British Journal of Anaesthesia 99 (1): 1017 (2007)

doi:10.1093/bja/aem140

Cerebral protection
S. Fukuda1 and D. S. Warner1 3*
1
Department of Anesthesiology, 2Department of Neurobiology and 3Department of Surgery, Duke University
Medical Center, Box 3094, Durham, NC 27710, USA
*Corresponding author: Department of Anesthesiology, Duke University Medical Center, Box 3094,
Durham, NC 27710, USA. E-mail: david.warner@duke.edu
Ischaemic/hypoxic insults to the brain during surgery and anaesthesia can result in long-term
disability or death. Advances in resuscitation science encourage progress in clinical manage-
ment of these problems. However, current practice remains largely founded on extrapolation
from animal studies and limited clinical investigation. A major step was made with demon-
stration that rapid induction of mild sustained hypothermia in comatose survivors of out-
of-hospital ventricular fibrillation cardiac arrest reduces death and neurological morbidity with
negligible adverse events. This provides the first irrefutable evidence that outcome can be
favourably altered in humans with widely applicable neuroprotection protocols. How far
hypothermic protection can be extended to global ischaemia of other aetiologies remains to
be determined. All available evidence suggests an adverse response to hyperthermia in ischae-
mic or post-ischaemic brain. Management of other physiological values can have dramatic
effects in experimental injury models and this is largely supported by available clinical data.
Hyperoxaemia may be beneficial in transient focal ischaemia but deleterious in global ischae-
mia. Hyperglycaemia causes exacerbation of most forms of cerebral ischaemia and this can be
abated by restoration of normoglycaemia. Studies indicate little, if any, role for hyperventilation.
There is little evidence in humans that pharmacological intervention is advantageous.
Anaesthetics consistently and meaningfully improve outcome from experimental cerebral
ischaemia, but only if present during the ischaemic insult. Emerging experimental data portend
clinical breakthroughs in neuroprotection. In the interim, organized large-scale clinical trials
could serve to better define limitations and efficacy of already available methods of interven-
tion, aimed primarily at regulation of physiological homeostasis.
Br J Anaesth 2007; 99: 1017
Keywords: brain, ischaemia; complications, cerebral ischaemia; recovery, neurological
Cerebral ischaemia/hypoxia can occur in a variety of peri- However, it has become increasingly clear that an
operative circumstances. Outcomes from such events range ischaemic/hypoxic insult does not simply constitute energy
from sub-clinical neurocognitive deficits to catastrophic failure with consequent interruption of ongoing metabolic
neurological morbidity or death. Although certain surgical events. Indeed this does occur. In addition, though, ischae-
procedures present greater risk for ischaemic/hypoxic mia and hypoxia stimulate active responses in the brain,
brain injury, most insults are not presaged but instead arise which persist long after substrate delivery has been
as unintended complications of either the surgical pro- restored. These responses include activation of transcrip-
cedure or the anaesthetic. tion factors which up-regulate expression of genes contri-
It has been the investigative interest of surgeons and buting to apoptosis and inflammation, inhibition of protein
anaesthesiologists to reduce perioperative brain injury synthesis, sustained oxidative stress, and neurogenesis.
for more than 60 yr.12 Classically, such intervention has Although some of these responses may have a teleological
been categorized as either neuroprotection or neuro- advantage [e.g. elimination of dead or dysfunctional
resuscitation. Neuroprotection was defined as treatment tissue or increased tolerance to a subsequent insult ( pre-
initiated before onset of ischaemia, intended to modify conditioning)], most responses aggravate damage caused
intra-ischaemic cellular and vascular biological responses by the primary insult. Consequently, the concept that neu-
to deprivation of energy supply so as to increase tolerance roprotection can be extended well into the reperfusion
of tissue to ischaemia resulting in improved outcome. phase seems appropriate, albeit with different targets other
Neuroresuscitation, in contrast, implied treatment begun than preservation of energy stores. This possibility may, in
after the ischaemic insult had occurred with the intent of part, explain the efficacy of various experimental post-
optimizing reperfusion. ischaemic interventions, which have manifested either as

# The Board of Management and Trustees of the British Journal of Anaesthesia 2007. All rights reserved. For Permissions, please e-mail: journals.permissions@oxfordjournals.org
Downloaded from https://academic.oup.com/bja/article-abstract/99/1/10/269751/Cerebral-protection
by guest
on 20 October 2017
 Cerebral protection

Table 1 Evidence-based status of plausible interventions to reduce perioperative ischaemic brain injury. , Repeated physiologically controlled studies in
animals/randomized, prospective, adequately powered clinical trials; , consistent suggestion by case series/retrospective or prospective small sample size trials,
or data extrapolated from other paradigms; 2/, inconsistent findings in clinical trials; may be dependent on characteristics of insult; 2, well-defined absence of
benefit; 22, absence of evidence in physiologically controlled studies in animals/randomized, prospective, adequately powered clinical trials; 222, evidence
of potential harm; *, out-of-hospital ventricular fibrillation cardiac arrest

Intervention Pre-ischaemic Post-ischaemic Pre-ischaemic Post-ischaemic Sustained Sustained

efficacy in efficacy in efficacy in efficacy in protection in protection
experimental experimental humans humans experimental in humans
animals animals animals

Moderate 2/ *
hypothermia
Mild 222 222 22 22 222 22
hyperthermia
Hyperventilation 22 22 22 22 22 22
Normoglycaemia 22 22
Hyperbaric 22 22 2/ 22 22
oxygen
Barbiturates 2 2 22 22
Propofol 2 22 22 2
Etomidate 222 22 22 22 22 22
Nitrous oxide 2 22 22 22 22 22
Isoflurane 22 22 22 22
Sevoflurane 22 22 22 22
Desflurane 22 22 22 22 22
Lidocaine 22 22 22 22
Ketamine 22 22 22 22 22
Glucocorticoids 222 22 22 22 22 22

clinically available therapies (e.g. mild hypothermia) or
instead as promising candidates for future clinical use tar-
geting events, such as oxidative stress, apoptosis, and
neurogenesis.
