From the Departments of Neurosurgery, Neurology (M.A.Y.), Anesthesiology
(J.E.L.), and Stanford Stroke Center, Stanford University, Stanford, Calif.
Correspondence to Gary K. Steinberg, MD, PhD, Professor and Chairman, SUMC, Department of Neurosurgery, 300 Pasteur Dr, Room R281, Stanford, CA 94305. E-mail mk.gks{at}forsythe.stanford.edu
Methods—Rats underwent a 2-hour occlusion of the left middle
cerebral artery. In the first study (I) animals were kept
(intraischemically) at either 37°C (n=8), 33°C (n=8), or
30°C (n=8). Study II consisted of 4 groups: (1) controls (37°C,
n=10), (2) 30 minutes of hypothermia started at ischemic onset
(33°C, n=9), (3)1 hour (33°C, n=8), and (4) 2 hours (33°C, n=8).
Brain temperature was measured by a thermocouple probe placed in the
contralateral cortex. After suture removal, all animals were rewarmed
and reperfused for 22 hours (I) or 70 hours (II).
Results—Mild hypothermia to 33°C or 30°C was neuroprotective
(17±7% and 27±6%, respectively) relative to controls (53±8%,
P<0.02), but 33°C was better tolerated and recovery
from anesthesia was faster. The neurological score of
hypothermic animals was significantly better than that of controls (I &
II) at both 24 and 72 hours postischemia except for the
30-minute group (II), which showed no improvement. In Study II, 2 hours
of hypothermia reduced injury by 59%, 1 hour reduced injury by 84%
whereas 30 minutes did not reduce injury. Normalized for infarct
size, 2 hours of mild hypothermia decreased neutrophil accumulation by
57% whereas both 1 hour and 30 minutes had no effect. At 72 hours, 1
and 2 hours of mild hypothermia decreased transferase dUTP nick-end
labeling (TUNEL) staining by 78% and 99%, respectively, and 30
minutes of hypothermia had no effect.
Conclusions—Intraischemic mild hypothermia must be
maintained for 1 to 2 hours to obtain optimal neuroprotection against
ischemic cell death due to necrosis and apoptosis.
Small decreases in brain temperature (2°C to 6°C) are well
tolerated and have been shown to confer a significant degree of
neuroprotection in some animal models of cerebral
ischemia.1 2 3 4 5 Although research in this
area has been conducted for more than 30 years, many questions
regarding mild hypothermia remain unanswered, including the degree of
temperature reduction needed, when it should be instituted, and for how
long a period it should be maintained. The mechanism of cerebral
protection by mild brain hypothermia is also unclear and still a source
of controversy. Its neuroprotective effects have been ascribed to a
decrease in cerebral metabolic rate, restoration of
cerebral blood flow, and preservation of the blood-brain barrier. Other
potential protective mechanisms include alterations in neurotransmitter
release, activity of protein kinases, and resynthesis of cellular
repair proteins.6 More recently, there have also
been reports that hypothermia may lead to specific inhibition of
apoptosis7 and the attenuation of the
inflammatory response that often follows an ischemic
insult.8
Mild or moderate hypothermia has been shown to reduce neurological
deficits if started before, during, or after cerebral ischemia,
but few studies have examined functional outcome in detail after
experimental cerebral ischemia with
hypothermia.9 10 11 12 13 Furthermore, much of the work
with mild hypothermia has centered around global models of cerebral
ischemia and less is known about focal ischemia. Here
we report the results of two studies designed to determine the optimal
depth and duration of mild hypothermia in a model of middle cerebral
artery occlusion (MCAO) with special attention directed to the
animal's intraoperative systemic physiological
parameters, postanesthesia recovery time, and
behavioral assessment over a 24- to 72-hour period. We also examined
the effects of mild hypothermia on infarct size, apoptosis
using TUNEL staining plus morphologic criteria and analysis of
DNA fragmentation, and inflammation by myeloperoxidase (MPO) staining
of neutrophils.
Stroke Model
Behavioral Analysis
Infarct Analysis
Transferase dUTP Nick-End Labeling (TUNEL) Staining
Myeloperoxidase (MPO) Staining
Animals that died before the desired endpoint were excluded from
any histologic outcome determination (ie, infarct size quantification,
and TUNEL and MPO staining) to avoid potentially biasing the data. Six
coronal TUNEL-stained sections were counted on 2 animals from each
group (total of 8 animals) to identify the coronal level(s) with the
highest number of stained cells/nuclei. Then this coronal level was
assessed for TUNEL and MPO staining in every surviving animal (N, n=5;
H0.5, n=6; H1, n=8; H2, n=8). All TUNEL-staining data refer to the
average number of TUNEL-stained nuclei detected in two high power
fields at the level of the anterior commissure for each animal
that reached the desired 72-hour endpoint in Study II. All data from
MPO staining refer to the number of polymorphonuclear leukocytes
(PMNLs) detected in an entire left hemisphere section at the level of
the AC (for each surviving animal).
