Professional Documents
Culture Documents
Br.J.Anaesth. 2007 Helmy 32 42
Br.J.Anaesth. 2007 Helmy 32 42
net/publication/6281168
CITATIONS READS
204 2,806
3 authors, including:
Some of the authors of this publication are also working on these related projects:
All content following this page was uploaded by Marcela P Vizcaychipi on 25 January 2016.
Traumatic brain injury (TBI) has a dramatic impact on the intensive care management of patients with severe TBI
health of the nation: it accounts for 15– 20% of deaths in and reviews the rationale for the use of specific neuroin-
people aged 5 – 35 yr old, and is responsible for 1% of all tensive care interventions.
adult deaths.44 Approximately 1.4 million people in the
UK suffer a head injury every year38 resulting in nearly
150 000 hospital admissions per year.40 Of these, approxi-
mately 3500 patients require admission to ICU. The Natural history of severe TBI
overall mortality in severe TBI, defined as a post- An appreciation of the severity of the injury on admission
resuscitation Glasgow Coma Score (GCS) !8, is 23%.42 is beneficial as it predicts the likely prognosis as well as
In addition to the high mortality, approximately 60% of giving some indication of natural history. The most useful
survivors have significant ongoing deficits including cog- classification of TBI is based on the best GCS89 after
nitive competency, major activity, and leisure and recrea- resuscitation, as it has prognostic significance.58 After
tion.28 This has a devastating financial, emotional, and injury to the brain, an inflammatory cascade is initiated
social impact on survivors left with lifelong disability and which results in worsening oedema with vasogenic, cyto-
on their families. toxic, and osmotic components.93 This results in increased
It is well established that the major determinant of pressure within the confines of a fixed intracranial
outcome from TBI is the severity of the primary injury, compartment.
which is irreversible. However, secondary injury, primarily It is important to realize that the type and mechanism of
cerebral ischaemia, occurring in the post-injury phase, injury has an important bearing on the likely clinical
may be due to intracranial hypertension, systemic hypoten- course after TBI. High velocity injuries involving rapid
sion, hypoxia, hyperpyrexia, hypocapnia and hypoglycae- acceleration and deceleration, particularly if there is a
mia, all of which have been shown to independently rotational element, result in shearing forces at the bound-
worsen survival after TBI. In 1996 and 2000 (updated in ary between neocortical grey and white matter. This shear-
2003), the Brain Trauma Foundation published guidelines ing force can lead to widespread disruption of axonal
on the management of severe TBI,9 accepted by the processes which can be visualized histologically as ‘retrac-
American Association of Neurosurgeons and endorsed by tion balls’.81 This type of injury has been termed as
the World Health Organization Committee in diffuse axonal injury (DAI),1 and although features of this
Neurotraumatology. Although many of the recommen- injury are not readily appreciated on CT scan, significant
dations from these guidelines are incorporated into proto- brain swelling is a common consequence. Clinically, DAI
cols for the management of head-injured patients in is recognized by a triad of a consistent mechanism of
individual ICU, there is still wide variation between Units. injury (rapid acceleration/deceleration/rotation, typically
This article outlines the basic principles of the general seen in road traffic accidents), GCS ,8 after resuscitation
# The Board of Management and Trustees of the British Journal of Anaesthesia 2007. All rights reserved. For Permissions, please e-mail: journals.permissions@oxfordjournals.org
Traumatic brain injury
and a CT scan without focal mass lesion but signs of Even with focal lesions, one must always be suspicious of
brain swelling. DAI is a histological diagnosis and the co-existent diffuse injury in the presence of a compatible
term diffuse brain injury is preferred in the ante-mortem mechanism of injury.81 Furthermore, some types of focal
setting. Fine petechial haemorrhages at the grey/white lesion, such as subdural haematoma (Fig. 1B), are associated
junction can sometimes be visualized on CT scan
(Fig. 1A), but magnetic resonance imaging (MRI) is a
Table 1 CT grading system for diffuse brain injury after Marshall and
more sensitive imaging modality, and being increasingly colleagues.52 The cisterns referred to are the ones surrounding the midbrain as
used as a diagnostic tool, in this patient population. assessed on CT head scan, that is, the interpeduncular, ambient, and
The CT grading system for diffuse injury is shown in quadrigeminal plate cisterns
33
Helmy et al.