The above logic is presented as a taste of where we are
going with investigations aimed at ameliorating long-term
improvement from an ischaemic/hypoxic insult that may
occur in the perioperative period. However, the rest of this
article will focus on the opportunities and limitations of
currently available interventions (Table 1).
Anaesthetics
Barbiturates
It has been postulated for more than 50 yr that anaesthetics
increase the tolerance of brain to an ischaemic insult.28 The
logic is simple. Most drugs selected to be anaesthetics sup-
press neurotransmission. This suppression reduces energy
requirement, and reduction in energy requirement should
allow tissue better to preserve energy balance during a tran-
sient interruption of substrate delivery. Since adenosine tri-
phosphate (ATP) synthesis recovers rapidly after restoration
of substrate delivery, anaesthetics would be expected to be
protective if present during ischaemia but not if given after
restoration of substrate delivery. It would also follow that
efficacy of an anaesthetic is dependent upon the severity of
the ischaemic insult. If the insult were sufficiently severe to
cause loss of all electrical activity, there would be no
activity for anaesthetics to suppress and thus no mechanism
for such drugs to increase tolerance to ischaemia. In
contrast, in less severe insults, suppression of activity by
the anaesthetic before onset of ischaemia should delay
decay of ATP concentrations and thus also delay loss of
ionic gradients and calcium influx.
Many studies have supported this logic. Indeed, during
abrupt onset of hypoxaemia, barbiturates and isoflurane
slow deterioration of ATP concentrations.43 48 Furthermore,
post-ischaemic treatment with either barbiturates or volatile
anaesthetics has no effect on outcome.1 59 Surprisingly,
irrefutable data supporting efficacy of pre-treatment with
anaesthetics have proved difficult to acquire.
Early work testing intra-ischaemic anaesthetic efficacy
was confounded by poor physiological control of exper-
imental subjects. It was recognized later in the evolution of
anaesthetic efficacy studies that factors such as blood
glucose, brain temperature, and perfusion pressure were
important determinants of ischaemic outcome and that
anaesthetics independently modulated these factors. In
addition, early studies typically compared one anaesthetic
against another. The assumption was that the control
anaesthetic was not protective and thus failure to improve
outcome by the test anaesthetic indicated lack of a pro-
tective state. However, little work was done to confirm that
the control anaesthetic was not protective. Subsequent
studies, which became feasible as experimental models
evolved, often found considerable protection from the
control anaesthetic when compared with an awake state.
Thus, the field remained confused for more than a
decade and insufficient data were generated to warrant
human trials of anaesthetic efficacy when employed intra-
operatively. Even then, the early results were mixed. One

11
Downloaded from https://academic.oup.com/bja/article-abstract/99/1/10/269751/Cerebral-protection
by guest
on 20 October 2017
 Fukuda and Warner
study found efficacy from thiopental when given in
cardiac surgical patients, whereas another did not.50 67
However, only short-term outcomes were assessed, which
prevented assessment of the full evolution of the ischae-
mic injury. Furthermore, surgical procedures and cardio-
pulmonary bypass conditions were markedly different
between the two trials. Numerous other explanations have
been offered, but perhaps the overall potency of barbitu-
rates as neuroprotective agents is weak in the face of
severe ischaemic insults.65
One problem with barbiturates is their prolonged dur-
ation of action. It was believed that optimal protection
would be present only when massive doses were adminis-
tered to abolish electroencephalographic (EEG) activity,
thereby eliciting maximal suppression of cerebral meta-
bolic rate (CMR) before onset of the insult. Some
practitioners still adhere to this principle when using bar-
biturates to protect the brain but such large doses can
markedly delay anaesthesia emergence, which has limited
their clinical application. Although it is unlikely that these
massive doses are necessary to obtain maximal efficacy,65
recognition that volatile anaesthetics can also produce
EEG isoelectricity at doses which still allow rapid anaes-
thesia emergence was greeted with optimism because such
compounds could be more widely applied in clinical
settings.
