вход по аккаунту


Focal cerebral ischemia in the rat Topography of hemodynamic and histopathological changes.

код для вставкиСкачать
Focal Cerebral Ischemia in the Rat:
Topography of Hernodynamic and
Histopathological Changes
George W. Tyson, MD,*§ Graham M. Teasdale, MRCP, FRCS,? David I. Graham, PhD, FRCPathJ
and James McCulloch, PhD"
We studied local cerebral blood flow, as measured by autoradiography with digital image processing and by tissue
morphology, in six rats 4 hours after occlusion of the proximal middle cerebral artery. A consistent, three-dimensional
pattern of graded reductions in local cerebral blood flow involved the affected hemisphere, with a densely ischemic
zone (local cerebral blood flow less than 3 mMOO g d m i n ) in the dorsolateral caudate putamen and the adjacent
frontoparietal cortex. In the frontoparietal cortex, the normal laminar pattern of local cerebral blood flow was disrupted, and there was a transcortical gradient in flow, with pronounced ischemia in deeper layers and relatively
preserved superficial flow. Comparisons of autoradiographic findings with histopathological abnormalities in adjacent
frozen sections showed that the region of ischemic damage corresponded closely with the area of greatest reduction in
blood flow. Although around this region local cerebral blood flow increased centrifugally, a striking finding was that
flow density changed abruptly (a tenfold variation in flow within a 1 to 2 mm interval) at the edge of the pathological
lesion. Penumbral conditions may therefore exist in only a very narrow zone 4 hours after onset of focal ischemia. After
occlusion of a major cerebral artery, the pattern of local cerebral blood flow changes appears to depend on interactions
among vascular architecture, reductions in perfusion pressure, alterations in metabolic demands, and variations in local
vascular resistance.
Tyson GW, Teasdale GM, Graham DI, McCulloch J: Focal cerebral ischemia in the rat: topography of
hemodynamic and histopathological changes. Ann Neurol 15:559-567, 1984
Materials and Methods
Detailed maps of blood flow levels are needed to discover more about precise patterns of alterations in tissue perfusion in focal cerebral ischemia. These patterns
provide the basis for understanding the distribution of
certain electrophysiological r3, 51 and biochemical [I,
3, 41 derangements, the development of which depends on the density of residual local cerebral blood
flow (ICBF) and which foreshadow the development of
structural ischemic brain damage Y2, 12, 151.
In the study reported here, high-resolution flow density maps were produced by a scanning microdensitometer and digital image processing system {91 to analyze
cerebral autoradiographs from rats in which one middle
cerebral artery had been occluded 4 hours prior to
tracer injection. We also determined the distribution of
histopathological alterations in the brains of the same
animals to corroborate the existence of an ischemic
lesion in each animal and to identify the relationship
between its anatomy and patterns of cerebral blood
Experiments were performed on six adult male SpragueDawley rats weighing 340 to 385 gm. Tracheostomy was
performed with the rats anesthetized by 2% halothane, after
which the animals were paralyzed by administration of gallamine triethiodide, 20 mg per kg of body weight, intraperitoneally, and mechanically ventilated with a nitrous oxide-oxygen-halothanemixture (70:30:0.5%). Cannulas were
inserted into the femoral arteries and veins to monitor
mean arterial blood pressure, sample arterial blood, and administer radioactive tracer. Normothermia ( 37°C)was maintained by a heating system regulated by rectal temperature.
Arterial blood gases were measured after middle cerebral
artery occlusion and immediately prior to tracer administration. Ventilation was adjusted to maintain arterial carbon
dioxide tension (PaCO2) between 32 and 40 mmHg. Animals with an arterial oxygen tension (PaOz)of less than 80
mmHg at any time were discarded.
Following subtemporal craniectomy, the main trunk of the
left middle cerebral artery was focally electrocoagulated just
medial to the olfactory tract, a point proximal to the origin of
the most lateral lenticulostriate branches [ l 6 ] .
From the 'Wellcome Surgical Institute, and the Departments of
tNeurosurgery and fNeuropathology, University of Glasgow, Scotland.
