
35. Chen JJ, Smith MR, Frayne R. The impact
of partial-volume effects in dynamic sus-
ceptibility contrast magnetic resonance
perfusion imaging. J Magn Reson Imaging
36. Konstas AA, Goldmakher GV, Lee TY, Lev
MH. Theoretic basis and technical imple-
mentations of CT perfusion in acute ische-
mic stroke: II. Technical implementations.
AJNR Am J Neuroradiol 2009;30(5):885–
37. Konstas AA, Goldmakher GV, Lee TY, Lev
MH. Theoretic basis and technical imple-
mentations of CT perfusion in acute ischemic
stroke: I. Theoretic basis. AJNR Am J Neuro-
radiol 2009;30(4):662–668.
38. Bristow MS, Simon JE, Brown RA, et al.
MR perfusion and diffusion in acute ische-
mic stroke: human gray and white matter
have different thresholds for infarction.
J Cereb Blood Flow Metab 2005;25(10):
39. Arakawa S, Wright PM, Koga M, et al. Ische-
mic thresholds for gray and white matter: a
diffusion and perfusion magnetic resonance
study. Stroke 2006;37(5):1211–1216.
40. Calamante F, Christensen S, Desmond PM,
Ostergaard L, Davis SM, Connelly A. The
physiological significance of the time-to-max-
imum (Tmax) parameter in perfusion MRI.
Stroke 2010;41(6):1169–1174.
41. Kudo K, Sasaki M, Yamada K, et al. Dif-
ferences in CT perfusion maps generated
by different commercial software: quan-
titative analysis by using identical source
data of acute stroke patients. Radiology
42. Kamalian S, Maas MB, et al. CT cerebral
blood flow maps optimally correlate with
admission diffusion-weighted imaging in
acute stroke but thresholds vary by postpro-
cessing platform. Stroke 2011;42(7):1923–
43. Galinovic I, Brunecker P, Ostwaldt AC, So-
emmer C, Hotter B, Fiebach JB. Fully au-
tomated postprocessing carries a risk of
substantial overestimation of perfusion
deficits in acute stroke magnetic reso-
nance imaging. Cerebrovasc Dis 2011;31(4):
44. Soares BP, Dankbaar JW, Bredno J, et al.
Automated versus manual post-processing
of perfusion-CT data in patients with acute
cerebral ischemia: influence on interobserver
variability. Neuroradiology 2009;51(7):
45. Salottolo KM, Fanale CV, Leonard KA, Frei
DF, Bar-Or D. Multimodal imaging does
not delay intravenous thrombolytic therapy
in acute stroke. AJNR Am J Neuroradiol
    • "However, standardized, reliable, and validated thresholds have not been definitively established[97,105]. Despite having reasonable sensitivity and high specificity for detecting infarcts, limited spatial resolution and motion artifact remain important limitations of CTP [106]. In addition, some conditions like proximal intracranial stenosis and extracranial carotid stenosis, atrial fibrillation and heart failure can simulate perfusion patterns seen in acute cerebral ischemia, contributing to false-positive evaluations[107]. "
    [Show abstract] [Hide abstract] ABSTRACT: Introduction: Stroke is the third leading cause of death and disability in Canada. In the hyperacute stroke setting, the treating physician must make a time critical decision on the treatment of each patient. Recent advances in imaging help the treating physician identify the subgroup of patients eligible for acute treatment of ischaemic stroke. Areas covered: In this review we will discuss Non-Contrast Computed Tomography (NCCT), CT-Angiography (CTA), and CT-Perfusion (CTP) in assessment of patients with acute ischaemic stroke and intracerebral haemorrhage. Intravenous tPA was the only proven therapy for acute ischaemic stroke presenting within 4.5 hours, until the five recent trials proved the efficacy of EVT for acute ischaemic stroke with proximal arterial occlusion. Imaging played a major role in patient selection in all five trials. Expert commentary: The challenge of rapid clinical assessment, review of imaging and timely treatment will continue to be made easier as the development and understanding of imaging progresses.
    Full-text · Article · Jun 2016
    • "These issues are receding with the fullbrain coverage of modern scanners and availability of standardized fully automated software. Multiphase CTA lacks the temporal resolution of CTP but may provide similar pathophysiologic information at a decision-making level [44,55,56]. To FIGURE 1. Futile endovascular thrombectomy of a patient with acute carotid T occlusion. "
    [Show abstract] [Hide abstract] ABSTRACT: Purpose of review: With the positive results of recent endovascular thrombectomy (EVT) trials, intravenous thrombolysis (IVT) and EVT provide physicians with two majorly effective acute treatment options for patients with acute ischemic stroke. IVT and EVT can be used as a single treatment or as a combined IVT/EVT treatment approach. This review summarizes how imaging findings can help in selecting stroke patients who are likely to benefit from these revascularization therapies. Recent findings: IVT applied within 4.5 h from symptom onset remains the mainstay of acute stroke therapy and was also applied to most patients in the randomized EVT trials. Recent studies have failed to demonstrate the effectiveness of IVT in later time windows. Vascular imaging is crucial to identify patients with a target intracranial occlusion prior to EVT. Patients with a small ischemic core, with good leptomeningeal collaterals or with evidence of penumbral tissue may particularly benefit from EVT. These imaging findings may also identify patients who benefit from EVT if applied more than 6 h from symptom onset. Summary: Pretherapeutic imaging findings help in identifying stroke patients who are likely to benefit from endovascular stroke therapies, and may identify patients who benefit from revascularization therapies in later time windows.
    Full-text · Article · Dec 2015
    • "Cerebral computed tomography perfusion (CTP) scans are acquired in patients with acute stroke [1] or subarachnoid hemorrhage [2]. Although there is some debate2345 about the prognostic value of CT perfusion, many studies [1, 2,6789 have shown that it provides valuable information about the cerebral hemodynamics, especially with the introduction of 320-slice CT scanners (16 cm coverage) that enable the acquisition of whole-brain CTP scans and provide an option to derive 4D dynamic CT angiography (dCTA) images from the CTP scans9101112 . These 4D dCTA images show great potential for the assessment of collaterals [13], the measurement of cerebral circulation times [14], and arteriovenous shunting lesion assessment [15]. "
    [Show abstract] [Hide abstract] ABSTRACT: Background and Purposes. The 320-detector row CT scanner enables visualization of whole-brain hemodynamic information (dynamic CT angiography (CTA) derived from CT perfusion scans). However, arterial image quality in dynamic CTA (dCTA) is inferior to arterial image quality in standard CTA. This study evaluates whether the arterial image quality can be improved by using a total bolus extraction (ToBE) method. Materials and Methods. DCTAs of 15 patients, who presented with signs of acute cerebral ischemia, were derived from 320-slice CT perfusion scans using both the standard subtraction method and the proposed ToBE method. Two neurointerventionalists blinded to the scan type scored the arterial image quality on a 5-point scale in the 4D dCTAs in consensus. Arteries were divided into four categories: (I) large extradural, (II) intradural (large, medium, and small), (III) communicating arteries, and (IV) cerebellar and ophthalmic arteries. Results. Quality of extradural and intradural arteries was significantly higher in the ToBE dCTAs than in the standard dCTAs (extradural , large intradural , medium intradural , and small intradural ). Conclusion. The 4D dCTAs derived with the total bolus extraction (ToBE) method provide hemodynamic information combined with improved arterial image quality as compared to standard 4D dCTAs.
    Full-text · Article · Oct 2014
Show more