Section 1 Etiology, pathophysiology, and imaging
Chapter
Neuropathology and pathophysiology
1 of stroke
Konstantin-A. Hossmann and Wolf-Dieter Heiss
The vascular origin of cerebrovascular
disease
All cerebrovascular diseases (CVDs) have their origin
in the vessels supplying or draining the brain. There-
fore, the knowledge of pathological changes occurring
in the vessels and in the blood are essential for under-
standing the pathophysiology of the various types of
CVD and for planning of efficient therapeutic strat-
egies. Changes in the vessel wall lead to obstruction of
blood flow, by interacting with blood constituents
they may cause thrombosis and blockade of blood
flow in this vessel. In addition to vascular stenosis or
occlusion at the site of vascular changes, disruption of
blood supply and consecutive infarcts can also be
produced by emboli arising from vascular lesions
situated proximally to otherwise healthy branches
located more distal in the arterial tree or from a
source located in the heart. At the site of occlusion,
opportunity exists for thrombus to develop in ante-
rograde fashion throughout the length of the vessel,
but this event seems to occur only rarely.
Changes in large arteries supplying the brain,
including the aorta, are mainly caused by atheroscler-
osis. Middle-sized and intracerebral arteries can also
be affected by acute or chronic vascular diseases of
inflammatory origin due to subacute to chronic infec-
tions, e.g. tuberculosis and lues, or due to collagen
disorders, e.g. giant cell arteritis, granulomatous
angiitis of the central nervous system, panarteritis
nodosa, and even more rarely systemic lupus erythe-
matosus, Takayasu’s arteritis, Wegener granulomato-
sis, rheumatoid arteritis, Sjögren’s syndrome, or
Sneddon and Behçet’s disease. In some diseases
affecting the vessels of the brain the etiology and
pathogenesis are still unclear, e.g. moyamoya disease
Textbook of Stroke Medicine, Second Edition, ed. Michael Brainin and Wolf-Dieter Heiss. Published by Cambridge University
Press. © Michael Brainin and Wolf-Dieter Heiss 2014.
Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 17 Apr 2018 at 16:27:57, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781107239340.002
and fibromuscular dysplasia, but these disorders are
characterized by typical locations of the vascular
changes. Some arteriopathies are hereditary, such as
CADASIL (cerebral autosomal dominant arteriopa-
thy with subcortical infarcts and leukoencephalopa-
thy), in some such as cerebral amyloid angiopathy a
degenerative cause has been suggested. All these vas-
cular disorders can cause obstruction, and lead to
thrombosis and embolizations. Small vessels of the
brain are affected by hyalinosis and fibrosis; this
“small-vessel disease” can cause lacunes and, if wide-
spread, is the substrate for vascular cognitive impair-
ment and vascular dementia.
Atherosclerosis is the most widespread disorder
leading to death and serious morbidity including
stroke [1]. The basic pathological lesion is the ather-
omatous plaque, and the most commonly affected
sites are the aorta, the coronary arteries, the carotid
artery at its bifurcation, and the basilar artery.
Arteriosclerosis, a more generic term describing
hardening and thickening of the arteries, includes as
additional types Mönkeberg’s sclerosis which is char-
acterized by calcification in the tunica media and
arteriolosclerosis with proliferative and hyaline
changes affecting the arterioles. Atherosclerosis starts
at young age, lesions accumulate and grow through-
out life and become symptomatic and clinically evi-
dent when end organs are affected [2].
Atherosclerosis: atheromatous plaques, most commonly
in the aorta, the coronary arteries, the bifurcation of the
carotid artery and the basilar artery.
The initial lesion of atherosclerosis has been attrib-
uted to “fatty streaks” and the “intimal cell mass.”
Those changes already occur in childhood and ado-
lescence and do not necessarily correspond to the
future sites of atherosclerotic plaques. Fatty streaks
1
 Section 1: Etiology, pathophysiology, and imaging

Figure 1.1. The stages of development of an atherosclerotic plaque. (1) LDL moves into the subendothelium and (2) is oxidized by
macrophages and smooth muscle cells (SMC). (3) Release of growth factors and cytokines (4) attracts additional monocytes. (5) Macrophages
and (6) foam cell accumulation and additional (7) SMC proliferation result in (8) growth of the plaque. (9) Fibrous cap degradation and plaque
rupture (collagenases, elastases). (10) Thrombus formation. IL-1 ¼ interlenkin-1; MCP-1 ¼ Monocyte chemotactic protein-1. (Modified with
permission from Faxon et al. [5].)

are focal areas of intra cellular lipid collection in both
macrophages and smooth muscle cells. Various con-
cepts have been proposed to explain the progression
of such precursor lesions to definite atherosclerosis
[2, 3], most remarkable of which is the response-to-
injury hypothesis postulating a cellular and molecular
response to various atherogenic stimuli in the form of
an inflammatory repair process [4]. This inflamma-
tion develops concurrently with the accumulation of
minimally oxidized low-density lipoproteins (LDLs)
[5, 6], and stimulates vascular smooth muscle cells
(VSMCs), endothelial cells, and macrophages [7], and
as a result foam cells aggregate with an accumulation
of oxidized LDL. In the further stages of arthero-
sclerotic plaque development VSMCs migrate, prolif-
erate, and synthesize extracellular matrix components
on the luminal side of the vessel wall, forming the
fibrous cap of the atherosclerotic lesion [8]. In this
complex process of growth, progression, and finally
rupture of an atherosclerotic plaque a large number of
matrix modulators, inflammatory mediators, growth
factors, and vasoactive substances are involved. The
complex interactions of these many factors are dis-
cussed in the special literature [5–9].
The fibrous cap of the atherosclerotic lesion
covers the deep lipid core with a massive accumula-
tion of extracellular lipids (atheromatous plaque), or
fibroblasts and extracellular calcifications may con-
tribute to a fibrocalcific lesion. Mediators from inflam-
2 matory cells at the thinnest portion of the cap surface
of a vulnerable plaque – which is characterized by a
larger lipid core and a thin fibrous cap – can lead to
plaque disruption with formation of a thrombus or
hematoma or even to total occlusion of the vessel.
During the development of atherosclerosis the entire
vessel can enlarge or constrict in size [10]. However,
once the plaque covers >40% of the vessel wall, the
artery no longer enlarges, and the lumen narrows as
the plaque grows. In vulnerable plaques thrombosis
forming on the disrupted lesion further narrows
the vessel lumen and can lead to occlusion or be
the origin of emboli. Less commonly, plaques
have reduced collagen and elastin with a thin and
weakened arterial wall, resulting in aneurysm forma-
tion which when ruptured may be the source of
intracerebral hemorrhage (Figure 1.1).
Injury hypothesis of progression to atherosclerosis: fatty
streaks (focal areas of intra cellular lipid collection) !
inflammatory repair process with stimulation of vascular
smooth muscle cells ! atheromatous plaque.
Thromboembolism: immediately after plaque rupture
or erosion, subendothelial collagen, the lipid core, and
procoagulants such as tissue factor and von Wille-
brand factor are exposed to circulating blood. Platelets
rapidly adhere to the vessel wall through the platelet
glycoproteins (GP) Ia/IIa and GP Ib/IX [11] with
subsequent aggregation to this initial monolayer
through linkage with fibrinogen and the exposed GP
IIb/IIIa on activated platelets. As platelets are a source
of nitric oxide (NO), the resulting deficiency of bioac-
tive NO, which is an effective vasodilator, contributes

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 17 Apr 2018 at 16:27:57, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781107239340.002
 Chapter 1: Neuropathology and pathophysiology of stroke
to the progression of thrombosis by augmenting
platelet activation, enhancing VSMC proliferation
and migration, and participating in neovasculariza-
tion [12]. The activated platelets also release adeno-
sine diphosphate (ADP) and thromboxane A2 with
subsequent activation of the clotting cascade. The
growing thrombus obstructs or even blocks the
blood flow in the vessel. Atherosclerotic thrombi
are also the source for embolisms, which are the
primary pathophysiological mechanism of ischemic
strokes, especially from carotid artery disease or of
cardiac origin.
Rupture or erosion of atheromatous plaques ! adhesion
of platelets ! thrombus ! obstruction of blood flow
and source of emboli.
Small-vessel disease usually affects the arterioles and is
associated with hypertension. It is caused by suben-
dothelial accumulation of a pathological protein, the
hyaline, formed from mucopolysaccharides and
matrix proteins, which leads to narrowing of the
lumen or even occlusion of these small vessels. Often
it is associated with fibrosis, which affects not only
arterioles, but also other small vessels and capillaries
and venules. Lipohyalinosis also weakens the vessel
wall, predisposing for the formation of “miliary
aneurysms.” Small-vessel disease results in two patho-
logical conditions: status lacunaris (lacunar state) and
status cribrosus (état criblé). Status lacunaris is char-
acterized by small irregularly shaped infarcts due to
occlusion of small vessels; it is the pathological sub-
strate of lacunar strokes and vascular cognitive
impairment and dementia. In status cribrosus small
round cavities develop around affected arteries due
to disturbed supply of oxygen and metabolic sub-
strate. These “criblures” together with miliary aneur-
ysms are the sites of vessel rupture causing typical
hypertonic intracerebral hemorrhages [13–16].
A second type of small-vessel disease is characterized
by the progressive accumulation of congophilic, βA4
immunoreactive, amyloid protein in the walls of
small- to medium-sized arteries and arterioles. Cere-
bral amyloid angiopathy is a pathological hallmark
of Alzheimer’s disease and also occurs in rare genet-
ically transmitted diseases, e.g. CADASIL and Fabry’s
disease [17]. For a more detailed discussion of the
etiology and pathophysiology of the various specific
vascular disorders see [18–20].
Small-vessel disease: subendothelial accumulation of
hyaline in arterioles.
Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 17 Apr 2018 at 16:27:57, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781107239340.002
Types of acute cerebrovascular diseases
Numbers relating to the frequency of the different
types of acute CVD are highly variable depending
on the source of data. The most reliable numbers
come from the in-hospital assessment of stroke in
the Framingham study determining the frequency
of completed stroke: 60% were caused by athero-
thrombotic brain infarction, 25.1% by cerebral
embolism, 5.4% by subarachnoid hemorrhage,
8.3% by intracerebral hemorrhage, and 1.2% by
undefined diseases. In addition, transient ischemic
attacks (TIAs) accounted for 14.8% of the total
cerebrovascular events [21].
Ischemic strokes result from a critical reduction of
regional cerebral blood flow (rCBF) lasting beyond a
critical duration, and are caused by atherothrombotic
changes of the arteries supplying the brain or by
emboli from sources in the heart, the aorta, or the
large arteries. The pathological substrate of ischemic
stroke is ischemic infarction of brain tissue, the loca-
tion, extension, and shape of which depend on the
size of the occluded vessel, the mechanism of arterial
obstruction, and the compensatory capacity of the
vascular bed. Occlusion of arteries supplying defined
brain territories by atherothrombosis or embolization
lead to territorial infarcts of variable size: they may be
large – e.g. the whole territory supplied by the middle
cerebral artery (MCA) – or small, if branches of large
arteries are occluded or if compensatory collateral
perfusion – e.g. via the circle of Willis or leptomenin-
geal anastomoses – is efficient in reducing the area of
critically reduced flow (Figure 1.2) [14, 16]. In a
smaller number of cases infarcts can also develop at
the borderzones between vascular territories, when
several large arteries are stenotic and the perfusion
in these “last meadows” cannot be constantly main-
tained above the critical threshold of morphological
integrity [22]. Borderzone infarctions are a subtype of
the low-flow or hemodynamically induced infarctions
which are the result of critically reduced cerebral
perfusion pressure in far-downstream brain arteries.
The more common low-flow infarctions affect sub-
cortical structures within a vascular bed with pre-
served but marginal irrigation [23]. Lacunar infarcts
reflect disease of the vessels penetrating the brain to
supply the capsule, the basal ganglia, thalamus, and
paramedian regions of the brainstem [24]. Most often
they are caused by lipohyalinosis of deep arteries 3
(small-vessel disease); less frequent causes are stenosis
 Section 1: Etiology, pathophysiology, and imaging

Figure 1.2. Various types and sizes of infarcts due to different hemodynamic patterns
a. Total territorial infarct due to defective collateral supply
b. Core infarct, meningeal anastomosis supply peripheral zones
c. Territorial infarct in center of supply area, due to branch occlusion
d. Borderzone infarction in watershed areas due to stenotic lesions in arteries supplying neighboring areas
e. Lacunar infarctions due to small-vessel disease. (Modified with permission from Zülch [14].)

of the MCA stem and microembolization to penetrant
arterial territories. Pathologically these lacunes are
defined as small cystic trabeculated scars about
5 mm in diameter, but they are more often observed
on magnetic resonance images, where they are
accepted as lacunes up to 1.5 cm diameter. The classic
lacunar syndromes include pure motor, pure sensory,
and sensorimotor syndromes, sometimes ataxic hemi-
paresis, clumsy hand, dysarthria, and hemichorea/
hemiballism, but higher cerebral functions are not
involved. A new classification of stroke subtypes is
mainly oriented on the most likely cause of stroke:
atherosclerosis, small-vessel disease, cardiac source,
or other cause [25].
Territorial infarcts are caused by an occlusion of arteries
supplying defined brain territories by atherothrombosis
or embolizations.
Borderzone infarcts develop at the borderzone between
vascular territories and are the result of a critically
reduced cerebral perfusion pressure (low-flow
infarctions).
Lacunar infarcts are mainly caused by small-vessel
disease.
4 Hemorrhagic infarctions, i.e. “red infarcts” in contrast
to the usual “pale infarcts,” are defined as ischemic
infarcts in which varying numbers of blood cells are
found within the necrotic tissue. The size can range
from a few petechial bleeds in the gray matter of
cortex and basal ganglia to large hemorrhages involv-
ing the cortical and deep hemispheric regions. Hem-
orrhagic transformation frequently appears during
the second and third phase of infarct evolution, when
macrophages appear and new blood vessels are
formed in tissue consisting of neuronal ghosts and
proliferating astrocytes. However, the only significant
difference between “pale” and “red infarcts” is the
intensity and extension of the hemorrhagic compon-
ent, since in at least two-thirds of all infarcts petechial
hemorrhages are microscopically detectable. Macro-
scopically, red infarcts contain multifocal bleedings
which are more or less confluent and predominate
in cerebral cortex and basal ganglia, which are richer
in capillaries than the white matter[26]. If the hemor-
rhages become confluent intrainfarct hematomas
might develop, and extensive edema may contribute
to mass effects and lead to malignant infarction. The
frequency of hemorrhagic infarctions in anatomical
studies ranged from 18% to 42% [27], with a high
incidence (up to 85% of hemorrhagic infarcts) in
cardioembolic stroke [28].