with more underlying cerebral parenchymal injury than neuroscience critical care unit.65 Most clinically adopted
others, such as extradural haematoma (Fig. 1C). protocols for management of TBI are based around provid-
ing good basic intensive care and interventions to target
cerebral perfusion pressure (CPP) and intracranial pressure
(ICP). An example of such a protocol is given in Figure 2.
Intensive care management of TBI
General approach
Secondary brain insults arise from both systemic and intra- Ventilatory support, sedation, analgesia,
cranial causes and may occur at any time during initial and paralysis
resuscitation and stabilization and during intensive care. Patients with severe head injury require mechanical venti-
Management of TBI in intensive care is targeted at opti- lation70 to maintain an arterial PO2 above 11 kPa and an
mizing cerebral perfusion, oxygenation and avoiding sec- arterial PCO2 between 4.5 and 5 kPa.63 There is no absol-
ondary insults. There is good evidence that protocolized ute contraindication to the use of positive end expiratory
management leads to improved outcome after TBI32 66 and pressure in hypoxaemic patients unless the increase in
Fig 2 Addenbrooke’s NCCU (neurocritical care unit) TBI Protocol. ICP, intracranial pressure; CPP, cerebral perfusion pressure; SvO2 jugular venous
oxygen saturation; SOL, space-occupying lesion; PtO2 partial pressure of tissue oxygen; LPR, lactate pyruvate ratio; EVD, external ventricular drain;
PAC, pulmonary artery catheter.
34
Traumatic brain injury
in ICP. Permissive hypercapnea should be avoided because artery catheter or non-invasive cardiac output monitor
of its cerebral vasodilatory effect that increases ICP. should be considered. Adrenal insufficiency is not uncom-
Adequate sedation minimizes pain, anxiety, and agita- mon after severe TBI,6 and in patients with high inotropic
tion, reduces the cerebral metabolic rate of oxygen con- requirements, a short synacthen should be carried out
sumption, and facilitates mechanical ventilation. This is before initiation of empirical steroid replacement.6
achieved with sedative drugs and opioids.63 A short acting Before ICP monitoring is instituted, hypertension
benzodiazepine such as midazolam is commonly used, should not be treated unless the mean arterial pressure
which is very effective both as a sedative and as an anti- (MAP) is above 120 mm Hg because the high systemic
convulsant, although accumulation is a problem. Propofol blood pressure may be maintaining CBF. After ICP moni-
may have benefits over midazolam because of its superior toring is instituted, the target MAP is determined by the
metabolic suppressive effects, and favourable short half- CPP as discussed later. For the treatment of hypertension
life. However, it is not recommended in hypothermic to achieve CPP targets, an infusion of short acting beta-
patients as it has a tendency to accumulate and precipitate blockers should be titrated against blood pressure. These
hyperlipidaemia. Other reported problems with propofol agents do not cause cerebral vasodilation, when compared
include precipitous cardiovascular collapse95 and the pro- with nitrates and calcium channel blockers, and therefore
35
Helmy et al.
