www.thelancet.com/neurology Vol 7 August 2008
Moderate and severe traumatic brain injury in adults
Andrew I R Maas, Nino Stocchetti, Ross Bullock
Traumatic brain injury (TBI) is a major health and socioeconomic problem that aff ects all societies. In recent
years, patterns of injury have been changing, with more injuries, particularly contusions, occurring in older
patients. Blast injuries have been identifi ed as a novel entity with specifi c characteristics. Traditional approaches to
the classifi cation of clinical severity are the subject of debate owing to the widespread policy of early sedation and
ventilation in more severely injured patients, and are being supplemented with structural and functional
neuroimaging. Basic science research has greatly advanced our knowledge of the mechanisms involved in
secondary damage, creating opportunities for medical intervention and targeted therapies; however, translating
this research into patient benefi t remains a challenge. Clinical management has become much more structured
and evidence based since the publication of guidelines covering many aspects of care. In this Review, we summarise
new developments and current knowledge and controversies, focusing on moderate and severe TBI in adults.
Suggestions are provided for the way forward, with an emphasis on epidemiological monitoring, trauma
organisation, and approaches to management.
Traumatic brain injury (TBI) constitutes a major health
and socioeconomic problem throughout the world.1,2 It
is the leading cause of mortality and disability among
young individuals in high-income countries, and
globally the incidence of TBI is rising sharply, mainly
due to increasing motor-vehicle use in low-income and
middle-income countries. WHO has projected that, by
2020, traffi c accidents will be the third greatest cause of
the global burden of disease and injury.3 In higher
income countries, traffi c safety laws and preventive
measures have reduced the incidence of TBI due to
traffi c accidents,4 whereas the incidence of TBI caused
by falls is increasing as the population ages, leading to
a rise in the median age of TBI populations (table 1).
This has consequences for the type of brain damage
currently seen, and contusions (falls in older patients)
are becoming more frequent than diff use injuries
(high-velocity traffi c accidents in younger patients).
Violence is now reported as the cause of closed head
injury in approximately 7–10% of cases,9,10 a substantial
increase from earlier studies. The incidence of
penetrating brain injury is also increasing, particularly
in the USA, due to the use of fi rearms in violence-
related injuries. Worldwide, armed confl icts and
terrorist activities are causing more brain injuries, often
due to improvised explosive devices, to the extent that
blast injuries of the brain are now recognised as a
specifi c entity. The changing patterns of injury and
treatment approaches have challenged current concepts
of classifi cation. Moreover, basic research has greatly
advanced our knowledge of what happens in the brain
after TBI, off ering opportunities to limit processes
involved in secondary brain damage. However,
translating advances from basic research into clinical
benefi t has proven complex. Here, we discuss current
knowledge and novel insights and controversies in the
study of adults with moderate and severe TBI, with the
aim of integrating basic science and clinical research to
provide guidelines on the epidemiological monitoring
of TBI, trauma organisation, and management at the
Epidemiology and cost
In the USA, monitoring by the Centers for Disease
Control and Prevention shows the annual incidence of
emergency department visits and hospital admissions
for TBI to be 403 per 100 000 and 85 per 100 000,
respectively.11 Epidemiological data on TBI from the
European Union are scarce, but do indicate an annual
aggregate incidence of hospitalised and fatal TBI of
approximately 235 per 100 000,12 similar to that found
in Australia,13 although substantial variation exists
between European countries. Most patients with TBI
have milder injuries, but residual defi cits in these
patients are not infrequent.14 Approximately 10–15% of
patients with TBI have more serious injuries, requiring
TBI is more common in young adults, particularly
men (75%), which causes high costs to society because
of life years lost due to death and disability. In Europe,
TBI accounts for the greatest number of total years
lived with disability resulting from trauma, and is
among the top three causes of injury-related medical
Lancet Neurol 2008; 7: 728–41
Department of Neurosurgery,
University Hospital Antwerp,
(A I R Maas MD); Neuroscience
Intensive Care Unit, Ospedale
Maggiore Policlinico, Milan
University, Milan, Italy
(N Stocchetti MD); and
Department of Neurosurgery,
University of Miami, Miller
School of Medicine, Miami, FL,
USA (R Bullock MD)
Andrew I R Maas, Department of
Hospital Antwerp, Wilrijkstraat
10, 2650 Edegem, Belgium
aged >50 years
UK four centre study61986–1988
Core Data Survey7
Austrian Severe TBI
1984–1987 746 25 15%
*Unpublished (Maas, AIR).
Table 1: Increasing age in TBI studies
www.thelancet.com/neurology Vol 7 August 2008 729
costs.15,16 In the USA, the fi nancial burden has been
estimated at over US$60 billion per year.17 These
numbers stand in stark contrast to the amount of
funding for TBI research, which has one of the highest
unmet needs within the already severely underfunded
fi eld of brain research.18
TBI can be isolated, but is associated with extracranial
injuries (limb fractures, thoracic or abdominal injuries)
in about 35% of cases,19 which increases the risk of
secondary brain damage due to hypoxia, hypotension,
pyrexia, and coagulopathy. The recording of the severity
of extracranial injuries should therefore form an
integral part of TBI classifi cation (panel).
Traditionally, TBI has been classifi ed by mechanism
(closed vs penetrating), by clinical severity (Glasgow
coma scale [GCS]20), and by assessment of structural
damage (neuroimaging;21 panel). The GCS has evolved
into a universal classifi cation system for the severity of
TBI, and consists of the sum score (range 3–15) of the
three components (eye, motor, and verbal scales). For
assessment of severity in individual patients, the three
components should be reported separately. A
standardised approach to assessment is advocated. If
painful stimuli are required to elicit a response, nail-
bed pressure and supraorbital pressure (to test for
localising) are recommended. Increasingly, in modern
practice, classifi cation of TBI by clinical severity is
limited. The level of consciousness might be obscured
in the acute setting by confounders such as medical
sedation, paralysis, or intoxication.22,23
Assessment of structural damage by neuroimaging is
not infl uenced by these confounders. Marshall and
colleagues21 proposed a descriptive system of CT
classifi cation, which focuses on the presence or absence
of a mass lesion, and diff erentiates diff use injuries by
signs of increased intracranial pressure (ICP; ie,
compression of basal cisterns, midline shift). However,
the Marshall classifi cation has limitations, such as the
broad diff erentiation between diff use injuries and mass
lesions, and the lack of specifi cation of the type of mass
lesion (eg, epidural vs subdural). Thus, this classifi cation
system might mask patients who have diff use axonal
injury (DAI) or signs of raised ICP in addition to a mass
lesion, and does not fully use the prognostic information
contained in the individual CT characteristics scored.24
Furthermore, CT can only capture momentary
snapshots of the dynamically evolving process of TBI,
and important lesions that occur at the microscopic
level, such as DAI and ischaemic damage, cannot be
visualised. Therefore, new surrogate markers for these
processes are needed. Such markers have revolutionised
cardiology (eg, troponin) and AIDS therapy, but have
not yet been established for TBI.
An alternative approach is to classify patients by
prognostic risk. Although not new, this is an approach
under development. Recent, well validated models,
developed on large patient samples, have become
available to facilitate this approach.25,26 Prognostic
classifi cation can serve as an objective basis for
comparison of diff erent TBI series, form the basis for
quality assessment of the delivery of health care, and
aid the analysis of clinical trials.27,28
Types of brain damage
TBI is a heterogeneous disorder with diff erent forms of
presentation. The unifying factor is that brain damage
results from external forces, as a consequence of direct
impact, rapid acceleration or deceleration, a penetrating
object (eg, gunshot), or blast waves from an explosion.
The nature, intensity, direction, and duration of these
forces determine the pattern and extent of damage.
On the macroscopic level, damage includes shearing
of white-matter tracts, focal contusions, haematomas
Panel: Approaches to classifi cation of TBI
Closed; penetrating; crash; blast.
Clinical severity: level of consciousness (Glasgow coma scale)
The GCS score comprises the values from three component tests (eye, motor, and verbal
scales). Injuries are classifi ed as severe (GCS 3–8), moderate (GCS 9–13), or mild (GCS 14–15).
• Eyes: 1=no response; 2=open in response to pain; 3=open in response to speech;
• Motor: 1=no response; 2=extension to painful stimuli; 3=abnormal fl exion to painful
stimuli; 4=fl exion/withdrawal to painful stimuli; 5=localises painful stimuli; 6=obeys
• Verbal: 1=no response; 2=incomprehensible sounds; 3=inappropriate utterances;
4=disoriented, confused; 5=oriented, converses normally
Clinical severity (injury severity score)
An abbreviated injury scale (range 0–6) is obtained for six body regions. The injury
severity score (range 0–75) is the sum of quadratic scores for each of the six body regions.
• Body regions: external (skin); head/neck (includes brain); thorax; abdomen/pelvis;
• Scores: 0=none; 1=minor; 2=moderate; 3=serious; 4=severe; 5=critical; 6=virtually
Structural damage (CT)
• Diff use injury I: no visible pathology
• Diff use injury II: cisterns present, midline shift 0–5 mm and/or lesion densities present
or no mass lesion >25 mL
• Diff use injury III (swelling): cisterns compressed or absent with midline shift 0–5 mm
or no mass lesion >25 mL
• Diff use injury IV (shift): midline shift >5 mm, no mass lesion >25 mL
• Evacuated mass lesion: any lesion surgically evacuated
• Non-evacuated mass lesion: High or mixed-density lesion >25 mL, not surgically
Classifi cation by expected outcome as calculated from prognostic models. Examples of
prognostic models can be found at the CRASH and IMPACT websites.
