Resuscitation 42 (1999) 221–229
Normoxic ventilation during resuscitation and outcome from
asphyxial cardiac arrest in rats
Christopher A. Lipinskia, Shawn D. Hicksb, Clifton W. Callawayb,*
aThe Department of Emergency Medicine, Wayne State Uni?ersity, Detroit Recei?ing Hospital, 4201 St. Antoine, Detroit, MI 48201, USA
bThe Department of Emergency Medicine, Uni?ersity of Pittsburgh, 230 McKee Place, Suite 400, Pittsburgh, PA 15213, USA
Received 9 December 1998; received in revised form 28 May 1999; accepted 28 May 1999
The formation of reactive oxygen species during reperfusion is one trigger for neuronal injury after global cerebral ischemia.
Because formation of reactive oxygen species requires delivery of molecular oxygen to ischemic tissue, restricting inspired oxygen
during reperfusion may decrease neurological damage. This study examined whether ventilation with room air rather than pure
oxygen during resuscitation would improve neurological recovery after cardiac arrest in rats. Adult, male rats were subjected to
8 min of asphyxia resulting in cardiac arrest. During resuscitation, rats were ventilated either with hyperoxia (FiO2=1.0) or
normoxia (FiO2=0.21, room air). Neurobehavioral deficits were scored daily for 72 h after resuscitation, after which brains were
collected for histology. Normoxia decreased arterial oxygen content. Other physiological parameters and mortality did not differ
between groups. All surviving rats exhibited behavioral and histological signs of brain damage. Neurological deficit scores did not
differ between normoxia and hyperoxia conditions at any time point. The number of ischemic neurons in the hippocampus also
did not differ between groups. These data indicate neither benefit nor detriment of reducing inspired oxygen concentration during
resuscitation from asphyxial cardiac arrest in rats. © 1999 Elsevier Science Ireland Ltd. All rights reserved.
Keywords: Asphyxia; Brain ischemia; Heart arrest; Neurologic dysfunction; Resuscitation
Successful resuscitation from cardiac arrest oc-
curs in approximately 70000 people per year in
the USA, but the prognosis for these patients is
poor. In fact, some estimates indicate that less
than 1% of those who survive after cardiac arrest
will return to their original level of functioning .
The advent of treatments that have greater rates of
resuscitation (e.g. early defibrillation) increases the
Oxidative damage plays a prominent role in
neuronal injury after global ischemia [2,3]. More-
over, it is known that cellular antioxidants includ-
ing glutathione, ascorbate, and vitamin E are
depleted within 10 min of return of spontaneous
circulation in rats . Interestingly, antioxidant
depletion occurs only during reperfusion of previ-
ously ischemic brain and not during the ischemic
event [4–7]. Lipid peroxidation results from this
oxidative stress and has been used as a marker of
reperfusion injury after brain ischemia [3,6,7].
Lipid peroxides appear in brain during ischemia,
and increase further during early reperfusion .
It has been proposed that reperfusion of is-
chemic brain with oxygenated blood stimulates the
generation of reactive oxygen species, and that
reducing the oxygen tension during reperfusion
decreases this component of reperfusion injury. In
fact, reduced oxygen concentrations during reper-
fusion improve recovery of ischemic spinal cord
and cerebral cortex [9,10], and reducing inspired
oxygen during resuscitation from cardiac arrest in
dogs improves neurological outcome when as-
* Corresponding author. Tel.: +1-412-647-9047; fax: +1-412-578-
E-mail address: firstname.lastname@example.org (C.W. Callaway)
0300-9572/99/$ - see front matter © 1999 Elsevier Science Ireland Ltd. All rights reserved.
C.A. Lipinski et al. / Resuscitation 42 (1999) 221–229222
sessed after 24 h [8,11,12]. In dogs, the improved
neurological recovery after 24 h of reperfusion is
associated with a decrease in lipid peroxidation
products . However, neuronal injury progresses
for several days after resuscitation from cardiac
arrest, and longer observational periods are neces-
sary to accurately assess functional and structural
neuronal injury [5,13]. Thus, this study compared
the effects of normoxic (FiO2=0.21) and hyper-
oxic (FiO2=1.0) ventilation during resuscitation
on neurobehavioral deficits and histological injury
for 3 days after asphyxial cardiac arrest in rats.