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 17 Apr 2018 at 16:27:57, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781107239340.002
 Chapter 1: Neuropathology and pathophysiology of stroke
Mechanisms for hemorrhagic transformation are
manifold and vary with regard to the intensity of
bleeding. Petechial bleeding results from diapedesis
rather than vascular rupture. In severe ischemic tissue
vascular permeability is increased and endothelial
tight junctions are ruptured. When blood circulation
is spontaneously or therapeutically restored, blood
can leak out of these damaged vessels. This can also
happen with fragmentation and distal migration of
an embolus (usually of cardiac origin) in the damaged
vascular bed, explaining delayed clinical worsening
in some cases. For the hemorrhagic transformation
the collateral circulation might also have an impact:
in some instances reperfusion via pial networks
may develop with the diminution of peri-ischemic
edema at borderzones of cortical infarcts. Risk
of hemorrhage is significantly increased in large
infarcts, with mass effect supporting the importance
of edema for tissue damage and the deleterious
effect of late reperfusion when edema resolves. In
some instances also the rupture of the vascular wall
secondary to ischemia-induced endothelial necrosis
might cause an intra-infarct hematoma. Vascular
rupture can explain very early hemorrhagic infarcts
and early intrainfarct hematoma (between 6 and
18 hours after stroke), whereas hemorrhagic trans-
formation usually develops within 48 hours to 2
weeks.
Hemorrhagic infarctions are defined as ischemic infarcts
in which varying amounts of blood cells are found
within the necrotic tissue. They are caused by leakage
from damaged vessels, due to increased vascular
permeability in ischemic tissue or vascular rupture
secondary to ischemia
Intracerebral hemorrhage (ICH) occurs as a result of
bleeding from an arterial source directly into the
brain parenchyma and accounts for 5–15% of all
strokes [29, 30]. Hypertension is the leading risk
factor, but in addition advanced age, race and also
cigarette smoking, alcohol consumption, and high
serum cholesterol levels have been identified. In a
number of instances ICH occurs in the absence of
hypertension, usually in atypical locations. The causes
include small vascular malformations, vasculitis,
brain tumors, and sympathomimetic drugs (e.g.
cocaine). ICH may also be caused by cerebral amyloid
angiopathy and rarely damage is elicited by acute
changes in blood pressure, e.g. due to exposure to
cold. The occurrence of ICH is also influenced
by the increasing use of antithrombotic and
Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 17 Apr 2018 at 16:27:57, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781107239340.002
thrombolytic treatment of ischemic diseases of the
brain, heart, and other organs [31, 32].
Spontaneous ICH occurs predominantly in the
deep portions of the cerebral hemispheres (“typical
ICH”) [33]. Its most common location is the putamen
(35–50% of cases). The subcortical white matter is
the second most frequent location (approx. 30%).
Hemorrhages in the thalamus are found in 10–15%,
in the pons in 5–12%, and in the cerebellum in 7%
[34]. Most ICHs originate from the rupture of small,
deep arteries with diameters of 50 to 200 μm which
are affected by lipohyalinosis due to chronic hyper-
tension. These small-vessel changes lead to weakening
of the vessel wall and miliary microaneurysm and
consecutive small local bleedings, which might be
followed by secondary ruptures of the enlarging
hematoma in a cascade or avalanche fashion [35].
After active bleeding has started it can continue
for a number of hours with enlargement of hema-
toma that is frequently associated with clinical
deterioration [36].
Putaminal hemorrhages originate from a lateral
branch of the striate arteries at the posterior angle,
resulting in an ovoid mass pushing the insular cortex
laterally and displacing or involving the internal
capsule. From this initial putaminal-claustral loca-
tion a large hematoma may extend to the internal
capsule and lateral ventricle, into the corona radiata,
and into the temporal white matter. Putaminal
ICHs are considered the typical hypertensive
hemorrhages.
Caudate hemorrhage, a less common form of
bleeding from distal branches of lateral striate arter-
ies, occurs in the head of the caudate nucleus. This
bleeding soon connects to the ventricle and usually
involves the anterior limb of the internal capsule.
Thalamic hemorrhages can involve most of this
nucleus and extend into the third ventricle medially
and the posterior limb of the internal capsule laterally.
The hematoma may press on or even extend into the
midbrain. Larger hematomas often reach the corona
radiata and the parietal white matter.
Lobar (white matter) hemorrhages originate at the
cortico-subcortical junction between gray and white
matter and usually spread along the fiber bundles in
the parietal and occipital lobes. The hematomas are
close to the cortical surface and usually not in direct
contact with deep hemisphere structures or the ven-
tricular system. As atypical ICHs they are not neces- 5
sarily correlated with hypertension.
 Section 1: Etiology, pathophysiology, and imaging
Cerebellar hemorrhages usually originate in the
area of the dentate nucleus from rupture of distal
branches of the superior cerebellar artery and extend
into the hemispheric white matter and into the fourth
ventricle. The pontine tegmentum is often com-
pressed. A variant, the midline hematoma, originates
from the cerebellar vermis, always communicates
with the fourth ventricle, and frequently extends
bilaterally into the pontine tegmentum.
Pontine hemorrhages from bleeding of small para-
median basilar perforating branches cause medially
placed hematomas involving the basis of the pons.
A unilateral variety results from rupture of distal
long circumferential branches of the basilar artery.
These hematomas usually communicate with the
fourth ventricle, and extend laterally and ventrally
into the pons.
The frequency of recurrent ICHs in hypertensive
patients is rather low (6%) [37]. Recurrence rate is
higher with poor control of hypertension and also in
hemorrhages due to other causes. In some instances
multiple simultaneous ICHs may occur, but also in
these cases the cause is other than hypertension.
In ICHs, the local accumulation of blood des-
troys the parenchyma, displaces nervous structures,
and dissects the tissue. At the bleeding sites fibrin
globes are formed around accumulated platelets.
After hours or days extracellular edema develops at
the periphery of the hematoma. After 4 to 10 days
the red blood cells begin to lyse, granulocytes and
thereafter microglial cells arrive, and foamy macro-
phages are formed, which ingest debris and hemosi-
derin. Finally, the astrocytes at the periphery of the
hematoma proliferate and turn into gemistocytes
with eosinophylic cytoplasma. When the hematoma
is removed, the astrocytes are replaced by glial
fibrils. After that period – extending to months –
the residue of the hematoma is a flat cavity with a
reddish lining resulting from hemosiderin-laden
macrophages [34].
Intracerebral hemorrhage (ICH) occurs as a result of
bleeding from an arterial source directly into the brain
parenchyma, predominantly in the deep portions of
the cerebral hemispheres (typical ICH). Hypertension
is the leading risk factor, and the most common location
is the putamen.
Cerebral venous thrombosis (CVT) can develop from
many causes and due to predisposing conditions.
6 CVT is often multifactorial, when various risk factors
Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 17 Apr 2018 at 16:27:57, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781107239340.002
and causes contribute to the development of this dis-
order [38]. The incidence of septic CVT has been
reduced to less than 10% of cases, but septic cavernous
sinus thrombosis is still a severe, however, rare prob-
lem. Aseptic CVT occurs during puerperium and less
frequently during pregnancy, but may also be related
to use of oral contraceptives. Among the non-
infectious causes of CVT congenital thrombophilia,
particularly prothrombin and factor V Leiden gene
mutations, as well as antithrombin, protein C, and
protein S deficiencies must be considered. Other con-
ditions with risk for CVT are malignancies, inflamma-
tory diseases, and systemic lupus erythematosus.
However, in 20–35% of CVT the etiology remains
unknown. The fresh venous thrombus is rich in red
blood cells and fibrin and poor in platelets. Later on,
it is replaced by fibrous tissue, occasionally with reca-
nalization. The most common location of CVT is the
superior sagittal sinus and the tributary veins.
Whereas some thromboses, particularly of the
lateral sinus, may have no pathological consequences
for the brain tissue, occlusion of large cerebral veins
usually leads to a venous infarct. These infarcts are
located in the cortex and adjacent white matter and
often are hemorrhagic. Thrombosis of the superior
sagittal sinus may lead only to brain edema, but
usually causes bilateral hemorrhagic infarcts in both
hemispheres. These venous infarcts are different from
arterial infarcts: cytotoxic edema is absent or mild,
vasogenic edema is prominent, and hemorrhagic
transformation or bleeding is usual. Despite this
hemorrhagic component heparin is the treatment of
choice.
Cerebral venous thrombosis can lead to a venous infarct.
Venous infarcts are different from arterial infarcts:
cytotoxic edema is absent or mild, vasogenic edema is
prominent, and hemorrhagic transformation or bleeding
is usual.
Cellular pathology of ischemic stroke
Acute occlusion of a major brain artery causes a
stereotyped sequel of cellular alterations which evolve
over a protracted period of time and which depend on
the topography, severity, and duration of ischemia
[39]. The most sensitive brain cells are neurons,
followed – in this order – by oligodendrocytes, astro-
cytes, and vascular cells. The most vulnerable brain
regions are hippocampal subfield CA1, neocortical
 Chapter 1: Neuropathology and pathophysiology of stroke
layers 3, 5, and 6, the outer segment of striate nucleus,
and the Purkinje and basket cell layers of cerebellar
cortex. If blood flow decreases below the threshold of
energy metabolism, the primary pathology is necrosis
of all cell elements, resulting in ischemic brain infarct.
If ischemia is not severe enough to cause primary
energy failure, or if it is of so short duration that
energy metabolism recovers after reperfusion, a
delayed type of cell death may evolve which exhibits
the morphological characteristics of necrosis, apop-
tosis, or a combination of both. In the following,
primary and delayed cell death will be described
separately.
Primary ischemic cell death
In the core of the territory of an occluded brain
artery the earliest sign of cellular injury is neuronal
Light-microscopical characteristics of rat infarction
Acute ischemic changes
Control swelling
sham surgery 4 hours
Necrotic changes
red neuron ghost neuron Dark neuron artifact
1 day 3 days
Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 17 Apr 2018 at 16:27:57, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781107239340.002
swelling or shrinkage, the cytoplasm exhibiting
microvacuolation (MV) which ultrastructurally has
been associated with mitochondrial swelling [40].
These changes are potentially reversible if blood flow
is restored before mitochondrial membranes begin
to rupture. One to two hours after the onset of ische-
mia, neurons undergo irreversible necrotic alterations
(red neuron or ischemic cell change). In conven-
tional hematoxylin-eosin-stained brain sections such
neurons are characterized by intensively stained eosi-
nophilic cytoplasma, formation of triangular nuclear
pyknosis, and direct contact with swollen astrocytes
(Figure 1.3). Electron-microscopically mitochondria
exhibit flocculent densities which represent denatu-
rated mitochondrial proteins. Ischemic cell change
must be distinguished from artifactual dark neurons
which stain with all (acid or basic) dyes and are not
surrounded by swollen astrocytes [41].
Figure 1.3. Light-microscopical
evolution of neuronal changes after
experimental middle cerebral occlusion.
shrinkage (Modified with permission from Garcia
et al. [126].)
2 hours
sham surgery 7
 Section 1: Etiology, pathophysiology, and imaging
With ongoing ischemia, neurons gradually loose
their stainability with hematoxylin, they become
mildly eosinophilic, and, after 2–4 days, transform
to ghost cells with hardly detectable pale outline.
Interestingly, neurons with ischemic cell change are
mainly located in the periphery and ghost cells in the
center of the ischemic territory, which suggests that
manifestation of ischemic cell change requires some
residual or restored blood flow whereas ghost cells
may evolve in the absence of flow [39].
Primary ischemic cell death induced by focal
ischemia is associated with reactive and secondary
changes. The most prominent alteration during the
initial 1–2 hours is perivascular and perineuronal
astrocytic swelling, after 4–6 hours the blood–brain
barrier breaks down, resulting in the formation of
vasogenic edema, after 1–2 days inflammatory cells
accumulate throughout the ischemic infarct, and
within 1.5–3 months cystic transformation of the
necrotic tissue occurs together with the development
of a peri-infarct astroglial scar (Figure 1.4)
Delayed neuronal death
The prototype of delayed cell death is the slowly
8 progressing injury of pyramidal neurons in the CA1
sector of the hippocampus after a brief episode of
Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 17 Apr 2018 at 16:27:57, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781107239340.002
Figure 1.4. Transformation of acute
ischemic alterations into cystic infarct.
Note pronounced inflammatory reaction
prior to tissue cavitation. PMN ¼
polymorphonuclear leukocyte. (Modified
with permission from Petito [39].)
global ischemia [42]. In focal ischemia delayed neur-
onal death may occur in the periphery of cortical
infarcts or in regions which have been reperfused
before ischemic energy failure becomes irreversible.
Cell death is also observed in distant brain regions,
notably in the substantia nigra and thalamus.
The morphological appearance of neurons during
the interval between ischemia and the manifestation
of delayed cell death exhibits a continuum that
ranges from necrosis to apoptosis with all possible
combinations of cytoplasmic and nuclear morph-
ology that are characteristic for the two types of cell
death [43]. In its pure form, necrosis combines
karyorrhexis with massive swelling of endoplasmic
reticulum and mitochondria, whereas in apoptosis
mitochondria remain intact and nuclear fragmenta-
tion with condensation of nuclear chromatin gives
way to the development of apoptotic bodies.
A frequently used histochemical method for the
visualization of apoptosis is terminal deoxyribonu-
cleotidyl transferase (TdT)-mediated dUTP-biotin
nick-end labeling (TUNEL assay), which detects
DNA strand breaks. However, as this method may
also stain necrotic neurons, a clear differentiation is
not possible [44].
A consistent ultrastructural finding in neurons
undergoing delayed cell death is disaggregation of
 Chapter 1: Neuropathology and pathophysiology of stroke
ribosomes, which reflects the inhibition of protein
synthesis at the initiation step of translation [45].
Light-microscopically, this change is equivalent to
tigrolysis, visible in Nissl-stained material. Disturb-
ances of protein synthesis and the associated endo-
plasmic reticulum (ER) stress are also responsible for
cytosolic protein aggregation and the formation of
stress granules [46]. In the hippocampus, stacks
of accumulated ER may become visible but in other
areas this is not a prominent finding.
Pathology of the neurovascular unit
The classical pathology of ischemic injury differenti-
ates between the sensitivity of the various cell types
of brain parenchyma with the neurons as the most
vulnerable elements. The molecular analysis of injury
evolution, however, suggests that ischemia initiates a
coordinated multi-compartmental response of brain
cells and vessels, also referred to as the neurovascular
unit [47]. This unit includes microvessels (endothelial
cells, basal lamina matrix, astrocytic endfeet, peri-
cytes, and circulating blood elements), the cell body
and main processes of astrocytes, the nearby neurons
together with their axons, and supporting cells,
notably microglia and oligodendrocytes. It provides
the framework for the bi-directional communication
between neuron and supplying microvessel. Under
physiological conditions, the most prominent func-
tion is the neurovascular coupling for maintaining
adequate supply of brain nutrients and clearance of
waste products. Pathophysiological disturbances of
microcirculation, conversely, provoke coincidental
microvessel–neuron responses, possibly mediated
by alterations in the matrix of the vascular and non-
vascular compartments of the ischemic territory.
Severe ischemia induces primary cell death due to
necrosis of all cell elements. Not so severe or short-term
ischemia induces delayed cell death with necrosis,
apoptosis, or a combination of both. The neurovascular
unit provides the conceptual framework for the
propagation of injury from microvessels to neurons.
Animal models of stroke
According to the Framingham study, 65% of strokes
that result from vascular occlusion present lesions in
the territory of the MCA, 2% in the anterior and 9%
in the posterior cerebral artery territories; the rest is
located in brainstem, cerebellum, in watershed, or
Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 17 Apr 2018 at 16:27:57, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781107239340.002
multiple regions. In experimental stroke research, this
situation is reflected by the preferential use of MCA
occlusion models.
Transorbital middle cerebral artery occlusion: this
model was introduced in the 1970s for the production
of stroke in monkeys [48] and later modified for use
in cats, dogs, rabbits, and even rats. The procedure is
technically demanding and requires microsurgical
skills. The advantage of this approach is the possibility
to expose the MCA at its origin from the internal
carotid artery without retracting parts of the brain.
Vascular occlusion can thus be performed without the
risk of brain trauma. On the other hand, removal
of the eyeball is invasive and may evoke functional
disturbances which should not be ignored. Surgery
may also cause generalized vasospasm which may
interfere with the collateral circulation and, hence,
induce variations in infarct size. The procedure there-
fore requires extensive training before reproducible
results can be expected.
The occlusion of the MCA at its origin interrupts
blood flow to the total vascular territory, including
the basal ganglia which are supplied by the lenticulo-
striate arteries. These MCA branches are end-arteries
which in contrast to the cortical branches do not form
collaterals with the adjacent vascular territories. As a
consequence, the basal ganglia are consistently part of
the infarct core whereas the cerebral cortex exhibits
a gradient of blood flow which decreases from the
peripheral towards the central parts of the vascular
territory. Depending on the steepness of this gradient,
a cortical core region with the lowest flow values
in the lower temporal cortex is surrounded by a
variably sized penumbra which may extend up to
the parasagittal cortex.
Transcranial occlusion of the middle cerebral
artery: post- or retro-orbital transcranial approaches
for MCA occlusion are mainly used in rats and mice
because in these species the main stem of the artery
appears on the cortical surface rather close to its
origin from the internal carotid artery [49]. In con-
trast to transorbital MCA occlusion, transcranial
models do not produce ischemic injury in the basal
ganglia because the lenticulostriate branches origin-
ate proximal to the occlusion site. Infarcts, therefore,
are mainly located in the temporo-parietal cortex with
a gradient of declining flow values from the periph-
eral to the central parts of the vascular territory.
Filament occlusion of the middle cerebral artery: 9
the presently most widely used procedure for MCA
 Section 1: Etiology, pathophysiology, and imaging
occlusion in rats and mice is the intraluminal filament
occlusion technique, first described by Koizumi et al.
[50]. A nylon suture with an acryl-thickened tip is
inserted into the common carotid artery and ortho-
gradely advanced, until the tip is located at the origin
of the MCA. Modifications of the original technique
include different thread types for isolated or com-
bined vascular occlusion, adjustments of the tip size
to the weight of the animal, poly-L-lysine coating of
the tip to prevent incomplete MCA occlusion, or the
use of guide-sheaths to allow remote manipulation of
the thread for occlusion during polygraphic record-
ings or magnetic resonance imaging.
The placement of the suture at the origin of the
MCA obstructs blood supply to the total MCA sup-
plied territory, including the basal ganglia. It may also
reduce blood flow in the anterior and posterior cere-
bral arteries, particularly when the common carotid
artery is ligated to facilitate the insertion of the thread.
As this minimizes collateral blood supply from these
territories, infarcts are very large and produce massive
ischemic brain edema with a high mortality when
experiments last for more than a few hours. For this
reason, threads are frequently withdrawn 1–2 hours
following insertion. The resulting reperfusion sal-
vages the peripheral parts of the MCA territory, and
infarcts become smaller [51]. However, the pathophy-
siology of transient MCA occlusion differs basically
from that of the clinically more relevant permanent
occlusion models, and neither the mechanisms of
infarct evolution nor the pharmacological responsive-
ness of the resulting lesions replicate that of clinical
stroke [52].
Transient filament occlusion is also an inappro-
priate model for the investigation of spontaneous or
thrombolysis-induced reperfusion. Withdrawal of the
intraluminal thread induces instantaneous reperfu-
sion whereas spontaneous or thrombolysis-induced
recanalization results in slowly progressing recir-
culation. As post-ischemic recovery is greatly influ-
enced by the dynamics of reperfusion, outcome and
pharmacological responsiveness of transient filament
occlusion is distinct from most clinical situations of
reversible ischemia, where the onset of reperfusion is
much less abrupt.
Clot embolism of middle cerebral artery: MCA
embolism with autologous blood clots is a clinically
highly relevant but also inherently variable stroke
10 model which requires careful preparation and place-
ment of standardized clots to induce reproducible
Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 17 Apr 2018 at 16:27:57, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781107239340.002
brain infarcts [53]. The most reliable procedure for
clot preparation is thrombin-induced clotting of auto-
logous blood within calibrated tubings, which results
in cylindrical clots that can be dissected in segments
of equal length. Selection of either fibrin-rich (white)
or fibrin-poor (red) segments influences the speed
of spontaneous reperfusion and results in different
outcome. Clots can also be produced in situ by
microinjection of thrombin [54] or photochemically
by ultraviolet illumination of the MCA following
injection of rose Bengal [55].
The main application of clot embolism is for the
investigation of experimental thrombolysis. The drug
most widely used is human recombinant tissue plas-
minogen activator (rtPA) but the dose required in
animals is much higher than in humans, which must
be remembered when possible side-effects such as
rtPA toxicity are investigated. The hemodynamic
effect, in contrast, is similar despite the higher dose
and adequately reproduces the slowly progressing
recanalization observed under clinical conditions.
A recent development of clinical stroke treatment
and possibly the central challenge for future animal
research is interventional thrombectomy [56]. The
animal most widely used for this research is the swine
but as in this species the carotid access to the anterior
cerebral vasculature is impeded by a rete mirabile, clot
embolism and retrieval is carried out via the internal
maxillary or lingual artery [57]. Angiographic studies
confirm that clot retrieval using either aspiration or
removable stent devices results in immediate recana-
lization but a detailed pathophysiological analysis of
post-ischemic reperfusion is not yet available. It is,
therefore, premature to speculate whether this treat-
ment and its effect on post-ischemic recovery can also
be replicated in smaller animals by technically simpler
mechanical occlusion models, such as transient fila-
ment occlusion.
Various procedures for artery occlusion models, mostly
middle cerebral artery occlusion models, were developed
to study focal ischemia in animals.
Hemodynamics of stroke
Normal regulation of blood flow
In the intact brain, CBF is tightly coupled to the
metabolic requirements of tissue (metabolic regula-
tion) but the flow rate remains essentially constant
over a wide range of blood pressures (autoregulation).
 Chapter 1: Neuropathology and pathophysiology of stroke
An important requirement for metabolic regula-
tion is the responsiveness of blood vessels to carbon
dioxide (CO2 reactivity), which can be tested by the
application of carbonic anhydrase inhibitors or CO2
ventilation. Under physiological conditions, blood
flow doubles when CO2 rises by about 30 mmHg,
and is reduced by one-third when CO2 declines by
15 mmHg. The vascular response to CO2 depends
mainly on the changes of extracellular pH but it is
also modulated by other factors such as prostanoids,
NO, and neurogenic influences.
Autoregulation of CBF is the remarkable capacity
of the vascular system to adjust its resistance in such a
way that blood flow is kept constant over a wide range
of cerebral perfusion pressures (80–150 mmHg). The
range of autoregulation is shifted to the right, i.e. to
higher values, in patients with hypertension and to
the left during hypercarbia.
The mechanism of autoregulation is complex [58].
The dominating factor is a pressure-sensitive direct
myogenic response initiated by the activation of
stretch-sensitive cation channels of vascular smooth
muscle. In addition, a flow-sensitive indirect smooth
muscle response is initiated by changes in the shear
stress of endothelial cells, which result in activation of
various signal transduction pathways. Other influ-
ences are mediated by metabolic and neurogenic
factors but these may be secondary effects and are of
lesser significance.
Metabolic regulation: cerebral blood flow is coupled
to metabolic requirements of tissue by a vascular
response to changes in CO2. Autoregulation: cerebral
blood flow is kept constant over a wide range
of cerebral perfusion pressures.
Disturbances of flow regulation
Focal cerebral ischemia is associated with tissue acid-
osis which leads to vasoparalysis and, in consequence,
to a severe disturbance of the regulation of blood flow
[59]. In the center of the ischemic territory, CO2
reactivity is abolished or even reversed, i.e. blood flow
may decrease with increasing arterial pCO2. This
paradoxical “steal” effect has been attributed to the
rerouting of blood to adjacent non-ischemic brain
regions in which CO2 reactivity remains intact.
Stroke also impairs autoregulation but the dis-
turbance is more severe with decreasing rather than
with increasing blood pressure. This is explained by
the fact that in the ischemic tissue a decrease of local
Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 17 Apr 2018 at 16:27:57, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781107239340.002
brain perfusion pressure cannot be compensated by
further vasorelaxation whereas an increase may shift
the local perfusion pressure into the autoregulatory
range and cause vasoconstriction. An alternative
explanation is “false autoregulation” due to brain
edema which causes an increase in local tissue pres-
sure that precludes a rise of the actual tissue perfusion
pressure. Failure of cerebral autoregulation can be
demonstrated in such instances by dehydrating the
brain in order to reduce brain edema.
After transient ischemia, vasorelaxation persists
for some time, which explains the phenomenon of
post-ischemic hyperemia or luxury perfusion. During
luxury perfusion, oxygen supply exceeds oxygen require-
ments of the tissue, as reflected by the appearance of
red venous blood. With the cessation of tissue acido-
sis, vascular tone returns, and blood flow declines to
or below normal. At longer recirculation times auto-
regulation – but not CO2 reactivity – may recover,
resulting in persisting failure of metabolic regulation.
This is one of the reasons why primary post-ischemic
recovery may be followed by delayed post-ischemic
hypoxia and secondary metabolic failure [60].
Disturbances of flow regulation through ischemia:
tissue acidosis leads to vasorelaxation, CO2 reactivity is
abolished or even reversed, and autoregulation is
impaired.
Disturbances of microcirculation
With the increasing understanding of the pathobiol-
ogy of the neurovascular unit, microcirculatory
disturbances are recognized to contribute to the evo-
lution of ischemic brain injury [61]. Such disturb-
ances develop at the capillary level within the first
hour of focal ischemia and may persist even after full
reversal of vascular occlusion (incomplete microcir-
culatory reperfusion). The dominating pathology is
the narrowing of the capillary lumen, induced by
constriction of pericytes and swelling of pericapillary
astrocytic endfeet. The capillaries are filled with
aggregated red blood cells, leukocytes, and fibrin/
platelet deposits, the high viscosity of which adds to
the increased vascular resistance of the reduced
capillary lumen.
The mechanism of microcirculatory impairment
is multifactorial. Pericytes constrict in response to the
generation of reactive oxygen species (ROS), swelling
of astrocytic endfeet is due to cytotoxic brain edema, 11
and leukocyte adhesion to the vessel wall is part of the
 Section 1: Etiology, pathophysiology, and imaging
inflammatory response mediated by the generation of
chemoattractants, cytokines, and chemokines. Finally,
the activation of proteolytic enzymes contributes
to the dismantlement of basal lamina and results in
damage of the blood–brain barrier, an increase
in interstitial tissue pressure, and the risk of hemor-
rhagic transformation.
The impairment of microcirculation is equivalent
to a reduction of nutritional blood flow. During per-
manent vascular occlusion it aggravates the effect of
primary ischemia, particularly in the borderzone of
the infarct, and after transient vascular occlusion it
prevents adequate reoxygenation despite recanaliza-
tion of the supplying artery. It is still unresolved to
what extent microcirculatory impairment contributes
to or originates from ischemic injury but there is
general consent that microvascular protection is a
requirement for successful stroke treatment [62].
Focal brain ischemia is aggravated by microcirculatory
disturbances which may persist despite recanalization.
Anastomotic steal phenomena
The brain is protected against focal disturbances of
blood flow by the collateral circulation, which pro-
vides a subsidiary network of vascular channels when
principal conduits fail [63]. However, the connection
of ischemic and non-ischemic vascular territories by
anastomotic channels may divert blood from one
brain region to another, depending on the magnitude
and direction of the blood pressure gradients across
the anastomotic connections (for review see Tode and
McGraw [64]). The associated change of regional
blood flow is called “steal” if it results in a decrease in
flow, or “inverse steal” if it results in an improvement
in flow. Inverse steal has also been referred to as the
Robin Hood syndrome in analogy to the legendary
hero who took from the rich and gave to the poor.
Steals are not limited to a particular vascular ter-
ritory and may affect both the extra- and intracerebral
circulation. Examples of extracerebral steals are the
subclavian, the occipital-vertebral, and the ophthal-
mic steal syndrome. Intracerebral steal occurs across
collateral pathways of brain, notably the circle of
Willis and Heubner’s network of pial anastomoses.
The pathophysiological importance of steal has been
disputed but as it depends on the individual hemody-
namic situation it may explain unintended effects
12 when flow is manipulated by alterations of arterial
pCO2 or vasoactive drugs. Most authors, therefore, do
Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 17 Apr 2018 at 16:27:57, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781107239340.002
not recommend such manipulations for the treatment
of stroke.
“Steal”: decrease in focal blood flow when blood is
diverted from one brain region to another by
anastomotic channels; “inverse steal” if that results in an
improvement in flow.
The concept of ischemic penumbra
Energy requirements of brain tissue
The energy demand of the central nervous tissue is
very high and therefore sufficient blood supply to the
brain must be maintained constantly. A normal adult
male’s brain containing about 130 billion neurons
(21.5 billion in the neocortex) [65] comprises only
2% of total body mass, yet consumes at rest approxi-
mately 20% of the body’s total basal oxygen consump-
tion supplied by 16% of the cardiac blood output. The
brain’s oxygen consumption is almost entirely for the
oxidative metabolism of glucose, which in normal
physiological conditions is the almost exclusive sub-
strate for the brain’s energy metabolism (Table 1.1)
[66]. Glucose metabolized in neuronal cell bodies is
mainly to support cellular vegetative and house-
keeping functions, e.g. axonal transport, biosynthesis
of nucleic acids, proteins, and lipids, as well as other
energy-consuming processes not related directly to
the generation of action potentials. Therefore, the rate
of glucose consumption of neuronal cell bodies is
essentially unaffected by neuronal functional acti-
vation. Increases in glucose consumption (and
regional blood flow) evoked by functional activation
are confined to synapse-rich regions, i.e. the neuropil,
which contains axonal terminals, dendritic processes,
and also the astrocytic processes that envelop the
synapses. The magnitudes of these increases are lin-
early related to the frequency of action potentials in
Table 1.1. Cerebral blood flow (CBF), oxygen utilization
(CMRO2), and metabolic rates of glucose (CMRGIc) in man
(approximated values)
Cortex White Global
matter
CBF (ml/100 g/min) 65 21 47
CMRO2 230 80 160
(μmol/100 g/min)
CMRGlc 40 20 32
(μmol/100 g/min)
 Chapter 1: Neuropathology and pathophysiology of stroke

