KONTROVERSIAL DALAM NEUROANAESTHESIA

When Dr. John W. Michenfelder coined the term Neuroanaesthesia in his pioneering review published
in 19691, he would have least believed where the specialty has headed for in the past 40 years. From a
service that provided anaesthesia to patients undergoing neurosurgery, it evolved to a subspecialty that
combines the rapidly advancing basic and clinical neuroscience knowledge with the knowledge of
anaesthesia to improve the outcomes of neurological patients. The speciality also has been able to
contribute to the understanding of some of the basic phenomena in neurosciences. Seminal changes have
occurred in the concepts in many clinical and research arenas as discussed below(what a new
neuroanesthesia)

As 'Neuroanaesthesia' continues to expand and develop, the basic principles remain

unchanged - provision of optimal operative conditions, maintenance of cerebral
perfusion pressure (CPP), and cerebral oxygenation. However, despite advances in
drugs and monitoring modalities, many controversies remain regarding the clinical
practice of neuroanaesthesia the choice of anaesthetic agent and fluids, and
intraoperative hypothermia. [1],[2],[3],[4],[5] To address some of these issues such as the
use of nitrous oxide, intraoperative hypotension and anaesthetic technique for
carotid endarterectomy [6],[7],[8],[9],[10],[11],[12] .1

Intraoperative concerns during craniotomy generally revolve around intracranial pressure (ICP)
reduction and protection against inadvertent cerebral ischemia. Maintenance of optimal systemic
physiology still remains the mainstay to achieve both the goals. Mannitol and hyperventilation
have been the most commonly used tools for intraoperative ICP reduction. Hypertonic saline has
emerged as a more physiological alternative to mannitol. But, the limited data available has not
convincingly proved that it offers any advantage over mannitol for operative conditions of the
brain during craniotomy2. With regards to hyperventilation, a recent study in patients undergoing
surgery for supratentorial tumors has shown that moderate hyperventilation to a PaCO2 of 32-35
mmHg effects a significant change in subdural ICP and the surgeon's assessment of brain
relaxation3. Whether such hyperventilation raises the same concerns of cerebral ischemia as in
traumatic brain injury (TBI), remains to be evaluated. (what a new neuroanesthesia)

Anaesthetics have been relied upon for intraoperative cerebral protection. Controversy continues
regarding the superiority of one agent over others4. Inhalational agents have been claimed to
offer more predictable protection than intravenous agents. Given the Experimental evidence that
cerebral vascular occlusion caused under halothane anaesthesia produced smaller infarct volumes
than under awake state raises the possibility that it is the anaesthetic state rather than the
individual agent that offers protection. Apart from their effects on cerebral metabolism,
anaesthetics seem to offer protection through other mechanisms such as ischemic
preconditioning. (what a new neuroanesthesia)
Isoflurane has been a commonly used agent for neuroanesthesia, but newer agents,
sevoflurane and desflurane, have a quicker onset and shorter emergence from
anesthesia and are increasingly preferred for general pediatric anesthesia. But their
effects on intracranial pressure (ICP) and cerebral perfusion pressure (CPP),
especially in pediatric patients with already increased ICP, have not been well
documented(Effects of 0.5 and 1.0 MAC isoflurane)

Isoflurane 0.5 and 1.0 MAC, sevoflurane and desflurane in N2O all increased ICP and
reduced MAP and CPP in a dose-dependent and clinically similar manner. There were
no baseline dependent increases in ICP from 0 to 1.0 MAC with isoflurane or
sevoflurane, but ICP increased somewhat more, although statistically insignificant,
with higher baseline values in patients given desflurane. The effect of MAP on CPP is
3-4 times higher than the effect of the increases in ICP on CPP and this makes MAP
the most important factor in preserving CPP. In children with known increased ICP,
intravenous anesthesia may be safer. However, maintaining MAP remains the most
important determinant of a safe CPP. (Effects of 0.5 and 1.0 MAC isoflurane)

Sevoflurane is less vasoactive than halothane, enflurane, isoflurane, or desflurane. The context
sensitive half-life is short and similar to that of desflurane, which translates into fast on and
offset. Compared with propofol, sevoflurane decreases cerebral blood flow to a lesser extent,
while cerebral metabolism is suppressed to the same degree. Sevoflurane does not increase
intracranial pressure, while propofol decreases intracranial pressure.

In neurosurgical patients with normal intracranial pressure, sevoflurane might be a good

alternative to propofol. In patients with reduced intracranial elastance, caused by space
occupying lesions, with elevated intracranial pressure or complex surgical approaches, propofol
should remain first choice.

(Inhalational or intravenous anesthetics for craniotomies?

Pro inhalational)
The results of this study indicate that there are no variations on ICP in normocapnic patients
undergoing removal of supratentorial brain tumors without midline shift, as they were
anesthetized with isoflurane or desflurane. CPP and cerebral AVDo(2) decreased with both
agents. (The effects of isoflurane and desflurane on intracranial pressure)

Evolution of neurosurgery mainly trends towards minimally invasive and functional procedures
including endoscopies, small-size craniotomies, intraoperative imaging and stereotactic
interventions. Consequently, new adjustments of anaesthesia should aim at providing brain
relaxation, minimal interference with electrophysiological monitoring, rapid recovery, patients'
cooperation during surgery and neuroprotection.
RECENT FINDINGS:

In brain tumour patients undergoing craniotomy, propofol anaesthesia is associated with lower
intracranial pressure and cerebral swelling than volatile anaesthesia. Hyperventilation used to
improve brain relaxation may decrease jugular venous oxygen saturation below the critical
threshold. It decreases the cerebral perfusion pressure in patients receiving sevoflurane, but not
in those receiving propofol. The advantage of propofol over volatile agents has also been
confirmed regarding interference with somatosensory, auditory and motor evoked potentials.
Excellent and predictable recovery conditions as well as minimal postoperative side-effects make
propofol particularly suitable in awake craniotomies. Finally, the potential neuroprotective effect
of this drug could be mediated by its antioxidant properties which can play a role in apoptosis,
ischaemia-reperfusion injury and inflammatory-induced neuronal damage.

SUMMARY:

Although all the objectives of neurosurgical anaesthesia cannot be met by one single
anaesthetic agent or technique, propofol-based intravenous anaesthesia appears as the
first choice to challenge the evolution of neurosurgery in the third millennium. (Why we still
use intravenous drugs as the basic regimen for neurosurgical anaesthesia)

The aim of this study was to compare the cerebral protective effects of two known
protective anesthetics, isoflurane and propofol, when these were used in
combination with moderate hypothermia (33-34 degrees C) after diffuse traumatic
brain injury (TBI) in the rat. We assessed cerebral protection by measuring local
cerebral blood flow (LCBF), mean arterial blood pressure (MABP), cerebral perfusion
pressure (CPP) and intracranial pressure (ICP).

In the isoflurane group, MABP and CPP decreased significantly from baseline to 180 minutes (p
< 0.05 and p < 0.01, respectively), and MABP was significantly lower than the pressure in the
propofol group from 45 minutes through 180 minutes (p < 0.05, p < 0.01). ICP and LCBF
remained unchanged in this group. In the propofol group, from baseline to 180 minutes, CPP
increased to maximum 120 +/- 8 mmHg at 75 minutes from 98 +/- 5 mmHg (p < 0.05) and ICP
fell from 18 +/- 2 mmHg to 7 +/- 1 mmHg (p < 0.01); and the latter was significantly lower than
ICP in the isoflurane group (p < 0.05, p < 0.01, p < 0.001). LCBF in this group was significantly
higher than LCBF in the isoflurane group in the last 30 minutes of the experiment (p < 0.05). The
propofol group showed no change in MABP over the course of the experiment.

CONCLUSION:
In the clinical setting, propofol anesthesia may be better for use in combination with
hypothermia in cases of traumatic brain injury, as it reduces ICP and increases CPP under these
conditions. (Propofol versus isoflurane anesthesia) Different anesthetic agents have different
effects on cerebrovascular physiology. However, the importance of these differences in
neuroanesthetic practice are unclear. In an effort to determine whether important clinical
differences are present, the authors compared three anesthetic techniques in 121 adults
undergoing elective surgical removal of a supratentorial, intracranial mass lesion.

Although there are modest differences among the three tested anesthetics, short-term outcome
was not affected. These results indicate that, despite their respective cerebrovascular effects, all
of the anesthetic regimens used were acceptable in these patients undergoing elective surgery.

(A prospective, comparative trial of three anesthetics for

elective supratentorial craniotomy. Propofol/fentanyl,
isoflurane/nitrous oxide, and fentanyl/nitrous oxide)
In neurosurgery, anesthesiologists and surgeons focus on the same target - the brain. The nature
of anesthetics is to interact with brain physiology, leading to favorable and adverse effects.
Research in neuroanesthesia over the last three decades has been dedicated to identifying the
optimal anesthetic agent to maintain coupling between cerebral blood flow and metabolism, keep
cerebrovascular autoregulation intact, and not increase cerebral blood volume and intracranial
pressure.

(Inhalational or intravenous anesthetics for craniotomies?

Pro inhalational)
Desflurane elicits a dose-dependent increase in cerebral blood flow velocities and heart rate, but
does not change mean arterial pressure, suggesting that its cerebrovascular action is independent
of its systemic vascular action. CO2 reactivity is maintained at one MAC. The results in children
are similar to those seen in adults. (The effect of desflurane on cerebral blood flow)

Desflurane is a volatile anesthetic agent with low solubility whose use in neurosurgery has been
debated because of its effect on intracranial pressure and cerebral blood flow. The purpose of this
study was to determine the variations on intracranial pressure (ICP) and cerebral perfusion
pressure (CPP) as well as on cerebral arteriovenous oxygen content difference (AVDo(2)) in
normocapnic patients scheduled to undergo removal of supratentorial brain tumors with no
evidence of mass effect during anesthesia with isoflurane or desflurane.

There were no significant differences between groups in heart rate, mean arterial pressure, ICP,
and CPP. ICP measurements throughout the study did not change within each group compared to
baseline values. Mean arterial pressure decreased significantly in all patients compared to
baseline values, changing from 105 +/- 14 mmHg (mean +/- SD) to 85 +/- 10 mmHg in the
isoflurane group and from 107 +/- 11 mmHg to 86 +/- 10 mmHg in the desflurane group (P <
0.05 in both groups). CPP also decreased within each group compared with baseline values,
changing from 95 +/- 15 mmHg to 74 +/- 11 mmHg in the isoflurane group and from 95 +/- 16
mmHg to 74 +/- 10 mmHg in the desflurane group (P < 0.05 in both groups). Cerebral AVDo(2)
decreased significantly in both groups throughout the study, changing from 2.35 +/- 0.77 mm to
1.82 +/- 0.61 mm (mmol/l) in the isoflurane group (P < 0.05) and from 2.23 +/- 0.72 mm to 1.94
+/- 0.76 mm in the desflurane group (P < 0.05), without differences between groups.

The results of this study indicate that there are no variations on ICP in normocapnic patients
undergoing removal of supratentorial brain tumors without midline shift, as they were
anesthetized with isoflurane or desflurane. CPP and cerebral AVDo(2) decreased with both
agents. (The effects of isoflurane and desflurane on intracranial pressure)

Summary This review considers the current position of nitrous oxide in anaesthetic
practice and balances potential beneficial and disadvantageous effects. The classic
adverse characteristics of nitrous oxide, such as diffusion hypoxia, expansion of gas-
filled spaces, and postoperative nausea and vomiting, are often cited as reasons to
avoid this old drug. Recent concerns regarding neurotoxicity, adverse cardiovascular
outcomes, and wound complications have further hardened many practitioners
against nitrous oxide. New evidence and underpinning mechanistic data, however,
suggest potential beneficial effects on the central nervous system, cardiovascular
system, and acute and chronic pain. While we await the outcome of large studies
including ENIGMA-II, many clinicians have already decided against this agent. The
authors argue that this abandonment may be premature.

Traditionally, nitrous oxide (N 2 O) has been an integral component of general

anesthesia. However, in recent years, evidence on complications associated with N 2
O has raised arguments against the use of N 2 O in anesthesia, particularly for
patients with cerebral pathology. In this issue of Neurology India, Singh et al. [1]
challenge the need to withhold N 2 O from anesthesia for brain tumour surgery by
demonstrating that N 2 O did not significantly influence the duration of intensive
care stay or mortality. The study raises a debate on the propriety of using N 2 O
during neurosurgery.

By contributing to the depth of anesthesia, [2] N 2 O decreases the need for

the primary anesthetic agent, and thus helps in rapid awakening of the
patient, which facilitates immediate post-operative neurological examination.
Arguments for avoiding N 2 O in anesthesia, in general, are based on its
adverse effects: (1) Potential to interfere with vitamin B 12 and folate
metabolism, leading to megaloblastic anemia; [3] (2) expansion of the volume of
gas-filled cavities (eg, pneumothorax, distended bowel); [4] and (3) post-
operative nausea and vomiting (PONV). [5] The ENIGMA trial, [6] comprising 2,050
patients, showed that avoidance of N 2 O during anesthesia decreased the
incidence of major post-operative complications such as severe PONV, wound
infection, pneumonia, and atelectasis. The hospital stay was not significantly
affected by N 2 O; however, among the patients admitted to the intensive care
unit (ICU), the probability of discharge from ICU, on any given day, was
higher in the N 2 O-free group.

There are additional concerns also for using N 2 O in neurosurgery. Studies

have documented increase in cerebral blood flow (CBF), [7] cerebral blood
volume (CBV), and cerebral metabolism. [8] Background anesthetic and the degree
of ventilation seem to influence the magnitude of these changes. Intravenous
anesthetic agents like propofol [9] and hyperventilation [10] limit the increase
in CBF and CBV. N 2 O-induced PONV may confound post-operative vomiting caused
by raised intracranial pressure (ICP) due to acute intracranial complications
(eg, post-operative hematoma). Venous air embolism during neurosurgery, with
the patient in a sitting position and post-operative tension pneumocephalus
due to subdural air collection during craniotomy may be accentuated by N 2 O.
[11]
N 2 O can decrease the amplitude of cerebral electrophysiological signals,
[12],[13]
dissuading its use when crucial electrophysiological monitoring is
required. It has been implicated in the reversal of neuronal protection
offered by thiopentone in experimental cerebral injury. [14] N 2 O interferes
with homocysteine metabolism by oxidising the cobalt atom in vitamin B 12 ,
which, in turn, leads to defects of neuronal myelination. [15],[16] Neurological
manifestations such as numbness, decreased deep tendon reflexes, cognitive
dysfunction with histological evidence of neuronal mitochondrial swelling,
vacuolation, and apoptosis have been reported. [17] N 2 O has also been
implicated as a cause of post-operative dementia and cognitive deficits. [18],
[19]
By increasing serum homocysteine concentration, N 2 O may increase platelet
aggregation, cause vasoconstriction, and impair endothelial function. [20]

Do all the aforementioned effects of N 2 O influence the outcome of

neurosurgery? Two recent publications address this issue, both of which are
posthoc analyses of the Intra-operative Hypothermia for Aneurysm Surgery trial
(IHAST) data. [21] The first study [22] analysed the effect of N 2 O on
neurologic and neuropsychological function after intracranial aneurysm
surgery. Of the 1,000 patients studied, 373 received N 2 O. N 2 O did not
influence the development of delayed ischaemic neurological deficit (DIND).
Outcomes at three months, as assessed by Glasgow Outcome Scale (GOS), National
Institutes of Health (NIH) stroke scale, Rankin Disability Score, Barthel
Index and neuropsychological testing, were similar between the groups. The
authors speculate on the cause for the lack of difference between the groups:
First, the study is posthoc in nature. Second, many of the patients may not
have suffered ischemia that could have been influenced by N 2 O. In a
subsequent study, [23] where only patients subjected to temporary vascular
occlusion intraoperatively were included, N 2 O increased the risk of DIND
[odds ratio (OR) = 1.78; 95% confidence interval (CI) = 1.08-2.95, P = 0.025).
However, at three months, there was no difference in the neurological outcome
scores. The data suggest that N 2 O may influence the short-term, if not the
long-term neurological morbidity.

Although Singh et al [1] . showed that there was no effect of N 2 O on post-operative

ICU stay or mortality, the data obtained are not reassuring with regard to the use of
N 2 O during craniotomy. The major limitation is the grossly inadequate sample size.
The study data do not have any measure of raised ICP or cerebral ischemia in the
study population. The subjects of the study being elective brain tumour patients,
who would have been substantially optimised with pre-operative ICP control
measures, the possibility of finding adverse influence of any therapeutic
intervention is low. An adequately powered outcome trial, where raised ICP or
cerebral ischemia is objectively quantified, is required to address the issue. The
issue of using ICU stay and survival rates as the primary outcome measures of
clinical interventions of low, but not negligible, risk also needs to be further debated
(Neurosurgery and nitrous oxide) In recent years, isolated concerns regarding the safety
profile of nitrous oxide have grown into a chorus of criticism. Increasingly, modern anaesthetists
view nitrous oxide as an anachronism; a relic from the bad old days of anaesthesia. It is
therefore reasonable to ask whether there is a role for nitrous oxide in modern anaesthetic
practice. The answer to this question requires a two-pronged approach: first, does nitrous oxide
have a unique selling proposition that warrants its specific use and, secondly, does its side-effect
profile justify continued use?

Nitrous oxide has several advantages. Its physicochemical properties, especially its relatively
low solubility in blood, allow for rapid, reliable changes in depth of anaesthesia/analgesia and
rapid recovery. Its molecular mechanism of action, as predominantly an N-methyl-d-aspartate
(NMDA) receptor antagonist, differs from the majority of our conventional anaesthetic agents
which are predominantly gamma-aminobutyric acid (GABA) agonists. This review highlights its
analgesic effects, potential to reduce awareness, role in neuroprotection, and haemodynamic
effects.

Pragmatically, nitrous oxide is an agent with which we are familiar, easy to use, and easy to
monitor: all advantages in real-world anaesthesia. Nitrous oxide has been used for more than 150
yr without leaving an obvious trail of death and destruction in its wake. It is thus clearly safe for
most patients.1 Outstanding questions are whether subtle adverse effects have been missed over
the years, and if there are specific vulnerable populations? We discuss potential effects on acute
and chronic pain, neurological and cardiovascular outcomes, and wound infection as these
remain controversial and are the focus of current research. Certain characteristics of nitrous
oxide, including expansion of gas-filled cavities, the second-gas effect, diffusion hypoxia, and its
propensity to postoperative nausea and vomiting, are well known and are therefore not the focus
of this review.

The acute analgesic effect of nitrous oksida is used as a component of balanced anaesthesia. The
magnitude of this effect is however unclear. Given the pharmacokinetic profile of nitrous oksida,
a relevant comparison is with remifentanil where 6670% nitrous oksida is equivalent to
remifentanil 0.0850.17 g kg1 min1, or a whole-blood concentration of 2 ng ml1.2,3 The
analgesic effect of nitrous oksida may be smaller when co-administered with GABAergic agents;
however, these studies used either animal models or sub-anaesthetic concentrations of
sevoflurane and nitrous oksida.46

Nitrous oksida may reduce postoperative pain when compared with remifentanil and attenuate
remifentanil-induced hyperalgesia.79 Nitrous oksida may also have utility in the prevention and
treatment of chronic pain syndromes. In a follow-up study of participants in the ENIGMA trial,
nitrous oksida use was associated with a significant reduction in chronic postsurgical pain which
was maintained after multivariate analysis.10,11 Although the methodology was not robust
(telephonic survey), this requires further investigation. The findings are biologically plausible
though, because even a single exposure to nitrous oksida results in a prolonged reduction in pain
hypersensitivity in an animal model of peripheral neuropathy.12

The amnestic and analgesic effects of nitrous oksida have a similar doseresponse profile, and
there are sound pharmacokinetic and pharmacodynamic reasons for it to decrease anaesthetic
awareness.13 Hopkins13 suggests that the number need to treat (NNT) to prevent awareness with
nitrous oksida compares favourably with monitoring with bispectral index (BIS). Tramer and
colleagues14 reported an NNT of 46 with nitrous oksida, whereas Myles and colleagues15 reported
an NNT of 138 with BIS monitoring in high-risk patients.In contrast, ENIGMA reported two
cases with awareness, both in the nitrous oksida group.10 Currently then, the effectiveness of
nitrous oksida as a tool to prevent anaesthetic awareness remains controversial although it is an
attractive proposition.16

What is the effect of nitrous oksida on commonly used depth of anaesthesia monitors? NMDA
receptor antagonists, such as ketamine, xenon, and nitrous oksida, suppress the cortical
electroencephalogram less than GABAergic agents, so BIS and spectral entropy are relatively
insensitive to nitrous oksida.17 Although the magnitude of this effect is controversial, using these
monitors to titrate a nitrous oksida-based anaesthetic may result in an inappropriately deep
anaesthetic, potentially leading to morbidity or mortality.18 Failure to take this into account or to
prevent or control for differences in depth of anaesthesia may explain some of the adverse
outcomes seen in recent studies, such as the ENIGMA trial.10

Previous SectionNext Section

Potential adverse neurological effects include myelinopathies, neurotoxicity/hypoxic-ischaemic

injury, neurodevelopment disturbances, postoperative cognitive dysfunction, and alterations in
intracranial dynamics.

Myelinopathies, such as sub-acute combined degeneration of the cord (SACD) feature

prominently on most anaesthetic trainees list of nitrous oksida-related complications. While there
is a sound biochemical basis for nitrous oksida to induce myelinopathy, this complication is
limited to case reports and usually involves prolonged exposure, either occupationally or as a
result of nitrous oksida abuse, that exceeds clinical anaesthetic exposure.1922 However, patients
with untreated vitamin B12 or folate deficiency may be at some risk from medical exposure, as
are patients with genetic disorders such as methylene tetrahydrofolate reductase deficiency.2325
As SACD is a potentially devastating complication if undiagnosed and untreated, patients with
risk factors, such as untreated B12 deficiency, should receive appropriate treatment with B-
vitamins, or nitrous oksida should be avoided.

Does nitrous oksida cause direct cerebral neurotoxicity or potentiate hypoxic-ischaemic injury?
Animal studies are contradictory. Rat studies have shown worsening of ischaemic injury and
direct neurotoxic changes. The former, however, was only seen with total ischaemia and not
partial ischaemia, the latter only with hyperbaric exposure, was short-lived, and was prevented
by co-administration with a GABAergic agent such as a volatile anaesthetic, as occurs in clinical
anaesthesia.26,27 These effects are thus inconsistent and do not reflect real-world scenarios. In
addition, nitrous oksida may in fact have a neuroprotective effect via the reduction of NMDA-
induced glutamate excitoxicity, and in support of this animal studies have reported a smaller
cortical infarct volume in ischaemic stroke with the use of nitrous oksida.28

So, while the animal data muddy the water, are there any human data to guide us? Unfortunately,
there is little good-quality evidence that focuses primarily on this issue. The Intraoperative
Hypothermia in Aneurysm Surgery Trial (IHAST) randomized 1001 patients undergoing cerebral
aneurysm clipping to either of mild hypothermia or normothermia.29 A post hoc analysis by
McGregor and colleagues30 found no difference in early or late neurological deficits between
those who had received nitrous oksida and those who had not. More patients in the nitrous oksida
group were however able to be discharged home. In an additional post hoc analysis, Pasternak
and colleagues31 evaluated only the subgroup of patients who had temporary aneurysm clipping,
a neurologically high-risk group. While the nitrous oksida group in this analysis had an increased
risk of delayed ischaemic neurological deficit, an early adverse neurological outcome, there
was a lower risk of impairment on neuropsychological testing at 3 months, and a greater chance
of being discharged home. While there are many methodological concerns and confounding
factors when it comes to using post hoc analyses of a trial of hypothermia to answer questions
regarding anaesthetic management, this is the best clinical evidence that we have in this regard.
IHAST reflects real-world anaesthetic practice and is of a magnitude and quality that is unlikely
to be repeated specifically to examine the role of nitrous oksida in this context. Finally, by
conducting separate analyses on both the whole cohort and those who had temporary aneurysm
clipping the use of nitrous oksida in patients at both standard and high risk of cerebral
ischaemia was evaluated. The best clinical evidence, therefore, suggests that nitrous oksida is
safe to use in patients at risk of cerebral ischaemic injury.

Concerns have been raised about possible adverse neurodevelopmental effects of nitrous oksida.
NMDA receptor antagonists have been associated with widespread neuronal apoptosis in rat
pups.32 However, with respect to nitrous oksida, this has been demonstrated when nitrous oksida
was used in combination with isoflurane and may have reflected more an exacerbation of
isoflurane-induced neurodegenerative changes than primarily a nitrous oksida effect.33 A more
recent study however showed that this combination has no direct neurotoxic effect.34 The effects
seen in animal studies may therefore be related to other components of the anaesthetic
management of these animals. Pragmatically, the realities of animal studies preclude the detailed
metabolic, haemodynamic, and respiratory management given to human patients. In addition, the
timing and duration of anaesthetic exposure in animal models may not correlate with, and
probably far exceeds, clinical anaesthetic exposure.35,36 Finally, despite the controversy generated
by animal studies, and recent concerns regarding an association between anaesthesia and
learning disabilities, no human clinical trials have linked nitrous oksida exposure specifically to
adverse neurodevelopmental outcomes.37

Nitrous oksida has been linked to postoperative cognitive dysfunction in elderly rats.38,39 Current
human data suggest that the aetiology is multifactorial, including varied risk factors such as the
neuroinflammatory response to surgery, environmental factors, and sleep disturbances, with little
evidence to suggest a role for nitrous oksida.4042

Contemporary neuroanaesthesia teaching often suggests that nitrous oksida adversely affects
intracranial dynamics, with increased cerebral metabolic rate (CMR), cerebral blood flow (CBF),
cerebral blood volume (CBV) and intracranial pressure (ICP), and also impaired autoregulation;
with recommendations that its use be avoided in the neurologically at risk patient. An evidence-
based approach shows that the reality is more complex.43

While some studies show an increase in CBF, others report no significant effect.4451 CBF effects
may depend on which hypnotic is co-administered with nitrous oksida, for example, no change in
CBF is seen with desflurane.49 The findings with propofol are inconsistent, but a number of
studies report no effect.48,50,51 The effect with sevoflurane seems to be concentration dependent.52

Effects on autoregulation are also inconsistent. The addition of nitrous oksida to propofol
anaesthesia does not seem to impair autoregulation.51 The effect with sevoflurane is
concentration dependent, with nitrous oksida impairing autoregulation when added to 1
minimum alveolar concentration (MAC) sevoflurane, but not when added to 1.5 MAC
sevoflurane.52 These findings suggest that the maintenance hypnotic exerts more effect on
cerebral haemodynamics than nitrous oksida itself.

Early animal data suggested nitrous oksida increased cerebral metabolic activity.53,54 Based on
these data, it made sense to avoid nitrous oksida in the neurologically at risk patient. This
metabolic effect is however complex. While nitrous oksida does not in fact increase global
cerebral metabolic rate in humans, it may alter the regional distribution of metabolic activity.46
The global effect on human cerebral metabolic rate may be dependent on depth of anaesthesia
and the agent co-administered with nitrous oksida.47,55 Nitrous oksida increased CMR during
propofol-induced electrical silence, but it did not increase CMR when added to clinically
relevant concentrations of isoflurane.47,55 In non-surgical volunteers, CMR was however higher
with an equi-MAC combination of nitrous oksida and isoflurane than with isoflurane alone.47
However, the cerebral metabolic effect of nitrous oksida in patients exposed to surgical stimuli
may be reduced if it is used to achieve an appropriate level of hypnosis and analgesia, an
optimal anaesthetic fit hypothesis.56,57 It is perhaps time to re-evaluate the role of nitrous oksida
in this context, as is being done with ketamine, another NMDA antagonist.58

Another factor to consider is carbon dioksida reactivity. While some studies report a reduction in
carbon dioksida reactivity when nitrous oksida is co-administered with a volatile anaesthetic
agent, this appears most significant at high concentrations of the volatile agent.59,60 During
clinical use, carbon dioksida reactivity is largely maintained and any potential adverse effects on
cerebral haemodynamics could be countered by an appropriate use of mild hypocapnia.6165

The common theme in the examples above is that the effects of nitrous on intracranial dynamics
are highly dependent on the anaesthetic milieu. Adverse effects are seen when nitrous oksida is
added to other agents to achieve deep anaesthesia, >1.5 MAC, or when subjects are too lightly
anaesthetized. Both these extremes are inappropriate and minimal effects are seen when nitrous
oksida is used to achieve an appropriate depth of anaesthesia.