We hypothesized that reducing oxygen concentra-
tions during resuscitation would reduce reperfu-
sion injury to the ischemic brain.
The asphyxial cardiac arrest model has been
described in detail previously [4,14,15]. The cur-
rent protocol was performed in complete compli-
ance with the US Department of Health and
Human Services ‘Guide for the Care and Use of
Laboratory Animals’ (NIH Publication No. 86-23)
and was approved by the Institutional Animal
Care and Use Committee of the University of
A total of 22 male, Sprague–Dawley rats (300-
350 g) (Harlan Sprague–Dawley Inc., Indiana-
polis, IN) were prepared and instrumented in the
same fashion with random assignment to treat-
ment groups. Rats were housed two per cage
under humidity and temperature-controlled condi-
tions. Food and water were available at all times.
A standard 12 h:12 h light:dark schedule was
maintained, with experiments performed during
halothane in oxygen) was induced by way of a
modified face-mask. Rats were orotracheally intu-
bated via direct laryngoscopy with a 2 inch×14
gauge catheter, after which halothane was reduced
to 0.8% and titrated to the lightest possible surgi-
cal plane. Mechanical ventilation was maintained
with a Harvard rodent ventilator (tidal volume of
9 ml/kg, rate of 40 breaths/min) (Harvard Appara-
tus, South Natick, MA). The femoral artery and
vein were cannulated via cutdown with catheters
fashioned from PE-50 tubing. These catheters al-
lowed central venous drug administration, contin-
uous arterial blood pressure monitoring, and
arterial blood sampling. Temperature was moni-
tored via a tympanic probe (YSI-500, Cole-
Parmer, Indianapolis, IN).
Two minutes prior to asphyxia, animals were
paralyzed with vecuronium (2 mg/kg), halothane
was discontinued, and FiO2was reduced to 0.21
(room air). In other studies we have noted that
halothane-anesthetized rats that are not paralyzed
require an interval longer than this washout period
to emerge from anesthesia. Asphyxia was induced
by disconnecting the ventilator circuit at the end-
expiration for a total of 8 min. Apnea produced
immediate hypertension followed by bradycardia
and hypotension that progressed within 3 min to
pulseless electrical activity or asystole. The hyper-
tensive bout at the onset of asphyxia is associated
with hypoxemia .
Resuscitation was accomplished by restarting
mechanical ventilation (tidal volume of 9 ml/kg,
rate of 60 breaths/min) with either an FiO2of 0.21
(room air, normoxia group) or 1.0 (hyperoxia
group). The investigator was not blinded to treat-
ment group. Epinephrine (0.005 mg/kg) and bicar-
bonate (1 mEq/kg) were given intravenously and
chest compressions were performed at 200 com-
formed using two fingers tapping the lower
one-half of the sternum with sufficient force to
displace the sternum at least 5 mm. The opposite
hand was used to stabilize the torso of the rat.
This regimen reliably restored spontaneous circu-
lation in all rats. In pilot studies, we confirmed
that the alveolar–arterial oxygen gradient in-
creases in rats resuscitated from asphyxia , and
that if severe hypoxemia develops during reperfu-
sion, rats usually will experience a second cardiac
arrest. To prevent the potential confound of this
secondary ischemic event in the normoxia group,
an FiO2of 0.21 was maintained strictly for 10 min.
If the 10 min blood gas demonstrated severe hy-
poxemia associated with hypotension that was un-
responsive to a fluid bolus, supplemental oxygen
was titrated to achieve normoxemia (paO2of 95–
110). Only two rats in the normoxia group actu-
ally required supplemental oxygen. This protocol
provided a difference in arterial oxygen tension
between the hyperoxia and normoxia groups while
Arterial blood gases were sampled during the
preparation and at 10, 30 and 60 min after resusci-
tation. Base deficits greater than 5 mEq/l were
C.A. Lipinski et al. / Resuscitation 42 (1999) 221–229223
replaced with intravenous sodium bicarbonate (1
mEq/ml, 0.05 ml×base deficit). Persistent hy-
potension (MAP?60 mmHg) was corrected with
intravenous boluses of normal saline (1 ml/kg
every 10 min). After stabilization, cannulated ves-
sels were ligated and wounds were closed.