(a) (b)
Glutamate-releasing
Astrocyte Capillary
presynaptic terminal

Glucose
34% postsynaptic ion fluxes

Glycolysis
Lactate

Oxidative Glucose 10% neuronal resting

phosphorylation potential
Gln
Glu Gln
3% glial resting potential
+
Glu K 3% glutamate recycling
EAAT Na
+ Na+/K+-ATPase 3% presynaptic Ca2+
+
Na

47% action potential propagation

Postsynaptic site lonotropic glutamate receptor
NMDAR

Figure 1.5. (a) Schematic representation of the mechanism for glutamate-induced glycolysis in astrocytes during physiological activation
[127]. (b). Distribution of energy expenditure in rat cortex at a mean spike rate of 4 Hz: most energy is required for activity, only 13% is
used for maintenance of resting potentials of neurons and glial cells [69,128]. EAAT ¼ excitatory amino-acid transporter; NMDAR ¼ N-methyl-
D-aspartate receptor.

the afferent pathways, and increases in the projection

zones occur regardless of whether the pathway is
excitatory or inhibitory. Energy requirements of func-
tional activation are due mostly to stimulation of the
Na+/K+-ATPase activity to restore the ionic gradients
across the cell membrane and the membrane poten-
tials following spike activity, and are rather high com-
pared to the basal energy demands of neuronal cell
bodies (Figure 1.5) [67].
In excitatory glutamatergic neurons, which
account for 80% of the neurons in the mammalian
cortex, glucose utilization during activation is medi-
ated by astrocytes which by anaerobic glycolysis pro-
vide lactate to the neurons where it is used for
oxidative metabolism [68]. Overall, 87% of the total
energy consumed is required for signaling, mainly
action potential propagation and postsynaptic ion
fluxes, and only 13% is expended in maintaining
membrane resting potential (Figure 1.5) [69].
The mechanisms by which neurotransmitters
other than glutamate influence blood flow and energy
metabolism in the brain are still not understood [70].
A normal adult male’s brain comprises only 2% of total
body mass, yet consumes at rest approximately 20% of
the body’s total basal oxygen consumption. Glucose is
the almost exclusive substrate for the brain’s energy
metabolism; 87% of the total energy consumed is
required for signaling, mainly action potential
propagation and postsynaptic ion fluxes.
Viability thresholds of brain ischemia
The different amounts of energy required for the
generation of membrane potential and the propaga-
tion of electrical activity are reflected by different
thresholds of oxygen consumption and blood
flow that must be maintained to preserve neuronal
function and morphological integrity. Flow values
below normal but above the threshold of neuronal
function are referred to as “benign oligemia.”
The flow range between the thresholds of neuronal
function and morphological integrity is called the
“ischemic penumbra” [71]. It is characterized by
the preservation of membrane polarization and the
potential of functional recovery without morpho-
logical damage, provided that local blood flow can
be re-established [72, 73]. The “infarct core” is the
area in which blood flow declines below the thresh-
old of morphological integrity and in which tissue
necrosis evolves.
According to the classical concept of viability
thresholds, functional activity – reflected by the
amplitudes of spontaneous and evoked electrical
activity – begins to decline at flow values below 50%
of control and is completely suppressed at about 30%
of control [71]. In awake monkeys these values cor-
respond to the progression of neurological injury
from mild paresis at 22 ml/100 g/min to complete 13
paralysis at 8 ml/100 g/min. Morphological damage

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 17 Apr 2018 at 16:27:57, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781107239340.002
 Section 1: Etiology, pathophysiology, and imaging
evolves as soon as cell membranes depolarize (“ter-
minal” depolarization) and occurs at flow values
below 15–20% of control. Biochemically, functional
suppression is associated with the inhibition of pro-
tein synthesis at about 50% and the development of
lactacidosis at 30–40% of control, whereas membrane
depolarization and morphological injury correspond
to the breakdown of energy metabolism and the loss
of adenosine triphosphate (ATP) at about 18% of
control (Figure 1.6).
A more detailed picture of the dynamics of injury
evolution is obtained by the simultaneous recording
of local blood flow and spontaneous unit activity of
cortical neurons [74]. According to these measure-
ments, unit activity disappeared at a mean value of
18 ml/100 g/min but the large variability of the func-
tional thresholds of individual neurons (6–22 ml/
100 g/min) indicates differential vulnerability even
within small cortical sectors. This explains the gradual
development of neurological deficits, which may be
related to differences in single cell activity with regu-
lar or irregular discharges at flow levels above the
threshold of membrane failure.
Whereas neuronal function is impaired immedi-
ately when flow drops below the threshold, the devel-
opment of irreversible morphological damage is time
14 dependent. Based on recordings from a considerable
number of neurons during and after ischemia of
Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 17 Apr 2018 at 16:27:57, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781107239340.002
Figure 1.6. Diagrammatic
representation of viability thresholds
of focal brain ischemia.
EEG ¼ electroencephalogram; OEF ¼
oxygen extraction fraction; SEP ¼
somatosensory evoked potential.
varying degree and duration it was possible to con-
struct a discriminant curve representing the worst
possible constellation of residual blood flow and dur-
ation of ischemia still permitting neuronal recovery
(Figure 1.7). These results broaden the concept of the
ischemic penumbra: the tissue fate – potential of
recovery or irreversible damage – is determined not
only by the level of residual flow but also by the
duration of the flow disturbance. Each level of
decreased flow can, on average, be tolerated for a
defined period of time; flow between 17 and 20 ml/
100 g/min can be tolerated for prolonged but yet
undefined periods of time. As a rule used in many
experimental models, flow rates of 12 ml/100 g/min
lasting for 2–3 hours lead to large infarcts, but indi-
vidual cells may become necrotic after shorter periods
of time and at higher levels of residual flow.
The ischemic penumbra is the range of perfusion
between the flow threshold for preservation of function
and the flow threshold for preservation of morphological
integrity. It is characterized by the potential for
functional recovery without morphological damage.
Imaging of penumbra
Based on the threshold concept of brain ischemia, the
penumbra can be imaged on quantitative flow maps
using empirically established flow thresholds. A more
 Chapter 1: Neuropathology and pathophysiology of stroke

Figure 1.7. (A) Activity of a single neuron during graded ischemia before, during, and after reversible MCA occlusion. (B) Recovery of neuronal
function after a limited period of ischemia. (C) Diagram of CBF thresholds required for the preservation of function and morphology of
brain tissue. The activity of individual neurons is blocked when flow decreases below a certain threshold (upper dashed line) and returns
when flow is raised again above this threshold. The fate of a single cell depends on the duration for which CBF is impaired below a certain level.
The solid line separates structurally damaged from functionally impaired, but morphologically intact tissue, the “penumbra.” The upper dashed 15
line distinguishes viable from functionally impaired tissue. EP ¼ evoked potentials. (Modified with permission from Heiss and Rosner [129]).