For more on the CRASH study
For more on the IMPACT study
www.thelancet.com/neurology Vol 7 August 2008
(intracerebral and extracerebral), and diff use swelling
(fi gure 1). At the cellular level, early neurotrauma events
(which can occur minutes to hours after initial injury)
include microporation of membranes, leaky ion
channels, and stearic conformational changes in
proteins. At higher shear rates, blood vessels can be
torn, causing (micro)haemorrhages.
DAI is characterised by multiple small lesions in
white-matter tracts. Patients with DAI are usually in
profound coma as a result of the injury, do not manifest
high ICP, and often have a poor outcome. Focal cerebral
contusions are the most common traumatic lesion, are
more frequent in older patients, and usually arise from
contact impact. Traumatic intracranial haematomas
occur in 25–35% of patients with severe TBI and in
5–10% of moderate injuries.
In static crush injuries and focal blows, much of the
energy is absorbed by the skull; thus, brain damage
might remain superfi cial, often with a depressed skull
fracture. Blast injuries have been identifi ed as a novel
entity within TBI.30,31 The pathological mechanism is
much less understood, but the injuries are characterised
by severe early brain
haemorrhage, and often prominent vasospasm.32,33
Outcome of severe blast injuries, even with aggressive
management, is still unknown, but has been encouraging
after debridement of wounds and aggressive control of
ICP, including decompressive surgery.
Ischaemic brain damage is often superimposed on
the primary damage (fi gure 1), and can be widespread
or, more commonly, perilesional. Impaired cerebral
perfusion and oxygenation, excitotoxic injury, and focal
microvascular occlusion can be contributing factors.34,35
Each type of head injury might initiate diff erent
pathophysiological mechanisms, with variable extent
and duration (fi gure 1). These mechanisms (acting
concurrently and often with synergising eff ects) and
the intensity of systemic insults determine the extent of
secondary brain damage. Secondary processes develop
over hours and days, and include neurotransmitter
release, free-radical generation, calcium-mediated
damage, gene activation, mitochondrial dysfunction,
and infl ammatory responses.
Glutamate and other excitatory neurotransmitters
exacerbate ion-channel leakage, worsen astrocytic
swelling, and contribute to brain swelling and raised
ICP. Neurotransmitter release continues for many days
after TBI in human beings, paralleling the course of
high ICP, and, with free-radical and calcium-mediated
damage, is a major cause of early necrotic cell death.
Early gene activation and release of proapoptotic
molecules (eg, caspases) induce apoptotic neuronal
loss. A third potential cause of cell death, autophagy,
might also play an important part.36,37
Infl ammatory response is an important component
of TBI, particularly around
(micro)haemorrhages. The maximum response occurs
within a few days, but cytokines are released from
microglia, astrocytes, and polymorphonuclear cells
within hours after TBI, leading to opening of the blood–
brain barrier, complement-mediated activation of cell
death, and the triggering of apoptosis. Although the
infl ammatory response can be deleterious in excess, it
is necessary in order to clean up cellular debris after
injury, and infl ammatory signals might also trigger
Contribution to secondary damage (%)
Figure 1: Components of TBI and importance of diff erent pathophysiological mechanisms
(A) The diff erent components of TBI with ischaemic damage are superimposed on the primary types of injury (haematoma, contusion, and diff use axonal injury).
Systemic insults and brain swelling contribute to ischaemic damage, which might in turn cause more swelling. (B) The relative importance of diff erent
pathophysiological mechanisms in various types of TBI. CPP=cerebral perfusion pressure. ICP=intracranial perssure. SDH=acute subdural haematoma. DAI=diff use
axonal injury. Adapted from Graham et al,29 with permission from Hodder Arnold.
www.thelancet.com/neurology Vol 7 August 2008 731
regeneration. Hence, infl ammation in TBI can be
thought of as both helpful and harmful.38
Recent research has raised new insights that challenge
existing concepts of pathophysiology. Mitochondrial
dysfunction can cause energy failure after TBI, with a
decrease in production of ATP and consumption of
oxygen by 40–50%. This can trigger opening of the
mitochondrial transition pore, setting off an autodestruct
phenomenon that induces both apoptosis and necrosis.39
Mitochondrial dysfunction might also lead to axonal
disruption. The classical concept that DAI is due to
mechanical rupture of axons, incompatible with
regeneration or repair, has now been abandoned.
Neurons can at least partially regenerate their axonal
anatomy. This accords with clinical observations that
patients with the hallmark features on CT of DAI can
recover with modern neurocritical care. Furthermore,
laboratory studies have shown that DAI can take up to
48 hours to become fully established and is thus
amenable to therapeutic interventions.40
Whether decreased cerebral blood fl ow (CBF) after
trauma is indicative of ischaemia or is secondary to
metabolic depression remains the subject of debate.
PET studies have attempted to clarify the eff ects of
hyperventilation, which lowers ICP by reducing CBF.
Some investigators have found that oxygen metabolism
is preserved,41 whereas others are more concerned that
ischaemia could be induced.42 A combined PET and
microdialysis study showed an increase in the oxygen
extraction fraction in only 1% of cases, whereas lactate
increases were seen in 25% of cases.43 These observations
are not consistent with the classical concept of
ischaemia, and indicate that other mechanisms
contribute to derangements of fl ow and metabolism.
Energy failure due to mitochondrial dysfunction and
diff usion barriers to oxygen delivery resulting from
cytotoxic oedema are likely to be involved. The role of
vasospasm is unclear. Although transcranial doppler
studies report fl ow velocity values suggestive of
vasospasm in up to 35% of patients,44 the correlation
with CBF is relatively poor.45
Head injury does not always implicate TBI. A diagnosis
of TBI is established on the basis of clinical symptoms:
for example, the presence of any documented loss of
consciousness and/or amnesia (retrograde or post-
traumatic). Additional clinical investigations can be
driven by the patient’s level of awareness, presence of
risk factors, and mechanisms of injury (fi gure 2).
CT is the preferred method of assessment on
admission to determine structural damage and to detect
(developing) intracranial haematomas. Traumatic
intracranial lesions occur frequently in severe and
moderate injuries, but are also reported in 14% of
patients with a GCS of 14.46 The risk of intracranial
lesions in patients with a GCS of 15 is generally low,
unless risk factors are present. Therefore, current
guidelines advocate CT examinations in all TBI patients
with a GCS of 14 or lower and in patients with a GCS of
15 in the presence of risk factors.47–49 Some studies have
successfully identifi ed risk factors for intracranial
lesions, such as vomiting, age, duration of amnesia,
injury mechanism, neurological
anticoagulant therapy.46,50,51 In the past, much attention
was focused on conventional radiographs of the skull to
triage indications for CT, but these are no longer
thought to be required. Fractures of the skull can be
defi cit, and
Findings consistent with
ICH; major vessel
Figure 2: Diagnostic approaches in TBI
GCS=Glasgow coma scale. ICH=intracerebral haematoma.
www.thelancet.com/neurology Vol 7 August 2008
seen adequately on CT in the bone-window setting, and
rendering techniques provide a far superior insight into
MRI studies are seldom done in the acute phase of
TBI because they are logistically complex and more
time consuming, and do not necessarily provide any
further information for clinical decisions. However,
MRI can be more informative if a penetrating injury
with a wooden object is suspected.52 In the subacute
and chronic phases of TBI, MRI is more informative
than CT, off ering better detection of white-matter
lesions in patients with DAI.53 Neuroimaging techniques
are rapidly progressing from purely structural
assessments to more functional imaging, off ering a
potentially better understanding of TBI.54
Because TBI is a dynamic process and pathology
evolves, follow-up CT is advisable if lesions were present
on the initial CT or if indicated by clinical deterioration.
New lesions develop in approximately 16% of patients
with diff use injuries,55 and 25–45% of cerebral
contusions will enlarge
occurrences are reported if the initial CT scan is done
within 2 hours of injury.57
Many clinics will usually follow up the patient on the
day after admission, but recent studies indicate that CT
progression will generally occur within 6–9 hours after
injury.58 Follow-up CT is always indicated if larger
lesions are present or if there is clinical deterioration or
increasing ICP.59 Indiscriminate and too frequent use
of CT follow-up are thought to be inappropriate. There
signifi cantly.55,56 Higher
is growing concern about the cancer risk connected to
CT radiation exposure, which is held responsible for
2% of all cancer cases in the USA,60–62 where more than
62 million CT scans, including 4 million in children,
are done each year. The need for CT follow-up should
thus be balanced against awareness of the cancer risk.
Specifi c injury mechanisms might cause vascular
lesions. Arterial dissections
intracranial) have been recognised in up to 17% of
patients with cervical spine injuries, and in those with
skull-base fractures involving the carotid canal.
Traumatic aneurysms are seen in about 15% of patients
with penetrating TBI.63 Consequently, in penetrating
brain injury, CT angiography is indicated in the
presence of an intracranial haematoma, or when a
missile tract, for example, crosses a major vessel
Guidelines and individualised management
Over the past 10 years, much of the treatment of TBI
has evolved towards standardised approaches that
follow international and national guidelines (table 2).
