Review Article

Address correspondence to Dr Halinder Mangat, Weill Cornell Medical Center, Department of Neurology and Neuroscience, 525 E. 68th St., Ste F-610, New York, NY 10021, hsm9001@med.cornell.edu. Relationship Disclosure: Dr Mangat reports no disclosure. Unlabeled Use of Products/Investigational Use Disclosure: Dr Mangat reports no disclosure. * 2012, American Academy of Neurology.

Severe Traumatic Brain Injury

Halinder S. Mangat, MD ABSTRACT
Purpose of Review: Although adherence to traumatic brain injury (TBI) guidelines has been associated with improved patient outcomes, guideline adherence remains suboptimal in practice. With neurologists becoming increasingly involved in specialized neurointensive care units and in the care of patients with severe TBI, familiarization with these guidelines is essential. Recent Findings: Intracranial monitoring of different physiologic variables has increased in the past few years. Intracranial pressure (ICP)Ydriven therapy has been replaced by ICPYcerebral perfusion pressure (CPP)Ydriven therapy. More recently, the importance of brain oxygen optimization in addition to ICP-CPP has been recognized, and clinical trials are underway to study the effect of this approach. Surgical management of patients with TBI is also evolving rapidly with further studies on decompressive craniectomy. These are significant advances to improve TBI outcomes. Summary: This article summarizes the routine monitoring of patients with severe TBI and offers insight into some novel physiologic monitoring devices available. The guidelines for management of patients with severe TBI are summarized along with outcome measures.
Continuum Lifelong Learning Neurol 2012;18(3):532546.

INTRODUCTION Injuries are the leading cause of death between the ages of 1 and 44 years.1 Traumatic brain injury (TBI) accounts for 30% of the mortality associated with injury-related death.2 TBI is defined as an alteration in brain function or other evidence of brain pathology caused by an external force.3 A severe TBI is one in which the patient presents with loss of consciousness and a Glasgow Coma Scale (GCS) score of less than 9. Neuroimaging usually shows diffuse or focal lesion(s). Severe TBI results in high mortality whereas mild TBI is increasingly recognized as causing significant functional morbidity. EPIDEMIOLOGY Approximately 1.7 million Americans have a TBI every year.2 This leads to

1.3 million visits to the emergency department, 275,000 hospitalizations, and 52,000 deaths. Fortunately, 75% of all TBIs are mild. The common causes of TBI are falls (35%), motor vehicle accidents (17%), struck-by events (17%), assaults (10%), and unknown causes (21%). The highest incidence of TBI occurs among children younger than 4 years and adults older than 65 years. In these ages, falls are the most common cause. Between ages 14 and 34, there is a high incidence of motor vehicleY related TBI. Patients older than 75 have the highest TBI-related mortality. Men are four times more likely to have a TBI than women in the same age group. Direct medical costs and indirect costs of TBI, such as loss of productivity, totaled $60 billion in the United States in the year 2000.4
June 2012

532

www.aan.com/continuum

Copyright American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

CEREBROVASCULAR PHYSIOLOGY We will first review the cerebrovascular physiology that forms the basis of therapeutic strategies in severe TBI. Monro-Kellie Doctrine The contents of the cranial cavity are brain, CSF, and blood (arterial and venous). In an adult, the brain makes up 80% of the contents, and the CSF and the blood each make up 10%. Because the cranial vault cannot expand, the total intracranial volume remains constant. Therefore, an increase in the volume of one compartment or appearance of a new mass lesion must result in a decrease in the volume of the other compartments. To compensate for any increase in volume due to a lesion, compartments containing blood and CSF experience decreases in volume. Cerebral blood volume is distributed between venous sinuses (two-thirds) and arterioles (onethird). To compensate for the volume of a new lesion (such as a hematoma, cerebral contusion, or cerebral edema),

the venous sinuses collapse readily and push venous blood into the systemic circulation. The arteriolar bed then remains the main blood volume regulator and can compensate for up to 75 mL in volume. The most important blood volume regulator is arteriolar carbon dioxide. Similarly, the CSF compartment compensates by decreasing CSF production and increasing absorption, and some CSF is also pushed into the subarachnoid space from the ventricles.

KEY POINTS

h In an adult, the brain

makes up 80% of the contents of the cranial cavity, and the CSF and the blood each make up 10%.

h Once intracranial
compensatory mechanisms are exhausted, small increases in intracranial volume can cause large increases in ICP, leading to herniation. It is therefore imperative for the neurointensivist to know the compliance status of the brain.

Brain Compliance Once intracranial compensatory mechanisms are exhausted, small increases in intracranial volume can cause large increases in ICP, leading to herniation. It is therefore imperative for the neurointensivist to know the compliance status of the brain. Brain compliance is divided into the following stages, as shown in Figure 2-1:5 1. High compliance + normal ICP 2. Low compliance + normal ICP 3. Low compliance + increased ICP

FIGURE 2-1

Brain compliance curve demonstrating different stages of brain compliance.

Reprinted from Brunicardi FC, Andersen DK, Billiar TR, et al, editors. Schwartzs principles of surgery. 9th ed. New York, NY: McGraw-Hill Professional, 2009.5 B 2009, with permission from The McGraw-Hill Companies, Inc.

Continuum Lifelong Learning Neurol 2012;18(3):532546

www.aan.com/continuum

533

Copyright American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

Traumatic Brain Injury

KEY POINTS

h The Monro-Kellie
doctrine states that intracranial volume remains constant, and addition or increase in the volume of one compartment must cause a decrease in another or result in an increase in intracranial pressure. Cerebral blood flow remains constant at mean arterial pressures between 60 mm Hg and 160 mm Hg by autoregulatory mechanisms.