Analysis of DNA Fragmentation
Statistical Analysis
In Study 1, neurological outcome showed that hypothermic animals
tended to score better than the normothermic animals,
although this difference was not statistically significant
(P=0.08 by Kruskal-Wallis test). In Study 2, neurologic
scores at both 24 and 72 hours postischemia were
significantly better for animals subjected to 1 or 2 hours of
hypothermia. In particular, these animals were more alert and more
responsive to handling; showed normal eating, drinking and grooming;
and had increased exploratory behavior; whereas
normothermic animals were often very lethargic and showed
no interest in their surroundings. Treatment with thirty minutes of
hypothermia showed no neurologic improvement when compared with the
normothermic controls. Results from the neurological
deficit scores are shown in Figure 2a
In Study 1, histopathology revealed an 68% and 51% reduction in the
area of cortical infarct for the 33°C group (17±7%,
P<0.005) and the 30°C group (27±6%,
P<0.02), respectively, compared with the
normothermic animals (53±8%) as seen in Figure 3
Data from a representative
normothermic and a 2-hour hypothermic animal (Study II) are
displayed in Figure 5
After 70 hours of reperfusion, neutrophils could be easily detected in
both cortex and striatum and had a similar distribution to the
TUNEL-stained nuclei; however, a higher concentration of PMNLs was
found within the ischemic core compared with
TUNEL-stained cells. At 72 hours postischemia, 1 hour of
mild hypothermia reduced neutrophil accumulation by 75%, and 2 hours
of mild hypothermia decreased it by 72%. Thirty minutes of hypothermia
reduced PMNLs infiltration by 39%, but this was not statistically
significant (Table 3
Our first study demonstrates that both mild (33°C) and moderate
(30°C) intraischemic hypothermia are protective against
ischemic neuronal damage in a rat model of transient focal
cerebral ischemia. Although we find no significant differences
between the two hypothermic groups, Goto et al20
have reported temperature-dependent reductions in cerebral infarction.
In that study, infarct volume was reduced by 22.4% in mildly
hypothermic animals (33°C) and by 49.5% in moderately hypothermic
animals (29°C). This suggests that a more intense ischemic
challenge, such as the one used in that study (3 hours MCAO plus
bilateral carotid artery occlusion), may require larger temperature
reductions to obtain significant neuroprotection. In fact, the 3-hour
model of reversible ischemia may be more analogous to permanent
focal ischemia, since the size of infarction obtained in that
study is not different from what others have found using permanent
MCAO.21 In the present study, both
temperatures afford a marked and comparable degree of neuroprotection;
however, mild (33°C) hypothermia is devoid of the systemic
complications, including decreased respiratory rate and cardiac
arrhythmias, which were observed in animals treated with
moderate (30°C) hypothermia. Furthermore, average
postanesthesia recovery time showed that animals in the
33°C group recovered at a significantly faster rate than
normothermic controls, whereas animals in the 30°C group
showed no improvement in recovery time. Therefore, mild hypothermia
(33°C) may be a safer, more manageable alternative to moderate
hypothermia. Based on these results, 33°C was used as the target
temperature in the second study that was designed to determine the
optimal duration of intraischemic hypothermia.
Results from Study 2 show that both 1 and 2 hours of
intraischemic hypothermia, started at the ischemic
onset, can reduce the behavioral and histopathological deficits
associated with transient focal cerebral ischemia. Thirty
minutes of mild hypothermia, however, does not afford any protection
against neuronal damage and does not improve neurological outcome.
Others who have studied this question have found conflicting results.
Kader et al22 found that 33°C or 34.5°C for 1
hour, induced at ischemic onset, reduced infarction due to
permanent MCAO with a 24-hour endpoint. Another
study,23 however, found that 33°C for 1 hour
reduced infarct size at 96 hours after transient but not permanent
focal ischemia. Karibe et al24 found a
significant reduction in infarction if mild intraischemic
hypothermia was introduced within 30 minutes of transient (2 hours)
MCAO, but the protection was lost if hypothermic induction was delayed
by more than 60 minutes. Thus, although it may not be necessary to
maintain intraischemic hypothermia for more than 1 hour if
started within 30 minutes of the focal ischemic onset, longer
periods of hypothermia may be needed to achieve neuroprotection
following permanent vessel occlusion or if initiated in a delayed
fashion after ischemic onset.
The observation that reducing the temperature for 1 hour but not 30
minutes immediately after MCAO results in cerebroprotection in our
model, suggests that mild hypothermia affects some of the injury
mechanisms that occur early in the ischemic process. During
ischemia there is an uncoupling of cerebral blood flow and
metabolism that results in cerebral energy failure and a
concomitant disruption of ion homeostasis that leads to the
accumulation of intracellular calcium.25 This
activates several enzymatic systems that, if severe and
prolonged, can lead to irreversible cell damage. Mild hypothermia may
work early on by reducing the energy demands of the cell (ie, slowing
down energy failure), thereby delaying activation of deleterious
catabolic processes.26 Hypothermia may also act
by preventing adenosine triphosphate
depletion27 and intracellular
acidosis,6 28 by reducing the release of
excitatory amino acids and
catecholamines,29 and by preventing
the ischemic inhibition of protein
kinases.30
Cytotoxic edema also occurs early in the ischemic process. By
lowering the rates of catabolic reactions, mild hypothermia may help
maintain tissue integrity and therefore reduce edema
formation.31 In Study 2 we show that 1 hour of
mild hypothermia reduces cerebral edema by 85%. Based on the
behavioral and histological data, the lack of
significant protection against brain swelling with 30 minutes of
hypothermia is not surprising. However, 2 hours of mild hypothermia
does not significantly reduce brain edema. This may be explained by the
variability in infarct sizes and edema observed in that group. One
animal in the 2-hour group had a large infarct with a correspondingly
large edema component (4% higher than the largest value in the
normothermic group). That animal had a low (near normal)
neurological score at 24 hours postischemia but, unlike the
other animals in that group, showed increased behavioral deficits at 72
hours.
