Download The Relationship Between Cerebrospinal Fluid Creatine Kinase and

The Relationship Between Cerebrospinal Fluid
Creatine Kinase and Morphologic Changes
in the Brain After Transient Cardiac Arrest
PER VAAGENES, M.D., JOHN KJEKSHUS, M.D.,
AND
ANSGAR TORVIK, M.D.
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
SUMMARY Recent studies have shown that the increase in creatine kinase (CK) in lumbar spinal fluid
(CSF) can effectively predict the outcome of cerebral ischemia after cardiac arrest. In the present investigation
maximum CSF-CK concentrations in 40 patients who died more than 3 days after successful cardiopulmonary
resuscitation were related to the histologic brain damage. The level of consciousness was assessed in all
patients during the time of survival. Neuronal necrosis was determined semiquantitatively according to a scoring system in sections from the frontal cortex, thalamus, hippocampus and the cerebellum. A curvilinear
relationship was found between maximum CSF-CK and extent and magnitude of defined histologic brain
damage. Maximum CSF-CK activity was found 48-72 hours after the anoxic episode. No cerebral damage
was found when maximum CSF-CK remained below 4 U/l.
The frontal cortex was less severely damaged than the other regions examined. CSF-CK increased to
between 4-10 U/I in patients without damage of the frontal cortex but with slight-to-severe damage of the
other regions. In deeply comatose patients, the CSF-CK activity always exceeded 10 U/I and neuronal
necrosis was found in all regions.
Approximately one-half of the patients had a diffuse loss of nerve cells in the cerebral cortex and hippocampus, while the other half had a patchy necrosis. The former patients generally had a rapid rise and fall in CSFCK, with a peak at 48 hours, while the latter group had a delayed rise and fall, with a maximum value at 72
hours. This difference may be due to a delayed necrosis of nerve cells caused by a no-reflow phenomenon in the
latter group.
Despite a selective vulnerability in certain parts of the brain, and a delayed necrosis of nerve cells in some
patients, we conclude that maximum CK activity in the lumbar CSF accurately and reliably measures the extent of the permanent brain damage and the functional outcome after cerebral ischemia.
SUCCESSFUL cardiopulmonary resuscitation after
transient cardiac arrest in man is limited by the
susceptibility of the central nervous system (CNS) to
ischemia and anoxia. Recent studies have shown that
the activity of creatine kinase (CK) in lumbar spinal
fluid (CSF) gives a good indication of the degree of
cerebral damage and prognosis in patients resuscitated after cardiac arrest." 2 Substantial increase in
CSF-CK was associated with severely depressed brain
function and death of all patients, while patients with
negligible rise in CSF-CK recovered without gross intellectual or sensory-motor defects.
Different regions of the brain show different sensitivity to anoxia.3 It is therefore conceivable that
ischemic damage to small but functionally important
areas of the brain may give only a small rise in CSFCK. Further, variations in the speed of release of CK
into the CSF and its appearance in the lumbar region
might influence the reliability of the method as an indicator of brain damage. The present study was undertaken to correlate maximum levels of CK in the lumbar CSF with the degree of ischemic damage in the
whole brain and in different brain regions at autopsy.
Materials and Methods
Forty patients who had survived more than 3 days
after successful cardiopulmonary resuscitation after
cardiac arrest were included in the study. None of
them had sustained hypotension before the arrest.
They were all intubated and ventilated with oxygen.
Sodium bicarbonate was given for correction of
metabolic acidosis during resuscitation. When indicated, lidocaine, 100 mg/min i.v., was given after a
priming dose of 75-100 mg. The patients were electrocardiographically monitored in a coronary care
unit and the arterial blood pressure was measured at
6-hour intervals. Cardiac decompensation was treated
with digitalis and diuretics and three patients received
dopamine for impending cardiogenic shock. A
neurologic examination was done daily and the level
of consciousness assessed according to a grading
system (table 1). Patients who had reacted to a mild
painful stimulus (grade II) were considered to have
been able to recover complete consciousness if they
had survived.