Address reprint requests to Dr Teasdale, Department of
Neurosurgery, Institute of Neurological Sciences, Southern General
Hospital, Glasgow, G51 4TF, Scotland.
Received Apr 14, 1983, and in revised form Oct 31. Accepted for
publication Nov 13, 1983.
$Present address: Department of Neurological Surgery, School of
Medicine, Health Sciences Center, T12-080, State University of
New York at Stony Brook, Stony Brook, NY 11794.
Autoradiography and lCBF Measurements
Local cerebral blood flow was measured by the [“C)iodoantipyrine quantitative autoradiographic technique described
by Sakurada and colleagues { 131. After infusion of the tracer,
multiple timed samples of arterial blood were collected from
one arterial catheter, and mean arterial blood pressure was
simultaneously recorded from the other. The animals were
decapitated and their brains removed, rapidly frozen in isopentane (cooled to - 45”C), and then sectioned (20 pm) in a
cryostat ( - 22°C). Six consecutive coronal sections were processed for autoradiography 117). The next 2 were reserved
for examination by light microscopy, and the following 12
were discarded. This sequence was repeated until the entire
cerebrum had been sectioned.
Autoradiographs obtained by exposing the brain sections
to radiographic film for 7 days were processed according to
the method described by Goochee and co-workers 19). By
this method, a computer-coupled scanning microdensitometer and a digital image-processing system translated the digitized optical density measurements into equivalent tissue
tracer concentrations; then the operational equation was
solved for the iodoantipyrine method [ 131. The tissue blood
flow values were obtained after the blood-brain partition coefficient for iodoantipyrine (0.8) was entered, and the time
course of the arterial tracer concentration was color coded
according to a preselected series of blood flow ranges and
displayed on a color monitor. The pseudocolor reconstruction of ICBF in the section was then photographed directly
from the video monitor.
Three representative coronal sections through the caudate
putamen were studied in detail. The patterns of optical density (which does not relate linearly to measured cerebral
blood flow) observed in the original autoradiographs were
compared with data from the photographs of the pseudocolor
transformations. The three sections and their approximate
stereotaxic planes [ l l ) were (1) the widest portion of the
caudate nucleus (A8500p); (2) the portion of the caudate
nucleus adjacent to the globus pallidus (A6500p); and (3) the
most posterior portion of the nucleus (A4500p).
Histological Studies
Frozen tissue sections adjacent to those used for autoradiography were fixed in 40% formaldehyde, glacial acetic acid,
and methanol (1:1:8); stained with cresyl violet, and examined by light microscopy. For each animal, the distribution of
histopathological changes in the neocortex and in the caudate
putamen was carefully plotted by one observer (D. I. G.) on
line diagrams of coronal sections of the rat brain. It was more
difficult to define the cellular changes of early necrosis in
frozen sections than in paraffin sections. Nevertheless, by this
method it was possible to obtain a map of ischemic damage.
These histopathological maps were constructed without reference to the autoradiographs or the pseudocolor reconstruction of the ICBF data. The distribution of histopathological
changes was compared visually with that of the alterations in
Immediately before administration of the tracer, the
following values (mean +- standard deviation) were
560 Annals of Neurology Vol 15 No 6 June 1984
noted: PaOz was 141 k 19 mmHg; PaCo2 was 36 2 4
mmHg; and p H was 7.46 & 0.07. Mean arterial blood
pressure at the time of middle cerebral artery occlusion
was 107 +- 7 mmHg and at the end of tracer infusion
was 100 ? 10 mmHg.
Alteration in lCBF
The autoradiographs showed foci of marked reductions
in optical density in the normal distribution of the left
middle cerebral artery and, to a lesser extent, in that of
the left anterior cerebral artery (Figs 1-3). The distribution of ICBF indicated by the autoradiographs was
verified by the pseudocolor transformation maps (Fig
each animal there was a distinct focus in which optical
density was reduced and lCBF was extremely low (less
than 10 mV100 gdmin). This focus involved the
deeper layers of the frontoparietal neocortex and
the dorsolateral portion of the caudate putamen. In
the cortex, the severity of the abnormality varied at
different depths, and lCBF always increased from
deep to superficial layers. This variation created a pattern of bands of uniform rates of blood flow arranged
parallel to the cortical surface, so that there was a
twofold to fourfold increase of lCBF from the deep to
superficial layers within the cortex. This was more
marked at the periphery of the ischemic area (see Fig 3;
Fig 4, lower portion) than at the core (see Fig 1; Fig 4,
upper portion).
