(Acta Anaesth. Belg.
, 2006, 57, 239-247)
Head trauma
V. BONHOMME (*), P. HANS (**) and J. Fr. BRICHANT (***)
Abstract : The aim of this review is to provide the read-
er with the most commonly accepted principles for the
management of head trauma patients. The initial clinical
evaluation and resuscitation, radiological evaluation,
monitoring, intracranial pressure and cerebral perfusion
pressure management, brain protection, associated organ
dysfunctions and complications, anaesthetic manage-
ment and the singularities of paediatric head trauma
patients are described, either for the acute phase and the
secondary phase of management.
Key words : Head trauma ; management ; review.
Optimal management of head trauma patients
is challenging. Good knowledge of brain physiolo-
gy and traumatic brain injury physiopathology is
essential to successfully manage patients with head
trauma. Success or failure depends on several fac-
tors including the initial severity of brain and asso-
ciated lesions and their adequate clinical evaluation,
efficient and non harmful early resuscitation, avail-
ability of a multidisciplinary neuro-trauma centre,
and prevention and early detection of complica-
tions. In this paper, the currently accepted general
principles governing the management of head trau-
ma patients are reviewed, either for the acute phase
and for the secondary phase of management.
However, many of the frequently proposed thera-
pies are not supported by class I evidence in the lit-
erature. Their benefit still need to be demonstrated
by large randomized controlled trials (1). The initial
evaluation, making decision regarding orientation
towards a neuro-trauma centre, early resuscitation,
radiological evaluation, monitoring modalities,
intracranial pressure (ICP) and cerebral perfusion
pressure (CPP) management, brain protection, non-
neurological organ dysfunction of central origin,
indications for surgery, anaesthetic management,
and paediatric head trauma patient management
will be described.
INITIAL CLINICAL EVALUATION AT THE ACUTE PHASE OF
MANAGEMENT
When taking care of head trauma patients on
the scene of the accident, the very first step consists
of a rapid and efficient evaluation of the severity of
brain aggression and associated lesions. This initial
evaluation is of utmost importance for guiding
patient orientation and future therapeutic decisions.
The most widely used clinical neurological evalua-
tion is the Glasgow Coma Score (GCS) associated
to the determination of pupil size and reactivity to
light (2). It is based on the best eye (1 = no eye
opening, 2 = opening to pain, 3 = opening to verbal
command, 4 = spontaneous opening), verbal (1 =
no verbal response, 2 = incomprehensible sounds,
3 = inappropriate words, 4 = confused response, 5 =
orientated response) and motor (1 = no motor
response, 2 = extension to pain, 3 = flexion to pain,
4 = withdrawal from pain, 5 = orientated response
to pain, 6 = obeys to command) responses of the
patient to stimulation (3). Over a maximal score of
15, mild brain injury corresponds to a GCS
13,
moderate to a GCS between 9 and 12 whereas
severe brain injury is defined as a GCS  8. The
initial GCS has a prognostic value and serves as a
reference for subsequent evaluation. It must be
evaluated after the initial correction of vital func-
tions. The motor part of the score is the most perti-
nent. Several other neurological symptoms may
also help in evaluating the patient. A brainstem
lesion can be suspected in the absence of fronto-
orbicular, oculo-cephalic, oculo-vestibular and
oculo-cardiac reflexes (4). Caution is advised when
looking for those reflexes. Moving the head can be
V. BONHOMME ; P. HANS ; J. Fr. BRICHANT.
(*) Chef de Clinique associé, Service Universitaire
d’Anesthésie-Réanimation, CHR de la Citadelle, CHU de
Liège, Bd du 12ème de Ligne, 1, 4000 Liège, Belgium.
(**) Chef de Service, Service Universitaire d’Anesthésie-
Réanimation, CHR de la Citadelle, CHU de Liège, Bd du
12ème de Ligne, 1, 4000 Liège, Belgium.
(***) Chef de Service associé, Service Universitaire
d’Anesthésie-Réanimation, CHR de la Citadelle, CHU de
Liège, Bd du 12ème de Ligne, 1, 4000 Liège, Belgium.
Correspondence address : V. Bonhomme, Service Univer-
sitaire d’Anesthésie-Réanimation, CHR de la Citadelle,
Bd du 12ème de Ligne, 1, 4000 Liège, Belgium.
E-mail : vincent.bonhomme@chu.ulg.ac.be.
© Acta Anæsthesiologica Belgica, 2006, 57, n° 3
240 V. BONHOMME
dangerous in case of associated cervical spine
lesion. Pupil size and reactivity to light anomalies
may not be easy to interpret, particularly when the
patient is intoxicated with alcohol, cocaine,
amphetamines or other psychotropic substances, or
in case of orbital trauma.
Special attention should be paid to concomi-
tant lesions to other parts of the body, namely the
face, the cervical spine, the thorax, the abdomen, as
well as the upper and lower limbs. Lesions to the
face and mouth may render tracheal intubation dif-
ficult. Concomitant lesions of the cervical spine are
very common in head trauma patients. Management
of such lesions will not be detailed here and inter-
ested readers are referred to our previously pub-
lished review paper (5). Lesions to the thorax and
the abdomen can be life threatening by compromis-
ing ventilation (pneumo-hemothorax, costal frac-
tures), and haemodynamic stability (heart failure
caused by myocardial contusion, tamponade, aortic
isthmus rupture, silent intra-abdominal bleeding).