Lumbar puncture was performed in the third or
fourth lumbar interspace with the patient in the lateral
recumbent position, and CSF pressure was recorded.
When consent for repeated punctures was obtained,
4-5-ml serial samples were taken after 6, 24, 48 and 72
hours. The samples were collected in precooled vials
refrigerated at 4°C. Immediate cooling of the sample
to 4°C prevents denaturation of the CK-BB isoenzyme.' Analysis was done at 37°C on a LKB 8600
Reaction Rate Analyser within 24 hours. CK was
From the Departments of Medicine (VIII), Anesthesiology and
Pathology, Ulleval Hospital, and the Department of Medicine (B),
Rikshospitalet, Oslo, Norway.
Address for correspondence: John Kjekshus, M.D., Department
of Medicine B, Rikshospitalet, Oslo, Norway.
Received June 20, 1979; revision accepted November 11, 1979.
Circulation 61, No. 6, 1980.
1194
CSF-CK AND ISCHEMIC BRAIN INJURY/Vaagenes et al.
TABLE 1.
Grade
0
I
II
III
IV
Level of Consciousness
Description
Awake
Drowsy but responding to verbal commands
Responding to mild painful stimulation
Minimum response to maximal painful stimulation
No response to maximal painful stimulation
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
assayed according to the method of Oliver4 and
Rosalki.5 Activity below 2 U/l was considered normal.' All samples with CK more than 2 U/l were examined for CK isoenzymes according to the method of
Somer and Konttinen.8 Samples of CSF contaminated
with blood or containing MM or MB dimers were excluded.
At autopsy, the brain was fixed in formalin, sectioned and processed for light microscopy. Hematoxylin and eosin were used as routine stains. Histologic
sections were examined from the frontal pole, the
pyramidal layer of the hippocampus, the medial part
of the midthalamus and the cerebellar cortex. The
assessment of the ischemic damage in the sections was
made without knowledge of the clinical history or the
CSF-CK activities.
Semiquantitative evaluation of necrosis or loss of
nerve cells was made separately in the four regions acno
cording to the following system: grade 0
necrosis or cell loss of less than
damage; grade 1
25% of the neurons; grade 2 damage of 25-50% of
the neurons; and grade 3 damage of more than 50%
of the neurons. In the hippocampus, only the
pyramidal neurons were examined, and in the
cerebellum only the Purkinje cells.
In the frontal cortex and the hippocampus, a further
distinction was made between patchy necrosis and
diffuse necrosis. This was done to evaluate the role of
vascular factors, such as the no-reflow phenomenon
postulated by Ames et al.,7 or the possible effects of
focal edema. A similar subdivision was found to be
impractical in the thalamic and cerebellar sections. To
test the reproducibility of the grading, the material
was divided into four equal groups before the examination, and each group was examined successively.
After the first group was assessed, four cases unknown
to the examiner were mixed into the next group and so
on. In this way, four cases were graded separately four
times. The four gradings from these four cases were
nearly identical for all sections.
As expected, the sensitivity to ischemia in the
cerebral cortex was less than in the other regions examined (table 2). Because the neurons in the hippocampus, thalamus and cerebellum were about
equally affected, the gradings from these areas were
averaged from each case and denoted A and graded
from 0-3. Similarly, the changes in the frontal cortex
were denoted B and graded from 0-3. Region B
presumably reflected the damage of the whole cerebral
cortex and region A the most sensitive, but quan-
1195
titatively less extensive areas. A tentative
histopathologic score-list of the total brain damage
was then made as follows: total score 0 - A0 BO;
score I - Al BO, A2 BO, A3 BO; score II - Al Bl,
A2 Bl, A3 Bl; score III - A2 B2, A3 B2; score IV
A3 B3.