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 17 Apr 2018 at 16:27:57, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781107239340.002
 Section 1: Etiology, pathophysiology, and imaging
precise approach is the imaging of threshold-
dependent biochemical disturbances to demarcate
the mismatch between disturbances which occur only
in the infarct core and others which also affect the
penumbra (Figure 1.8) [75]. Under experimental con-
ditions the most reliable way to localize the infarct core
is the loss of ATP on bioluminescent images of tissue
ATP content. A biochemical marker of core plus pen-
umbra is tissue acidosis or the inhibition of protein
synthesis. The penumbra is the difference between the
respective lesion areas. The reliability of this approach
is supported by the precise colocalization of gene tran-
scripts that are selectively expressed in the penumbra,
such as the stress protein hsp70, or the documentation
of the gradual disappearance of the penumbra with
increasing ischemia time [76].
Non-invasive imaging of the penumbra is possible
using positron emission tomography (PET) or mag-
netic resonance imaging (MRI). Widely used PET
parameters are the increase in oxygen extraction or
the mismatch between reduced blood flow and the
preservation of vitality markers, such as flumazenil
binding to central benzodiazepine receptors [77]. An
alternative PET approach is the use of hypoxia
markers such as 18F-misonidazole (F-MISO), which
is trapped in viable hypoxic but not in normoxic or
necrotic tissue [78].
The best-established MRI approach for penumbra
imaging is the calculation of mismatch maps between
the signal intensities of perfusion (PWI) and diffusion-
weighted images (DWI), but its reliability has been
questioned [79]. An alternative method is quanti-
16 tative mapping of the apparent diffusion coefficient
(ADC) of water, which reveals a robust correlation
Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 17 Apr 2018 at 16:27:57, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781107239340.002
Figure 1.8. Biochemical imaging of
infarct core, penumbra, and benign
oligemia after experimental middle
cerebral artery occlusion. The core is
identified by ATP depletion, the
penumbra by the mismatch between the
suppression of protein synthesis and ATP
depletion (top) or by the mismatch
between tissue acidosis and ATP
(bottom), and benign oligemia by the
reduction of blood flow in the absence of
biochemical alterations. (Modified
with permission from Hossman and
Mies [130]).
with the biochemically characterized penumbra for
ADC values between 90% and 77% of control [80].
Recently MR stroke imaging has been performed by
combining PWI, DWI, and pH-weighted imaging
(pHWI) [81]. The mismatch between DWI and
pHWI detects the penumbra, and that between PWI
and pHWI the area of benign oligemia, i.e. a region in
which flow reduction is not severe enough to cause
metabolic disturbances. Diffusion kurtosis imaging
(DKI), an extension of diffusion imaging, demarcates
the regions with structural damage that cannot be
salvaged upon reperfusion [82].
Finally, new developments in MR molecular
imaging are of increasing interest for stroke research
[83]. These methods make use of contrast-enhanced
probes that trace gene transcription or of intracellular
conjugates that reflect the metabolic status and/or
bind to stroke markers. The number of molecules that
can be identified by these methods rapidly expands
and greatly facilitates the regional analysis of stroke
injury.
Non-invasive imaging of the penumbra is possible using
positron emission tomography (PET) or magnetic
resonance imaging (MRI).
Mechanisms of infarct expansion
With the advent of non-invasive imaging evidence has
been provided that brain infarcts grow (Figure 1.9).
This growth is not due to the progression of ischemia
because the activation of collateral blood supply and
spontaneous thrombolysis tend to improve blood
flow over time. Infarct progression can be differenti-
ated into three phases. During the acute phase tissue
 Chapter 1: Neuropathology and pathophysiology of stroke
injury is the direct consequence of the ischemia-
induced energy failure and the resulting terminal
depolarization of cell membranes. At flow values
below the threshold of energy metabolism this injury
is established within a few minutes after the onset of
ischemia. During the subsequent subacute phase, the
infarct core expands into the peri-infarct penumbra
until, after 4–6 hours, core and penumbra merge. The
reasons for this expansion are peri-infarct spreading
depressions and a multitude of cell biological disturb-
ances, collectively referred to as molecular cell injury.
Moreover, a delayed phase of injury evolves which
may last for several days or even weeks. During this
phase secondary phenomena such as vasogenic
edema, inflammation, and possibly programmed cell
death may contribute to a further progression of
injury.
The largest increment of infarct volume occurs
during the subacute phase in which the infarct core
expands into the penumbra. Using multiparametric
imaging techniques for the differentiation between
core and penumbra, evidence could be provided that
in small rodents submitted to permanent occlusion of
the MCA at its origin, the penumbra equals the
volume of the infarct core at 1 hour, but after 3 hours
more than 50% and between 6 and 8 hours almost all
of the penumbra has disappeared and is now part of
the irreversibly damaged infarct core [76]. In larger
Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 17 Apr 2018 at 16:27:57, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781107239340.002
Figure 1.9. Relationship between peri-
infarct spreading depressions (above) and
infarct growth (below) during permanent
focal brain ischemia induced by occlusion
of the middle cerebral artery in rat. The
effect of spreading depressions on
electrical brain activity (EEG) and blood
flow (LDF) are monitored by DC recording
of the cortical steady potential, and infarct
growth by MR imaging of the apparent
diffusion coefficient (ADC) of brain water.
(Modified with permission from
Hossmann [131,132]).
animals the infarct core may be heterogeneous with
multiple mini-cores surrounded by multiple mini-
penumbras but these lesions also expand and eventu-
ally progress to a homogeneous defect with a similar
time course [84].
Brain infarcts evolve in three phases:
 acute phase, within a few minutes after the onset of
ischemia; terminal depolarization of cell
membranes;
 subacute phase, within 4–6 hours; spreading
depression and molecular cell injury, the infarct core
expands into the peri-infarct penumbra;
 delayed phase, several days to weeks; vasogenic
edema, inflammation, and possibly programmed
cell death.
In the following, the most important mediators of
infarct progression will be discussed.
Peri-infarct spreading depression
A functional disturbance contributing to the growth
of the infarct core into the penumbra zone is the
generation of peri-infarct spreading depression-like
depolarizations (Figure 1.9) [85]. These depolariza-
tions are initiated at the border of the infarct core
and spread over the entire ipsilateral hemisphere.
During spreading depression the metabolic rate of 17
the tissue markedly increases in response to the
 Section 1: Etiology, pathophysiology, and imaging
greatly enhanced energy demands of the activated ion
exchange pumps. In the healthy brain the associated
increase of glucose and oxygen demands are coupled
to a parallel increase of blood flow but in the peri-
infarct penumbra this flow response is suppressed or
even reversed [86]. As a result, a misrelationship
arises between the increased metabolic workload
and the low oxygen supply, leading to transient epi-
sodes of hypoxia and the stepwise increase in lactate
during the passage of each depolarization.
The pathogenic importance of peri-infarct depo-
larizations for the progression of ischemic injury is
supported by the linear relationship between the
number of depolarizations and infarct volume. Cor-
relation analysis of this relationship suggests that
during the initial 3 hours of vascular occlusion each
depolarization increases the infarct volume by more
than 20%. This is probably one of the reasons that
glutamate antagonists, which are potent inhibitors of
spreading depression, reduce the volume of brain
infarcts [87].
Peri-infarct spreading depressions are depolarizations
initiated at the border of the infarct core and may
contribute to progression of ischemic injury.
Molecular mechanisms of injury progression
In the borderzone of permanent focal ischemia or in
the core of the ischemic territory after transient vas-
cular occlusion, cellular disturbances may evolve that
18 cannot be explained by a lasting impairment of blood
flow or energy metabolism. These disturbances are
Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 17 Apr 2018 at 16:27:57, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781107239340.002
Figure 1.10. Schematic representation
of molecular injury pathways leading to
necrotic or apoptotic brain injury after
focal brain ischemia. Injury pathways can
be blocked at numerous sites, providing
multiple approaches for the amelioration
of both necrotic and apoptotic cell death.
DAG ¼ diacylglycerol; IP3 ¼ inositol 1, 4,
5-trisphosphate; PARP ¼ poly (ADP-
ribose) polymerase; TPA ¼ tissue
plasminogen activator.
referred to as molecular injury, where the term
“molecular” does not anticipate any particular injury
pathway (for reviews see [88], [89]). The molecular
injury cascades (Figure 1.10) are interconnected in
complex ways, which makes it difficult to predict their
relative pathogenic importance in different ischemia
models. In particular, molecular injury induced by
transient focal ischemia is not equivalent to the alter-
ations that occur in the penumbra of permanent
ischemia. Therefore, the relative contribution of the
following injury mechanisms differs in different types
of ischemia.
Acidotoxicity: during ischemia oxygen depletion
and the associated activation of anaerobic glycolysis
cause an accumulation of lactic acid which, depending
on the severity of ischemia, blood glucose levels, and
the degree of ATP hydrolysis, results in a decline
of intracellular pH to between 6.5 and below 6.0.
As the severity of acidosis correlates with the
severity of ischemic injury, it has been postulated
that acidosis is neurotoxic. Recently, evidence has
been provided that ASICs (acid-sensing ion channels)
are glutamate-independent vehicles of calcium
flux, and that blockade of ASICs attenuates stroke
injury. This suggests that acidosis may induce calcium
toxicity, and that this effect is the actual mechanism
of acidotoxicity [90].
Excitotoxicity: shortly after the onset of ischemia,
excitatory and inhibitory neurotransmitters are
released, resulting in the activation of their specific
receptors. Among these neurotransmitters, particular
attention has been attributed to glutamate, which
 Chapter 1: Neuropathology and pathophysiology of stroke
under certain experimental conditions may produce
excitotoxic cell death [91]. The activation of ionotro-
pic glutamate receptors results in the inflow of cal-
cium from the extracellular into the intracellular
compartment, leading to mitochondrial calcium
overload and the activation of calcium-dependent
catabolic enzymes. The activation of metabotropic
glutamate receptors induces the inositol 1,4,5-
trisphosphate (IP3)-dependent signal transduction
pathway, leading among others to the stress response
of ER, and by induction of immediate-early genes
(IEGs) to adaptive genomic expressions. At very
high concentration, glutamate results in primary
neuronal necrosis. However, following pharmaco-
logical inhibition of ionotropic glutamate receptors,
an apoptotic injury mechanism evolves that may
prevail under certain pathophysiological conditions.
The importance of excitotoxicity for ischemic cell
injury has been debated, but this does not invalidate
the beneficial effect of glutamate antagonists for the
treatment of focal ischemia. An explanation for this
incongruity is the above-described pathogenic role
of peri-infarct depolarizations in infarct expansion.
As glutamate antagonists inhibit the spread of these
depolarizations, the resulting injury is also reduced.
Calcium toxicity: in the intact cell, highly efficient
calcium transport systems assure the maintenance
of a steep calcium concentration gradient of approxi-
mately 1:10 000 between the extra- and the intracel-
lular compartment on the one hand, and between the
cytosol and the ER on the other. During ischemia
anoxic depolarization in combination with the acti-
vation of ionotropic glutamate and acid-sensing ion
channels causes a sharp rise of cytosolic calcium [92].
At the onset of ischemia this rise is further enhanced
by activation of metabotropic glutamate receptors
which mediate the release of calcium from ER,
and after recovery from ischemia by activation of
transient receptor potential (TRP) channels which
perpetuate intracellular calcium overload despite
the restoration of ion gradients (Ca2+ paradox)
[93]. The changes in intracellular calcium activity
are highly pathogenic. Prolonged elevation of cyto-
solic calcium causes mitochondrial dysfunction and
induces catabolic changes, notably by activation of
Ca2+-dependent effector proteins and enzymes such
as endonucleases, phospholipases, protein kinases,
and proteases that damage DNA, lipids, and pro-
teins. The release of calcium from the ER evokes an
ER stress response, which mediates a great number
Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 17 Apr 2018 at 16:27:57, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781107239340.002
of ER-dependent secondary disturbances, notably
inhibition of protein synthesis. Calcium-dependent
pathological events are therefore complex and con-
tribute to a multitude of secondary molecular injury
pathways.
Free radicals: in brain regions with low or intermit-
tent blood perfusion, reactive oxygen species (ROS)
are formed which produce peroxidative injury of
plasma membranes and intracellular organelles [94].
The reaction with NO leads to the formation of per-
oxynitrite, which also causes violent biochemical reac-
tions. Secondary consequences of free radical
reactions are the release of biologically active free fatty
acids such as arachidonic acid, the induction of ER
stress and mitochondrial disturbances, the initiation
of an inflammatory response, breakdown of the
blood–brain barrier, and fragmentation of DNA.
The last may induce apoptosis and thus enhance
molecular injury pathways related to mitochondrial
dysfunction. The therapeutic benefit of free radial
scavengers, however, is limited, as documented by
the therapeutic failure of the free-radical-trapping
agent NXY-059 [95].
Nitric oxide toxicity: NO is a product of NO
synthase (NOS) acting on arginine. There are at least
three isoforms of NOS: eNOS is constitutively
expressed in endothelial cells, nNOS in neurons, and
the inducible isoform iNOS mainly in macrophages.
Pathophysiologically, NO has two opposing effects
[96]. In endothelial cells the generation of NO leads
to vascular dilatation, an improvement of blood flow
and the alleviation of hypoxic injury, whereas in
neurons it contributes to glutamate excitotoxicity
and – by formation of peroxynitrite – to free-
radical-induced injury. The net effect of NO thus
depends on the individual pathophysiological situ-
ation and is difficult to predict.
Zinc toxicity: zinc is an essential catalytic and
structural element of numerous proteins and a sec-
ondary messenger which is released from excitatory
synapses during neuronal activation. Cytosolic zinc
overload may promote mitochondrial dysfunction
and generation of ROS, activate signal transduction
pathways such as mitogen-activated protein kinase
(MAPK), enhance calcium toxicity, and promote
apoptosis [97]. However, at low concentration zinc
may also exhibit neuroprotective properties, indicat-
ing that cells may possess a specific zinc set-point by
which too little or too much zinc can promote ische- 19
mic injury [98].
 Section 1: Etiology, pathophysiology, and imaging
ER stress and inhibition of protein synthesis: a
robust molecular marker for the evolution of ische-
mic injury is inhibition of protein synthesis, which
persists throughout the interval from the onset of
ischemia until the manifestation of cell death [45]. It
is initiated by a disturbance of the calcium homoeo-
stasis of the ER, which results in ER stress and various
cell biological abnormalities such as un- or misfolding
of proteins, expression of stress proteins, and a global
inhibition of the protein synthesizing machinery.
The last is due to the activation of protein kinase
R (PKR), which causes phosphorylation and inactiva-
tion of the alpha subunit of eukaryotic initiation
factor eIF2. This again leads to selective inhibition of
polypeptide chain initiation, disaggregation of ribo-
somes, and inhibition of protein synthesis at the level
of translation.
To restore ER function un- or misfolded proteins
must be refolded (by activation of the unfolded pro-
tein response, UPR) or degraded (by ER-associated
degradation, ERAD). Cells in which UPR and ERAD
fail to restore ER function die by apoptosis [99].
Obviously, persistent inhibition of protein synthe-
sis is incompatible with cell survival but as the mani-
festation of cell death greatly varies in different cell
populations, other factors must also be involved.
Mitochondrial disturbances: to mitigate metabolic
or environmental stress, functional mitochondria
are maintained by fission and fusion [100]. However,
the concurrence of an increased cytosolic calcium
activity with the generation of ROS may lead to an
increase in permeability of the inner mitochondrial
membrane (mitochondrial permeability transition,
MPT), which has been associated with the formation
of a permeability transition pore (PTP). The PTP is
a Ca2+-, ROS- and voltage-dependent, cyclosporine
A-sensitive high-conductance channel, located in
the inner mitochondrial membrane. It is also a