Volatile anaesthetics
The efficacy of volatile anaesthetics as neuroprotective
agents has undergone more than 30 yr of scrutiny and still
no human outcome trials have been conducted to guide
clinical practice. We know the following facts from the
laboratory. Volatile anaesthetics provide major improve-
ment in ischaemic outcome. The dose required to obtain
this protection is within a clinically relevant range, with
higher doses potentially worsening outcome.46 Volatile
anaesthetics protect against both focal (e.g. obstruction of
flow distal to the circle of Willis) and global (e.g. com-
plete cessation of blood flow to the brain or forebrain)
ischaemia. However, the improvement in outcome is tran-
sient in global ischaemia,23 whereas it is persistent in
focal ischaemia.58 Sevoflurane has also been shown to
provide long-term protection in one experimental model.51
The mechanism by which volatile anaesthetics protect is,
in part, attributable to suppression of energy require-
ments.47 Both inhibition of excitatory neurotransmission
and potentiation of inhibitory receptors are likely to be
involved.15 22 30 It is also likely that volatile anaesthetics
have other important effects that include regulation of
intracellular calcium responses during ischaemia,29 and
activation of TREK-1 two-pore-domain K channels.25
Although a great deal has been learned from the labora-
tory, in the absence of human outcome data, it cannot be
stated that volatile anaesthetics improve outcome from
perioperative ischaemic insults. However, if an anaesthetic
12
Downloaded from https://academic.oup.com/bja/article-abstract/99/1/10/269751/Cerebral-protection
by guest
on 20 October 2017
is required for a surgical procedure, inclusion of volatile
anaesthetics can be considered. Isoflurane and sevoflurane
carry the largest data set to this decision. Desflurane also
offers promise,33 38 but has been insufficiently studied to
determine whether it should be equally considered in this
class of potential neuroprotective compounds.
Other anaesthetics
Other anaesthetics possess properties that suggest potential
for intra-ischaemic neuroprotection. These include propo-
fol, etomidate, and lidocaine. Study of these drugs has not
been as extensive as for either barbiturates or volatile
anaesthetics. The principle feature of propofol and etomi-
date is suppression of CMR by inhibition of synaptic
activity.19 35 Propofol may also have free radical scaven-
ging and anti-inflammatory properties.57 Propofol appears
unique among anaesthetics in the laboratory setting
because it offers efficacy with post-ischaemic therapy
onset, although such treatment provides only transient pro-
tection.9 Propofol appears to offer efficacy similar to bar-
biturates but a dose-dependent study of its efficacy has not
been completed, leaving little guidance for potential clini-
cal use. Furthermore, propofol infused to induce EEG
burst suppression failed to improve outcome in cardiac
valve surgery patients.56 Etomidate, although initially her-
alded as a substitute for barbiturates,8 has never met rigor-
ous evaluation for neuroprotective properties. In fact,
some work has indicated that etomidate may paradoxically
exacerbate ischaemic injury by inhibiting nitric oxide
synthase, thereby intensifying the ischaemic insult.21 As a
result of this and other studies, the use of etomidate for
neuroprotection has fallen out of favour in clinical
settings.
Lidocaine also suppresses CMR, but this effect is only
meaningful at doses beyond those typically employed in
clinical environments. Numerous laboratory studies have
found efficacy for lidocaine, with perhaps its principle
mechanism of action relating to inhibition of apoptosis.39
The efficacy of lidocaine appears dependent on dose, with
doses in the range used to manage cardiac dysrhythmias
having greatest efficacy.61 There have been no long-term
outcome studies of lidocaine efficacy in experimental
stroke. One small human trial found benefit from low-dose
lidocaine infusion during cardiac surgery on long-term
neuropsychological impairment.44 Lidocaine should be
further evaluated for neuroprotective properties since its
use is supported by a litany of laboratory successes such
as short-duration of action and ease of use. However,
because it has not been evaluated in a large-scale clinical
trial, efficacy in clinical environments remains speculative.
Ketamine offers potent inhibition of glutamatergic
neurotransmission at the N-methyl-D-aspartate (NMDA)
receptor. There is a long history of NMDA receptor antag-
onists as potential neuroprotective agents but, overall, such
compounds offer little or no protection against global
 Cerebral protection
insults. Protection against focal insults is substantial, but
only if the drug is given before ischaemia onset. Because
ketamine is clinically available, it is tempting to argue that
it should be considered when a focal ischaemic insult is
anticipated. To date, however, there are no human data
supporting this practice. Little is also known about dose
response properties, even in animals. Thus, it is difficult to
recommend ketamine for the purposes of neuroprotection
in the clinical environment at this time.
Physiological management
Temperature
Hypothermia has been proposed to offer therapeutic
benefit for more than 60 yr.24 Early investigators examined
its effects in both neurosurgery and cardiac surgery
patients. In the same era, it was also considered to offer
benefit in survivors of cardiac arrest and hypoxic insults.10
It remains unclear why hypothermia fell out of favour
in subsequent decades. One factor may have been its
apparent lack of efficacy, which reduced enthusiasm for
the logistical issues necessary routinely to cool and
re-warm a large patient population. Another factor may
have been the influence of mechanistic studies conducted
in the laboratory.42 That work examined effects of
hypothermia on brain energy metabolism and found
hypothermia to reduce CMR in a temperature-dependent
fashion, which became the presumed mechanism of
action. The most impressive effects on CMR were at very
low temperatures, and those temperatures required use of
cardiopulmonary bypass. The effects of mild (32 358C)
hypothermia on CMR were negligible. In contrast, barbitu-
rates can reduce CMR by 50 60% without the use of car-
diopulmonary bypass and were therefore viewed as having
a greater potential benefit. Perhaps for those reasons, the
use of perioperative hypothermia persisted only in the
context of caring for some cardiac surgical patients.