One hour after resuscitation, rats were weaned
from mechanical ventilation, and all supplemental
oxygen. Rats were returned to individual cages
where they had free access to food and water.
During recovery and until resumption of sponta-
neous activity, rats were warmed by an overhead
lamp adjusted to maintain a tympanic temperature
of 36–37°C. Supplemental normal saline (20 ml/
kg) was administered subcutaneously each day.
Rat behavior was assessed using a standardized
neurological deficit score (NDS) at 24, 48, and 72
h after resuscitation. In previous studies, the NDS
was positively correlated with global histological
damage (Table 1) . The NDS ranges from 0%
(no deficit, normal animal) to 100% (brain dead).
paradigms could not be performed because of the
spasticity and motor deficits exhibited by the rats
After 72 h of recovery, rats were deeply anes-
thetized with halothane and transcardially per-
fused with 100 ml of lactated ringers solution
followed by 3% paraformaldehyde. Brains were
removed and fixed in 1.5% paraformaldehyde
prior to paraffin embedding and sectioning. An
investigator who was unaware of the treatment
examined 6 ?m, cresyl violet-stained sections of
bregma −3.3 according to the atlas of Paxinos
and Watson ). Viable and ischemic neurons
were counted in the CA1, CA2, CA3 regions of
the hippocamus and in the upper and lower blades
of the dentate gyrus. Ischemic neurons were recog-
nized by their pyknotic nuclei which lacked a clear
Physiological variables for the two groups were
compared using unpaired t-tests. Neurological
deficit scores were compared using Mann–Whit-
ney U tests for nonparametric data. Mortality and
the proportions of ischemic neurons were were
compared using chi-square tests. The significance
level was set at P?0.05 for all tests. Based upon
the variances observed in prior studies using this
model [4,14,15], a total of ten rats per experimen-
tal group were expected to provide 80% power to
detect a 10% difference in neurological deficit
scores with P?0.05.
Mortality did not differ between hyperoxia and
normoxia groups. A total of ten rats were resusci-
tated with hyperoxic ventilation, and 12 rats were
resuscitated with normoxic ventilation. Total anes-
thesia time (97?30 min), time from onset of
asphyxia to circulatory arrest (206?12 s) and
resulting duration of complete ischemia (274?12
s) were not significantly different between groups.
Total duration of chest compressions required to
restore circulation (100?73 s) did not differ be-
tween groups or between survivors and nonsur-
vivors. Of the rats resuscitated with hyperoxic
ventilation, 70% were alive after 24 h, and 60%
were alive after 72 h. Of the rats resuscitated with
normoxic ventilation, 58% were alive after 24 h,
and 50% were alive after 72 h.
Physiological measures other than arterial oxy-
gen tension did not differ between groups (Table
2). Mean hemoglobin concentrations (12.6?1.0
Neurological deficit score for ratsa
0, 10, 20%Consciousness(absent, abnormal,
Respiration 0, 10, 20%
0, 5, 10%Limb movement(absent, abnormal,
Limb sensation0, 5, 10%
Stops at edge
aNeurological deficit score=100%–total score.
C.A. Lipinski et al. / Resuscitation 42 (1999) 221–229 224
10 min30 min Group60 min Baseline
72 h sur?i?ors
? 72 h sur?i?ors
72 h sur?i?ors
?72 h sur?i?ors
72 h sur?i?ors
?72 h sur?i?ors
72 h sur?i?ors
?72 h sur?i?ors
72 h sur?i?ors
?72 h sur?i?ors
72 h sur?i?ors
?72 h sur?i?ors
* Significantly different from the hyperoxia group, P?0.05.
g/dl) did not differ between groups. Consistent
with prior studies using this model, reperfusion
was associated with a large metabolic acidosis at
10 min that resolved after 30 min. The brief hyper-
tension observed immediately after resuscitation
was followed by a fall in mean arterial blood
pressure that reached its nadir between 20 and 30
min. All rats were stabilized after 60 min, and
could be weaned from the ventilator.
All rats exhibited neurological deficits during
the recovery period after asphyxia. Consistent with
prior studies using this model, rats were globally
injured, exhibiting abnormal posture, difficulty
with limb coordination, and spasticity. Rats re-
C.A. Lipinski et al. / Resuscitation 42 (1999) 221–229225
quired supplemental fluids and hand-feeding dur-
ing recovery. Despite these measures, both groups
lost weight (mean change?S.D.= −62.3?6.8 g)
over the 72 h recovery.
