(Stroke. 1995;26:2127-2134.)
© 1995 American Heart Association, Inc.
Articles |
Regional Prevalence and Distribution of Ischemic Neurons in Dog Brains 96 Hours After Cardiac Arrest of 0 to 20 Minutes
From the Safar Resuscitation Research Center and the Departments of Anesthesiology/Critical Care Medicine and Pathology, University of Pittsburgh (Penn) Medical Center.
Correspondence to Ann Radovsky, DVM, PhD, National Institute of Environmental Health Sciences, PO Box 12233, Mail Drop A0-01, Research Triangle Park, NC 27709. E-mail radovsky@niehs.nih.gov.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Background and Purpose In this established outcome model of cardiac arrest in dogs, we have used total (summed regional) brain histopathologic damage scores. The present study describes the regional progression of necrotic (ischemic) neuron prevalence with increasing duration of cardiac arrest. It tests the hypothesis that increases in the total prevalence of necrotic neurons better correspond to increasing arrest duration and better correlate with neurological deficit than do any individual regional scores.
Methods Blinded evaluation with light microscopy was used to score the prevalence (five categories) and note the distribution of necrotic neurons in dog brains 96 hours after normothermic ventricular fibrillation cardiac arrest followed by standard reperfusion and control of extracerebral variables. Six coronal brain sections including 19 regions were examined from dogs subjected to 0 (n=2), 5 (n=5), 10 (n=6), 12.5 (n=12), 15 (n=8), 17 (n=5), or 20 (n=1) minutes of cardiac arrest. Dogs were neurologically evaluated before death.
Results Necrotic neurons were widespread and scattered
among normal neurons. Individual regions varied in their sensitivity to
different durations of cardiac arrest. There were consistent
increases in the mean prevalence of necrotic neurons with increased
arrest duration in the hippocampal dentate gyrus and for cerebellar
granule neurons. Regionally, the caudate nucleus had the best
correlation with clinical neurological deficit (
=+.85,
P<.01).
Conclusions Compared with total (summed regional) necrotic neuron prevalence scores, increased regional prevalence scores for cerebellar granule neurons with increasing arrest duration were equally significant, and scores for the caudate nucleus had nearly the same correlation with individual clinical neurological deficit.
Key Words: animal models • cerebral ischemia • heart arrest • neuronal death • dogs
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Many different clinical and histopathologic scoring systems are used to compare the relative severity of postischemic brain damage. Recent investigations have focused intensely on histopathologic scoring of individual neurons in postischemic rat1 2 brain cortex at relatively short periods after insult. Such intense focus on individual neurons necessarily limits evaluation to relatively small areas of the brain. Typically, a brain region is chosen for close scrutiny because of its vulnerability to ischemia and clinical relevance (eg, CA1 hippocampal pyramidal neurons3 4 ). Different regions have been studied in different models of ischemia. In rodent models of forebrain ischemia, the CA1 region of the hippocampus is frequently used to compare therapies.5 6 A recent test of therapeutic hypothermia in cats after 15 minutes of cardiac arrest used morphometry of neuronal damage on the cingulate gyrus, parietal cortex, and hippocampus at one coronal level.7 Counting techniques used in dog brain after 15 minutes of cardiac arrest have included parietal cortex, hippocampus, and cerebellum.8 9
The outcome model in dogs of the present study has been used by our group for many years to evaluate special therapies to mitigate brain damage. Previous studies have reported total brain histopathologic damage scores consisting of a weighted overall combined ischemic neuron prevalence and infarct score.10 11 12 13 14 Overall brain damage, rather than damage in any one specific region, has been assumed to have most significance.
The present study investigates whether there is a region or neuronal population in the brain that has particular sensitivity to the duration of ischemia or that has particular correlation with clinical neurological deficit in this dog outcome model of cardiac arrest. Identification of such a region might enable the use of a more rigorous method of quantitation of histopathologic brain damage than the one currently used.