There is no doubt that deep hypothermia (e.g. 18
228C) is highly neuroprotective. We know that only a few
minutes of complete global ischaemia will cause neuronal
death in normothermic brain. This has been best examined
in the laboratory, but human evidence is consistent with
those findings.53 In contrast, it is widely observed that
induction of deep hypothermia before circulatory arrest
routinely allows the brain to tolerate intervals of no-flow
exceeding 40 min, and substantially greater intervals of
arrest with complete or near-complete neurological recov-
ery are frequently reported. As a result of this prima facie
evidence, the efficacy of deep hypothermia has not been
subjected to randomized controlled trials. However, there
is still much to be learned with respect to optimizing
cooling and re-warming methods, optimal magnitude of
hypothermia, determination of brain temperature using sur-
rogate sites, and defining within individual patients when
13
Downloaded from https://academic.oup.com/bja/article-abstract/99/1/10/269751/Cerebral-protection
by guest
on 20 October 2017
the duration of circulatory arrest approaches the limits of
deep hypothermic neuroprotection.
The story might have ended there had it not been for
several laboratory studies that ignored the CMR hypoth-
esis. Those studies re-visited the possibility that mild
hypothermia could protect the brain against ischaemia
insults.14 40 To most peoples surprise, reduction in brain
temperature by only a few degree Celsius provided major
protection. These findings stimulated numerous clinical
trials in both adults and newborns, which have since pro-
vided a scientific basis defining the opportunities and
limitations of using off-bypass hypothermia to provide
meaningful neuroprotection.
The first reported work related to traumatic brain injury
(TBI). Three pilot studies provided suggestive evidence
that mild hypothermia improved either brain physiology or
outcome. However, those studies employed small sample
sizes and more definitive evidence was needed. Thus, a
large-scale prospective human trial was conducted, but
disappointing results were obtained.18 Cooling TBI
patients within the first several hours after injury failed to
improve outcome. The design and conduct of this trial
have been vigorously debated but what is clear is that
induced hypothermia is not a panacea for TBI. If it is
proven effective in later trials, it will probably be shown
to have efficacy only in certain patient populations and
only when conducted with specific protocols. Such work
is ongoing.
If the TBI study had been performed in isolation,
perhaps off-bypass hypothermia would have been aban-
doned in the clinic again. However, other studies were
already underway, two of which markedly altered the
mood of the investigative community. Both studies were
reported simultaneously and used similar experimental
designs wherein comatose survivors of out-of-hospital
cardiac arrest were randomized to normothermia or mild
hypothermia, which involved rapid surface cooling as
soon as spontaneous circulation was restored.2 11 Both
studies found significantly more patients with good
outcome in the hypothermia group and negligible adverse
events. Finally, convincing evidence is available that off-
bypass hypothermia can appreciably improve outcome
from at least cardiac arrest in humans.
These findings have prompted publication of guidelines
recommending that comatose survivors of out-of-hospital
cardiac arrest undergo cooling after restoration of spon-
taneous circulation.3 49 The extent to which the efficacy of
induced hypothermia can be extrapolated to other con-
ditions of cardiac arrest (loss of airway, asphyxia, and
drowning) may never be known given the sporadic and
relatively rare nature of those events. However, such inter-
vention may be considered.41
In addition, there is an increasing evidence that peripar-
tum neonatal asphyxial brain injury favourably responds to
treatment with hypothermia. Two trials have been
reported. The first employed selective head cooling and
 Fukuda and Warner
could only find a beneficial effect of hypothermia in a
subset of the study population.27 The second employed
total body cooling.60 In this study, the benefit of induced
mild hypothermia was clear. Despite this, some feel
additional trials are required before such intervention can
be widely advocated.32
In the course of defining hypothermia efficacy, it has
also become apparent that hyperthermia has adverse effects
on post-ischaemic brain. Spontaneous post-ischaemic
hyperthermia is common4 and, in animals, intra-ischaemic
or even delayed post-ischaemic hyperthermia dramatically
worsens outcome. Spontaneous hyperthermia has also been
associated with poor outcome in humans.36 These facts
provide sufficient evidence to advocate frequent tempera-
ture monitoring in patients with cerebral injury (and those
at risk for cerebral injury). Aggressive treatment of
hyperthermia should be considered.
Glucose
Glucose is a fundamental substrate for brain energy metab-
olism. Deprivation of glucose in the presence of oxygen
can result in neuronal necrosis, but the presence of
glucose in the absence of oxygen carries a worse fate. The
mechanistic basis for this dichotomy remains unclear. The
most persistent hypothesis is that glucose, in the absence
of oxygen, undergoes anaerobic glycolysis resulting in
intracellular acidosis, which amplifies the severity of other
deleterious cascades initiated by the ischaemic insult.
Many animal studies have demonstrated adverse effects of
hyperglycaemia from a wide variety of brain insults.
Human studies remain principally correlative in nature,
that is, patients having worse outcomes from stroke, TBI,
etc. also tend to have higher blood glucose concentrations
on hospital admission. For some time, it was unclear
whether admission hyperglycaemia simply represented a
stress response to the brain insult, or instead was contribut-
ing to a worsened injury. The animal data clearly favour
the latter interpretation. More importantly, human research
has demonstrated more rapid expansion of ischaemic
lesions in hyperglycaemic, compared with normoglycaemic
patients.6 52 In addition, there is accumulating evidence that
regulation of blood glucose yields a higher incidence of
good outcome in stroke patients.26 For all of these reasons,
it is rational to maintain normoglycaemia in all patients at
risk for, or recovering from acute brain injury.