In recent years two other delayed injury mechanisms have come into
focus: activation of apoptosis, and inflammation. Several
investigators have recently shown that apoptosis may be a
contributing factor in neuronal death following both
global32 33 34 and
focal33 35 36 37 38 39 ischemia. Results from our
second study show that at 3 days postischemia, 1 and 2
hours of mild hypothermia is sufficient to decrease the number of
apoptotic cells, as determined by TUNEL staining and
morphology, by 78% and 99%, respectively. Thirty minutes of
hypothermia, however, has no effect. This is particularly significant
because the only other study that has demonstrated robust effects of
hypothermia on the number of cells undergoing apoptosis (in
transient global hypoxia-ischemia) used a 12-hour
cooling period commenced postinsult.7 In that
study, mild hypothermia (34.9°C mean tympanic membrane temperature)
had no effect on the fraction of cells undergoing necrosis, although
the fraction of apoptotic cells was significantly reduced,
suggesting that hypothermia specifically inhibited apoptosis.
The authors speculate that, although the trigger of programmed cell
death may occur during the ischemic insult, cellular commitment
to death occurs later, thus allowing hypothermia to interrupt the
biochemical cascade leading to apoptosis.
In our study we find that TUNEL staining is absent before 6 hours of
recirculation but is clearly visible at 24 hours. By 72 hours
postischemia, TUNEL staining is very prominent in both
normothermic and hypothermic animals, although to a
significantly different degree. Results from the DNA fragmentation
analysis confirm the latter findings. After 3 days of
reperfusion we see evidence of DNA laddering in the cortex and striatum
(ipsilateral to the MCAO) of both normothermic and
hypothermic animals. At first this would appear to contradict results
from a study by Charriaut-Marlangue et al37 in
which the initial stages of DNA fragmentation were observed as early as
3 hours into the recirculation period following 1 hour MCAO and
bilateral CCA occlusion. However, full DNA laddering was not evident
until 18 hours postischemia. Li et
al40 also investigated the temporal profile of in
situ DNA fragmentation following 2 hours MCAO and found that
apoptosis first appears within 30 minutes of reperfusion, peaks
at 24 to 48 hours, and persists for 4 weeks after the insult. The
differences observed in those studies and ours may be attributed to
variations in the ischemic model, as well as in the duration of
vessel occlusion. It is possible that apoptosis is triggered by
various mechanisms that themselves may have unique temporal profiles
and be differentially activated depending on infarct severity.
On the other hand, the variability between studies may simply reflect
differences in the sensitivity of the assays used in each study. The
anatomical distribution of TUNEL-stained cells is another issue that
deserves attention. In their studies, Li and
colleagues40 41 showed that apoptotic
cells are primarily localized to the inner boundary zone of the
ischemic infarct and that they appear earlier in the preoptic
area and striatum than in the cortex. This is consistent with
the idea that apoptosis is an active process that requires
energy, so the pericore regions of the affected hemisphere may provide
the necessary conditions for the induction of programmed cell death. In
our study, however, we find a somewhat different distribution pattern
of TUNEL-stained nuclei. In both normothermic animals and
those treated with 30 minutes of mild hypothermia, we find that
apoptotic cells, although dispersed throughout the MCA
territory, are concentrated primarily in the boundaries of the entire
striatum and are also prominent in the amygdala. Although the
significance of this spatial distribution is unclear, it is possible
that apoptotic cells appear first in the boundaries of the
ischemic core and are still detectable 72 hours after the
initial insult even though the lesion has already expanded to pericore
regions by this time. No distinct distribution pattern of
TUNEL-stained cells was found in animals treated with either 1 or 2
hours of mild hypothermia.
Although the significance of the inflammatory response in infarct
development has attracted the attention of many researchers over the
years, little work has been done to assess the effects of mild
hypothermia on leukocyte accumulation in ischemically-injured
neuronal tissue. Polymorphonuclear leukocytes are at the center of
the inflammatory process and may play an important role in the
development/progression of an infarct.42 PMNLs
can adhere to the endothelium of the microvasculature
within the ischemic territory, thereby plugging vessels ("no
reflow" phenomenon) and worsening ischemic
damage.43 Activated PMNLs may also
mediate injury of the vascular endothelium through the
generation of free radicals, thus intensifying the ischemic
condition.44 Finally, once leukocytes have
infiltrated the neuropil, they can injure neurons and glia directly via
proteolytic enzyme release.45 In 1994, Garcia et
al,46 using a rat model of permanent MCAO, showed
that PMNLs could be detected in capillaries and venules within 30
minutes of ischemic onset in areas where the microvascular
patency was impaired. Others47 48 49 50 have studied
the role of leukocytes in inducing neutropenia or blocking adhesion
molecules involved in neutrophil migration, particularly ICAM-1, CD11b,
and CD18. Those studies show that inhibition of these molecules can
reduce infarct volume, and that this reduction in damage is dose
dependent and associated with significant functional improvement. More
recently, a study by Toyoda et al8 examined the
effects of intraischemic moderate hypothermia (30°C) on
neutrophil infiltration following transient (3 hours) MCAO. Results
from that study showed that, at 24 hours postischemia,
hypothermic animals exhibited significantly less MPO activity in the
pericore region compared with normothermic controls,
although no differences were observed in the core. In Study 2 we find
that 1 hour of mild hypothermia is sufficient to reduce neutrophil
accumulation by 75% and 2 hours of hypothermia decreases it by 72%.