Trauma to the limbs and wounds to the scalp can
also cause severe bleeding.
In each case, the initial evaluation of the head
trauma patient must be fast and efficient in order to
avoid delaying early resuscitation. Early resuscita-
tion and adequate decision on the need for hospital
admission is essential. Minimal criteria for non-
delayed hospital admission include transient or per-
sisting loss of consciousness, post-traumatic amne-
sia, persisting nausea and vomiting, or difficulty
with assessment (alcohol ...) (6).
EARLY RESUSCITATION AND MANAGEMENT ON THE SCENE
OF TRAUMA
The goal here is the restoration of vital func-
tions. Early resuscitation always start with the so
called ABC’s. There is a consensus to say that all
severe brain trauma patients (GCS  8) must be
mechanically ventilated, as well as those with asso-
ciated facial injuries (7). Management of the airway
can reveal difficult, particularly in patients with
associated cervical spine and facial injuries, and
careful sedation is often required. A rapid sequence
induction is mandatory and necessitates skilled care
providers (8). Stability of the cervical spine must
always be insured and cricoid pressure will be
applied using a two-hand technique (5).
The choice of anaesthetic medication will
depend on their potential side effects and on the
experience of the practitioner. It must always be
kept in mind that, outside the hospital and before
© Acta Anæsthesiologica Belgica, 2006, 57, n° 3
et al.
ICP can be monitored, cerebral perfusion pressure
must be maintained at all costs. It means that any
comatose patient will be considered as a patient
with raised ICP and, therefore, a need for maintain-
ing adequate CPP.
Possible medications to induce anaesthesia
include propofol, barbiturates, etomidate, ketamine
or benzodiazepines. Propofol and barbiturates
reduce the cerebral metabolic rate, and hence ICP.
They also have neuroprotective properties (9).
However, they may induce profound hypotension in
hypovolaemic patients. Their use will therefore be
limited to those patients whose volaemia is sup-
posed to be adequate or after adequate fluid resus-
citation. The same is true, although to a lesser
extent, for benzodiazepines. Etomidate is a good
first choice, as it has brain protective effects and
does not induce hypotension. However, it can not be
used for maintenance of anaesthesia due to its
depressing effects on endogenous adrenal secre-
tions. Ketamine is brain protective, has strong anti-
nociceptive properties and helps maintaining
haemodynamic stability. However, it increases the
cerebral metabolic rate and increases ICP. Caution
should therefore be paid in patients suspect of
increased ICP. However, this effect is strongly
attenuated by the concomitant use of propofol or
barbiturates.
Analgesia should be insured using opioid
derivatives. It is legitimate to use either alfentanil,
sufentanil or fentanil. However, one must keep in
mind that intermittent neurological examination
will be necessary during the next hours. Opioid
derivatives poorly alter the motor response to stim-
ulation, but may alter consciousness at high doses.
Choosing a medication with unfavourable pharma-
cokinetic properties may impede those evaluations.
It is also the reason why remifentanil may be pre-
ferred in some instances, although the haemo-
dynamic consequences can be more marked when
using this medication than when using other
opioids (9).
Neuromuscular blocking agents are useful to
facilitate tracheal intubation. Although it induces
muscular fasciculation and increases ICP, succinyl-
choline may be of great help when intubation is
expected to be difficult. The alternative is the use of
a high dose of rocuronium (0.9 mg kg-1). In that
case, prolonged muscle relaxation may be danger-
ous if tracheal intubation is impossible. Further-
more, it may delay reliable neurological evaluation.
The upcoming availability of the new steroid neuro-
muscular blocking agents antagonist, cyclodextrin,
may soon overcome the problem of prolonged
 HEAD TRAUMA
neuromuscular blockade (10). However, studies are
still needed to determine the effect of cyclodextrin-
induced neuromuscular blockade antagonism on
ICP.
Other agents may help reducing the sympa-
thetic response to tracheal intubation, and hence the
raise in ICP at that time. In that respect, lidocaine at
the dose of 1.5 mg kg-1 has been shown to attenuate
the cardiovascular, coughing and ICP responses to
intubation (11).
Maintenance of sedation to allow for trans-
portation to the trauma care centre can be per-
formed using a continuous infusion of propofol or
midazolam and sufentanil or remifentanil, once
haemodynamic stability is achieved. Halogenated
inhaled anaesthetic agents are not easy to use out-
side the operating theatre. Some of them have
vasodilating properties on the cerebral vasculature
and hence raise ICP. For those reasons, they are not
frequently used for sedation of head trauma
patients. Due to its effects on ICP and its potential
neurotoxic effects, nitrous oxide should always be
avoided (9). Intermittent muscle relaxation may be
necessary, particularly if coughing on the tracheal
tube cannot be alleviated by the intravenous anaes-
thetic regimen. Non-depolarising intermediate
duration agents such as atracurium, cis-atracurium
or rocuronium will be used. Continuous muscle
relaxation is not recommended, as it impede neuro-
logical evaluation. Long term muscle relaxation
may also have neuromuscular trophic conse-
quences.
Beside clinical signs of brain herniation such
as anisocoria and Cushing’s reaction (bradycardia
and hypertension), no real signs of intracranial
hypertension can be recognised in a ventilated and
sedated patient. The clinician will therefore be
mainly guided by the initial GCS. Moderate hyper-
ventilation to a target end-tidal CO2 partial pressure
between 30 and 35 mmHg will be the rule until
brain imaging and/or direct measurement can rule
out increased ICP. Indeed, prolonged hypocapnia
has a limited duration of action on ICP and can be
deleterious for the brain as it can favour brain
ischaemia through excessive vasoconstriction (12).