Progression beyond the third stage causes precipitous rise in ICP and brain herniation. Herniation, however, may also occur at an earlier stage if there are focal lesions with compartmental high ICP. This can occur with temporal lobe contusions and result in uncal herniation. Cerebral Blood Flow Autoregulation Brain perfusion is maintained by a constant cerebral blood flow (CBF) over a wide range of systemic blood pressure (mean arterial pressure [MAP] 60 mm Hg to 160 mm Hg or CPP 50 mm Hg to 150 mm Hg) (Figure 2-2).6 This constant CBF is achieved by a mechanism called pressure autoregulation. This occurs at the arteriolar level by vasodilatation at low perfusion pressures and vasoconstriction at higher perfusion pressures. Beyond these autoregulation thresholds, the arterioles are no longer able to compensate, resulting in abnormal CBF and severe ischemia. The autoregulation curve shifts to the left in states

of high ICP and to the right in longstanding systemic arterial hypertension. Flow-metabolism coupling is the mechanism by which CBF is coupled to cerebral metabolism and is therefore proportional to it. Lowering brain metabolism consequently decreases CBF. This serves as a basis for reducing ICP by sedation and hypothermia. PATHOPHYSIOLOGY OF SEVERE TRAUMATIC BRAIN INJURY Primary Injury The forces of impact determine the nature of primary injury from TBI. The mechanisms implicated are accelerationdeceleration (motor vehicle accidents, falls), rotational (motor vehicle accidents, falls), crush (struck-by events, motor vehicle accidents), and missile (gun shot, shrapnel). Focal lesions include subdural hematomas (SDHs), cerebral contusions with intracerebral hemorrhage, epidural hematoma (EDH), and traumatic subarachnoid hemorrhage (tSAH). Diffuse lesions include tSAH and diffuse axonal injury due to shear forces. The mainstay of reducing primary injury is prevention. Secondary Injury Numerous factors cause secondary injury, and it is imperative to anticipate these and minimize their detrimental effect. Frequently, the initial injury is severe but treatable, such as an acute SDH or EDH. However, patients often deteriorate further and become severely ill due to secondary injuries, such as ischemia, hypoxia, edema, and resultant intracranial hypertension. Hypoxia-ischemia is one of the most important secondary factors that influences outcome after TBI. The most critical time is during resuscitation and immediately post-injury, when systemic hypoxia and hypotension are likely to occur due to either inability to establish an airway or volume loss from hemorrhage.
June 2012

h In addition to initial
injury, secondary injury is caused by hypoxia, hypotension, ischemia, seizures, and metabolic impairment such as mitochondrial injury.

FIGURE 2-2

Autoregulation curve with vessel caliber. CBF = cerebral blood flow; CVR = cerebrovascular resistance.

Reprinted with permission from Rosenthal G, Sanchez-Mejia RO, Phan N, et al. Incorporating a parenchymal thermal diffusion cerebral blood flow probe in bedside assessment of cerebral autoregulation and vasoreactivity in patients with severe traumatic brain injury. J Neurosurg 2011;114(1):62Y70.6 thejns.org/doi/abs/10.3171/2010.6.JNS091360?url_ver=Z39.88-2003& rfr_id=ori:rid:crossref.org&rfr_dat=cr_pub%3dpubmed.

534

www.aan.com/continuum

Copyright American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

Subsequent hypoxia-ischemia also results from vascular injury, systemic causes, cerebral edema (which causes a diffusion gradient for oxygen), intracranial hypertension, and exaggerated hyperventilation. Disordered cerebral metabolism may be due to cellular as well as mitochondrial injury. Mitochondrial membranes and oxidative metabolic processes are disrupted, which results in toxic calcium influx and cellular death. In SDH, PET imaging shows that metabolic impairment occurs well beyond the extent of injury visible on CT imaging.7 CBF dysautoregulation also occurs as a result of severe TBI and is associated with poor outcome.8 Dysautoregulation can be focal or diffuse, and its presence predisposes tissue to hyperperfusion and vasogenic edema. The state of autoregulation can be determined by simple observation of changes in ICP with spontaneous changes in MAP. With loss of autoregulation, ICP increases linearly, with increases in MAP even within the postulated autoregulation range. The pressure reactivity index (PRx) is a moving index of correlation between ICP and MAP and is used as an index of autoregulation. It ranges from j1 to +1. A negative value (including 0) indicates good autoregulation, and a positive value indicates decreased autoregulation. Inflammatory mechanisms are being increasingly recognized as mediators of secondary injury by inducing inflammatory cytokines as well as microglial activation in the brain. Electrophysiologic phenomena such as seizures and epileptiform discharges occur in the postinjury period and lead to detrimental metabolic derangements. Monitoring for these phenomena is important because they can be subclinical.9 Cortical spreading depression (CSD) is an electrophysiologic phenomenon that occurs after TBI and is associated with poor prognosis. CSD
Continuum Lifelong Learning Neurol 2012;18(3):532546