It is unclear whether hyperglycaemia or lack of insulin prophylaxis which all carry side-effects or complications
during the metabolic stress response affects outcome, but that may negatively influence the outcome after TBI. Most
it is clear that an adequate level of glucose in plasma is authors agree that after 72 h post-TBI, UH or LMWH
associated with lower morbidity and better outcome. Van should be commenced, but few support the use of DVT
den Berghe and colleagues94 in 2001 randomized a large chemothromboprophylaxis as early as 24 h after blunt
group of surgical intensive care patients and demonstrated closed head injuries.62 67 Although LMWH seems better
that tight glycaemic control with intensive insulin therapy than UH in preventing DVT, the incidence of adverse
reduced the number of deaths from multiple organ failure events is low with either option.43
with sepsis regardless of whether or not there was a
history of diabetes or hyperglycaemia. In patients with Miscellaneous
TBI, hyperglycaemia was associated with higher ICP, a
As for all intensive care patients, chest physiotherapy, fre-
longer stay in hospital, worse neurological outcome, and
quent turning, eye care, and full hygiene care must be pro-
reduced survival.79 In 2004, Clayton and colleagues18
vided. Laxatives are prescribed to ensure regular bowel
showed a relative reduction in intensive care mortality of
opening to reduce the risk of intra-abdominal hypertension
around 30% in patients with severe head injury after the
and its systemic repercussions. Frequent dressing of cath-
36
Traumatic brain injury
TBI, and 70 mm Hg was adopted as a CPP target in the CBF is reduced and aggressive hyperventilation can com-
first Brain Trauma Foundation guidelines published in pound cerebral ischaemia.57 For this reason, hyperventila-
1996,8 although the ideal CPP target has been a source of tion should not be applied outside a dedicated
contention ever since. Although increasing CPP may be neurointensive care setting when appropriate monitoring,
seen as a useful way of increasing oxygen delivery to the such as jugular bulb oxygen saturation, can be employed.
brain, it comes at a cost. Loss of vascular autoregulation in A PaCO2 target of 4.5 – 5 kPa63 should be used in the first
the injured brain is a common consequence of TBI and instance for those patients with raised ICP, with hyperven-
leads to dissociation between CBF and metabolic require- tilation to 4.0 – 4.5 kPa reserved for those with intractable
ments. In this circumstance, increasing CPP can lead to intracranial hypertension.22
passive increases in blood vessel diameter, increasing cer-
ebral blood volume and consequently ICP. Increased Hyperosmolar therapy
hydrostatic pressure across the cerebral capillary bed can
Hyperosmolar therapy is a key intervention for the man-
also lead to vasogenic oedema, which again increases ICP.
agement of cerebral oedema and raised ICP after TBI. It is
An alternative approach, the Lund protocol, has been
particularly indicated for acute rises in ICP as it has a
suggested which aims to minimize the CPP target to a
rapid effect. Mannitol, an osmotic diuretic, is commonly
37
Helmy et al.
amino acids, to attenuate the osmotic gradient across the evidence suggesting that moderate hypothermia of 338C
BBB.25 47 If the serum osmolarity then falls, the brain is instituted at admission was associated with significantly
susceptible to rebound oedema. During the course of treat- improved outcome at 3 and 6 months post-injury.49
ment of raised ICP, the serum osmolarity can be raised to However, effectiveness of this therapy was not reproduced
145 –155 mmol litre21 and the serum osmolarity to in a multicentre phase 3 clinical trial. However, post hoc
320 mOsm litre21, but if this hyperosmolar state is analysis by the authors demonstrated that older people
reached, osmolarity must not be brought down rapidly may have worse outcomes with hypothermia and patients
with hypotonic fluids. The maintenance fluid of choice is hypothermic on arrival in the Emergency Department have
normal saline with supplemental potassium. more severe injuries thus confounding the results.19 A
TBI patients are susceptible to disorders of salt and water Cochrane review in 2004 analysed 14 trials with a total of
balance. Causes include central diabetes insipidus (CDI), 1094 patients and did not find any evidence supporting the
cerebral salt wasting (CSW) syndrome, and syndrome of use of hypothermia during the treatment of TBI, but did
inappropriate anti-diuretic hormone (SIADH) secretion. find a statistically significant increased risk of pneumonia
Table 2 lists the differentiating features between these syn- and other potentially harmful side-effects.3 However,
dromes. The situation is further complicated by the large delayed hypothermia in the case of uncontrollable intracra-
38
Traumatic brain injury
distinguish between using anticonvulsants in the acute dura opened to allow the brain to expand out of the con-
phase after TBI (first 7 days) and their continued use in fines of the rigid skull. It can be performed unilaterally
the longer term. Anticonvulsant medication in the acute during evacuation of a specific space-occupying lesion or,
phase does not reduce the incidence of post-traumatic sei- in diffuse injury, a bifrontal craniectomy can be used to
zures in the long term14 and is therefore not rec- remove the most anterior part of the skull. There is evi-
ommended.9 For prevention of seizures in the short term, dence to demonstrate that this procedure reduces ICP,97
a Cochrane review suggests a number needed to treat of but a beneficial effect on outcome is yet to be proven. To
10 to benefit one with no impact on outcome or mor- this end, the RESCUEicp study75 is examining decompres-
tality.82 These drugs are not without their side-effects, sive craniectomy vs barbiturate coma for raised refractory
anti-epileptics should therefore not be prescribed unless intracranial hypertension. As the evidence base is currently
there is documented clinical or EEG evidence of seizures.9 limited, this method of ICP control is retained for when
Some neurosurgeons advocate the use of anti-epileptics in other techniques have failed.