Arterial carbon dioxide partial pressure (PaCO2)
Because cerebral blood flow and PaCO2 are linearly related
within physiologically relevant ranges, hyperventilation
had become an entrenched practice in cerebral resuscita-
tion. Reduction in PaCO2 was presumed to augment cer-
ebral perfusion pressure favourably by reducing the
cross-sectional diameter of the arterial circulation and thus
cerebral blood volume. This would offset increases in
intracranial pressure. Although the logic behind this
14
Downloaded from https://academic.oup.com/bja/article-abstract/99/1/10/269751/Cerebral-protection
by guest
on 20 October 2017
practice can be appreciated, in fact, it is contradicted by
direct examination of cerebral well being. The most salient
evidence is derived from TBI investigations. These studies
support a different concept, that being worsening of per-
fusion by hyperventilation-induced vasoconstriction in
ischaemic tissue. Indeed, the volume of ischaemic tissue,
elegantly assessed with positron emission tomography in
TBI patients, was markedly increased when moderate hypo-
capnia was induced.20 This is consistent with the only pro-
spective trial of hyperventilation on TBI outcome, which
observed a decreased number of patients with good or mod-
erate disability outcomes when chronic hyperventilation
was employed.45 It remains unevaluated whether acute
hyperventilation improves outcome from pending transten-
torial herniation or when rapid surgical decompression of a
haematoma (e.g. epidural) is anticipated. Within the
context of focal ischaemic stroke, clinical trials have found
no benefit from induced hypocapnia,17 62 although hyper-
ventilation is sometimes employed in cases of refractory
brain oedema. Use of hyperventilation during cardiopul-
monary resuscitation may serve to increase mean intrathor-
acic pressure thereby decreasing perfusion pressure and is
not advocated.5 Consequently, there are few data to support
use of hyperventilation in the context of cerebral
resuscitation.
Arterial oxygen partial pressure
It makes sense that optimization of oxygen delivery to
ischaemic tissue should improve outcome. Indeed, oxygen
deprivation is the fundamental fault leading to tissue
demise. However, reperfusion presents deranged oxygen
metabolism with the opportunity to increase formation of
reactive oxygen species that plausibly induce secondary
insults, thereby worsening outcome. There are few human
data regarding the effects of normobaric hyperoxaemia in
human resuscitation. One retrospective perinatal resuscita-
tion analysis found worse long-term outcome in children
when either hyperoxaemia or hypocapnia was present
during resuscitation or early recovery.37 Others found more
rapid normalization of Apgar scores when 40% oxygen
compared with 100% oxygen was used for resuscitation.31
In animal models, it is becoming evident that the effect
of hyperoxaemia is dependent on the nature of the ischae-
mic insult. Rats subjected to middle cerebral artery occlu-
sion had smaller infarcts when normobaric hyperoxaemia
was present during both ischaemia and reperfusion. This is
consistent with the demonstrated efficacy of hyperbaric
oxygen (HBO) in rats undergoing a similar focal ischaemic
insult.63 Evidence for HBO efficacy in humans is weak.16
In contrast, in dogs subjected to cardiac arrest, it has been
repeatedly observed that outcome is worsened by normoba-
ric hyperoxaemia present during early recirculation.64 This
has been attributed to oxidation and decreased pyruvate
dehydrogenase activity, the enzymatic link between anaero-
bic and aerobic glycolysis.55 Management of oxygen
 Cerebral protection
delivery after restoration of spontaneous circulation, so as
to maintain pulse oximeter values within the range of
94 96, optimized short-term neurological outcome.7 These
compelling data should serve as a stimulus for a random-
ized clinical trial and stimulates re-consideration of the
necessity for hyperoxaemia in the early post-resuscitation
interval.
Steroids
Steroids such as dexamethasone reduce oedema surround-
ing brain tumours. Beyond that, evidence for benefit from
the use of steroids is weak. Evidence that methylpredniso-
lone improves outcome from acute spinal cord trauma is
controversial,13 but some surgeons have extended this
observation to intraoperative use in spinal cord surgery.
There is insufficient evidence to define the role of gluco-
corticoids in focal ischaemic stroke.54 A large retrospec-
tive analysis found no benefit from glucocorticoid
treatment in patients with cardiac arrest.34 In fact, there is
animal evidence that such glucocorticoids exacerbate
injury from global ischaemia by increasing plasma glucose
concentration.66 Given the potential adverse effects of
steroids and lack of demonstrable efficacy in ischaemic
brain, their use cannot be advocated.
Conclusion
Ischaemic brain injury remains a potentially devastating
disorder, although progress is being made in resuscitation
science. Two key advances occurred in the past decade.
The first was repeated demonstration that induced mild
hypothermia reduces neurological morbidity and mortality
associated with out-of-hospital ventricular fibrillation
cardiac arrest. Beyond the immediate potential to apply
this intervention is the larger message that post-ischaemic
intervention can favourably influence outcome in humans.
The second advance was recognition that efficacy of mild
hypothermia depends at least in part upon the type of
ischaemic lesion being treated. Trauma and focal ischae-
mia could not be shown to be amenable to hypothermic
Table 2 Considerations when anticipating or managing a perioperative
ischaemic insult
Assure absence of hyperthermia
Manage blood glucose with insulin to induce normoglycaemia
Optimize haemoglobin-oxygen saturation (increasing concern that
hyperoxaemia may be adverse in global ischaemia)
Establish normocapnia
Consider the use of volatile anaesthetics if surgery ongoing (consistent
sustained benefit in experimental animal studies, reversible allowing
neurological examination, human trials not performed)
Resist the use of glucorticoids (no evidence of efficacy, preclinical evidence
of adverse effect in global ischaemia)
Consider the use of postoperative sustained induced moderate hypothermia if
global ischaemia (not tested by clinical trials in perioperative environment,
but supported by consistent evidence of efficacy when used in out-of-hospital
ventricular fibrillation cardiac arrest)
15
Downloaded from https://academic.oup.com/bja/article-abstract/99/1/10/269751/Cerebral-protection
by guest
on 20 October 2017
intervention, at least within the bounds of the clinical trial
protocols employed.