occurs over the injured cortex, involving waves of depolarization that spread at 2 mm/min to 5 mm/min and cause ionic fluxes, metabolic derangements, and vascular flow abnormalities.10 Secondary injury results from a composite of the above processes and leads to cell injury, death, and apoptosis. Resultant inflammation and edema, along with evolution of primary lesions such as cerebral contusions, cause intracranial hypertension. This causes further cellular hypoxia and injury and cascades into a positive feedback cycle. While it is essential to arrest all of these processes, the occurrence of each cannot yet be determined clinically or by bedside monitoring. Therefore, the mainstay of therapy targets treatment of intracranial hypertension, cerebral hypotension, and cerebral hypoxia. EVALUATION AND DIAGNOSIS Early evaluation is essential for early intervention and begins at the scene of the injury. Airway, breathing, and circulation must be assessed. Preliminary neurologic examination at the scene should focus on the GCS (Table 2-1)11 and examination of the pupils. Pupillary size, reactivity, and asymmetry should be noted. Pupillary asymmetry is defined as a size difference greater than 1 mm. A fixed pupil is defined as one that has a response to light of less than 1 mm. Orbital trauma must be noted. The GCS and pupils must be assessed regularly and in the absence of medications that may interfere with an accurate response. A direct transfer must be made to a Level I or Level II trauma center. Mortality increases by 50% when patients are not transferred directly to the appropriate hospital, even if it is not the nearest hospital.12 A complete primary survey by a certified traumatologist is essential to rule out other potentially life-threatening injuries. A neurologic examination must

KEY POINT

h Direct transfer to a Level

I or Level II trauma center improves outcomes, even if such a hospital is not the nearest one.

www.aan.com/continuum

535

Copyright American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

Traumatic Brain Injury

KEY POINTS

h Hypotension and
hypoxemia are independent predictors of poor outcome after severe traumatic brain injury.

TABLE 2-1 Glasgow Coma Scale Response Type

Eye response (E)

Response
Open spontaneously Open to verbal command Open in response to pain No response

Score
4 3 2 1 5 4 3 2 1 6 5 4 3 2 1

h Hypoxia (oxygen
saturation less than 90%) and hypotension (systolic blood pressure less than 90 mm Hg) should be avoided and treated aggressively to improve outcomes after traumatic brain injury. Verbal response (V)

Talking and oriented Confused speech/disoriented Inappropriate words Incomprehensible sounds No response

Motor response (M)

Obeys commands Localizes to pain Flexion/withdrawal Abnormal flexion Extension No response

Total

3 to 15

Reprinted from Jeannett B, Teasdale G, Braakman R, et al. Predicting outcome in individual patients after severe head injury. Lancet 1976;15(7968):1031Y1034.11 B 1976, with permission from Elsevier. www.thelancet.com/ journals/lancet/article/PIIS0140-6736(76)92215-7/abstract.

be done regularly to evaluate for signs of cerebral herniation. These signs include anisocoria, pupillary dilatation, nonreactive pupils, motor extensor posturing, lack of motor response, or a drop in GCS score of 2 points. As soon as hemodynamic stabilization is achieved, the patient must undergo emergent CT of the brain. However, if the patient is comatose or has a focal neurologic deficit, neuroimaging must be expedited. If a vascular injury is suspected, CT angiography must be included. A high index of suspicion for vascular dissection must be maintained, depending on the nature of injury (eg, fracture of carotid canal, motorbike helmet strap injury on neck, fracture dislocation of upper cervical spine). Spine imaging

should also be performed if motor and autonomic system examination reveals evidence of spinal cord injury. CSF leak from nose or ears is associated with a fracture of the base of the skull. Expeditious transfer to an intensive care unit or operating room must be made following investigations and imaging. PREHOSPITAL MANAGEMENT Guidelines are available for the prehospital management of TBI.13 Numerous studies have shown that hypotension14 and hypoxemia15 are independent predictors of poor outcome after severe TBI. An airway must be established, and arterial oxygen saturation below 90% and systolic blood pressure below 90 mm Hg must be avoided and treated
June 2012

536

www.aan.com/continuum

Copyright American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

immediately, as illustrated in Case 2-1. End-tidal carbon dioxide must be maintained between 35 mm Hg and 40 mm Hg. Volume status should be assessed and resuscitation begun. In pediatric patients, critical low threshold for systolic blood pressure is less than 60 mm Hg for 0 to 28 days of age; less than 70 mm Hg for 1 to 12 months of age; less than 70 mm Hg plus 2 times age in years for 1 to 10 years; and less than 90 mm Hg for older than 10 years. Cerebral herniation can be treated with mild hyperventilation to target end-tidal carbon dioxide 30 mm Hg to 35 mm Hg. This is a temporizing measure and should be discontinued once signs resolve and other therapies have been delivered. Hyperventilation should be delivered by administering 20 breaths/ min in an adult, 25 breaths/min in a child, and 30 breaths/min in an infant. INTENSIVE CARE UNIT MONITORING OF A PATIENT WITH SEVERE TRAUMATIC BRAIN INJURY On arrival at the trauma bay, a complete primary survey must be done, followed by management of the most life-threatening injuries. Once diagnostic and immediate

surgical interventions are completed, the patient must be transported to an intensive care unit (ICU). Patients cared for in a specialized neurocritical care unit have improved outcomes.16 Precautions should be taken to assess spinal injury and to document any suspicion thereof. The Brain Trauma Foundation endorses very elaborate yet clear guidelines for the critical care management of a patient with severe TBI.17 The main principles of therapy are to ensure adequate cerebral perfusion and oxygenation while brain swelling and injury improve. The next sections discuss the parameters that can be monitored as surrogate indices of brain ischemia and hypoxia. (A below/above critical threshold value is defined as one that remains abnormal for 5 consecutive minutes and is considered normalized when it has remained within normal range for 5 consecutive minutes.) Blood Pressure and Systemic Oxygenation As reviewed earlier, blood oxygen saturation should remain at greater than 90% throughout the critical care of a patient with severe TBI. In addition, arterial blood sampling must be done frequently to maintain a PaO2 greater

KEY POINT

h Very elaborate yet clear

guidelines for the critical care management of a patient with severe TBI are endorsed by the Brain Trauma Foundation. The main principles of therapy are to ensure adequate cerebral perfusion and oxygenation while brain swelling and injury improve.