Local cerebral blood flow in the caudate putamen
increased from lateral to medial parts and also from
anterior to posterior portions. In the most medial portion of the nucleus, at the level of the globus pallidus,
lCBF was approximately 100 mV100 g d m i n , which
was more than 30 times the flow in the most lateral
portions. In animals with larger lesions, the gradient
across the nucleus was less marked anteriorly, because
blood flow in the medial portion was also severely reduced.
A sharp transition between the central focus of extremely low cerebral blood flow and surrounding areas
was seen in each of the six animals. In both the cortex
and the caudate putamen there was a zone where lCBF
changed by an order of magnitude (i.e., tenfold) within
a distance of only 1 to 2 mm.
Although there was some overlap between the blood
flow ranges in the right and left dorsomedial (anterior cingulate) cortical regions, ICBF tended to be
lower and more heterogeneous (ranging from 34 to
190 mV100 g d m i n ) on the left in all three brain sections. Within the left anterior cingulate cortex, lCBF
also increased from deep to superficial layers. This pat-
Fig I . {'4C}iodoantipyrineautoradiographs (right) at the level of
the caudate nucleus in each animal and diagrammatic mapping
(left) of the extent of histological change at the same level. The
ischemic hemisphere is on the right side of the diagrams in this
figure and in Figures 2-4. The images are ordered, according to
extent of histological change,from most widespread (animalI ,
top) to least extensive (animal 6, bottom). Regions of severe hypoperfusion corresponded with regions of histological change. For
example, in animal 3, the area of reduced bloodjmu in the most
dorsomedial aspect of the caudzte nucleus matches precisely a region in which histological alterations were present.
Tyson et al: Focal Cerebral Ischemia
Fig 2. Autoradiographs (right) and histological alterations (left)
at the level of the globus pallidus. The images are ordered as in
Figure 1. Note particular4 the cowespondence of low tracer uptake and structural change in the caudate nucleus of animals 2
562 Annals of Neurology Vol 15 No 6 June 1984
and 5 and the sharpness of the gradient at the boundary between
histologically normal and abnormal brain.
Fig 3 . Autoradiographs (right) and histological alterations (left)
at the level of the posterior caudate putamen. The images are or&red as in Figure 1 , Diffwences among animals are more pro-
nounced at this level. Bloodjou, is greater and histological
changes less pronounced in the superficial layers of the a f f t e d
Tyson et al: Focal Cerebral Ischemia
tern was similar to that in the distribution of the middle
cerebral artery, but blood flow was greater in the anterior cingulate cortex.
Fig 4. Pseudocolor reconstructions of the autoradiograph, showing topogruphiculdistribution of lCBF (mV100 gmlmin).
Animal 1 (top, left) has the largest histologicul lesion.
the dorsolateral and dorsomedial cortical regions were
similar. Both showed the normal laminar pattern, with
highest blood flow in the middle cortical layers. There
were no areas of gray matter in the right hemisphere in
which ICBF was less than that in corresponding regions
on the left.
Cells were smaller in the affected regions. The lesion
invariably involved the dorsolateral portion of the caudate putamen and the adjacent, deeper layers of the
cortex of the frontoparietal convexity. In the more posterior sections, smaller lesions were confined to a small
region in the dorsolateral cortex and often absent in the
most superficial layers. The dorsomedial (anterior cingulate) portion of the cortex was not involved in any
animal. Larger lesions in the caudate putamen extended
more medially but never involved the globus pallidus.
The most posterior portion of the caudate putamen
was similarly spared.
The second lesion in the left hemisphere was a small,
well-circumscribed focus of hemorrhagic necrosis in
Histopathological Findings
Two discrete lesions were observed histologically in
the left hemisphere of all six animals. The principal
abnormality was loss of normal laminar and columnar
organization of neocortex, along with diminished staining by cresyl violet in both neocortex and neostriatum.