In the present study, dead (ischemic) neurons at the time point studied are referred to as necrotic, but no inferences about the mechanism of neuronal cell death are intended. Whether neurons died via apoptosis, oncosis,15 or some other mechanism is presently unknown. The descriptions of the regional progression of distribution of ischemic neurons with increasing duration of cardiac arrest in the present study will help in selecting appropriate durations of cardiac arrest and regions of interest to facilitate future studies of therapeutic interventions in this model.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Microscopic slides and clinical data used in this retrospective study came from studies made between 1988 and 1993 in young adult male coonhound dogs weighing 18 to 25 kg. These studies were conducted by the same core research team, using the same type of dogs, type of insult (ventricular fibrillation), reperfusion method, and details of resuscitation and life support. Neurological deficit scoring criteria were the same. Only the duration of cardiac arrest varied.10 11 12 13 14 16
Dogs were normothermic controls in studies evaluating special therapies, such as resuscitative hypothermia. Data came from dogs with durations of total circulatory arrest as follows: 0 minutes (n=2); 5 minutes (n=5); 10 minutes (n=6); 12.5 minutes (n=12); 15 minutes (n=8); 17 minutes (n=5); and 20 minutes (n=1). One author (A.R.) was responsible for determining brain histopathologic damage scores without knowing the insult or treatment used. Details of the protocol of these studies (circulatory arrest, resuscitation, and control of extracerebral variables by intensive care) have been previously published.10 11 12 13 14 16 All projects were approved by the University of Pittsburgh's Animal Care and Use Committee in accordance with the National Research Council's Guide for the Care and Use of Laboratory Animals.
Dogs were anesthetized with inhaled nitrous oxide and halothane, had external transthoracic electric shock to induce cardiac ventricular fibrillation for the specified time period, and underwent resuscitation with 3 to 5 minutes of total cardiopulmonary bypass, followed by external electric defibrillation and a short period of partial bypass for temperature control. As has been previously reported,11 electrically induced ventricular fibrillation resulted in a decrease of mean arterial pressure from 100 to 0 mm Hg (no blood flow) within 30 seconds. After the arrest period, total cardiopulmonary bypass enabled the return to normal mean arterial pressure within 1 minute of reperfusion and defibrillation and restoration of heartbeat within 1 to 3 minutes. Of the 2 dogs with 0-minute arrest, one was an "operated" sham, and the other was perfusion-fixed under anesthesia without being instrumented.
The abrupt onset of the ischemic insult and its equally abrupt
reversal eliminated a prolonged low blood flow condition that could
complicate interpretation of the effects of the applied insult
duration. Routinely, standardized controlled ventilation with
normotension and normothermia was maintained for 20 hours, and
intensive care was maintained for 96 hours after the arrest. All dogs
were provided the level of nursing, hydration, and nutritional support
that their condition required. Clinical neurological deficit testing as
previously reported (Table 1
)17 was performed at 96
hours after insult. Dogs were examined for level of consciousness,
cranial nerve function, and overall gross and fine motor and sensory
capabilities. Clinical neurological deficit was graded from 0% to
100%, with 0% representing no deficit and 100% deficit
brain death. Significant overall functional impairment was seen with
scores greater than 25%. Clinical and histopathologic evaluations were
performed by different investigators who did not exchange information
until all data had been generated.
|
At 96 hours after the insult, the dogs were reanesthetized and killed by retrograde infusion of 3 L of 3% buffered paraformaldehyde solution into the cranial vasculature through the aortic arch, using a peristaltic pump at a pressure of approximately 100 mm Hg. The brains were removed from the skulls 24 hours later and immersion-fixed in 3% paraformaldehyde for at least another week. Brains were then sliced coronally at 3-mm intervals. Six coronal sections were chosen for processing (levels R25, R20, R15, R5, CR0, and C10, rostral and caudal to the interaural line).18 Sections included the following 19 anatomic regions: frontal, parietal, occipital, insular, and temporal cortex; pyramidal neurons and neurons of the dentate gyrus of the hippocampus; caudate nucleus, putamen, globus pallidus, and amygdala; thalamus; midbrain and substantia nigra; pons; cerebellar dentate nucleus, Purkinje cells, and granule cells; and medulla. When regions were represented at more than one level, scoring was done at one level consistently (eg, the hippocampus was evaluated at the most rostral level [level R15] where damage was more extensive than that on level R5). Sections were processed, paraffin-embedded, sectioned at 6-µm thickness, and stained with hematoxylin-eosin-phloxine.