Case 2-1
A 24-year-old man was found after his car struck a tree. He was unconscious with no visible external injury. Emergency medical services was called and arrived in 3 minutes. On arrival, paramedics extricated him from the car and stabilized his neck and spine. His airway was assessed as inadequate, his blood pressure was 80/40 mm Hg, his pulse rate was 110 beats/min, and pulse oximetry was 80%. His pupils were asymmetric; the left was 5 mm and unreactive and the right was 3 mm and reactive. He was intubated at the scene and given IV fluids and 100% oxygen. He was mildly hyperventilated en route to the nearest trauma center. Comment. This patient has a severe traumatic brain injury. He has an inadequately secured airway and is hypotensive and hypoxic, all of which need to be treated emergently on the scene. He also has signs of cerebral herniation as evidenced by a dilated and fixed pupil. He should have mild hyperventilation to reverse herniation until arrival at the hospital.

Continuum Lifelong Learning Neurol 2012;18(3):532546

www.aan.com/continuum

537

Copyright American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

Traumatic Brain Injury

KEY POINT

h A sustained ICP greater

than 20 mm Hg is considered harmful and is associated with poor outcome. Higher ICP causes secondary ischemia to the brain. ICP-directed therapy should be used to maintain ICP less than 20 mm Hg.

than 60 mm Hg. Systolic blood pressure must be maintained above 90 mm Hg. Hypotonic and dextrose-containing IV fluids should not be used. Intracranial Pressure ICP can be measured either by an intraventricular device (external ventricular drain connected to an external strain gauge transducer) or by an intraparenchymal probe (strain gauge or fiberoptic). The latter can be either tunneled or placed via a cranial bolt. A sustained ICP greater than 20 mm Hg is considered harmful and is associated with poor outcome.18 Higher ICP causes secondary ischemia to the brain. ICP-directed therapy should be used to maintain ICP less than 20 mm Hg. TBI guidelines recommend that ICP be monitored in all patients with a severe TBI (GCS score less than 9) and an

abnormal CT scan of the brain (Level II recommendation). An abnormal scan is defined as one with hematoma, contusion, swelling, herniation, or effaced basal cisterns. ICP monitoring is also indicated in patients with a normal CT scan of the brain and GCS score less than 9 if they have two of the following three characteristics: age older than 40 years, motor posturing, or systolic blood pressure less than 90 mm Hg. Cerebral Perfusion Pressure CPP is derived from MAP and ICP. CPP equals MAP minus ICP. It can only be determined when ICP measurement is also being performed. The current guideline recommendation is to use ICPand CPP-guided therapy, as illustrated in Case 2-2A. A CPP less than 50 mm Hg is associated with poor outcome even if it

Case 2-2A
A 50-year-old man was admitted to the neurointensive care unit following a motor vehicle accident. His admission Glasgow Coma Scale (GCS) score was 5. His CT scan showed a skull fracture and a large epidural hematoma with contralateral traumatic subarachnoid hemorrhage (Figure 2-3). He had no extracranial injuries. He had been adequately resuscitated and was normotensive and normoxic. The epidural hematoma was operated on and removed. The patient remained comatose with a GCS score of 5. An intracranial pressure (ICP) monitor was placed, and the ICP was 24 mm Hg and the cerebral perfusion pressure (CPP) was 48 mm Hg. Comment. This is an example of a patient with a severe traumatic brain injury who presented with an intracranial lesion, intracranial hypertension, and a poor GCS FIGURE 2-3 Patient with right-side skull fracture, epidural score. Such a patient requires ICP hematoma, and and CPP monitoring. If possible, contralateral traumatic subarachnoid hemorrhage. brain oxygenation should also be monitored.

538

www.aan.com/continuum

June 2012

Copyright American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

occurs only periodically.19 The ischemic threshold seems to lie below a CPP of 50 mm Hg. However, in patents with poor cerebral autoregulatory function, a lower CPP of 50 mm Hg to 60 mm Hg is advised and tolerated because of a higher blood flow at the same CPP. In patients with intact or minimally disturbed autoregulation, a slightly higher CPP of 60 mm Hg to 70 mm Hg is recommended. Overall goal for CPP should be 50 mm Hg to 70 mm Hg. Brain Oxygenation Two different methods are commonly used to measure brain oxygenation. Jugular oximetry is a global measure of cerebral oxygen consumption and extraction. It is measured by sampling blood from a catheter in the internal jugular vein, positioned cranially to the drainage of the facial vein. It serves as a good monitoring marker for hyperventilation. Maintaining the catheter in an appropriate position is technically difficult, which has led to its falling out of favor in clinical practice. When done vigorously, hyperventilation leads to severe vasoconstriction and ischemia and is manifested by low jugular venous saturation (less than 50%). Brain tissue oximetry is a focal measurement (13 mm2) of oxygen tension and is used as a marker for cerebral hypoxia. This is measured via a microprobe in the brain parenchyma and can be tunneled or inserted via a bolt. The measurement is in a small focal brain volume and may not detect events in regions removed from the probe. A drift may occur over longer periods of monitoring (more than 7 days) and at higher temperatures. Per TBI guidelines, there is only a Level III recommendation regarding monitoring and thresholds for jugular venous saturation and brain tissue oxygenation, with critical low thresholds of 50% and 15 mm Hg, respectively. More
Continuum Lifelong Learning Neurol 2012;18(3):532546