reversible fast Ca2+ release channel, facilitated by the
mitochondrial matrix protein cyclophilin D [101].
The increase in permeability of the inner mito-
chondrial membrane has two pathophysiologically
important consequences. The breakdown of the
electrochemical gradient interferes with mitochon-
drial oxidative phosphorylation and, in consequence,
with aerobic energy production. Furthermore, the
equilibration of mitochondrial ion gradients causes
swelling of the mitochondrial matrix, which eventu-
20 ally will cause disruption of the outer mitochon-
drial membrane and the release of pro-apoptotic
Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 17 Apr 2018 at 16:27:57, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781107239340.002
mitochondrial proteins (see below). Ischemia-induced
mitochondrial disturbances thus contribute to delayed
cell death both by impairment of the energy state and
the activation of apoptotic injury pathways [102].
A large number of molecular disturbances are involved
in the progression of ischemic damage.
Inflammation
Brain infarcts evoke a strong inflammatory response
which is thought to contribute to the progression of
ischemic brain injury. Gene expressions related to this
response have, therefore, been extensively investigated
to search for possible pharmacological targets (for
review see Rothwell and Luheshi [103]). The inflamma-
tory response of the ischemic tissue has been associated,
among others, with the generation of free radicals in
reperfused or critically hypoperfused brain tissue. The
prostaglandin synthesizing enzyme cyclo-oxygenase-2
(COX-2) and nuclear factor-kappa B (NF-kappa B), a
transcription factor that responds to oxidative stress, are
strongly upregulated and may be neurotoxic, as sug-
gested by the beneficial effect of COX-2 inhibitors.
Infarct reduction was also observed after genetic or
pharmacological inhibition of matrix metalloproteinase
(MMP)-9 but this effect has been disputed.
A key player in the intracellular response to cyto-
kines is the JAK (janus kinase)/STAT (signal trans-
ducer and activator of transcription) pathway, which
induces alterations in the pattern of gene transcription.
These changes are associated with either cell death or
survival and suggest that inflammation may be both
neurotoxic and neuroprotective [104]. Inflammatory
reactions and the associated free-radical-mediated pro-
cesses are, therefore, important modulators of ischemic
injury but the influence on the final outcome is difficult
to predict.
Inflammatory reactions are important modulators of
ischemic injury.
Brain edema
Ischemic brain edema can be differentiated into two
pathophysiologically different types: an early cytotoxic
type, followed after some delay by a late vasogenic type
of edema. The cytotoxic type of edema is threshold
dependent. It is initiated at flow values of similar to
30% of control when stimulation of anaerobic metab-
olism causes an increase of brain tissue osmolality and,
hence, an osmotically obliged cell swelling. At flow
 Chapter 1: Neuropathology and pathophysiology of stroke
values below 20% of control, anoxic depolarization
and equilibration of ion gradients across the cell mem-
branes further enhance intracellular osmolality and the
associated cell swelling. The intracellular uptake of
sodium is also associated with a coupled movement
of water that is independent of an osmotic gradient
and which is referred to as “anomalous osmosis.”
In the absence of blood flow, cell swelling occurs at
the expense of the extracellular fluid volume, leading
to the shrinkage of the extracellular compartment, but
not to a change in the net water content. The shift of
fluid is reflected by a decrease of the ADC of water,
which is the reason for the increase of signal intensity
in diffusion-weighted MR imaging [80]. However, if
some residual blood flow persists, water is taken up
from the blood, and the net tissue water content
increases. After vascular occlusion this increase starts
within a few minutes after the onset of ischemia and
causes a gradual increase in brain volume.
With the evolution of tissue necrosis and the deg-
radation of basal lamina, the blood–brain barrier
breaks down [105], and after 4–6 hours serum pro-
teins begin to leak from the blood into the brain. This
disturbance initiates a vasogenic type of edema, which
further enhances the water content of the tissue.
Vasogenic edema reaches its peak at 1–2 days after
the onset of ischemia and may cause an increase of
tissue water by more than 100%. If brain infarcts are
large, the volume increase of the edematous brain
tissue may be so pronounced that transtentorial her-
niation results in compression of the midbrain. Under
clinical conditions, this “malignant” form of brain
infarction is by far the most dangerous complica-
tion of stroke and an indication for decompressive
craniectomy [106].
Vasogenic edema, in contrast to the early cyto-
toxic type of edema, is isoosmotic and accumulates
mainly in the extracellular compartment. This
reverses the cytotoxic narrowing of the extracellular
space and explains the “pseudonormalization” of the
signal intensity observed in diffusion-weighted MR
imaging [107]. However, as the total tissue water
content is increased at this time, the high signal
intensity in T2-weighted images clearly differentiates
this situation from a “real” recovery to normal.
The formation of cytotoxic and, to a lesser extent,
also vasogenic edema requires the passage of water
through aquaporin channels located in the plasma
membrane [108]. Inhibition of aquaporin water
conductance may, therefore, reduce the severity of
Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 17 Apr 2018 at 16:27:57, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781107239340.002
ischemic brain edema. Similarly, the inhibition of
sodium transport across sodium channels has been
suggested to reduce edema formation. However, as
the driving force for the generation of edema is the
gradient of osmotic and ionic concentration differ-
ences built up during ischemia, aquaporin and
sodium channels may modulate the speed of edema
generation but not the final extent of tissue water
accumulation. Their pathophysiological importance
is, therefore, limited.
Early cytotoxic edema is caused by osmotically induced
cell swelling; the later vasogenic edema is isoosmotic,
caused by breakdown of the blood–brain barrier, and
accumulates in the extracellular compartment.
Apoptosis
Apoptosis is an evolutionary conserved form of pro-
grammed cell death that in multicellular organisms
matches cell proliferation to preserve tissue homoeo-
stasis [109]. It is an active process that requires intact
energy metabolism and protein synthesis, and it is
initiated essentially by two pathways: an extrinsic
death receptor-dependent route, and an intrinsic
pathway which depends on the mitochondrial release
of pro-apoptotic molecules such as apoptosis-
inducing factor (AIF) and cytochrome C. In focal
ischemia pro-apoptotic pathways are also initiated
by activation of toll-like receptors 2 and 4, the
NOTCH-1 receptor, and the adiponectin receptor 1.
Apoptotic pathways involve a series of enzymatic
reactions and converge in the activation of caspase-
3, a cystine protease, which contributes to the execu-
tion of cell death. An end stage of this process is the
ordered disassembly of the genome, resulting in a
laddered pattern of oligonucleosomal fragments as
detected by electrophoresis or TUNEL.
Although apoptosis is mainly involved in physio-
logical cell death, it is widely assumed to contribute to
the pathogenesis of diseases, including cerebral ische-
mia [110]. In the context of stroke this is difficult to
understand because in areas with primary cell death
the obvious cause is energy failure, and in regions
with delayed injury the dominating biochemical dis-
turbance is the irreversible suppression of protein
synthesis. However, ischemia induces a multitude of
biochemical reactions that are reminiscent of apop-
tosis, such as the expression of p53, JNK, c-jun, p38,
cyclin-dependent kinase 5, or caspase 3, all of which 21
correlate to some degree with the severity of injury.
 Section 1: Etiology, pathophysiology, and imaging
Moreover, inhibition of these reactions by gene
manipulation or pharmacological interventions
reduces the volume of brain infarcts. It has, therefore,
been suggested that ischemic cell death is a hybrid of
necrosis and apoptosis, appearing on a continuum
with the two forms of cell death at its poles [111].
Apoptosis, an active form of programmed cell death,
may contribute to a certain extent to ischemic cell death.
Pre- and postconditioning of ischemic injury
The molecular signaling cascades initiated by brain
ischemia are not solely destructive but may also exert
a neuroprotective effect. In fact, most of the above-
described injury pathways including ischemia itself
induce a transient state of increased ischemic toler-
ance, provided the initial injury remains subliminal
for tissue destruction. This effect is called “ischemic
preconditioning” and can be differentiated into three
phases: during the induction phase molecular sensors
which respond to the preconditioning stimulus are
activated by transcription factors; the transduction
phase results in the amplification of the signal; and
during the effector phase proteins with a protective
impact are switched on [112]. The increase in ische-
mia tolerance appears 2 to 3 days after the precondi-
tioning stimulus, and it slowly disappears after
1 week.
An important preconditioning pathway is the upre-
gulation of the hypoxia-inducible factor 1 (HIF-1)
in astrocytes. HIF-1 is a transcription factor that among
others induces the expression of erythropoietin (EPO)
which binds to the neuronal EPO receptor and
which exhibits potent neuroprotective effects. Another
putative mechanism is the ER stress response. Depletion
of ER calcium stores causes accumulation of unfolded
proteins in the ER lumen and induces the activation
of two highly conserved stress responses, the ER over-
load response (EOR) and the unfolded protein response
(UPR). EOR triggers activation of the transcription
factor NF-kappa B, and UPR causes a suppression of
the initiation of protein synthesis. As the latter contri-
butes to delayed ischemic injury (see above) its reduc-
tion may have a neuroprotective effect.
Evidence has also been provided that ischemic
injury can be alleviated by repeated mechanical inter-
ruptions of blood reperfusion after a period of transi-
ent focal ischemia [113]. This phenomenon, termed
22 “ischemic postconditioning,” has been associated
with the phosphorylation of several prosurvival
Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 17 Apr 2018 at 16:27:57, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781107239340.002
protein kinases, such as extracellular signal-regulated
kinase (ERK), p38 MAPK, and Akt. The possibility to
influence ischemic injury after the primary impact
is challenging but it remains to be shown for which
kind of clinical situation this finding is of practical
relevance.
Short episodes of ischemia can improve the
tolerance of brain tissue for subsequent blood flow
disturbance.
Regeneration and cell therapy
Brain infarcts produced by focal ischemia are seem-
ingly irresolvable in agreement with Cajal’s classical
statement that in the adult brain “everything may die,
nothing may be regenerated.” This dogma was
reversed by the discovery of three permanently neuro-
genic regions, i.e. the subventricular zone (SVZ), the
subgranular zone (SGZ), and the posterior perireticu-
lar (PPr) area, which provide lifelong supply of
newly generated neurons to the hippocampus and
olfactory bulb. After stroke, neurogenesis increases
in these areas, and some of the newly formed cells
migrate into the infarct penumbra, differentiate into
glia and mature neurons, and survive for at least
several weeks [114]. Neurogenesis may also occur
through the neurovascular unit. After ischemia peri-
cytes strongly migrate into the peri-infarct surround-
ing and contribute to tissue repair by controlling
neurogenesis, angiogenesis, and blood–brain barrier
function [115].
Ischemia-induced neurogenesis is enhanced by
growth factors, NO, inflammation, and various hor-
mones and neurotransmitters, notably estradiol and
dopamine, but it is repressed by activation of the N-
methyl-D-aspartate (NMDA) subtype of glutamate
receptors. The functional consequences of spontan-
eous or drug-enhanced neurogenesis are modest but
optimism is building up for targeted interventions.
Similarly, considerable expectations are placed on the
transplantation of neural progenitor cells, particularly
in combination with growth factors and/or strategies
that permit recruitment of transplanted cells to the
site of injury [116]. However, cell therapy carries the
risk of tumorigenesis, and as major breakthroughs
have not yet been achieved, further research is neces-
sary to explore the actual potentials of stroke regen-
erative medicine.
Several brain regions may provide lifelong supply of
newly generated neurons.
 Chapter 1: Neuropathology and pathophysiology of stroke
Translation of experimental concepts
to clinical stroke
Experimental research has advanced our knowledge
about the pathophysiology of brain disorders, but
the transfer of this knowledge into clinical appli-
cation is difficult and often lacks behind. One
of the reasons are differences between the brain
of experimental animals and humans with respect
to evolutionary state (non-gyrencephalic vs. gyrence-
phalic anatomy (amount of gray vs. white matter),
relative size, cellular density, blood supply, and
metabolism). Additionally, experimental models in
animals cannot be easily compared to complex
human diseases often based on a different patho-
physiology and affecting multimorbid patients. The
other problem arises from the investigative pro-
cedures, which cannot be equally applied in animals
and patients. This is especially true when pathophy-
siological changes obtained by invasive procedures in
animals, e.g. by analysis of tissue samples, by auto-
radiography, or by histology, should be related to the
course of a disease, but cannot be assessed repeatedly
and regionally. To facilitate the transfer of knowledge
from experimental neuroscience to clinical neurology,
it is necessary to develop methods which can be
equally applied in patients and animal models, and
which are not invasive and can be performed repeat-
edly without affecting or harming the object. To this
task of transferring experimental results into clinical
application functional imaging modalities are success-
fully applied.
Detection of the penumbra by PET
PET is still the only method allowing quantitative
determination of various physiological variables in
the brain and was applied extensively for studies in
patients with acute, subacute, or chronic stages of
ischemic stroke (review in Heiss [77]). The introduc-
tion of scanners with high resolution (2.5–5 mm for
human, 1 mm for animal application) made PET a
tool for studying animal models and to compare
repeat examinations of various variables from experi-
ments to the course of disease in humans. The
regional decrease of CBF can be directly observed in
PET as in other studies (single-photon emission com-
puted tomography [SPECT], PWI-MRI, perfusion
computed tomography [PCT]). However, already in
early PET studies [117] preserved glucose consumption
Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 17 Apr 2018 at 16:27:57, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781107239340.002
was observed in regions with decreased flow in the
first hours after the ictus. Since the 1980s, PET with
oxygen-15 tracers became the gold standard for the
evaluation of pathophysiological changes in early
ischemic stroke [118]. The quantitative measurement
of CBF, oxygen utilization (CMRO2), oxygen extrac-
tion fraction (OEF), and cerebral blood volume
(CBV) permitted the independent assessment of per-
fusion and energy metabolism, and demonstrated the
uncoupling of these usually closely related variables.
These studies provided data on flow and metabolic
variables predicting final infarction on late CTs (rCBF
less than 12 ml/100 g/min, CMRO2 less than 65 μmol/
100 g/min). Relatively preserved CMRO2 indicated
maintained neuronal function in regions with
severely reduced CBF; this pattern was coined “misery
perfusion” and served as a definition for the penum-
bra, which is characterized by increased oxygen
extraction fraction (up to more than 80% from the
normal 40–50%). Late CT or MRI often showed these
regions as morphologically intact.
Sequential PET studies of CBF, CMRO2, and
metabolic rate of glucose (CMRGlc) before and
repeatedly up to 24 hours after MCA occlusion in
cats could demonstrate the development and growth
of irreversible ischemic damage. Immediately after
MCA occlusion CBF within the supplied territory
dropped, but CMRO2 was less diminished and was
preserved at an intermediate level. As a consequence,
OEF was increased, indicating misery perfusion, i.e.
penumbra tissue. However, as OEF is also increased
in benign oligemia, demarcation of the penumbra
from normal tissue is not possible. With time, OEF
was decreased, a process which started in the center
and developed centrifugally to the borderline of the
ischemic territory, indicating the conversion into irre-
versible damage and the growth of the MCA infarct.
In experiments with transient MCA occlusion it could
be demonstrated that an infarct did not develop when
reperfusion was initiated to tissue with increased
OEF. Comparable to patients with early thrombolysis,
reperfusion could salvage ischemic tissue in the con-
dition of “penumbra” (Figure 1.11). Similar results
were obtained in ischemia models of baboons. PET
thus permits the differentiation of various tissue com-
partments within an ischemic territory: irreversible
damage by decreased flow and oxygen consumption
below critical thresholds; misery perfusion, i.e. pen-
umbra, by decreased flow, but preserved oxygen util- 23
ization above a critical threshold, expressed by
 Section 1: Etiology, pathophysiology, and imaging