Light microscopic evaluations of all regions of all brains were made using a Reichert Microstar Series 10 microscope. Additionally, 10 of the 39 brains were examined for cytoplasmic fluorescence of necrotic neurons with a Nikon Optiphot fluorescence microscope with a 480-nm filter. Manual counting of fluorescent neurons in the rostral hippocampus (level R15) of one case from each duration arrest group was performed, and results were compared with regional severity scores based on prevalence. For consistency with previous methods,19 20 the prevalence of necrotic neurons in each region was given a numerical severity score in multiples of 2. A score of 0 signified that no ischemic neurons were seen in the region, 2 (minimal), 4 (mild), 6 (moderate), and 8 (marked). A minimal score (2) included the range from a single necrotic neuron to approximately 25% of the susceptible population. The worst necrotic neuron prevalence score in a given region (both sides) was 16. The sum of categorical scores for regional prevalence throughout the brain was the total prevalence of necrotic neurons.
Criteria for scoring relative prevalence of necrotic neurons varied in different anatomic regions because of regional differences in the number of neurons observed to be susceptible to postischemic necrosis. Since even with the longest arrest duration the neurons in any region were never observed to be all necrotic, the susceptible population was defined as the maximum seen to be necrotic by that reviewer in the most damaged example of that region.
Jonckheere's test, a nonparametric dose-response test, was used to assess the significance of arrest duration–related changes in prevalence, both for each region and for the summed total prevalence throughout the brain. Spearman's correlation was used to compare the necrotic neuron prevalence scores of each region and of the total score with individual clinical neurological deficit scores.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Necrotic neurons were characterized by cytoplasmic hypereosinophilia and hyperreflective "glassiness," no visible Nissl substance, and nuclear condensation (usually pyknosis) and no visible nucleolus. Except in the thalamus, dentate gyrus, and among cerebellar Purkinje and granule neurons where they remained rounded, necrotic neurons generally were shrunken and had triangular cytoplasmic and nuclear outlines. The absence of neurons (apparently disintegrated, homogenized, or shrunken out of the plane of section) was inferred as necrosis. Such inference was possible only for the large, relatively evenly spaced and aligned hippocampal pyramidal cells and cerebellar Purkinje cells.
In the cases evaluated with fluorescence microscopy, necrotic
neurons in all anatomic regions had brightly fluorescent
cytoplasm (Figs 1 and 2). Manual counts
of all the fluorescent neurons in the rostral hippocampus of
one case with average severity of damage from the 5-, 10-, 12.5-, 15-,
and 17-minute arrest groups did not correspond to the rank of either
the duration of arrest or of neurological deficit of the animals.
"Dark" neurons with hyperchromic basophilic cytoplasm, marginated
Nissl substance, and a visible nucleolus were considered artifacts and
not necrotic.
|
|
Necrotic neurons were scattered among normal neurons, and different
regions had characteristic patterns of distribution with increased
arrest duration (Fig 3). There were
increases in prevalence of necrotic neurons with increase in duration
of cardiac arrest for all regions, but most of the mean scores for
individual regions had plateaus or inconsistent changes
(Fig 4). Necrotic neurons were most
prevalent in the lateral caudate, putamen, superior and lateral
neocortex, hippocampus, and among cerebellar Purkinje cells. Neurons of
the substantia nigra, midbrain, medulla, and pons were very rarely
necrotic, even in the longest arrest groups. The two dogs with
0-minute arrests had mean summed scores of 6 (sham at 96 hours)
and 2 (control), resulting from unilateral scores of minimum in the
caudate nucleus of both dogs and in parietal cortex and putamen of the
sham.
|
|
The number of dogs showing damage in a particular region increased with increased duration of arrest. After a 5-minute arrest, there were occasional necrotic neurons scattered among large neocortical neurons (in 5 of 5 dogs), among hippocampal pyramidal cells (CA1 sector) (3 of 5), among small superior-lateral neurons in the caudate and putamen (2 of 5), and in cerebellar Purkinje cells (2 of 5). With increasing duration of arrest, necrotic neurons in the areas affected in the 5-minute arrest group were more prevalent, and additional areas developed necrotic neurons. In the 10-minute arrest group, all 6 of 6 dogs had increased prevalence of necrotic neurons in neocortical regions, hippocampal pyramidal neurons, and in the caudate and putamen; 5 of 6 had necrotic Purkinje cells, and 4 of 6 had necrotic neurons in the dorsimedial thalamus. After 12.5-minute arrest, increased damage included 9 of 12 dogs with necrotic cerebellar granule cell neurons and 7 of 12 dogs with necrotic neurons in the dentate gyrus of the hippocampus. In the 15-, 17-, and 20-minute arrest groups, increased brain damage scores reflected increases in the regional scores of the dentate gyrus, the cerebellar granule neurons, and some hippocampal pyramidal neurons.