With these contradictory underpinnings are there any clinical data that integrate these results?
Good surgical conditions have been reported with the use of nitrous oksida-based regimes in
brain tumour surgery despite many of these patients having a significant mass effect before
operation.67,68 More recently Singh and colleagues69 reported on a comparison between a nitrous
oksidaisoflurane regime and an isoflurane-only regime for supratentorial tumour surgery.
Although this was only a small pilot study, they found no differences in surgical conditions and
intraoperative or postoperative complications between the groups, and the nitrous oksida group
had more stable haemodynamics and lower analgesic and neuromuscular blocking agent
requirements.

From a practical point of view, the effect of nitrous oksida on intracranial dynamics appears
more benign than is often claimed. Holistically, the relative haemodynamic stability of
anaesthetic techniques including nitrous oksida may be advantageous. In the neurologically at
risk patient, avoiding secondary insults, such as hypotension, may be more important to the
patient than any specific choice of hypnotic.

Finally, two additional matters of relevance from a neurosurgical point of view need to be
addressed: these both relate to the ability of nitrous oksida to expand gas-filled spaces.

In head-injured or neurosurgical patients, nitrous oksida may theoretically convert a

pneumocephalus into a tension pneumocephalus. There appears, however, to be no difference in
the volume of intracranial gas post-craniotomy in patients who have received nitrous oksida vs
those who have had a nitrous-free anaesthetic.70 In fact it has been reported that patients who
received nitrous oksida during dural closure had lower ICPs than those who did not receive
nitrous oksida.71 It appears that the rapid washout of nitrous oksida may actually decrease the
pneumocephalus and that the risk of tension pneumocephalus with nitrous oksida is overstated.

Also of concern here is the risk that nitrous oksida may expand venous air emboli (VAE). The
timeframe involved is in all likelihood too short to have any clinically significant effect though.
In support of this, Losasso and colleagues72 found no evidence that nitrous oksida increased the
risk, volume, or clinical consequences of VAE. Although it is not rational then to omit nitrous
oksida-based solely on the fear of a VAE, it is still prudent to stop its administration if an air
embolus is suspected.

Previous SectionNext Section

Cardiovascular effects
The cardiovascular effects of nitrous oksida have generated much discussion since the
publication of the original ENIGMA trial and a subsequent post hoc secondary analysis of long-
term morbidity and mortality.10,73 ENIGMA demonstrated a trend to a lower risk of myocardial
infarction in the nitrous oksida-free group [adjusted OR=0.58 (95% CI: 0.221.50; P=0.26)]. The
long-term follow-up, over a median of 3.5 yr, further demonstrated a statistically significant
increase in the risk of myocardial infarction in patients exposed to nitrous oksida [adjusted OR
1.59 (95% CI: 1.012.51; P=0.04)]; however, 17% of patients were lost to follow-up.
Nevertheless, these data have raised serious concerns and have generated a clear hypothesis that
is currently being tested in the ENIGMA-II study.74

Should we then avoid the use of nitrous oksida in those with cardiovascular risk factors?

Given the statistical equipoise in ENIGMA, the next question is whether a sound biological basis
exists for increased cardiovascular risk caused by nitrous oksida? Nitrous oksida inactivates
vitamin B12, inhibiting methionine synthase, preventing the conversion of homocysteine to
methionine, and resulting in elevation of plasma homocysteine levels. Hyperhomocysteinaemia
in turn creates a milieu for acute coronary syndrome via endothelial dysfunction and
prothrombotic effects. However, the cardiovascular risk posed by chronically elevated
homocysteine levels (that is the basis for our acute concerns) is being questioned. Recent studies
have questioned whether this association truly exists and even if it does, if it is causal, or merely
an association, as recent studies using folate/B-vitamins to reduce homocysteine levels failed to
reduce cardiovascular adverse events.7577

There are, however, a number of concerns with this attractive paradigm. First, there is a
significant variation in the reported magnitude and time-scale of the postoperative increase in
plasma homocysteine. Badner and colleagues described a 22% increase in the post-anaesthesia
care unit and a 48% increase at 48 h after operation in one trial, and a 74% increase after
operation in an earlier trial; Myles and colleagues reported a 45% increase within 24 h; and
Nagele and colleagues reported a 228% increase within the first few hours after operation, which
returned to baseline by 24 h.7881 In addition, Nagele and colleagues showed that the magnitude
of the increase varied from 14 to 567% amongst different individuals.81 The rapid increase raises
questions as to whether this is really the direct effect of a reduction in methionine synthase
activity or whether there is another mechanism at play. The swift return to preoperative levels
leads one to question whether this could really explain the effects on long-term cardiovascular
events. Elevations in plasma homocysteine from preoperative levels are also seen after operation
in patients not exposed to nitrous oksida.82 The inter-individual variability raises questions as to
what other risk factors are involved. Duration of nitrous exposure seems significant, vitamin B12
and folate deficiency may play a role, and genetic polymorphisms may be relevant: at present,
we have insufficient data to properly explain this variability.81,8385

Perhaps the most powerful risk factor for postoperative hyperhomocysteinaemia is ASA III or IV
status.85 Hyperhomocysteinaemia increased the risk of major postoperative complications,
independent of nitrous oksida use. Does this imply that homocysteine levels are simply a marker
of cardiovascular risk and that any effect of nitrous oksida may serve to unmask an underlying
risk? Certainly there is no easily discernible linear cause and effect between nitrous oksida use
and cardiac risk.
While Myles and colleagues reported that nitrous oksida exposure was associated with both an
increase in homocysteine and a reduction in flow-mediated vasodilation (a marker of endothelial
dysfunction) and while others have reported a direct association between homocysteine exposure
and endothelial dysfunction, the clinical significance is uncertain.80,86 Badner and colleagues, for
example, reported that nitrous oksida increased the risk of postoperative ischaemia in patients
undergoing carotid endarterectomy.78 There was an increase in the number of patients with
ischaemia, the number of episodes of ischaemia, and the number of episodes of >30 min
duration. However, the number of patients with total ischaemia of >2 h was not significantly
increased. Again, it is difficult to interpret the clinical impact of these findings as clinical
outcomes were not reported and it is currently unclear what duration of ischaemia is associated
with adverse clinical outcomes. In contrast, Kozmary and colleagues87 had previously found a
trend towards a reduction in intra- and postoperative myocardial ischaemia/infarction in patients
undergoing carotid surgery who received nitrous oksida. In terms of recent clinical data, a
subgroup analysis of the General Anaesthetic vs Local Anaesthetic for carotid surgery (GALA)
trial showed no association between nitrous oksida and an increased risk of the composite
primary endpoint of stroke, death, or myocardial infarction within 30 days of carotid
endarterectomy.88 Furthermore, a post hoc subanalysis of the Perioperative Ischemic Evaluation
(POISE) study trial showed no increase in adverse cardiovascular events in the nitrous oksida
group and a recent large retrospective cohort analysis showed no difference in cardiac
complications between those who received nitrous oksida and those who did not.1,89 The presence
and clinical significance of acute effects of nitrous oksida on endothelial function and adverse
cardiovascular events thus remain unclear.

Might the excess cardiovascular risk implied by ENIGMA then be because of confounding
factors? In ENIGMA, the median volatile anaesthetic concentration was 0.87 MAC in the N2O-
free group and 0.67 MAC in the N2O group, while in Badner and colleagues' study the
comparison was 0.48% fractional end-tidal (Fet) isoflurane in the N2O group vs 0.67% Fet
isoflurane in the nitrous oksida-free group.10,78 The ENIGMA trial used 70% nitrous oksida,
whereas Badner and colleagues' study used >50% nitrous oksida. Thus, the nitrous oksida groups
in both studies have a greater depth of anaesthesia, as quantified by total MAC fraction, than the
nitrous oksida-free groups. Might the greater depth of anaesthesia experienced by patients given
nitrous oksida account for the adverse cardiovascular events reported in both sets of patients?90

While the association between nitrous oksida, homocysteine, and acute coronary syndromes
remains a matter of debate, can we hedge our bets and play it safe but still use nitrous oksida?
Vitamin B12 and folate supplementation has been investigated with conflicting results. Badner
and colleagues showed that oral supplementation for a week before operation prevented nitrous
oksida-associated increases in homocysteine levels, but Rao and colleagues82 failed to show an
effect with a single i.v. dose in the pre-anaesthetic holding area.91 Thus, further work needs to be
done to define the optimal dose, duration, and timing of this therapy and to evaluate its clinical
efficacy. It may also prove impractical clinically.

While most of the recent debate on the cardiovascular effects of nitrous oksida has focused on
the issues raised above, are we ignoring low-tech factors that have real-world benefits? The
haemodynamic stability of nitrous oksida is an example. While nitrous oksida may have a direct
myocardial-depressant effect, via the reduction in calcium release from the sarcoplasmic
reticulum, this is generally counteracted by indirect sympathetic stimulation; the net effect being
minimal cardiovascular depression.92 In support of this, Fernandes and colleagues reported stable
arterial pressure despite a greater functional depth of anaesthesia when nitrous oksida was added
to sevoflurane in patients undergoing laparoscopic cholecystectomy; Inada and colleagues
showed a trend towards a reduced heart rate and increased MAP when 0.65 MAC nitrous oksida
was substituted for equi-MAC concentrations of isoflurane or sevoflurane; and Shiga and
colleagues showed that 70% nitrous oksida caused little cardiovascular depression when added to
clinically applicable target concentrations of propofol.9395 The clinical implication is that the
anaesthetist at the coal-face can use nitrous oksida to facilitate the balancing act of achieving an
adequate depth of anaesthesia while maintaining haemodynamic stability without excessive use
of inotropes/vasopressors. This may be particularly useful in the elderly, those on cardiovascular-
depressant drugs (e.g. calcium channel blockers), and those with cardiovascular disease. While
there is no evidence for an outcome benefit in this regard, we know that sustained intraoperative
hypotension increases the risk of perioperative adverse cardiovascular events.96 Unfortunately,
the patients who would most benefit from the enhanced haemodynamic stability are a similar
group that may be at increased risk of the homocysteine-related adverse cardiac effects of
nitrous oksida. How we tease out the competing influences and effects remains a major challenge
for future research. ENIGMA-II may provide some answers.74

Finally, some lesser-known facets of the nitrous oksidacardiovascular interaction warrant

further exploration. Nitrous oksida, when administered as an adjunct to isoflurane anaesthesia,
attenuated the vascular hyporeactivity seen after haemorrhagic shock.97 This implies that our
anaesthetic choice may affect postoperative cardiovascular function and, as vasomotor
dysfunction is central to the development of organ dysfunction and death, potentially influence
perioperative outcome.

Previous SectionNext Section

Wound infection
While there are multiple potential mechanisms by which nitrous oksida may impair immune
function and wound healing the clinical effect on wound infection is uncertain. ENIGMA
highlighted a significant increase in wound infection in patients exposed to nitrous oksida.10
However, previous research, with wound infection as the primary outcome, showed no increase
in the risk of wound infection in the nitrous oksida group.98 A large retrospective cohort analysis
also showed no difference in wound disruption, and infectious complication in general, between
the nitrous oksida and nitrous oksida-free groups.1 Recent data on the subject suggest that nitrous
oksida increases deoxyribonucleic acid damage which may predispose patients to a higher risk of
wound infection.99 These conflicting data highlight the enigma of nitrous oksida: after 150 yr of
use we are left with more questions than answers, even regarding as simple a complication as
wound infection. It also highlights the conundrums that underpin the entire article: are there
specific benefits to the use of nitrous oksida, and if so, can we identify specific groups in which
these benefits outweigh the risks or identify strategies to mitigate these risks?

This review has focused on the most current and controversial issues around the use of nitrous
oksida, issues that would be either deal breakers or unique selling propositions for many
anaesthetists. There are of course a multitude of other considerations that may influence the
individual practitioner or institution. These are beyond the scope of the review, but include
clinical considerations such as the influence of nitrous oksida on postoperative nausea and
vomiting; the impact of which requires careful consideration of available evidence including the
compensatory effects of multi-modal anti-emetic prophylaxis.14,100 There are also complex
economic considerations, with nitrous oksida not necessarily reducing healthcare costs, as is
often claimed by its proponents.101 Finally, we must consider the environmental impact of our
anaesthetic choices. While the contribution of nitrous oksida used for anaesthesia may be low, a
recent review on the environment impact of anaesthetic gases is thought provoking.102,103

Previous SectionNext Section

Conclusion
Nitrous oksida should remain an option in contemporary anaesthesia. There are potential
advantages in pain control and prevention, reduction of awareness with recall, and use in
neurologically and cardiovascularly at risk patients. With respect to its side-effect profile,
recent data suggest that nitrous oksida is safe (and possibly beneficial) in an unselected
heterogenous patient population.1 In addition, certain conventional concerns have been addressed
(e.g. post-craniotomy pneumocephalus) or appear less of an issue than previously thought (e.g.
intracranial dynamics). New concerns regarding matters such as neurotoxicity and adverse
cardiovascular events have however emerged. Thus, while anaesthetists can rest assured that they
are, in general, not doing their patients an injustice with the use of nitrous oksida, it remains
incumbent upon the practitioner to utilize the data presented above, and any new data (e.g.
ENIGMA-II), to evaluate the riskbenefit profile for the individual patient and make the use of
nitrous oksida as safe as possible. (Nitrous oksida are we still in equipoise)

Abstract
Nitrous oksida is a widely used analgesic agent, used also in combination with anaesthetics
during surgery. Recent research has raised concerns about possible neurotoxicity of nitrous
oksida, particularly in the developing brain. Nitrous oksida is an N-methyl-d-aspartate (NMDA)-
antagonist drug, similar in nature to ketamine, another anaesthetic agent. It has been linked to
post-operative cardiovascular problems in clinical studies. It is also widely known that exposure
to nitrous oksida during surgery results in elevated homocysteine levels in many patients, but
very little work has investigated the long term effect of these increased homocysteine levels.
Now research in rodent models has found that homocysteine can be linked to neuronal death and
possibly even cognitive deficits. This review aims to examine the current knowledge of
mechanisms of action of nitrous oksida, and to describe some pathways by which it may have
neurotoxic effects.

Keywords: nitrous oksida, neurotoxicity, homocysteine, NMDA antagonist

Go to:

1. Introduction
Nitrous oksida (N2O) has been used alone or in combination with other agents to produce
analgesia and anaesthesia for over 150 years [1]. It shows anaesthetic properties at very high
concentrations, with a minimum alveolar concentration needed to produce anaesthesia in 50% of
subjects (MAC) of 104% [2]. To reach near this level without compromising oxygenation,
hyperbaric conditions are necessary which are impractical in a surgical setting. During the 1940s
however, doctors began administering N2O in combination with a number of other non-volatile
anaesthetic agents to allow for lower N2O concentrations to be used [1]. Nowadays, nitrous
oksida is still used in combination with various anaesthetic agents such as isoflurane and
ketamine for anaesthetic sparing, to allow lower concentrations of volatile or non-gaseous
anaesthetics to be used [3,4].

Nitrous oksida is also commonly administered in a 50:50 mixture with oxygen to give analgesia
during labour as it has no effect on awareness, and can be self-administered by the mother
allowing for more personalised pain relief during contractions. In recent years, however, the
safety and efficacy of nitrous oksida has been questioned [5,6]. While many studies show
adverse effects of nitrous oksida anaesthesia, there is still no general consensus as to whether
N2O is dangerous enough to warrant discontinuation as an anaesthetic or analgesic [7]. This
review article aims to summarise the current evidence for toxicity of nitrous oksida. However,
with the limited clinical data presently available on nitrous oksida toxicity it is, as of yet, too
soon to draw conclusions.

Go to:

2. Neurotoxicity
It has been documented through a series of clinical studies that nitrous oksida administration is
associated with post-operative cardiac problems [6,8]. Further evidence is now mounting which
implies nitrous oksida may also cause neurotoxicity. Often, neurological damage may not have
overt symptoms and vulnerable patients, such as the elderly, may experience cognitive changes
which may go unnoticed. It is often the most vulnerable patients exposed to anaesthetic agents,
so one must be sure that the stress of surgery is not exacerbated by the anaesthetic agent which
should be providing relief.

Many of the neurotoxic effects of nitrous oksida are dependent on exposure at a certain age or
developmental stage. Extensive research indicates that the main periods of vulnerability to
nitrous oksida neurotoxicity are during the perinatal period and again in the aged brain. The
foundation work has mostly been carried out in rat models but more recent research work has
extended into non-human primate models. In rats, perinatal development extends to postnatal day
(PND) 7, juvenile rats are classed PND 2228, adolescence begins around PND 3035 while
adulthood is reached at PND 60 [9]. This demonstrates the rapid development of rats in
comparisons to primates and humans.

2.1. Perinatal Brain

Rat neurodevelopment is confined to a short period directly after birth, from postnatal day (PND)
0 to PND 7. This roughly translates to a period spanning the third trimester of pregnancy to
approximately the 6th month of age in humans [10]. During this period there is a massive
increase in programmed cell death as excess neurons are cleared and synapses of remaining
neurons are strengthened, known as synaptogenesis. As can be seen in Table 1, rats exposed to
N2O in combination with other clinical anaesthetics during this period have a consistent,
excessive increase in apoptosis in various brain regions, most notably the retrosplenial cortex
(RSC) and thalamus [11]. It was also found that these animals have long term impairment of
cognitive function [12,13]. It has been shown that at PND 7 in rats there is peak N-methyl-d-
aspartate (NMDA) receptor expression in the developing brain, which may explain the increased
sensitivity to N2O [14,15]. This period is approximately equivalent to 2022 weeks in humans.

Table 1

An outline of major in vivo studies, regarding the neurotoxicity and mechanisms of

nitrous oksida in combination with other anaesthetics, spanning the past 25 years.
The studies cover a wide range of ages, anaesthetic concentrations, duration of
anaesthesia ...

Using non-human primates, these models can be taken a step closer to clinical relevance,
achieving that which would not be ethical or feasible in a human study. No current studies have
looked at N2O alone, but one study assessed neurotoxicity after a N2O/isoflurane mixed
anaesthesia protocol in PND 5 or 6 rhesus monkeys [23]. Similar to rodent models, there was
widespread cell death in the young monkeys, but interestingly they found a different pattern of
distribution in comparison to rodent models. While rodents usually had cell death in the posterior
cingulate and retrosplenial cortex (PC-RSC) and thalamus [11,14], this study found widespread
apoptosis in the temporal gyrus, hippocampus and frontal cortex [23]. The primate study also
found evidence of both necrotic and apoptotic cell death occurring, as opposed to simply
apoptotic in the rat. Other studies involving administration of ketamine, another NMDA
antagonist anaesthetic, to PND 36 rhesus monkeys demonstrated similar patterns of cell death
and cognitive dysfunction as rodent models [24,25,26]. These studies also found that by PND 35,
the neurotoxic effects of ketamine were no longer present.

Loepke et al. [27] undertook a review of all general anaesthetics administered to children in the
perinatal period and found a wide range of variability in neurotoxic potential of anaesthetic
agents. Nitrous oksida itself had not been subjected to any clinical trials but it was reported that
in utero or perinatal exposure to N2O was correlated with short term neurological problems such
as resistance to smiles and increased muscle tone [28]. This indicates that further research into
the effects of N2O on infants should be undertaken, considering how popular N2O is as an
induction agent and anaesthetic.

2.2. Aged Brain

A study by Noguchi et al. [29] compared a range of different aged rats, from PND 2060, and
discovered that at PND 30, MK801 started producing NMDA antagonist toxicity, with PND 60
rats having the highest level of cell death. This indicates that once past the early vulnerable
stages after birth, juvenile rats are not as susceptible to the neurotoxic effects of NMDA
antagonists, while adolescents are less vulnerable than adults. A number of rodent studies found
that N2O administration alone or in combination with other anaesthetic agents produced
cognitive deficits in aged mice (1820 months old) [20,21,22]. As shown in Table 1, all studies
from this lab used the radial arm maze (RAM) to test cognitive function, which is known to
involve hippocampal and cortical memory circuits. One interesting study found that older rats
were more susceptible to N2O in combination with ketamine than younger (6 month old) rats
[19]. They hypothesised that this was due to reduced hepatic function in older rats, which
resulted in slower clearance of ketamine from the body. This highlights the fact that N2O may not
always be reliably compared to ketamine or other non-inhalational anaesthetics, due to their
different metabolisation processes.

There is also a trend, as seen in Table 2, that in adult rats, high concentrations of N2O given
under hyperbaric conditions can result in neurotoxicity [31], however due to the fact that these
concentrations are unfeasible in a clinical setting, this may not be altogether relevant except to
help understand the toxic potential of N2O. Together, these results infer that there is a period,
beginning in the weeks just after birth, extending until adulthood, where rats appear to have a
less severe reaction to NMDA antagonist toxicity, except at clinically irrelevant concentrations.
This may be due to changes in the brain during this period, where there is a high level of
development but less programmed cell death. In line with these findings, Yon et al. [16] showed
that between PND7 and PND14, rats become desensitized to the damage induced by an
isoflurane/N2O/midazolam cocktail, and showed a significant increase in expression of Bcl-XL,
an anti-apoptotic protein. Clearly, this warrants further study.

Table 2

Summary of current papers available which study the effects of N 2O alone, including
case studies implicating N2O. Abbreviations: N2Onitrous oksida; NMDAN-methyl-
d-aspartate; PC-RSCposterior cingulate-retrosplenial cortex; MSmethionine ...

Go to:
3. Molecular Mechanisms of Action
Despite widespread use for many years, the mechanisms by which nitrous oksida achieves its
anaesthetic and analgesic properties have still not fully been elucidated. It has been suggested
that opioid receptors are responsible for the analgesic properties of nitrous oksida. Research
revealed that administration of Nalaxone, an opioid reverse agonist, inhibits the analgesic effects
of nitrous oksida [37]. It is well known that there are a range of opioid receptors so it is difficult
to pinpoint one specific receptor as being responsible. Work done on the abdominal muscles of
mice found that the endogenous ligand for the -opioid receptor, dynorphin, may be the mediator
of N2O antinociception [38,39]. However, both the - and -opioid receptors were found to have
involvement in the rat hot plate test, which involves more peripheral nerves [40]. A study in the
guinea-pig brain served to compare binding by nitrous oksida to opioid receptors in the brain and
discovered N2O acts differently on - and -opioid receptors. -receptors were competitively
inhibited by N2O while -receptors were non-competitively bound [41]. Another mechanism of
analgesia appears to be via indirect T-type calcium channel inhibition by N2O [42]. This shows
there is a high level of variance in how nitrous oksida modulates receptor activity to produce its
analgesic effects.

For a drug to have an anaesthetic effect it must decrease excitatory output or increase inhibitory
signals to result in a net loss of neuronal activation. In terms of nitrous oksida anaesthesia, the
glutamatergic N-methyl-d-aspartate (NMDA) receptors have been implicated as a major site of
action. NMDA receptors are the natural receptors for endogenous glutamate and are excitatory in
nature. In this way nitrous oksida, as an NMDA antagonist, may inhibit excitatory signalling in
the CNS. At the simplest neurological level, it was found that NMDA receptors were necessary
for the behavioural effects of N2O in the nematode Caenorhabditis elegans, while volatile
anaesthetics such as isoflurane or halothane had another, unspecified mechanism of action [43].
While we cannot directly translate findings in this organism to rodents or humans, NMDA
receptors are a highly conserved structure through phyla, allowing for some level of comparison.
Jevtovic-Todorovic et al. [30] looked at the mechanistic similarities between N2O and other
NMDA receptor antagonists in an in vivo rodent model to discover that N2O acted via NMDA
receptor antagonism. N2O was found to produce neurotoxicity, afford neuroprotection, and
induce blockade of NMDA currents, as well as work in the same age-dependent manner as other
NMDA receptor antagonists such as MK801. It also provides the same sort of dissociative
anaesthesia as ketamine, an NMDA receptor antagonist, overall suggesting that N2O likely works
through this receptor.

Further work has revealed that N2O also has some actions on the two-pore domain TREK-1
potassium channel [44]. This channel functions as a leak channel to release potassium from the
cell, stabilising the resting membrane potential in neurons [45]. Previously, it has been shown
that TREK-1 channels are important for anaesthesia and knockout mice for the channel are
resistant to volatile anaesthetics [46]. TREK-1 has also been found to be important in various
types of pain perception [47]. This ion channel could then be a factor in both the anaesthetic and
analgesic actions of N2O.

Go to:

4. Mechanisms of Neurotoxicity
There are various mechanisms which are responsible for its neurotoxic effects, such as NMDA
antagonism, enzyme inhibition and alteration of cerebral blood flow. Different brain conditions
have different vulnerabilities to each form of toxicity, with neonatal brains more susceptible to
NMDA antagonism, vitamin B12 deficient patients more prone to homocysteine mediated
problems and the damaged brain often more vulnerable to changes in cerebral blood flow.
Because of this, there is a wide range of patients to whom N2O may have some form of toxicity,
with different groups being at greater risks than others. In this way it is extremely important to
understand all the needs of a patient before giving nitrous oksida. The danger arises when nitrous
oksida is given during dental procedures or as emergency analgesia, e.g., en route to hospital,
where underlying problems such as vitamin B12 deficiencies may be undetected. One case study
details a patient who presented with weakness in her lower limbs as well as peripheral numbness
[36]. MRI scans showed abnormalities on the cervical spinal cord consistent with small lesions.
The patient was found to be deficient in vitamin B12 and had been exposed to nitrous oksida for
dental surgeries 23 months previously. Following 10 months of vitamin B12 injections the
symptoms had abated, yet this could have been avoided altogether if nitrous oksida had been
avoided for this patient. This highlights the need to fully elucidate nitrous oksida mechanisms of
toxicity, so that clinicians can make informed decisions regarding N2O use. This case is reflected
in further case reports involving patients with no prior N2O abuse experiencing myelopathies
following N2O anaesthesia [48], as well as patients with a history of N2O abuse [49,50,51,52].