CAT HUMAN

CBF

CMRO2

OEF

Reperfusion spontaneous course rtPA treatment

OEF CMRGlc CBF

CMRGlc

Figure 1.11 Sequential PET images of CBF, CMRO2, and OEF of MCA occlusion in cats compared to images of patients after stroke: Left
columns: In the right cat, the progressive decrease of CMRO2 and the reduction of OEF predict infarction that cannot benefit from reperfusion.
Only if OEF is increased until the start of reperfusion can the infarct be salvaged (left cat). Middle columns: In the patient the areas with
preserved OEF are not infarcted and can survive in spontaneous course (posterior part of ischemic cortex in left, anterior part in right patient,
as indicated on late MRI and CT). Right columns: In patients receiving rtPA treatment measurements of CMRO2 and OEF are not feasible, but
flow determinations show the effect. If reperfusion occurs early enough and before tissue damage, tissue can be salvaged (left patient). If
reperfusion is achieved too late, tissue cannot be salvaged despite hyperperfusion in some parts (right patient).

increased OEF; luxury perfusion by flow increased
above the metabolic demand; anaerobic glycolysis by
a change in the ratio between glucose metabolism and
oxygen utilization. However, PET has severe disad-
vantages limiting its routine application in patients
with stroke: it is a complex methodology, requires
multitracer application, and quantitative analysis
necessitates arterial blood sampling.
Positron emission tomography (PET) is the most
reliable non-invasive method to identify irreversible
tissue damage and penumbra.
Prediction of irreversible tissue damage
24 The prediction of the portion of irreversibly damaged
tissue within the ischemic area early after the stroke
is of utmost importance for the efficiency of treat-
ment. Meticulous analyses of CBF and CMRO2 data
indicated that CMRO2 below 65 μmol/100 g/min
predicted finally infarcted tissue, but also large
portions with flow and oxygen utilization in the pen-
umbra range were included in the final cortical–
subcortical infarcts. Determination of oxygen
utilization additionally requires arterial blood sam-
pling, which limits clinical applicability. These facts
stress the need for a marker of neuronal integrity
that can identify irreversibly damaged tissue irre-
spective of the time elapsed since the vascular attack,
and irrespective of the variations in blood flow
over time.
Central benzodiazepine receptor (BZR) ligands
can be used as markers of neuronal integrity as
they bind to the gamma-aminobutyric acid (GABA)

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 17 Apr 2018 at 16:27:57, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781107239340.002
 Chapter 1: Neuropathology and pathophysiology of stroke

receptors abundant in cerebral cortex that are sensi-
tive to ischemic damage. After successful testing in
the cat MCA occlusion model, cortical binding of
flumazenil (FMZ) was investigated in patients with
acute ischemic stroke [119]. In all patients, defects
in FMZ binding were closely related to areas with
severely depressed oxygen consumption and pre-
dicted the size of the final infarcts, whereas preserved
FMZ binding indicated intact cortex. Additionally,
FMZ distribution within 2 minutes after tracer injec-
tion was highly correlated with CBF measured by
H2O and therefore can be used as a relative flow
tracer yielding reliable perfusion images. PET with
FMZ therefore can be used as a non-invasive proce-
dure to image irreversible damage and critically
reduced perfusion (i.e. penumbra) in early ischemic
stroke.
Special tracers, e.g. ligands to benzodiazepine
receptors, can be used as early markers of irreversible
neuronal damage, since they bind to GABA receptors
of cortical neurons, which are very sensitive to
ischemic damage.

Figure 1.12. (a, b) Volumetric comparison of time to peak (TTP) (MRI) and OEF (PET) images in two patients measured in the chronic
phase of stroke. In both patients a TTP delay of >4 seconds indicates a considerable mismatch volume (red contour on TTP images).
The mismatch volumes were 473 cm3 for patient a and 199.7 cm3 for patient b. However, only patient b had a corresponding volume
of penumbra (260 cm3). (c) Volumes of penumbra defined by increased OEF (black) and mismatch defined by TTP > 4 (gray) in 13 patients:
all 13 patients showed mismatch, only 8 patients showed penumbra, which comprised 1–75% of the mismatch volume. (Modified with 25
permission from Sobesky et al. [124].)