recent studies, however, suggest that therapy that includes a brain oximetry target greater than 20 mm Hg, in addition to recommended ICP and CPP targets, improves outcomes.20 A randomized controlled clinical trial (the BOOST 2 trial) is underway to examine the benefit of ICP-, CPP- and oximetrydirected therapy (http://clinicaltrials. gov/ct2/show/NCT00974259). Cerebral Microdialysis Microdialysis is a technique of measuring brain metabolism. It is done by placing a catheter with a filtration membrane at the tip into the brain parenchyma. Using dialysis technology, artificial CSF is perfused through the catheter and interstitial solutes are dialyzed out in microliter volumes. These solutes are then analyzed using an online high-performance liquid chromatography technique. Interstitial concentrations of lactate, pyruvate, glucose, glutamate, and glycerol are obtained. The most common values analyzed are the ratio between lactate and pyruvate (L/P ratio) and the ratio between lactate and glucose (L/G ratio). These ratios provide insight into cerebral energy metabolism, albeit focal, and provide an understanding of the pathophysiology of ongoing secondary injury to the brain. The L/P ratio indicates the balance between aerobic and anaerobic metabolism. Glucose levels signify adequate substrate delivery, and the trends of these values are frequently more important than the values themselves. Abnormal levels of lactate and pyruvate and abnormal L/P and L/G ratios are indicative of both ischemic and nonischemic metabolic stress. While these values are directly proportional, pyruvate is inversely related to the extent of metabolic disorder. Further, pathologic values that are refractory to therapeutic maneuvers are likely due to mitochondrial injury or substrate diffusion gradient due to edema.

KEY POINT

h A cerebral perfusion
pressure less than 50 mm Hg is associated with poor outcome even if it occurs only periodically. Overall goal for cerebral perfusion pressure should be 50 mm Hg to 70 mm Hg.

www.aan.com/continuum

539

Copyright American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

Traumatic Brain Injury

KEY POINT

h Patients with severe

traumatic brain injury have improved outcomes when cared for in a specialized neurointensive care unit. Per guidelines, intracranial pressure and cerebral perfusion pressure should be monitored and treated if they are abnormal. Brain oxygenation, microdialysis, and EEG are other modalities used to monitor brain injury and can help direct therapies that improve outcomes.

The threshold for critical metabolic disorder is thought to be an L/P ratio of 25 to 40. A recent study shows that an L/P ratio greater than 25 is correlated with unfavorable outcome or death.21 The probe is most commonly placed in the frontal lobe on the nondominant side and is assessed on CT by visualizing its gold tip, which is radio-opaque. It is considered to be perilesional if the tip is within 0.5 cm to 1.5 cm of a radiographic lesion, or in normal brain if it is greater than 1.5 cm away. Perilesional values tend to be higher than those in normal brain.22 Just like brain oximetry, the microdialysis measurement is also focal and may miss ischemia distant from the probe location. No prospective randomized trials have evaluated the role of microdialysis in affecting patient outcomes, likely because few patients undergo this monitoring and the patient groups are heterogeneous. However, an International Consensus statement does recommend the use of microdialysis in the management of patients with severe TBI.23 Continuous EEG The risk of early posttraumatic seizures ranges between 10% and 20%. In patients with depressed skull fractures or penetrating injury, continuous EEG may be useful to screen for subclinical seizures. Many different epileptiform discharges are seen in patients with TBI, and they precede actual seizures.9 TREATMENT OF SEVERE TRAUMATIC BRAIN INJURY As discussed above, the aims of ICU therapy of severe TBI are to maintain adequate cerebral perfusion by targeting ICP and CPP. In addition, therapies should be instituted to avoid cerebral hypoxia. In one study, patients who responded to ICP-lowering therapy had a 64% lower risk of death at 2 weeks.24 The target value for ICP is less than 20

mm Hg, for CPP is 50 mm Hg to 70 mm Hg, and for brain oxygenation is greater than 15 mm Hg. Therapies to Lower Intracranial Pressure Head end elevation. The head of the bed should be elevated 30 to 45 degrees. This maximizes cerebral venous drainage via jugular veins. However, the patient should be inclined only if the spine has been deemed stable. Inclination also helps reduce the incidence of ventilatorassociated pneumonia. Ventriculostomy. Insertion of a ventriculostomy drain can be done in the trauma bay of the emergency department, in the operating room, or at the bedside in the ICU. CSF drainage reduces intracranial volume and thus pressure. An external ventricular drain may also be used for ICP measurement by connecting it to an external transducer setup. This transducer must be zeroed at the level of the pinna, which corresponds to the level of the foramen of Monro. The drainage circuit must remain closed at all times, as it is in direct communication with the intracranial cavity. Routine CSF sampling is not necessary because it increases risk of infection. CSF sampling should be done only in the presence of fevers or elevated white blood cell count or if infection is suspected. Ventriculitis is treated by drain replacement, systemic antibiotics, and, on occasion, intrathecal antibiotics. If intrathecal antibiotics are administered, they must be specially prepared for intrathecal use and contain no preservatives, as these can cause chemical meningitis. Hyperosmolar therapy. A dose of 1 g/kg to 1.5 g/kg of 20% mannitol may be administered for cerebral herniation or raised ICP, and doses may be repeated if required. Mannitol should be given only in the setting of raised ICP or herniation, as prophylactic therapy with
June 2012