4.1. NMDA Antagonism

NMDA receptors are excitatory receptors in the body which respond to the endogenous agonist
glutamate. NMDA antagonists are known to have both protective and toxic effects depending on
their activation. As glutamate is an excitatory neuromodulator, excessive release, for example
after traumatic brain injury, can lead to excitotoxicity due to high influx of Ca2+ into neurons. In
this way, NMDA antagonists can provide protection against excitotoxic damage [30,53]. This
was shown in a rat model of middle cerebral artery occlusion (MCAO), where 75% N2O
provided a reduction in cortical, but not striatal, infarcts [54]. This was associated with increased
performance in motor coordination tasks compared to MCAO animals with no treatment [55].
While this might suggest some use for N2O as a treatment for stroke due to its NMDA antagonist
features, it has also been shown that N2O has the ability to inactivate tissue plasminogen
activator (tPA), as well as increase haemorhage and blood-brain barrier dysfunction [56]. This
would preclude its use for stroke as the NMDA antagonist benefits are outweighed by the
negative effects.
If administered to the nave brain, however, N2O has the ability to cause neurotoxicity itself. N2O
has been shown to induce cell death in neurons after prolonged exposure, and shorter term
exposure also leads to a more reversible vacuolisation [30,31]. Another potent NMDA
antagonist, MK-801 [57], like nitrous oksida, leads to antagonism and thus reduction of signal
from excitatory glutamatergic neurons. Despite showing anti-convulsive actions [58], it has not
been introduced clinically due to the finding it can form lesions in the brain [59]. It has also been
found to alter the structure and function of hippocampal synapses [60]. Jevtovic-Todorovic et al.
[30] used MK-801 in a comparative study to investigate the possibility that N2O was a NMDA
antagonist, with N2O showing identical physiological outcomes to MK-801. They found that
both drugs induce a similar age-dependent toxicity in older rats. They also discovered that
administration of GABAergic or muscarinic agents was successful in reversing the vacuolisation
of neurons after N2O or MK-801 administration [30]. N2O is not as potent an antagonist as MK-
801 but there appear to be numerous similarities between the two drugs in terms of toxicity.
Similar to N2O, another NMDA receptor antagonist, ketamine, is often used as an anaesthetic
agent. Ketamine is also coming under scrutiny since it was discovered that, like N2O, it has the
ability to induce reversible or irreversible vacuolisation of neurons [19,25]. Ketamine has also
been implicated in causing neuronal cell death by increasing NMDA NR1 subunit expression,
leading to increased cytosolic calcium and increased cell death. However, this was after
prolonged (24 h) expression and there is little evidence of this mechanism being involved in N2O
neurotoxicity.

The neurotoxic actions of NMDA antagonists have been attributed to modulation of GABAergic
inhibition of various neuronal pathways. The PC/RSC has been associated with learning and
memory, as well as pain and awareness. Studies have found that NMDA antagonist
administration can result in increased acetylcholine (ACh) release in the PC/RSC as well as the
septohippocampal pathway, also involved in learning and memory [61,62]. Normally, NMDA
receptors on GABAergic neurons act as an upregulating mechanism to ensure constant inhibitory
GABA release. GABA acts upon receptors on cholinergic neurons in the PC/RSC such that ACh
release is tonically inhibited. NMDA antagonists release this GABAergic inhibition of
cholinergic neurons, allowing ACh release for as long as the NMDA receptor is antagonised.
Both studies suggested that NMDA antagonists act not at the region where ACh release is
recorded, but instead at some separate site, with GABAergic projections between both sites. In
the case of the PC/RSC this was shown to be the basal ganglia [62], while for the hippocampus it
appears to be the medial septum. These findings have been extended to show increased ACh
release in the cerebral cortex of rats exposed for 1 h with 75% N2O [63]. The area postrema is
one of the major centres involved in emesis and can be stimulated by acetylcholine [64]. This
increased cholinergic output has been postulated to underlie the increased nausea and vomiting
often accompanying N2O administration. It has yet to be studied if N2O can have similar effects
on other pathways.
The group led by John Olney [65], who has carried out a mass of work in this field, has referred
to this NMDA hypofunction as a cause of complex excitotoxicity. Antagonism of NMDA
receptors on GABAergic neurons can release other pathways from the inhibitory control of
GABA. These other pathways are usually excitatory in nature, such as the cholinergic pathway
investigated above [62]. This excitatory disinhibition has now been implicated in a range of
disorders such as schizophrenia and Alzheimers disease [66,67,68,69]. While N2O most likely
does not have as severe an effect as MK801 or ketamine due to its shorter duration of action, it is
nevertheless important to consider how these changes in learning and memory centres may affect
the very young or old brain.

4.2. Homocysteine Imbalance

One side effect of N2O which may mediate its toxic effects is indirect inactivation of methionine
synthase, an important enzyme in the remethylation pathway converting homocysteine to
methionine. Nitrous oksida irreversibly binds to the cobalt atom in vitamin B12, also known as
cobalamin, via mechanisms which are not well understood. This leads to oxidation of the enzyme
[70], eventually causing inactivation of vitamin B12. Vitamin B12 is an essential cofactor for
methionine synthase so inactivation leads to a loss of function of the enzyme. In the normal
methionine cycle, methionine is converted to homocysteine (Hcy) via the intermediary molecules
S-adenosyl-methionine (SAM) and S-adenosyl-homocysteine (SAH). From here, homocysteine
can either be irreversibly converted to cystathione (eventually becoming glutathione) via the
trans-sulfuration pathway, or reversibly converted back to methionine by methionine synthase.
Homocysteine, a sulphur-containing amino acid, does not appear to have any inherent function in
the body except as a part of this methionine pathway. However, it is known to have various toxic
effects in the body so any accumulation can be detrimental. A randomised double-blind study
into the effects of anaesthesia post-surgery found that N2O administration was associated with
higher rates of heart attack, even with patients having non-cardiac surgeries [8]. Homocysteine
has been associated with a high rate of cardiac problems [71,72] and patients are found to have
elevated homocysteine levels post-surgery [73,74]. This cardiovascular dysfunction appears to be
regulated by increasing coagulation and endothelial adhesion, promoting atherosclerosis, as well
as altering vascular responses to certain molecules such as argenine via oxidative mechanisms
[75].

Recent interest in homocysteine has revealed multiple pathways by which it causes neurotoxicity
at a cellular and cognitive level. Homocysteine has been shown to act as an agonist on the
glutamate binding site on NMDA receptors, having an opposing effect to N2O (see Figure 1).
While this might suggest that N2O may counteract homocysteine excitotoxicity, in reality N2O is
cleared from the system very quickly following cessation of anaesthesia, while homocysteine is
known to stay elevated in humans serum for days. In adolescents, homocysteine levels return to
baseline between 12 and 24 h [74], while in adults this post-exposure increase is still high at 24 h
[73] and continued elevation has been noted for up to one week [76,77]. Lipton et al. [78] also
report on the dual actions on the NMDA receptor that homocysteine can have. As well as being
an agonist, Hcy can also act as a partial antagonist on the glycine binding site of the NMDA
receptor. While this might imply that the two binding sites for Hcy would cancel each other out,
one being excitatory and one inhibiting the excitatory potential, the reality is more complex,
particularly in a brain injury setting. During brain damage such as stroke or traumatic brain
injury, glycine levels in the brain become elevated and will overpower the partial homocysteine
binding on the NMDA receptor, leading to an elevated excitatory output. Adding to this the
agonistic effect of homocysteine on the glutamate binding site, this achieves an even higher level
of excitotoxic damage [78].

Figure 1

An overview of the homocysteine-mediated pathway of cell death induced by N 2O

exposure. N2O inhibits the action of vitamin B12, an essential cofactor in the
conversion of homocysteine to methionine. This inhibition of vitamin B 12 leads to a
buildup of ...

It is known that methionine synthase inactivation is very fast in rats as compared with humans,
with a half-time in rats of 5.4 min vs. 40 min in humans when exposed to 50% N2O [32] (See
Table 2). However, 40 min is still well within the time frame of human exposure during surgery
so this should not be taken lightly. Often, these increases in homocysteine may not reach
detrimental levels in normal patients, but patients at high risk for hyperhomocysteinemia (HHcy;
classed as Hcy levels > 15 mol/L) could be severely affected if given N2O anaesthesia. There
are numerous risk factors for HHcy such as Alzheimers disease [79], vitamin B12 deficiency
[80], MTHFR gene mutation [81], age and gender [82]. One striking example of this is a case
report involving a young child (3 months old) with an MTHFR gene mutation, leading to an
MTHFR enzyme deficiency [35]. This enzyme is important in the remethylation pathway and
deficiencies have been linked to HHcy. This particular patient was administered 60% N2O on
two occasions during surgery, and within 3 weeks after surgery was admitted to hospital
suffering from seizures. Less than 2 months post-surgery the patient had died and was found to
have severe lesions in the brain, as well as nerve demyelination. At such a young age it is
probable that the brain was extremely sensitive to molecular changes and the rapid and extreme
increase in homocysteine levels appears to have been involved in the patients death. This again
highlights the need for clinicians to be vigilant in ensuring their patients are not at risk if exposed
to N2O. It also showcases the range of physiological parameters which are important to be aware
of before administering N2O, which dentists and paramedics, who routinely use N2O as an
analgesic and anxiolytic, do not normally have access to.

4.3. Reactive Oxygen Species and Mitochondrial Dysfunction

A range of molecules involved in apoptotic mechanisms in neurons have been found associated
with increased homocysteine levels. One of homocysteines main mechanisms of cellular
damage is oxidative stress, which involves the formation of reactive oxygen species (ROS). ROS
are strongly involved in apoptosis and cell death so any increase in levels will be detrimental. It
was discovered that NMDA receptor activation leads to production of O2 free radicals in
cerebellar granular cells [83], peroxyinitryte (ONOO) in the midbrain [84] and various ROS in
the forebrain [85,86]. Since homocysteine can act as an NMDA agonist this may cause increases
in ROS. Increased intracellular Ca2+ following NMDA receptor activation may account for the
increased ROS, whilst ROS can themselves cause a rise in intracellular Ca2+ [87,88]. As seen in
Figure 1, this increased Ca2+ can lead to disturbances in mitochondrial function, resulting in the
production of ROS [85,89]. This mitochondrial dysfunction may be a major pathway involved in
homocysteine mediated neurotoxicity. It is interesting to note that oxidative stress and
mitochondrial ROS formation play a role in Alzheimers disease (AD) pathogenesis [90], and it
has been shown that high plasma homocysteine is a reliable biomarker for AD [91], although
there is no clear consensus as to any causal relationship [92]. AD treatments which act as NMDA
antagonists (e.g., memantine) have been shown to reduce homocysteine mediated
neurodegeneration [93]. This implies a common underlying mechanism between the two and
highlights the damage that can result from homocysteine overload in the brain.

Increased intra-mitochondrial Ca2+ also induces formation of mitochondrial permeability

transition pores (MPTP), which allows release of cytochrome C from the mitochondria.
Cytochrome C can then bind with APAF (Apoptotic Protease Activating Factor) to form an
apoptosome, leading to downstream activation of caspase 3, resulting in apoptosis and cell death
[94]. It has been shown that the vacuolisation resulting from N2O exposure is in fact massive
swelling of mitochondria [31]. Drugs increasing mitochondrial membrane stability have been
shown to be protective against the neurotoxic effects of N2O when combined with midazolam
and isoflurane [95]. This membrane stabilisation was associated with improved cognition in rats
tested [95].

4.4. In Combination with Other Anaesthetics

While N2O induced anaesthesia may not show convincing evidence of danger to some, it is also
prudent to assess the toxicity of N2O in combination with clinically relevant anaesthetic agents to
more closely mimic real world scenarios. There are a series of papers which combine N2O with
isoflurane which consistently show an increase in neuroapoptosis when the two are combined
over either agent alone [17,20,96]. These findings have even been replicated in a non-human
primate model, the rhesus monkey [23]. It appears that this neurotoxicity is correlated with age;
younger animals are susceptible to increased neurodegeneration with isoflurane addition, while
adults are less prone to neuronal damage [18,23]. This may be related to the dual function of
GABAergic neurons. In young animals, GABAergic neurons are excitatory in nature for a short
period postpartum, while in older animals they take on their normal inhibitory function [97]. This
may mean that alongside N2O induced excitotoxicity, isoflurane, a GABA receptor agonist, can
induce extra excitotoxicity, while in adults isoflurane may counteract the excitotoxicity. This
excitatory GABAergic action is also found in humans in the few weeks after birth [98], which
would suggest that this same enhancement of N2O excitotoxicity by isoflurane or any GABA
agonist could be present in humans.

Go to:

5. Strategies to Minimize Toxicity of N2O

The primary concern in medicine is to cause no harm; therefore it would not be possible to
perform procedures without anaesthesia. The stress and damage caused by this would be greater
than any deleterious side effects from N2O anaesthesia. However, although N2O has been used
for over a century, it should not be excluded from examination, and if similar or better
alternatives are available, they should perhaps be utilised.

A number of possible adjuncts have been put forward. Xenon, another gaseous anaesthetic agent,
has been found to be neuroprotective in comparison with other anaesthetics, including N2O [13]
and has already begun clinical trials for neonates at risk for hypoxic brain damage (CoolXenon2
ISRCTN75602528; Toby XeISRCTN08886155). Melatonin also shows neuroprotective
promise when combined with anaesthetics [18]. This may be even more relevant for N2O due to
the proven effect of melatonin in decreasing homocysteine mediated neurotoxicity in animal
studies [99,100].

In terms of possible replacements, a few studies have looked at remifentanil, a fast acting opioid
analgesic. Due to its speed of recovery, it has been suggested as a replacement for N2O in
neurosurgery, as it does not adversely affect cerebral blood flow, unlike N2O [101]. Remifentanil
has also been suggested as a labour analgesic agent if administered intravenously [102].

Go to:

6. Conclusions
At the moment, it is premature to suggest that N2O should be discontinued as an anaesthetic
agent. However, the growing body of evidence does support the theory that N2O has some
neurotoxic effects and these results should not be taken lightly. Nitrous oksida is regularly used
for neonatal surgery and, as shown, this is a high risk period for neurodevelopment. It is difficult
to assess the long term cognitive outcomes in humans, but rat studies suggest long term
developmental issues such as memory impairment. Nitrous oksida is also often used in elderly or
brain damaged patients and it is clear from numerous studies, such as the ENIGMA trial, that
N2O is not as harmless as some might believe. It is important that further molecular work be
carried out to determine the pathways by which N2O has its toxic effects, as these pathways may
reveal areas for drug development to replace or work alongside N2O to mitigate its neurotoxic
effects. It would also be advisable to carry out further studies on non-human primates to
determine any differences between rodent studies. At the moment, from rodent studies, we can
only make educated assumptions on what might occur in humans. Non-human primates can help
bridge this gap in knowledge without compromising patient safety in clinical trials.
(The Neurotoxicity of Nitrous Oksida: The Facts and
Putative Mechanisms)
1. Introduction: The use of opioids has been increasing in operating room and intensive
care unit to provide perioperative analgesia as well as stable hemodynamics. However,
many authors have suggested that the use of opioids is associated with the expression of
acute opioid tolerance (AOT) and opioid-induced hyperalgesia (OIH) in experimental
studies and clinical observations in dose and/or time dependent exposure even when used
within the clinically accepted doses. Recently, remifentanil has been used for pain
management during anesthesia as well as in the intensive care units because of its rapid
onset and offset.

2. Objectives: Search of the available literature to assess remifentanil AOT and OIH based
on available published data.

3. Methods: We reviewed articles analyzing remifentanil AOT and OIH, and focused our
literature search on evidence based information. Experimental and clinical studies were
identified using electronic searches of Medline (PubMed, Ovid, Springer, and Elsevier,
ClinicalKey).

4. Results: Our results showed that the development of remifentanil AOT and OIH is a
clinically significant phenomenon requiring further research.

5. Discussions and Conclusions: AOT defined as an increase in the required opioid dose
to maintain adequate analgesia, and OIH defined as decreased pain threshold after
chronic opioid treatment, should be suspected with any unexplained pain report
unassociated with the disease progression. The clinical significance of these findings was
evaluated taking into account multiple methodological issues including the dose and
duration of opioids administration, the different infusion mode, the co-administrated
anesthetic drugs effect, method assessing pain sensitivity, and the repetitive and
potentially tissue damaging nature of the stimuli used to determine the threshold during
opioid infusion. Future studies need to investigate the contribution of remifentanil
induced hyperalgesia to chronic pain and the role of pharmacological modulation to
reverse this process.

6. Introduction
7. Even though remifentanil increases analgesia and respiratory depression in a dose-
dependent manner (Hughes et al., 1992; Egan et al., 1993; Glass et al., 1993;
Westmoreland et al., 1993; Kapila et al., 1995), these effects disappear rapidly after
discontinuing administration of the drug because of the extremely short elimination half-
life (9.5 4 min). Especially, remifentanil has the shortest context-sensitive half-time and
terminal elimination half-life among other opioid after 3-h infusion (Kapila et al., 1995).
 Therefore, remifentanil can be given in high doses throughout surgery without the risk of
delayed postoperative recovery or respiratory depression. Because of its
pharmacodynamic and pharmacokinetic effects, remifentanil has been used in clinical
anesthesia as an induction and maintenance agent, and postoperative pain management in
the intensive care units.

8. Most of the studies conducted with remifentanil showed cardiovascular responses during
perioperative manipulations. They recommended a bolus injection of remifentanil of 1
g/kg as more effective dose in reducing the pressor response during laryngoscopy and
tracheal intubation (McAtamney et al., 1998; OHare et al., 1999). However, while the
cardiovascular responses reaches a peak 12 min after laryngoscopy and intubation, and
usually subsides within 56 min (Singh et al., 1995), the context-sensitive half-time of
bolus remifentanil is only 3.2 min (Glass et al., 1993; Kapila et al., 1995). Therefore,
remifentanil bolus alone is not enough to attenuate the responses and the use of a bolus-
infusion regimen is required (McAtamney et al., 1998). The commonly accepted and
recommended dose of remifentanil is 1 g/kg followed by an infusion of 0.51 g/kg/min
for induction of anesthesia or 0.052.0 g/kg/min for maintenance of anesthesia (Burkle
et al., 1996; Hall et al., 2000; Sneyd et al., 2001). In postoperative period, remifentanil
continuous infusion (CI) also can be used for controlling the pain, and the final
remifentanil infusion rates have been reported as 0.050.26 g/kg/min for satisfactory
analgesia after surgery (Bowdle et al., 1996, 1997; Schuttler et al., 1997; Yarmush et al.,
1997; Sneyd et al., 2001).

9. Common concerns regarding the use of opioids are potential detrimental side effects,
physical dependence, and addiction. However, an additional concern has recently risen
that these opioids can induce an acute tolerance and hyperalgesia in dose and/or time
dependent manner even when used within clinical accepted doses. They provide straight
analgesic and antihyperalgesic effects originally, but subsequently are associated with
expression of hyperalgesia (Angst and Clark, 2006). The use of opioids may seem to be a
double-edged sword. In other words, patients receiving opioids to control their pain
somewhat paradoxically may become more sensitive to pain as a direct result of opioid
therapy.

10. Therefore, a review of literature was carried out to analyze acute tolerance and/or
hyperalgesia induced by remifentanil in perioperative period using electronic searches of
Medline (PubMed, Ovid, Springer, and Elsevier, ClinicalKey). The objective was to
address the following issues: (1) what is the definition of acute opioid tolerance (AOT)
and opioid-induced hyperalgesia (OIH)? (2) Dose remifentanil may induce the acute
tolerance and hyperalgesia? (3) Are AOT and OIH significant enough to consider
reducing the dose of remifentanil or adopting preventive modulations?
11. Definition of Acute Opioid Tolerance and Opioid-
Induced Hyperalgesia
12. Before we discuss AOT and OIH induced by remifentanil, we have to understand the
definitions of AOT and OIH because management for pain control is different according
to make a diagnosis one of them; While AOT can be overcome by increasing the dosage,
OIH even worsens the pain. Consequently, OIH is reduced by reducing or eliminating the
opioid.

13. Opioid-induced hyperalgesia is defined as a state of nociceptive sensitization, which is

characterized by a paradoxical response, whereby a patient receiving opioids for pain
treatment might have an increased sensitivity to painful stimuli (Angst and Clark, 2006;
Chu et al., 2008). Even though controlled preclinical experiments have defined OIH in
animals as a decrease in pain threshold from baseline after chronic administration of
opioids (Fishbain et al., 2009), there still is no accepted operational definition of OIH
among researchers in human clinical trials; hyperalgesia is defined as decrease in either
pain threshold or pain tolerance after chronic opioid exposure (Martinez and Fletcher,
2012). Pain threshold is the lowest intensity of stimulation at which pain is experienced,
and pain tolerance is the amount of pain from a given stimulus a person can handle before
seeking relief.

14. Opioid-induced hyperalgesia often is confused with opioid tolerance because of the
manifestations of similar symptoms. AOT is defined as an increase in the dose required
maintaining adequate analgesia in patients receiving opioid medication for the treatment
of pain in clinical settings (Angst and Clark, 2006; Chu et al., 2008). Clinical data
indicates that early postoperative pain scores and subsequent greater demand of opioids
could be attributed to tolerance, while the greater requirement for opioids at a later
recovery stage could be associated with OIH after high-dose remifentanil anesthesia.

15. General Consensus of AOT and OIH

16. Recently, many studies have focused on the development of AOT and OIH after using
opioids based on idea that OIH might be a potential risk factor for the development of
chronic pain after surgery (Perkins and Kehlet, 2000; Woolf and Salter, 2000; Wilder-
Smith and Arendt-Nielsen, 2006). The circumstances under which OIH may occur are
also not yet entirely understood but may include high doses, long-term treatment, or
abrupt changes in concentrations (Simonnet and Rivat, 2003). Most authors have
suggested that the higher dose of opioids induce the higher tolerance and/or hyperalgesia.
OIH was observed either to follow analgesia and lasted long after opioid exposure ended
(Celerier et al., 2000; Kissin et al., 2000; Li et al., 2001), or during continuous opioid
exposure (Laulin et al., 1999; Vanderah et al., 2000). It has been reported that high
intraoperative doses of opioids not only increase postoperative pain scores and acute
morphine consumption, but also induce significant nociceptive threshold changes (Chia
 et al., 1999; Guignard et al., 2000; Mao, 2002; Angst et al., 2003; Joly et al., 2005; Van
Elstraete et al., 2005; Cabanero et al., 2009). Celerier et al. (2000) showed that fentanyl
injection was associated with sustained lowering of the nociceptive threshold below
baseline value, and the higher the fentanyl dose used, the more pronounced was the
fentanyl-induced hyperalgesia. Laulin et al. (1999) reported that repeated once-daily
heroin injections induced a gradual lowering of the nociceptive threshold which
progressively masked a sustained heroin analgesic functional effect, which is suggested
as opiate tolerance.

17. If all types of opioids have been shown to induce such a dose-dependent hypersensitivity,
exposure to short-acting opioids, such as remifentanil, seems more likely to be
responsible for postoperative high pain scores, high morphine consumption, and
hypersensitivity to pain (Guignard et al., 2000; Angst et al., 2003; Hood et al., 2003;
Koppert et al., 2003a). A relatively large-dose of intraoperative remifentanil has been
shown to induce postoperative hyperalgesia more rapidly and more frequently as
compared with longer-acting opioids (Derrode et al., 2003; Joly et al., 2005; Koppert and
Schmelz, 2007). These results are in agreement with the clinical observation of increased
postoperative pain and morphine requirement (Zarate et al., 1999; Guignard et al., 2000,
2002).

18. As many papers suggest, remifentanil could induce AOT and/or OIH (Vinik and Kissin,
1998; Hayashida et al., 2003; Gomez de Segura et al., 2009; Aguado et al., 2011). Many
authors documented that remifentanil CI alone induced AOT in the dose-dependent
manner within first few hours in animal models (Hayashida et al., 2003; Aguado et al.,
2011). A study in healthy human volunteers performed by Vinik and Kissin (1998) also
suggested that AOT was profound and developed very rapidly, and remifentanil of 0.1
g/kg/min resulted in the maximum analgesic effect in 6090 min, and then began to
decline despite the constant-rate infusion, eventually reaching 1/4 of the peak value after
3 h of infusion, measuring by cold thermal and mechanical noxious stimulation (Vinik
and Kissin, 1998).

19. A small number of clinical studies have looked at AOT and OIH in the setting of acute
perioperative opioid exposure. A series of studies in patients undergoing surgery
suggested that exposure to a high rather than to a low intraoperative opioid dose was
associated with the increased opioid consumption and/or pain in the postoperative period
(Cooper et al., 1997; Chia et al., 1999; Guignard et al., 2000; Joly et al., 2005). A feasible
explanation for these findings is either the development of acute tolerance on the rescue
opioids for controlling the postoperative pain or a possible OIH in patients exposed to a
high intraoperative opioid dose (Angst and Clark, 2006). Guignard et al. (2000)
suggested that AOT as well as OIH might be induced by the acute exposure to large doses
of opioids. The patients received 0.3 0.2 g/kg/min of the intraoperative remifentanil
 required morphine significantly earlier and needed nearly twice more morphine in the
first 24 postoperative hours than those who remifentanil was kept constant at 0.1
g/kg/min. Furthermore, despite higher morphine requirement, higher pain scores were
observed in group that remifentanil was infused higher. Joly et al. (2005), in their
randomized, double-blind study, showed that intraoperative remifentanil of 0.4 g/kg/min
triggered the larger hyperalgesia as well as the larger morphine consumption for 48
postoperative hours, compared with remifentanil of 0.05 g/kg/min. In prospective,
randomized, double-blind study, they suggested that remifentanil of mean 0.28 g/kg/min
was associated with the development of clinically relevant AOT, in patients underwent
the general anesthesia using propofol infusion (Crawford et al., 2006).

20. Many authors showed that the OIH can develop differently for different types of pain
such as transdermal electrical stimulation (Angst et al., 2003; Hood et al., 2003; Koppert
et al., 2003a,b; Troster et al., 2006), cold pressor pain (Compton et al., 2003, 2004), and
pressure-evoked pain (Luginbuhl et al., 2003). Transdermal electrical stimulation used for
inducing mechanical hyperalgesia on an experimental skin lesion rendered hyperalgesia
in human volunteers. These investigators suggested that 3090 min of remifentanil
exposure significantly enlarge the skin area with pre-existing mechanical hyperalgesia by
transdermal electrical stimulation with direct relation of the infusion duration and the
opioid dose (Angst et al., 2003; Hood et al., 2003; Koppert et al., 2003a,b; Troster et al.,
2006). This hyperalgesia was sustained up to 4 h after stopping infusion (Hood et al.,
2003). They also documented that the pain score was increased and the pain threshold
was decrease, by dose-dependent manner, after the discontinuation of opioids (Angst et
al., 2003; Hood et al., 2003; Koppert et al., 2003a). In a model of acute physical opioid
dependence, which withdrawal was precipitated with the opioid antagonist naloxone after
a single injection of opioid (Compton et al., 2003, 2004), the authors showed that
sensitivity to cold pressor pain was significantly increased after injection of naloxone and
OIH may be triggered if opioid effect is suddenly disappeared or reversed. Finally,
Luginbuhl et al. (2003) suggested that remifentanil alone induced significant hyperalgesia
detected by test for pressure pain tolerance threshold in volunteers.

21. Suggested Doses of Remifentanil Inducing AOT and

OIH
22. Most articles documented that AOT and OIH were induced during and after remifentanil
infusion at 0.1 g/kg/min (Vinik and Kissin, 1998; Glass et al., 1999; Guignard et al.,
2000; Gustorff et al., 2002; Angst et al., 2003; Koppert et al., 2003a; Joly et al., 2005;
Angst et al., 2009; Cabanero et al., 2009). In a mouse model, they suggested that
remifentanil-induced hyperalgesia was induced by dose-dependent manner and the
calculated ED50s for thermal and mechanical hyperalgesia was 1.7 (95% confidence
interval, 1.32.1) and 1.26 (1.01.6) g/kg/min, respectively (Cabanero et al., 2009).
Infusion of remifentanil at a rate of 0.1 g/kg/min induced the very rapid and profound
 acute tolerance during infusion (Vinik and Kissin, 1998; Glass et al., 1999), while a rate
of 0.08 g/kg/min could not induce acute tolerance in the cold pressor test and in models
of electrical and heat pain (Gustorff et al., 2002). Angst et al. (2003) showed that pre-
existing mechanical hyperalgesia was significantly aggravated after discontinuation of
remifentanil at a rate of 0.10 g/kg/min. Koppert et al. (2003a) also suggested that even
though remifentanil dose-dependently reduced pain and mechanical hyperalgesia during
the infusion, after discontinuation of infusion, pain and mechanical hyperalgesia
increased significantly in group which was administered remifentanil at a rate of 0.10
g/kg/min compared with at a rate of 0.05 g/kg/min. Furthermore, in the patients
receiving higher intraoperative remifentanil, such as 0.3 0.2 and 0.4 g/kg/min, than
0.10 g/kg/min, the time of first morphine requirement was significantly earlier and the
morphine consumption in the first 24 or 48 postoperative hour was significantly lager,
which suggest that remifentanil causes AOT and hyperalgesia (Guignard et al., 2000; Joly
et al., 2005). Therefore, AOT as well as OIH might be induced by the acute exposure to
large doses of opioids.