Mean blood pressure will be maintained at least
above 70 mmHg, ideally 90 mmHg, through fluid
administration and, perhaps, the use of vasopres-
sors. At least isotonic fluids should be used, as
hyponatremia can be very detrimental for the
injured brain (13). Only dramatic situations of brain
herniation will require the infusion of mannitol
(0.25 to 1 g kg-1), hypertonic saline, furosemide
and/or barbiturates, while preserving perfusion
241
pressure. Prone position of the head to 30° is bene-
ficial to reduce ICP. Peripheral saturation in oxygen
will be maintained in the normal range, at least
above 95% and the stomach will be emptied using a
gastric tube. Caution is necessary when inserting
this probe if a skull base fracture is suspected (“sun-
glasses” haematoma). In that case, insertion
through the mouth, under laryngoscope-directed
vision is preferable.
BRAIN LESION IMAGING STRATEGY
Upon arrival in the trauma centre, it is neces-
sary to assess the severity of lesions. Beside imag-
ing of concomitant lesions to other parts of the body
(cervical spine, thorax, abdomen, limbs), a brain
CT scan must be performed as soon as possible
when GCS is low, in case of neurological deficits,
epileptic seizure, skull fracture and cerebrospinal
fluid (CSF) leak (6). It will allow identification of
intracranial lesions and may indicate the need for
surgical intervention (extra-dural, sub-dural or
intra-cerebral haematoma, brain contusion, acute
hydrocephaly, depressed skull fracture, open frac-
ture and penetrating lesions). Intracranial hyperten-
sion can be detected through visualisation of brain
oedema, mass lesion, reduced cistern size and mid-
line shift. The early brain CT scan has a strong
prognostic value (2).
MONITORING DURING THE ACUTE AND THE SECONDARY
PHASE OF MANAGEMENT
Basic monitoring
Basic monitoring including electrocardio-
gram, non-invasive blood pressure and peripheral
oxygen saturation must be started immediately.
End-tidal CO2 monitoring can be of great help to
adjust ventilation in mechanically ventilated
patients, allowing the indirect control of arterial
CO2 concentration. Invasive arterial blood pressure
monitoring will often be started as soon as possible,
that is once the patient is admitted to the emergency
care unit. This monitoring will allow faster detec-
tion of episodes of hypo- or hypertension, tight con-
trol of mean blood pressure (and hence of cerebral
perfusion pressure), guiding fluid administration
and the use of vasoactive medications, and perform-
ing blood gas analyses. Temperature and urine out-
put monitoring are also recommended.
Advanced monitoring techniques can be used
during the secondary phase of trauma patient care,
© Acta Anæsthesiologica Belgica, 2006, 57, n° 3
242 V. BONHOMME
either to guide therapeutic measures or to help in
evaluating prognosis.
ICP monitoring
Decision to monitor ICP depends in part on
CT results (14). Generally speaking, ICP monitor-
ing is required in patients with an initial GCS  8.
However, it is also recommended in patients with a
GCS
8 and severe brain oedema, reduced cistern
size, severe temporal lobe contusion, midline shift
and obliteration of the third ventricle. Conversely,
patients with a GCS  8 but a normal CT will
require ICP monitoring if they are more than
40 years of age, they display motor deficits or a sys-
tolic blood pressure lower than 90 mmHg (15). In
many instances, neurosurgeons favour ICP meas-
urement through an intra-ventricular catheter. This
method is accurate, allows CSF drainage, and recal-
ibrations. However, placement of those catheters
necessitates the environment of an operating theatre
and may be associated with infectious complica-
tions or lesions of brain parenchyma (16). It has
been known for a long time that the number of high
ICP episodes strongly determines the outcome of
head trauma patients (17). Combined to mean blood
pressure monitoring, it allows controlling the CPP.
A detailed description of the goals to achieve
regarding ICP and CPP management is provided
hereafter. Other ICP monitoring devices are avail-
able such as the subarachnoid bolts and intra-
parenchymal fiberoptic devices. Subarachnoid bolts
allow CSF drainage, but are less accurate.
Fiberoptic devices do not allow CSF drainage and
may also damage brain tissue.
Jugular venous oxygen saturation
Jugular venous oxygen saturation (SjO2) mon-
itoring can be instituted in severe cases to help for
therapeutic decisions. It necessitates the retrograde
insertion of a catheter into the jugular bulb. Usually,
this is not feasible in the emergency care unit and
will only be performed once the patient is admitted
to the intensive care unit. SjO2 reflects the global
balance between O2 delivery to the brain and its
metabolic needs. Normal SjO2 values range
between 55 to 71% (18). As a sustained SjO2 value
below 50% is associated to cerebral ischaemia, the
practitioner will adapt CPP, O2 delivery (cardiac
output, haemoglobin, arterial O2 partial pressure)
and cerebral metabolic rate to maintain SjO2 into
the normal range. This monitoring will also help
detecting vasospasm and deleterious effects of
© Acta Anæsthesiologica Belgica, 2006, 57, n° 3
et al.
hyperventilation. Noteworthy, excessive and pro-
longed hyperventilation can be harmful to localised
brain regions, and, as SjO2 reflects the global situa-
tion of the brain, this deleterious effect will not be
detected by SjO2 monitoring (19).