540

www.aan.com/continuum

Copyright American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

mannitol is of little benefit because of the risk of hypovolemia and hypotension. Hypertonic saline (HTS) may also be used and is available in concentrations ranging from 2% to 23.4%. Both of these osmotic agents work by increasing capillary perfusion by improving red blood cell rheology, augmenting cardiac output, causing water shifts from brain tissue, and improving laminar capillary flow by dehydrating endothelial cells. The effects of mannitol are typically shorter than those of HTS. A single 30 mL dose of 23.4% HTS has been shown to rapidly reverse transtentorial herniation.25 HTS also improves elevated ICP that is refractory to mannitol by improving cerebral oxygenation and hemodynamics.26 Whereas mannitol carries the risk of renal failure, HTS may precipitate pulmonary edema. Serum osmolality and sodium must be checked frequently, and hyperosmotics can be used with ICP spikes to maximal serum osmolality of 330 mOsm/kg (mannitol) or serum sodium 160 mEq/L (HTS).27 Sedation and analgesia. Analgesia is essential to prevent a rise in ICP during stimulation, movement, and nursing procedures such as endotracheal suctioning. In patients with severe intracranial hypertension, these procedures can provoke a sustained ICP crisis. Concomitant use of analgesia and sedation also reduces the total dose of sedative required. The preferred analgesics are opiates. Sedation provides anxiolysis, agitation control, and hypnosis and improves cerebral physiology. It lowers cerebral metabolic rate and thereby CBF because of flow-metabolism coupling. The reduction in CBF reduces ICP. Propofol and benzodiazepines such as midazolam preserve flow-metabolism coupling and are therefore desirable. Inhaled anesthetic agents do not maintain flowmetabolism coupling and are therefore less favored. Most sedative agents cause
Continuum Lifelong Learning Neurol 2012;18(3):532546

hypotension. In addition, the risk of propofol infusion syndrome and hypertriglyceridemia must be kept in mind. Sedation must be titrated to effect for ICP control, and prophylactic burst suppression with any agent is not advisable. Hyperventilation. Hyperventilation causes hypocarbia and vasoconstriction, which reduce CBF and therefore cerebral blood volume. There is a positive linear relationship between arterial carbon dioxide and CBF between a PaCO2 of 20 mm Hg to 80 mm Hg. However, below the threshold of a PCO2 of 28 mm Hg, severe vasoconstriction results in a decrease in CBF to ischemic levels in the brain. The target for mild hyperventilation is a PaCO2 of 32 mm Hg to 36 mm Hg and for moderate hyperventilation is 28 mm Hg to 32 mm Hg. Continuous monitoring of end-tidal carbon dioxide is essential to ensure regulated hypocarbia. Alternately, jugular venous saturations or brain tissue oxygenation must be monitored to avoid cerebral hypoxia and ischemia. Hyperventilation must be avoided in the first 24 hours after a severe TBI as the brain is hypermetabolic and CBF is depressed in relation to metabolic rate. Vasoconstriction in this time period causes ischemia.28 However, transient hyperventilation is an excellent tool to reduce ICP in emergent situations, such as a herniation syndrome, while other therapies to lower ICP are being implemented. Hyperventilation should not be used prophylactically as it increases the ischemic burden of the brain.7 Muscle relaxation. It is important to maintain a constant PaCO2 because of its effect on ICP. Muscle relaxants may be needed for this purpose. This is especially the case if hyperventilation needs to be maintained. Muscle relaxants may also be indicated if hypothermia or controlled normothermia is implemented and causes shivering. When

KEY POINT

h Hyperventilation must
be avoided in the first 24 hours after a severe TBI as the brain is hypermetabolic and CBF is depressed in relation to metabolic rate. Vasoconstriction in this time period causes ischemia.

www.aan.com/continuum

541

Copyright American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

Traumatic Brain Injury

KEY POINT

h Intracranial pressure
should be maintained at less than 20 mm Hg. This can be done using sedation, analgesia, CSF drainage, mild hyperventilation, hypothermia, barbiturate coma, and decompressive craniectomy. Each of these therapeutic modalities has adverse effects of which the physician should be aware.

muscle relaxants are used, all feeding tubes must be postpyloric. Hypothermia. Numerous trials have been conducted to evaluate the benefit of prophylactic hypothermia as a neuroprotective therapy in severe TBI. There appears to be a clear lack of benefit even when hypothermia is administered very early in severe TBI.29 Hypothermia is beneficial, however, in controlling high ICP refractory to other therapies. Mild hypothermia to 35-C for more than 48 hours effectively reduces high ICP and improves outcome in such patients.30 Hypothermia can be induced and maintained using surface cooling devices or IV cooling catheters. Hypothermia causes shivering, immunosuppression, arrhythmias, cold diuresis, myocardial suppression, hypokalemia, and increased vasopressor requirement. Paralysis may be required to control shivering. Rewarming must be done slowly at a rate of 1-C every 8 to 12 hours, and patients should be monitored for hyperkalemia and vasodilatation. Barbiturate coma. High-dose barbiturates used to achieve EEG burst suppression are recommended to control elevated ICP refractory to maximal medical and surgical therapy.27 Barbiturates maintain flow-metabolism coupling (albeit not very closely) and lower CBF and cerebral blood volume. They also acutely lower ICP and long-term ICP trends. Patients with refractory increased ICP who do not respond to barbiturates have a significantly higher mortality than those who respond. Hemodynamic stability is essential before and during therapy because of the hypotension caused by barbiturates. Every effort must be made to avoid systemic hypotension, which will further lower CPP. Pentobarbital is the most common agent used. After a bolus of 10 mg/kg given over 60 minutes, the dose is maintained at 1 mg/kg/h to 5 mg/kg/h to achieve burst suppression (3 to 6 bursts per minute). Con-