23. In studies on AOT and OIH, while most authors have used the CI mode, there are a few
reports using the target-controlled infusion (TCI) mode (Hood et al., 2003; Angst et al.,
2009; Shin et al., 2010; Richebe et al., 2011). Hood et al. (2003) showed that areas of
hyperalgesia continuously enlarged 4 h after remifentanil (targeted concentration of 3.1
1.2 ng/ml) was stopped, to 180% 47%. Shin et al. (2010) suggested that remifentanil
using TCI at 4 ng/ml induced the more increased cumulative morphine consumption and
postoperative hyperalgesia than that at 1 ng/ml during sevoflurane anesthesia. This result
is similar with that of previous studies using CI mode; higher dose of opioid, higher
development of AOT and OIH. The infusion rate for remifentanil 0.1 g/kg/min can
achieve a stable plasma concentration ranging between 2.7 and 2.9 ng/ml during the
infusion (Angst et al., 2003). A bolus of remifentanil 1 g/kg followed by infusion 0.2
g/kg/min will produce stable plasma concentrations of 45 ng/ml within a few minutes
(Minto et al., 1997). Angst et al. (2009) documented that target remifentanil
concentrations corresponding to infusion rates of 0.065 and 1.3 g/kg/min did not induce
tolerance in any of their pain models. Remifentanil concentrations of 1.6 and 3.2 ng/ml,
which are commonly used in a clinical setting to provide analgesia during surgery,
correspond to steady-state concentrations achieved when infusing remifentanil at a
constant rate of about 0.065 and 0.13 g/kg/min (Glass et al., 1999). Therefore, according
these references, 0.1 g/kg/min using CI mode and 2.7 ng/ml using TCI mode seem to
be sufficient to develop hyperalgesia.

24. Controversial Results on Development of AOT and

OIH by Remifentanil
25. There is controversial results on AOT and OIH induced by remifentanil at the
perioperative period, even though many authors suggested that remifentanil could induce
 them (Schraag et al., 1999; Cortinez et al., 2001; Hansen et al., 2005; Lahtinen et al.,
2008; Angst et al., 2009; Yeom et al., 2012). Angst et al. (2009) documented that 3 h
infusion of remifentanil of up to 4.0 ng/ml was not associated with the development of
significant tolerance to analgesic in placebo-controlled, double-blind study. Cortinez et
al. (2001) also suggested that remifentanil-based anesthesia (0.23 0.10 g/kg/min,
average duration; 116 min) did not induce the AOT when compared with sevoflurane-
based anesthesia in patients undergoing elective gynecologic surgery. Hansen et al.
(2005) investigated how remifentanil of 0.4 g/kg/min intra-operatively affected post-
operative pain and opioid consumption after major abdominal surgery. In a double-blind
study, although they found higher visual analog scale score in the remifentanil group
compared with placebo during the immediate postoperative period, this difference was no
longer significant 2 h after surgery or during the remainder of the 24-h observation
period. A prospective, randomized, double-blind study, showed that 3 h infusion of
remifentanil of 0.3 g/kg/min did not increase postoperative pain or opioid consumption
in cardiac surgery patients who underwent sufentanil/propofol-based general anesthesia
(Lahtinen et al., 2008). Yeom et al. (2012) suggested that remifentanil did not appear to
cause AOT and OIH in patients undergoing spinal fusion even though there was
significant differences in the mean intraoperative infusion rate of remifentanil (0.16
g/kg/min vs. 0.03 g/kg/min) for a short period of time (averaging 225 and 216 min,
respectively). Schraag et al. (1999) found that Remifentanil infusion using specially
designed patient-maintained TCI systems, which was controlled according to the patients
demand for postoperative pain control by up-and-down method (intervals 0.2 ng/ml) did
not show any evidence of rapid development of acute tolerance in patients maintained
hypnotic dose of remifentanil in the first six postoperative hours from end of total
intravenous anesthesia with remifentanil and propofol.

26. Factors that Lead to Discrepancies Regarding AOT

and OIH
27. The clinical relevance of above mentioned results is questionable and there is some
limitation in negative results concerning AOT and OIH. These discrepancies can be
explained by multiple methodological issues including the administrated dose and
duration of opioids administration, the different infusion mode, the co-administrated
anesthetic drugs effect, method assessing pain sensitivity, and the repetitive and
potentially tissue damaging nature of the stimuli used to determine the threshold during
opioid infusion (Angst and Clark, 2006; Lahtinen et al., 2008; Fishbain et al., 2009;
Simonnet, 2009).

28. First, we can explain these discrepant results by differences in exposed opioid doses and
administration duration. Studies reporting positive results have shown that AOT develops
in dose-dependent fashion and only becomes evident when total opioid exposure is quite
high. The non-significant increase of postoperative pain and opioid consumption in
 studies reporting negative results may be noticed because of lower total intraoperative
opioid exposure when compared with the positive results, suggesting a dose-dependent
effect of opioids on the development of OIH (Guignard et al., 2000; Cortinez et al., 2001;
Lee et al., 2005). Cabanero et al. (2009) agreed that remifentanil induced dose-dependent
pronociceptive effects for thermal and mechanical hyperalgesia, which lasted longer with
higher doses, but they suggested that the duration of infusion did not alter the
pronociceptive effects of remifentanil. This negative result might be explained by the
shorter exposed duration, just over 30 or 60 min, than the positive results studies, in
which Acute tolerance is typically investigated during CI over 23 h, whereas
hyperalgesia is usually assessed within 1 h post-infusion (Guignard et al., 2000, 2002;
Laulin et al., 2002; Hayashida et al., 2003; Gomez de Segura et al., 2009; Benito et al.,
2010; Aguado et al., 2011; Ishida et al., 2012; Ishii et al., 2013). In the recent animal
study, they showed that intravenous remifentanil infusion alone induced transient
hyperalgesia associated with the duration of exposure to remifentanil (Ishida et al., 2012).
While 30 min remifentanil infusion did not induce hyperalgesia, 120 min remifentanil
infusion induced the hyperalgesia regardless of dose. However, hyperalgesia was not
sustained more than 60 min. Other study also documented that remifentanil-induced
hyperalgesia started from 2 h after surgery and reached its peak at 2448 h after surgery
(Gu et al., 2009).

29. Also, it is not sufficient to explain the cause of these discrepancies by using the dose and
duration of remifentanil infusion. Some authors have reported the negative results even
though they used remifentanil infusion rate that was enough to develop hyperalgesia. It
can be partially explained by the effect of co-administrated anesthetic drugs, such as
propofol, sevoflurane and nitrous oksida. Fodale et al. (2006) suggested that AOT was
not induced when remifentanil was co-administered with propofol or sevoflurane, which
produced an inhibiting effect at NMDA receptors neutralizing the remifentanil
stimulation on these receptors. However, Solt et al. (2006) documented that sevoflurane
has only a minimal inhibitory effect on NMDA receptors, which are considered to be
involved in the development of opioid-related hypersensitivity (Tompkins and Campbell,
2011). Relatively low sevoflurane concentrations (1.0%) reverse OIH, but there was the
lack of effect of sevoflurane concentrations of 1.0 and 1.5% to oppose hyperalgesia
following high-dose opioids (Solt et al., 2006; Richebe et al., 2009). Shin et al. (2010)
also suggested that remifentanil-induced hyperalgesia was not apparent during propofol
anesthesia compared with the effect produced during sevoflurane anesthesia even though
dosage of remifentanil was increased from 1.0 to 4.0 ng/ml. This result can be supported
by that propofol may have some modulatory effect on OIH, through inhibition of the
NMDA subtype of the glutamate receptor (Orser et al., 1995; Kingston et al., 2006),
which is one of the potential mechanisms that induced the OIH, and possibly through
interactions with gamma-aminobutyric acid (GABA-A) receptors at the supraspinal level
(Wang et al., 2004; Singler et al., 2007). Specifically, sub-hypnotic dose of propofol has
 analgesic effects, which delayed the onset of anti-analgesia after remifentanil infusion
(Singler et al., 2007). However, the clinical significance of these findings, especially in
higher dosages used in the intraoperative setting, remains to be studied. In addition, some
authors ignored the impact that nitrous oksida might have against the AOT and
hyperalgesia (Cortinez et al., 2001; Yeom et al., 2012). Echevarria et al. (2011) reported
that the group using the 70% nitrous oksida with remifentanil of 0.3 g/kg/min showed a
greater decreased mechanical threshold than the group without nitrous oksida at
postoperative 1218 h, even though the postoperative pain scores and cumulative
morphine consumption was similar between the groups. Lee et al. (2005) suggested that
70% nitrous oksida showed comparable effect on the postoperative opioid consumption
similar to remifentanil at mean 0.17 g/kg/min.

30. Next, different infusion mode also can influence the development of AOT and OIH even
though they have an equipotential effect on the pain control. TCI has been shown not
only to improve intraoperative hemodynamic stability but also to decrease intraoperative
remifentanil requirements (Minto et al., 1997; De Castro et al., 2003). Interestingly,
Richebe et al. (2011) evaluated whether the use of TCI mode also would lead to decrease
in early postoperative period hyperalgesia after cardiac surgery. They suggested that an
infusion of intraoperative remifentanil using TCI mode (7 ng/ml) reduced postoperative
hyperalgesia, compared with that using CI mode (0.3 g/kg/min), and this decrease in
postoperative hyperalgesia was sustained and lasted throughout the 1st postoperative
week. The conclusion of the study was supported by the difference of intraoperative
infused total remifentanil dose, which was greater in CI than in TCI group.

31. Then, the possibility of cumulative tissue injury manifesting as AOT or OIH should be
carefully excluded (Petrenko et al., 2012). Especially, this is likely to occur with
repetitive testing protocols in a situation in which protective withdrawal reflexes are
impaired or abolished by opioid administration. In Luginbuhl et al.s (2003) study, they
exposed volunteers to significantly higher nociceptive input during remifentanil vs.
during saline placebo administration. It cannot be excluded that postinfusion hyperalgesia
resulted from more intense noxious stimulation during the remifentanil infusion rather
than the opioid administration itself. On the other hand, Ishii et al. (2013) documented
that neither acute tolerance nor hyperalgesia was observed even in the setting when they
used a tapered remifentanil infusion to rapidly attain maximum analgesic effect of
remifentanil and tried to minimize the repetitive and potentially tissue damaging nature of
the stimuli.

32. Finally, it might contribute to influence of development of OIH, in studies exposing

patients to high intraoperative opioid doses, whether assessment of pain sensitivity before
and after surgery was formally done or not. No causal relationship between acute
perioperative opioid exposure and development of OIH can be established without direct
 measurement of pain sensitivity. If patients have a comorbidity affecting sensory
thresholds preoperatively, this condition could distort postoperative measures on OIH.
The German Network on Neuropathic Pain established a standardized quantitative
sensory testing (QST) protocol to investigate the somatosensory thresholds in healthy
subjects and in patients with neuropathic pain (Rolke et al., 2006). Reference values from
healthy subjects could be used to establish normal sensory functioning in patients before
anesthesia. In a clinical setting, this direct measure could be used for distinguishing
between OIH and AOT, because of clinical importance, as AOT can be overcome by dose
increase, while OIH may be aggravated by the same intervention. Without direct
measures to assess hyperalgesia such as QST, the results are not easy to distinguish from
acute tolerance (Vinik and Kissin, 1998). Clinical studies examining remifentanil-induced
hyperalgesia by QST showed that, at a relatively high dose, remifentanil decreased the
pain threshold (Gustorff et al., 2002; Joly et al., 2005; Schmidt et al., 2007; Song et al.,
2011; Yalcin et al., 2012). These studies used remifentanil infusions at clinically standard
rates, and all of them showed clear hyperalgesia either shortly after discontinuing
infusion or 1 and 2 days postoperatively. However, most studies confirmed the reality of
this phenomenon using indirect evidences, such as greater postoperative pain and
morphine consumptions instead of using QST (Guignard et al., 2000; Rauf et al., 2005).
Treskatsch et al. (2014) did not demonstrate any influence on clinical outcome
parameters of pain after remifentanil of >0.2 g/kg/min, and they suggested that indirect
evidences might be less sensitive to detect OIH compared with QST. Therefore, we think
that a further study is needed to reveal the development of AOT and OIH using QST.

33. Importance of AOT and/or OIH in Considering the

Reduction of Remifentanil Dosage and Adopting
Preventive Modulations
34. Forty-one percentage of all surgical patients still experience moderate to severe acute
postoperative pain and that 24% experience inadequate pain relief (Dolin et al., 2002). It
has recently been highlighted that chronic pain as a direct result of surgery is more
common than previously recognized (Perkins and Kehlet, 2000). Furthermore, in cohort
study on postoperative remifentanil-induced hyperalgesia, its incidence was reported
16.1% of patients undergoing general anesthesia with remifentanil (Ma et al., 2011).
Hyperalgesia and increased pain in the postoperative period is now considered a major
candidate mechanism for the development of chronic pain (Perkins and Kehlet, 2000;
Woolf and Salter, 2000; Wilder-Smith and Arendt-Nielsen, 2006). Therefore, less
postoperative hyperalgesia results in better acute postoperative pain control (Joly et al.,
2005), and interventions associated with alterations of postoperative hyperalgesia are also
associated with changes in acute postoperative pain outcomes (Guignard et al., 2000;
Rauf et al., 2005).
35. However, there is a lack of good quality clinical research in this area, despite the fairly
extensive basic science evidence. In a structured evidence-based review for all levels of
evidence on OIH in humans (Fishbain et al., 2009), they suggested that there was not
sufficient evidence to support or refute the existence of OIH in humans except in the case
of normal volunteers receiving opioid infusions. There was consistent evidence that
opioid infusion in normal volunteers induced either an increase in secondary hyperalgesia
or allodynia, and there was inconsistent evidence on pain threshold and tolerance in
normal volunteers or chronic pain patients although the threshold decreased with opioid
infusion. They also documented that using of opioid in perioperative period increased the
postoperative pain or opioid requirements with inconsistent evidence. In other recent
evidence based systematic review (Rivosecchi et al., 2014), they suggested that
remifentanil might induce a degree of hyperalgesia, but it did not reach a level of clinical
significance that requires prevention.

36. There are some questions regarding the importance of the drug use to prevent AOT and
OIH in postoperative patients and whether the assessment of OIH at immediate
postoperative period is suitable. Clinical studies assessing the preventive effect of drugs
on OIH in the immediate postoperative period, showed that the clinical benefit is either
absent (Echevarria et al., 2011), limited to a moderate opioid-sparing effect (Joly et al.,
2005; Shin et al., 2010), or a slight reduction in pain scores (Shin et al., 2010; Song et al.,
2011). According to these results, Martinez and Fletcher (2012) suggested that the
immediate postoperative period may not be the optimal period to detect the preventive
effects on OIH, even though additional clinical data need to confirm it.

37. Furthermore, Simonnet and Rivat (2003) suggested that OIH should be considered as a
normal adaptive response counteracting the perturbations caused by administration of
analgesic opioids. However, it is tempting to speculate that the long lasting hyperalgesia
induced by endogenous or exogenous opioids may still facilitate learning processes and
memorization of drives so that environmental changes which might lead to pain may be
better avoided. From a medical viewpoint, OIH following a first opioid administration is
not a passive response but it might be considered as the first step of an active process
leading to pain sensitization. This suggests that opioids have reinforced a nociceptive
memory which could contribute to pain chronicization.

38. Conclusion
39. Current experimental and clinical data generally support the development of AOT and
OIH in specific settings such as acute remifentanil exposure in human volunteer and
postsurgical pain cohorts when remifentanil was infused at 0.1 g/kg/min either alone or
with inhalation anesthetics. Therefore, in these situations, clinicians need to be cautious
for the possibility of the development of AOT and OIH, which may impair treatment of
pain or even aggravate preexisting pain. Clinicians should suspect manifestation of OIH
 when opioid treatment effect seems to decline in the absence of disease progression, with
unexplained pain reports or allodynia unassociated with the site of injury. According to
the previous reported results, co-administrated anesthetic drugs, such as propofol and
nitrous oksida, and using of TCI model seem to be helpful to modulate the development
of the AOT and OIH. However, there are not sufficient data to support the evidence of
modulatory effect of them. Finally, we also cannot find any strong consistent evidence to
support the need to reduce the dose of remifentanil or apply the modalities for preventing
the AOT and the OIH. Consequently, further studies will need to investigate whether
remifentanil induce the AOT and the OIH after general anesthesia using propofol, nitrous
oksida, or TCI mode through high quality prospective trials. And the development of the
AOT and the OIH should be evaluated with direct measures such as QST. It is also
important to investigate if remifentanil-induced hyperalgesia may contribute to the
development of chronic pain, and if this contribution can be attenuated or even reversed
through pharmacologic modulation.

40. (Intraoperative use of remifentanil and opioid

induced hyperalgesia/acute opioid tolerance:
systematic review)
Background
Cardiac arrest outside the hospital is common and has a poor outcome. Studies in laboratory
animals suggest that hypothermia induced shortly after the restoration of spontaneous circulation
may improve neurologic outcome, but there have been no conclusive studies in humans. In a
randomized, controlled trial, we compared the effects of moderate hypothermia and
normothermia in patients who remained unconscious after resuscitation from out-of-hospital
cardiac arrest.

Full Text of Background ...

Methods
The study subjects were 77 patients who were randomly assigned to treatment with hypothermia
(with the core body temperature reduced to 33C within 2 hours after the return of spontaneous
circulation and maintained at that temperature for 12 hours) or normothermia. The primary
outcome measure was survival to hospital discharge with sufficiently good neurologic function
to be discharged to home or to a rehabilitation facility.

Full Text of Methods ...

Results
The demographic characteristics of the patients were similar in the hypothermia and
normothermia groups. Twenty-one of the 43 patients treated with hypothermia (49 percent)
survived and had a good outcome that is, they were discharged home or to a rehabilitation
facility as compared with 9 of the 34 treated with normothermia (26 percent, P=0.046). After
adjustment for base-line differences in age and time from collapse to the return of spontaneous
circulation, the odds ratio for a good outcome with hypothermia as compared with normothermia
was 5.25 (95 percent confidence interval, 1.47 to 18.76; P=0.011). Hypothermia was associated
with a lower cardiac index, higher systemic vascular resistance, and hyperglycemia. There was
no difference in the frequency of adverse events.

Full Text of Results ...

Conclusions
Our preliminary observations suggest that treatment with moderate hypothermia appears to
improve outcomes in patients with coma after resuscitation from out-of-hospital cardiac arrest.

Full Text of Conclusions ...

Read the Full Article...

(Treatment of Comatose Survivors of Out-of-Hospital

Cardiac Arrest with Induced Hypothermia)
OBJECTIVE:
To conduct a pilot trial of mild intraoperative hypothermia during cerebral aneurysm surgery.

METHODS:
One hundred fourteen patients undergoing cerebral aneurysm clipping with (n = 52) (World
Federation of Neurological Surgeons score < or =III) and without (n = 62) acute aneurysmal
subarachnoid hemorrhage (SAH) were randomized to normothermic (target esophageal
temperature at clip application of 36.5 degrees C) and hypothermic (target temperature of 33.5
degrees C) groups. Neurological status was prospectively evaluated before surgery, 24 and 72
hours postoperatively (National Institutes of Health Stroke Scale), and 3 to 6 months after
surgery (Glasgow Outcome Scale). Secondary outcomes included postoperative critical care
requirements, respiratory and cardiovascular complications, duration of hospitalization, and
discharge disposition.

RESULTS:
Seven hypothermic patients (12%) could not be cooled to within 1 degrees C of target
temperature; three of the seven were obese. Patients randomized to the hypothermic group more
frequently required intubation and rewarming for the first 2 hours after surgery. Although not
achieving statistical significance, patients with SAH randomized to the hypothermic group, when
compared with patients in the normothermic group, had the following: 1) a lower frequency of
neurological deterioration at 24 and 72 hours after surgery (21 versus 37-41%), 2) a greater
frequency of discharge to home (75 versus 57%), and 3) a greater incidence of good long-term
outcomes (71 versus 57%). For patients without acute SAH, there were no outcome differences
between the temperature groups. There was no suggestion that hypothermia was associated with
excess morbidity or mortality.

CONCLUSION:
Mild hypothermia during cerebral aneurysm surgery is feasible in nonobese patients and is well
tolerated. Our results indicate that a multicenter trial enrolling 300 to 900 patients with acute
aneurysmal SAH will be required to demonstrate a statistically significant benefit with mild
intraoperative hypothermia.

(Mild hypothermia as a protective therapy during

intracranial aneurysm surgery: a randomized prospective
pilot trial)
BACKGROUND:
The aim of this study was to compare the cerebral protective effects of two known protective
anesthetics, isoflurane and propofol, when these were used in combination with moderate
hypothermia (33-34 degrees C) after diffuse traumatic brain injury (TBI) in the rat. We assessed
cerebral protection by measuring local cerebral blood flow (LCBF), mean arterial blood pressure
(MABP), cerebral perfusion pressure (CPP) and intracranial pressure (ICP).

METHODS:
Sixteen female Wistar rats weighing 275 to 350 g were anesthetized and subjected to an
accelerated-impact weight-drop model of diffuse TBI. Hypothermia (33-34 degrees C) was
induced 45 minutes after TBI (baseline), and was maintained for 180 minutes. The isoflurane
group (n = 8) received 70% N(2)O in O(2), and isoflurane at 0.9 +/- 0.04%. The propofol group
(n = 8) received 70% N(2)O in O(2) and a propofol infusion (12 mg/kg/hr). LCBF was measured
by laser Doppler flowmeter. MABP, ICP, and brain and rectal temperatures were measured every
15 minutes from baseline through 180 minutes. Blood gas and hematocrit testing was also done
at baseline and every 60 minutes thereafter to assess the animals' physiological state.

RESULTS:
In the isoflurane group, MABP and CPP decreased significantly from baseline to 180 minutes (p
< 0.05 and p < 0.01, respectively), and MABP was significantly lower than the pressure in the
propofol group from 45 minutes through 180 minutes (p < 0.05, p < 0.01). ICP and LCBF
remained unchanged in this group. In the propofol group, from baseline to 180 minutes, CPP
increased to maximum 120 +/- 8 mmHg at 75 minutes from 98 +/- 5 mmHg (p < 0.05) and ICP
fell from 18 +/- 2 mmHg to 7 +/- 1 mmHg (p < 0.01); and the latter was significantly lower than
ICP in the isoflurane group (p < 0.05, p < 0.01, p < 0.001). LCBF in this group was significantly
higher than LCBF in the isoflurane group in the last 30 minutes of the experiment (p < 0.05). The
propofol group showed no change in MABP over the course of the experiment.

CONCLUSION:
In the clinical setting, propofol anesthesia may be better for use in combination with
hypothermia in cases of traumatic brain injury, as it reduces ICP and increases CPP under these
conditions.

(Propofol versus isoflurane anesthesia under hypothermic

conditions: effects on intracranial pressure and local
cerebral blood flow after diffuse traumatic brain injury in
the rat.)

BACKGROUND:
Simultaneous recordings of arterial blood pressure (ABP) and middle cerebral artery blood
velocity can be used to calculate the apparent zero flow pressure (aZFP). The inverse of the slope
of the pressure-velocity relationship is known as resistance area product (RAP) and is an index
of cerebrovascular resistance. There is little information available regarding the effects of
vasoactive drugs, arterial carbon dioksida (Paco(2)), and impaired cerebral autoregulation on
aZFP and RAP during general anesthesia. During isoflurane anesthesia, we investigated the
effects of hypocapnia and the effects of a phenylephrine infusion, on aZFP and RAP.

METHODS:
Radial ABP and transcranial Doppler middle cerebral artery blood velocity signals were recorded
in 11 adults undergoing isoflurane anesthesia. A phenylephrine infusion was used to increase
ABP and ventilation was adjusted to control Paco(2). Cerebral hemodynamic variables were
compared at two levels of mean ABP (approximately 80 and 100 mm Hg) and at two levels of
Paco(2): normocapnia (Paco(2) 38-43 mm Hg) and hypocapnia (Paco(2) 27-34 mm Hg). Two
aZFP analysis methods were compared: one based on linear regression and one based on Fourier
analysis of the waveforms.

RESULTS:
At the lower ABP, aZFP was 23 +/- 11 mm Hg and 30 +/- 13 mm Hg (mean +/- sd) with
normocapnia and hypocapnia, respectively (P < 0.001) and RAP was 0.76 +/- 0.97 mm Hg x s x
cm(-1) and 1.16 +/- 0.16 mm Hg x s x cm(-1) with normocapnia and hypocapnia, respectively (P
< 0.001). Similar effects of hypocapnia were seen at the higher ABP. With normocapnia,
isoflurane impaired cerebral autoregulation and aZFP did not change with the increase in ABP.
With hypocapnia, cerebral autoregulation was not significantly impaired and increasing ABP was
associated with increased aZFP (from 30 +/- 13 to 35 +/- 13 mm Hg, P < 0.01) and increased
RAP (from 1.16 +/- 0.16 to 1.52 +/- 0.20 mm Hg x s x cm(-1), P < 0.001). Calculation of the
relative contributions of aZFP and RAP to the cerebral hemodynamic responses indicated that
changes in RAP appeared to have a greater influence than changes in aZFP. The mean difference
between the two methods of determining aZFP (Fourier-regression) was 0.5 +/- 3.6 mm Hg
(mean +/- 2sd).

CONCLUSIONS:
During isoflurane anesthesia, two interventions that increase cerebral arteriolar tone, hypocapnia
and the autoregulatory response to increasing ABP, were associated with increased RAP and
increased aZFP. The effect of changes in RAP appeared to be quantitatively greater than the
effects of changes in aZFP. These results imply that arteriolar tone influences cerebral blood flow
by controlling both resistance and effective downstream pressure.

(The effects of hypocapnia and the cerebral autoregulatory

response on cerebrovascular resistance and apparent zero
flow pressure during isoflurane anesthesia.)