We observe a decrease in neutrophil accumulation within the core of the
infarct. Thirty minutes of hypothermia reduces PMNL infiltration by
39%, although this is not statistically significant. When normalized
for infarct size, only 2 hours of mild hypothermia significantly
reduces accumulation of PMNLs, whereas 1 hour shows a trend and 30
minutes is not effective.
There are some limitations to the present studies. Although we
showed that mild hypothermia is effective at reducing neuronal injury
when applied intraischemically, a more clinically relevant
issue is whether hypothermia can protect if applied hours after
ischemia or once reperfusion has occurred. However, other
studies1 51 52 on focal ischemia have
shown that delaying hypothermic onset up to 1.5 hours can still be
beneficial, provided hypothermia is maintained for at least 1 to 3
hours. Longer durations of mild hypothermia with 2 to 3 hours of delay
have not yet been studied in this model. Secondly, since we assessed
injury at 24 and 72 hours postischemia, we cannot exclude
the possibility that mild hypothermia simply delays neuronal damage.
Indeed, work by Dietrich et al53 on global
ischemia has shown that mild hypothermia (applied for 3 hours
at reperfusion onset) is very effective at reducing neuronal damage 3
days postischemia, but the protective effect is less
impressive at 7 days. By 2 months, cytoprotection is lost unless
hypothermia is applied intraischemically. Finally, we cannot
exclude the possibility that the neuroprotective effects observed with
hypothermia are due to changes in blood flow. Indeed, 30°C (but not
33°C) has been shown to decrease blood flow in the ischemic
brain following permanent MCAO.54
To date, no other single therapeutic approach has been developed that
can reduce ischemic neuronal injury to the extent that is
observed with hypothermia. As suggested by the present and other
studies, mild hypothermia's neuroprotective benefit may be secondary
to attenuation of several detrimental processes involved in both
necrotic and apoptotic cell death. Furthermore, this may be a
cost-effective therapy (particularly in the surgical setting) that is
easily implemented, and may prove to be of value by itself or in
combination with more traditional pharmaceutical approaches. Although
further work needs to be done to evaluate its long-term effects,
clinical studies of mild hypothermia in the treatment and prevention of
stroke should be considered.
Received June 10, 1998;
accepted June 26, 1998.
Department
of Anesthesiology and Critical Care Medicine,
Johns Hopkins Medical Institutions,
Baltimore, Maryland
The authors suggest that studies should be initiated to evaluate
efficacy of hypothermia in the clinical setting. They argue that it may
be preferable to use 33°C rather than 30°C, because it was
associated with fewer cardiovascular and respiratory
management challenges. However, the authors also argue that this
therapy may be of particular use in the surgical setting, where the
patient would more than likely have the benefit of controlled
ventilation and not suffer the cardiovascular and
respiratory consequences of more profound (30°C) hypothermia.
Therefore, the current study may not completely assess the
risk-to-benefit ratio of more profound hypothermia in the clinical
setting. In addition, in the clinical setting it is likely that the
anesthetic regimen would not include halothane because of its potential
for hepatoxicity and elevation of cerebral blood flow in patients. It
is even possible that in the setting of modern clinical anesthetic
practice, hypothermia may not provide significant neuroprotection
beyond that which is already being produced by modern anesthetic agents
(eg, propofol, etomidate, and isoflurane).
From a technical point of view, this study utilizes strict control of
physiological variables to allow for
straightforward interpretation of outcome variables. Although the
authors use a well-characterized model of transient focal
ischemia, it would have been better had they provided data
regarding their methods to ensure that the degree of ischemia
was similar between groups. Possibilities for standardizing the degree
of ischemia produced by this model include measuring cerebral
perfusion (eg, blood flow or laser-Doppler flowmetry),
electrical activity (EEG or somatosensory evoked potentials), or
evaluation of the neurological exam.
Received June 10, 1998;
accepted June 26, 1998.
© 1998 American Heart Association, Inc.
Original Contributions
Optimal Depth and Duration of Mild Hypothermia in a Focal Model of Transient Cerebral Ischemia
Effects on Neurologic Outcome, Infarct Size, Apoptosis, and Inflammation
Abstract
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
Introduction
Background and Purpose—Mild
hypothermia is possibly the single most effective method of
cerebroprotection developed to date. However, many questions regarding
mild hypothermia remain to be addressed before its potential
implementation in the treatment of human stroke. Here we report the
results of 2 studies designed to determine the optimal depth and
duration of mild hypothermia in focal stroke and its effects on infarct
size, neurological outcome, programmed cell death, and
inflammation.
Key Words: hypothermia • ischemia • neuroprotection • apoptosis • inflammation
Introduction
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
Introduction
Stroke is a major source of disability and thus
much research emphasis is being placed on the treatment and prevention
of stroke. Neuroprotective strategies include the use of glutamate
receptor antagonists, calcium channel blockers, and
free-radical scavengers. Thrombolytic agents have already
been shown to improve outcome by accelerating
recanalization. In recent years, there has also
been a resurgence of interest in mild hypothermia as a method of
cerebral protection.