Results
We studied 11 female and 29 male patients. The
patients are listed according to maximum CSF-CK in
table 2. The mean age was 66 years (range 45-84
years). Previous findings have indicated a slightly
better prognosis for cerebral function in young
patients.' In the present study the mean age of the
deeply comatose patients (consciousness grades 3 and
4) was 65.4 years (n = 28), whereas patients with consciousness grades 0-2 had a mean age of 67.2 years
(n = 12). Thus, age did not appear to influence the
outcome of transitory cardiac arrest.
In 27 patients the cardiac arrest was due to acute
myocardial infarction, another five patients had
chronic ischemic heart disease and one patient had
aortic stenosis. In seven patients the arrest was
primarily due to extracardiac causes (acute
respiratory failure in five patients and profuse
hemorrhage in two). Artificial ventilation was continued after the initial resuscitation period in 29
patients, and maintained for an average of 4 days
(range 2 hours to 4 weeks). The time of survival
ranged from 3-40 days. The cause of death was considered to be cardiac ischemia or arrhythmia in 27
patients and pulmonary embolism in three.
Bronchopneumonia developed in 18 of the 40 patients,
and was considered to be the cause of death in nine.
One patient died from liver rupture.
The level of consciousness in all patients is given in
table 2. Generally, maximum CSF-CK and level of
consciousness correlated well.
Necrotic neurons can be easily identified when survival after resuscitation is more than 24 hours.
Necrotic neurons have a characteristic cytoplasmic
eosinophilia and a decreased nuclear size and
basophilia (fig. 1). After 3-4 days, perineuronal
microglial proliferation indicates early phagocytosis
(fig. 2), and by 1-2 weeks, loss of nerve cells is evident.
None of the cases had histologic evidence of delayed
or progressive necrosis of nerve cells.
In the cerebral cortex, the middle layers generally
were most severely damaged and the cortex around
the depth of the sulci was somewhat more affected
than the superficial parts. Frank tissue necrosis (infarction) was not observed in any of the cases. In the
hippocampus, only the pyramidal cell layer was
necrotic and the Sommer sector and the end plate
were most consistently affected. In the cerebellum,
only the Purkinje cells showed clear-cut necrosis.
These regional variations are in general agreement
with previous observations on the distribution of anoxic damage of cerebral cells.8
Reliable and reproducible semiquantitative es-
1196
CIRCULATION
VOL 61, No 6, JUNE 1980
TABLE 2. Maximum CSF-CK, Corresponding Levels of Consciousness and Histological Changes in 40 Patients
Histologic changest
Age CSF-CK
Level of
B
A
Pathologic Survival
Frontal cortex
Pt (years) (U/1)
consciousness* Hipp Thal Cere
scoret
(days)
14
0
0
0
2
0
0
1
0
67
4
0
0
0
0
2
81
0
0
2
40
0
0
o
o
0
0
74
3
3
20
0
0
0
0
4
0
0
4
64
6
0
0
4
1
I
0
69
5
I
3
1
0
I
0
0
6
77
I
5
3
1
0
0
1
I
I
45
6
7
0
1
I
1
1
7
48
6
I
8
0
0
0
7
0
9
72
8
0
I
7
1
II
1
1
2
10
76
II
9
16
1
2
II
64
3
3
11
11
II
6
0
I
0
3
12
12
II
3
69
7
2
III
3
1
12
13
72
Ill
3
6
II
1
3
14
III
3
3
75
14
9
III
2
52
3
3
III
3
15
15
12
2
3
III
3
3
16
81
15
3
51
3
18
3
3
2
17
IV
25
2
III
III
3
18
3
3
65
19
18
IV
3
2
20
IV
3
52
III
19
8
II
1
3
76
21
20
3
3
Iv
10
3
3
3
21
73
21
3
7
1
1
3
3
22
64
26
lv
III
6
IV
Iv
3
3
3
28
23
68
IV
II
1
II
2
3
1
24
29
51
7
IV
II
3
3
73
29
3
3
Iv
III
25
II
2
1
2
1
53
30
26
4
74
1
3
3
3
36
IVI
27
6
2
40
3
3
3
28
55
IV
6
3
3
45
3
29
67
IV
3
IV
3
64
3
30
3
IV
51
3
3
Iv
IV
8
31
3
3
61
3
3
IV
57
8
IV
32
3
61
3
3
3
72
IV
10
ill
IV
33
3
3
60
3
3
81
III
13
2
34
III
54
3
3
3
82
III
7
2
III
35
56
3
3
3
89
IVI
3
36
IV
56
3
3
3
3
96
IV
3
79
37
IV
3
3
3
98
IV
a
76
38
3
IV
3
127
3
3
IV
8
39
76
3
130
3
IV
3
IV
4
40
84
132
3
3
3
3
IV
*See table 1.