Cerebral blood flow monitoring
Global or regional cerebral blood flow (CBF)
can be measured intermittently using several tech-
niques such as the Kety-Schmidt method, xenon
dilution (20), jugular thermodilution, single-photon
emission scan, xenon-enhanced scan, perfusion
scan, magnetic resonance angiography, and
positron emission tomography. These techniques
are not widely used in daily practice, essentially
because they are too complex. A step further, the
study of regional CBF distribution using positron
emission tomography may have prognostic value in
severe brain-injured patients such as vegetative or
minimally conscious patients (21). Transcranial
doppler (TCD) does not allow measuring CBF and
is essentially helpful to appreciate vasospasm, the
consequences of elevated ICP/low CPP and carotid
dissection. Through detection of cerebral circulato-
ry arrest, it is also useful for the diagnosis of brain
death (18).
Electrophysiological monitoring
Continuous monitoring of the electroen-
cephalogram (EEG) is necessary to evaluate barbi-
turate-induced coma in severely elevated ICP
patients. In that case, barbiturate infusion is target-
ed to obtain a flat or quasi-flat EEG. EEG monitor-
ing also allows detecting non-convulsive seizure
activity, which requires immediate therapeutic
intervention. Other electrophysiological monitoring
modalities can also be used. For example, the
Bispectral Index (BIS) allows individual titration of
sedation, detection of seizure activity and guidance
of barbiturate coma (22, 23). It can predict the prob-
ability of return to consciousness once sedation has
been withdrawn (24). Somatosensory-evoked
potentials are good predictors of outcome (25).
Brain oxygenation and metabolism
Brain oxygenation and metabolism, either
global or regional, can be assessed through several
monitoring modalities. Near infrared spectroscopy
has been proposed in that respect but seems to be
less reliable than other techniques (26). The inva-
sive cerebral tissue oxygen monitoring, although
 HEAD TRAUMA
more invasive, offers a promising alternative and
would be able to detect ischemic or hyperaemic
episodes (27-29). Cerebral microdialysis measures
metabolic substrates (glucose, lactate, pyruvate,
adenosin or xanthin), neurotransmitters (glutamate,
aspartate, GABA), cellular death witnesses
(potassium, glycerol), and exogenous substances
(medications). So far, evidence has not granted this
technique as a clinical tool (30, 31).
INTRACRANIAL PRESSURE AND CEREBRAL PERFUSION
PRESSURE
CPP maintenance is a determinant factor of
outcome in severely brain-injured patients (17).
CPP is the difference between mean arterial blood
pressure (MABP) and ICP. Normal values range
between 70 and 85 mmHg. In the normal brain, the
ischemic threshold of CCP is considered to be
50 mmHg. When pressure autoregulation of CBF is
altered, this threshold may shift to higher or lower
values. Intense debate has occurred on determining
the ideal CPP level to be targeted (32), and it is not
easy to define an optimal CPP threshold.
Maintaining CPP at too high levels (> 70 mmHg)
exposes to the risk of hyperaemia and raised ICP,
while maintaining it at too low levels to the risk of
ischemia. The optimal threshold will therefore be
defined on an individual basis, using estimates of
the adequacy of O2 delivery to the brain such as
SjO2, near infrared spectroscopy, invasive brain tis-
sue oxygenation monitoring or microdyalisis.
CPP can be controlled through modifications
of the two components of its equation, namely ICP
and MABP. MABP is easily modified using fluid
infusion and/or vasoactive medications (e.g. levore-
nine, dobutamine). Caution is advised concerning
the administered amount of fluids, as excess fluid
may lead to pulmonary oedema or acute respiratory
distress syndrome. Indeed, brain trauma patients are
prone to develop such problems (33). According to
the Monro-Kellie principle (any raise in the content
of the rigid skull box is associated to a pressure
increase), the reduction of ICP can be achieved
through reducing brain size (osmotherapy), the sur-
gical removal of a mass, reducing CSF volume
(drainage), reducing the blood volume of the brain
(hyperventilation, sedation, prone position), or
opening the rigid box (decompressive craniecto-
my). The therapeutic gradation can be schematized
as follows : moderate hyperventilation (PaCO2 at
35 mmHg first, 30 if not sufficient), CSF drainage,
0.25 to 0.5 g kg-1 mannitol (up to 2 g kg-1) (34) or
243
hypertonic saline (35) (150 mg kg-1) (intermittent
boluses every 4 hours, monitor natremia, plasma
osmolarity, and renal function), furosemide (20 mg
boluses, maintain volaemia), EEG-guided barbitu-
rate administration, decompressive craniectomy
and hypothermia (2). Decompressive craniectomy
seems to improve outcome in severe head trauma
patients with raised ICP that is refractory to other
treatments, although prospective studies are still
needed to confirm the role of this therapeutic meas-
ure and its indications (36, 37). Pentobarbital-
induced coma should be started using a loading
dose of 5-10 mg kg-1 administered over 30 minutes.
Maintenance can be achieved using a continuous
infusion at the rate of 1-3 mg kg-1 h-1, and titration
should be performed to obtain EEG-burst suppres-
sion and serum pentobarbital levels between 3 and
4 mg % (22). The maintenance of haemodynamic
stability is mandatory.