Abstract
Objective: To investigate effect of equal volumes (250 ml) of 7.2% hypertonic saline - 6%
hydroxyethyl starch (HS-HES) and 20% mannitol (M) on dural tension, serum osmolality and
hemodynamics in patients during elective neurosurgical procedures. Material and methods: Forty
ASA I-II patients scheduled for elective neurosurgical supratentorial procedures were randomly
assigned to two groups. About 30 min before skull opening, patients received either HS-HES or
M at infusion rate 750 ml/h. Dural tension score was used to evaluate the dural tension by
neurosurgeons. Serum osmolality was tested at following time points: before, 125 ml infused,
250 ml infused, 30 min and 60 min after infusion. Hemodynamic variables were measured by
FloTrac. Results: Patients who received HS-HES had a significant decrease in dural tension
scores (P < 0.05) and obtained more satisfactory brain relaxation for neurosurgeon (95% vs.
75%). In HS-HES group, the peak of serum osmolality occurred earlier and hyperosmolality
lasted for longer time. Transient decrease in mean arterial pressure was observed in M group at
10 min after the start infusion (P < 0.01). Heart rate significantly decreased after HS-HES
infusion, whereas no significant changes were observed in M group. In HS-HES group, stroke
volume variation significantly decreased from 9.7 3.5 at the initiation of infusion to 6.7 2.4 at
30 min after the infusion and remained decreased more than 60 min while it decreased from 6.8
3.1 to 5.3 1.5 in M group. Moreover, urine output in HS-HES group from initiation to 60 min
after the infusion was significantly less than those in M group.Conclusion: HS-HES might be an
alternative to mannitol in treatment of intracranial hypertension.
Keywords: Hypertonic saline, hydroxyethyl starch, mannitol, dural tension,
neurosurgery anesthesia

Go to:

Go to:

Introduction
Therapeutic efforts to decrease the brain size and to avoid injuring the dura mater and brain
tissue during intracranial procedures include cerebral vasoconstriction by means of anesthetic
intravenous agents and/or hypocapnia, volume reduction of cerebrospinal fluid (CSF), and brain
dehydration with osmotic agents [1]. The effect of hyperosmolar solutions on brain tissue was
first introduced by Weed and McKibben nearly 90 years ago [2]. Gradually, mannitol becomes
the most widely used agent for treating intracranial hypertension but can result in systemic
hypotension [3]. Currently, there is evidence that hypertonic saline solution may produce similar
effects based on various experimental and clinical studies [4-7]. Small volume resuscitation with
hypertonic saline showed obvious advantages including volume expansion and cerebral and
systemic hemodynamics improvement especially in patients undergoing traumatic brain injury
(TBI) [8-12]. Moreover, few clinical studies addressed the possible role of hypertonic saline for
patients undergoing elective neurosurgical procedures.

This prospective randomized clinical study was designed to compare the efficacy and safety of
intravenous administration of hypertonic saline 7.2% - hydroxyethyl starch 6% (HS-HES) and
20% mannitol (M) at equal volume (250 ml) for controlling intracranial pressure in patients
undergoing elective neurosurgical procedures. Our pervious study had evaluated the blood
coagulation and electrolytes changes after HS-HES infusion [13]. This study focuses on the
effects of both solutions on dural tension, hemodynamics and serum osmolality.

Go to:

Go to:

Material and methods

Study population
After informed consent was obtained, consecutive ASA I-II patients scheduled for supratentorial
elective procedures (resection of glioma, meningiomas or other brain tumors) were enrolled in
this prospective randomized study, which was approved by the ethics committee of Beijing
Sanbo Brain Hospital, Capital Medical University. The patients were randomized to receive
either 7.2% hypertonic saline - 6% hydroxyethyl starch (HperHAES, Fresenius Kabi
Deutschland GmbH, Bad Homburg, Germany) or 20% mannitol [Baxter, Baxter Healthcare
(Shanghai) Co., LTD, China]. The exclusion criteria included age < 18 years or > 65 years,
clinical signs of significantly increased ICP such as severe headache, blurred vision and/or
papilledema, history of cardiac, pulmonary and renal dysfunction, electrolyte disturbances,
preoperative coagulation disorders. The protocol has been registered at Chictr.org (ChiCTR-
TRC-12002357).

Anesthetic management
Patients were premedicated with midazolam (0.05 mg/kg, iv) and penehyclidine hydrochloride
injection (1 mg, im) at 15 min before anesthesia. Radial arterial catheters were inserted under
lidocaine local anesthesia and connected with FloTrac sensor (Edward Lifescience, USA) to
monitor hemodynamic variables. General anesthesia was induced with fentanyl 3 g/kg, propofol
2 mg/kg and vecuronium bromide (0.1 mg/kg). After endotracheal intubation, ventilation was
controlled by intermittent positive pressure ventilation to maintain the end-tidal CO2 pressure at
30-35 mmHg. Anesthesia was maintained with sevoflurane (1-1.3 minimum alveolar
concentration, MAC) combined with intravenous bolus of fentanyl and vecuronium
intermittently.

Intraoperative fluid management

About 30 min before the skull opening, patients were assigned by random number to receive 250
ml of hypertonic saline 7.2% - hydroxyethl starch 6% solution (Group HS-HES, 750 ml/h) or
20% mannitol (Group M, 750 ml/h). Maintenance fluid consisting of Ringers solution was
supplied at the rate 5-10 ml/kg/h during the study period, to maintain stable hemodynamics and
the mean of arterial pressure (MAP) 65 mmHg. Packed red blood cells (PRBC) were
transfused if necessary.

Measurements
Clinical measurements included age, sex, height, weight, brain lesion and urine output. Dural
tension score was estimated by the same team of neurosurgeons immediately after the skull
opening who was blinded to the groups. The reliability and validity of dural tension scores were
tested in our previous study [13]. Grade I, Normal dural tension, it was easy for the neurosurgeon
to open the dura mater. Grade II, Increased dural tension, the dura mater could be opened without
additional procedures to lower the ICP. Grade III, Pronouncedly increased dural tension, it was
necessary to apply additional procedures of lowering the ICP such as hyperventilation in order to
open the dura mater. We considered Grade I and II as the satisfactory level of brain relaxation for
neurosurgeons. Serum osmolality was measured by freezing point depression. Serum osmolality
was tested at the following time points: before (baseline), 125 ml infused, 250 ml infused, 30
min and 60 min after infusion. Hemodynamic measurements included mean arterial pressure
(MAP), heart rate (HR), cardiac index (CI) and stroke volume variation (SVV) which were
monitored continuously by the monitor (GE-Ohmeda S/5 and FloTrac, USA). These parameters
were documented at the 5 time points.
Statistical analysis
Statistical analysis was performed with SPSS 17.0 software. All data were presented as the mean
SD. Categorical data were compared by 2 test or Fishers exact test (types of brain tumors and
dural tension scores). Differences within groups were evaluated using paired student t test and
clinical variables differences between two groups were evaluated using unpaired student t test
(age, height, weight, operation duration, fluid input, urine output, blood loss and fluid balance).
For variables with multiple measurements (hemodynamic variables), a repeated measures
analysis of variance was used to evaluated the effects of time and group assignment, while two-
way analysis of variance was used to compare the difference within the groups. P value of < 0.05
was considered statistically significant.

Go to:

Go to:

Results
A total of 40 patients were included in the study, 20 in each group. Demographic data are
summarized in Table 1. There were no significant differences between two groups with respect to
sex, age, height, weight and brain lesion.

Table 1

Demographic data of patients treated with M (n = 20) or HS-HES (n = 20)

Dural tension in HS-HES group decreased significantly compared with M infusion (P < 0.05,
Table 2) and neurosurgeons were satisfactory with brain relaxation in 19 of 20 HS-HES treated
and in 15 of 20 mannitol treated patients. Diuresis was different between the two groups. Urine
output from the start of infusion to 30 min after the infusion was 156 202 ml in HS-HES group
and 489 306 ml in M group. And urine output in HS-HES group from the start of infusion to 60
min after the infusion was significantly less than that in M group (286 283 ml vs. 804 478
ml, P < 0.01).

Table 2
Clinical parameters of patients treated with M (n = 20) or HS-HES (n = 20)

In both groups, serum osmolality displayed a trend of rising from baseline level and then
decreased after the administration. The maximal serum osmolality appeared at 10 min after the
start of HS-HES administration, whereas occurred later in M group, at 20 min after the start of
mannitol administration. And the peak was significantly higher in HS-HES group (317 4.8 vs
308 5.5 mosm/kg, P < 0.01). Serum osmolality in HS-HES treated patients returned to control
level at 60 min after the end of infusion (304.7 3.6 mosm/kg), and 95% confidence interval
(CI) was 303.0-306.3 mosm/kg, whereas it returned earlier in M group, at 30min after the end of
mannitol infusion (302.5 5.4, 95% CI, 299.9-305.0 mosm/kg). It is obviously observed that
hyperosmolality of HS-HES lasted longer time compared with mannitol and showed a higher
increase in serum osmolality with HS-HES administration (Table 3).

Table 3

Changes of serum osmolality in two groups

In term of hemodynamics, transient decrease of MAP was observed in M group at 10 min after
the start of infusion (P < 0.01), while there was a slightly increase in HS-HES, but no statistical
significance. In the HS-HES group, heart rate significantly decreased from 71.9 8.8 bpm at
initiation of infusion to 61.6 8.8 bpm 30 min after terminating infusion, whereas it showed no
clinically relevant changes in M group. Cardiac index was slightly increased during the
administration of both solutions, but the difference was not statistically significant. In HS-HES
group, SVV significantly decreased from 9.7 3.5 at initiation of infusion to 7.1 3.0 at the end
of infusion and remained significantly decreased to 6.7 2.4 at 30 min after infusion. In M
group, SVV decreased from 6.8 3.1 at initiation of infusion to 5.3 1.5 during infusion, but
remained unchanged later (Table 4).

Table 4

Changes of HR, MAP, CI and SVV in two groups

Go to:

Go to:
Discussion
In the present study, we compared the effects of HS-HES and M in terms of dural tension, urine
output, hemodynamic and serum osmolality in patients undergoing the elective neurosurgical
procedures. The main findings were as follows: (1) 250 ml HS-HES could remarkably reduce the
dural tension scores and provide more satisfactory brain relaxation than mannitol for operation.
(2) Compared with M-treated patients, the peak of serum osmolality occurred earlier and
hyperosmolality lasted longer time in HS-HES group. (3) Patients in HS-HES group displayed an
upward trend of MAP and a decrease of SVV after the administration, whereas patients suffered
from transient decrease in MAP after mannitol infusion. These presented that HS-HES could
augment intravascular volume and improve hemodynamic states. (4) Though all hyperosmolar
agents cause diuresis, the diuretic effect of HS-HES was weaker in comparison with mannitol. It
might result from the stimulation of natriuretic peptide (ANP) release and not a direct osmotic
diuresis, which might assist in avoiding hypovolemia and hypotension. (5) In our previous and
present study, patients with HS-HES infusion did not show any other relative hyperosmolality
complication except for hypernatremia and hyperchloraemia. Plasma sodium decreased into the
normal range for 1 h and plasma chlorine normalized within 24 h. These findings provided the
substantial evidence that the HS-HES solution was effective and safe to decrease dural tension
and improve cardiovascular performance with small volume administration during the
neurosurgery.

In our study, we used dural tension scores as an alternative parameter for indirectly reflecting
ICP. We found HS-HES could remarkably reduce the dural tension scores and provide more
satisfactory brain relaxation for neurosurgeons. In the clinical practice, the ICP is not routinely
measured during elective neurosurgical procedure. Dural tension scores have shown strongly
positive correlation between the degree of cerebral edema and ICP [14,15]. In clinical,
neurosurgeons evaluated tension of dura mater based on their experiences before opening the
dura mater. If the tension of dura mater was high enough, brain tissue might protrude through the
craniotomy site, which increased the risk of cerebral ischemia with possible worsening outcome.
In the present study, we deduce the improvement in the dural tension scores in the HS-HES
group may arise from the effect of HS, which showed osmotic changes from HS-HES infusion
were significantly higher than mannitol. Infusion of HS creates an osmotic force that draws the
fluid back into the interstitial and intravascular area from the intracellular area due to the
impermeability of the blood-brain barrier (BBB) to sodium [6,9]. In addition, HS also decrease
the formation and/or increase the resistance to reabsorption of the CSF [16-18]. In an animal
model, Toung et al. [19] examined the effect of HS on cerebral edema secondary to tumor
effects. They found HS was more effective than mannitol in reducing both ipsilateral and
contralateral hemispheric water content as measured by wet-to-dry weight radios [19]. In clinical
studies, HS has also been shown to reduce ICP in different intracranial diseases, particular in
head trauma with increased ICP [7,20,21]. Additionally, researches have demonstrated that HS
remains effective in intracranial hypertension refractory to treatment with mannitol [22,23].
Moreover, our data showed that the peak osmolality with HS-HES administration occurred
earlier than mannitol administration (10 min vs. 20 min after the start of infusion). And serum
osmolality in HS-HES treated patients returned to normal range later than M-treated patients (60
min vs. 30 min after the end of infusion). These findings both revealed that osmotic effect of HS-
HES would last longer time. It might arise from HS-HES combination with colloids
(hydroxyethyl starch). Previous studies that compared changes in ICP with the use of HS or
mannitol in TBI patients found that both significantly reduced ICP and that HS had a longer
duration of action than mannitol [24-26].

Our hemodynamic data indicated that patients suffered from transient hypotension during the
infusion of mannitol, while patients with HS-HES administration displayed a upward trend of
MAP. In HS-HES group, SVV significantly decreased from 9.7 3.5 at initiation of infusion to
6.7 2.4 at 30 min after terminating infusion and remained significantly decreasing more than
60 min after the end of infusion. We deduced that these changes arised from expanding
intravascular volume of adiministration of HS-HES and improving cardiovascular performance.
Multiple mechanisms could explain this volume-expanding property, including compartment
redistribution with fluid shift to the vascular bed, positive effects in cardiac output [27,28], and
immunologic effects [29,30]. Relative studies also indicated that an infusion of 7.5% hypertonic
saline dextran could increase the intravascular volume by as much as four times the infused
volume within minutes of infusion, and after osmotic equilibrium, about 4 h of a bolus dose, the
net effect is to increase the plasma volume by 750 ml for every liter administered [31]. Besides,
HS-HES was able to exert these plasma expanding effects with administration of small volumes
(250 ml). Our previous study demonstrated that 250 ml HS-HES could reduce the volume of
intraoperative fluid [9] and other results indicated that HS could decrease postoperative fluid
input. A near zero fluid balance was observed in patients with hypertonic saline-dextran (HSD)
infusion during the first 48 hours after cardiac surgery [32] and HSD patients ran a slightly
negative fluid balance over the first 72 h following cardiopulmonary bypass [33], while control
patients ran a large positive fluid balance. Those data showed that effect of HS-HES on the
perioperative fluid treatment might not be neglected. Though all hyperosmolar agents could
cause diuresis, we found the diuretic effect of hypertonic saline solutions differed from that of
mannitol by monitoring urine output from the start of infusion to 30 and 60 min after the end of
infusion. We analyzed that it might result from the stimulation of atrial natriuretic peptide (ANP)
release and not in a direct osmotic diuresis, which might assist in avoiding hypovolemia and
hypotension [34].

Additionally, other important issues deserved attention with HS including changes of blood
coagulation function, electrolyte and serum osmolality after HS-HES infusion. These blood
variables were measured in our previous study [13] including plasma electrolyte concentration
(sodium, potassium, chloride, calcium), hemoglobin concentration (Hb), platelet (Plt), hematocrit
(Hct), and coagulation parameters [prothrombin time (PT), activated partial thromboplastin time
(APTT) and fibrinogen (Fbg)]. And previous results showed no or slight impairment on blood
coagulation function and no significant increase of blood loss. Electrolyte abnormalities such as
hypernatremia and hyperchloraemia occurred immediately after HS-HES infusion, then plasma
sodium returned normal level about 1 h after terminating infusion and plasma chloride decreased
to normal range within 24 h after HS-HES infusion. The plasma potassium level took on biphasic
changes with increasing at the beginning and then decreasing later, but did not exceed normal
range. At present guidelines recommend maintaining serum osmolality < 320 mosm/kg when
using mannitol and suggest that the maximal serum osmolality is 360 mosm/kg when using
hypertonic saline solutions [34-36]. In our study, the maximum of serum osmolality was 318
mosm/kg in M group and 330 mosm/kg in HS-HES group, within the recommended range. And
there were no potential side effects such as cardiac failure, phlebitis, central pontine myelinolysis
found in the whole study period. So we considered it was relatively safe to administer 250 ml
bolus dose of 7.2% hypertonic saline - 6% hydroxyethyl starch within 20 min.

In conclusion, 7.2% hypertonic saline - 6% hydroxyethyl starch infused before skull opening is
an attractive choice in reducing dural tension during neurosurgical procedures. HS-HES could
provide more satisfactory brain relaxation for neurosurgeon and this effect may last longer time
than mannitol. HS-HES received a more stable hemodynamic state with its volume-expanding
property and weaker-diuresis, which avoided transient hypotension of mannitol infusion. Our
results indicate HS-HES may represent a new avenue for osmotherapy the during the
neurosurgical procedures.

(Effects of hypertonic saline - hydroxyethyl starch and

mannitol on serum osmolality, dural tension and
hemodynamics in patients undergoing elective
neurosurgical procedure)
Glycemic control among critically-ill patients has been a topic of considerable attention for the
past 15 years. An initial focus on the potentially deleterious effects of hyperglycemia led to a
series of investigations regarding intensive insulin therapy strategies that targeted tight glycemic
control. As knowledge accumulated, the pursuit of tight glycemic control among critically-ill
patients came to be seen as counterproductive, and moderate glycemic control came to dominate
as the standard practice in intensive care units. In recent years, there has been increased focus on
the importance of hypoglycemic episodes, glycemic variability, and premorbid diabetic status as
factors that contribute to outcomes among critically-ill patients. This review provides a survey of
key studies on glucose control in critical care, and aims to deliver perspective regarding
glycemic management among critically-ill patients.

Keywords: Glycemic control, Critical care, Blood sugar in intensive care unit,
Diabetes in intensive care unit, Glycemic control

Core tip: Glucose control among critically-ill patients has been an area of active research and
considerable controversy in the past 15 years. This review provides a practical guide to the
evidence, with a survey of the key studies that have informed current perspectives and clinical
guidelines related to glycemic management among the critically ill. The article shows why initial
enthusiasm for tight glycemic control waned as evidence accumulated favoring more modest
glucose goals. The article also summarizes recent work investigating the importance of
hypoglycemic episodes, glycemic variability, and premorbid diabetic status on morbidity and
mortality in the intensive care unit.

Go to:

INTRODUCTION
In 2001, van den Berghe et al[1] reported results from a single-center, prospective, randomized
controlled trial in Leuven, Belgium, and changed the way that blood glucose was managed in
intensive care units (ICUs) throughout the world. Prior to the publication of this first Leuven
study, glycemic control among critically-ill patients received scant attention, either at the bedside
or in academic journals. The overwhelmingly favorable results of the study - which, among
critically-ill surgical patients, found a remarkable mortality benefit from the use of intensive
insulin therapy targeting normoglycemia - sparked strong interest in glycemic management in the
ICU. Intensive insulin therapy quickly became the standard of care in both medical and surgical
ICUs. However, as has been the experience in many facets of critical care, promising initial
single-center results were not duplicated in subsequent trials. The publication of the NICE-
SUGAR trial in 2009, which reported that intensive insulin therapy may actually result in
increased mortality among critically-ill patients, served as a major bookend to the era of tight
glycemic control as a pillar of ICU management[2].

Nonetheless, interest in defining optimal glycemic control among critically-ill patients has
continued. In the years that have followed the publication of the NICE-SUGAR trial,
investigations have focused on establishing which factors of glycemic control and dysregulation
most affect patient outcomes in the ICU. It has been increasingly recognized that hypoglycemia,
glycemic variability, and premorbid diabetic status are all important considerations to be taken
into account when approaching the glycemic management of a critically-ill patient.

This review aims to provide a survey of the key studies that have informed the changes in
thinking in the past 15 years as regards glucose control in critical care. It explores the basis of the
initial enthusiasm for, and subsequent skepticism of, intensive insulin therapy in the ICU. It also
aims to provide perspective regarding major issues of glycemic management among critically-ill
patients: hyperglycemia, hypoglycemia, glycemic variability, and premorbid diabetic status.

Go to:

HYPERGLYCEMIA
Elevated blood sugar levels are commonly seen among critically ill patients, including those
without a known history of diabetes. There are many reasons why patients undergoing treatment
for critical illness develop hyperglycemia, and these reasons include both effects of endogenous
stress responses and byproducts of medical interventions. Inflammatory cytokines and stress
hormones, including cortisol and epinephrine, serve to inhibit insulin release and promote
insulin resistance, thereby naturally increasing blood glucose levels by stimulating
gluconeogenesis and glycogenolysis while impeding glucose uptake by peripheral tissues[3,4].
Many medical therapies further promote hyperglycemia, including the administration of
exogenous catecholamines and corticosteroids, the infusion of dextrose for the purpose of
suspending intravenous medications or providing parenteral nutrition, and even bedrest, which
in and of itself may serve to impair glucose uptake in skeletal muscles[5,6].

Prior to the publication of the first Leuven trial[1], many practitioners viewed moderately severe
hyperglycemia among critically ill patients to be either an epi-phenomenon or an adaptive
response, not warranting significant concern or intervention[7]. However, as observational
studies accumulated linking hyperglycemia to negative in-hospital patient outcomes, this
permissive attitude began to change[8-11]. Hyperglycemia was coming to be seen as
complication worthy of physician attention. For example, a retrospective study of 1826 patients
admitted to a mixed ICU in Stamford, Connecticut serving medical, surgical, and coronary
patients reported reduced survival among those with elevated mean blood glucose levels, with a
stepwise effect resulting in higher mortality as mean blood glucose levels rose[8]. Compared to
patients who survived to hospital discharge, those who died had higher initial (175 mg/dL vs 151
mg/dL), mean (172 mg/dL vs 138 mg/dL), and maximum (258 mg/dL vs 177 mg/dL) blood
glucose levels. In-hospital mortality was 9.6% among those with a mean blood glucose of 80-99
mg/dL, 29.4% among those with a mean blood glucose of 180-199 mg/dL, and 42.5% among
those with a mean blood glucose greater than 300 mg/dL.

Observations such as these raised concern that acute hyperglycemia was itself contributing to
poor outcomes, potentially by leaving affected patients susceptible to at least some of the
consequences that have long been observed among chronic diabetics, including high infection
rates, poor wound healing, and polyneuropathy[1,5]. Laboratory studies have also raised
concerns about the possible deleterious effects of acute hyperglycemia, as hyperglycemia has
been shown to cause injury to a variety of cell types that exhibit insulin-independent glucose
uptake, including endothelial cells, hepatocytes, and renal tubular cells[12-16].

The repeated observation that hyperglycemia is associated with worse outcomes among
critically ill patients, together with the theoretical harms of acutely elevated blood glucose
levels, represents the basis for focusing on glycemic control in the intensive care setting.
However, the possibility remains that elevated blood glucose levels are actually beneficial to the
critically ill individual, and that stress hyperglycemia is an appropriate and adaptive response to
life-threatening illness, as no randomized trial investigating glycemic control has studied the
effect of truly permissive hyperglycemia[17]. Potential benefits of hyperglycemia in the critically
ill individual include promotion of glucose delivery in the face of ischemic insults (down an
enhanced glucose diffusion gradient), with insulin resistance favoring redistribution of available
glucose stores toward cells of the immune and nervous systems, and away from peripheral
tissues[17]. Recent observational studies have provided some support for this view, reasserting
the possibility that hyperglycemia is simply a marker of illness severity. For example, a recent
retrospective study of 7925 consecutive critically ill patients admitted to three mixed ICUs in
Australia showed that while hyperglycemia was associated with in-hospital mortality, once
lactate levels were considered, there was no independent association between hyperglycemia and
mortality[18]. This finding was consistent with a previous retrospective study, which found that
among a cohort of septic nondiabetic adult patients, hyperglycemia noted on initial presentation
did not increase mortality risk unless accompanied by concurrent hyperlactatemia[19]. Such
observations present a useful reminder that our understanding of the effects of hyperglycemia
remains incomplete.

Our ability to identify patients most likely to suffer harm from hyperglycemia also remains
incomplete. Several studies have concluded that the association between hyperglycemia and in-
hospital mortality is attenuated among those with pre-existing diabetes mellitus, with some even
failing to demonstrate any association at all[11,20-23].

Go to:

MAJOR INVESTIGATIONS OF GLYCEMIC CONTROL IN CRITICALLY

ILL PATIENTS
Concern about the potentially deleterious effects of hyperglycemia in critically ill patients has
motivated multiple randomized controlled trials investigating glycemic management in
ICUs[1,2,24-32]. This section serves to review the major trials regarding this subject, exploring
the evidence that underlay the initial enthusiasm for, and subsequent skepticism of, intensive
insulin therapy for glycemic normalization among critically ill patients. Key features of the trials
are summarized in Table Table11.
Table 1

Summary of major randomized controlled trials investigating the use of intensive vs

conventional insulin therapy in critically-ill patients

The original Leuven study, reported by van den Berghe et al[1] in 2001, was the first major
prospective trial to investigate the effects of tight glycemic control in critically ill patients. This
was a prospective, non-blinded, randomized controlled trial of 1548 mechanically ventilated
adult patients admitted to a single surgical ICU in Leuven, Belgium. A majority of the patients
(63%) had undergone cardiac surgery. Prior to admission, 13% of patients had been diagnosed
with diabetes mellitus, and 5% had been maintained on insulin therapy. Upon ICU admission,
patients were randomly assigned to receive either intensive or conventional insulin therapy.
For all patients, insulin was delivered via a continuous infusion, and glycemic monitoring was
performed via measurements of whole-blood glucose of arterial blood samples, collected every
one to four hours. For patients in the intensive insulin therapy group, insulin infusions were
started if measures of blood glucose exceeded 110 mg/dL, and the infusions were titrated to
maintain blood glucose in the range of 80 to 110 mg/dL. By contrast, for patients in the
conventional therapy group, insulin infusions were only started if measures of blood glucose
exceeded 215 mg/dL, and the infusions were titrated to maintain blood glucose in the range of
180 to 200 mg/dL. All patients received intravenous glucose for the first 24 h of ICU admission,
after which feeding continued via total parenteral, total enteral, or combined enteral and
parenteral nutrition. All patients reverted to conventional blood glucose management upon
discharge from the ICU. During their ICU stays, 98.7% of patients in the intensive insulin
therapy group required insulin infusions, and the targeted blood glucose level was achieved, with
a mean blood glucose of 103 mg/dL. Among patients in the conventional insulin therapy group,
only 39.2% required insulin infusions, and the mean blood glucose was 153 mg/dL. The results
of the study strongly favored the intensive insulin therapy group, with observed benefits in terms
of both morbidity and mortality. In-ICU mortality was 4.6% in the intensive insulin therapy
group compared to 8.0% in the conventional insulin therapy group (P < 0.04), and the survival
benefit persisted to hospital discharge, with an absolute risk reduction of in-hospital mortality of
3.7% (7.2% vs 10.9%; P = 0.01), largely due to a reduction in deaths attributed to sepsis.
Compared to patients in the conventional insulin therapy group, those receiving intensive insulin
therapy also experienced reduced rates of renal replacement therapy, prolonged mechanical
ventilation, and extended ICU stays. The overwhelmingly positive results from the first Leuven
study were in many ways practice-changing, and it informed investigations into glycemic
management of critically ill patients for the ensuing decade, and beyond.

The next major prospective trial came from the same group in Belgium, and was again a single-
ICU study[24]. In this second Leuven study, 1200 adult patients admitted to a medical ICU were
studied. The study included only patients who were unable to take oral nutrition upon ICU
admission, and who were anticipated to require at least 3 d of intensive care. Patients were
randomized to intensive vs conventional insulin therapy groups, with stratification according to
diagnostic categories. Thresholds for initiation of insulin therapy and target blood glucose levels
for the two groups were identical to what had been used in the first Leuven study[24]. In stark
contrast to the findings of the previous trial, the second Leuven study showed no overall
mortality benefit to intensive insulin therapy, as both ICU and in-hospital mortality rates were
similar among patients in the intensive and conventional insulin therapy groups. However, the
authors reported a statistical difference in in-hospital mortality among the subset of patients who
actually received at least 3 d of ICU care, as had been intended at the time of their inclusion in
the study. Among this subset of 767 patients who stayed in the ICU for at least 3 d (of whom 386
received intensive insulin therapy and 381 received conventional insulin therapy), in-hospital
mortality was 43.0% in the intensive therapy group, compared to 52.5% in the conventional
therapy group (P = 0.009). While an interesting finding, this subset analysis suffered from a lack
of real-world applicability (even the authors were unable to accurately predict which patients
would require extended ICU stays) and a loss of balanced diagnostic categorization (likely
biasing the results). While no mortality benefit to intensive insulin therapy was identified,
secondary analyses of patient morbidity found reduced rates of acquired kidney injury, reduced
durations of mechanical ventilation, and reduced lengths of ICU and hospital stay among patients
in the intensive insulin therapy group compared to those in the conventional insulin therapy
group.