Other than the use of mild hypothermia for ventricular
fibrillation cardiac arrest, practice of clinical neuroprotection
rests on extrapolation from animal studies and weak
clinical trials. Review of these data allows some recommen-
dations to be made (Table 2). Such recommendations are
likely to be advanced with increased understanding of
cellular responses to ischaemia and appropriately conducted
clinical trials.
References
1 Randomized clinical study of thiopental loading in comatose sur-
vivors of cardiac arrest. Brain Resuscitation Clinical Trial I Study
Group. N Engl J Med 1986; 314: 397 403
2 Hypothermia after Cardia Arrest Study Group. Mild therapeutic
hypothermia to improve the neurologic outcome after cardiac
arrest. N Engl J Med 2002; 346: 549 56
3 Part 7.5: Postresuscitation support. Circulation 2005; 112: IV84 8
4 Albrecht RF 2nd, Wass CT, Lanier WL. Occurrence of potentially
detrimental temperature alterations in hospitalized patients at
risk for brain injury. Mayo Clin Proc 1998; 73: 629 35
5 Aufderheide TP, Sigurdsson G, Pirrallo RG, et al.
Hyperventilation-induced hypotension during cardiopulmonary
resuscitation. Circulation 2004; 109: 1960 5
6 Baird TA, Parsons MW, Phanh T, et al. Persistent poststroke
hyperglycemia is independently associated with infarct expansion
and worse clinical outcome. Stroke 2003; 34: 2208 14
7 Balan IS, Fiskum G, Hazelton J, Cotto-Cumba C, Rosenthal RE.
Oximetry-guided reoxygenation improves neurological outcome
after experimental cardiac arrest. Stroke 2006; 37: 3008 13
8 Batjer HH, Frankfurt AI, Purdy PD, Smith SS, Samson DS. Use of
etomidate, temporary arterial occlusion, and intraoperative
angiography in surgical treatment of large and giant cerebral
aneurysms. J Neurosurg 1988; 68: 234 40
9 Bayona NA, Gelb AW, Jiang Z, Wilson JX, Urquhart BL,
Cechetto DF. Propofol neuroprotection in cerebral ischemia and
its effects on low-molecular-weight antioxidants and skilled
motor tasks. Anesthesiology 2004; 100: 1151 9
10 Benson DW, Williams GR Jr, Spencer FC, Yates AJ. The use of
hypothermia after cardiac arrest. Anesth Analg 1959; 38: 423 8
11 Bernard SA, Gray TW, Buist MD, et al. Treatment of comatose
survivors of out-of-hospital cardiac arrest with induced hypother-
mia. N Engl J Med 2002; 346: 557 63
12 Bigelow WG, Lindsay WK, Greenwood WF. Hypothermia; its
possible role in cardiac surgery: an investigation of factors gov-
erning survival in dogs at low body temperatures. Ann Surg 1950;
132: 849 66
13 Bracken MB, Shepard MJ, Holford TR, et al. Administration of
methylprednisolone for 24 or 48 hours or tirilazad mesylate for
48 hours in the treatment of acute spinal cord injury. Results of
the Third National Acute Spinal Cord Injury Randomized
Controlled Trial. National Acute Spinal Cord Injury Study. JAMA
1997; 277: 1597 604
14 Busto R, Dietrich WD, Globus M-T, Ginsberg MD. The import-
ance of brain temperature in cerebral ischemic injury. Stroke
1989; 20: 1113 14
15 Canas PT, Velly LJ, Labrande CN, et al. Sevoflurane protects rat
mixed cerebrocortical neuronal-glial cell cultures against transient
oxygen-glucose deprivation: involvement of glutamate uptake and
reactive oxygen species. Anesthesiology 2006; 105: 990 8
 Fukuda and Warner
16 Carson S, McDonagh M, Russman B, Helfand M. Hyperbaric
oxygen therapy for stroke: a systematic review of the evidence.