International guidelines for severe head injury are
mostly evidence based, and address specifi c aspects of
management. National guidelines focus more on
organisational issues, such as admission and referral
policy; however, these remain limited to the constraints
of existing trauma systems, and clear statements on the
best trauma organisation are often avoided.
Patel and colleagues71 unequivocally showed a 2·15
times increase in the odds of death (adjusted for case
Year of publicationDescription TypeTopics (n) Recommendations (n)
Class I Class II Class III
Maas and co-workers59
1997 European Brain Injury Consortium guidelines on management
of severe head injury in adults
Management of severe TBI (fi rst edition)
Management and prognosis of severe TBI
Penetrating brain injury
European Federation of Neurological Societies guidelines on
mild traumatic brain injury
Field management of combat-related head trauma
Surgical management of TBI
Revised guidelines for management of severe TBI
Italian guidelines for management of patients with minor
UK guidelines for the initial management of head injuries
Management of acute neurotrauma in rural and remote
locations of Australia
UK guidelines for triage, assessment, investigation, and
management of TBI
Consensus/expert opinion .... .. ..
Bullock and co-workers64*
Bullock and co-workers65
Brain Trauma Foundation*
Aarabi and co-workers63
Vos and co-workers47†
Adelson and co-workers66*
Brain Trauma Foundation*
Bullock and co-workers67*
Brain Trauma Foundation68*
Italian Society of Neurology46
Bartlett and co-workers69
Newcombe and Merry70
UK National Institute for Health
and Clinical Excellence48‡
*http://www.braintrauma.org/. †http://www.efns.org/. ‡In the NICE guidelines (http://www.nice.org/), the grading scheme for level of recommendations was adapted from the Oxford Centre for Evidence Based
Medicine levels of evidence as level A–D; for consistency, we considered grade A as class I, grade B as class II, and grades C and D as class III.
Table 2: Overview of international and national guidelines
www.thelancet.com/neurology Vol 7 August 2008 733
mix) for patients with severe head injury treated in non-
neurosurgical centres versus neurosurgical centres.
Their report makes a strong case for transferring and
treating all patients with severe head injury in a setting
with 24-hour neurosurgical facilities. Most neurosurgical
centres in the UK are not equipped to receive all patients
with TBI because of shortages in human and technical
resources,72 and patients with TBI are referred only for
selected surgical indications. In light of these data, we
suggest that all treatable patients with TBI should be
centralised in large neurotrauma centres that off er
surgical therapy and access to specialised neurocritical
care.73–75 However, this cannot be achieved without
profound re-design of national health systems and re-
allocation of resources.76 A similar case has been made
in the USA: for example, in the state of Florida, which
has 20 million inhabitants, fi ve neurotrauma centres
have been designated to receive patients with severe
TBI and spinal-cord injuries. The concept here is that
greater volume (and hence experience) will lead to
higher quality services and better outcome.77–79
The evidence-based guidelines show that the strength
of supporting data is relatively weak, underscoring the
need for more rigorous evidence (table 2). However, for
various aspects of care in TBI, such as surgery for
epidural haematomas in patients with deteriorating
consciousness, randomised controlled trials are unlikely
to be considered ethical. Notwithstanding the great
benefi ts of evidence-based approaches, guidelines and
practice recommendations should not be made on the
basis of the conclusions of literature reviews alone.
Furthermore, we should recognise that no single
treatment can be uniformly appropriate across the wide
range of conditions within TBI, and this vision would
support the search for more individualised treatment
approaches (ie, determined by monitoring, biomarkers,
and possibly genotype).80
Approaches to management
Pre-hospital emergency care
The main goals of prehospital management are to
prevent hypoxia and hypotension, because these
systemic insults lead to secondary brain damage.68,81
When assessed before hospital admission (by
ambulance or helicopter crews), oxygen saturation
below 90% is found in 44–55% of cases and hypotension
in 20–30%.81–83 Trauma renders the brain more
vulnerable to these insults,84 and hypoxia and
hypotension are strongly associated with poor outcome
(hypoxia: odds ratio [OR] 2·1, 95% CI 1·7–2·6;
hypotension: OR 2·7, 95% CI 2·1–3·4).81
In various settings, the introduction of a pre-hospital
system capable of normalising oxygenation and blood
pressure has been associated with improved outcome.85,86
However, ensuring adequate training of paramedics is
vital because intubation by poorly trained paramedics
has been associated with worse outcome.87 Arterial
hypotension is best prevented by early and adequate
fl uid resuscitation with normotonic crystalloids and
colloids. No benefi ts have yet been shown for hypertonic
solutions,88 or for albumin, which has been associated
with worse outcome.89
The primary aims of admission care are stabilisation
and diagnostic assessment with prioritisation according
to US Advanced Trauma Life Support standards. From
a neurosurgical perspective, the immediate priority is
rapid detection and treatment of operable lesions.
Indications for emergency surgery in closed severe TBI
are summarised in the evidence-based surgical
guidelines.67 CT criteria, including volume, thickness,
and signs of mass eff ect (midline shift), are as relevant
as signs of neurological deterioration. A shift of
emphasis is continuing from the conventional approach
of surgery after neurological deterioration to a more
pre-emptive approach in which the aim of surgery is to
prevent deterioration. In penetrating TBI, a superfi cial
debridement and dural closure is recommended as
general standard of care,63 but in patients with small
entry wounds, simple wound closure may be
considered. This approach is much more conservative
than earlier policies of extensive debridement and
repeated surgery for removal of bone fragments with
the aim of preventing infection. There is no evidence
to support such aggressive approaches, and routine
antibiotic treatment is usually eff ective for infection
Intensive care management
A major focus for neurointensive care is to prevent and
limit ongoing brain damage and to provide the best
conditions for natural brain recovery by reducing brain
swelling and raised ICP. Optimum oxygenation,
perfusion, nutrition, glycaemic control, and temperature
homoeostasis are indicated, as in general intensive
care. However, the concern that the injured brain
cannot tolerate hypoglycaemia, which might result as
an adverse event from over-enthusiastic glycaemic
control, is specifi c to neurointensive care.90 Furthermore,
the brain must be protected from overt or silent
seizures. The benefi ts of prophylactic antiseizure
activity should be balanced against the potential risks.
Opinions vary greatly about routine antiseizure
prophylaxis, but recommended indications include
penetrating brain injury and the presence of a depressed
skull fracture in patients with post-traumatic amnesia
for more than 24 hours in whom a dural lesion is
Sedation and artifi cial ventilation are used to reduce
brain swelling and raised ICP in patients with severe
head injuries. In the 1990s, hyperventilation was
commonly used, which, although eff ective in reducing
ICP, has the risk of enhancing ischaemia.91 Arterial
www.thelancet.com/neurology Vol 7 August 2008
carbon dioxide lower than 30 mm Hg should be
avoided, unless facilities exist for more advanced
monitoring; osmotherapy is currently preferred as the
fi rst agent in the medical management of raised ICP,
and interest in early and extensive decompressive
craniotomies is increasing.
Rapid infusion of mannitol, which creates an osmolar
gradient, mobilises water across an intact blood–brain
barrier,92 but also improves focal CBF.93 Hypertonic
saline infusion creates an analogous, sometimes
stronger, osmolar gradient
improvement in ICP.94–96 When used for medical
management of raised ICP, hyperosmolar agents should
be administered repeatedly (at least every 4–6 hours) or
even continuously, because a rebound phenomenon
might otherwise occur after reversal of the osmotic
gradients by passing into the extracellular space of the
brain. Few studies have compared mannitol with
hypertonic saline.68,97 In one study in which short-term
administration of equimolar amounts of mannitol and
hypertonic saline were compared, hypertonic saline was
associated with a larger and more lasting ICP
However, use of osmotherapy, particularly for long
periods, is associated with electrolyte abnormalities,
especially hypernatraemia, cardiac failure, bleeding
diathesis, and phlebitis. No benefi ts of small-volume
resuscitation with hypertonic saline have been shown
in patients with TBI.88 Consequently, in adults, there is
insuffi cient evidence to support the use of hypertonic
saline over mannitol for osmotherapy. We caution
against the use of hypertonic saline because of the risks
of severe hypernatraemia, and advise particular caution
with the combined use of mannitol and hypertonic
saline. Careful control of fl uid balance, electrolyte
status, and serum osmolarity (<320 mMol/L) is
mandatory in the use of hypertonic agents.