Materials and Methods
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
Introduction
The following animal protocols were approved by the Stanford
University Administrative Panel on Laboratory Animal Care.
Male Sprague Dawley rats weighing 290 to 350 g (Charles
Rivers, Wilmington, Del) were anesthetized with 3% halothane
(delivered by mask) and were maintained in surgical plane of
anesthesia with 1% halothane in 200 mL/min oxygen and 800
mL/min air without the use of paralytic agents. Depth of
anesthesia was assessed every 15 minutes by hind limb
pinch. A thermistor probe was inserted 50 mm into the rectum and
rectal temperature was maintained at 36.5°C to 37.5°C before
ischemia. ECG leads were placed to monitor heart rate and
respirations. The animal's head was immobilized in a
stereotactic frame and a small scalp incision was made. A
small burr hole was drilled to permit insertion of a 33-gauge
thermocouple temperature probe to measure cortical brain temperature in
the nonischemic hemisphere. The probe was inserted into the
cortex approximately 3 mm deep. The probe site was sealed with
dental cement, and the animal was placed in supine position. The right
femoral artery was catheterized with PE-50 tubing for monitoring blood
pressure, collecting blood samples, and infusion of normal saline and
sodium bicarbonate (8.4%), as necessary, to correct
arterial base deficit. Blood glucose and hematocrit values
were measured before and during the ischemic period. Blood
gases were measured (alpha-stat method) with an automatic pH/Blood Gas
Analyzer Model 178 (Ciba Corning Diagnostics Corp,
Medfield, Mass). A midline incision was made in the neck to expose the
common carotid (CCA), external carotid (ECA), internal carotid (ICA),
and pterygopalatine (PPA) arteries. The CCA, ECA, and PPA were ligated
with a 6-0 silk suture. The stroke was produced by inserting a 3-0
monofilament suture (with a flamed tip) 18 to 23 mm from the
bifurcation of the ICA and ECA, thus occluding the ostium of the middle
cerebral artery (MCA). The suture was kept in place for 2 hours. In the
hypothermic groups, total body cooling was achieved by spraying alcohol
onto the animal and cooling it to the desired temperature with a fan.
This method decreased brain temperature from 37°C to 30°C within 10
minutes. In the first study, brain temperatures were kept
(intraischemically) either normothermic (36.5°C
to 37.5°C; n=8), mildly hypothermic (32.5°C to 33.5°C; n=8), or
moderately hypothermic (29.5°C to 30.5°C; n=8). After the optimal
depth of hypothermia was determined (33°C), a second study was
carried out in which animals were randomized into 4 different
experimental groups: (1) normothermic ischemic
controls (N, n=10), (2) mildly hypothermic for 30 minutes started at
the onset of ischemia (H0.5, n=9), (3) mildly hypothermic for 1
hour (H1, n=8), and (4) mildly hypothermic for 2 hours (H2, n=8). To
avoid limb ischemia after catheterization of
the right femoral artery, an anastomosis was done (10-0 suture) just
before suture removal (Study 2). After suture removal, the animal was
returned to normothermia and allowed to reperfuse for 22 (Study 1) or
70 hours (Study 2). The brain thermocouple probe was removed just
before anesthesia was terminated, during early reperfusion.
Rewarming was achieved by a heating pad placed under the animal and a
lamp positioned over the animal's body. This method increased rectal
temperature from 30°C to 37°C within 10 minutes in all hypothermic
animals. Postanesthesia recovery time, defined as the time
from when anesthesia was discontinued until the animal
regained its righting reflex, was recorded. The animal was then
transported to the intensive care unit (ICU) at the Veterinary Services
Center at Stanford University, Stanford, Calif, where it was
closely monitored throughout the recovery period and evaluated for
neurological findings. Rectal temperature was monitored hourly for the
first 3 hours after recovery from anesthesia. Fluids (1 to
2 mL normal saline per 100 grams of body weight) were given as needed,
and the analgesic butorphanol tartrate (0.05 to 2.0 mg/kg) was
administered if the animal was perceived to be in pain or in undue
distress. The animal was allowed free access to food and water after
surgery.
The animals were monitored continuously and assessed for
neurological findings at the end of the experiment by an individual
blinded to the experimental groups. A neurological grading scale was
used (Table 1
). The
animal's weight was recorded at 24, 48, and 72 hours
postischemia, and mortality was also used as an
endpoint.
View this table:
[in a new window]
Table 1. Neurological Grading
Scale1
Only animals surviving the entire study were used for histologic
and histochemical analysis (24-hour survival for Study I,
72-hour survival for Study II). Animals were euthanized with a
halothane overdose, and the brains were quickly removed and sliced into
3-mm thick coronal sections. The brain slices were then incubated in
2% tetrazolium chloride (TTC) at 37°C for 15 minutes and fixed in
10% buffered formalin (pH 7.4) for 1 week. After paraffin embedding,
10-µ thick sections were stained with hematoxylin and eosin (H & E).
In Study 2, additional sections were cut for TUNEL and MPO staining.