PREVENTION OF SECONDARY BRAIN DAMAGE AND BRAIN
PROTECTION
The practitioner in charge of traumatic brain-
injured patients must always keep in mind the need
for preventing aggravation of the initial brain
lesions. The prevention of secondary brain damage
and brain protection occurs through two main guid-
ing principles : the prevention of secondary brain
damage of systemic origin (secondary brain aggres-
sion of systemic origin, SBASO), often referred to
as passive neuroprotection, and the instauration of
brain protecting therapies, or active neuroprotec-
tion. The prevention of SBASO relies on the main-
tenance of homeostasis. Any episode of hypoxia
(ventilation problems, pulmonary oedema, ...),
anaemia (concomitant haemorrhage), hypo- or
excessive hypertension, hyperglycaemia
(> 8.33 mmol l-1), hypo- or hypernatremia and
hyperthermia should be avoided or treated as fast as
possible (38). Seizure activity is common and pro-
phylaxis is recommended (phenytoin, valproate).
Direct brain protecting therapies are scarce,
although numerous laboratory investigations have
evidenced several possible therapeutic measures.
Direct proofs of the efficiency of those measures for
improving outcome in humans are not easy to
obtain. Several anaesthetic agents such as barbitu-
rates, propofol, halogenated compounds, xenon,
ketamine, magnesium and lidocaine have theoreti-
cal protective effects on the injured brain through
their action at various levels of the secondary neu-
ronal damage cascade, including apoptosis for
© Acta Anæsthesiologica Belgica, 2006, 57, n° 3
244 V. BONHOMME
some of them (38). However, the question of the
best agent to use is not resolved. Beside the anaes-
thetic armamentarium, most of the promising drugs
evidenced by laboratory and animal experimenta-
tion have revealed disappointing in clinical practice.
Research is still in progress for the most promising
ones such as magnesium, calcium channel blockers,
NMDA antagonists, anti-inflammatory medica-
tions, anti-proteases, free radical scavengers,
immunomodulators and neurotrophic factors. Other
tools include preconditioning, hyperoxia and
hypothermia. Preconditioning consists in exposing
the brain to minor insults (ischaemia, hypoxia,
hyperoxia ...) or to medications (sevoflurane, ery-
thropoietin, ...) that induce an increased tolerance to
further aggression. Although promising, clinical
evidence is still too poor to recommend their use
routinely. This is also true for eubaric hyper-
oxia (39). Finally, the National Acute Brain Injury
Study on hypothermia (NABISH) (40) has not per-
mitted to conclude that hypothermia is beneficial
following a traumatic brain injury, although it is
recommended not to rewarm TBI patients that are
hypothermic on admission. Noteworthy, hypother-
mia remains beneficial following cardiac arrest (41)
and may be interesting to control for intractable
raised ICP. It is now commonly admitted that
steroids have no place for brain protection in trau-
matic brain injury (2).
NEUROENDOCRINE AND NON-NEUROLOGICAL ORGAN DYS-
FUNCTIONS OF CENTRAL ORIGIN
Beside direct effect on cerebral function, brain
trauma may induce several neuroendocrine and non-
neurological organ dysfunctions. Among them, inap-
propriate anti-diuretic hormone secretion syndrome
(IADHS), diabetes insipidus, hypopituitarism, neu-
rogenic pulmonary oedema, and the dysautonomic
storm syndrome are the most common.
Inappropriate antidiuretic hormone secretion syn-
drome
IADHS is mainly characterised by the occur-
rence of hyponatremia that cannot be attributed to
another cause (hypotonic fluid infusions, renal salt
wasting syndrome, excessive osmotic or diuretic-
enhanced diuresis) and may be treated using fluid
intake restriction and, eventually, arginin vaso-
pressin-receptor antagonists (42). Hypertonic saline
infusion to correct hyponatremia may be danger-
ous, particularly in more than 48 hours hyponatrem-
ic patients, as it can induce central demyalination.
© Acta Anæsthesiologica Belgica, 2006, 57, n° 3
et al.
Diabetes insipidus
Diabetes insipidus is suspected in patients pre-
senting high volume hypotonic diuresis (urine den-
sity < 1010) that do not respond to fluid restriction.
Treatment consists of desmopressin subcutaneous
administration (2-4 µg every 12 hours).
Hypopituitarism
The incidence of various degrees of hypopitu-
itarism following TBI may be as high as 50%, and
this pathology may concern any of the hypothala-
mo-hypophyso-peripheral hormonal axis (43).
Systematic screening of pituitary function is recom-
mended for all patients with moderate to severe
TBI (44). Treatment consists of hormonal replace-
ment.
Neurogenic pulmonary oedema and dysautonomic
syndrome
The origin of the neurogenic pulmonary oede-
ma is not known with precision and may be due to
excessive adrenergic activity and/or cardiac failure,
and would therefore be a consequence of the dysau-
tonomic storm syndrome (45). This syndrome is
characterised by several neurological symptoms,
hypertension, hyperthermia, and tachycardia. These
elements may have cardiac and neurological conse-
quences (myocardial infarct, cardiac failure, raised
ICP). The treatment is symptomatic and may
require pulmonary artery catheterism, as well as
echocardiographic monitoring.