The neurological deficit scores for rats surviving
to the 72 h time point are illustrated in Fig. 1.
There was no significant difference between groups
at any time point. The large variance in NDS
observed in the normoxia group was due to a
single rat with an NDS at 24, 48 and 72 h of 70,
40, and 70%. The hyperoxia group had one outlier
at 72 h with an NDS of 50%. Autopsy and review
of physiological measures revealed no obvious
cause for the increased severity of neurological
injury in these two rats. Analysis of the NDS data
with and without these outliers demonstrated no
significant between-group difference at any time
point. Excluding these outliers, the mean?S.D.
NDS at 72 h for the hyperoxia group and for the
normoxia group were 10?7.1% and 12?9.1%
respectively. The 95% confidence interval for the
Ischemic neurons were recognized in CA1, CA2,
and CA3 areas of the hippocampus as shrunken,
darkened nuclei with loss of a clear nucleolus (Fig.
2). These morphological changes have been de-
scribed after cardiac arrest in rat  and in dog
. The dentate gyrus exhibited minimal histolog-
ical signs of damage in cresyl violet-stained sec-
tions. Total viable neurons per high power field
did not differ between groups (?2=0.63, NS) (Fig.
3). For comparison, the number of viable neurons
per high power field in the CA1 region of unoper-
ated sham rats (69?7) did not differ significantly
from the total number of neurons in the asphyxi-
ated rats (Fig. 3). Likewise, the number and pro-
portion of ischemic neurons per high power field
did not differ between groups.
This study addressed the effect of inspired oxy-
gen concentration on the neurological injury in-
curred after resuscitation from cardiac arrest in
rats. Our results indicate that reducing inspired
oxygen concentrations during reperfusion pro-
duces no significant improvement in either neuro-
logical deficit or histological injury at 24, 48, or 72
h after resuscitation. Thus, normoxic ventilation
Fig. 1. NDSs of rats resuscitated with normoxic ventilation or
hyperoxic ventilation at 24, 48, and 72 h. Scores were not
significantly different between groups at any time point.
Fig. 2. Photomicrographs of cresyl violet-stained sections
from the CA1 region of the hippocampus from a sham
operated rat (top) and from a rat sacrificed 72 h after 8 min
of asphyxial cardiac arrest and resuscitation (bottom) (mag-
C.A. Lipinski et al. / Resuscitation 42 (1999) 221–229 226
Fig. 3. Total number of viable neurons (top) and ischemic
neurons (bottom) in five regions of the hippocampus from
rats resuscitated with normoxic ventilation or hyperoxic venti-
lation. CA, cornu ammoni (areas 1, 2, and 3); DGU, upper
blade of the dentate gyrus; and DGL, lower blade of the
arterial oxygen content. Based upon the measured
hemoglobin concentrations at baseline, the arterial
oxygen content for both normoxia and hyperoxia
groups was estimated to be 18.0?1.7 ml/dl. How-
ever, the partial pressure of oxygen is more impor-
tant for affecting oxidative processes at the cellular
level, because diffusion of oxygen from capillaries
to individual cells is driven by the gradient of
partial pressures of oxygen. The rate of diffusion
is also a function of the distance of a given cell
from the capillary (d), and the diffusion coefficient
of the tissue (Kd) :
Oxygen reaching cell=Kd×d
Thus, the oxygen content of blood determines how
much oxygen is available to be extracted by the
tissue, but the partial pressure of oxygen actually
determines the concentration of oxygen available
to drive reactions outside the vasculature. The fact
that increasing FiO2from 0.21 to 1.0 during resus-
confirms that a change in partial pressures can
alter chemical reactions . A reduction in reactive
oxygen species may result in a greater ability of
injured neurons to translate new proteins by pre-
venting phosphorylation, or enhancing dephos-
phorylation of the eukaryotic initiation factor 2?
[10,21,22]. Furthermore, reducing oxidative injury
to lipid membranes could preserve cell integrity.
Several differences may account for the differ-
ences between the present study and prior studies.