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 17 Apr 2018 at 16:27:57, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781107239340.002
 Section 1: Etiology, pathophysiology, and imaging
Figure 1.12. (cont.)
Surrogate markers for penumbra and
irreversible damage
Although PET remains the imaging gold standard for
identification of the penumbra in stroke patients, MR
studies using DWI and perfusion-weighted imaging
(PWI) might provide a differentiation between the
core and the penumbra: the early DWI lesion might
define the ischemic core and adjacent critically hypo-
perfused tissue might be identified with PWI [120].
Therefore, brain regions with hypoperfusion assessed
by PWI but without restricted diffusion (PWI/DWI
mismatch) were assumed to represent the penumbra.
This surrogate definition of the penumbra has several
uncertainties [121]: the initial diffusion lesion does
not only consist of irreversibly infarcted tissue; diffu-
sion lesions may be reversed if blood flow is restored
at an early time point; critically perfused tissue (i.e.
penumbra) cannot be clearly differentiated from
tissue experiencing benign oligemia; the PWI abnor-
mality often overestimates the amount of tissue at
risk. These facts are further accentuated by methodo-
logical limitations, because perfusion techniques and
data evaluation are not quantitative and vary among
26 centers. Despite the attempts to correct the mismatch
area by subtracting the region of benign oligemia and
Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 17 Apr 2018 at 16:27:57, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781107239340.002
adding the portion of reversible diffusion abnormality
[121] the definition of the penumbra remains contro-
versial and suffers from lack of standardization of
methodological approaches to imaging, post process-
ing, and analysis, which restricts pooling of data and
cross-comparison of results across studies.
A validation of PWI/DWI results with quantitative
measurements of flow values and oxygen consumption
or FMZ uptake in the same patients early after stroke
is necessary for the assessment of the accuracy of
the applied signatures for predicting tissue outcome.
Several studies were performed in order to validate
mismatch as a surrogate of penumbra on PET-derived
discrimination of irreversibly damaged, critically per-
fused “at risk,” and oligemic “not at risk” tissue. The
studies demonstrated that the DWI lesion predicts
more or less the finally infarcted tissue [122], but
contains up to 25% false positive, i.e. surviving tissue.
The inaccuracy in defining the penumbra with
PWI/DWI mismatch is thought to be mainly related
to PW data acquisition, since the parameters used to
estimate perfusion are variable and somewhat arbi-
trary. As a consequence, perfusion lesion size differs
markedly depending on the parameters calculated
[123] and usually is overestimated and extends into
considerable areas with non-critical oligemia especially

when short delays are used. Overall, PWI is unable to
provide a reliable quantitative estimation of cerebal
perfusion when compared to gold-standards such as
PET, SPECT, or Xe-CT and overestimated the size of
the critically perfused tissue and therefore also the
volume of critically perfused but salvageable tissue,
i.e. the penumbra (Figure 1.12) [124]. Of 13 patients
showing considerable PWI/DWI mismatch only
8 had areas with elevated OEF typical for penumbra
tissue, and these areas were always smaller than MR
mismatch. Overall, the mismatch volume in PWI/
DWI as conventionally calculated does not reliably
reflect misery perfusion, i.e. the penumbra as defined
by PET. Recently, several methods have been pro-
posed to improve the reliability of assessment of
perfusion using MR methods [125], but they all need
to be validated by quantitative measures.
The mismatch between perfusion-weighted and
diffusion-weighted MRI can be used as a surrogate
marker of the penumbra, but the applied parameters
need to be validated.
Chapter summary
Atherosclerosis is the most widespread disorder
leading to death and serious morbidity including
stroke. It develops over years from initial fatty streaks
to atheromatous plaques with the risk of plaque
disruption and formation of thrombus, from which
emboli might originate. Lipohyalinosis affects small
vessels, leading to lacunar stroke. The vascular
lesions and emboli from the heart cause territorial
infarcts, whereas borderzone infarcts are due to low
perfusion in the peripheral parts of the vascular ter-
ritories (last meadows). Ischemic infarcts may be
converted into hemorrhagic infarctions by leakage
of vessels, whereas intracerebral hemorrhages
(5–15% of all strokes) result from rupture of arteries
typically in deep portions of the hemispheres.
Venous infarcts usually result from thrombosis of
References
1. Sanz J, Moreno PR, Fuster V.
The year in atherothrombosis.
J Am Coll Cardiol 2012;
60:932–42.
3.
2. Rajamani K, Fisher M, Fisher M.
Atherosclerosis – pathogenesis
and pathophysiology. In:
Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 17 Apr 2018 at 16:27:57, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781107239340.002
Chapter 1: Neuropathology and pathophysiology of stroke
sinuses or veins and are often accompanied by
edema, hemorrhagic transformation, and bleeding.
Primary ischemic cell death is the result of severe
ischemia; early signs are potentially reversible swell-
ing or shrinkage; irreversible necrotic neurons have
condensed acidophilic cytoplasm and pyknotic
nuclei. Delayed neuronal death can occur after
moderate or short-term ischemia; it goes along
with nuclear fragmentation and development of
apoptotic bodies. The pathophysiology of ischemic
cell damage in animal models reflects only certain
aspects of ischemia and cannot give a complete
picture of ischemic stroke in humans. However, from
these experimental models principles of regulation of
cerebral blood flow and flow thresholds for mainten-
ance of function and morphology can be deduced.
As the energy requirement of the brain is very high,
already mild decreases of blood supply lead to
potentially reversible disturbance of function and,
if the shortage is more severe and persists for cer-
tain periods, to irreversible morphological damage.
Tissue perfused in the range between the thresholds
of functional and morphological injury has been
called the penumbra, a concept which has great
importance for treatment. The ischemia-induced
energy failure triggers a complex cascade of electro-
physiological disturbances, biochemical changes,
and molecular dysfunction, which lead to progres-
sive cell death and growth of infarction. The pro-
gression of ischemic injury is further boosted by
inflammatory reactions and the development of
early cytotoxic and later vasogenic brain edema.
The translation of these experimental concepts into
clinical application and management of stroke
patients, however, is difficult. It can be achieved
in some instances by special functional imaging
techniques, such as positron emission tomography,
but requires further refinements to predict the
outcome of neuroprotective interventions. Solid
understanding of experimental and clinical stroke
pathophysiology is, therefore, needed to improve
the reliability of translational research.
Ginsberg MD, Bogousslavsky J, 4. Ross R. Atherosclerosis–an
eds. Cerebrovascular Disease: inflammatory disease. N Engl
Pathophysiology, Diagnosis and J Med 1999; 340:115–26.
Management. London: Blackwell
5. Faxon DP, Fuster V, Libby P, et al.
Science; 1998: 308–18. Atherosclerotic Vascular Disease
Willeit J, Kiechl S. Biology of Conference: Writing Group III:
arterial atheroma. Cerebrovas Dis pathophysiology. Circulation
2000; 10(Suppl 5):1–8. 2004; 109:2617–25. 27
 Section 1: Etiology, pathophysiology, and imaging
6. Aikawa M, Libby P. The
vulnerable atherosclerotic plaque:
pathogenesis and therapeutic
approach. Cardiovasc Pathol 2004;
13:125–38.
7. Koga J, Aikawa M. Crosstalk
between macrophages and
smooth muscle cells in
atherosclerotic vascular
diseases. Vascul Pharmacol
2012; 57:24–8.
8. Dzau VJ, Braun-Dullaeus RC,
Sedding DG. Vascular
proliferation and atherosclerosis:
new perspectives and therapeutic
strategies. Nat Med 2002;
8:1249–56.
9. Madden JA. Role of the vascular
endothelium and plaque in acute
ischemic stroke. Neurology 2012;
79:S58–62.
10. Glagov S, Weisenberg E, Zarins
CK, Stankunavicius R, Kolettis GJ.
Compensatory enlargement of
human atherosclerotic coronary
arteries. N Engl J Med 1987;
316:1371–5.
11. Rauch U, Osende JI, Fuster V,
et al. Thrombus formation on
atherosclerotic plaques:
pathogenesis and clinical
consequences. Ann Intern Med
2001; 134:224–38.
12. Lubos E, Handy DE, Loscalzo J.
Role of oxidative stress and nitric
oxide in atherothrombosis. Front
Biosci 2008; 13:5323–44.
13. Fisher CM. Cerebral miliary
aneurysms in hypertension.
Am J Pathol 1972; 66:
313–30.
14. Zülch K-J. The Cerebral Infarct.
Pathology, Pathogenesis, and
Computed Tomography. Berlin,
Heidelberg, New York, Tokyo:
Springer-Verlag; 1985.
15. Russell RWR. Observations on
intracerebral aneurysms. Brain
1963; 86:425–42.
16. Zülch K-J. Über die Entstehung
und Lokalisation der Hirninfarkte.
28 Zentralbl Neurochir 1961;
21:158–78.
Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 17 Apr 2018 at 16:27:57, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781107239340.002
17. Dichgans M. Genetics of
ischaemic stroke. Lancet Neurol
2007; 6:149–61.
18. Mohr JP, Choi DW, Grotta JC,
Weir B, Wolf PA (eds.). Stroke –
Pathophysiology, Diagnosis, and
Management, 4th edn.
Philadelphia: Churchill
Livingstone; 2004.
19. Pantoni L. Cerebral small vessel
disease: from pathogenesis and
clinical characteristics to
therapeutic challenges. Lancet
Neurol 2010; 9:689–701.
20. Ringelstein EB, Nabavi DG.
Cerebral small vessel diseases:
cerebral microangiopathies.
Curr Opin Neurol 2005;
18:179–88.
21. Wolf PA. Epidemiology of
stroke. In: Mohr JP, Choi DW,
Grotta JC, Weir B, Wolf PA,
eds. Stroke – Pathophysiology,
Diagnosis, and Management,
4th edn. Philadelphia:
Churchill Livingstone; 2004:
13–34.
22. Stochdorph O. Der Mythos
der letzten Wiese. Zentralbl
Allg Pathol Path Anat 1977;
121:554.
23. Ringelstein EB, Zunker P. Low-
flow infarction. In: Ginsberg MD,
Bogousslavsky J, eds.
Cerebrovascular Disease:
Pathophysiology, Diagnosis, and
Management. London: Blackwell
Science; 1998: 1075–89.
24. Fisher CM. Lacunes: small, deep
cerebral infarcts. Neurology 1965;
15:774–84.
25. Amarenco P, Bogousslavsky J,
Caplan LR, Donnan GA,
Hennerici MG. New approach to
stroke subtyping: the A-S-C-O
(phenotypic) classification of
stroke. Cerebrovasc Dis 2009;
27:502–8.
26. Beghi E, Bogliun G, Cavaletti G,
et al. Hemorrhagic infarction: risk
factors, clinical and tomographic
features, and outcome. A case-
control study. Acta Neurol Scand
1989; 80:226–31.
27. Lodder J, Krijne-Kubat B,
Broekman J. Cerebral
hemorrhagic infarction at
autopsy: cardiac embolic cause
and the relationship to the cause
of death. Stroke 1986; 17:626–9.
28. Fisher M, Adams RD.
Observations on brain embolism
with special reference to the
mechanism of hemorrhagic
infarction. J Neuropathol Exp
Neurol 1951; 10:92–4.
29. Mohr JP, Caplan LR, Melski JW,
et al. The Harvard Cooperative
Stroke Registry: a prospective
registry. Neurology 1978; 28:754–62.
30. Sacco RL, Wolf PA, Bharucha NE,
et al. Subarachnoid and
intracerebral hemorrhage: natural
history, prognosis, and precursive
factors in the Framingham Study.
Neurology 1984; 34:847–54.
31. Feldman E. Intracerebral
Hemorrhage. Armonk, New York:
Futura; 1994.
32. Schütz H. Spontane intrazerebrale
Hämatome. Pathophysiologie,
Klinik und Therapie. Berlin,
Heidelberg, New York: Springer-
Verlag; 1988.
33. Qureshi AI, Tuhrim S, Broderick
JP, et al. Spontaneous
intracerebral hemorrhage. N Engl
J Med 2001; 344:1450–60.
34. Kase CS, Mohr JP, Caplan LR.
Intracerebral hemorrhage. In:
Mohr JP, Choi DW, Grotta JC,
Weir B, Wolf PA, eds. Stroke –
Pathophysiology, Diagnosis, and
Management, 4th edn.
Philadelphia: Churchill
Livingstone; 2004: 327–76.
35. Fisher CM. Pathological
observations in hypertensive
cerebral hemorrhage.
J Neuropathol Exp Neurol 1971;
30:536–50.
36. Brott T, Broderick J, Kothari R,
et al. Early hemorrhage growth in
patients with intracerebral
hemorrhage. Stroke 1997; 28:1–5.
37. Gonzalez-Duarte A, Cantu C,
Ruiz-Sandoval JL,