Decompressive craniectomy has been used for more
than a century to treat brain swelling and reduce ICP,
and a resurgence of interest has recently been seen in
this practice, with 240 papers published between 1996
and 2006.99 Despite this renewed enthusiasm, the
technique remains controversial. It does not produce
improved outcome in all series,100 and has many side-
eff ects, some severe.101 The evidence is also confounded
by ambiguous defi nitions and a lack of focus. A clear
distinction should be made between true decompressive
craniectomy versus simple removal of a small bone
fl ap, and between procedures done in combination with
the evacuation of a mass lesion versus an isolated
However, there is consensus that the craniectomy
should be large enough (approximately 15 cm×15 cm)
in diff use injuries.
and should be done early and with duraplasty.102 Two
trials are currently ongoing:
(Randomised Evaluation of Surgery in Craniectomy for
Uncontrollable Elevation of Intracranial Pressure)
Study in Europe and the DECRA (Decompressive
Craniectomy) Study in Australia.103
Monitoring of the severely injured brain
Many patients develop progressive damage without
clear clinical signs; in others, damage develops at an
alarming pace, exemplifi ed by the so-called “talk and
die” syndrome.104,105 Neuroworsening, defi ned as a
deterioration of the GCS motor score, development of
pupillary abnormalities, or development of progressive
CT lesions, has been reported in 29–44% of patients.106–108
Clinical monitoring (level of consciousness and
pupillary reactivity) remains essential. However, more
severely injured patients are universally treated with
sedation and artifi cial ventilation, and advanced
monitoring of cerebral variables is required. Besides
standard systemic monitoring, ICP and cerebral
perfusion pressure monitoring are important. The
eff ectiveness of ICP monitoring in TBI has been
questioned;109,110 no monitoring technique can improve
outcome unless it can drive an appropriate intervention.
Intracranial hypertension develops in up to 77% of
cases, and raised ICP is related to poorer outcome.68
ICP monitoring carries a 0·5% risk of haemorrhage
and a 2% risk of infection. Intraventricular catheters
are preferable, because they are accurate, can be re-
calibrated, and allow drainage of CSF. Intraparenchymal
probes are user friendly and accurate.64 Less accurate
data are provided by subdural catheters,64,111 and epidural
probes are unreliable.112,113 The accuracy of ICP
monitoring can be enhanced by use of computer-
supported systems.114 Attempts to monitor ICP non-
invasively have been unsatisfactory.115,116
More advanced techniques include the monitoring of
microdialysis, and electrophysiological monitoring.
Cerebral oxygenation can be measured focally (brain
tissue oxygen tension) or, more globally, by measuring
the oxygen content in the cerebral venous outfl ow.
Brain tissue oxygen tension indicates the balance
between oxygen delivered to the tissue and its
consumption in a specifi c area, and can indicate
regional hypoxia if it falls below 15–20 mm Hg.117,118 The
diameter of microvascular vessels and diff usion barriers
might also infl uence recorded values.119,120 Venous
oxygen saturation is a more global approach to
monitoring oxygenation. Values below 55% indicate an
increased oxygen extraction relative to perfusion, and
suggest the presence of ischaemia.121,122 Non-randomised
studies have indicated benefi t of an oxygen-directed
Measurements of CBF have shown widespread zones
of brain ischaemia in a third of patients with severe
For more on the
RESCUEicp study see http://
www.thelancet.com/neurology Vol 7 August 2008 735
TBI, especially if done within 8 hours of TBI.124
Microsensor technology has recently been devised to
allow continuous blood fl ow monitoring within the
brain by use of thermal diff usion and laser doppler
Clinical microdialysis allows metabolic exploration of
the cerebral cortex in vivo. Within the fi rst few hours,
most patients with severe TBI have a high lactate:pyruvate
ratio, indicating ischaemia or hyperglycolysis.126–128 More
recently, microdialysis has been used to measure brain
penetration of drugs.129 Microdialysis, oxygen tension
catheters, and CBF probes can be used to assess events
in a small region, but possibly fail to detect harmful
events in other parts of the brain. Conversely, more
global approaches (venous oxygen saturation) fail to
detect regional abnormalities.41,130
Continuous electroencephalographic monitoring can
be used to identify occult seizure activity, which can
occur in 15–18% of patients with moderate and severe
TBI.131 In the research setting, interest exists in
monitoring cortical spreading depression. Traumatically
damaged neurons decrease
substantially in the early post-injury period. Waves of
depolarisation result in ionic fl ux and loss of resting
membrane potential, which worsens neurochemical
dysregulation, and places extra metabolic demands on
their fi ring rates
Outcome and prognosis
Outcome after head injury is generally assessed at
6 months after injury, representing a compromise
between true fi nal outcome and logistic limitations.
Experience shows that about 85% of recovery occurs
within this time period, but further recovery can occur
later. Accurate and consistent outcome determination
at a fi xed timepoint is a prerequisite for any TBI study.
The most frequently used global outcome measure in
TBI is the Glasgow outcome scale. More specifi c tools
are required for detailed examination, such as for
functional and neuropsychological assessment. These
are particularly relevant in the rehabilitation setting.
Early and intensive rehabilitation is recommended to
achieve the best possible functional outcome and social
re-integration. However, the optimum timing and
approach to rehabilitation of patients with TBI remains
to be determined.
In severe closed head injury, the outcome distribution
represents a U-shaped curve, with most patients either
dying or recovering to an independent lifestyle. This
has promoted dichotomisation of outcomes into
unfavourable (death, vegetative state, or severe
disability) versus favourable outcome (moderate
disability or good recovery). Recent studies using the
more detailed 8-point Glasgow outcome scale (extended
version), and a structured interview for its assessment,135
did not fi nd the typical
distribution.88,108 In non-military penetrating head
injury, mortality is consistently over 95% in patients
with a GCS of 3–5, raising ethical questions of whether
active treatment should be initiated, particularly if the
cause of injury was a suicide attempt.
The widespread belief that all patients in a vegetative
state are awake but not aware has been challenged.
Incidental reports on functional MRI studies show that
external stimuli can be processed in the human cortex
of some vegetative patients, and that even spoken
commands might elicit appropriate cortical responses
that are indistinguishable from normal human
responses.136 The implications are that caretakers should
be aware that the patient might hear, see, and realise a
lot more than is commonly thought, and the concept
that a vegetative state might be a worse outcome than
death has become uncertain.
Few TBI studies have used health-related quality of
life measures. Comprehensive
instruments for such assessment in people after TBI
are needed, such as the new disease-specifi c scale,
quality of life in brain injury (QOLIBRI).137
Many studies have reported on the univariate
association between predictors and outcome after TBI,
but few have used multivariate analyses, which adjust
for associations between predictors. Extensive work by
the IMPACT study group, which analysed individual
patient data from over 9000 patients with severe or
moderate TBI merged from 11 studies, confi rmed age,
GCS motor score, pupillary response, and CT
n Odds ratio (95% CI)
Abnormal fl exion
CT classifi cation
Traumatic subarachnoid haemorrhage
2·14 (2·00–2·28) ..
Data from Murray et al.138
Table 3: Predictors of outcome in TBI
www.thelancet.com/neurology Vol 7 August 2008
characteristics as the most powerful independent
prognostic variables (table 3).138 Other important
prognostic factors include hypotension, hypoxia, eye
and verbal components of the GCS, and laboratory
variables (glucose, platelets, and haemoglobin). Other
studies have shown an association between coagulopathy
and poorer outcome,35 indicating that more careful
attention to correcting these disturbances might be
For clinical application and research, the predictive
value of variables can be combined into prognostic
models. Many previously developed models had
shortcomings in development (in particular, a lack of
external validation),139,140 but more recently published
models have greater validity and generalisability.25,26
These models can provide a scientifi c basis for
informing relatives about the likely long-term outcome,
facilitate prognostic classifi cation and valid comparisons
of outcome between diff erent patient series, and enable
the setting of baselines for clinical audits. Furthermore,
prognostic models will have an important role in
randomised controlled trials for stratifi cation and
statistical analyses that explicitly consider prognostic
information, such as covariate adjustment.141
Neuroprotection and clinical trials
The original concept of neuroprotection involved the
initiation of treatment before the onset of the event,
and was aimed at minimising the intensity of an insult
or its immediate eff ects on the brain by interrupting the
harmful cascades of biochemical events. A major new
focus of neurobiology now revolves around cell
replacement, aimed at promoting neuroplasticity, and
regeneration or replacement of lost neuronal and glial
cells and neuronal circuits. Over the past 25 years, over
20 agents have been studied in phase III clinical trials
of severe TBI (table 4). Other strategies have focused on
the use of hypothermia and evaluation of a CBF-targeted
approach versus an ICP-directed approach.152
None of the phase III trials have convincingly shown
effi cacy in the overall study population (table 4).153,154
Various factors have contributed to these failures,
including uncertainty about validity and robustness of
preclinical phase II data, problems in translating results
from experimental studies to clinical practice,
uncertainty about whether
pathophysiological mechanisms targeted are indeed
active in individual patients, uncertainty about the
therapeutic window, and inadequacies in clinical trial
design and analysis. Multidisciplinary eff orts are
currently ongoing to implement approaches in trial
design for dealing with challenges posed by the
heterogeneity of TBI populations.155 Clinical trials in TBI
also pose specifi c ethical dilemmas that are not always
appreciated by policy-making authorities,156 and current
legislation is often perceived
Patients with TBI are often acutely incapacitated by
their injury and cannot always provide informed
consent. Proxy consent is generally substituted, but
does not serve to protect the patient’s autonomy, might
induce selection bias,157 and causes substantial delays
without off ering any real protection to the incapacitated
patient, often only providing some legal protection to
the trial sponsor.158 The validity of proxy consent in the
emergency situation of TBI is questionable because
many relatives are unlikely to be able to make a balanced
decision at a time of such emotional stress. Various
experts have therefore argued for deferred consent in
and when the
Here, we have illustrated the seriousness and complexity
of the problems posed by TBI to patients, relatives,
doctors, authorities, and society alike. Great
nYear of studyStart of treatment Results
1391996≤12 h 12% improvement in favourable
HIT I nimodipine143
HIT II nimodipine144
12 h of not obeying
24 h of injury
No signifi cant eff ect
No signifi cant eff ect in overall
HIT III nimodipine145
1231994 Signifi cant reduction in
No signifi cant eff ect
HIT IV nimodipine
4521993–1995≤12 hNo signifi cant eff ect
6931994–1996≤8 h and within 4 h
No signifi cant eff ect
No signifi cant eff ect
No signifi cant eff ect
Better outcome in men
Lipid peroxidation/free-radical damage
Tirilazad domestic trial
CRASH steroid trial150
No signifi cant eff ect
No signifi cant eff ect
No signifi cant eff ect
No signifi cant eff ect
No signifi cant eff ect
10 0082000–2004≤8 h Higher mortality
No signifi cant eff ect
Poorer outcome at low dose;
higher mortality at high dose
HIT=Head Injury Trial. PESGOD=polyethylene glycol conjugated superoxide dismutase.