Infarct was evaluated by light microscopy, in a blinded fashion, in
both the cortex and striatum. Histologic criteria for infarct included
areas of pan-necrosis with shrunken dark neurons and glial
pallor.14 The area of infarct (determined from
H&E-stained sections) was expressed as a percentage of the total area
of left hemisphere, cortex, or striatum (Study I). Infarct volume was
also calculated corrected for edema by use of the method of Swanson et
al.15 Infarct areas on TTC-stained slices were
traced and measured with an image analysis system (MCID,
Imaging Research Inc, Ontario, Canada). Infarct volume was calculated
as the sum of the infarcted areas multiplied by the distance between
brain slices. Brain swelling was calculated as follows: [(volume of
ipsilateral hemisphere-volume of contralateral hemisphere)/volume of
contralateral hemisphere (x100%)].
TUNEL staining was performed with the ApopTag Plus in situ
Apoptosis Detection Kit (Oncor, Gaithersburg, Md). Tissue
sections were deparaffinized in xylene and hydrated in a sequence of
ethanol washes, followed by a final wash in phosphate-buffered saline
(PBS). Nuclei of tissue sections were stripped of proteins by
incubation with 20 µg/mL of proteinase K (42°C) for 15 minutes. The
slices were then washed in distilled water and PBS and incubated in 3%
hydrogen peroxide to remove endogenous peroxidases. After
equilibration, the sections were incubated at 42°C in terminal
deoxynucleotidyl transferase (TdT) enzyme and
digoxigenin-labeled substrate for 1 hour. Antidigoxigenin was then
applied, and visualization was accomplished with a diaminobenzidine
(DAB) substrate solution. The sections were then counterstained with
methyl green, cleared, and mounted. Sections treated with the DNAse I
enzyme were used as positive controls (Amersham, Cleveland, Ohio), and
sections in which TdT treatment was omitted were used as negative
controls. TUNEL-labeled cells were counted in the cortex and
striatum.
Tissue sections were deparaffinized, hydrated, and washed in
PBS. The sections were then incubated with 20 µg/mL of proteinase K
for 15 minutes, washed with distilled water and PBS, and incubated with
3% hydrogen peroxide to remove endogenous peroxidases.
Blocking serum was applied for 20 minutes at 42°C. The sections were
then sequentially incubated in primary rabbit anti-human MPO antibody
(1:200) for 30 minutes, biotinylated goat anti-rabbit IgG antibody
(1:200) for 30 minutes, and ABC reagent (avidin and biotinylated
horseradish peroxidase complex) for 30 minutes (ABC kit, Vector,
Burlingame, Calif). Visualization was accomplished with a DAB substrate
solution until the desired staining intensity was obtained (Oncogene
Research Products, Cambridge, Mass). The sections were then
counterstained with H & E, cleared, and mounted.
Four animals (2 normothermics and 2 hypothermics,
33°C, 2-hour duration) underwent 2 hours of MCA occlusion followed by
72 hours of reperfusion as previously described. At the end of
reperfusion period, DNA was harvested from the cortex and striatum of
both the ipsilateral and contralateral hemispheres. The tissue samples
were immediately frozen on dry ice or immediately digested in 100
mmol/L Tris-HCl (pH 8.0), 200 mmol/L NaCl, 5 mmol/L EDTA, and
0.2% SDS containing 5 mg/mL proteinase K overnight at 55°C. The DNA
solution was incubated for 1 hour at room temperature with 5 mg/mL
RNase A and then extracted with phenol-chloroform. DNA was separated on
0.8% agarose gels and visualized with ethidium bromide
staining.16 17 A 123 bp DNA ladder (Life
Technologies, GibcoBRL, Gaithersburg, Md) was used as a control in all
gels.
Statistical analyses were done with ANOVA for
continuous data and with nonparametric tests for
noncontinuous data. All data were expressed as mean±SEM; a
P value < 0.05 was considered significant.
Results
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
Introduction
Figure 1 shows the overall
correlation coefficient between brain and rectal temperature (r=0.91,
P<0.0001), with brain temperature higher than rectal
temperature by 0.2°C to 0.7°C. There were no significant
differences between groups in systemic
physiological parameters or volume of
fluid infused with the exception of the desired changes in brain and
rectal temperature (Studies 1 and 2), and a lower intraischemic
pH value for the 2-hour hypothermic group compared with the 30-minute
hypothermic group (Study 2). Of note, however, was the increased
difficulty in controlling the systemic parameters of the
30°C group versus the 33°C group because of the former group's
decreased respiratory rate and occasional cardiac arrhythmias
(Study 1, Table 2). Average
postanesthesia recovery time showed that the animals in the
33°C group recovered at a significantly faster rate (21±4 minutes,
P<0.05) compared with the normothermic animals
(37±4 minutes). No significant improvement in recovery time was found
in the 30°C group (28±4 minutes). In Study 2,
normothermic animals had an average weight loss of
approximately 28 g by postischemic day 1, whereas the
average weight loss for hypothermic animals was significantly lower
(P<0.05): 25 g (H0.5), 20 g (H1), 18 g (H2).
By 3 days postischemia, normothermic animals
had lost, on average, 51 g, and hypothermic animals lost 30 g
(H0.5), 31 g (H1), and 32 g (H2), but these differences were
not statistically significant.
View larger version (26K):
[in a new window]
Figure 1. Graphic representation of linear
regression model relating brain temperature to rectal temperature in
animals undergoing 2 hours of MCA occlusion. Data points
represent individual values obtained 10 minutes before
occlusion, 30 minutes intraischemically, and 10 minutes into
reperfusion.