5 6 4 Annals of Neurology
Vol 15 No 6 June 1984
the region of the rhinal sulcus, immediately subjacent
to the craniectomy, and was a consequence of the local
effects of the operation.
The combination of autoradiographs and pseudocolor
maps has shown more clearly and comprehensively
than previously [S, 17, 20) the hemodynamic consequences of focal cerebral ischemia. The autoradiographs provided the most precise information yet
about patterns of local perfusion; the pseudocolor
maps presented objective lCBF values, although with
some loss of resolution with respect to local changes.
The three-dimensional patterns that we observed provide clues to what factors interact to determine ICBF
values after occlusion of a major cerebral artery.
The dense, ischemic core in the hemisphere ipsilateral to the occluded middle cerebral artery invariably
included the dorsolateral caudate putamen and the adjacent frontoparietal cortex, which are the terminal
fields of distribution of the ganglionic and cortical end
arteries distal to the occlusion. Cerebral blood flow in
superficial layers of the cortex may have been relatively
well preserved because these layers are the initial field
of disuibution of the pial collateral supply. By contrast,
there was not a consistent variation in ICBF along the
course of the lateral lenticulostriate arteries, and there
was a uniformly low lCBF along their distribution, perhaps reflecting the extremely poor anastomoses of
these vessels.
The less pronounced reductions in ICBF in the cingulate cortex reflect its supply by the anterior cerebral
artery 12 1). The graded reduction in ICBF from medial
to lateral portions of the cingulate cortex probably
reflects the redistribution of blood flow from the nor-
rCBF (rn1/100g/min)
25 29 34 2105
Cowekation Between LCBF and
Histopathological Findings
Figures 1-3 show the relationship between areas with
extremely low blood flow, as reflected by low optical
density on the autoradiographs, and the distribution of
ischemic damage detected in the adjacent brain sec-.
tions. The distribution of structural and hemodynamic
abnormalities closely corresponds. A striking feature
was that the zones of maximum change in ICBF, in
which it varied tenfold within 1 and 2 mm, bordered
the region of ischemic histological change.
The ranges of ICBF in histologically normal and abnormal portions of each of the three brain sections
listed earlier are shown in Figure 5. Where histological
abnormalities were detected, lCBF was always less than
25 d l 0 0 g d m i n . Histological abnormalities were not
seen in areas where lCBF was 35 mVlOO g d m i n or
greater. In regions with flow values between 25 and 34
mVlOO gmlmin, there was not a consistent correlation
between lCBF and histological appearances.
Fig 5 . Range of lCBF (mV100 gm/min) in histologacalIy abnormal (solid bars) and normal (open bars) portions of the ldt
neocortex and cauhte putamen. Values in the 24 to 34 mlil00
gmlmin range wee associated with both normal and abnormal
histological appearances, even in the same tissue section. (rCBF
= regional cerebral bloodfEoru.)
mal cortical distribution of the anterior cerebral artery
into that of the occluded middle cerebral artery.
We were surprised to find that changes in cerebral
blood flow were not consistently a function of distance
from the area of densest ischemia. This area was surrounded by a zone 1 to 2 mm wide, across which lCBF
varied by an order of magnitude or more: beyond this
zone, changes in flow were smaller and more gradual.
The sharp lCBF gradient occurred at the interface between tissue dependent on collateral blood flow (the
former middle cerebral artery distribution) and tissue
still perfused from its primary supply (the original anterior cerebral artery distribution). These patterns suggest that lCBF was not determined entirely by reductions in local perfusion pressure beyond the occlusion.
It appears likely that vascular resistances differ
within and around the main lesion. The low level of
blood flow at the lesion’s center may reflect local vasoconstriction 1191 induced by high local extracellular
potassium concentrations that occur as ionic homeostasis fails [l8}. The steep flow gradient may occur at the
interface between this area of increased cerebrovascular resistance and the surrounding brain, in which collateral vessels are dilated.