Table 4: Overview of phase III randomised clinical trials in TBI
www.thelancet.com/neurology Vol 7 August 2008 737
advancements have been achieved over the past
10–15 years, but advances in basic science have not yet
led to new treatments of clinically proven benefi t, and
advanced monitoring has not routinely resulted in
individualised management. Current classifi cation
systems are no longer suffi cient and not all patients
have access to the best care. Clinical trials in TBI have
had methodological problems posed by the inherent
heterogeneity of the TBI population, and we note a
disconnection between ethical directives and their
validity and applicability in the emergency situation of
acutely incapacitated patients with TBI. Therefore, a
great deal is still to be accomplished. From the
perspective of clinicians, we would suggest that the
following topics are prioritised.
First, standardised epidemiological monitoring should
be implemented to off er a sound basis for appropriate
targeting of prevention. Second, a specifi c focus should
be directed towards trauma organisation with
concentration of care for more severely injured patients
in specialised centres. The specifi cs of the best structure
might vary with the local setting, but we strongly feel
that there is now suffi cient evidence to prove the benefi ts
of centralisation. This centralisation should form a
continuum from emergency systems through to
rehabilitation care. Third, we suggest that further
dissemination of existing guidelines should form the
basis for standards of care, but recommend an additional
strong focus on more individualised treatment, targeting
the specifi c needs of a given patient. This approach will
require advanced diagnostics and monitoring, thus
allowing identifi cation of pathophysiological alterations.
These diagnostic procedures could include the use of
biomarkers and genotyping. As in the acute phase,
specialised facilities and concentration of care are
necessary in the post-acute setting, off ering appropriate
rehabilitation programmes, and exploring strategies for
promoting recovery and regeneration.
These goals cannot be achieved without more basic
and clinical research. Multiple mechanisms of TBI have
been identifi ed in the experimental setting, but much
more needs to be elucidated. From a research
perspective, attention might focus on the cerebrovascular
interface and mechanisms of energy failure after TBI.
Notwithstanding the importance of basic science,
clinically oriented research with therapeutic trials is
needed, perhaps based more on an integral concept or
combined approaches, rather than focusing on the
possible benefi t of agents that target a single
pathophysiological mechanism among the many other
processes. The methods used in TBI clinical trials
require refi nements, including the implementation of
approaches to deal with the inherent heterogeneity of
TBI populations. Whatever route is taken or priorities
chosen, advancing the fi eld further will require great
multidisciplinary eff orts
clinicians, facilitated by appropriate funding.
from researchers and
All authors contributed equally to the preparation of this Review.
Confl icts of interest
The authors are members of the European and American Brain Injury
Consortia. AIRM has been a consultant and a member of steering
committees for clinical trials in TBI for Pharmos, Solvay, Vasopharm,
and Novo Nordisk. NS has been a member of steering committees or
safety boards for clinical trials in TBI for Pharmos, Solvay, Vasopharm,
We dedicate this Review to the memory of Bryan Jennett (1926–2008),
who inspired us to devote a large part of our lives and work to TBI. He
was one of the founding leaders of neurotrauma research and a great
inspiration to all around him. Many of the concepts and insights
developed by him in the 1970s are still in use today. We thank our co-
workers and partners in research, both within their own institutes and
within the American and European Brain Injury Consortia. We are
grateful to Beverly Walters for help in editing the manuscript, to Paul
Parizel for advice on neuroradiological items, and to Anne-Claire van
Harderwijk for administrative support. AIRM is a recipient of a NIH
grant (NS-042691) supporting methodological and clinical research in
TBI, and RMB is funded by the US Department of Defense.
1 Ghajar J. Traumatic brain injury. Lancet 2000; 356: 923–29.
2 Cole TB. Global road safety crisis remedy sought: 1·2 million killed,
50 million injured annually. JAMA 2004; 291: 2531–32.
3 Finfer SR, Cohen J. Severe traumatic brain injury. Resuscitation
2001; 48: 77–90.
4 Redelmeier DA, Tibshirani RJ, Evans L. Traffi c-law enforcement
and risk of death from motor-vehicle crashes: case-crossover study.
Lancet 2003; 361: 2177–82.
5 Foulkes AM, Eisenberg MH, Jane AJ. The Traumatic Coma Data
Bank; design, methods and baseline characteristics. J Neurosurg
1991; 75 (suppl): S1–15.
6 Murray LS, Teasdale GM, Murray GD, Miller DJ, Pickard JD,
Shaw MD. Head injuries in four British neurosurgical centres.
Br J Neurosurg 1999; 13: 564–69.
7 Murray GD, Teasdale GM, Braakman R, et al. The European Brain
Consortium survey of head injuries. Acta Neurochir (Wien) 1999;
8 Rusnak M, Janciak I, Majdan M, Wilbacher I, Mauritz W, Austrian
Severe TBI Study Investigators. Severe traumatic brain injury in
Austria. I: Introduction to the study. Wien Klin Wochenschr 2007;
9 Butcher I, McHugh GS, Lu J, et al. The prognostic value of cause of
injury in traumatic brain injury: results from the IMPACT study.
J Neurotrauma 2007; 24: 281–86.
10 Jiang J, Feng H, Fu Z, et al. Violent trauma in China: report of
2254 cases. Surg Neurol 2007; 68 (suppl 2): 2–5.
11 Langlois JA, Rutland-Brown W, Thomas KE. Traumatic brain injury
in the United States: emergency department visits, hospitalizations,
and deaths. Atlanta: Centers for Disease Control and Prevention,
National Center for Injury Prevention and Control, 2006.
Search strategy and selection criteria
References for this Review were identifi ed through searches
of PubMed, by use of the search terms “traumatic brain
injury” or “head injury” and other appropriate targets, such as
“epidemiology”, “pathophysiology”, and “guidelines”, up to
May, 2008. Papers were also identifi ed from the authors’ own
fi les and from references cited in relevant articles. An
electronic search of resources, such as international and
national guidelines and book chapters, was also done. The
fi nal reference list was generated on the basis of relevance to
the topics covered in this Review.
www.thelancet.com/neurology Vol 7 August 2008
12 Tagliaferri F, Compagnone C, Korsic M, Servadei F, Kraus J.
A systematic review of brain injury epidemiology in Europe.
Acta Neurochir (Wien) 2006; 148: 255–68.
13 Hillier S, Hiller J, Metzer J. Epidemiology of traumatic brain injury
in South Australia. Brain Inj 1997; 11: 649–59.
14 Thornhill S, Teasdale GM, Murray GD, McEwen J, Roy CW,
Penny KI. Disability in young people and adults one year after head
injury: prospective cohort study. BMJ 2000; 320: 1631–35.
15 Polinder S, Meerding WJ, van Baar ME, et al. Cost estimation of
injury-related hospital admissions in 10 European countries.
J Trauma 2005; 59: 1283–91.
16 Polinder S, Meerding WJ, Mulder S, Petridou E, van Beeck E,
EUROCOST Reference Group. Assessing the burden of
injury in six European countries. Bull World Health Organ 2007;
17 Thurman DJ. Epidemiology and economics of head trauma. In:
Miler L, Hayes R, eds. Head trauma: basic, preclinical, and clinical
directions. New York: Wiley and Sons, 2001: 1193–202.
18 Sobocki P, Olesen J, Jönsson B. Brain research has high returns but
Europe is lagging behind. Eur J Neurol 2007; 14: 708–10.
19 Gennarelli TA, Champion HR, Sacco WJ, Copes WS, Alves WM.
Mortality in patients with head injury and extracranial injury treated
in trauma centers. J Trauma 1990; 29: 1193–202.
20 Teasdale G, Jennett B. Assessment of coma and impaired
consciousness. A practical scale. Lancet 1974; 2: 81–84.
21 Marshall LF, Marshall SB, Klauber MR, et al. A new classifi cation of
head injury based on computerized tomography. J Neurosurg 1991;
75 (suppl): s14–20.
22 Balestreri M, Czosnyka M, Chatfi eld DA, et al. Predictive value
of Glasgow Coma Scale after brain trauma: change in trend
over the past ten years. J Neurol Neurosurg Psychiatry 2004;
23 Stocchetti N, Pagan F, Calappi E, et al. Inaccurate early assessment
of neurological severity in head injury. J Neurotrauma 2004;
24 Maas AI, Hukkelhoven CW, Marshall LF, Steyerberg EW. Prediction
of outcome in traumatic brain injury with computed tomographic
characteristics: a comparison between the computed tomographic
classifi cation and combinations of computed tomographic
predictors. Neurosurgery 2005; 57: 1173–82.