Jonckheere's test assessed the relative significance of arrest
duration–related increases in prevalence scores with increases in
duration of cardiac arrest. The total necrotic neuron prevalence score
had a stronger arrest duration–related response than any
individual brain region except for the regional cerebellar granule
neuron scores, which were equally significant (P<.01). Mean
total (summed regional) and mean regional cerebellar granule neuron
prevalence scores, as well as mean neurological deficit scores for each
duration arrest group, are given in Table 2. Other
regional prevalence scores had less correspondence with changes in
arrest duration than did the total scores.
|
Spearman's correlation coefficients for major regional necrotic neuron
prevalence scores versus individual clinical neurological deficit
scores are given in Table 3. All regions examined had
significant (P<.01) positive correlation coefficients. The
=.86 correlation coefficient of the total (summed regional) scores
was nearly matched by the .85 coefficient of the regional caudate
nucleus scores.
|
Cortex
An increased severity score could result from either a wider
distribution of necrotic neurons or a greater density of necrotic
neurons in scattered sites. The susceptible population for most of the
neocortex consisted of approximately 50% of the visible neurons.
Necrotic neurons were seen primarily in the upper and middle laminae,
first in the superior and middle gyri and with increasing duration of
ischemia in inferior gyri, and finally, with some
inconsistency, in the cingulate gyri. With increasing
duration of arrest, more laminae became involved. In some cortical
areas of dogs with circulatory arrest of 12.5 minutes or longer, there
were segments with necrosis of neurons in all laminae. In comparison to
the neocortex, relatively few necrotic neurons were found in either the
insular cortex or in the temporal cortex. Both insular cortex and
ventral temporal cortex had relatively numerous "dark"
neurons.
Caudate Nucleus and Putamen
Necrotic neurons of the caudate nucleus and putamen had shrinkage,
angulation, and increased cytoplasmic eosinophilia similar to cortical
neurons, but they had more of a tendency to blend into the surrounding
neuropil because of their smaller size. Large neurons in the caudate
nucleus were relatively unaffected, even with long durations of arrest.
There were necrotic neurons in lateral portions of the caudate nucleus
adjacent to the internal capsule in all arrest groups. Neurons in the
middle and lastly subependymal areas became necrotic after longer
durations of arrest. The small neurons of the putamen became necrotic
at relatively short arrest durations. Their number increased rapidly
with increasing duration of arrest until most of the smaller neurons
were necrotic in the 15-minute and longer groups.
Hippocampal Formation
In the hippocampus, neurons of the CA1 sector were virtually all
necrotic at arrest times greater than 10 minutes, whereas other sectors
of the hippocampus were spared even after 20 minutes of arrest. At 96
hours after arrest, in all arrest groups except the 5-minute duration
group, many hippocampal pyramidal cells were
missing—disintegrated, homogenized with the neuropil,
or shrunken out of the plane of section—and thus were invisible
to conventional light microscopy. Because of the regular arrangement of
these large aligned neurons, spaces or "glassy" neuropil were
evident where pyramidal cells would exist normally, and
these spaces were evaluated as evidence of neuronal necrosis.
Postischemic necrosis of the very small neurons in the rostral hilar portion of the dentate gyrus of the hippocampus occurred after relatively short arrest durations, but with increasing duration of ischemia, neurons in the entire rostral gyrus became necrotic. These neurons became shrunken, had pyknotic or karyorrhectic nuclei, and had only a thin rim of hypereosinophilic cytoplasm.
Thalamus
In the superior medial thalamus, scattered large neurons became
necrotic and had hypereosinophilic cytoplasm and condensed nuclei.
Necrotic thalamic neurons, unlike postischemia necrotic
neurons in other regions, often remained rounded, and many appeared
swollen rather than becoming shrunken and angular. Because of the
relative density and eosinophilia of the thalamic neuropil, necrotic
thalamic neurons were not as readily distinguishable as were necrotic
cortical neurons.