tinuous EEG monitoring is essential to monitor burst suppression. The morbidity of high-dose barbiturates is high because of the immunosuppression, myocardial suppression, and hypotensive effects. Decompressive craniectomy. Studies illustrate improvement in ICP after craniectomy,31 and retrospective studies suggest favorable outcomes after decompressive craniectomy.32 However, the recently published Decompressive Craniectomy in Diffuse TBI (DECRA) trial concluded that early bifrontotemporoparietal decompressive craniectomy lowered ICP and length of stay in the ICU but increased mortality.33 In this study, however, all patients who had surgical removal of a mass lesion were excluded, and patients underwent decompressive craniectomy without maximal medical therapy. Another clinical trial, called Randomised Evaluation of Surgery with Craniectomy for Uncontrollable Elevation of Intra-Cranial Pressure (RESCUEicp), is underway and randomizes patients to receive either barbiturate therapy or decompressive craniectomy as the last step in management of intracranial hypertension (www.rescueicp. com). The results of this trial are eagerly awaited. Therefore, while no randomized controlled trials supporting the use of craniectomy in TBI in adults have been reported, nonrandomized trials using historic controls suggest it is beneficial when maximal medical therapy has failed. This is illustrated in Case 2-2B. Therapies to Increase Cerebral Perfusion Pressure Lowering intracranial pressure. The most common cause of low CPP is a high ICP, and lowering ICP often increases CPP. However, sometimes the cause of suboptimal CPP is systemic hypotension or hypovolemia. Fluids. Hypotonic and dextrosecontaining fluids must not be used.
June 2012

542

www.aan.com/continuum

Copyright American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

KEY POINTS

Case 2-2B
The patient presented in Case 2-2A was treated with sedation and 80 g of mannitol. His ICP decreased to 12 mm Hg and his CPP rose to 60 mm Hg. Over the next few days, however, he required frequent hypertonic saline boluses and mild hyperventilation to a Paco2 of 32 mm Hg. His serum sodium was 155 mEq/L. Because his ICP remained above 20 mm Hg, hypothermia (to 35-C) was initiated, and thereafter he underwent a right-sided hemicraniectomy. Following the hemicraniectomy, his ICP was controlled and his neurologic examination started to improve after 10 days. Comment. This patient experienced significant secondary injury and edema, which produced refractory intracranial hypertension requiring maximal medical and surgical therapy.

h Cerebral perfusion
pressure should be maintained at 50 mm Hg to 70 mm Hg. Normovolemia is essential and occasionally induced hypertension may be required.

h Brain oxygenation of
less than 20 mm Hg should be avoided and may be treated with hyperoxia, increased cerebral perfusion pressure, or blood transfusion.

Isotonic or hypertonic fluids are effective for resuscitating patients with severe TBI and shock. Colloids may also be used, although they must be given in judicious quantities to avoid a coagulopathy. Vasopressors. After volume resuscitation, a patient may still have a low CPP, necessitating use of a vasopressor. Noradrenaline has some beneficial effects over dopamine. The former has a more predictable and efficient augmentation of CPP and CBF as estimated by transcranial Doppler.34 Johnston and colleagues35 demonstrated a reduction in oxygen extraction and a significant increase in brain tissue oxygenation when CPP was augmented with noradrenaline. Caution must be exercised if vasopressin is used as it may cause cerebral vasodilatation and increase in ICP. Drugs. Sedatives, analgesics, and barbiturates all cause hypotension, and their doses must be optimized to minimize systemic hypotension. Optimal cerebral perfusion pressure. Optimal CPP is defined as the CPP at which autoregulation status is most intact in a patient. This is measured by plotting CPP against PRx. Patients treated with optimal CPP have improved outcomes.36 Therapies to Increase Brain Oxygenation Lowering intracranial pressure. High ICP causes tissue hypoxia and isContinuum Lifelong Learning Neurol 2012;18(3):532546

chemia. Lowering ICP increases tissue oxygenation. Increasing cerebral perfusion pressure. Low CPP also causes tissue ischemia and hypoxia. A trial of increase in CPP must be done to examine whether cerebral hypoxia is responsive to perfusion pressures. This may require the use of a vasopressor. Increasing fraction of inspired oxygen. Increase in inspired oxygen increases blood oxygen content and delivery to tissue at risk. Hematocrit. Blood transfusion increases the oxygen-carrying capacity and delivery to brain tissue.37 No defined hemoglobin target exists. Hypertonic saline. In patients with ICP refractory to mannitol, HTS increases brain oxygenation by improving capillary perfusion and CBF.26 Boluses of 250 mL of 7.5% HTS may be given. Management of Other Critical Care Issues Seizure prophylaxis. Early seizures cause metabolic stress and acute increases in ICP. Left untreated, seizures can cause hippocampal atrophy.38 While guidelines do not recommend routine seizure prophylaxis to prevent late-onset posttraumatic epilepsy, if a patient is at high risk for early seizures (eg, has a penetrating TBI, depressed skull fracture, cortical contusion, SDH, or EDH),

www.aan.com/continuum

543

Copyright American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

Traumatic Brain Injury

KEY POINTS

h Bilateral unreactive
pupils predict poor outcome and, when present, are the strongest predictor.