25 Steyerberg EW, Mushkudiani N, Perel P, et al. Predicting outcome
after traumatic brain injury: development and international
validation of prognostic scores based on admission characteristics.
PLoS Med 2008 (in press).
26 MRC CRASH Trial Collaborators. Predicting outcome after
traumatic brain injury: practical prognostic models based on large
cohort of international patients. BMJ 2008; 336: 425–29.
27 Menon D, Harrison D. Prognostic modelling in traumatic brain
injury. BMJ 2008; 336: 397–98.
28 Maas AI, Lingsma HF. New approaches to increase statistical
power in TBI trials: insights from the IMPACT study group.
Acta Neurochir (Wien) 2008 (in press).
29 Graham DI, Gennarelli TA, McIntosh TK. Trauma. In: Graham DI,
Lantos PL, eds. Greenfi eld’s Neuropathology, vol 1, 7th edn. London:
Hodder Arnold, 2002: 823–898.
30 Okie S. Traumatic brain injury in the war zone. N Engl J Med 2005;
31 Zoroya G. Scientists: brain injuries from war worse than thought.
USA Today, Nov 5, 2007. http://www.usatoday.com/news/world/
June 12, 2008).
32 Taber KH, Warden DL, Hurley RA. Blast-related traumatic brain
injury: what is known? J Neuropsychiatry Clin Neurosci 2006;
33 Armonda RA, Bell RS, Vo AH, et al. Wartime traumatic cerebral
vasospasm: recent review of combat casualties. Neurosurgery 2006;
34 Stein SC, Chen XH, Sinson GP, Smith DH. Intravascular
coagulation: a major secondary insult in nonfatal traumatic brain
injury. J Neurosurg 2002; 97: 1373–77.
35 Harhangi, S, Kompanje EJO, Leebeek FWG, Maas AIR. Coagulation
disorders after traumatic brain injury. Acta Neurochir (Wien) 2008;
36 Clark RS, Bayir H, Chu CT, Alber SM, Kochanek PM, Watkins SC.
Autophagy is increased in mice after traumatic brain injury and is
detectable in human brain after trauma and critical illness.
Autophagy 2008; 4: 88–90.
37 Lai Y, Hickey RW, Chen Y, et al. Autophagy is increased after
traumatic brain injury in mice and is partially inhibited by the
antioxidant gamma-glutamylcysteinyl ethyl ester. J Cereb Blood Flow
Metab 2008; 28: 540–50.
38 Morganti-Kossmann MC, Rancan M, Stahel PF, Kossmann T.
Infl ammatory response in acute traumatic brain injury: a double-
edged sword. Curr Opin Crit Care 2002; 8: 101–05.
39 Kroemer G, Galluzzi L, Brenner C. Mitochondrial membrane
permeabilization in cell death. Physiol Rev 2007; 87: 99–163.
40 Büki A, Koizumi H, Povlishock JT. Moderate posttraumatic
hypothermia decreases early calpain-mediated proteolysis and
concomitant cytoskeletal compromise in traumatic axonal injury.
Exp Neurol 1999; 159: 319–28.
41 Diringer MN, Videen TO, Yundt K, et al. Regional cerebrovascular
and metabolic eff ects of hyperventilation after severe traumatic
brain injury. J Neurosurg 2002; 96: 103–08.
42 Coles JP, Minhas PS, Fryer TD, et al. Eff ect of hyperventilation
on cerebral blood fl ow in traumatic head injury: clinical
relevance and monitoring correlates. Crit Care Med 2002;
43 Vespa P, Bergsneider M, Hattori N, et al. Metabolic crisis without
brain ischemia is common after traumatic brain injury: a combined
microdialysis and positron emission tomography study. J Cereb
Blood Flow Metab 2005; 25: 763–74.
44 Oertel M, Boscardin WJ, Obrist WD, et al. Posttraumatic
vasospasm: the epidemiology, severity, and time course of an
underestimated phenomenon: a prospective study performed in
299 patients. J Neurosurg 2005; 103: 812–24.
45 Chieregato A, Sabia G, Tanfani A, Compagnone C, Tagliaferri F,
Targa L. Xenon-CT and transcranial Doppler in poor-grade or
complicated aneurysmatic subarachnoid hemorrhage patients
undergoing aggressive management of intracranial hypertension.
Intensive Care Med 2006; 32: 1143–50.
46 Smits M, Dippel DW, Steyerberg EW, et al. Predicting intracranial
traumatic fi ndings on computed tomography in patients with
minor head injury: the CHIP prediction rule [summary for patients
in Ann Intern Med 2007; 146: I55]. Ann Intern Med 2007;
47 Study Group on Head Injury of the Italian Society for Neurosurgery.
Guidelines for minor head injured patients’ management in adult
age. J Neurosurg Sci 1996; 40: 11–15.
48 Vos PE, Battistin L, Birbamer G, et al. EFNS guideline on mild
traumatic brain injury: report of an EFNS task force. Eur J Neurol
2002; 9: 207–19.
49 UK National Institute for Health and Clinical Excellence. Head
injury. Triage, assessment, investigation and early management of
head injury in infants, children and adults. Partial update of NICE
clinical guideline 4 (June, 2003); September, 2007. http://www.nice.
June 12, 2008).
50 Haydel MJ, Preston CA, Mills TJ, Luber S, Blaudeau E, DeBlieux
PM. Indications for computed tomography in patients with minor
head injury. N Engl J Med 2000; 343: 100–05.
51 Stiell IG, Wells GA, Vandemheen K, et al. The Canadian CT Head
Rule for patients with minor head injury. Lancet 2001;
52 Green BF, Kraft SP, Carter KD, Buncic JR, Nerad JA, Armstrong D.
Intraorbital wood. Detection by magnetic resonance imaging.
Ophthalmology 1990; 97: 608–11.
53 Huisman TA, Schwamm LH, Schaefer PW, et al. Diff usion
tensor imaging as potential biomarker of white matter
injury in diff use axonal injury. ANJR Am J Neuroradiol 2004;
54 Parizel PM, Van Goethem JW, Ozsarlak O, Maes M, Phillips CD.
New developments in the neuroradiological diagnosis of
craniocerebral trauma. Eur Radiol 2005; 15: 569–81.
55 Servadei F, Murray GD, Penny K, et al. The value of the “worst”
computed tomographic scan in clinical studies of moderate and
severe head injury. European Brain Injury Consortium.
Neurosurgery 2000; 46: 70–77.
www.thelancet.com/neurology Vol 7 August 2008 739
56 Chang EF, Meeker M, Holland MC. Acute traumatic
intraparenchymal hemorrhage: risk factors for progression in the
early post-injury period. Neurosurgery 2006; 58: 647–56.
57 Oertel M, Kelly DF, McArthur D, et al. Progressive hemorrhage
after head trauma: predictors and consequences of the evolving
injury. J Neurosurg 2002; 96: 109–16.
58 Narayan RK, Maas AIR, Servadei F, et al. Progression of traumatic
intracerebral hemorrhage: a prospective observational study.
J Neurotrauma 2008; 25: 629–639.
59 Maas A, Dearden M, Teasdale GM, et al. EBIC guidelines for
management of severe head injury in adults. European Brain Injury
Consortium. Acta Neurochir (Wien) 1997; 139: 286–94.
60 Brenner DJ, Hall EJ. Computed tomography—an increasing source
of radiation exposure. N Engl J Med 2007; 357: 2277–84.
61 Birnbaum S. CT scanning: too much of a good thing. BMJ
2007; 334: 1006.
62 Kmietowicz Z. Better safe than sorry? BMJ 2007; 335: 1182–84.
63 Aarabi B, Alden TD, Chestnut RM, et al. Management and
prognosis of penetrating brain injury—guidelines. J Trauma 2001;
51 (suppl): S1–86.
64 Bullock R, Chesnut RM, Clifton G, et al. Guidelines for the
management of severe head injury. J Neurotrauma 1996;
65 Bullock RM, Chesnut RM, Clifton GL, et al. Management and
prognosis of severe traumatic brain injury. Guidelines for the
management of severe traumatic brain injury. J Neurotrauma 2000;
66 Adelson PD, Bratton SL, Carney NA, et al. Guidelines for the acute
medical management of severe traumatic brain injury in infants,
children, and adolescents. Pediatr Crit Care Med 2003;
4 (suppl 3): S1–75.
67 Bullock RM, Chesnut R, Ghajar J, et al. Guidelines for the surgical
management of traumatic brain injury. Neurosurgery 2006;
58 (suppl 2): 1–58.
68 Brain Trauma Foundation, American Association of Neurological
Surgeons (AANS), Congress of Neurological Surgeons (CNS),
AANS/CNS Joint Section on Neurotrauma and Critical Care.
Guidelines for the management of severe traumatic brain injury,
3rd edition. J Neurotrauma 2007; 24 (suppl 1): S1–106.