Cerebellum
Cerebellar Purkinje cells were considered necrotic when they had
cytoplasmic hypereosinophilia and nuclear pyknosis or were absent in
the plane of section. In groups of 12.5-minute arrest or longer, sites
of apparently disintegrated Purkinje cells were marked by vacuoles.
Absent Purkinje cells increased with arrest duration until, in the 17-
and 20-minute arrest groups, essentially all Purkinje cells were
missing, and only a margin of vacuoles separated the junction of the
molecular and granular cell layer of the cerebellum.
Necrotic cerebellar granule neurons were first noted in superior folia. They remained rounded but were shrunken and had pyknotic nuclei and a small rim of hypereosinophilic cytoplasm. With increasing duration of circulatory arrest, necrotic cerebellar granule cells became more widespread but never were observed to be the majority of the population.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The hypothesis that total (summed regional) necrotic neuron prevalence scores for the brain would better correspond to arrest duration and better correlate with clinical neurological deficit scores than prevalence scores for any individual region in this model was proved only marginally correct. Furthermore, the expected finding that longer durations of transient cardiac arrest would result in increased prevalence of necrotic neurons in the brain was not demonstrated for most regions and not for arrest durations of longer than 12.5 minutes for total prevalence.
The 96-hour survival period with a standardized period of controlled
ventilation and intensive care enabled effective clinical neurological
deficit scoring that was unimpaired by extracerebral variables. By
96 hours after arrest, neuronal necrosis was easily observed using
conventional light microscopy, but at this time point there was
appreciable loss of two regional neuronal populations: cerebellar
Purkinje neurons (after arrests of 12.5 minutes and longer) and
hippocampal pyramidal neurons (primarily in the CA1 region
after arrests of 10 minutes and longer). Disappearance of necrotic
hippocampal pyramidal neurons after longer arrest durations
may have caused the lack of correspondence of the manual count of
fluorescent neurons to either arrest duration or neurological
deficit. Loss of CA1 hippocampal neurons and Purkinje neurons by 3 days
after an 18-minute ischemic insult in dogs has previously been
reported.21 Because in rodents the speed with which
neurons die varies in different neuronal populations22 23
and with different severities of insult,24 the speed with
which necrotic neurons disappear regionally in dogs might also be
related to insult severity (duration of cardiac arrest). Loss of
irregularly spaced or small neurons could not be appreciated in the
present scoring method and may account for apparent decreases (Fig 4
) in visible necrotic neurons in regions such as the
caudate or putamen after arrest durations of 12.5 minutes or longer. In
the present study, the prevalence scores included necrotic neurons
that had disappeared from the hippocampal pyramidal neurons
and cerebellar Purkinje neurons because the loss of these neurons
caused appreciable gaps in the architecture of the region.
The prevalence of necrotic cerebellar granular neurons had the strongest regional correspondence with increasing arrest duration, which was equally significant to that of the total prevalence score. Because cerebellar granule neurons were minimally affected until arrest durations were 10 minutes or longer, and maximum prevalence (after 20-minute arrest) was less than 25% of the population, this population can be considered relatively resistant to ischemic insult. This insensitivity may be reflected in less disappearance of necrotic neurons after longer duration arrests when evaluated at 96 hours after arrest and resuscitation, thus making the prevalence of necrotic neurons in the region more consistent with arrest duration than regions in which necrotic neurons disappeared. Nuclear pyknosis in cerebellar granule cells in human brain has previously been considered as artifact (part of the complex of artifacts that includes dark, hyperchromatic neurons) by Cammermeyer.25 In the present study, fluorescence of the cytoplasm of these cerebellar granule neurons, perhaps due to lipofuscin accumulation, was considered confirmation of their necrosis.
The positive (r=.86; P<.01) correlation of total necrotic neuron prevalence in the brain with individual clinical neurological deficit is consistent with our previously reported studies.10 11 12 13 14 The finding that a single region, the caudate nucleus, had nearly as strong a correlation was unexpected. Neurological deficit scoring in dogs has a strong motor component. The correlation of regional necrotic neuron prevalence scores for the caudate nucleus with neurological deficit is consistent with a report that this region, associated with the control of voluntary movement, has a strong influence on the motor behavior of dogs.26 Ideally, neurobehavioral tests specific to each region and independent of damage in other regions should have been made. In the present study, dogs were untrained, and facilities and expertise for sophisticated neurobehavioral tests were not available. Strong correlation of a regional severity of brain damage with the total neurological deficit suggests that damage in that region strongly contributed to total deficit.