In this study rats were observed for 72 h after
reperfusion, which is longer than in past subacute
experiments [8,11,12]. It is known that the mani-
festation of neuronal injury often requires days,
due in part to delayed neuronal death that is
triggered by a multitude of signals [5,13]. Histolog-
ical signs of neurological injury develop over days
after reperfusion [17,23–25]. These data were the
basis for our choice of survival time. Recently,
several studies have illustrated that neuroprotec-
tion by drugs or hypothermia that is apparent at
short survival times is often undetectable after
longer survival times . Even clinical studies
advise against prognosticating about the neurolog-
ical recovery of patients in anoxic or other
metabolic coma before 48 or 72 h after their insult
. The present data confirm that neurological
during resuscitation from asphyxial cardiac arrest
is not neuroprotective by itself. Whether normoxic
ventilation during resuscitation could enhance
other neuroprotective interventions, such as free
radical scavengers, is still unknown.
Previous studies have advocated reduction of
the partial pressure of oxygen during resuscitation
from cardiac arrest [8,10–12,19]. In these studies,
hyperoxic ventilation resulted in similar hyperox-
emia. These studies proposed that normoxic venti-
lation and resultant normoxemia might reduce the
rate of formation of reactive oxygen species in
reperfused brain. Because arterial hemoglobin is
nearly saturated with oxygen even during nor-
moxic ventilation, increasing inspired oxygen con-
centration probably does not significantly alter
C.A. Lipinski et al. / Resuscitation 42 (1999) 221–229227
deficits continue to stabilize over the first 3 days
after resuscitation (Fig. 1).
Differences in species (rat versus dog), model
(asphyxia versus ventricular fibrillation) and anes-
thesia (halothane versus barbiturate) also may
contribute to the difference in results between this
study and previously reported studies in dogs
[8,11,12]. Even within a single species, differences
have been noted between cardiac arrest induced by
asphyxia and cardiac arrest induced by ventricular
fibrillation . In particular, asphyxial cardiac
arrest produces more severe functional deficits and
histological damage than a comparable duration
of ventricular fibrillation cardiac arrest. Further-
more, drug therapies that are effective for amelio-
rating brain damage after ventricular fibrillation
cardiac arrest are sometimes ineffective for im-
proving recovery from asphyxial cardiac arrest
. Interestingly, hypoxia during reperfusion also
does not improve recovery from cerebral ischemia
in swine .
Asphyxia does produce more severe histological
injury in dogs than does immediate cessation of
blood flow such as in ventricular fibrillation ,
perhaps making it a more difficult model in which
to demonstrate any improvement after an inter-
vention. However, the neurological deficits ob-
served in the present study were moderate and
comparable to the deficits observed in other stud-
ies that demonstrated effects of therapeutic inter-
ventions after asphyxial cardiac arrest [15,30].
Finally, delayed formation of free radicals may
occur days after reperfusion. For example, super-
oxide radicals are increased three to five days after
reperfusion in the hippocampus of gerbils sub-
jected to cerebral arterial occlusion . Thus, the
reduction of oxygen tension during the first hour
of reperfusion may not have been the appropriate
duration of therapy. Against this possibility, data
obtained from the same rat model employed in
this study indicates that oxidative stress, as as-
sessed by depletion of endogenous tissue antioxi-
dants, is maximal during early reperfusion .
Although some oxidative damage may occur dur-
ing ischemia, even greater lipid peroxidation oc-
curs during reperfusion . Furthermore, the
depletion of cellular antioxidants is restricted to
early reperfusion, suggesting that these cellular
defenses against oxidative damage are not over-
whelmed before or after this interval . The
depletion of hippocampal antioxidants observed
during the first 10 min of reperfusion after asphyx-
ial cardiac arrest returns to normal after 2 h and
remains normal for 72 h.
This study did not determine whether reducing
inspired oxygen concentrations actually reduces
tissue oxidative stress. A future study could assess
the effect of inspired oxygen concentrations on
production of lipid peroxides or reduction in an-
tioxidant activity as indices of oxidative damage.
Accurate assay of tissue antioxidants requires
techniques beyond the scope of this study .
However, if reduced oxygen concentrations during
resuscitation do reduce oxidative stress, this inter-
vention may prove useful in conjunction with ther-
apies directed at other mechanisms of injury.