Clin Rehabil 2005; 19: 819 33
17 Christensen MS, Paulson OB, Olesen J, et al. Cerebral apoplexy
(stroke) treated with or without prolonged artificial
hyperventilation. 1. Cerebral circulation, clinical course, and
cause of death. Stroke 1973; 4: 568 631
18 Clifton GL, Miller ER, Choi SC, et al. Lack of effect of induction
of hypothermia after acute brain injury. N Engl J Med 2001; 344:
556 63
19 Cold GE, Eskesen V, Eriksen H, Blatt Lyon B. Changes in CMRO2,
EEG and concentration of etomidate in serum and brain tissue
during craniotomy with continuous etomidate supplemented with
N2O and fentanyl. Acta Anaesthesiol Scand 1986; 30: 15963
20 Coles JP, Fryer TD, Coleman MR, et al. Hyperventilation following
head injury: effect on ischemic burden and cerebral oxidative
metabolism. Crit Care Med 2007; 35: 568 78
21 Drummond JC, McKay LD, Cole DJ, Patel PM. The role of nitric
oxide synthase inhibition in the adverse effects of etomidate in
the setting of focal cerebral ischemia in rats. Anesth Analg 2005;
100: 841 6
22 Elsersy H, Mixco J, Sheng H, Pearlstein RD, Warner DS. Selective
gamma-aminobutyric acid type A receptor antagonism reverses iso-
flurane ischemic neuroprotection. Anesthesiology 2006; 105: 8190
23 Elsersy H, Sheng H, Lynch JR, Moldovan M, Pearlstein RD,
Warner DS. Effects of isoflurane versus fentanyl-nitrous oxide
anesthesia on long-term outcome from severe forebrain ischemia
in the rat. Anesthesiology 2004; 100: 1160 6
24 Fay T. Observations on generalized refrigeration in cases of severe
cerebral trauma. Assoc Res Nerv Ment Dis Proc 1943; 24: 61119
25 Franks NP, Honore E. The TREK K2P channels and their role in
general anaesthesia and neuroprotection. Trends Pharmacol Sci
2004; 25: 601 8
26 Gentile NT, Seftchick MW, Huynh T, Kruus LK, Gaughan J.
Decreased mortality by normalizing blood glucose after acute
ischemic stroke. Acad Emerg Med 2006; 13: 174 80
27 Gluckman PD, Wyatt JS, Azzopardi D, et al. Selective head
cooling with mild systemic hypothermia after neonatal encephalo-
pathy: multicentre randomised trial. Lancet 2005; 365: 663 70
28 Goldstein A, Jr, Wells BA, Keats AS. Increased tolerance to
cerebral anoxia by pentobarbital. Arch Int Pharmacodyn Ther 1966;
161: 138 43
29 Gray JJ, Bickler PE, Fahlman CS, Zhan X, Schuyler JA. Isoflurane
neuroprotection in hypoxic hippocampal slice cultures involves
increases in intracellular Ca2 and mitogen-activated protein
kinases. Anesthesiology 2005; 102: 606 15
30 Harada H, Kelly PJ, Cole DJ, Drummond JC, Patel PM. Isoflurane
reduces N-methyl-D-aspartate toxicity in vivo in the rat cerebral
cortex. Anesth Analg 1999; 89: 1442 7
31 Hellstrom-Westas L, Forsblad K, Sjors G, et al. Earlier Apgar
score increase in severely depressed term infants cared for in
Swedish level III units with 40% oxygen versus 100% oxygen
resuscitation strategies: a population-based register study.
Pediatrics 2006; 118: e1798 804
32 Higgins RD, Raju TN, Perlman J, et al. Hypothermia and perinatal
asphyxia: executive summary of the National Institute of Child
Health and Human Development workshop. J Pediatr 2006; 148:
170 5
33 Hoffman WE, Charbel FT, Edelman G, Ausman JI. Thiopental and
desflurane treatment for brain protection. Neurosurgery 1998; 43:
1050 3
34 Jastremski M, Sutton-Tyrrell K, Vaagenes P, Abramson N,
Heiselman D, Safar P. Glucocorticoid treatment does not
Downloaded from https://academic.oup.com/bja/article-abstract/99/1/10/269751/Cerebral-protection
by guest
on 20 October 2017
improve neurological recovery following cardiac arrest. Brain
Resuscitation Clinical Trial I Study Group. JAMA 1989; 262:
3427 30
35 Kaisti KK, Langsjo JW, Aalto S, et al. Effects of sevoflurane, pro-
pofol, and adjunct nitrous oxide on regional cerebral blood flow,
oxygen consumption, and blood volume in humans. Anesthesiology
2003; 99: 603 13
36 Kammersgaard LP, Jorgensen HS, Rungby JA, et al. Admission
body temperature predicts long-term mortality after acute
stroke: the Copenhagen Stroke Study. Stroke 2002; 33: 1759 62
37 Klinger G, Beyene J, Shah P, Perlman M. Do hyperoxaemia and
hypocapnia add to the risk of brain injury after intrapartum
asphyxia? Arch Dis Child Fetal Neonatal Ed 2005; 90: F49 52
38 Kurth CD, Priestley M, Watzman HM, McCann J, Golden J.
Desflurane confers neurologic protection for deep hypothermic
circulatory arrest in newborn pigs. Anesthesiology 2001; 95: 95964
39 Lei B, Popp S, Capuano-Waters C, Cottrell JE, Kass IS. Lidocaine
attenuates apoptosis in the ischemic penumbra and reduces
infarct size after transient focal cerebral ischemia in rats.
Neuroscience 2004; 125: 691 701
40 Leonov Y, Sterz F, Safar P, et al. Mild cerebral hypothermia during
and after cardiac arrest improves neurologic outcome in dogs.
J Cereb Blood Flow Metab 1990; 10: 57 70
41 McDonagh DL, Allen IN, Keifer JC, Warner DS. Induction of
hypothermia after intraoperative hypoxic brain insult. Anesth
Analg 2006; 103: 180 1
42 Michenfelder J, Terry HJ, Daw E, Uihlein A. Induced hypothermia:
physiologic effects, indications, and techniques. Surg Clin North
Am 1965; 45: 889
43 Michenfelder JD, Theye RA. Cerebral protection by thiopental
during hypoxia. Anesthesiology 1973; 39: 510 7
44 Mitchell SJ, Pellett O, Gorman DF. Cerebral protection by lido-
caine during cardiac operations. Ann Thorac Surg 1999; 67:
1117 24
45 Muizelaar JP, Marmarou A, Ward JD, et al. Adverse effects of pro-
longed hyperventilation in patients with severe head injury: a ran-
domized clinical trial. J Neurosurg 1991; 75: 731 9
46 Nasu I, Yokoo N, Takaoka S, et al. The dose-dependent effects of
isoflurane on outcome from severe forebrain ischemia in the rat.