Correlation between regional necrotic neuron prevalence and
neurological deficit scores (Table 3) shows
exceptionally poor (=.24) correlation for the dentate gyrus. The
accuracy of scoring of the dentate gyrus region is supported by
consistent correspondence of regional prevalence scores with
changes in duration of cardiac arrest. These findings suggest that
neuronal necrosis in the rostral dentate gyrus of dogs has limited
functional consequence.
The experiments described in the present article are unique. The cardiac arrest insults were carefully controlled (abrupt-onset cessation and immediate restoration of circulation) global ischemic insults. Electrically induced cardiac ventricular fibrillation and immediate reinstatement of postinsult blood flow using brief cardiopulmonary bypass eliminated the low blood flow states that characterize many of the naturally occurring ischemic states. A "watershed" pattern of neuronal necrosis based on worsened ischemia in the boundary zones of the major arteries was neither expected nor detected in the present model of abrupt onset and recovery of cerebral blood flow. Additionally, individual variation in arterial supply and extensive intracranial collateral circulation27 make defining boundary zones in dogs difficult.
The widespread distribution of neuronal necrosis in this model is consistent with descriptions of humans28 29 and of dogs after experimental transient global ischemia.30
Additional studies at different time points after cardiac arrest with this dog model will determine the time course of regional necrotic neuron prevalence. Although most necrotic neurons at the 96-hour time point examined in the present study were shrunken, many necrotic neurons in the thalamus were swollen. This regional difference in morphology may be important, since it has been suggested that swelling versus shrinkage of dying cells is an important distinction in determining the mechanism of neuronal cell death.15
Specific markers to enable practical automated counting of individual necrotic neurons are not available, and manual counting of individual necrotic neurons throughout all affected regions in the large brain of the dog is not feasible. Identification of the caudate nucleus as clinically relevant, and of cerebellar granule neurons as having consistent responses to duration of arrest, suggests that manual counting of individual neurons in these regions might be worthwhile. However, the model of the present study is used to test therapeutic interventions, and different therapeutic interventions might affect different regions differently. Rigorous counts in limited regions have the disadvantage of not revealing changes in the severity of damage in other (uncounted) regions.
In conclusion, the present study demonstrated regional variability in the morphology and vulnerability of necrotic neurons after different durations of cardiac arrest; fluorescence of hematoxylin-eosin-phloxine–stained necrotic neurons; obvious disappearance of necrotic hippocampal pyramidal neurons and cerebellar Purkinje neurons by 96 hours after longer arrest durations; and identification of the caudate nucleus as the region where neuronal prevalence scores are most closely correlated with observed neurological deficit and identification of cerebellar granule neuron prevalence scores as being as sensitive to increases in cardiac arrest duration as total prevalence scores.
|
Acknowledgments |
|---|
This study was supported by the A.S. Laerdal Foundation and National Institutes of Health grant NS-24446. Thanks to Joe Haseman for statistical consultation and guidance; Patrick Kochanek, Joel Mahler, Mike Elwell, and Deryck Read for valuable suggestions; Feng Xiao, Kah Ming Sim, and Antonio Capone for their help with the experiments; Greg Fisher and the Carnegie Mellon Light Microscopy Imaging Facility for use of fluorescence microscopes interfaced with computer counting equipment; and David Lennard, Hayes Brown, and John Horton for help with photomicrography.
Received March 27, 1995; revision received August 8, 1995; accepted August 8, 1995.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1. Eke A, Conger KA, Anderson M, Garcia JH. Histologic assessment of neurons in rat models of cerebral ischemia. Stroke. 1990;21:299-304.
2. Eke A, Conger KA. Classifying cells from light microscopic bit features by binary logic. Lab Invest. 1989;61:243-252. [Medline] [Order article via Infotrieve]
3. Schmidt-Kastner R, Freund TF. Selective vulnerability of the hippocampus in brain ischemia. Neuroscience. 1991;40:599-636. [Medline] [Order article via Infotrieve]
4. Auer RN, Jensen ML, Whishaw IQ. Neurobehavioral deficit due to ischemic brain damage limited to half of the CA1 sector of the hippocampus. J Neurosci. 1989;9:1641-1647. [Abstract]
5. Busto R, Dietrich WD, Globus MY, Ginsberg MD. Postischemic moderate hypothermia inhibits CA1 hippocampal ischemic neuronal injury. Neurosci Lett. 1989;101:299-304. [Medline] [Order article via Infotrieve]
6.