tSee text for grading system.
Abbreviations: CSF-CK = creatine kinase activity in cerebrospinal fluid; Hipp - hippocampus; Thal =
thalamus; Cere = cerebellum.
-
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
timates of the nerve cell damage could be made. The
neuronal necrosis was often patchy, with necrotic and
normal areas alternating. The size of each damaged
area was only 1-2 mm in diameter and could not be
explained on the basis of regional variations. In other
specimens the neuronal damage was evenly distributed. Of the 40 patients examined, patchy necrosis
was found in 19. The patchy neuronal necrosis was
suggestive of the no-reflow phenomenon.7' 9 No common clinical denominator that could explain this
CSF-CK AND ISCHEMIC BRAIN INJURY/Vaagenes et al.
4*
CSF-CK and type of cortical damage
C
100
;:0. .
p
Ae
4'
S9
1197
Diffuse
Patchy
4:
4
*
80-
D
U
60-
40-
IL
o
t; *i
7
:
,Q
x0ir0;f:5
20-
0
q*
24
48
72
96
120
144
HOURS
t
S
;
. ;
';S
*.X
;
eD:0z0:
:-fS
0;s+.
ai
M: "_
--i.^
@
4'
i
,.
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
9
_its
*s
k
;.
..
',..;.
... ...
o
A4f.
a ,ei^
e
0'.
* ,
.@
0:.
e*a.*.
t.
:
FIGURE 1. Group of necrotic neurons in the hippocampus
(arrows) 3 days after cardiac resuscitation. Hematoxylin
and eosin; magnification X 200.
feature was found. However, the CSF-CK timeactivity curves separated the patients with diffuse from
those with patchy necrosis, especially in the cortex
(fig. 3). Patients with diffuse necrosis generally had a
rapid rise in CSF-CK, reaching the peak 48 hours
after the resuscitation, followed by a rapid return to
basal activity. In contrast, patients with patchy
necrosis had a delayed rise in CSF-CK, reaching a
maximum value at 72 hours, with a poorly defined
peak and a slowly declining tail. Patients with diffuse
? V~
4
4'
*:N'0
.S
:f
X
4
....
'
a t$ *-
X *'t
4ff
ffWSsv
pNi04,
necrosis had a higher activity at 48 hours than those
with patchy necrosis (68.5 + 9 and 37.5 ± 8.5 U/1,
respectively, p < 0.05). The different evolution of the
enzyme release strongly suggests that patchy necrosis
is associated with a delayed ischemic condition. An
exponential curve-fit of pathologic score and CSF-CK
gave a coefficient of determination of r2 = 0.87 in
patients with exclusively diffuse type of injury, while a
poorer fit was obtained when all patients were pooled.
As expected, damage in the pyramidal layer of
the hippocampus was generally more severe than in
the frontal cortex. Maximum damage could occur in
the hippocampus in cases with slight or no damage in
the frontal cortex (table 2). The changes in the
thalamus, cerebellum and hippocampus were approximately of the same magnitude and occurred at
considerably lower CSF-CK values than those in the
frontal cortex (table 3).