The mortality benefits realized in the first Leuven study and the morbidity benefits realized in
the second sustained considerable enthusiasm for tight glycemic control in critically ill patients
for the next several years, with widespread adoption of intensive insulin protocols in medical and
surgical ICUs, despite occasional voices urging caution[33,34]. However, a series of studies
published in 2008 and 2009, culminating with the NICE-SUGAR trial, severely tempered this
enthusiasm[2,25-28]. The first of these trials, reported by Brunkhorst et al[26], involved patients
with severe sepsis or septic shock admitted to multidisciplinary ICUs in 18 academic tertiary
hospitals in Germany. This was a two-by-two factorial trial, and patients were randomized to
receive either intensive or conventional insulin therapy for glycemic control (with protocols
similar to those used in the two Leuven studies[1,24]) and either hydroxyethyl starch or modified
Ringers lactate for fluid resuscitation. The use of intensive insulin therapy was terminated after
the first safety analysis, due to a nearly six-fold increased frequency of hypoglycemia in the
intensive insulin group, including a high proportion of severe hypoglycemic events that were
classified as life-threatening and requiring prolonged hospitalization. Among the patients studied,
there was no documented benefit to intensive insulin therapy, as there were no statistical
differences in rates of mortality, rates of acute renal failure or renal replacement therapy, use of
vasopressor medications, number of ventilator-free days, or length of ICU stay.

Several subsequent studies conducted in a variety of settings similarly failed to demonstrate clear
benefits to tight glycemic control in critically ill patients, but consistently highlighted an
increased risk of hypoglycemia among patients treated with intensive insulin
protocols[2,25,27,28]. Arabi et al[25] reported a prospective trial wherein they randomized 523
medical, surgical, and trauma patients admitted to a single ICU in Riyadh, Saudi Arabia to
intensive or conventional insulin therapy, and found no between-group differences in mortality,
ICU or hospital lengths of stay, rates of renal replacement therapy, durations of mechanical
ventilation, or frequencies of infectious complications, but patients in the intensive insulin group
experienced much higher rates of hypoglycemia. Similar negative findings with respect to
measures of mortality and morbidity were reported by De La Rosa Gdel et al[27] in their study
of 504 medical, surgical, and trauma patients admitted to a single ICU in Medellin, Colombia
and randomized to intensive or conventional insulin therapy, though again, rates of
hypoglycemia were much higher in the intensive insulin group. A subsequent multinational trial,
involving patients admitted to 21 medico-surgical ICUs in 7 countries, also failed to identify
meaningful benefits to tight glycemic control[28]. This study, which again randomized patients
to intensive or conventional insulin therapy, was ultimately underpowered, as it was prematurely
stopped due to a high rate of unintended protocol violations. However, among the 1078 patients
studied, there were no between-group differences in mortality, and the only differences in
measures of morbidity were higher rates of hypoglycemia among patients in the intensive insulin
therapy group and a slight reduction in vasopressor/inotrope use in the conventional insulin
therapy group.

On the heels of these four consecutive negative studies[25-28], the landmark NICE-SUGAR trial
was reported, which remains the most comprehensive study of glycemic control strategies among
ICU patients performed to date[2]. The NICE-SUGAR study included 6104 medical and surgical
patients admitted to ICUs at 42 hospitals in Australia, New Zealand, Canada, and the United
States. All patients were anticipated to require at least 3 d of ICU care, were expected to be
unable to eat for at least 2 d, and had an arterial line in place as part of their routine ICU
management. As with previous studies, patients were randomized to intensive or conventional
insulin therapy groups, but the target blood glucose range of the conventional insulin therapy
group was lower than it had been in the Leuven studies[1,24], based on updated practice surveys.
In the intensive insulin therapy group, the target blood glucose range was 81 to 108 mg/dL, while
in the conventional insulin therapy group, the target blood glucose was 180 mg/dL or less, with
insulin administration reduced and then discontinued if blood glucose levels fell below 144
mg/dL. As had been the case in Leuven studies[1,24], blood glucose monitoring was performed
every one to four hours, and the use of arterial rather than capillary blood samples for this
purpose was encouraged. The majority of patients in both treatment groups received insulin
therapy (97.2% of those in the intensive insulin therapy group and 69.0% of those in the
conventional insulin therapy group). The mean time-weighted blood glucose level in the
intensive group was 115 mg/dL, while it was 144 mg/dL in the conventional group. The primary
study endpoint was 90-d all-cause mortality, which was 2.6% higher in the intensive than in the
conventional insulin therapy group (27.5% vs 24.9%, P = 0.02). Subgroup analyses suggested no
differences in treatment effects for comparisons of medical and surgical patients, patients with
and without preexisting diabetes, and patients with and without severe sepsis. With the exception
of rates of severe hypoglycemia, markers of morbidity did not differ according to treatment
groups, as there were similar between-group ICU and hospital lengths of stay, durations of
mechanical ventilation, frequencies and durations of renal replacement therapy, rates of new
organ failure, and occurrences of positive blood cultures. Severe hypoglycemia (defined as a
blood glucose level less than or equal to 40 mg/dL) occurred in 6.8% of the patients in the
intensive insulin therapy group vs 0.5% of those in the conventional therapy group (P < 0.001).

Following the overwhelmingly negative results of the NICE-SUGAR study, Annane et al[29]
reported on the use of intensive vs conventional insulin therapy in patients with septic shock
being treated with corticosteroids, hypothesizing that this subset of ICU patients may benefit
from intensive insulin therapy, even if a general ICU population does not. A total of 509 patients
treated in 11 ICUs in France were randomized to intensive or conventional insulin therapy,
according to the treatment protocols used in the first Leuven study[1]. Here again, there were no
between-group differences in measures of patient mortality or morbidity, with the exception of
an increased rate of severe hypoglycemia among patients in the intensive insulin therapy group.
Subsequently, randomized controlled trials investigating intensive insulin therapy among
mechanically ventilated neurologic patients, patients with severe traumatic brain injuries, and
critically ill pediatric patients have all failed to demonstrate a clinical benefit to tight glycemic
control[30-32].

In summary, following the publication of the two single-center Leuven studies[1,24], the
preponderance of evidence has strongly indicated that the use of intensive insulin treatment with
the goal of tight glycemic management in critically-ill patients at best provides no benefit over
moderate or lax glycemic control, and at worst results in markedly increased rates of severe
hypoglycemia and possibly even increased mortality[2,25-29].
Go to:

HYPOGLYCEMIA
As clinicians and investigators have grappled with the results of the NICE-SUGAR trial and of
other negative studies regarding the use of intensive insulin therapy in critically-ill patients[2,25-
32], several potential explanations have been proposed to account for the lack of demonstrable
benefit for tight glucose control. The proposed explanations have targeted either the rationale for
intensive insulin therapy (positing that hyperglycemia may be beneficial, or that exogenous
insulin may be harmful), or the execution of the strategy (suggesting that the labor-intensive
focus on tight glycemic control distracts from other considerations, or that the benefits of
normoglycemia have been obscured by an inability to avoid hypoglycemia)[4,35]. This final
consideration-that hypoglycemic complications negate the potential benefits of tight glycemic
control-has gained widespread acceptance, and has important implications for future study of
glycemic management among critically-ill patients. Hypoglycemia has been a commonly-
reported occurrence among the patients treated with intensive insulin therapy in major trials, and
severe hypoglycemia (defined as a blood glucose level less than 40 mg/dL) has occurred in up to
28% of these patients[4]. It was not initially clear whether the increased rate of hypoglycemia
experienced among patients treated with a tight glycemic control strategy was problematic. In the
first Leuven study, severe hypoglycemia was reported to have occurred 6.6-fold more commonly
among patients in the intensive insulin therapy group, but no clinically-significant outcomes
were associated with its occurrence in any of the patients, and the issue of hypoglycemia was not
addressed in the manuscripts discussion[1].

By the time the NICE-SUGAR trial was reported, the frequency of hypoglycemic episodes among
patients treated with intensive insulin regimens had become a significant concern. It was
recognized that hypoglycemia could theoretically be harmful to patients by means of a number of
different mechanisms, including irreversible neuronal damage, autonomic instability, cardiac
arrhythmia, and alteration of inflammatory responses[36,37]. The relationship between
hypoglycemia and mortality was examined in a post-hoc analysis of the NICE-SUGAR trial[37].
For the purpose of this analysis, severe hypoglycemia was defined as a recorded blood glucose
level of 40 mg/dL or less, while moderate hypoglycemia was defined as a recorded blood glucose
level in the range of 41 to 70 mg/dL. Among the 6026 patients analyzed, severe hypoglycemia
occurred in 3.7% of individuals, while moderate hypoglycemia occurred in an additional 45.0%.
Hypoglycemic episodes were much more common among those patients in the intensive insulin
therapy group, with this group accounting for 93.3% of severe hypoglycemia and 82.4% of
moderate hypoglycemia. The occurrence of hypoglycemia was strongly associated with an
increased risk of death, with moderate hypoglycemia associated with a 40% increase in adjusted
mortality risk, and severe hypoglycemia associated with a doubling of this risk. While these data
do not prove a causal relationship between hypoglycemia and mortality, they do support the
possibility that it was the increased frequency of iatrogenic hypoglycemic episodes that
accounted in some measure for the excess mortality observed among patients treated with
intensive insulin therapy in the NICE-SUGAR trial.

This possibility has been supported by other retrospective studies investigating the relationship
between hypoglycemic episodes and mortality among ICU patients. In a review of 4946 patients
admitted to two ICUs in Australia, Egi et al[38] found that 22.4% of patients experienced at least
one episode of hypoglycemia, defined as recorded blood glucose of less than 82 mg/dL. The
patients were analyzed in six bands, according to the level of their lowest recorded blood
glucose, and it was shown that the severity of hypoglycemia was independently associated with
in-hospital mortality. In a separate single-center review of 5365 consecutive patients admitted to
a mixed medical-surgical ICU, the occurrence of even one episode of severe hypoglycemia was
seen to be independently associated with mortality, both by case-control and by multivariable
logistic regression analyses[39].

To a significant extent, a desire to avoid inducing hypoglycemia has motivated the move away
from treating ICU patients with intensive insulin protocols[40]. It should be noted that the focus
on avoiding hypoglycemia leaves the door open to future reconsideration of the benefits of tight
glycemic control. If the problem with intensive insulin therapy is mainly an inability to avoid
hypoglycemic episodes, one can imagine that the development of better glucose monitoring
technologies and glycemic control algorithms (if they allow for severe reductions in the
incidence of hypoglycemia) could result in improved outcomes with a tight glycemic control
strategy. In recent years, the development of continuous glucose monitoring systems has received
significant attention along these lines, but the benefits of continuous glucose monitoring have not
yet been established[41-43].

Go to:

GLYCEMIC VARIABILITY
In recent years, it has increasingly been recognized that glycemic variability is a dimension of
significant importance among critically-ill patients, independent of the acute highs and lows of
blood glucose measurements in the ICU. The potential significance of glycemic variability
among ICU patients was first raised by Egi et al[44], in a retrospective observational study of
7049 patients who had been admitted to four hospitals in Australia. For the purposes of this
study, a patients glycemic variability was defined as the standard deviation of the arithmetical
mean of the entire set of glucose measurements during that individuals ICU stay. The authors
found that glycemic variability was an independent predictor of mortality, and that the glycemic
variability was actually a stronger predictor of ICU mortality than the mean glucose
concentration. A subsequent single-center retrospective observational study of 3252 ICU patients
in the United States confirmed and extended these findings, again demonstrating that this
measure of glycemic variability was a strong independent predictor of mortality, even after
excluding patients who had experienced severe hypoglycemia[45].
As glycemic variability has been further considered among ICU patients, definitions have
changed. Defining glycemic variability as the standard deviation of the mean of all blood glucose
measurements has fallen out of favor, as starkly different glycemic patterns can generate
identical mean glucose and standard deviations[46]. Multiple other measures of glycemic
variability have been described, including coefficient of variation, glycemic lability index, mean
absolute glucose change, and mean amplitude of glycemic excursion[47,48]. No gold standard
for measuring glycemic variability has been established, but multiple studies utilizing these more
complicated metrics have confirmed that glycemic variability is independently associated with
mortality among ICU patients[23,46,48,49].

Whether glycemic variability is a cause of poor patient outcomes or is simply a marker of severe
illness is not known. However, several lines of evidence have suggested that glycemic variability
causes oxidative stress, enhances cell apoptosis, and impairs endothelial function[45,46].
Therefore, it is plausible that glycemic variability causes harm to critically-ill patients, and that
optimal glycemic control in the ICU would aim to minimize glycemic variability. As with
avoiding hypoglycemia in the ICU, it is hoped that advances in glycemic monitoring and
corresponding glucose control algorithms will reduce the extent of glycemic variability, but at
least one early study has failed to show that existing means of continuous glucose monitoring
would reduce glycemic variability[47].

Go to:

PREMORBID DIABETIC STATUS

From the first Leuven study to the NICE-SUGAR trial, all of the major investigations of
intensive insulin therapy in critically-ill patients utilized glycemic-control protocols that did not
differentiate between diabetic and nondiabetic patients[1,2,24-28]. Similarly, recent guidelines
regarding the use of insulin infusions in the ICU do not advocate altering the approach to
glycemic management on the basis of patients premorbid diabetic status[40]. However, there is
growing evidence that diabetic and nondiabetic patients respond differently to dysglycemia
experienced in the ICU.

Krinsley et al[49] performed a retrospective observational study of 44964 patients admitted to 23

ICUs in 9 countries to determine how diabetic status affected the associations of hyperglycemia,
hypoglycemia, and glycemic variability with mortality. While hypoglycemia was associated with
an increased risk of mortality among all patients, the diabetic status modulated the impact of
both hyperglycemia and glycemic variability. In nondiabetic patients, maintenance of
euglycemia was independently associated with a reduced mortality risk, but among diabetic
patients, those with a mean glucose of 80 to 110 mg/dL actually had an increased risk of
mortality, even when compared only to those with a mean glucose greater than 179 mg/dL. The
significance of glycemic variability also seemed to differ according to diabetic status, as a high
level of glycemic variability (defined as a coefficient of variability greater than 20%) was
independently associated with an increased risk of mortality among nondiabetic patients, but not
among those with diabetes.

Similar findings were reported in a subsequent single center retrospective observational study
that analyzed glucose and outcome data from 10320 ICU patients[23]. Again, hypoglycemia was
associated with mortality in both diabetic and nondiabetic patients, but outcomes associated with
hyperglycemia and glycemic variability differed according to premorbid diabetic status. While
hyperglycemia was associated with increased mortality among the nondiabetic patients, no clear
pattern relating elevated mean glucose levels with mortality could be found among the diabetic
patients. In addition, glycemic variability (as measured by mean absolute glucose change) was
only associated with increased mortality among the nondiabetic patients.

Such differences among diabetic and nondiabetic patients have raised the possibility that future
glycemic control protocols for critically-ill patients will differ according to premorbid diabetic
status, or other markers of insulin resistance, such as metabolic syndrome or non-alcoholic fatty
liver disease[50,51]. However, further studies are needed to better define optimal glycemic
management among diabetic patients in the ICU.

Go to:

CONCLUSION
In the past two decades, glycemic management among critically-ill patients has been a topic of
extensive study, leading to significant changes in clinical practice. Intensive insulin therapy was
widely adopted following the publication of the first Leuven study[1], only to be largely
abandoned as further knowledge accumulated questioning the benefits of this approach,
ultimately culminating with the NICE-SUGAR trial, which found an increased risk of mortality
among patients treated with tight, as compared to moderate, glucose control strategies[2].
Current guidelines regarding glycemic management of critically-ill patients advocate initiating
insulin infusions for blood glucose measurements in excess of 150 mg/dL, with the goal of
maintaining blood glucose less than 180 mg/dL[40]. While targeting a blood glucose level less
than 180 mg/dL is now widespread (and consistent with the control group in NICE-SUGAR), it
should be noted that evidence supporting this goal, as opposed to an even more permissive
glycemic control strategy, is lacking.

In recent years, there has been an increased focus on the potential deleterious effects of glycemic
variability, though it remains unclear how best to avoid fluctuations in blood glucose levels. In
addition, there has been increasing attention given to differences among the glycemic control
needs of diabetic and nondiabetic patients.

In coming years, we expect that new glucose monitoring systems will emerge, and that new
strategies for maintaining euglycemia (while avoiding hypoglycemic episodes and glycemic
variability) will follow. Glycemic management among critically-ill patients remains an area of
unsettled medicine.

Introduction: Only few studies have investigated pain, nausea, sedation and analgesic strategies
in post-craniotomy patients. The aim of this observational study was to explore pain, nausea,
sedation and analgesic procedures after craniotomy, and to evaluate the quality of current
analgesic therapy administered to post-craniotomy patients.

Material and methods: A total of 59 patients undergoing supratentorial or infratentorial

craniotomy were included over a three-month period. The intensity of pain, nausea and sedation
was evaluated at one, two, four, eight and 24 h after extubation. Post-operative analgesic
consumption at 0-48 h after extubation was noted. Post-operative morphine consumption in
relation to gender, surgical procedure, administration of preoperative steroids and application of
surgical drains was evaluated.

Results: Fifty patients completed the study. After the first post-operative hour, 56% suffered from
moderate-to-severe pain, which decreased to 38% at 24 h post-operatively. Patients receiving
preoperative steroids experienced significantly less pain than patients who did not receive
preoperative steroids (p = 0.04). The mean post-operative morphine consumption 0-48 post-
operatively was 28.8 mg ( 23.6 mg). Only 52% of the patients received the planned amount of
acetaminophen of 4,000 mg/day.

Conclusion: Pain following craniotomy is moderate to severe in a substantial number of patients.

The quality of the analgesic treatment leaves room for improvement. Administration of
preoperative steroids may reduce post-craniotomy pain.

Funding: not relevant.

Trial registration: not relevant.

In 1996, De Benedittis et al performed a pilot study demonstrating that 60% of post-craniotomy

patients suffered from moderate-to-severe post-operative pain [1]. Prospective observational
studies have subsequently confirmed these findings [2-5]. However, despite the growing interest
in post-craniotomy pain, a gold standard for analgesic therapy in these patients is still lacking [6,
7]. An optimal analgesic therapy is important, considered the fact that sub-optimal postoperative
analgesia is not only of discomfort to the patient, but may also lead to an increase in the
incidence of post-operative complications, prolonged hospital stay and, in turn, to an increase in
health expenses [8]. Additionally, intense post-operative pain following craniotomy can result in
sympathetically induced arterial hypertension, which increases the risk of developing secondary
intracranial haemorrhage [9].
The aim of this prospective observational study was to explore pain intensity, nausea, sedation
and analgesic management after craniotomy. In particular, we hypothesized that administration
of preoperative steroids, supratentorial surgery, avoidance of surgical drains and male gender
would be associated with reduced post-operative pain scores. Furthermore, we wanted to
evaluate the current analgesic therapy administered to the post-craniotomy patients.

MATERIAL AND METHODS

This prospective observational study was carried out at the Department of Neurosurgery,
Rigshospitalet, Denmark. Data from 59 patients undergoing elective supratentorial or
infratentorial craniotomy in the period from November 2010 to January 2011 were collected. The
local research ethics committee was informed, but deemed it unnecessary to consider the study
for approval due to its strictly observational character.

Informed consent was, however, obtained from all patients.

The inclusion criteria were patients over the age of 18 years who presented for elective
craniotomy at Rigshospitalet, were able to speak and understand Danish, did not have a history
of substance abuse and were not diagnosed with chronic pain. The exclusion criteria were age
under 18 years, difficulties in communication, Burrhole trephination, shunt implantation,
emergency surgery, brain biopsies, transphenoidal surgery of the pituitary gland and extubation
later than 1 h after the end of surgery.

On the day before surgery, patients were interviewed and received information regarding
perioperative analgesic and antiemetic use. Furthermore, they were instructed in the use of the
11-point numeric rating scale (NRS) for pain assessment, and a four-point nausea and sedation
score (0: no nausea/sedation, 1: light nausea/sedation, 2: moderate nausea/sedation, 3: severe
nausea/sedation). After instruction patients were scored for NRS and nausea at rest.

ction with propofol 2 mg/kg, followed by maintenance of anaesthesia with remifentanil 30

microgram/kg/h and propofol 5 mg/kg/h. All patients received 0.1 mg/kg intravenous morphine
30 minutes before wound closure.

Data regarding surgical procedure (supratentorial versus infratentorial), anaesthetic, and

administered analgesics and antiemetics were documented. Post-operative pain, nausea and
sedation were recorded at one, two, four, eight, and 24 h after extubation. Post-operative
analgesic and antiemetic consumption was recorded 48 h post-operatively. For comparison, post-
operative opioids were converted into morphine equivalents. All patients were prescribed
standard analgesic treatment consisting of oral acetaminophen 4,000 mg/day with supplemental
morphine (intravenous or oral) as needed.

Trial registration: not relevant.

Statistics

All statistical analyses were performed with SPSS for Windows, version 18.0.

Mean and median values and standard deviations of post-operative NRS score, nausea and
sedation scores were calculated. We calculated the mean value of post-operative morphine
consumption 48 h post-operatively and the area under the curve (AUC) of NRS scores from
patients stratified according to gender and to administration of preoperative steroids versus no
preoperative steroids.

The Mann-Whitney U test was used to compare NRS scores, nausea, and sedation scores, and
post-operative morphine consumption between genders along with type of surgical procedure,
administration of preoperative steroids and application of surgical drains.

RESULTS

Fifty-nine patients were included in the study. A total of nine patients were excluded. Five
patients were excluded due to cancellation of surgery, one was intubated 8 h post-operatively and
three patients developed aphasia. In ten patients, data on post-operative pain, nausea, vomiting
and sedation were incomplete (i.e. measurements were not made at all five time points). The
collected data from these patients were included in the analysis. Data of the included patients are
presented in Table 1.

After the first post-operative hour, 56% of the included patients had moderate to severe pain
(NRS 4-10), which decreased to 38% at 24 h post-operatively. The pain intensity during the
observational period is illustrated in Figure 1. The median NRS score during the 24-h study
period ranged from two to four.

Patients receiving preoperative steroids experienced significantly less pain than patients
receiving no steroids (p = 0.04) (Figure 2). No significant differences in pain score or post-
operative analgesic consumption were observed when comparing surgical site (infratentorial
versus supratentorial), application of surgical drains, or gender. However, we observed a
tendency towards higher pain scores in females and in patients undergoing infratentorial surgery.

The incidence of post-operative nausea and vomiting was low with 86% of the patients
experiencing no nausea after the first post-operative hour and 78% experiencing no pain at 24 h.
Likewise, the incidence of sedation in the first 24 post-operative hours was low as 62% of the
patients experienced no sedation and 22% experienced only light sedation at 24 h post-
operatively. The mean post-operative morphine consumption in the first 48 h post-operatively
was 28.8mg ( 23.6 mg.). Only 52% of the patients received the planned 4,000 mg/day of
acetaminophen (Table 2).

DISCUSSION

The study demonstrated that 56% of the patients had moderate-to-severe pain after the first post-
operative hour with a median NRS score of four. Furthermore, the study demonstrated significant
differences in post-operative pain when NRS scores from patients receiving preoperative steroids
were compared with scores from patients who did not receive preoperative steroids. A rather low
incidence of nausea, vomiting and sedation was demonstrated. Finally, only 52% of the patients
received the planned 4,000 mg/day of acetaminophen (Table 2).

Insufficient analgesic treatment causes discomfort to the patient and may also lead to an increase
in the incidence of post-operative complications. In the post-craniotomy patient, post-operative
pain resulting in an increase in arterial blood pressure may increase the risk of intracranial
haemorrhage [9]. The rather high number of patients suffering from moderate-to-severe pain
during the first 24 post-operative hours demonstrates the need for an improved analgesic strategy
in post-craniotomy patients.

Steroids have a well-known effect on post-operative nausea and vomiting (PONV) and pain [10].
Mordhorst et al demonstrated that use of intraoperative steroids reduced the risk of post-
craniotomy pain [5]. In accordance with our hypothesis, administration of preoperative steroids
to patients mainly with malignancies was associated with significantly lower pain scores up to
24 h post-operatively (Figure 2). These observational findings are encouraging, but have to be
documented together with the potential side effects in further randomised controlled trials.

In contrast to our initial hypothesis, no significant differences in pain scores were demonstrated
in this study when comparing gender, type of surgery and application of surgical drains. Several
studies have demonstrated significant differences in pain scores when comparing gender and
surgical procedure. However, the insignificant result in this trial was most probably due to the
rather low number of patients included.

The mean cumulated post-operative morphine consumption of 28.80 mg ( 23.6 mg) 0-48 h
post-operatively may seem low considering that a rather large number of patients suffered from
moderate-to-severe pain. The use of opioids in post-craniotomy patients is controversial. Their
use in this particular setting may be limited by the potential risk of respiratory depression,
sedation, nausea and vomiting, and by the fact that opioids may mask neurological assessment.
As discussed in a recent review [7], only one double-blinded, placebo-controlled randomised
trial (RCT) investigating morphine has been performed in this setting [11]. The trial
demonstrated no significant opioid-related side-effects of patient-controlled morphine (PCA-
morphine).

In our department, the basic analgesic regimen in post-craniotomy patients consists of

acetaminophen 4,000 mg/day with supplemental intravenous or oral morphine as needed. Fear of
opioid-related side-effects combined with insufficient amounts of acetaminophen means that the
risk of moderate-to-severe post-operative pain is imminent. The reason why only 52% of the
patients received the prescribed amount of acetaminophen of 4,000 mg/day remains unclear, but
similar results were recently documented by other departments at our institution [12].

In addition to pain, PONV is a major concern after craniotomy. PONV is a serious complication
that can lead to a potential elevation of intracranial pressure which may, in turn, increase the risk
of haemorrhage. Contrary to our initial hypothesis, the incidence of PONV was rather low in this
study. The limited use of morphine and successful antiemetic strategies may have contributed to
this. Administration of preoperative steroids with a known antiemetic effect had no significant
effect on nausea in our rather small observational study, but this result should not be considered
conclusive.

In conclusion, 56% of the patients included in this descriptive study experienced moderate to
severe pain during the first post-operative hours, and 38% at 24 h post-operatively. A rather low
incidence of nausea, vomiting and sedation was documented. Only 52% of the patients received
the prescribed amount of acetaminophen in the first 48 h. Our observations indicate that
administration of preoperative steroids may reduce post-craniotomy pain. Our study thus adds to
the evidence that post-craniotomy pain as well as its handling seems to be a significant problem
which leaves room for improvement. In addition, there is an urgent need for well-performed
RCTs on pain therapy following craniotomy [7].

(Suboptimal pain treatment after craniotomy, controversial opioid for craniotomy))

Introduction

Fragile X syndrome (FXS) is the most common cause of inherited mental retardation1,2
accounting for 30% of the cases,3 with an associated connective tissue disorder that leads to
mitral valve prolapse (MVP) in most subjects.4 The goal in our clinical case was to individualize
anaesthetic management in this syndrome given difficult airway control and the associated
cardiovascular pathology. Non-invasive cardiac output monitoring was used during the
emergency craniotomy in order to reduce morbidity with the help of goal-directed therapy and
avoid delaying the surgery, considering that this approach only requires placement of a finger
device that provides all the relevant information pertaining to the patient's haemodynamic status.