certain high-risk groups, such as those with depressed
skull fractures,90 but these must be considered on a
case-by-case basis.
39
Helmy et al.
7 Borzotta AP, Pennings J, Papasadero B, et al. Enteral versus par- frequency of major gastrointestinal bleeding. Pharmacotherapy
enteral nutrition after severe closed head injury. J Trauma 1994; 1999; 19: 452 – 60
37: 459 – 68 28 Dikmen SS, Machamer JE, Powell JM, Temkin NR. Outcome 3 to
8 Brain_Trauma_Foundation. Guidelines for cerebral perfusion 5 years after moderate to severe traumatic brain injury. Arch
pressure. Brain Trauma Foundation. J Neurotrauma 1996; 13: 693–7 Physical Med Rehab 2003; 84: 1449 – 57
9 Brain_Trauma_Foundation. The Brain Trauma Foundation. The 29 Eastern Association for the Surgery of Trauma (EAST). EAST
American Association of Neurological Surgeons. The Joint Practice Management Guidelines Work Group. Practice Management
Section on Neurotrauma and Critical Care. J Neurotrauma 2000; Guidelines for Nutritional Support of the Trauma Patient. Allentown,
17: 457 – 553 PA 2001
10 Brain_Trauma_Foundation. The Brain Trauma Foundation. The 30 Eisenberg HM, Frankowski RF, Contant CF, Marshall LF, Walker
American Association of Neurological Surgeons. The Joint Section MD. High-dose barbiturate control of elevated intracranial
on Neurotrauma and Critical Care. Nutrition. J Neurotrauma pressure in patients with severe head injury. J Neurosurg 1988;
2000; 17: 539 – 47 69: 15 – 23
11 Brooks D. Acid suppression in the critically ill patient: an evi- 31 Eker C, Asgeirsson B, Grande PO, Schalen W, Nordstrom CH.
dence based medicine approach. Medscape Gastroenterol 2004; 6: Improved outcome after severe head injury with a new therapy
http://www.medscape.com/gastroenterology based on principles for brain volume regulation and preserved
12 Bullock R. Mannitol and other diuretics in severe neurotrauma. microcirculation. Crit Care Med 1998; 26: 1881 – 6
40
Traumatic brain injury
46 Kumar MA, Urrutia VC, Thomas CE, Abou-Khaled KJ, 66 Patel HC, Menon DK, Tebbs S, Hawker R, Hutchinson PJ,
Schwartzman RJ. The syndrome of irreversible acidosis after pro- Kirkpatrick PJ. Specialist neurocritical care and outcome from
longed propofol infusion. Neurocrit Care 2005; 3: 257 – 9 head injury. Intensive Care Med 2002; 28: 547 – 53
47 Lee JH, Arcinue E, Ross BD. Brief report: organic osmolytes in 67 Patel NY, Hoyt DB, Nakaji P, et al. Traumatic brain injury: patterns
the brain of an infant with hypernatremia. N Engl J Med 1994; of failure of nonoperative management. J Trauma 2000; 48:
331: 439 – 42 367 – 74; discussion 74 – 5
48 Manninen PH, Lam AM, Gelb AW, Brown SC. The effect of high- 68 Perel P, Yanagawa T, Bunn F, Roberts I, Wentz R, Pierro A.