69 Bartlett J, Kett-White R, Mendelow AD, et al. Guidelines for the
initial management of head injuries. Recommendations from the
Society of British Neurological Surgeons. Br J Neurosurg 1998;
70 Newcombe R, Merry G. The management of acute neurotrauma in
rural and remote locations: a set of guidelines for the care of head
and spinal injuries. J Clin Neurosci 1999; 6: 85–93.
71 Patel HC, Bouamra O, Woodford M, et al. Trends in head injury
outcome from 1989 to 2003 and the eff ect of neurosurgical care: an
observational study [published erratum in Lancet 2006; 367: 816].
Lancet 2005; 366: 1538–44.
72 Thillai M. Neurosurgical units working beyond safe capacity. BMJ
2000; 320: 399.
73 Varelas PN, Conti MM, Spanaki MV, et al. The impact of a
neurointensivist-led team on a semiclosed neurosciences intensive
care unit. Crit Care Med 2004; 32: 2191–98.
74 Suarez JI, Zaidat OO, Suri MF, et al. Length of stay and mortality in
neurocritically ill patients: impact of a specialized neurocritical care
team. Crit Care Med 2004; 32: 2311–17.
75 Visca A, Faccani G, Massaro F, et al. Clinical and neuroimaging
features of severely brain-injured patients treated in a
neurosurgical unit compared with patients treated in peripheral
non-neurosurgical hospitals. Br J Neurosurg 2006; 20: 82–86.
76 Seeley HM, Maimaris C, Carroll G, Kellerman J, Pickard JD.
Implementing the Galasko Report on the management of head
injuries: the Eastern Region approach. Emerg Med J 2001;
77 Heros RC. Case volume and mortality. J Neurosurg 2003;
78 Trunkey DD. Trauma centers and trauma systems. JAMA 2003;
79 MacKenzie EJ, Rivara FP, Jurkovich GJ, et al. A national evaluation
of the eff ect of trauma-center care on mortality. N Engl J Med 2006;
80 Miller JD, Piper IR, Dearden NM. Management of intracranial
hypertension in head injury: matching treatment with cause.
Acta Neurochir Suppl (Wien) 1993; 57: 152–59.
81 McHugh GS, Engel DC, Butcher I, et al. Prognostic value of
secondary insults in traumatic brain injury: results from the
IMPACT study. J Neurotrauma 2007; 24: 287–93.
82 Silverstone P. Pulse oxymetry at the roadside: a study of pulse
oxymetry in immediate care. BMJ 1989; 298: 711–13.
83 Stocchetti N, Furlan A, Volta F. Hypoxemia and arterial hypotension
at the accident scene in head injury. J Trauma 1996; 40: 764–67.
84 DeWitt DS, Jenkins LW, Prough DS. Enhanced vulnerability to
secondary ischemic insults after experimental traumatic brain
injury. New Horiz 1995; 3: 376–83.
85 Winchell RJ, Hoyt DB. Endotracheal intubation in the fi eld
improves survival in patients with severe head injury. Arch Surg
1997; 132: 592–97.
86 Rudehill A, Bellander BM, Weitzberg E, Bredbacka S, Backheden
M, Gordon E. Outcome of traumatic brain injuries in 1,508
patients: impact of prehospital care. J Neurotrauma 2002;
87 Davis DP, Hoyt DB, Ochs M, et al. The eff ect of paramedic rapid
sequence intubation on outcome in patients with severe traumatic
brain injury. J Trauma 2003; 54: 444–53.
88 Cooper DJ, Myles PS, McDermott FT, et al. Prehospital hypertonic
saline resuscitation of patients with hypotension and severe
traumatic brain injury: a randomized controlled trial. JAMA 2004;
89 SAFE Study Investigators. Saline or albumin for fl uid resuscitation in
patients with traumatic brain injury. N Engl J Med 2007; 357: 874–84.
90 Vespa P, Boonyaputthikul R, McArthur DL, et al. Intensive insulin
therapy reduces microdialysis glucose values without altering
glucose utilization or improving the lactate/pyruvate ratio after
traumatic brain injury. Crit Care Med 2006; 34: 850–56.
91 Stocchetti N, Maas AI, Chieregato A, van der Plas AA.
Hyperventilation in head injury: a review. Chest 2005;
92 Polderman KH, Van De Kraats G, Dixon JM, Vandertop WP,
Girbes AR. Increases in spinal fl uid osmolarity induced by
mannitol. Crit Care Med 2003; 31: 584–90.
93 Muizelaar JP, Wei EP, Kontos HA, Becker DP. Mannitol causes
compensatory cerebral vasoconstriction and vasodilation in
response to blood viscosity changes. J Neurosurg 1983; 59: 822–28.
94 Shackford SR, Bourguignon PR, Wald SL, Rogers FB, Osler TM,
Clark DE. Hypertonic saline resuscitation of patients with head
injury: a prospective, randomized clinical trial. J Trauma 1998;
95 Suarez JI, Qureshi AI, Bhardwaj A, et al. Treatment of refractory
intracranial hypertension with 23·4% saline. Crit Care Med 1998;
96 Ware ML, Nemani VM, Meeker M, Lee C, Morabito DJ, Manley GT.
Eff ects of 23·4% sodium chloride solution in reducing intracranial
pressure in patients with traumatic brain injury: a preliminary
study. Neurosurgery 2005; 57: 727–36.
97 Wakai A, Roberts I, Schierhout G. Mannitol for acute traumatic
brain injury. Cochrane Database Syst Rev 2005; 4: CD001049.
98 Battison C, Andrews PJ, Graham C, Petty T. Randomized,
controlled trial on the eff ect of a 20% mannitol solution and a 7·5%
saline/6% dextran solution on increased intracranial pressure after
brain injury. Crit Care Med 2005; 33: 196–202.
99 Citerio G, Andrews PJ. Refractory elevated intracranial pressure:
intensivist’s role in solving the dilemma of decompressive
craniectomy. Intensive Care Med 2007; 33: 45–48.
100 Münch E, Horn P, Schurer L, Piepgras A, Paul T, Schmiedek P.
Management of severe traumatic brain injury by decompressive
craniectomy. Neurosurgery 2000; 47: 315–23.
101 Servadei F, Compagnone C, Sahuquillo J. The role of surgery in
traumatic brain injury. Curr Opin Crit Care 2007; 13: 163–68.
102 Polin RS, Shaff rey ME, Bogaev CA, et al. Decompressive bifrontal
craniectomy in the treatment of severe refractory posttraumatic
cerebral edema. Neurosurgery 1997; 41: 84–94.
103 National Trauma Research Institute. Clinical research projects.
Decompressive craniectomy (DECRA) study. http://www.ntri.com.
au/research/clinical/projects/ (accessed June 13, 2008).
www.thelancet.com/neurology Vol 7 August 2008
104 Reilly PL, Graham DI, Adams JH, Jennett B. Patients with head
injury who talk and die. Lancet 1975; 2: 375–77.
105 Rockswold GL, Pheley PJ. Patients who talk and deteriorate.
Ann Emerg Med 1993; 22: 1004–07.
106 Morris GF, Juul N, Marshall SB, Benedict B, Marshall LF.
Neurological deterioration as a potential alternative endpoint in
human clinical trials of experimental pharmacological agents for
treatment of severe traumatic brain injuries. Neurosurgery 1998;
107 Juul N, Morris GF, Marshall SB, Marshall LF. Intracranial
hypertension and cerebral perfusion pressure: infl uence on
neurological deterioration and outcome in severe head injury.
J Neurosurg 2000; 92: 1–6.
108 Maas AI, Murray G, Henney H 3rd, et al. Effi cacy and safety of
dexanabinol in severe traumatic brain injury: results of a phase III
randomised, placebo-controlled, clinical trial. Lancet Neurol 2006;
109 Cremer OL, van Dijk GW, van Wensen E, et al. Eff ect of intracranial
pressure monitoring and targeted intensive care on functional
outcome after severe head injury. Crit Care Med 2005; 33: 2207–13.
110 Shafi S, Diza-Arrastia R, Madden C, Gentilello L. Intracranial
pressure monitoring in brain-injured patients is associated with
worsening of survival. J Trauma 2008; 64: 335–40.
111 Bhatia A, Gupta AK. Neuromonitoring in the intensive care unit. I.
Intracranial pressure and cerebral blood fl ow monitoring.
Intensive Care Med 2007; 33: 1263–71.
112 Citerio G, Andrews PJ. Intracranial pressure. Part two: clinical
applications and technology. Intensive Care Med 2004; 30: 1882–85.
113 Poca MA, Sahuquillo J, Topczewski T, Penarrubia MJ, Muns A. Is
intracranial pressure monitoring in the epidural space reliable? Fact
and fi ction. J Neurosurg 2007; 106: 548–56.
114 Zanier ER, Ortolano F, Ghisoni L, Colombo A, Losappio S,
Stocchetti N. Intracranial pressure monitoring in intensive care:
clinical advantages of a computerized system over manual
recording. Crit Care 2007; 11: R7.
115 Czarnik T, Gawda R, Latka D, Kolodziej W, Sznajd-Weron K,
Weron R. Noninvasive measurement of intracranial pressure: is it
possible? J Trauma 2007; 62: 207–11.
116 Stocchetti N. Could intracranial pressure in traumatic brain injury
be measured or predicted noninvasively? Almost. Intensive Care Med
2007; 33: 1682–83.