Kirino T, Tamura A, Sano K. A reversible type
of neuronal injury following ischemia in the gerbil
hippocampus. Stroke. 1986;17:455-459.
7. Horn M, Schlote W, Henrich HA. Global cerebral ischemia and subsequent selective hypothermia. Acta Neuropathol (Berl). 1991;81:443-449. [Medline] [Order article via Infotrieve]
8. Kumar K, White BC, Krause GS, Indrieri RJ, Evans AT, Hoehner TJ, Garritano AM, Koestner A. A quantitative morphological assessment of the effect of lidoflazine and deferoxamine therapy on global brain ischemia. Neurol Res. 1988;10:136-140. [Medline] [Order article via Infotrieve]
9. Kumar K, Krause G, Koestner A, Hoehner T, White B. Effect of flunarizine on global brain ischemia in the dog: a quantitative morphologic assessment. Exp Neurol. 1987;97:115-127. [Medline] [Order article via Infotrieve]
10. Sterz F, Leonov Y, Safar P, Radovsky A, Stezoski SW, Reich H, Shearman GT, Greger TF. Effect of excitatory amino acid receptor blocker MK-801 on overall, neurologic, and morphologic outcome after prolonged cardiac arrest in dogs. Anesthesiology. 1989;71:907-918. [Medline] [Order article via Infotrieve]
11.
Sterz F, Leonov Y, Safar P, Radovsky A, Tisherman SA,
Oku K-I. Hypertension with or without hemodilution after cardiac
arrest in dogs. Stroke. 1990;21:1178-1184.
12. Leonov Y, Sterz F, Safar P, Radovsky A, Oku K-I, Tisherman S, Stezoski SW. Mild cerebral hypothermia during and after cardiac arrest improves neurologic outcome in dogs. J Cereb Blood Flow Metab. 1990;10:57-70. [Medline] [Order article via Infotrieve]
13.
Leonov Y, Sterz F, Safar P, Radovsky A. Moderate
hypothermia after cardiac arrest of 17 minutes in dogs.
Stroke. 1990;21:1600-1606.
14. Kuboyama K, Safar P, Radovsky A, Tisherman SA, Stezoski SW, Alexander H. Delay in cooling negates the beneficial effect of mild resuscitative cerebral hypothermia after cardiac arrest in dogs: a prospective, randomized study. Crit Care Med. 1993;21:1348-1358. [Medline] [Order article via Infotrieve]
15. Majno G, Joris I. Apoptosis, oncosis, and necrosis: an overview of cell death. Am J Pathol. 1995;146:3-15. [Abstract]
16. Angelos M, Safar P, Reich H. External cardiopulmonary resuscitation preserves brain viability after prolonged cardiac arrest in dogs. Am J Emerg Med. 1991;9:436-443. [Medline] [Order article via Infotrieve]
17. Safar P, Sterz F, Leonov Y, Radovsky A, Tisherman S, Oku K. Systematic development of cerebral resuscitation after cardiac arrest. Three promising treatments: cardiopulmonary bypass, hypertensive hemodilution, and mild hypothermia. Acta Neurochir Suppl (Wien). 1993;57:110-121. [Medline] [Order article via Infotrieve]
18. Lim RKS, Liu C-N, Moffit RL. Stereotaxic Atlas of the Dog's Brain. Springfield, Ill: Charles C Thomas Publishing; 1960.
19.
Nemoto EM, Bleyaert AL, Stezoski SW, Moossy J, Rao G,
Safar P. Global brain ischemia: a reproducible monkey
model. Stroke. 1977;8:558-564.