View this table:
[in a new window]
Table 2. Physiological Variables at Baseline and During
Ischemic Period
(Study 1), 2b (Study 2 at 24 hours postischemia), and 2c
(Study 2 at 72 hours postischemia). A maximum score of 70
indicates that the animal died before the 24 or 72-hour endpoint.
View larger version (13K):
[in a new window]
Figure 2. Neurological deficit scores for each animal. A,
Study 1 at 24 hours postischemia in
normothermic (37°C), mildly hypothermic (33°C), and
moderately hypothermic (30°C) groups. B, Study 2 at 24 hours
postischemia in the normothermic group (N) and
in mildly hypothermic animals with hypothermic duration of 30 minutes
(H0.5), 1 hour (H1), and 2 hours (H2). C, Study 2 at 72 hours
postischemia (*P<0.05 by Kruskal-Wallis and
Mann-Whitney U test). . In the second study, 2 hours of mild
hypothermia decreased infarct area by 59% (hemispheric), 72%
(cortical), and 25% (striatal); 1 hour of mild hypothermia reduced
injury by 84% (hemispheric), 94% (cortical), and 60% (striatal). At
first these results may appear to show that 1-hour duration of mild
hypothermia is more protective than 2 hours, mainly because of a single
animal in the 2-hour duration group that had a very large infarct.
Therefore, this is a very conservative analysis that does not
exclude the outlier in the 2-hour group. Furthermore, there is no
statistically significant difference between the 1- and 2-hour duration
of hypothermia groups at any level studied. Thirty minutes of mild
hypothermia did not protect against neuronal damage. Figure 4 shows the infarct volumes obtained in
the various groups as determined from TTC-stained slices. There was a
good correlation between TTC- and H & E-stained tissue both at 24
(r=0.843) and 72 hours (r=0.917) postischemia. One hour of
hypothermia significantly reduced cerebral edema (1.9±1.3%,
P=0.002) compared with normothermic controls
(12.6±2.1%). Two hours or 30 minutes of hypothermia did not reduce
brain swelling to a significant degree (8.8±3% and 11.8±2.1%,
respectively). The mortality results for Study 2 were as follows: 50%
for the normothermic group, 44% for the 30 minutes
hypothermic group, and 0% for both 1- and 2-hour hypothermic
groups.
View larger version (20K):
[in a new window]
Figure 3. Percentage area of infarct for the hemisphere
(hem), cortex (ct), and striatum (st) at 24 hours
postischemia in normothermic (N), mildly
hypothermic (33°C), and moderately hypothermic (30°C) animals with
2 hours MCA occlusion. Data are expressed as mean±SEM. Hypothermic
animals showed significantly reduced neuronal damage compared with
controls by ANOVA and Dunnett's t test
(*P<0.05).
View larger version (23K):
[in a new window]
Figure 4. Quantification of infarct volume from
TTC-stained coronal sections for the hemisphere (hem), cortex (ct), and
striatum (st) in the normothermic group (N) and in mildly
hypothermic (33°C) animals with hypothermia duration of 30 minutes
(H0.5), 1 hour (H1), and 2 hours (H2) at 3 days
postischemia. Data are expressed as mean±SEM. Two hours of
mild hypothermia significantly decreased infarct volume in the cortex;
1 hour of mild hypothermia reduced injury to a significant degree in
the hemisphere, cortex, and striatum; whereas 30 minutes of mild
hypothermia did not protect against neuronal damage compared with
normothermic controls by ANOVA and the Student
t test (*P<0.05). , which shows that
for both MPO and TUNEL staining, the level of the AC (coronal level 3)
and coronal level 4 had the highest numbers of stained cells/nuclei.
Cells were counted as apoptotic only if they were
TUNEL-positive and showed characteristic nuclear morphology typical of
apoptosis (ie, cells containing pyknotic nuclei plus
apoptotic bodies). TUNEL staining was absent in animals that
survived 6 hours or less of reperfusion, but was prominent starting at
24 hours postischemia (data not shown). At 72 hours, 1 and
2 hours of mild hypothermia decreased TUNEL staining by 78% and 99%,
respectively, compared with normothermic controls. Thirty
minutes of mild hypothermia reduced the number of TUNEL-stained nuclei
by 17%, however, this was not statistically significant. Although
TUNEL-stained nuclei were found scattered throughout the
ischemic area, they tended to concentrate in the boundaries of
the striatum and were also prominent in the amygdala (Table 3, Figures 5 and 6). Results from DNA extraction at 72
hours postischemia showed evidence of internucleosomal
laddering in both ipsilateral cortex and striatum of
normothermic and hypothermic (2-hour duration) animals, but
no laddering was observed in the contralateral cortex or striatum in
either group (Figure 7). Although
visualization with ethidium bromide is not quantitative, it is
important to note that visualization of DNA laddering in hypothermic
animals was difficult, presumably because of the reduced number of
apoptotic cells observed in those animals.
View larger version (83K):
[in a new window]
Figure 5. TTC-stained coronal sections from
representative normothermic (N) and 2-hour
mildly hypothermic (H2) animals showing typical distribution of both
TUNEL (green) and MPO (yellow) staining. The level of the anterior
commissure (coronal level 3) and coronal level 4 contain the highest
numbers of stained cells/nuclei, and these numbers fall off in anterior
and posterior cuts. TUNEL-stained nuclei are found scattered throughout
the ischemic area, but concentrate in the boundaries of the
striatum, in the amygdala, and in the preoptic area.