Anesth Analg 2006; 103: 413 8
47 Nellgard B, Mackensen GB, Pineda J, Wellons JC 3rd, Pearlstein RD,
Warner DS. Anesthetic effects on cerebral metabolic rate predict
histologic outcome from near-complete forebrain ischemia in the
rat. Anesthesiology 2000; 93: 4316
48 Newberg LA, Michenfelder JD. Cerebral protection by isoflurane
during hypoxemia or ischemia. Anesthesiology 1983; 59: 29 35
49 Nolan JP, Morley PT, Vanden Hoek TL, et al. Therapeutic
hypothermia after cardiac arrest: an advisory statement by the
advanced life support task force of the International Liaison
Committee on Resuscitation. Circulation 2003; 108: 118 21
50 Nussmeier NA, Arlund C, Slogoff S. Neuropsychiatric compli-
cations after cardiopulmonary bypass: cerebral protection by a
barbiturate. Anesthesiology 1986; 64: 165 70
51 Pape M, Engelhard K, Eberspacher E, et al. The long-term effect
of sevoflurane on neuronal cell damage and expression of apop-
totic factors after cerebral ischemia and reperfusion in rats.
Anesth Analg 2006; 103: 173 9
52 Parsons MW, Barber PA, Desmond PM, et al. Acute hyperglyce-
mia adversely affects stroke outcome: a magnetic resonance
imaging and spectroscopy study. Ann Neurol 2002; 52: 20 8
53 Petito CK, Feldmann E, Pulsinelli WA, Plum F. Delayed hippocam-
pal damage in humans following cardiorespiratory arrest.
Neurology 1987; 37: 1281 6
16
 Cerebral protection
54 Qizilbash N, Lewington SL, Lopez-Arrieta JM. Corticosteroids
for acute ischaemic stroke. Cochrane Database Syst Rev 2002:
CD000064
55 Richards EM, Rosenthal RE, Kristian T, Fiskum G. Postischemic
hyperoxia reduces hippocampal pyruvate dehydrogenase activity.
Free Radic Biol Med 2006; 40: 1960 70
56 Roach GW, Newman MF, Murkin JM, et al. Ineffectiveness of
burst suppression therapy in mitigating perioperative cerebrovas-
cular dysfunction. Multicenter Study of Perioperative Ischemia
(McSPI) Research Group. Anesthesiology 1999; 90: 1255 64
57 Rodriguez-Lopez JM, Sanchez-Conde P, Lozano FS, et al.
Laboratory investigation: effects of propofol on the systemic
inflammatory response during aortic surgery. Can J Anaesth 2006;
53: 701 10
58 Sakai H, Sheng H, Yates RB, Ishida K, Pearlstein RD, Warner DS.
Isoflurane provides long-term protection against focal cerebral
ischemia in the rat. Anesthesiology 2007; 106: 92 99
59 Sarraf-Yazdi S, Sheng H, Brinkhous AD, Pearlstein RD, Warner DS.
Effects of postischemic halothane administration on outcome from
transient focal cerebral ischemia in the rat. J Neurosurg Anesthesiol
1999; 11: 316
60 Shankaran S, Laptook AR, Ehrenkranz RA, et al. Whole-body
hypothermia for neonates with hypoxic-ischemic encephalopathy.
N Engl J Med 2005; 353: 1574 84
17
Downloaded from https://academic.oup.com/bja/article-abstract/99/1/10/269751/Cerebral-protection
by guest
on 20 October 2017
61 Shokunbi MT, Gelb AW, Wu XM, Miller DJ. Continuous lidocaine
infusion and focal feline cerebral ischemia. Stroke 1990; 21:
107 11
62 Simard D, Paulson OB. Artifical hyperventilation in stroke. Trans
Am Neurol Assoc 1973; 98: 309 10
63 Veltkamp R, Sun L, Herrmann O, et al. Oxygen therapy in perma-
nent brain ischemia: potential and limitations. Brain Res 2006;
1107: 185 91
64 Vereczki V, Martin E, Rosenthal RE, Hof PR, Hoffman GE, Fiskum G.
Normoxic resuscitation after cardiac arrest protects against hippo-
campal oxidative stress, metabolic dysfunction, and neuronal death.
J Cereb Blood Flow Metab 2006; 26: 82135
65 Warner DS, Takaoka S, Wu B, et al. Electroencephalographic
burst suppression is not required to elicit maximal neuroprotec-
tion from pentobarbital in a rat model of focal cerebral ischemia.
Anesthesiology 1996; 84: 1475 84
66 Wass CT, Scheithauer BW, Bronk JT, Wilson RM, Lanier WL.
Insulin treatment of corticosteroid-associated hyperglycemia and
its effect on outcome after forebrain ischemia in rats.
Anesthesiology 1996; 84: 644 51
67 Zaidan JR, Klochany A, Martin WM, Ziegler JS, Harless DM,
Andrews RB. Effect of thiopental on neurologic outcome follow-
ing coronary artery bypass grafting. Anesthesiology 1991; 74:
406 11