20. Vaagenes P, Cantadore R, Safar P, Mossy J, Rao G, Diven W, Alexander H, Stezoski W. Amelioration of brain damage by lidoflazine after prolonged ventricular fibrillation cardiac arrest in dogs. Crit Care Med. 1984;12:846-855. [Medline] [Order article via Infotrieve]
21. Sato M, Hashimoto H, Kosaka F. Histological changes of neuronal damage in vegetative dogs induced by 18 minutes of complete global brain ischemia: two-phase damage of Purkinje cells and hippocampal CA1 pyramidal cells. Acta Neuropathol (Berl). 1990;80:527-534. [Medline] [Order article via Infotrieve]
22. Pulsinelli WA, Brierly JB, Plum F. Temporal profile of neuronal damage in a model of transient forebrain ischemia. Ann Neurol. 1982;11:491-498. [Medline] [Order article via Infotrieve]
23. Smith M-L, Auer RN, Siesjö BK. The density and distribution of ischemic brain injury in the rat following 2-10 min of forebrain ischemia. Acta Neuropathol (Berl). 1984;64:319-332. [Medline] [Order article via Infotrieve]
24. Araki T, Kato H, Kogure K. Selective neuronal vulnerability following transient cerebral ischemia in the gerbil: distribution and time course. Acta Neurol Scand. 1989;80:548-553. [Medline] [Order article via Infotrieve]
25. Cammermeyer J. A critique of neuronal hyperchromatosis. J Neuropathol. 1960;19:141-142.
26. Shapovalova KB, Iakunin IV, Boiko MI. Participation of the head of the caudate nucleus of the dog in mechanisms of conditioned reflex postural reorganization. Zh Vyssh Nerv Deiat Im I P Pavlova. 1984;34:669-677. [Medline] [Order article via Infotrieve]
27. Arai T, Andou Y, Tabo E, Fujitani T, Nagaro T. A multi-arterial clamping method with arterial hypotension does not bring about complete brain ischemia in dogs. Resuscitation. 1991;22:237-244. [Medline] [Order article via Infotrieve]
28. Brierley JB, Graham DI, Adams JH, Simpsom JA. A clinical, neurophysiological, and neuropathological report of two cases. Lancet. 1971;2:560-565. [Medline] [Order article via Infotrieve]
29.
Kinney HC, Korein J, Panigrahy A, Dikkes P, Goode
R. Neuropathological findings in the brain of Karen Ann
Quinlan. N Engl J Med. 1994;330:1469-1475.
30. Vanicky I, Marsala M, Murar J, Marsala J. Prolonged postischemic hyperventilation reduces acute neuronal damage after 15 min of cardiac arrest in the dog. Neurosci Lett.. 1992;135:167-170. [Medline] [Order article via Infotrieve]
This article has been cited by other articles:
 |
|
 |
|
  A. Nozari, P. Safar, S. W. Stezoski, X. Wu, S. Kostelnik, A. Radovsky, S. Tisherman, and P. M. Kochanek Critical Time Window for Intra-Arrest Cooling With Cold Saline Flush in a Dog Model of Cardiopulmonary Resuscitation Circulation, June 13, 2006; 113(23): 2690 - 2696. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Radovsky, L. Katz, U. Ebmeyer, and P. Safar Ischemic Neurons in Rat Brains After 6, 8, or 10 Minutes of Transient Hypoxic Ischemia Toxicol Pathol, September 1, 1997; 25(5): 500 - 505. [Abstract] [PDF]
|
|
|
|
|
|
A. H. Idris, L. B. Becker, J. P. Ornato, J. R. Hedges, N. G. Bircher, N. C. Chandra, R. O. Cummins, W. Dick, U. Ebmeyer, H. R. Halperin, et al. Utstein-Style Guidelines for Uniform Reporting of Laboratory CPR Research: A Statement for Healthcare Professionals From a Task Force of the American Heart Association, the American College of Emergency Physicians, the American College of Cardiology, the European Resuscitation Council, the Heart and Stroke Foundation of Canada, the Institute of Critical Care Medicine, the Safar Center for Resuscitation Research, and the Society for Academic Emergency Medicine Circulation, November 1, 1996; 94(9): 2324 - 2336. [Full Text]
|
|
|
|
 |
|
 P. Safar, F. Xiao, A. Radovsky, K. Tanigawa, U. Ebmeyer, N. Bircher, H. Alexander, and S. W. Stezoski Improved Cerebral Resuscitation From Cardiac Arrest in Dogs With Mild Hypothermia Plus Blood Flow Promotion Stroke, January 1, 1996; 27(1): 105 - 113. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||