View this table:
[in a new window]
Table 3. Quantification of Ipsilateral Apoptotic Nuclei and
Neutrophils at 72 Hours Reperfusion after 2-Hour
MCAO
View larger version (99K):
[in a new window]
Figure 6. Photomicrographs. A, DNA nick-end labeling of
cells in the cortex of a normothermic animal 72 hours after
transient focal cerebral ischemia (x100). Yellow arrows
indicate apoptotic cells (deep brown), black arrow shows a
necrotic cell with diffuse brown staining (counterstaining with methyl
green). B, High magnification of a labeled cell displaying the
characteristic features of chromatin condensation and apoptotic
bodies (arrow). C, TUNEL staining followed by a methyl green
counterstain in a respective cortical region of interest in a mildly
hypothermic animal (1-hour duration, x100); note absence of
TUNEL-positive cells. D, DNA nick-end labeling of apoptotic
cells (arrows) in mammary tissue (positive control) counterstained with
methyl green. E, Neutrophils (arrows) stained dark blue in
paraffin-embedded tissue of a normothermic animal 72 hours
postischemia. F, Respective region of interest in a mildly
hypothermic animal (2-hour duration, x250); note absence of
neutrophils.
View larger version (111K):
[in a new window]
Figure 7. Representative agarose gels
exhibiting DNA fragmentation in the cortex (Ctx) and striatum (St)
after 2 hours of MCA occlusion followed by 72 hours of reperfusion.
Lane 1, 123 bp standard DNA ladder; lane 2, right (contralateral
hemisphere to MCAo) cortex (R Ctx); lane 3 left (ipsilateral) cortex (L
Ctx); lane 4, right (contralateral) striatum (R St); lane 5, left
(ipsilateral) striatum (L St). 
, Figures 5 and 6). Normalized for infarct size, 2
hours of mild hypothermia decreased neutrophil accumulation by 57% in
the hemisphere, 27% in the cortex, and 76% in the striatum. One hour
or 30 minutes of hypothermia did not decrease PMNL infiltration after
normalization.
Discussion
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
Introduction
Mild hypothermia is already being used in the treatment of
traumatic brain injury18 and there is renewed
interest at the clinical level in developing guidelines for its
therapeutic use in stroke patients.19 Two key
issues that need to be addressed are the optimal depth and duration of
mild hypothermia. Although we find no differences in the degree of
neuroprotection afforded by either hypothermia group, in the
present studies we show that 33°C is associated with more stable
intraoperative hemodynamic and respiratory status as
well as with improved recovery from anesthesia compared
with 30°C. Furthermore, we show that just 1 hour of
intraischemic mild hypothermia is sufficient to obtain very
significant neuroprotection. Neurologic outcome and mortality are the
two most important endpoints in any clinical study on stroke. Most
experimental studies, however, use histologic outcome to determine the
efficacy of the therapeutic strategy in question. Here we report the
results of two studies that show a strong association between
histopathologic findings and neurobehavioral outcome.
Acknowledgments
We would like to thank the staff at the Department of
Comparative Medicine, Veterinary Services Center, Stanford
University, Stanford, Calif, for care and assessment of all animal
subjects; Karen Johnson for histological figures
preparation; and David Kunis for technical assistance. This research
was supported by NIH-NIMH Individual National Research Service Award
MH10748 (C.M.M.), by NIH-NINDS Grant RO1 NS27292 (G.K.S.), by K08
NS01860 (M.A.Y.), and by funding provided by Bernard and Ronni
Lacroute (G.K.S.).
References
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
Introduction
Editorial Comment
Effects on Neurologic Outcome, Infarct Size, Apoptosis, and Inflammation
Introduction
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
Introduction
The study by Maier et al is important in providing further
information regarding possible mechanisms for neuroprotection resulting
from mild hypothermia in a well-characterized model of transient focal
ischemia. Indeed, this study clearly demonstrates that even a
mild degree of hypothermia, when administered for the appropriate
duration of time, provides neuroprotection by preventing both
apoptosis and necrosis. It is not surprising that there is some
minimal amount of hypothermic time (dose) which is required for
efficacy; however, the degree by which transient, mild, hypothermia
prevents TUNEL staining and neutrophil accumulation is quite
impressive. In this study, the maximum duration of hypothermia was 2
hours, to coincide to the total duration of ischemia. Questions
that were not answered by this study relate to the therapeutic window
for this therapy and whether a longer duration of hypothermia would
have altered overall efficacy. As this therapy moves toward clinical
trials, it is also possible that details related to the rate and
mechanism of cooling and rewarming may become important in determining
ultimate efficacy.
This article has been cited by other articles:
 |
|
 |
|
  B. P. Meloni, F. L. Mastaglia, and N. W. Knuckey Review: Therapeutic applications of hypothermia in cerebral ischaemia Therapeutic Advances in Neurological Disorders, September 1, 2008; 1(2): 75 - 98. [Abstract] [PDF]
|
|
|
|
 |
|
 H. B. van der Worp, E. S. Sena, G. A. Donnan, D. W. Howells, and M. R. Macleod Hypothermia in animal models of acute ischaemic stroke: a systematic review and meta-analysis Brain, December 1, 2007; 130(12): 3063 - 3074. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Bardutzky and S. Schwab Antiedema Therapy in Ischemic Stroke Stroke, November 1, 2007; 38(11): 3084 - 3094. [Abstract] [Full Text] [PDF]
|
|
|
|