Barinagarrementeria F. Recurrent
primary cerebral hemorrhage:
frequency, mechanisms, and
prognosis. Stroke 1998; 29:
1802–5.
38. Bousser MG, Barnett HJM.
Cerebral venous thrombosis. In:
Mohr JP, Choi DW, Grotta JC,
Weir B, Wolf PA, eds. Stroke –
Pathophysiology, Diagnosis, and
Management, 4th edn.
Philadelphia: Churchill
Livingstone; 2004: 301–25.
39. Petito CK. The Neuropathology of
Focal Brain Ischemia. Basel: ISN
Neuropath; 2005.
40. Brown AW, Brierley JB. Anoxic-
ischaemic cell change in rat brain
light microscopic and fine-
structural observations. J Neurol
Sci 1972; 16:59–84.
41. Jortner BS. The return of the dark
neuron. A histological artifact
complicating contemporary
neurotoxicologic evaluation.
Neurotoxicology 2006; 27:628–34.
42. Kirino T, Sano K. Selective
vulnerability in the gerbil
hippocampus following transient
ischemia. Acta Neuropathol 1984;
62:201–8.
43. Martin LJ. Neuronal cell death in
nervous system development,
disease, and injury (Review). Int
J Mol Med 2001; 7:455–78.
44. Charriaut-Marlangue C, Ben-Ari
Y. A cautionary note on the use of
the TUNEL stain to determine
apoptosis. Neuroreport 1995;
7:61–4.
45. Hossmann KA. Disturbances of
cerebral protein synthesis and
ischemic cell death. Prog Brain Res
1993; 96:161–77.
46. DeGracia DJ, Jamison JT,
Szymanski JJ, Lewis MK.
Translation arrest and ribonomics
in post-ischemic brain: layers and
layers of players. J Neurochem
2008; 106:2288–301.
47. Stanimirovic DB, Friedman A.
Pathophysiology of the
neurovascular unit: disease cause
Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 17 Apr 2018 at 16:27:57, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781107239340.002
Chapter 1: Neuropathology and pathophysiology of stroke
or consequence? J Cereb Blood
Flow Metab 2012; 32:1207–21.
48. Hudgins WR, Garcia JH.
Transorbital approach to the
middle cerebral artery of the
squirrel monkey: a technique for
experimental cerebral infarction
applicable to ultrastructural
studies. Stroke 1970; 1:107–11.
49. Tamura A, Graham DI,
McCulloch J, Teasdale GM. Focal
cerebral ischaemia in the rat: 1.
Description of technique and early
neuropathological consequences
following middle cerebral artery
occlusion. J Cereb Blood Flow
Metab 1981; 1:53–60.
50. Koizumi J, Yoshida Y, Nakazawa
T, Ooneda G. Experimental
studies of ischemic brain
edema. 1. A new experimental
model of cerebral embolism
in rats in which recirculation
can be introduced in the
ischemic area. Jpn J Stroke 1986;
8:1–8.
51. Rogers DC, Campbell CA,
Stretton JL, Mackay KB.
Correlation between motor
impairment and infarct volume
after permanent and transient
middle cerebral artery occlusion
in the rat. Stroke 1997; 28:2060–5;
discussion 2066.
52. Hossmann KA. The two
pathophysiologies of focal brain
ischemia: implications for
translational stroke research.
J Cereb Blood Flow Metab 2012;
32:1310–16.
53. DiNapoli VA, Rosen CL,
Nagamine T, Crocco T. Selective
MCA occlusion: a precise embolic
stroke model. J Neurosci Methods
2006; 154:233–8.
54. Orset C, Macrez R, Young AR, et al.
Mouse model of in situ
thromboembolic stroke and
reperfusion. Stroke 2007; 38:2771–8.
55. Chen F, Suzuki Y, Nagai N, et al.
Rodent stroke induced by
photochemical occlusion of
proximal middle cerebral artery:
evolution monitored with MR
imaging and histopathology. Eur
J Radiol 2007; 63:68–75.
56. Mehra M, Henninger N, Hirsch
JA, et al. Preclinical acute
ischemic stroke modeling.
J Neurointerv Surg 2012; 4:307–13.
57. Mordasini P, Frabetti N, Gralla J,
et al. In vivo evaluation of the first
dedicated combined flow-
restoration and mechanical
thrombectomy device in a swine
model of acute vessel occlusion.
AJNR Am J Neuroradiol 2011;
32:294–300.
58. Koller A, Toth P. Contribution of
flow-dependent vasomotor
mechanisms to the autoregulation
of cerebral blood flow. J Vasc Res
2012; 49:375–89.
59. Symon L. Regional vascular
reactivity in the middle cerebral
arterial distribution. An
experimental study in baboons.
J Neurosurg 1970; 33:532–41.
60. Hata R, Maeda K, Hermann D,
Mies G, Hossmann KA. Evolution
of brain infarction after transient
focal cerebral ischemia in mice.
J Cereb Blood Flow Metab 2000;
20:937–46.
61. del Zoppo GJ. The neurovascular
unit in the setting of stroke.
J Intern Med 2010; 267:156–71.
62. Gursoy-Ozdemir Y, Yemisci M,
Dalkara T. Microvascular
protection is essential for
successful neuroprotection in
stroke. J Neurochem 2012; 123
(Suppl 2):2–11.
63. Liebeskind DS. Collateral
circulation. Stroke 2003;
34:2279–84.
64. Toole JF, McGraw CP. The steal
syndromes. Annu Rev Med 1975;
26:321–9.
65. Pakkenberg B, Gundersen HJ.
Neocortical neuron number in
humans: effect of sex and age.
J Comp Neurol 1997; 384:312–20.
66. Clarke DD, Sokoloff L.
Circulation and energy
metabolism of the brain. In: Siegel 29
G, Agranoff B, Albers RW, Fisher
 Section 1: Etiology, pathophysiology, and imaging
S, eds. Basic Neurochemistry:
Molecular, Cellular, and Medical
Aspects, 6th edn. Philadelphia:
Lippincott-Raven; 1999: 637–69.
67. Sokoloff L. Energetics of functional
activation in neural tissues.
Neurochem Res 1999; 24:321–9.
68. Magistretti PJ, Pellerin L.
Astrocytes couple synaptic activity
to glucose utilization in the brain.
News Physiol Sci 1999; 14:177–82.
69. Laughlin SB, Attwell D. The
metabolic cost of neural
information: from fly eye to
mammalian cortex. In:
Frackowiak RSJ, Magistretti PJ,
Shulman RG, Altman JS, Adams
M, eds. Neuroenergetics: Relevance
for Functional Brain Imaging.
Strasbourg: HFSP Workshop XI;
2001: 54–64.
70. Frackowiak RSJ, Magistretti PJ,
Shulman RG, Altman JS, Adams
M (eds.). Neuroenergetics:
Relevance for Functional Brain
Imaging. Strasbourg: HFSP
Workshop XI; 2001.
71. Astrup J, Siesjö BK, Symon L.
Thresholds in cerebral ischemia –
the ischemic penumbra. Stroke
1981; 12:723–5.
72. Heiss WD. Experimental evidence
of ischemic thresholds and
functional recovery. Stroke 1992;
23:1668–72.
73. Hossmann KA. Viability
thresholds and the penumbra of
focal ischemia. Ann Neurol 1994;
36:557–65.
74. Heiss W-D, Hayakawa T, Waltz
AG. Cortical neuronal function
during ischemia. Effects of
occlusion of one middle cerebral
artery on single-unit activity in
cats. Arch Neurol 1976; 33:813–20.
75. Hossmann KA. Cerebral
ischemia: models, methods and
outcomes. Neuropharmacology
2008; 55:257–70.
76. Hata R, Maeda K, Hermann D,
Mies G, Hossmann KA. Dynamics
30 of regional brain metabolism and
gene expression after middle
Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 17 Apr 2018 at 16:27:57, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781107239340.002
cerebral artery occlusion in mice.
J Cereb Blood Flow Metab 2000;
20:306–15.
77. Heiss WD. Ischemic penumbra:
evidence from functional imaging
in man. J Cereb Blood Flow Metab
2000; 20:1276–93.
78. Takasawa M, Beech JS, Fryer TD,
et al. Imaging of brain hypoxia in
permanent and temporary middle
cerebral artery occlusion in the rat
using 18F-fluoromisonidazole and
positron emission tomography: a
pilot study. J Cereb Blood Flow
Metab 2007; 27:679–89.
79. Kane I, Sandercock P, Wardlaw J.
Magnetic resonance perfusion
diffusion mismatch and
thrombolysis in acute ischaemic
stroke: a systematic review of the
evidence to date. J Neurol Neurosurg
Psychiatry 2007; 78:485–91.
80. Hoehn-Berlage M, Norris DG,
Kohno K, et al. Evolution of
regional changes in apparent
diffusion-coefficient during focal
ischemia of rat brain: the
relationship of quantitative
diffusion NMR imaging to
reduction in cerebral blood flow
and metabolic disturbances.
J Cereb Blood Flow Metab 1995;
15:1002–11.
81. Sun PZ, Zhou J, Sun W, Huang J,
van Zijl PC. Detection of the
ischemic penumbra using pH-
weighted MRI. J Cereb Blood Flow
Metab 2007; 27:1129–36.
82. Cheung JS, Wang E, Lo EH, Sun
PZ. Stratification of
heterogeneous diffusion MRI
ischemic lesion with kurtosis
imaging: evaluation of mean
diffusion and kurtosis MRI
mismatch in an animal model of
transient focal ischemia. Stroke
2012; 43:2252–4.
83. Kidwell CS, Heiss WD. Advances
in stroke: imaging. Stroke 2012;
43:302–4.
84. del Zoppo GJ, Sharp FR, Heiss
WD, Albers GW. Heterogeneity in
the penumbra. J Cereb Blood Flow
Metab 2011; 31:1836–51.
85. Dreier JP. The role of spreading
depression, spreading
depolarization and spreading
ischemia in neurological
disease. Nat Med 2011; 17:
439–47.
86. Strong AJ, Anderson PJ, Watts
HR, et al. Peri-infarct
depolarizations lead to loss of
perfusion in ischaemic
gyrencephalic cerebral cortex.
Brain 2007; 130:995–1008.
87. Mies G, Iijima T, Hossmann K-A.
Correlation between periinfarct
DC shifts and ischemic neuronal
damage in rat. Neuroreport 1993;
4:709–11.
88. Moskowitz MA, Lo EH, Iadecola
C. The science of stroke:
mechanisms in search of
treatments. Neuron 2010;
67:181–98.
89. Tymianski M. Emerging
mechanisms of disrupted cellular
signaling in brain ischemia. Nat
Neurosci 2011; 14:1369–73.
90. Simon R, Xiong Z. Acidotoxicity
in brain ischaemia. Biochem Soc
Trans 2006; 34:1356–61.
91. Choi DW. Excitotoxicity,
apoptosis and ischemic stroke.
J Biochem Mol Biol 2001; 34:8–14.
92. Szydlowska K, Tymianski M.
Calcium, ischemia and
excitotoxicity. Cell Calcium 2010;
47:122–9.
93. MacDonald JF, Xiong ZG, Jackson
MF. Paradox of Ca2+ signaling,
cell death and stroke. Trends
Neurosci 2006; 29:75–81.
94. Chan PH. Reactive oxygen
radicals in signaling and damage
in the ischemic brain. J Cereb
Blood Flow Metab 2001; 21:2–14.
95. Shuaib A, Lees KR, Lyden P, et al.
NXY-059 for the treatment of
acute ischemic stroke. N Engl
J Med 2007; 357:562–71.
96. Dalkara T, Moskowitz MA. The
complex role of nitric oxide in the
pathophysiology of focal cerebral
ischemia. Brain Pathol 1994;
4:49–57.

97. Shuttleworth CW, Weiss JH. Zinc:
new clues to diverse roles in brain
ischemia. Trends Pharmacol Sci
2011; 32:480–6.
98. Sensi SL, Jeng JM. Rethinking the
excitotoxic ionic milieu: the
emerging role of Zn(2+) in
ischemic neuronal injury. Curr
Mol Med 2004; 4:87–111.
99. Paschen W. Endoplasmic
reticulum dysfunction in brain
pathology: critical role of protein
synthesis. Curr Neurovasc Res
2004; 1:173–81.
100. Youle RJ, van der Bliek AM.
Mitochondrial fission, fusion, and
stress. Science 2012; 337:1062–5.
101. Gouriou Y, Demaurex N, Bijlenga
P, De Marchi U. Mitochondrial
calcium handling during
ischemia-induced cell death in
neurons. Biochimie 2011;
93:2060–7.
102. Norenberg MD, Rao KV. The
mitochondrial permeability
transition in neurologic disease.
Neurochem Int 2007; 50:983–97.
103. Rothwell NJ, Luheshi GN.
Interleukin 1 in the brain: biology,
pathology and therapeutic
target. Trends Neurosci 2000;
23:618–25.
104. Planas AM, Gorina R, Chamorro
A. Signalling pathways mediating
inflammatory responses in brain
ischaemia. Biochem Soc Trans
2006; 34:1267–70.
105. Wang CX, Shuaib A. Critical role
of microvasculature basal lamina
in ischemic brain injury. Prog
Neurobiol 2007; 83:140–8.
106. Walz B, Zimmermann C, Bottger
S, Haberl RL. Prognosis of
patients after hemicraniectomy in
malignant middle cerebral artery
infarction. J Neurol 2002;
249:1183–90.
107. Lansberg MG, Thijs VN, O’Brien
MW, et al. Evolution of apparent
diffusion coefficient, diffusion-
weighted, and T2-weighted signal
intensity of acute stroke. AJNR Am
J Neuroradiol 2001; 22:637–44.
Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 17 Apr 2018 at 16:27:57, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781107239340.002
Chapter 1: Neuropathology and pathophysiology of stroke
108. Badaut J, Lasbennes F, Magistretti
PJ, Regli L. Aquaporins in brain:
distribution, physiology, and
pathophysiology. J Cereb Blood
Flow Metab 2002; 22:367–78.
109. Kerr JF, Wyllie AH, Currie AR.
Apoptosis: a basic biological
phenomenon with wide-ranging
implications in tissue kinetics. Br
J Cancer 1972; 26:239–57.
110. Niizuma K, Yoshioka H, Chen H,
et al. Mitochondrial and apoptotic
neuronal death signaling
pathways in cerebral ischemia.
Biochim Biophys Acta 2010;
1802:92–9.
111. MacManus JP, Buchan AM.
Apoptosis after experimental
stroke: fact or fashion?
J Neurotrauma 2000; 17:899–914.
112. Dirnagl U, Simon RP, Hallenbeck
JM. Ischemic tolerance and
endogenous neuroprotection.
Trends Neurosci 2003; 26:248–54.
113. Zhao H, Sapolsky RM, Steinberg
GK. Interrupting reperfusion as a
stroke therapy: ischemic
postconditioning reduces infarct
size after focal ischemia in rats.
J Cereb Blood Flow Metab 2006;
26:1114–21.
114. Wiltrout C, Lang B, Yan Y,
Dempsey RJ, Vemuganti R.
Repairing brain after stroke: a
review on post-ischemic
neurogenesis. Neurochem Int
2007; 50:1028–41.
115. Kamouchi M, Ago T, Kuroda J,
Kitazono T. The possible roles of
brain pericytes in brain ischemia
and stroke. Cell Mol Neurobiol
2012; 32:159–65.
116. Abe K, Yamashita T, Takizawa S,
et al. Stem cell therapy for cerebral
ischemia: from basic science to
clinical applications. J Cereb
Blood Flow Metab 2012; 32:
1317–31.
117. Kuhl DE, Phelps ME, Kowell AP,
et al. Effects of stroke on local
cerebral metabolism and
perfusion: mapping by emission
computed tomography of 18FDG
and 13NH3. Ann Neurol 1980;
8:47–60.
118. Baron JC, Frackowiak RS, Herholz
K, et al. Use of PET methods for
measurement of cerebral energy
metabolism and hemodynamics in
cerebrovascular disease. J Cereb
Blood Flow Metab 1989; 9:723–42.
119. Heiss WD, Grond M, Thiel A,
et al. Permanent cortical damage
detected by flumazenil positron
emission tomography in acute
stroke. Stroke 1998; 29:454–61.
120. Baird AE, Benfield A, Schlaug G,
et al. Enlargement of human
cerebral ischemic lesion volumes
measured by diffusion-weighted
magnetic resonance imaging. Ann
Neurol 1997; 41:581–9.
121. Kidwell CS, Alger JR, Saver JL.
Beyond mismatch: evolving
paradigms in imaging the
ischemic penumbra with
multimodal magnetic resonance
imaging. Stroke 2003; 34:2729–35.
122. Heiss WD, Sobesky J, Smekal U,
et al. Probability of cortical
infarction predicted by flumazenil
binding and diffusion-weighted
imaging signal intensity: a
comparative positron emission
tomography/magnetic resonance
imaging study in early ischemic
stroke. Stroke 2004; 35:1892–8.
123. Kane I, Carpenter T, Chappell F,
et al. Comparison of 10 different
magnetic resonance perfusion
imaging processing methods in
acute ischemic stroke: effect on
lesion size, proportion of patients
with diffusion/perfusion
mismatch, clinical scores, and
radiologic outcomes. Stroke 2007;
38:3158–64.
124. Sobesky J, Weber OZ, Lehnhardt
FG, et al. Does the mismatch
match the penumbra? Magnetic
resonance imaging and positron
emission tomography in early
ischemic stroke. Stroke 2005;
36:980–5.
125. Olivot JM, Albers GW. Diffusion-
perfusion MRI for triaging 31
transient ischemic attack and
 Section 1: Etiology, pathophysiology, and imaging

acute cerebrovascular syndromes.
Curr Opin Neurol 2011; 24:44–9.
126. Garcia JH, Liu KF, Ho KL.
Neuronal necrosis after middle
cerebral artery occlusion in Wistar
rats progresses at different time
intervals in the caudoputamen
and the cortex. Stroke 1995;
26:636–42.
127. Magistretti PJ. Coupling synaptic
activity to glucose metabolism. In:
Frackowiak RSJ, Magistretti PJ,
Shulman RG, Altman JS, Adams
M, eds. Neuroenergetics: Relevance
for Functional Brain Imaging.
Strasbourg: HFSP Workshop XI;
2001: 133–42.
128. Attwell D, Laughlin SB. An energy
budget for signaling in the grey
matter of the brain. J Cereb Blood
Flow Metab 2001; 21:1133–45.
129. Heiss WD, Rosner G. Functional
recovery of cortical neurons as
related to degree and duration of
ischemia. Ann Neurol 1983;
14:294–301.
130. Hossmann KA, Mies G.
Multimodal Mapping of the
Ischemic Penumbra in Animal
Models. New York: Marcel
Dekker, Inc.; 2007.
131. Hossmann KA. Periinfarct
depolarizations. Cerebrovasc
Brain Metab Rev 1996; 8:
195–208.
132. Hossmann KA. Pathophysiology
and therapy of experimental
stroke. Cell Mol Neurobiol 2006;
26:1057–83.

32

Downloaded from https://www.cambridge.org/core. University of Sussex Library, on 17 Apr 2018 at 16:27:57, subject to the Cambridge Core terms of use, available at
https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781107239340.002