No ischemic damage was observed when CSF-CK
activity remained below 4 U/1. If patients survived for
more than 6-10 days, there was no residual neurologic
dysfunction.' Damage exclusively in the thalamus,
cerebellum or hippocampus was associated with a
modest rise in CSF-CK, the maximum activity never
exceeding 11-12 U/1. Long-term survivors in this
group remained invariably with mnemonic and
reading difficulties, inconsistently incoordinate
movements, but never with loss of motor function.' Involvement of.the frontal cortex was always associated
with CSF-CK activity above 10 U/1 and persistence of
depressed consciousness and motor functions. These
findings correspond closely to the previous observations' that the functional outcome of the cerebral
ischemia was poor when CSF-CK rose above 10 U/1,
while intellectual functions remained when CSF-CK
did not exceed 10 U/1. A rise in CSF-CK below 10
U/1 generally involved only a fraction of the cells in
the thalamus, hippocampus and cerebellum, while at a
CSF-CK activity above 10 U/1, the damage invariably
encompassed the majority of the cells in these regions.
Maximum CSF-CK changes and histologic scores,
demonstrated good agreement (fig. 4). A nearly cur-
:X'.~X.0,~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~.............
;a:
FIGURE 3. Sequential changes of cerebrospinal fluid
creatine kinase activity (CSF-CK) in patients with exclusively diffuse (n - 11) or patchy (n = 13) cortical
necrosis (mean ± SEM).
: *: f:Ftff.
Ei: :S:Xw a t:......
.a. 4~.~ . .....
Aa
FIGURE 2. Necrotic neurons surrounded by microglial
cells 9 days after cardiac resuscitation. Hematoxylin and
eosin; magnification X 200.
CIRCULATION
1198
VOL 61, No 6, JUNE 1980
TABLE 3. Mean CSF-CK Activity (U/l) at Different Grades of Damage in the Four Regions
Frontal cortex
Thalamus
Cerebellum
Neuronal damage
Hippocampus
5.1 1
3.8 1
4 1
5.2 1
0: No damage
n = 9
n =6
n = 6
n =10
1:
< 25%
22 3.5
n= 8
11.3 6
n=4
13.4 3
n= 7
7.5 2
n=2
2:
25-50%o
36.5 11
n =8
16.7 - 7
n = 3
29
n =1
24 - 6
n=2
3:
> 50%
70.4 11
n = 14
51.3 - 8
n = 27
52.7 9
n = 25
n = 26
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
vilinear relationship exists between the levels of consciousness and CSF-CK (fig. 4).
The lumbar pressure, which presumably reflects the
intracranial pressure, was measured in 30 patients.
The pressure rose slightly in all patients, but gross
cerebral edema was only registered in four cases. In
eight patients with maximum CSF-CK below 10 U/l,
the pressure rose to 206 ± 25 mm H2O, compared
with 231 ± 20 mm H20 in 11 patients with CSF-CK
above 30 U/I. This difference was significant
(p < 0.05), but the groups did not differ with respect
to gross cerebral edema.
Discussion
All patients in the present study had a short and
definite period of anoxia, and there was no prolonged
hypoxemia or hypotension before or after the cardiac
arrest. Generally, a close relationship was found
between enzymatic and histologic evidence of ischemic
brain damage. Patients with maximum CSF-CK less
than 4-5 U/l regained consciousness without
neurologic sequelae or histologic damage. Enzymatic
[o levelof conciousness
*
pathological
score
c)
0
0
1
11
Ill
lV
level/score
FIGURE 4. Maximum cerebrospinal fluid creatine kinase
activity (CSF-CK) as related to the level of consciousness
and pathologic score (mean ± SEM).
49
7
activity of 4-9 U/l was associated with neuronal
damage restricted to the hippocampus, thalamus and
cerebellum, while the frontal cortex was largely
spared. The hippocampal areas have important functions in the memory process and even modest
elevations in CSF-CK may indicate permanent functional disturbances in accordance with the clinical
findings. Although restricted and important areas of
the brain are more sensitive than others to anoxia, the
enzymatic method apparently is sufficiently sensitive
to detect damage even in those areas. Enzyme activity
above 10 U/l was invariably associated with severe
necrosis in the same regions and with variable damage
of the frontal cortex. All these patients were in a deep
stupor or comatose, confirming previous studies.'