Because of the extreme variability in ICBF over
short distances, flow at a given anatomical site can difTyson et al: Focal Cerebral Ischemia 565
fer greatly among animals. This presents difficulties if
cerebral blood flow is determined by using electrodes
to monitor hydrogen clearance, in which case only a
limited number of sites can be sampled. Even the depth
by which the electrode penetrates the cortex alters the
observed cerebral blood flow.
Examination of the frozen sections confirmed ischemic damage in each animal and enabled us to compare
its distribution with the pattern of tissue perfusion in
the same animal. In the past, such comparisons have
depended on parallel studies in two groups of animals.
Autoradiographic blood flow studies were performed
on cryostat sections of frozen brain in one series, and
histological studies were undertaken on separate animals killed by perfusion fixation in vivo {l6, 171.
Even though the subtle histological features of the
ischemic neuronal process [GI could not be identified
in the frozen sections, the boundaries of the ischemic
area were delineated by the distinct change in the ability of cresyl violet to stain ischemic nerve cells. The
delineation of lesions by staining was similar to the
delineation of areas of ischemic cell change apparent in
identically prepared but perfusion-fixed animals {16].
Cerebral blood flow and his tological abnormality
matched closely even within a given structure. Thus,
ischemic alterations were often inconspicuous or even
absent in the superficial cortical layers, a finding noted
previously in the primate f 7). This finding correlates
with our demonstration of smaller reductions in cerebral blood flow in the outer part of the cortex.
The studies were not intended to investigate the
lCBF “threshold” for the production of ischemic damage. The changes in cerebral blood flow that produce
ischemic damage probably occur within the first few
minutes after occlusion { l Z , 15, 17). Nevertheless, our
finding that there was a consistent relationship between
cerebral blood flow less than 25 mVlOO g d m i n and
structural alterations implies that, even after 4 hours of
permanent ischemia, lCBF continues to be linked with
the metabolic activity of ischemic cells.
The zone in which lCBF varied tenfold within 1 to 2
mrn was adjacent to or sometimes appeared to overlap
the border of the histopathologically defined lesion.
Blood flow in this narrow, perifocal zone was within
the range that is reputed, at least in other species, to
define the ischemic penambra [ 2 ) .
The concept of an ischemic penumbra is derived
largely from experiments in which blood flow was measured repeatedly at the same site while the severity of
ischemia was increased progressively. These studies
showed that, above flow levels that lead to ischemic
damage {12, 14, 151, different and distinct thresholds
oflCBF suppress electrical activity [ 3 , 5 , 101 and cause
loss of ionic homeostasis 11, 3, 41. In these experiments, however, the proportion of penumbral to ischemic brain tissue could not be determined. The ex566 Annals of Neurology
Vol 15 No 6 June 1984
tremely abrupt change in blood flow seen at the edges
of the lesions we studied suggests that 4 hours after
middle cerebral artery occlusion ischemic penumbra in
the rat is very narrow.
Our observations suggest that when a major cerebral
artery is occluded, local blood flow is determined by
the complex interaction of several factors. These include reductions in local perfusion pressure, the architecture of primary and collateral vasculature, the occurrence of vasodilation and vasoconstriction in the pial
collateral network, and local metabolic needs. The net
effect is that the ischemic lesion is not bounded by
gradual increments in ICBF but instead is delineated
by a steep “escarpment” in which cerebral blood flow
changes abruptly. Studies designed to discover how the
breadth and location of this escarpment are altered by
maneuvers that may modify the outcome of an ischemic insult will increase pathophysiological knowledge
and facilitate treatment of cerebral ischemia.