Clinical case

A 38 year-old male patient with FXS, severe mental retardation, and refractory epileptic
encephalopathy who suffered head injury following a tonicclonic seizure, with acute left
frontoparietal subdural haematoma requiring emergent craniotomy.

Once monitoring with arterial blood pressure (ABP), electrocardiography (EKG), pulse
oximetry, bispectral index (BIS) and non-invasive cardiac output (Nexfin BMEYE,
Amsterdam, The Netherlands) was established, cefazoline 2g and midazolam 5mg were given IV
through a peripheral venous line because the patient was uncooperative. Propofol 120mg,
remifentanil 0.1g/kg/min and 50mg of rocuronium were used for anaesthesia induction, with
subsequent successful endotracheal intubation with the help of Glidescope videolaryngoscopy.
Mechanical ventilation was started and right internal jugular venous access was established
under ultrasound guidance.

Propofol and remifentanil were used for anaesthesia maintenance. Mannitol 25g and furosemide
20mg were required for lowering intracranial pressure (ICP) (from 22mmHg initially down to
4mmHg after haematoma removal). ICP was monitored using a subdural intracranial pressure
sensor. Goal directed therapy (GDT) was implemented during the intraoperative period based on
cardiac output values obtained after the initial administration of 1000mL, increased to 4000mL
in response to signs of hypovolemia. An infusion of 0.1g/kg/min of noradrenaline was required
in order to normalize extremely low initial values of systemic vascular resistance and to maintain
mean arterial pressure at around 90mmHg, promoting adequate cerebral perfusion (Fig. 1).
Haemodynamic stability was maintained, vasoactive amine perfusion was removed and the
patient was transferred to the intensive care unit with adequate sedation and analgesia, and under
mechanical ventilation.

Fig. 1.

Non-invasive cardiac output monitoring display showing the various intraoperative

haemodynamic parameters.

Source: Authors.
Discussion

Described by Martin and Bell in 1943,3 FXS has typical physical and behavioural
characteristics.1 Prevalence in males (1:3.600) is higher than in females (1:8.000), it is being
associated with the X chromosome.1 It is caused by an abnormal expansion of the cytosine
guanineguanine triplet (CGG) in the FMR1 gene (Fragile X Mental Retardation 1 gene) on
chromosome X(Xq27.3), blocking the production of the FMR1 gene protein (FMRP).5

The phenotypic characteristics of this syndrome may have significant anaesthetic implications,
among other findings, because of craniofacial abnormalities. Physical characteristics include
macrocephaly,3,6 hyperteolirsm, strabismus, prognathism,3 large prominent ears, and
postpubertal marcroorchidism.7

Dental implantation is abnormal, with abraded dental surfaces as well as large crowns that create
severe boneteeth discrepancies,8 limiting mouth opening and impairing the placement of the
endotracheal tube.

Our patient had prognathism as well as abnormal dental implantation limiting mouth opening to
3cm. Because of an anticipated difficult airway, we decided to use the Glidescope
videolaryngoscope,9 with successful endotracheal intubation and minimum stimulus, thus
avoiding an increase in intracranial pressure.

FXS patients have excess joint laxity,7 flat feet, pectus excavatum and scoliosis. Patient
positioning on the operating table with adequate support points is essential in order to prevent
joint dislocations and gonadal compression.4

In 8090% of FXS cases there is moderate to severe mental retardation.3 Autism, hyperactivity,
agitation and anxiety are also frequent,3 hence the need for adequate sedation before induction,
because they are usually uncooperative.

In 1520%3 of cases there are partial complex and generalized tonic clonic seizures that are
usually benign3 and disappear before 20 years of age. The right premedication may reduce
surgical stress by raising the convulsive threshold perioperatively.

From the cardiovascular standpoint, 80% of cases may be associated with MVP with no previous
episodes of chest pain or palpitations4 but which may give rise to intraoperative arrhythmias.4
Occasionally, there is also aortic root dilatation,4 in which case it is advisable to use adequate
monitoring in order to implement GDT and create better conditions for improved results in major
surgery.10 Because our patient was transferred from another hospital as a vital emergency, we
were unable to confirm the presence of a MVP. However, given the high incidence of this
disorder and the fact that symptoms become exacerbated by anaesthesia induction, leading to
cardiovascular collapse,11 we decided to perform non-invasive cardiac output monitoring.
The main objective of the anaesthetic management was to prevent any reduction in left
ventricular volume during systole in order to reduce mitral valve prolapse.11

Reduced venous return and vascular resistance, tachycardia and increased contractility are not
well tolerated.11 During general anaesthesia, the use of vasopressors is recommended in order to
maintain ABP, together with short-acting beta-blockers for heart rate control, end-diastolic and
systolic volume preservation, and mitral prolapse control.11

Management is different in cases associated with mitral regurgitation because maintenance of

higher heart rates shortens diastolic time and reduces regurgitation volumes.12

Pulse pressure variation (PPV) and systolic volume variation (SVV) are dynamic predictors of
response to fluids in patients under mechanical ventilation. Their measurement is usually
invasive by means of the signal derived from the ABP curve. Non-invasive CO monitoring using
an inflatable cuff with an in-built photoelectric plethysmographic device provides continuous
ABP measurement13 based on the development of a pulsatile discharge from the arterial walls in
the finger.13

SVV and PPV measurements without the need to use an intra-arterial catheter offers an
advantage in emergency surgery, particularly in neurosurgery, where any delay in starting the
intervention may determine worse clinical outcomes. GDT allows for early detection of
pathophysiological changes and individualized adjustment of intraoperative haemodynamic
management.10 Invasive ABP monitoring with an intra-arterial catheter13 is considered the gold
standard.14 However, non-invasive measurement of dynamic predictors of response to fluid
replacement has been shown to have high specificity and sensitivity,13 with improved safety and
comfort.15 This system can be used on top of non-invasive ABP monitoring in
haemodynamically stable patients under general anaesthesia, with the benefit of providing beat-
to-beat ABP measurements.14 However, its concomitant use with the PICCO transpulmonary
thermodilution monitor for measuring cardiac output is controversial.15,16

In summary, we discuss the anaesthetic management of patients with FXS, given the low
incidence of this disorders and the very few reports found in the anaesthesia literature. We
believe that non-invasive cardiac output monitoring is a new option for emergency neurosurgical
procedures and in patients with heart disease, considering that it shortens anaesthesia time and
provides reliable parameters for GDT. Moreover, we believe that videolaryngoscopy is the first
choice for managing a predictably difficult airway in which endotracheal intubation might result
in a significant increase in ICP.

(Haemodynamic management using non-invasive cardiac output monitoring for

urgent craniotomy in fragile X syndrome: Case report)
Carotid endarterectomy reduces the risk of stroke among patients with
symptomatic1, 2 and asymptomatic3, 4 carotid stenosis. The American Heart
Association (AHA) guidelines for carotid endarterectomy 5 laid the foundation for
recent practice in carotid surgery. Improvements in endovascular techniques and
the development of carotid artery stenting (CAS) have caused a paradigm shift in
the treatment of carotid stenosis.

Although none of the randomised trials comparing CAS and CEA 6, 7, 8, 9, 10 could
demonstrate any significant benefit of stenting over surgery regarding stroke and
death, endovascular therapy may have some possible advantages, such as
avoidance of general anaesthesia and surgical trauma.

Although CEA was often performed under LA in the 1960's, 11 most surgeons prefer
to operate on a fully anaesthetised and relaxed patient. Reports from centres of
excellence12 as well as a recent large meta-analysis13 suggesting improved outcome
of CEA performed under LA are changing the attitudes of many surgeons. The
hypothesis of the completed but unpublished randomized GALA (General versus
local anesthesia) trial14 was to show a 50% reduction in stroke and death rates in
favour of LA. Recruitment of patients for the trial is completed and results are
expected this year.

Local anaesthesia may have some advantages compared to GA, such as a. risk of
myocardial infarction rates and pulmonary complications. 13 Direct neurological
monitoring of the awake patient eliminates the necessity for equipment for that
purpose. The aim of our study was to analyse the impact of the different
anaesthetic techniques (local and general) in CEA on neurological outcome (primary
endpoint) and mortality (secondary endpoint).

Patients and Methods

Over a 10 year period (January 1995December 2004) the hospital records of all
1341 patients consecutively operated on in the two units were collected for
analysis. 876 operations under GA and 465 under LA were performed in the two
teaching hospitals of Giessen and Dessau. Indications for surgery were
asymptomatic carotid stenosis >80% or symptomatic carotid stenosis of >70%.
Hospital volume was about sixty operations per year in both units.

Duplex imaging of the carotids was performed in all cases and findings confirmed by
angiography (n=830) or magnetic resonance angiography (MRA) (n=438). MRA was
more frequently performed during the later years of our study.

The choice of anaesthetic method was based on surgeon and patients' preferences.
For the analysis, the patients were divided into two groups based on anaesthetic
procedure. Comparisons were made with respect to preoperative risk factors,
intraoperative events such as use of intraluminal shunt, and postoperative
complications. Operating time and clamping time of the carotid artery were
recorded. All patients were examined by an independent neurologist pre- and
postoperatively.

Postoperatively, major stroke was defined as neurologic deficit lasting beyond 30

days and leading to handicap. Minor stroke was defined as any transient focal deficit
not leaving handicap (combining transient ischemic attack (TIA) and prolonged
reversible ischemic neurologic deficit (PRIND)). All patients with new central
neurological deficits underwent computed tomography postoperatively. Another
complication noted was significant neck haematoma requiring surgical evacuation.

LA was performed using 0.75% ropivacain. Altogether 1518ml were used to

infiltrate the skin and underlying tissue. Additional xylocaine was supplemented by
the surgeon as needed (15ml). Intravenous sedatives and anxiolytics were used as
suggested by the anaesthetist. At the time of carotid clamping, neurology was
tested by the patient's ability to squeeze a duck making a sound. Shunting was
performed, if the response was inappropriate.

In the GA group, the patients were intubated. Etomidate and propofol were used for
anaesthesia with remifentanyl infusion. Somato-sensoric evoked potential (SSEP)
monitoring was routinely used. Carotid shunting was performed if SSEP showed
slowing to 50% of pre-clamping response.

CEA was performed conventionally, using patch closure in all cases. Heparin was
administered before carotid clamping (100I.E. per kg) and was not routinely
reversed.

All patients were admitted to the intensive care unit (ICU) for 24 hours. Blood
pressure (BP) was recorded directly using a radial artery catheter. Postoperative
hypertension was reported when systolic blood pressure remaining higher than 160
mmHg for >2 hours in spite of medical treatment.

The patients in the LA group were older (mean age was 68.5 vs. 66.5 years,
p<0.001). There were no significant differences with regard to coronary or
peripheral occlusive arterial disease, hyperlipidemia, renal disease or stenosis on
the contralateral side.

There were more symptomatic patients (Combined transient ischemic event and
stroke) in the GA group (63.2% vs. 48.6%, p<0.001). The number of previous
strokes was not significantly different (23.7% vs. 19.7%). Hypertension and diabetes
were more common in the LA group (88.1% vs. 79.7%, p<0.01 and 35.2% vs.
22.6%, p<0.001 respectively). Smoking was more common in the GA group (40.1%
vs. 33.5%, p<0.02).

Operative variables in the groups are shown in Table 2.

There was no statistically significant difference in shunt use (GA=15.9% vs.

LA=13.6%). Total operating time and clamping time were shorter under LA (103min
vs. 111min, p<0.001 and 33min vs. 39.4min, p<0.001, respectively).

There was no statistically significant difference in perioperative mortality between

the groups (GA 0.5%, LA 0.8%). Cerebral complications were significantly more
common under GA (6.9% vs. 3.4%, p<0.009, relative risk 0.478, 95% confidence
interval (CI) 0.2720.839). More strokes (3.5% vs. 2.3%) and transient ischemic
attacks (3.4% vs. 1.0%), were seen in the GA group. Combined mortality and stroke
rate was not significantly different (4.1% vs. 3.2%, p=0.31, relative risk 0.715,
95%CI 0.3741.369).

A subgroup analysis of the asymptomatic and symptomatic patients showed that

there was no specific advantage for the different anesthetic procedures. For
asymptomatic patients the incidence of any postoperative neurologic event was
1.7% (4/238 patients) after LA and 4.3% (14/322) after GA (p=0.077; relative risk
0.514, 95% CI 0.22-1.22). For the symptomatic patients it was 5.3% (12/225) after
LA and 8.5% (47/554) after GA (p=0.132; relative risk 0.687, 95% CI 0.411.153).

Episodes of hypertension were significantly more common in the LA group during

the first 24 hours (47.7% vs. 20.4%, p<0.001).

Haematomas requiring surgical evacuation were more commonly seen in the GA

group (6.4% vs. 3.0%, p<0.02).

To compare and quantify the possible influence of all risk factors and the operation
technique on the occurrence of a cerebrovascular event, a backward stepwise
logistic regression analysis was performed (Table 3). Hosmer-Lemeshow-Test gave
0.345. The result has to be viewed with caution as the regression analysis, although
statistically significant (p<0.001) explains 10.4% (Nagelkerkes R square=0.104) of
the variability. Preoperative neurological events (odds ratio 1.3), operation time
(odds ratio 1.02 per minute) and type of anaesthesia (odds ratio 2.3) had a
significant impact on postoperative neurological events. In order to predict
morbidity and mortality (defined as postoperative mortality or stroke) out of the
preoperative and intraoperative variables, the best model (Hosmer-Lemeshow-
Significance=0.498) explained 9% of the variability. Significant predictor of outcome
were preoperative cerebrovascular events and total operating time.
The introduction of carotid stenting has resulted in a paradigm shift in the
management of carotid stenosis, although no clear benefits of CAS has been
demonstrated.6, 7, 8, 9, 10 One of the most obvious advantages of CAS is the possibility
to perform the procedure under local anaesthesia. If the aim is a fair comparison
between CEA and CAS, it probably should be made between CEA performed under
LA rather than GA. Our aim was to investigate whether CEA performed under the
two types of anaesthesia differ in outcome.

We included 1341 consecutive patients in our study, making this one of the largest
of its kind. The study was retrospective, the choice of anaesthetic method was
based on surgeon and patients' preference.

There were slight differences between the groups with regard to clinical
characteristics. The patients in the LA group were older (p<0.001). There were more
diabetics (p=0.001) and hypertensive patients (p<0.01) in the LA group. There were
more symptomatic patients in the GA group (p<0.001), but the number of patients
with previous stroke was not significantly different (Table 1).

One of the most important advantages of CEA performed under LA is thought to be

the monitoring of the awake patient. Patient co-operation is important during an
operation lasting for up to two hours and neurological monitoring calls for an
experienced anaesthetist. Operating on an awake patient becomes more difficult
with a restless patient and in case of a sudden neurologic deficit immediate
shunting becomes necessary. Nevertheless there may be the advantage of accurate
selection of the patients' requiring shunts which may explain the lower incidence of
cerebrovascular complications (p<0.009). Shunt use is thought to be reduced under
LA.15, 16 Differences in carotid shunting have also been reported by Watts et al.17 In
their study, shunt use was significantly more common in the GA group (83% vs.
9%). In our study, there was no difference in shunt use (16% vs. 14%), which
reflects the use of SSEP monitoring in the GA group and very selective shunting.
Routine shunting is naturally an option, but there may be advantages in selective
shunting,16 such as fewer embolic events due to minimised manipulation of the
artery. It has been postulated that more shunt use means longer clamping and
operating times,17 but shunt use has not been shown to be related to neurologic
complications. Shunt use and monitoring methods are currently debated. 16

Gabelman et al.18 have confirmed the benefits of performing CEA in awake patients
by showing less operative time and intensive care unit stay compared to those
operated on under GA. In our study, operative time and clamping time were
significantly shorter in the LA group. Connections between operating or clamping
time and neurological outcome have not been reported so far, but in our series total
operating time had a significant impact on neurological outcome in regression
analysis (odds ratio 1.014).
Other assumed advantages of operating under LA include fewer cardiac and
pulmonary complications.13 In our study, the incidence of myocardial infarction (MI)
and pneumonia were lower than 1% in both groups.

Some surgeons fear that sudden conversion from LA to GA may be dangerous.

However, conversion has been reported to be very rare. 19 In our study the rate was
1.3% (n=6), and conversion did not lead to added mortality or morbidity.

Blood pressure variability between the anaesthetic regimes has been reported in
several studies.18, 20, 21 Most studies report higher BP variability in patients operated
on under GA. However, in our study the LA group had higher systolic BP values in
the ICU (48% vs. 20%, p<0.001). The same finding was also reported by Forsell
et al.22 In our study, all pressure recordings were performed continuously and intra-
arterially on an intensive care unit, making the data very reliable. However, there
were more hypertensive patients in the LA group, which may at least partially
explain our findings. Another factor is that the patients in the LA group were older.
On the other hand, patients recovering from general anaesthesia may simply get
more attention from the ICU personnel.

In the study by Watts et al.,17 there was no difference in postoperative bleeding

between the groups. Rerkasem et al.13 found local haemorrhage significantly more
associated with general anaesthesia in the meta-analysis of the reviewed studies.
Surgical evacuation of a haematoma was more commonly performed in patients
operated on under GA in our study (6% vs. 3%, p<0.02). There was no difference in
heparin use between the groups.

The most relevant result of our study is the significantly better neurological
outcome in favour of local anaesthesia. Cerebral complications were significantly
more common after GA (6.9% vs. 3.4%), More strokes (3.5% vs. 2.3%) and transient
ischemic attacks (3.4% vs. 1.0%), were seen in the GA group. Regression analysis
shows that preoperative neurological events, operation time and type of
anaesthesia have a significant impact on postoperative neurological events. The
result should be viewed with caution as the regression analysis, although
statistically significant, explains 10.4% of the variability. Clearly, other factors
influence the neurological outcome as well, but using this less known but well-
accepted statistical method, GA can be shown to increase the odds of a
postoperative cerebrovascular event by a factor of 2.3.

A retrospective study by Watts et al.17 reported no difference between LA and GA

with regard to neurologic complications among 582 patients. Our experience in a
much larger series suggests LA is safer for CEA.

Our study has some obvious limitations. The form of anaesthesia used was not
randomised. However, prospective data collection was used. The data is based on a
rigorous analysis of a large population of patients operated on consecutively.
Certain patient selection bias is possible as well, but a wide range of variables were
considered.

Conclusion

Carotid endarterectomy performed under local anaesthesia can be performed safely

and may lead to a better neurological outcome as compared to general anesthesia.
Risk factor analysis did not reveal specific risk groups.

(Local versus General Anaesthesia for Carotid

Endarterectomy Improving the Gold Standard ?)
At present, no satisfactory reports on the monitoring of cerebral function to predict functional
outcomes after brain damage such as traumatic brain injury (TBI) and stroke. The middle latency
auditory-evoked potential index (MLAEPi) monitor (aepEX plus, Audiomex, UK) is a mobile
MLAEP monitor measuring the degree of consciousness that is represented by numerical values.
Hence, we hypothesized that MLAEPi predicts neurological outcome after emergency
craniotomy among patients with disturbance of consciousness (DOC), which was caused by
brain damage.

Methods
The afore-mentioned patients who underwent emergency craniotomy within 12 h of brain
damage and were subsequently monitored using MLAEPi were enrolled in this study. DOC was
defined as an initial Glasgow Coma Scale score<8. MLAEPi was measured for 14 days after
craniotomy. Neurological outcome was evaluated before discharge using a cerebral performance
category (CPC) score and classified into three groups: favorable outcome group for a CPC score
of 1 or 2, unfavorable outcome group for a score of 3 or 4, and brain dead (BD) group for a score
of 5.

Results
Thirty-two patients were included in this study (17 with TBIs and 15 with acute stroke).
Regarding outcome, 10 patients had a favorable outcome, 15 had an unfavorable outcome, and 7
were pronounced BD. MLAEPi was observed to be significantly higher on day 5 than that
observed immediately after craniotomy in cases of favorable or unfavorable outcome (633.5
vs. 362.5 in favorable outcome; 633.5 vs. 341.8 in unfavorable outcome). MLAEPi was
significantly lower in BD patients than in those with a favorable or unfavorable outcome on day
3 (244.2 in BD vs. 525.2 and 452.7 in favorable and unfavorable outcome, respectively)
and after day 4. MLAEPi was significantly higher in patients with a favorable outcome than in
those with a favorable or unfavorable outcome after day 6 (682.3 in favorable outcome vs. 48
2.3 in unfavorable outcome).
Conclusion
We believe that MLAEPi satisfactorily denotes cerebral function and predicts outcomes after
emergency craniotomy in patients with DOC, which was caused by acute brain damage.

Keywords: Intensive care unit, Critical care, Emergency medicine, Traumatic brain
injury, Stroke, Monitoring

Go to:

Background
Monitoring cerebral function is crucial in surgical critical care. However, to date, there have been
no satisfactory reports on the monitoring of cerebral function to predict functional outcome after
brain damage, i.e., traumatic brain injury (TBI) and stroke.

Several studies have examined the clinical application of the auditory-evoked potentials (AEPs)
as well as the bispectral (BIS) index that provide a good indication of the degree of
consciousness under anesthesia in a surgical setting [1]. In particular, cerebral function can be
monitored noninvasively by measuring middle latency (ML) AEPs [2]. MLAEPs are derived
from AEPs, which reflect the morphology of MLAEP curves, originating from the part of the
auditory pathway from the medial geniculate body to the primary auditory cortex. MLAEPs are
also less affected by age than other AEP components.

The aepEX plus (Audiomex, Glasgow, Scotland, UK) is the first mobile and battery-operated
MLAEP index (MLAEPi) monitor that can evaluate the level of anesthesia by the numerical
value acquired using MLAEP, which is a current generated by excitement of the cranial nerves
during information processing in the brain [3]. This monitor is already available worldwide. The
aepEX plus monitoring system is being increasingly used to measure both the level of
anesthesia and the cerebral function in intensive care units (ICUs) [1, 46]. The recommended
range of MLAEPi in general anesthesia settings is 30 to 45. We previously reported the
effectiveness of MLAEPi monitoring at the emergency department (ED) in patients with cardiac
arrest and disturbance of consciousness (DOC), which was defined as an initial Glasgow Coma
Scale (GCS) score<8, and the recommended range of MLAEPi in non-sedated patients with
DOC was between 35 and 61 [7, 8]. In this study, we speculated that MLAEPi predicts
neurological outcome after emergency craniotomy among patients with DOC, which was caused
by brain damage.

Go to:

Materials and methods

Patients and study design
This prospective study was conducted in the emergency department of Tokyo Medical University
Hachioji Medical Center between September 2010 and March 2013. The ethics committees at the
institution approved the study design, and written informed consent was obtained from the next
of kin of the patients or a posteriori from the patients themselves where possible.

Patients who were admitted to our ED and underwent emergency craniotomy within 12 h of
brain damage and were subsequently monitored using MLAEPi were enrolled in this study. DOC
was defined as an initial GCS score<8. We excluded comatose patients younger than 15 years of
age, those who had experienced DOC before the onset of brain damage, and those who had drug
intoxication, alcoholism, tympanic injury, or terminal diseases during the study.

AepEX and middle latency auditory-evoked potential index

The aepEX plus monitoring system was activated with bilateral click stimuli through
earphones at an intensity<90 dB for a nominal frequency of 6.9 Hz through a pair of
headphones to provide MLAEP. It continuously generates MLAEPi, which is a dimensionless
number scaled between 99 (wide awake) and 0 (no brain activity), with differences between
successive segments of the curve constructed from its amplitude. The aepEX value is updated
every 0.3 s instead of 36.9 s (256 sweeps) by a moving time-average technique. The aepEX
plus identifies the brainstem and cortical components of MLAEP following auditory stimuli,
particularly positive Pa and negative Nb waves. MLAEPi is calculated from consistent decreases
in amplitude and increases in latency, resulting in individual waves within 144 ms [9]. Detected
AEPs were consecutively extracted from the raw electroencephalogram (EEG) signal reflecting
the brainstem AEP and MLAEP by an internal processor. The aepEX plus values were closely
related to the AEP waveforms and calculated as the sum of the square root of the absolute
difference between every two successive 0.56-ms segments of the AEP waveforms (Fig. 1).

Fig. 1

The aepEX plus monitor

Intervention
Using the aepEX plus, MLAEPi was continuously calculated from information provided by
disposable sensor EEG electrodes affixed to the patients mid (ground electrode) and right (active
electrode) forehead as well as the right mastoid (active electrode) after cleaning the skin with 70
% isopropanol. In addition, an emergency medical physician used earphones to determine
auditory stimuli. MLAEPi was measured for 14 days after craniotomy every morning and an
average of 3 s MLAEPi was recorded for each MLAEPi.

All patients were administered sedatives for no longer than 3 days after the onset of brain
damage. All procedural sedations were performed by expert emergency physicians, and the
choice of sedatives was at the discretion of the emergency physician. In our ICUs, sedation was
achieved using midazolam (maximum 0.2 mg/kg/h) or propofol (maximum 3 mg/kg/h) as a
sedative and fentanyl (maximum 1.5 g/kg/h) for pain management. Neurological outcome was
evaluated before discharge using a cerebral performance category (CPC) score 1 month after
brain damage and classified into the following three groups: favorable outcome group for a CPC
score of 1 or 2; unfavorable outcome group for a score of 3 or 4; and brain dead (BD) group,
which equated to a score of 5 (Table 1).

Table 1

Cerebral performance category

Data collection
The following characteristics were noted from the charts of the comatose patients with brain
damage: age, gender, initial GCS, vital signs, clinical history, value of MLAEPi, and etiologies
of brain damage. Continuous MLAEPi monitoring did not affect standard intensive care
treatment and nursing in the ICU.

Statistical analyses
Data from all eligible patients were analyzed. Continuous variables were shown as median
values with interquartile ranges. Intergroup differences were statistically assessed using the
KruskalWallis test and one-way ANOVA with repeated measures, depending on the distribution
of measured variables using Prism version 6.0a statistical software (GraphPad Software, San
Diego, CA, USA).

Go to:
Results
Clinical characteristics
Thirty-two comatose patients with brain damage (median age was 55 years) were included in this
study. The demographics and clinical characteristics of patients are shown in Table 2. The GCS
score of E1V1M2 was common in this study. In the cohort, 17 patients had TBI and 15 had acute
stroke. The etiologies of brain damage of patients are shown in Table 3. Regarding outcome, 10
patients had neurologically favorable outcome, 15 had unfavorable outcome, and 7 were
pronounced BD.

Table 2

The demographics and clinical characteristics of patients

Table 3

Etiologies of brain damage of patients

Changes in MLAEPi at ICU

MLAEPi values measured until 14 days after injury are shown in Table 4. During patient
sedation for 3 days after injury, MLAEPi did not significantly differ among the favorable
outcome, unfavorable outcome, and BD groups. However, MLAEPi was observed to be
significantly higher on day 5 than that observed immediately after craniotomy in the favorable
outcome (p<0.01) and unfavorable outcome groups (p <0.01). MLAEPi was also observed to
be significantly lower in the BD group than that in the favorable outcome and unfavorable
outcome groups after day 3 (p<0.01). Furthermore, MLAEPi was significantly higher in the
favorable outcome group compared with that in the unfavorable outcome group after day 6 (p<
0.01). MLAEPi in the BD group did not show any significant increase during study periods
(Fig. 2).