Finally, this study failed to find a difference in
outcome between two groups. Because prolonged
hemodynamic monitoring of rats was not per-
formed, the present data cannot exclude that nor-
hemodynamic events after the first hour of recov-
ery. Any delayed hypoperfusion might cause sec-
neurological improvement produced by normoxic
ventilation. Alternatively, the present sample size
might be insufficient to detect a real difference.
However, the confidence interval estimated for the
difference between the mean NDS for each group
at 72 h suggests a low likelihood of any large
behavioral difference. Likewise, the small group
differences between the mean numbers of viable
and ischemic neurons for each region suggests that
oxygen concentration made no large difference in
the histological damage to the hippocampus. The
hippocampus was examined in this study because
it has a laminar structure that allows for quantifi-
cation of injury and is particularly sensitive to
ischemia. However, the present data cannot ex-
clude the possibility that other brain regions were
differentially affected by manipulations of inspired
Improvement in neurological outcome after re-
suscitation from global ischemia becomes increas-
ingly important as the rates of resuscitation
continue to improve. The need for effective neuro-
protective treatment paradigms for cardiac arrest,
head injury, stroke, and subarachnoid hemorrhage
is well known to the physician practicing acute
care medicine. The present findings illustrate that
single interventions are unlikely to be successful
for treatment of neuronal injury because of the
C.A. Lipinski et al. / Resuscitation 42 (1999) 221–229 228
multifactorial nature of ischemic neuronal dam-
age. A multifaceted approach that addresses the
many known mechanisms of neuronal death will
Preliminary data from this study were previ-
ously presented at the Society for Neuroscience
27th Annual Scientific Meeting, October 1997,
New Orleans, LA and the American College of
Emergency Physicians Annual Meeting, October
1997, San Francisco, CA. This work was sup-
ported by a resident research grant (CAL) from
the Emergency Medicine Foundation and Bristol-
 Becker L. The epidemiology of sudden death. In: Paradis
NA, Halperin HR, Nowak RM, editors. Cardiac Arrest:
The Science and Practice of Resuscitation Medicine.
Baltimore, MD: Williams and Wilkins, 1996:28–46.
 Wolbarsht ML, Fridovich I. Hyperoxia during reperfu-
sion is a factor in reperfusion injury. Free Rad Biol Med
 Watson BD, Busto R, Goldberg WJ, Santiso M, Yoshida
S, Sinsberg MD. Lipid peroxidation in vivo induced by
reversible global ischemia in rat brain. J Neurochem
 Katz LM, Callaway CW, Kagan VE, Kochanek PM.
Decreased hippocampal antioxidant activity after resusci-
tation from asphyxial cardiac arrest in rats. Neuroreport
 Dietrich WD. Morphological manifestations of reperfu-
sion injury in brain. Ann N Y Acad Sci 1994;723:15–24.
 Shivakumar BR, Kolluri SV, Ravindranath V. Glu-
tathione and protein thiol homeostasis in brain during
reperfusion after cerebral ischemia. J Pharmacol Exp
 Lazzarino G, Vagnozzi R, Tavazzi B, Pastore FS, Di
Pierro D, Siragusa P, Belli A, Giuffre R, Giardina B.
MDA, oxypurines, and nucleosides relate to reperfusion
in short-term incomplete cerebral ischemia in the rat.
Free Rad Biol Med 1992;13:489–98.
 Liu Y, Rosenthal RE, Haywood Y, Miljkovic-Lolic M,
Vanderhoek JY, Fiskum G. Normoxic ventilation after
cardiac arrest reduces oxidation of brain lipids and im-
proves neurological outcome. Stroke 1998;29:1679–86.
 Danielisova V, Marsala M, Chavko M, Marsala J.
Postischemic hypoxia improves metabolic and functional
recovery of the spinal cord. Neurology 1990;40:1125–9.
 Burda J, Marsala M, Radonak J, Marsala J. Graded
postischemic reoxygenation ameliorates inhibition of
cerebral cortical protein synthesis in dogs. J Cereb Blood
Flow Metab 1991;11:1001–5.
 Zwemer C, Whitesall SE, D’Alecy LG. Cardiopul-
monary–cerebral resuscitation with 100% oxygen exacer-
bates neurological dysfunction following nine minutes of
normothermic cardiac arrest in dogs. Resuscitation
 Zwemer C, Whitesall SE, D’Alecy LG. Hypoxic car-
diopulmonary-cerebral resuscitation fails to improve
neurological outcome following cardiac arrest in dogs.