Although individual variations occurred, the extent of
the ischemic injury generally paralleled the increase in
CSF enzyme activity. Only one section was examined
from the cerebral cortex and none from the basal
ganglia. Further, the arterial boundary zones were
only examined grossly. Regional variations in the distribution of the damage may therefore have occurred
and not been detected by the applied procedure. Such
variations could explain individual discrepancies
between the CK values and the grading of damage.
Accordingly, the best conformity was observed in
patients with the diffuse type of ischemic damage.
The cause of the delayed appearance of CK in the
CSF compared with the rapid release of enzymes into
the blood in myocardial infarction' is uncertain. In
contrast to the delayed appearance after anoxic
damage, cerebral contusion gives a rapid increase of
CK in the CSF.10 The tissue architecture of the brain
is largely intact after anoxia, whereas contusion gives
a complete and superficial tissue necrosis. Apparently,
the diffusion caused by anoxia of protein molecules
through the intercellular space toward the CSF is a
very slow process, whereas the exchange occurs more
freely from areas of complete tissue necrosis. However, the rapid in vivo denaturation of the enzyme'
makes it unlikely that this is the sole explanation for
the slow increase of the CSF-CK. Alternatively,
delayed appearance of CK could be due to a
protracted cell death caused by unknown metabolic or
vascular mechanisms. Ames has described a postischemic no-reflow phenomenon caused by small-
CSF-CK AND ISCHEMIC BRAIN INJURY/Vaagenes et al.
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
vessel obstruction.7 More recent demonstration that
the phenomenon was markedly blunted by flushing
with serum" suggests the occurrence of platelet plugs.
The patchy distribution of the loss of nerve cells in the
hippocampus and frontal cortex in almost 50% of the
patients in the present study might suggest a delayed
vascular factor in addition to the systemic ischemic
episode. This is strongly supported by the finding that
patchy necrosis was associated with more delayed and
protracted CSF-CK release than was diffuse necrosis.
In the latter group, maximum CSF-CK was higher,
peaked 24 hours earlier and returned more rapidly to
baseline than in patients with patchy necrosis. Because
blood pressure was restored similarly in both groups,
the patchy necrosis is probably caused by regional
cerebral ischemia persisting after successful cardiopulmonary resuscitation. To our knowledge, this is the
first time that evidence for a no-reflow phenomenon
has been shown in man. The present study suggests
that the no-reflow phenomenon exerts its effect several
hours after the general ischemic episode. However, the
histologic changes did not indicate a protracted or
progressive necrosis of nerve cells beyond 24 hours
after the resuscitation.
This investigation confirmed our previous observation that increase of enzymes in the lumbar CSF
reflects a permanent cerebral injury. Considering the
high activity of CK in normal brain tissue,". 12 only
small amounts are recovered in the CSF after anoxia.
It is therefore likely that only a small fraction of enzymes released from the brain reaches the lumbar
space. It has been shown that it takes 2 hours for substances injected into the lateral ventricle to reach
equilibrium with the CSF in the lumbar space.'3 The
denaturation of CK in the CSF in vitro at 37°C is very
rapid, with a fall in activity to 50% after 2 hours.' This
might explain the relatively low values observed in the
lumbar CSF. The brain-specific isoenzyme fraction
(CK-BB) did not appear in peripheral blood, and loss
of enzymes into the blood can therefore be disregarded.