Table 4

Measures of middle latency auditory-evoked potential index

Fig. 2

Changes in MLAEPi value with a standard error at the intensive care unit. *=non-
significance; =p<0.01 vs. post-craniotomy; =p<0.01 ...
Go to:

Discussion

(Middle latency auditory-evoked potential index

monitoring of cerebral function to predict functional
outcome after emergency craniotomy in patients with
brain damage)
To our knowledge, this is the first prospective study to evaluate the changes of MLAEPi in
patients with brain damage in an ICU. Several studies have found that white matter tracts within
the cerebral hemisphere and brain stem are injured following TBI or acute stroke [10, 11].
MLAEP is usually obtained intermittently and derived from AEPs, which reflects the
morphology of MLAEP curves. Previous studies have failed to demonstrate the usefulness of
single components of the early somatosensory-evoked potentials (SEP) and short latency AEP
for predicting the clinical outcome after TBI or acute ischemic injuries following stroke or
cardiac arrest [12, 13]. Although middle latency activities are considered to be good predictors of
prognosis in comatose patients, the predictive value of MLAEP remains uncertain. Therefore, we
hypothesized that MLAEPi decreases among patients with DOC caused by head injury or acute
stroke and can predict the neurological outcomes after emergency craniotomy.

Most studies have evaluated MLAEPi as an indicator of the state of anesthesia with 100 %
specificity using an MLAEPi cut-off value of 37 for unconsciousness during anesthesia [14]. We
previously demonstrated that the recommended range of MLAEPi in non-sedated patients with
DOC between 35 and 61 [8]. MLAEPi is profoundly affected by the decreasing amplitudes and
increasing latencies induced by hypnotic drugs such as midazolam and propofol. Although the
recommended range of MLAEPi in surgical anesthesia settings is 30 to 45, the adequate range of
MLAEPi for critical care patients in ICUs remains unknown [3]. Our results show that MLAEPi
did not differ significantly in patients with brain damage across the three outcome groups during
sedative periods. However, we were able to demonstrate that MLAEPi increased significantly
between patients with favorable outcome or unfavorable outcome compared with patients with
BD immediately after finishing sedation. Moreover, MLAEPi increased significantly among
patients with favorable outcome compared with patients with unfavorable outcome within
1 week after the onset of brain damage.

Several studies found that the GCS score and BIS value were relatively correlated in critically ill
patients, and the processed EEG is a non-invasive method for monitoring consciousness during
anesthesia or critical care sedation [9, 15]. BIS is the numerical value acquired using
spontaneous EEG. Although a BIS value of 0 is useful in severe TBI or ischemic brain injury for
the early detection and confirmation of BD with a GCS score of 3, there has been no satisfactory
report on BIS monitoring of cerebral function to predict functional outcome after brain damage
[16, 17]. Middle latency SEP may be valuable for increasing sensitivity without any loss of
specificity for predicting unfavorable outcome in patients after stroke, ischemic brain injury or
TBI [13, 18]. However, the prognostic value of SEPs remains controversial and should never be
considered in isolation but should be integrated with other neurophysiological tools and clinical
examination [19]. It has been reported that aepEX monitoring is a more effective indicator for
determining the state of consciousness than BIS or any other EEG-based monitoring method [4,
5, 14, 20].

The major difference between any other EEG-based monitoring method and the aepEX method
is that MLAEPi is the numerical value acquired using MLAEP. The strong correlation between
GCS scores and MLAEPi in patients with DOC has been previously reported [7, 8].
Furthermore, because of its small size and battery operation facility, aepEX monitoring can
provide a consistent assessment of MLAEP during life-saving procedures while transporting
patients within the hospital and in patients admitted to ICUs. In this study, we demonstrated that
MLAEPi may be a reasonable indicator of neurological outcomes in patients with brain damage
who have underwent emergency craniotomy in an acute care setting.

This study has several limitations, particularly the small number of evaluated patients sustaining
different types of brain damage such as focal TBI, diffuse TBI, and cerebral stroke. Second, we
did not measure initial MLAEPi in ED and did not use other monitors such as BIS, short latency
AEP, early SEP, or MLAEP to evaluate the degree of DOC in the ICU. Third, the measurement
of MLAEPi was performed by a single emergency physician (Dr. TJ). Thus, the patients with
DOC were not enrolled sequentially in this study. Fourth, we only obtained MLAEPi data for a
period of up to two weeks and had no records from the late phase following discharge from the
hospital. The purpose of this study was to assess MLAEPi monitoring for predicting functional
outcomes in the acute phase in patients admitted to ICUs. Thus, we limited the study endpoint to
the initial evaluation in the acute phase and did not perform long-term MLAEPi follow-up.

Go to:

Conclusion
MLAEPi can satisfactorily denote cerebral function as represented by simple numerical values,
and predict functional outcomes after emergency craniotomy among patients with DOC, which
was caused by brain damage. Large studies are essential for further evaluating the reliable cut-
offs of MLAEPi in patients with acute brain damage.
(Middle latency auditory-evoked potential index
monitoring of cerebral function to predict functional
outcome after emergency craniotomy in patients with
brain damage)
BACKGROUND:
Asleep-awake craniotomy presents challenges for the anesthetist who has to provide adequate
sedation and analgesia but also requires an awake and cooperative patient for neurological
testing. In this setting, we hypothesized that Bispectral Index (BIS) monitoring might be helpful
in shortening the patient's awakening and in predicting recovery of consciousness in order to
initiate reliable intraoperative brain mapping.

METHODS:
An observational prospective study was performed on 27 consecutive asleep-awake
craniotomies, in which BIS was monitored and BIS data collected offline. Nine critical
intraoperative time points were defined and analyzed [preinduction, start of surgery, termination
of hypnotic drug, eye opening, obeying simple commands, laryngeal mask airway (LMA)
removal, initiation of brain mapping, initiation of closure, and end of surgery].

RESULTS:
A shorter time to LMA removal was associated with a higher BIS at the termination of the
hypnotic drug (P=0.016, Mann-Whitney U test). From the initiation of surgery to the time of
LMA removal, BIS was significantly lower than the preinduction values, whereas at the initiation
of brain mapping, BIS returned to the preinduction values (Friedman test P<0.0001, Dunns
multiple comparisons test). Compared with LMA removal, BIS values >85 predicted the
initiation of brain mapping with a sensitivity of 44% (95% confidence interval, 25.5%-64.7%)
and a specificity of 74% (95% confidence interval, 53.7%-89%).

CONCLUSIONS:
During asleep-awake craniotomies, higher BIS values at the end of the asleep phase are
associated with shorter time to LMA removal, suggesting that BIS monitoring may be beneficial
in shortening recovery from anesthesia. During the awake phase, the return of BIS to the
preinduction values appeared to indicate full recovery of consciousness, thereby allowing a
reliable language testing.

(Bispectral index during asleep-awake craniotomies.)

The awake craniotomy is an important technique used for brain tumour excision from eloquent
cortex, epilepsy surgery, and deep brain stimulation surgery. It has been used, less commonly, in
the management of mycotic aneurysms and arterio-venous malformations near critical brain
areas.

The benefits are considered to be of increased lesion removal, with growing evidence of
improved survival benefit,1 whilst minimizing damage to eloquent cortex and resulting
postoperative neurological dysfunction. Other advantages include a shorter hospitalization time,
hence reduced cost of care, and a decreased incidence of postoperative complications such as
nausea and vomiting.

The concept of an awake craniotomy predates the existence of anaesthesia and in ancient times,
trephining of the skull was used to get rid of evil air.2 It is a procedure that has gained in
popularity because of advances in diagnosis, intra-operative functional neurosurgical technology,
developments in anaesthetic agents and monitoring, and the patient's expectations.

The term awake craniotomy is misleading as the patient is not fully awake for the entirety of
the procedure. The more surgically stimulating parts of the procedure require varying levels of
sedation, or anaesthesia. The patient is fully awake during the mapping procedure during which
lesion resection takes place.

The common anaesthetic techniques used are sedation only or general anaesthesia, and awaking
the patient for cortical mapping and resection, with the option of re-anaesthetizing for closure.
The patient has a scalp block inserted for pain relief usually for all anaesthetic approaches.
Occasionally the anaesthetic technique of awake with a scalp block alone is utilized, this can be
useful in elderly patients.

The anaesthetic considerations for an awake craniotomy can make it challenging, and this article
will consider the issues involved.

Previous SectionNext Section

Preoperative
One of the most important considerations is careful patient selection. There are both physical and
psychological prerequisites to avoid failure intra-operatively. All patients should have
consultations with the neurosurgeon and the anaesthetist.

These consultations allow the assessment of patient suitability (see Table 1 for absolute and
relative contraindications), and preparation of the patient for the procedure. This involves a full
assessment of the patient's co-morbidities, which should be optimized before operation, in order
to decrease intra-operative failure of the awake technique. It is also important to be aware of how
the patient's presenting problem for surgery affects them, for example, seizure type and
frequency or the presence of preoperative neurological deficits.
View this table:

In this window

In a new window

Table 1
Anaesthetic contraindications

Preoperative preparation includes providing detailed information to the patient. The patient must
know what to expect and the anaesthetic risks involved. This usually includes verbal and written
information. Occasionally the patient is helped by visiting the theatre and visualizing the
environment and equipment involved.

Patients may be seen by the neuropsychologist before operation if the lesion involves speech and
language centres, and their baseline responses to picture cards are assessed and recorded.3 In
some neurosurgical centres, the neuroanaesthetist performs the role of functional assessment in
theatre.

These preoperative visits provide an invaluable opportunity for the multidisciplinary team to
create a rapport with the patient and therefore encourage trust and familiarity.

Previous SectionNext Section

Theatre preparation
It is vital that communication between the anaesthetist and surgeon is effective and this is often
aided by ensuring familiarity and discussion of the operative plan in detail. It is imperative that
the plan for anaesthesia and surgery is well communicated to all the members of the theatre team.

As in all surgical cases, equipment should be checked and patient scans should be available
before commencing. The operating table must be made as comfortable as possible as the patient
may be lying in one position for several hours. The temperature of the operating theatre should
be comfortable for the patient, and staff numbers should be minimized to alleviate unneeded
noise and patient anxiety.

Consideration should be given to the operating theatre layout and position of the patient. The
ability to communicate with the patient should be maintained at all times and of equal
importance is access to the patient during adverse incidents.4

The patient position is dictated by the location of the lesion. This is usually a lateral or supine
position, but with occipital lesions and testing the visual cortex, a sitting position may be used. In
any position, it is important that when the patient is fully awake during mapping that they are
able to see and communicate with the anaesthetist or neuropsychologist. Sterile drapes used must
not encroach over the patients face as this can cause patient claustrophobia and difficulty
communicating.

A typical operating theatre layout is shown in Figure 1.

View larger version:

In this page

In a new window

Download as PowerPoint Slide

Fig 1
Theatre layout for an awake craniotomy.

Previous SectionNext Section

General anaesthetic principles

Premedication is not common, but consideration must be given to acid reflux prophylaxis, and
patients must take their usual steroid, anti-epileptic, or anti-hypertensive medication. Some
neurosurgical centres may load the patient with anti-convulsants on the day of surgery or check
for therapeutic plasma levels of anti-convulsants if patients are already on them.

Standard anaesthetic monitoring is applied as according to the Association of Anaesthetists of

Great Britain and Ireland guidelines. Large-bore intravenous access is gained and the majority of
anaesthetists insert an arterial line, usually sedated or asleep.

The use of other forms of monitoring is variable. Depth of anaesthesia monitors, for example
bispectral index monitoring (BIS), is sometimes utilized and there is some suggestion that its
use reduces the amount of anaesthetic agents administered, and hence the time taken for patient
emergence and co-operation for cortical mapping.5
Urinary catheterization can cause discomfort and intolerance of the procedure; some centres use
urinary convenes.6 When urinary catheterization is not used, the judicious use of fluids must be
considered.

Capnography whilst under a general anaesthetic is considered basic monitoring, but carbon
dioksida monitoring for patients who are sedated or awake during mapping is also common
practice. Although the carbon dioksida levels may be inaccurate, it is used to confirm ventilation.

Anaesthetic drug choice even within a preferred anaesthetic technique varies, but the general
principles are common to all; the need to maximize patient comfort, the prevention of nausea and
vomiting which can increase intracranial pressure, the need for haemodynamic stability, and the
use of short-acting drugs that allow acute control of a patient's conscious level.3

All patients will receive prophylactic antibiotics pre-incision, and usually one or more anti-
emetics. The most common choices are ondansetron, cyclizine, and dexamethasone.6
Dexamethasone may also be used to aid the condition of the brain intra-operatively. Intra-
operatively paracetamol and rarely non-steroidal anti-inflammatory drugs are used for analgesia.6

Previous SectionNext Section

Anaesthetic method
There is no recognized consensus on the best anaesthetic approach to an awake craniotomy.7 This
is often because the anaesthetist varies the technique dependent on the surgeon, pathology, length
of surgery, and patient factors. Some neurosurgical centres have developed local practice
pathways which allow selected patients to have an awake craniotomy as day surgery.

In a recent UK survey, 35% of anaesthetists reported their preferred anaesthetic approach as

asleep-awake, 35% preferred asleep-awake-asleep, and the remaining 30%; used an awake
throughout approach.6

Previous SectionNext Section

Sedation only awake throughout

The aim of this technique is to vary the levels of sedation according to the stage of surgery whilst
maintaining spontaneous ventilation without any airway device. Sedation is deepened during the
application of the Mayfield Pins, skin incision, removal of the bone flap, and dura mater. It is
then decreased or stopped for neurocognitive testing and mapping for resection of the lesion. The
sedation can then again be increased for closure.

If the level of sedation is not titrated correctly, there are risks of airway obstruction which can
lead to hypercapnia, hypoxia, and increased intracranial pressure with a tight brain, or
conversely a patient who is not adequately sedated will be uncomfortable and anxious.
The advantages of this technique are the avoidance of airway manipulation and its inherent risks.
Some studies state that the level of sedation during the crucial stage of intra-operative cortical
mapping is less and others have shown advantages in avoiding a general anaesthetic, for
example, decreased nausea and vomiting.4

Currently, within the UK, the most commonly used drugs in this setting are propofol and
remifentanil target-controlled infusions. There is also some use of clonidine infusions.6 Other
agents that are used include benzodiazepines, droperidol, and other short-acting opioids such as
fentanyl.8

Recently, dexmedetomidine has become available for use in the UK, and is routinely used in our
institution. It is a highly selective 2-receptor agonist and is unique in that it provides anxiolysis
and analgesia in addition to sedation without causing respiratory depression, even at very high
levels. It has anaesthetic sparing properties and does not have any effect on intracranial pressure.3
Patients are easily arousable despite sedation. However, it can cause hypotension and
bradycardia, which are dose-dependent.3

Dexmedetomidine is usually used as a sole agent and administered intravenously. Many studies
document the advantageous use of dexmedetomidine for awake craniotomies. A loading dose of
0.51.0 g kg1 over 20 min is then followed by an infusion rate of 0.20.7 g kg1 h1 depending
on the level of sedation required.9

Previous SectionNext Section

General anaesthetic asleep/awakeasleep

This technique involves induction of general anaesthesia and control of the airway with either a
supraglottic device or intubation. When neurocognitive testing and intra-operative mapping
needs to commence, the anaesthetic drugs are either reduced or stopped and the airway device is
removed, when the patient has regained upper airway reflexes and it is safe to do so. Once
resection of the lesion is complete, general anaesthesia can be re-introduced and with re-insertion
of the airway device.

The advantages of this technique include the ability to control ventilation and therefore control
carbon dioksida concentrations and prevent airway obstruction and hypoventilation. It also
facilitates greater depth of anaesthesia during the painful parts of the surgery.3,7

The anaesthetic drugs used for this technique are varied, but often are the same as those used in
the awake throughout technique. In the UK, propofol and remifentanil TCI are the most
common, followed by the use of a volatile anaesthetic and a remifentanil infusion.6 There has
been some reported use of dexmedetomidine in this technique, but usually for the awake stage of
the surgery and for closure.
The airway device most often used is the standard laryngeal mask (LMA); however others seek
the advantage of the gastric access and high pressure seal of the Proseal LMA or the integral
bite block of the Supreme LMA or the rigidity of the iGel.3 A minority, about 10%, of
surveyed anaesthetists in the UK, opt to intubate the patient.6

Controlled ventilation is most commonly used after the airway device is inserted, with the
advantages of preventing hypoventilation and hypercapnia. Neuromuscular blocking agents can
be used.6

Previous SectionNext Section

Scalp block
Patient tolerance of an awake craniotomy relies on effective analgesia of the surgical field, and
cannot rely on sedation or anaesthesia alone, especially as these are at a minimal level during
neurocognitive testing and intra-operative mapping. A scalp block also provides haemodynamic
stability and decreases the stress response to painful stimuli.

Occasionally, when sedation or general anaesthesia is not used, it is the scalp block that is used
as the sole technique.

Most anaesthetists will insert a bilateral scalp block before pinning of the head in Mayfield Pins.
Occasionally, a scalp block is not inserted and there is a reliance on the surgeon's local
anaesthetic infiltration.

The total local anaesthetic available to use with and without epinephrine must be calculated for
individual patients. Studies have shown that the increase in local anaesthetic levels of
levobupivicaine and ropivicaine are rapid compared with other regional blocks and similar in all
patients. Despite the rapid increase of plasma levels, there were no signs of cardiovascular or
central nervous system toxicity.10

The amount and type of local anaesthetic must be recorded and communicated with the surgeon
as they will often supplement the scalp block with further infiltration of the skin, temporalis
fascia and dura mater. The brain tissue is not painful to stimuli.

Bupivacaine, levobupivacaine, and ropivacaine of varying concentrations with and without

epinephrine have been described for use in a scalp block. The addition of epinephrine, usually
1:200 000, increases the total amount of local anaesthetic that can be used, decreases localized
bleeding, and maximizes the duration. However, systemic absorption may cause tachycardia and
hypertension and intra-arterial injection into the superficial temporal artery is possible when
blocking the auriculotemporal nerve.
The scalp block technique includes infiltrating local anaesthetic to seven nerves on either side.
This is an anatomical block, and not just a ring block. A ring block will require large volumes of
local anaesthetic, increases the risk of toxicity, and will not provide anaesthesia deep to the
temporalis fascia.4 At the end of the scalp block; further local anaesthetic can be infiltrated
locally to the pin sites.

Previous SectionNext Section

An approach to a scalp block3,4 (Fig. 2)

This block may be inserted with the patient sedated or after the induction of anaesthesia. The
skin is cleaned using either chlorhexidine or betadine. The amount inserted at each site will be
dependent upon the concentration and local anaesthetic used and whether it is mixed with
epinephrine. Using sterile gloves and a 23 gauge needle, the local anaesthetic is infiltrated into
the following sites:

Supraorbital nerve, a branch of the Trigeminal nerve, V1 distribution

View larger version:

In this page

In a new window

Download as PowerPoint Slide

Fig 2
Scalp innervation. Reproduced from Costello and Cormack4 with permission from Elsevier and
the authors.

It innervates the forehead, anterior part of the scalp, and top part of the head. Palpate the
supraorbital notch and insert needle perpendicularly and inject.

Supratrochlear nerve, a branch of the Trigeminal nerve, V1 distribution

It innervates the forehead and anterior part of the scalp. Just medial to the supraorbital nerve
injection site, above the eyebrow line, inject local anaesthetic to spread the block medially.
Zygomaticotemporal nerve, a branch of the Trigeminal nerve, V2
distribution
It innervates a small area of the forehead and the temporal area. The nerve passes through the
temporalis muscle to enter the temporalis fascia. Therefore, the local anaesthetic needs to be
infiltrated deep and superficial to the temporalis muscle. Infiltration begins at the lateral edge of
the supraorbital margin and continues to the distal aspect of the zygomatic arch.

Auriculotemporal nerve, a branch of the Trigeminal nerve, V3 distribution

It innervates the temporal areas, lower lip, lower face, auricle, and the scalp above the auricle.
Inject local anaesthetic about 1 cm anterior to the auricle, above the level of the
temporomandibular joint. This nerve crosses over the root of the zygomatic process of the
temporal bone and lies deep to the superficial temporal artery, which should be palpated to avoid
intra-arterial injection.

Lesser occipital nerve, a branch of the second or third cervical spinal

nerve
It ascends along the posterior border of the sternocleidomastoid muscle. It innervates the scalp in
the lateral area of the head posterior to the auricle. Infiltrate local anaesthetic subcutaneously
behind the auricle starting from the top-down to the auricular lobule and then continue to
infiltrate along the superior nuchal line to the greater occipital nerve.

Greater occipital nerve, a branch of the second cervical spinal nerve

It arises from the first and second cervical vertebrae. It ascends to innervate the skin along the
posterior part of the scalp. It can also innervate the scalp at the top of the head and over the
auricle. It is located by initially palpating the occipital artery, which is found about 34 cm
lateral to the external occipital protuberance along the superior nuchal line and then inject the
local anaesthetic, medial to the occipital artery.

Greater auricular nerve, a branch of the second and third cervical spinal
nerves
It is the largest of the ascending branches and emerges around the posterior border of the
sternocleidomastoid muscle. It divides into an anterior and a posterior branch and provides
sensory innervation for the skin over the parotid gland and mastoid process and the auricle. Inject
the local anaesthetic about 2 cm posterior to the auricle, at the level of the tragus.

Previous SectionNext Section

Neuropsychology/cortical mapping/resection
The patient is awoken for this stage and time is needed to allow the patient to adjust to the
environment to ensure that they are ready for clinical evaluation.
A calm, quiet environment is required for patient awakening. It is upon emergence that several
complications can occur, for example, pain from either the pins or discomfort from the prolonged
immobility, agitation or nausea, and vomiting. It is important to address these issues quickly and
effectively as it can lead to poor surgical conditions.

Cortical stimulation, also known as cortical or brain mapping, aims to localize the eloquent areas
of the brain through direct electrical stimulation of the cerebral cortex by electrodes. These areas
are those involved in speech, language, and motor abilities. Specifically, the Broca's area is
needed for speech production and language processing and the Wernicke's area is used for
language comprehension. It is also important to identify the motor and sensory cortex.3 Any
alteration of speech, language, and motor function by stimulation is communicated to the
surgeon.8 Resection takes place only after the cortex has been functionally mapped by this
process.

Previous SectionNext Section

Adverse incidents
Seizures, either focal or generalized, are most likely to occur during cortical mapping. They are
treated by irrigating the brain tissue with ice-cold saline. They usually cease with this treatment
alone, but occasionally benzodiazepines, anti-epileptic drugs, or re-sedation with airway control
are required.

An emergency plan for airway control has to be in place at all times and this can be challenging
as the patient's head is fixed in head pins and often away from the ventilator. The options include
the insertion of an LMA which may be easier than oro-tracheal intubation.

Awake craniotomy is generally a well-tolerated procedure with a low rate of conversion to

general anaesthesia and a low rate of complications. One of the most frequent complications is
patient intolerance of the procedure, often because of the urinary catheter or prolonged
positioning and intra-operative seizures.3,6 Table 2 lists intra-operative adverse incidents.

View this table:

In this window

In a new window

Table 2
Intra-operative adverse incidents

Previous SectionNext Section

Closure
Once resection is completed, the patient can be re-sedated or re-anaesthetized with re-insertion of
the airway device, even if in the lateral position. Closure of the dura mater, the bone flap, and the
scalp are then performed, the pins removed and the patient woken up. If remifentanil has been
used, it can be run at low infusion rates to aid a smooth wake-up and avoidance of coughing.

Previous SectionNext Section

Postoperative
Following an awake craniotomy the patient returns to either a neurosurgical ward or a high-
dependency unit bed.6 It is imperative that close neurological monitoring continues as
postoperative haematomas can develop, especially in the first 6 h after operating. This may
require an urgent repeat craniotomy for evacuation of the clot.

Some neurosurgical centres, including within the UK, carefully select appropriate patients for
day case awake craniotomy surgery. There are stringent inclusion and exclusion criteria and the
use of routine postoperative imaging for haematomas and access to advice after discharge if
needed. Most patients are in hospital for 12 days after operation.

After the scalp block has worn off, systemic pain relief is used. The use of postoperative pain
relief can be decreased in patients who have received a scalp block. Regular paracetamol and
opioids, such as codeine, morphine, or oxycodone, are used.

Previous SectionNext Section

Future developments
There is increasing evidence that an awake craniotomy would be an appropriate choice for
removal of all supratentorial tumours non-selectively. It can maximize lesion resection, which
can be linked to improved survival rates, and has low complication rates.7,11

The development of techniques that allow functional brain mapping are rapidly developing and
may supersede the awake craniotomy. For example intra-operative magnetic resonance (MR)
scanners are currently in use in some neurosurgical centres. In development are new technologies
that will enable preoperative brain mapping and these may have significant role in the future.3,4,12

(Anaesthesia for awake craniotomy)

Limited data are available on the influence of sedation for critical care therapy with the widely
used anesthetic propofol on recovery from acute traumatic brain injury. To establish the influence
of propofol on endogenous neurogenesis and functional recovery after traumatic brain injury, rats
were sedated with propofol either during or 2 hours after experimental traumatic brain injury.
Randomized controlled animal study.
University research laboratory.
One hundred sixteen male Sprague Dawley rats.
Mechanical brain lesion by controlled cortical impact.
This study investigated the dose-dependent influence of propofol (36 or 72 mg/kg/hr) either
during controlled cortical impact induction or in a delayed application protocol 2 hours after
experimental traumatic brain injury. Infusion of propofol resulted in 1) aggravation of
neurological dysfunction, 2) increased 28-day mortality rate, and 3) impaired posttraumatic
neurogenesis (5-bromo-2-deoxyuridine + NeuN-positive cells). Application of propofol during
trauma induction afforded a significant stronger effect in the high-dose group compared with
low-dose propofol. In the posttrauma protocol, animals were sedated with sevoflurane during the
controlled cortical impact injury, and propofol was given after an awake phase. In these animals,
propofol increased mortality rate and impaired neurological function and neurogenesis compared
with animals without delayed propofol anesthesia. The results show that propofol may prevent or
limit reparative processes in the early-phase postinjury. The results therefore indicate that
anesthetics may be potentially harmful not only in very young mammalians but also in adult
animals following acute cerebral injuries. The results provide first evidence for an altered
sensitivity for anesthesia-related negative effects on neurogenesis, functional outcome, and
survival in adult rats with brain lesions.24

PURPOSE OF REVIEW:

A major challenge in the treatment of brain-injured patients is the decision on indication and
timing of prophylactic anticoagulation. In addition, an increasing number of patients suffering
from traumatic brain injury (TBI) are on preinjury anticoagulation therapy. Despite clear
evidence for an increased risk of venous thromboembolic events and pulmonary embolism in
traumatized patients without prophylactic anticoagulation, there is a lack of distinct
recommendations and standardized clinical practice guidelines. This review summarizes current
research evidence regarding post-traumatic prophylactic anticoagulation and management of
patients with prehospital use of anticoagulants.

RECENT FINDINGS:
In addition to nonpharmacological techniques like compression stockings, use of low-dose
unfractionated heparin or low-molecular-weight heparin is effective in different studies in terms
of thromboprophylaxis. If follow-up computed tomography scans and clinical neurological
examinations do not show progression within 24h after initial evaluation, prophylactic
anticoagulation does not increase risk for hemorrhage progression and therefore seems to be safe
after TBI.

SUMMARY:
Stratification scores for identification of TBI patients with low, moderate, or high risk for
spontaneous cerebral bleeding may help to allow early thromboprophylaxis while maintaining a
good risk-benefit ratio. So far, these scores require validation by prospective trials. Therefore,
current evidence requires control computed tomography scans prior to early pharmacological
thromboprophylaxis.

(Anticoagulation in patients with traumatic brain injury.)