 Safar P. Prevention and therapy of postresuscitation
neurologic dysfunction and injury. In: Paradis NA,
Halperin HR, Nowak RM, editors. Cardiac Arrest: The
Science and Practice of Resuscitation Medicine. Balti-
more, MD: Williams and Wilkins, 1996, pp 859-887.
 Katz L, Ebmeyer U, Safar P, Radovsky A, Neumar R.
Outcome model of asphyxial cardiac arrest in rats. J
Cereb Blood Flow Metab 1995;15:1032–9.
 Neumar RW, Bircher NG, Sim KM, Xiao F, Zadach
KS, Radovsky A, Katz L, Ebmeyer E, Safar P.
Epinephrine and sodium bicarbonate during CPR fol-
lowing asphyxial cardiac arrest in rats. Resuscitation
 Paxinos G, Watson C. The Rat Brain in Stereotaxic
Coordinates. San Diego, CA: Academic Press, 1996.
 Kawai K, Nitecka L, Ruetzler CA, Nagashima G, Joo F,
Mies G, Nowak TS, Saito N, Lohr JM, Klatzo I. Global
cerebral ischemia associated with cardiac arrest in the
rat. I. Dynamics of early neuronal changes. J Cereb
Blood Flow Metab 1992;12:238–49.
 Radovsky A, Safar P, Sterz F, Leonov Y, Reich H,
Kuboyama K. Regional prevalence and distribution of
ischemic neurons in dog brains 96 hours after cardiac
arrest of 0 to 20 minutes. Stroke 1995;26:2127–34.
 Marsala M, Danielisova V, Chavko M, Hornakova A,
Marsala J. Improvement of energy state and basic mod-
ifications of neuropathological damage in rabbits as a
result of graded postischemic spinal cord reoxygenation.
Exp Neurol 1989;105:93–103.
 West JB. Respiratory Physiology—The Essentials. Balti-
more, MD: Williams and Wilkins, 1990.
 Burda J, Chavko M. Effect of ischaemia on protein
synthesis in neuron and glia-enriched fractions from the
rabbit spinal cord. Physiol Res 1991;40:39–47.
 Burda J, Chavko M. Mechanism of protein synthesis
inhibition in CNS during postischaemic reperfusion.
Physiol Res 1991;40:395–402.
 Ito U, Spatz M, Walker JT, Klatzo I. Experimental
cerebral ischemia in Mongolian gerbils. Acta Neu-
 Pulsinelli WA, Brierly JB, Plum F. Temporal profile of
neuronal damage in a modelof transient forebrain is-
chemia. Ann Neurol 1982;11:491–8.
 Colbourne F, Li H, Buchan AM. Continuing postis-
chemic neuronal death in CA1 Influence of ischemia
duration and cytoprotective doses of NBQX and SNX-
111 in rats. Stroke 1999;30:662–7.
 Dietrich WD, Busto R, Alonso O, Globus MY, Ginsberg
MD. Intraischemic but not postischemic brain hypother-
mia protects chronically following forebrain ischemia in
rats. J Cereb Blood Flow Metab 1993;13:541–9.
C.A. Lipinski et al. / Resuscitation 42 (1999) 221–229229 Download full-text
 Levy DE, Bates D, Caronna JJ, Cartlidge NEF, Knill-
Jones RP, Lapinski RH, Singer BH, Shaw DA, Plum F.
 Vaagenes P, Safar P, Moosy J, Rao G, Diven W, Ravi
C, Arfors K. Asphyxiation versus ventricular fibrillation
cardiac arrest in dogs: difference in cerebral resuscitation
effects—a preliminary study. Resuscitation 1997;35:41–
 Ulatowski JA, Kirsch JR, Traystman RJ. Hypoxic
reperfusion after ischemia in swine does not improve
acute brain recovery. Am J Physiol 1994;267:H1880–
 Xiao F, Safar P, Radovsky A. Mild protective and
resuscitative hypothermia for asphyxial cardiac arrest in
rats. Am J Emerg Med 1998;16:17–25.
 Yamaguchi S, Ogata H, Hamaguchi S, Kitajima T.
Superoxide radical generation and histopathological
changes in hippocampal CA1 after ischaemia/reperfusion
in gerbils. Can J Anaesth 1998;45:226–32.