ADDITIONAL CONCERNS OF THE SECONDARY PHASE OF
MANAGEMENT
Among the other concerns related to the man-
agement of head trauma patients, sepsis, hyperca-
tabolism, stress ulcer, and thromboembolic events
are the most common. TBI patients are immuno-
compromised and sepsis is frequent, particularly of
respiratory origin (46). Increased caloric and pro-
tein requirements of head trauma patients should be
met without delay, using the following formula :
RME = 152 – 14(GCS score) + 0.4(HR) + 7(DSI),
where RME is the resting metabolic expenditure,
GCS is the Glasgow coma scale, HR is the heart
rate and DSI is the number of days since the
injury (47). A stress ulcer prophylaxis should be
immediately started using either type II histaminic
 HEAD TRAUMA
receptors antagonists or hydrogen ion pump
inhibitors. Antithrombotic prophylaxis should also
be started using stocking or intermittent calf com-
pression, as well as low molecular weight heparins.
Caution should be paid to the risk of intracranial
bleeding in case of haemorrhagic traumatic
lesions (48, 49).
ANAESTHETIC MANAGEMENT WHEN SURGERY IS REQUIRED
Anaesthetic management, either for intracra-
nial or peripheral surgery, is again governed by the
prevention of SBASO and the protection of the
brain as much as possible (34). Surgical priorities
will be determined on a case by case basis, the con-
trol of menacing bleeding often being the first con-
cern. Large venous access is mandatory in those
conditions. A central venous catheter may be of
help to guide fluid administration but is not manda-
tory, and must not delay surgery. Beside classical
anaesthetic monitoring, invasive arterial blood pres-
sure monitoring will often be necessary. Bladder
catheterisation and temperature monitoring will be
instituted. The management of ICP and CPP during
surgery will be easier if ICP can be directly moni-
tored, but the placement of the ICP monitoring
device is not always possible before or at the begin-
ning of surgery. Throughout the procedure, nor-
moxia, normothermia, and normoglycaemia will be
maintained, and end-tidal PCO2 will be in the range
of 30-35 mmHg. Haemoglobin concentration
and/or haematocrit will be checked at regular inter-
vals. If an ICP monitoring device is available, the
management of arterial blood pressure will target a
CPP of 70 mmHg through adjustments of MABP
(fluids, levorenine) or ICP (drainage, mannitol,
lasix, hypertonic saline). Otherwise, MABP should
be maintained at 90 mmHg as much as possible if
raised ICP is suspected. If the patient is not already
intubated and sedated upon arrival in the operating
theatre, a rapid sequence induction is the rule, using
the same technique as the one described above. The
same rules also apply for the choice of anaesthetic
agents used to maintain anaesthesia. Caution should
be paid to the adequate positioning of the head, alle-
viating jugular compression and favouring a 30°
prone position.
PAEDIATRIC BRAIN TRAUMA
Specific concerns relate to the management of
paediatric head trauma patients. The volume of the
245
head compared to the volume of the body is propor-
tionally more important in children than in adults.
The physical mechanism responsible for brain
lesions in children is therefore often related to high
energy deceleration. Diffuse axonal lesions and
generalised oedema are more common than intrac-
erebral haematoma and contusion, with the conse-
quence that children will present more frequently
with an altered state of consciousness and seizures
rather than focal deficits (50). It is worth to note
that this decelerative mechanism of injury will often
be responsible for cervical spinal cord injuries with-
out radiological abnormalities (SCIWORA). The
skull of children is also immature and is more prone
to fractures than the adult. It is also more compliant
when sutures and fontanels are not closed. The neu-
rosurgical emergency to be fear of in children is the
extradural haematoma, which necessitates immedi-
ate surgical drainage. It is frequently associated to
parietal skull fracture, but not in all cases. It must be
suspected in children whose neurological status
deteriorates rapidly : an aggravating cephalalgy, nau-
sea and vomiting and progressive stupor must warn
the clinician. Hemiparesia and anisocoria are signs
of temporal herniation, which jeopardizes the imme-
diate vital prognosis. Those extradural haematomas
may also manifest in the form of a hypovolaemic
shock in very young children. Traumatic subdural
haematomas are frequent in 5 month children (shak-
en babies), in which the subdural spaces are large
and the brain more mobile. The management princi-
ples of head trauma in children are the same as those
described for the adult (51), except that the target
CPP to be maintained is lower and ranges between
40 and 50 mmHg instead of 70 (50).
CONCLUSIONS
The management of head trauma patients is
challenging and requires tight collaboration
between emergency, intensive care, radiology,
anaesthesiology and neurosurgery practitioners.
Careful initial evaluation, efficient early resuscita-
tion, targeted imaging, rapid surgical interventions,
and purposeful advanced life support are determi-
nant factors of the outcome.
References
1. Adamides A. A., Winter C. D., Lewis P. M., Cooper D. J.,
Kossmann T., Rosenfeld J. V., Current controversies in the
management of patients with severe traumatic brain injury,
ANZ J. SURG., 76, 163-174, 2006.
© Acta Anæsthesiologica Belgica, 2006, 57, n° 3
246 V. BONHOMME
2. Vincent J. L., Berre J., Primer on medical management of
severe brain injury, CRIT. CARE MED., 33, 1392-1399,
2005. 23.
3. Teasdale G., Jennett B., Assessment of coma and impaired
consciousness. A practical scale, LANCET, 2, 81-84, 1974.
4. Born J. D., Albert A., Hans P., Bonnal J., Relative prognos-
tic value of best motor response and brain stem reflexes in 24.
patients with severe head injury, NEUROSURGERY, 16, 595-
601, 1985.
5. Bonhomme V., Hartstein G., Hans P., The cervical spine in
trauma : implications for the anaesthesiologist, ACTA
ANAESTHESIOL. BELG., 56, 405-411, 2005. 25.