dose mannitol on serum and urine electrolytes and osmolality in Nutritional support for head-injured patients. Cochrane Database
neurosurgical patients. Can J Anaesth 1987; 34: 442 – 6 Syst Rev (Online) 2006: CD001530
49 Marion DW, Penrod LE, Kelsey SF, et al. Treatment of traumatic 69 Pharmacists ASoHs. ASHP Therapeutic Guidelines on Stress
brain injury with moderate hypothermia. New Engl J Med 1997; Ulcer Prophylaxis. ASHP Commission on Therapeutics and
336: 540 – 6 approved by the ASHP Board of Directors on November 14,
50 Marmarou A. Pathophysiology of traumatic brain edema: current 1998. Am J Health Syst Pharm 1999; 56: 347 – 79
concepts. Acta Neurochir 2003; 86: 7 – 10 70 Piek J. Guidelines for the pre-hospital care of patients with
51 Marmarou A, Signoretti S, Fatouros PP, Portella G, Aygok GA, severe head injuries. Working Group for Neurosurgical Intensive
Bullock MR. Predominance of cellular edema in traumatic brain Care of the European Society of Intensive Care Medicine.
swelling in patients with severe head injuries. J Neurosurg 2006; Intensive Care Med 1998; 24: 1221– 5
41
Helmy et al.
86 Sperry RJ, Bailey PL, Reichman MV, Peterson JC, Petersen PB, 94 van den Berghe G, Wouters P, Weekers F, et al. Intensive insulin
Pace NL. Fentanyl and sufentanil increase intracranial pressure in therapy in the critically ill patients. N Engl J Med 2001; 345:
head trauma patients. Anesthesiology 1992; 77: 416– 20 1359– 67
87 Stiefel MF, Spiotta A, Gracias VH, et al. Reduced mortality rate 95 Warden JC, Pickford DR. Fatal cardiovascular collapse following
in patients with severe traumatic brain injury treated with brain propofol induction in high-risk patients and dilemmas in the
tissue oxygen monitoring. J Neurosurg 2005; 103: 805 – 11 selection of a short-acting induction agent. Anaesth Intensive Care
88 Suarez JI, Qureshi AI, Bhardwaj A, et al. Treatment of refractory 1995; 23: 485 – 7
intracranial hypertension with 23.4% saline. Critical Care Med 96 Weekers F, Giulietti AP, Michalaki M, et al. Metabolic,
1998; 26: 1118 – 22 endocrine, and immune effects of stress hyperglycemia in a rabbit
89 Teasdale G, Jennett B. Assessment of coma and impaired con- model of prolonged critical illness. Endocrinology 2003; 144:
sciousness. A practical scale. Lancet 1974; 2: 81 – 4 5329– 38
90 Temkin NR. Risk factors for posttraumatic seizures in adults. 97 Whitfield PC, Patel H, Hutchinson PJ, et al. Bifrontal decompres-
Epilepsia 2003; 44(Suppl 10): 18 – 20 sive craniectomy in the management of posttraumatic intracranial
91 Temkin NR, Dikmen SS, Wilensky AJ, Keihm J, Chabal S, Winn HR. hypertension. Br J Neurosurg 2001; 15: 500 – 7
A randomized, double-blind study of phenytoin for the prevention 98 Wise BL, Perkins RK, Stevenson E, Scott KG. Penetration of
of post-traumatic seizures. N Engl J Med 1990; 323: 497–502 C14-labelled mannitol from serum into cerebrospinal fluid and
92 Tolias CM, Bullock MR. Critical appraisal of neuroprotection trials brain. Exp Neurol 1964; 10: 264 – 70
42