117 Van den Brink WA, van Santbrink H, Steyerberg EW, et al. Brain
oxygen tension in severe head injury. Neurosurgery 2000; 46: 868–78.
118 Zauner A, Daugherty WP, Bullock MR, Warner DS. Brain
oxygenation and energy metabolism: part I—biological function
and pathophysiology. Neurosurgery 2002; 51: 289–302.
119 Carmona Suazo JA, Maas AIR, van den Brink WA,
van Santbrink H, Steyerberg EW, Avezaat CJJ. CO2 reactivity and
brain oxygen pressure monitoring in severe head injury.
Crit Care Med 2000; 28: 3268–74.
120 van Santbrink H, Van den Brink WA, Steyerberg EW,
Carmona Suazo JA, Avezaat CJ, Maas AI. Brain tissue oxygen
response in severe traumatic brain injury. Acta Neurochir (Wien)
2003; 145: 429–38.
121 Robertson CS, Narayan RK, Gokaslan ZL, et al. Cerebral
arteriovenous oxygen diff erence as an estimate of cerebral blood
fl ow in comatose patients. J Neurosurg 1989; 70: 222–30.
122 Chieregato A, Calzolari F, Trasforini G, Targa L, Latronico N.
Normal jugular bulb oxygen saturation. J Neurol Neurosurg
Psychiatry 2003; 74: 784–86.
123 Stiefel MF, Spiotta A, Gracias VH, et al. Reduced mortality rate in
patients with severe traumatic brain injury treated with brain tissue
oxygen monitoring. J Neurosurg 2005; 103: 805–11.
124 Bouma GJ, Muizelaar JP, Stringer WA, Choi SC, Fatouros P,
Young HF. Ultra-early evaluation of regional cerebral blood fl ow in
severely head-injured patients using xenon-enhanced computerized
tomography. J Neurosurg 1992; 77: 360–68.
125 Vajkoczy P, Roth H, Horn P, et al. Continuous monitoring of
regional cerebral blood fl ow: experimental and clinical validation of
a novel thermal diff usion microprobe. J Neurosurg 2000; 93: 265–74.
126 Persson L, Hillered L. Chemical monitoring of neurosurgical
intensive care patients using intracerebral microdialysis.
J Neurosurg 1992; 76: 72–80.
127 Bergsneider M, Hovda DA, Shalmon E, et al. Cerebral
hyperglycolysis following severe traumatic brain injury in
humans: a positron emission tomography study. J Neurosurg 1997;
128 Glenn TC, Kelly DF, Boscardin WJ, et al. Energy dysfunction as a
predictor of outcome after moderate or severe head injury: indices of
oxygen, glucose, and lactate metabolism. J Cereb Blood Flow Metab
2003; 23: 1239–50.
129 Hillered L, Vespa PM, Hovda DA. Translational neurochemical
research in acute human brain injury: the current status and
potential future for cerebral microdialysis. J Neurotrauma 2005;
130 Imberti R, Bellinzona G, Langer M. Cerebral tissue PO2 and SjvO2
changes during moderate hyperventilation in patients with severe
traumatic brain injury. J Neurosurg 2002; 96: 97–102.
131 Vespa PM, Nuwer MR, Nenov V, et al. Increased incidence and
impact of nonconvulsive and convulsive seizures after traumatic
brain injury as detected by continuous electroencephalographic
monitoring. J Neurosurg 1999; 91: 750–60.
132 Strong AJ, Fabricius M, Boutelle MG, et al. Spreading and
synchronous depressions of cortical activity in acutely injured
human brain. Stroke 2002; 33: 2738–43.
133 Dreier JP, Woitzik J, Fabricius M, et al. Delayed ischaemic
neurological defi cits after subarachnoid haemorrhage are
associated with clusters of spreading depolarizations. Brain 2006,
134 Fabricius M, Fuhr S, Bhatia R, et al. Cortical spreading depression
and peri-infarct depolarization in acutely injured human cerebral
cortex. Brain 2006; 129: 778–90.
135 Wilson JTL, Pettigrew LEL, Teasdale GM. Structured interviews for
the Glasgow Outcome Scale and the extended Glasgow Outcome
Scale: guidelines for their use. J Neurotrauma 1998; 15: 573–85.
136 Owen AM, Coleman MR. Functional neuroimaging of the
vegetative state. Nat Rev Neurosci 2008; 9: 235–43.
137 von Steinbuechel, N, Petersen C, Bullinger M. Assessment of
health-related quality of life in persons after traumatic brain
injury—development of the Qolibri, a specifi c measure.
Acta Neurochir Suppl 2005; 93: 43–49.
138 Murray GD, Butcher I, McHugh GS, et al. Multivariable prognostic
analysis in traumatic brain injury: results from the IMPACT study.
J Neurotrauma 2007; 24: 329–37.
139 Mushkudiani NA, Hukkelhoven CWPM, Hernández AV, et al.
A systematic review fi nds methodological improvements necessary
for prognostic models in determining traumatic brain injury
outcomes. J Clin Epidemiol 2008; 61: 331–43.
140 Perel P, Edwards P, Wentz R, Roberts I. Systematic review
of prognostic models in traumatic brain injury. BMC Med
Inform Decis Mak 2006; published online Nov 14. DOI:10.1186/
141 Hernandez AV, Steyerberg EW, Butcher I, et al. Adjustment for
strong predictors of outcome in traumatic brain injury trials: 25%
reduction in sample size requirements in the IMPACT study.
J Neurotrauma 2006; 23: 1295–303.
142 Marmarou A, Nichols J, Burgess J, et al. Eff ects of bradykinin
antagonist bradycor (Deltibant, CP-1027) in severe traumatic brain
injury: results of a multi-center, randomized, placebo-controlled
trial. J Neurotrauma 1999; 16: 431–44.
143 Bailey I, Bell A, Gray J, et al. A trial of the eff ect of nimodipine on
outcome after head injury. Acta Neurochir (Wien) 1991;
144 European Study Group. A multicenter trial of the effi cacy of
nimodipine on outcome after severe head injury. The European
Study Group on Nimodipine in Severe Head Injury. J Neurosurg
1994; 80: 797–804.
145 Harders A, Kakarieka A, Braakman R. Traumatic subarachnoid
hemorrhage and its treatment with nimodipine. J Neurosurg 1996;
146 Morris GF, Bullock R, Bowers Marshall S, et al. Failure of the
competitive N-methyl-D-aspartate antagonist Selfotel (CGS 19755)
in the treatment of severe head injury: results of two phase III
clinical trials. J Neurosurg 1999; 91: 737–43.
147 Marshall LF, Maas AI, Marshall SB, et al. A multicenter trial on the
effi cacy of using tirilazad mesylate in cases of head injury.
J Neurosurg 1998; 89: 519–25.
www.thelancet.com/neurology Vol 7 August 2008 741
148 Grumme T, Baethmann A, Kolodziejczyk D, et al. Treatment of
patients with severe head injury by triamcinolone: a prospective,
controlled multicenter clinical trial of 396 cases. Res Exp Med (Berl)
1995; 195: 217–29.
149 Gaab MR, Trost HA, Alcantara A, et al. “Ultrahigh” dexamethasone
in acute brain injury. Results from a prospective randomized
double-blind multicenter trial (GUDHIS). Zentralbl Neurochir 1994;
150 Edwards P, Arango M, Balica L, et al. Final results of MRC CRASH,
a randomized placebo-controlled trial of intravenous corticosteroid
in adults with head injury-outcomes at 6 months. Lancet 2005;
151 Temkin NR, Anderson GD, Winn HR, et al. Magnesium sulfate for
neuroprotection after traumatic brain injury: a randomized
controlled study. Lancet Neurol 2007; 6: 29–38.
152 Robertson CS, Valadka AB, Hannay HJ, et al. Prevention of
secondary ischemic insults after severe head injury. Crit Care Med
1999; 27: 2086–95.
153 Maas AI, Steyerberg EW, Murray GD, et al. Why have recent trials
of neuroprotective agents in head injury failed to show convincing
effi cacy? A pragmatic analysis and theoretical considerations.
Neurosurgery 1999; 44: 1286–98.
154 Narayan RK, Michel ME, Ansell B, et al. Clinical trials in head
injury. J Neurotrauma 2002; 19: 503–57.
155 Maas AIR, Marmarou A, Murray GD, Teasdale GM, Steyerberg EW.
Prognosis and clinical trial design in traumatic brain injury: the
IMPACT study. J Neurotrauma 2007; 24: 232–38.
156 Visser HK. Non-therapeutic research in the EU in adults incapable
of giving consent? Lancet 2001; 357: 818–19.
157 Tu JV, Willison DJ, Silver FL, et al. Impracticability of informed
consent in the Registry of the Canadian Stroke Network.
N Engl J Med 2004; 350: 1414–21.
158 Corrigan OP, Williams-Jones B. Consent is not enough—putting
incompetent patients fi rst in clinical trials. Lancet 2003;
159 Kompanje EJ, Maas AI. ‘Treat fi rst, ask later?’ Emergency research
in acute neurology and neurotraumatology in the European Union.
Intensive Care Med 2004; 30: 168–69.
160 Annane D, Outin H, Fisch C, Bellissant E. The eff ect of waiving
consent on enrollment in a sepsis trial. Intensive Care Med 2004;