6. Bavetta S., Benjamin J. C., Assessment and management of
the head-injured patient, HOSP. MED., 63, 289-293, 2002.
7. Chesnut R. M., Management of brain and spine injuries, 26.
CRIT. CARE CLIN., 20, 25-55, 2004.
8. Fakhry S. M., Scanlon J. M., Robinson L., Askari R.,
Watenpaugh R. L., Fata P., Hauda W. E., Trask A.,
Prehospital rapid sequence intubation for head trauma :
conditions for a successful program, J. TRAUMA, 60, 997- 27.
1001, 2006.
9. Albanese J., Bonhomme V., Hans P., Les agents
anesthésiques hypnotiques, les opiaces et les myorelaxants 28.
en neuroanesthesie. In : Ravussin P. Le Patient Neuro-
chirurgical, 1, p. 33. Paris, Springer, 2004.
10. Sorgenfrei I. F., Norrild K., Larsen P. B., Stensballe J.,
Ostergaard D., Prins M. E., Viby-Mogensen J., Reversal of 29.
rocuronium-induced neuromuscular block by the selective
relaxant binding agent sugammadex : a dose-finding and
safety study, ANESTHESIOLOGY, 104, 667-674, 2006. 30.
11. Roppolo L. P., Walters K., Airway management in neuro-
logical emergencies, NEUROCRIT. CARE, 1, 405-414, 2004.
12. Kelly D. F., Becker D. P., Advances in management of neu- 31.
rosurgical trauma : USA and Canada, WORLD J. SURG., 25,
1179-1185, 2001.
13. Soreide E., Deakin C. D., Pre-hospital fluid therapy in the 32.
critically injured patient – a clinical update, INJURY, 36,
1001-1010, 2005.
14. Newberg A. B., Alavi A., Neuroimaging in patients with 33.
head injury, SEMIN. NUCL. MED., 33, 136-147, 2003.
15. The Brain Trauma Foundation. The American Association
of Neurological Surgeons. The Joint Section on
Neurotrauma and Critical Care. Recommendations for 34.
intracranial pressure monitoring technology, J. NEURO-
TRAUMA, 17, 497-506, 2000.
16. Fahy B. G., Sivaraman V., Current concepts in neurocriti-
cal care, ANESTHESIOL. CLIN. NORTH AMERICA, 20, 441-62, 35.
viii, 2002.
17. Miller J. D., Becker D. P., Ward J. D., Sullivan H. G., 36.
Adams W.E., Rosner M.J., Significance of intracranial
hypertension in severe head injury, J. NEUROSURG., 47, 503-
516, 1977.
18. Haddad S., Arabi Y., Al Shimemeri A., Intensive care man- 37.
agement of traumatic brain injury, MIDDLE EAST J.
ANESTHESIOL., 18, 69-84, 2005.
19. Imberti R., Bellinzona G., Langer M., Cerebral tissue PO2
and SjvO2 changes during moderate hyperventilation in 38.
patients with severe traumatic brain injury, J. NEUROSURG.,
96, 97-102, 2002. 39.
20. Cook D. J., Anderson R. E., Michenfelder J. D.,
Oliver W. C., Jr., Orszulak T. A., Daly R. C., Bryce R. D., 40.
Cerebral blood flow during cardiac operations : compari-
son of Kety-Schmidt and xenon-133 clearance methods,
ANN. THORAC. SURG., 59, 614-620, 1995.
21. Laureys S., Faymonville M. E., Luxen A., Lamy M.,
Franck G., Maquet P., Restoration of thalamocortical con-
nectivity after recovery from persistent vegetative state, 41.
LANCET, 355, 1790-1791, 2000.
22. Bader M. K., Arbour R., Palmer S., Refractory increased
intracranial pressure in severe traumatic brain injury :
© Acta Anæsthesiologica Belgica, 2006, 57, n° 3
et al.
barbiturate coma and bispectral index monitoring, AACN
CLIN. ISSUES, 16, 526-541, 2005.
Deogaonkar A., Gupta R., DeGeorgia M., Sabharwal V.,
Gopakumaran B., Schubert A., Provencio J. J., Bispectral
Index monitoring correlates with sedation scales in brain-
injured patients, CRIT. CARE MED., 32, 2403-2406, 2004.
Fabregas N., Gambus P. L., Valero R., Carrero E. J.,
Salvador L., Zavala E., Ferrer E., Can bispectral index
monitoring predict recovery of consciousness in patients
with severe brain injury ?, ANESTHESIOLOGY, 101, 43-51,
2004.
Carter B. G., Butt W., Are somatosensory evoked potentials
the best predictor of outcome after severe brain injury ? A
systematic review, INTENSIVE CARE MED., 31, 765-775, 2005.
Lewis S. B., Myburgh J. A., Thornton E. L., Reilly P. L.,
Cerebral oxygenation monitoring by near-infrared spec-
troscopy is not clinically useful in patients with severe
closed-head injury : a comparison with jugular venous
bulb oximetry, CRIT. CARE MED., 24, 1334-1338, 1996.
Vespa P. M., Multimodality monitoring and telemonitoring
in neurocritical care : from microdialysis to robotic tele-
presence, CURR. OPIN. CRIT. CARE, 11, 133-138, 2005.
Mulvey J. M., Dorsch N. W., Mudaliar Y., Lang E. W.,
Multimodality monitoring in severe traumatic brain
injury : the role of brain tissue oxygenation monitoring,
NEUROCRIT. CARE, 1, 391-402, 2004.