There are conflicting views regarding the usefulness
of enzyme studies in the CSF of patients with diseases
of the CNS."4-16 Different methods and errors may be
partly responsible for the different results. We consider the time of sampling of the lumbar fluid after an
insult to the brain to be particularly important,
because the maximum enzyme activity is found 48-72
hours after the incident. This is particularly important
in cases with relatively low activity, because there are
narrow limits between the groups with cerebral
1199
damage and those with satisfactory cerebral restitution. Hypothermia reduces the rate of denaturation of
CSF-CK' and may lead to overestimation of the exdamage using the limits reached in the
present study. Taking due regard of these factors, the
present investigation shows that the brain-specific CKBB isoenzyme determined in the CSF is a useful index
of ultimate brain damage after global cerebral
ischemia, and is of potential value in clinically assessing the no-reflow phenomenon.
tent of brain
References
1. Kjekshus J, Vaagenes P, Hetland 0: Assessment of cerebral
injury with spinal fluid creatine kinase (CSF-CK) in patients
after cardiac resuscitation. Scand J Clin Lab Invest. In press
2. Maas AIR, Cerebrospinal fluid enzymes in acute brain injury.
I. Dynamics of changes in CSF enzyme activity after acute experimental brain injury. J Neurol Neurosurg Psych 40: 655,
1977
3. Brierly JB, Meldrum BS, Brown AW: The threshold and
neuropathology of cerebral "anoxic-ischemic" cell change.
Arch Neurol 29: 367, 1973
4. Oliver IT: A spectrometric method for determination of
creatine phosphokinase and myokinase. Biochem J 61: 116,
1955
5. Rosalki SB: An improved procedure for serum creatine
phosphokinase determination. J Lab Clin Med 69: 696, 1967
6. Somer H, Konttinen A: Demonstration of serum creatine
kinase isoenzymes by fluorescence technique. Clin Chem Acta
40: 133, 1972
7. Ames A, Wright L, Kowada M, Thurston JM, Majno G:
Cerebral ischemia. II. The no-reflow phenomenon. Am J
Pathol 52: 437, 1968
8. Brierley JB: Cerebral hypoxia. In Greenfield's Neuropathology,
3d ed, edited by Blackwood W, Corsell JAN. London, Edward
Arnold, 1976, pp 43-85
9. Fischer EG, Ames A, Hedley-Whyte ET, Gorman S: Reassessment of cerebral capillary changes in acute global ischemia and
their relationship to the "no-reflow phenomenon." Stroke 8: 36,
1977
10. Nordby HK, Tveit B, Ruud I: Creatine kinase and lactate
dehydrogenase in cerebrospinal fluid in patients with head injuries. Acta Neurochir (Wien) 32: 209, 1977
11. Dawson DM, Fine IH: Creatine-kinase in human tissues. Arch
Neurol 16: 175, 1967
12. Kleine TO: Zur Lokalisation der Creatinkinase (CK) in
Mitochondrien und Microsomen von Skelemuskel, Hertz und
Hirnrinde des Menschen. Klin Wochenschr 43: 504, 1965
13. Bowser D: Cerebrospinal Fluid Dynamics in Health and
Disease. Springfield, Illinois, Charles C Thomas, 1960
14. Lisak RP, Graig FA: Lack of diagnostic value of creatine
phosphokinase assay in spinal fluid. JAMA 199: 750, 1967
15. Sherwin AL, Norris JW, Bulcke JA: Spinal fluid creatine
kinase in neurologic disease. Neurology 19: 993, 1969
16. Hildebrand J, Levin S: Enzymatic activities in cerebrospinal
fluid in patients with neurological disease. Acta Neurol BeIg 73:
229, 1973
The relationship between cerebrospinal fluid creatine kinase and morphologic changes in
the brain after transient cardiac arrest.
P Vaagenes, J Kjekshus and A Torvik
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
Circulation. 1980;61:1194-1199
doi: 10.1161/01.CIR.61.6.1194
Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1980 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7322. Online ISSN: 1524-4539
The online version of this article, along with updated information and services, is located on
the World Wide Web at:
http://circ.ahajournals.org/content/61/6/1194.citation
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally
published in Circulation can be obtained via RightsLink, a service of the Copyright Clearance Center, not the
Editorial Office. Once the online version of the published article for which permission is being requested is
located, click Request Permissions in the middle column of the Web page under Services. Further
information about this process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation is online at:
http://circ.ahajournals.org//subscriptions/