1. Astrup J, Blennow G, Nilsson B: Effects of reduced cerebral
blood flow on EEG pattern, cerebral extracellular potassium, and
energy metabolism in the rat cortex during bicuculline-induced
seizure. Brain Res 177:115-126, 1979
2. Astrup J, Siesjo BK, Symon L Thresholds in cerebral ischemia-the ischemic penumbra. Stroke 12:723-725, 1981
3. Astrup J, Symon L, Branston NM, Lassen NA: Cortical evoked
potential and extracellular Kf and H+ at critical levels of brain
ischemia. Stroke 8:51-57, 1977
4. Branston NM, Strong AJ, Symon L Extracellular potassium activity, evoked potential and tissue blood flow: relationship during progressive ischemia in baboon cerebral cortex. J Neurol Sci
32:305-321, 1977
5. Branston NM, Symon L, Crockard HA, Pasztor E: Relationship
between the cortical evoked potential and local cortical blood
flow following acute middle cerebral artery occlusion in the baboon. Exp Neurol45:195-208, 1974
6. Brown AW, Brierley JB: The nature, distribution and earliest
stages of anoxic-ischemic nerve cell damage in the rat brain as
defined by the optical microscope. Br J Exp Path01 4937-106,
7. Garcia JH, Kamijyo Y:Cerebral infarction. Evolution of histopathological changes after occlusion of a middle cerebral artery
in primates. J Neuropathol Exp Neurol 33:408-421, 1974
8. Ginsberg MD, Reivich M, Frinak S , Harbig K Pyridine nucleotide redox state and blood flow of the cerebral cortex following
middle cerebral artery occlusion in the cat. Stroke 7:126-131,
9. Goochee C, Rasband W, Sokoloff L Computerized densitometry and color coding of ~’4C~deoxyglucose
autoradiographs. Ann Neurol 7:359-370, 1980
10. Heiss W-D, Hayakawa T, Waltz AG: Cortical neuronal function
during ischemia. Arch Neurol 33313-820, 1976
11. Konig JFR, Klippel RA: The Rat Brain: A Stereotaxic Atlas.
Baltimore, Williams & Wilkins, 1963
12. Morawetz lU3, De Girolami U, Ojemann RG, et al: Cerebral
blood flow determined by hydrogen clearance during middle
cerebral artery occlusion in anesthetized monkeys. Stroke
9:143-149, 1978
13. Sakurada 0, Kennedy C, Jehle J, et al: Measurement of local
cerebral blood flow with iodo 1’*C) antipyrine. Am J Physiol
234:H59-H66, 1978
14. Symon L, Dorsch NWC, Crockard HA, et al: Clinical features,
local CBF and vascular reactivity in a chronic (3 year) stroke in
baboons. In Harper AM, Jennett B, Miller JD, Rowan JO (eds):
Blood Flow and Metabolism in the Brain. Edinburgh, ChurchillLivingstone, 1975, pp 12.10-12.14
15. Tamura A, Asano T, Sano K Correlation between rCBF and
histological changes following temporary middle cerebral artery
occlusion. Stroke 11:487-493, 1980
16. Tamura A, Graham DI, McCulloch J, Teasdale GM: Focal cerebral ischemia in the rat: 1. Description of technique and early
neuropathological consequences following middle cerebral artery occlusion. J Cereb Blood Flow Metab 1:53-60, 1981
17. Tamura A, Graham DI, McCulloch J, et al: Focal cerebral ischemia in the rat: 2. Regional cerebral blood flow determined by
('*C) iodoantipyrine autoradiography following middle cerebral
artery occlusion. J Cereb Blood Flow Metab 1:61-69, 1981
18. Teasdale G, Legrain Y , Graham D, MacKenzie E: Potassium
release and vascdar events in focal cerebral ischemia. J Cereb
Blood Flow Merab 3 [Suppl 1):S395-S396, 1983
19. Walt2 AG, Sundt TM Jr: The microvasculature and microcirculation of the cerebral cortex after arterial occlusion. Brain
90:681-696, 1967
20. Yamaguchi T, Waltz AG, Okazaki H: Hyperemia and ischemia
in experimental cerebral infarction: correlation of histopathology
and regional blood flow. Neurology 21:565-578, 1971
21. Yamori Y,Horie R, Handa H, et al: Pathogenetic similarity of
strokes in stroke-prone spontaneously hypertensive rats and humans.Stroke 7:46-53, 1976
Tyson et al: Focal Cerebral Ischemia
Без категории
Размер файла
2 799 Кб
ischemia, hemodynamics, change, focal, rat, histopathological, topography, cerebral
Пожаловаться на содержимое документа