Ng I., Lee K. K., Wong J., Brain tissue oxygenation moni-
toring in acute brain injury, ACTA NEUROCHIR. SUPPL., 95,
447-451, 2005.
Hutchinson P. J., Microdialysis in traumatic brain injury –
methodology and pathophysiology, ACTA NEUROCHIR.
SUPPL., 95, 441-445, 2005.
Cantais E., Boret H., Carre E., Pernod G., [Clinical use of
bedside cerebral microdialysis : a review], ANN. FR.
ANESTH. REANIM., 25, 20-28, 2006.
Vespa P., What is the optimal threshold for cerebral perfu-
sion pressure following traumatic brain injury ?,
NEUROSURG. FOCUS, 15, E4, 2003.
Contant C. F., Valadka A. B., Gopinath S. P., Hannay H. J.,
Robertson C. S., Adult respiratory distress syndrome : a
complication of induced hypertension after severe head
injury, J. NEUROSURG., 95, 560-568, 2001.
Audibert G., Steinmann G., Charpentier C., Mertes P. M.,
[Anaesthetic management of the patient with acute
intracranial hypertension], ANN. FR. ANESTH. REANIM., 24,
492-501, 2005.
Bhardwaj A., Ulatowski J. A., Hypertonic saline solutions
in brain injury, CURR. OPIN. CRIT. CARE, 10, 126-131, 2004.
Fourcade O., Fuzier R., Daboussi A., Gigaud M.,
Tremoulet M., Samii K., [Decompressive craniectomy and
intracranial hypertension], ANN. FR. ANESTH. REANIM., in
press, 2006.
Aarabi B., Hesdorffer D. C., Ahn E. S., Aresco C.,
Scalea T. M., Eisenberg H. M., Outcome following decom-
pressive craniectomy for malignant swelling due to severe
head injury, J. NEUROSURG., 104, 469-479, 2006.
Hans P., Bonhomme V., The rationale for perioperative
brain protection, EUR. J. ANAESTHESIOL., 21, 1-5, 2004.
Longhi L., Stocchetti N., Hyperoxia in head injury : thera-
peutic tool ?, CURR. OPIN. CRIT. CARE, 10, 105-109, 2004.
Clifton G. L., Miller E. R., Choi S. C., Levin H. S.,
McCauley S., Smith K. R., Jr., Muizelaar J. P.,
Wagner F. C., Jr., Marion D. W., Luerssen T. G.,
Chesnut R. M., Schwartz M., Lack of effect of induction of
hypothermia after acute brain injury, N. ENGL. J. MED.,
344, 556-563, 2001.
Oddo M., Schaller M. D., Feihl F., Ribordy V., Liaudet L.,
From evidence to clinical practice : Effective implementa-
tion of therapeutic hypothermia to improve patient outcome
after cardiac arrest, CRIT. CARE MED., in press, 2006.
 HEAD TRAUMA
42. Rabinstein A. A., Vasopressin antagonism : potential
impact on neurologic disease, CLIN. NEUROPHARMACOL., 29,
87-93, 2006.
43. Samadani U., Reyes-Moreno I., Buchfelder M., Endocrine
dysfunction following traumatic brain injury : mechanisms,
pathophysiology and clinical correlations, ACTA
NEUROCHIR. SUPPL., 93, 121-125, 2005.
44. Ghigo E., Masel B., Aimaretti G., Leon-Carrion J.,
Casanueva F. F., Dominguez-Morales M. R., Elovic E.,
Perrone K., Stalla G., Thompson C., Urban R., Consensus
guidelines on screening for hypopituitarism following trau-
matic brain injury, BRAIN INJ., 19, 711-724, 2005.
45. Hans P., Audibert G., Berre J., Bruder N., Ravussin P.,
Ter Minassian A., Puybasset L., Beydon L., Boulard G.,
Bonafe A., Kersaint-Gilly A., Gabrillargues J.,
Lejeune J. P., Proust F., Dufour H., [Cardiovascular and
pulmonary consequences of severe subarachnoid haemor-
rhage], ANN. FR. ANESTH. REANIM., 24, 734-738, 2005.
46. Dziedzic T., Slowik A., Szczudlik A., Nosocomial infec-
tions and immunity : lesson from brain-injured patients,
CRIT. CARE, 8, 266-270, 2004.
247
47. Clifton G. L., Robertson C. S., Choi S. C., Assessment of
nutritional requirements of head-injured patients,
J. NEUROSURG., 64, 895-901, 1986.
48. Payen J. F., Faillot T., Audibert G., Vergnes M. C.,
Bosson J. L., Lestienne B., Bernard C., Bruder N.,
[Thromboprophylaxis in neurosurgery and head trauma],
ANN. FR. ANESTH. REANIM., 24, 921-927, 2005.
49. Kim J., Gearhart M. M., Zurick A., Zuccarello M.,
James L., Luchette F.A., Preliminary report on the safety of
heparin for deep venous thrombosis prophylaxis after
severe head injury, J. TRAUMA, 53, 38-42, 2002.
50. Bissonnette B., Vinchon M., [Traumatic head injury in
children : physiopathology and clinical management],
ANN. FR. ANESTH. REANIM., 21, 133-140, 2002.
51. Trabold F., Meyer P., Orliaguet G., [Severe head injuries in
the young child : early management], ANN. FR. ANESTH.
REANIM., 21, 141-147, 2002.
© Acta Anæsthesiologica Belgica, 2006, 57, n° 3