h Age, Glasgow Coma

Scale score, pupillary reaction, and CT findings are predictors of outcome after severe traumatic brain injury.

antiepileptics should be used for 7 days. Levetiracetam or phenytoin may be used in usual doses with cost-utility analysis favoring phenytoin. Thromboprophylaxis. TBI is an independent risk factor for deep vein thrombosis. The risk of developing a deep vein thrombosis in the absence of thromboprophylaxis is estimated to be 20%, with an incidence of 0.38% for pulmonary embolism. Mechanical methods such as compression stockings or pneumatic compression stockings should be initiated early unless injury to lower extremities precludes their use. No reliable recommendation exists regarding when to begin pharmacologic prophylaxis. Generally, this may be started once sequential CT of the brain shows no further evolution of hemorrhage and at least 24 hours have elapsed since the last neurosurgical intervention. Nutrition. Patients often experience a systemic and cerebral hypermetabolic state after severe TBI, which may be related to a stress response. Patients not fed in the first 7 days have a fourfold increase in mortality at the end of 2 weeks. Patients given early nutrition have lower mortality and 55% lower infection rates. Current guidelines recommend nutrition at 140% of resting requirement in nonparalyzed patients and at 100% in paralyzed patients. As stated earlier, feeding tubes must be postpyloric in all patients receiving paralytic agents. Glycemic control. While strict glycemic control is recommended, its association with hypoglycemic episodes has been increasingly recognized. The neurologic outcomes among patients with strict control (less than 110 mg/dL) and conventional control (less than 180 mg/dL) are no different.39 Moreover, cerebral microdialysis reveals increased episodes of cerebral hypoglycemia with strict control.40 This deserves further study.

Steroids. Steroids have no role in the treatment of acute TBI. Tracheostomy. Early tracheostomy reduces days on ventilator but does not reduce ventilator-associated pneumonias or mortality. Endocrine abnormalities. A common neuroendocrine abnormality is hypopituitarism, most commonly manifesting as increased vasopressor requirement or cerebral salt wasting. This can be treated in the acute phase with replacement-dose hydrocortisone or fludrocortisone. Diabetes insipidus must be managed aggressively to avoid high serum sodium. Once hypernatremia is present, serum sodium should be lowered gradually so as not to worsen or cause malignant rebound cerebral edema. OUTCOMES The major predictors of outcome after severe TBI are GCS score after full resuscitation, age, pupillary reactivity, CT findings, and presence of major extracranial injury.41 The GCS has a 70% predictive value. Patients with lower GCS scores have a higher probability of poor outcomes. Only up to 20% of patients with a GCS score of 3 will survive. Mortality and outcome worsen with increasing age. Above 60 years, poor outcome increases dramatically. Bilateral unreactive pupils predict poor outcome and, when present, are the strongest predictor. Dilated pupils (greater than 4 mm) are also a poor prognostic indicator. The following CT findings are associated with poor outcome: compressed or absent basal cisterns, tSAH, midline shift, and intracranial lesions. A web-based outcome prediction model is available at the CRASH2 website (crash2.lshtm.ac.uk/Risk%20calculator/ index.html ).41 Adherence to formulated guidelines improves outcomes and decreases hospital expenses.42 Having a guideline-driven institutional ICP-CPP
June 2012

544

www.aan.com/continuum

Copyright American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

protocol helps around-the-clock care of the patient and decreases subjective variations in care.

Spreading depolarizations have prolonged direct current shifts and are associated with poor outcome in brain trauma. Brain 2011; 134(pt 5):1529Y1540. 11. Jeannett B, Teasdale G, Braakman R, et al. Predicting outcome in individual patients after severe head injury. Lancet 1976; 15(7968):1031Y1034. rtl R, Gerber LM, Iacono L, et al. Direct 12. Ha transport within an organized state trauma system reduces mortality in patients with severe traumatic brain injury. J Trauma 2006;60(6):1250Y1256; discussion 1256. 13. Badjatia N, Carney N, Crocco TJ, et al; Brain Trauma Foundation; BTF Center for Guidelines Management. Guidelines for prehospital management of traumatic brain injury 2nd edition. Prehosp Emerg Care 2008;12(suppl 1):S1YS52. 14. Chesnut RM, Marshall LF, Klauber MR, et al. The role of secondary brain injury in determining outcome from severe head injury. J Trauma 1993;34(2):216Y222. 15. Stocchetti N, Furlan A, Volta F. Hypoxemia and arterial hypotension at the accident scene in head injury. J Trauma 1996;40(5): 764Y767. 16. Patel HC, Menon DK, Tebbs S, et al. Specialist neurocritical care and outcome from head injury. Intensive Care Med 2002; 28(5):547Y553. 17. Brain Trauma Foundation, American Association of Neurological Surgeons, Congress of Neurological Surgeons. Guidelines for the management of severe traumatic brain injury [erratum published in J Neurotrauma 2008;25(3):276Y278]. J Neurotrauma 2007;24(suppl 1):S1YS106. 18. Marmarou A, Anderson RL, Ward JD. Impact of ICP instability and hypotension on outcome in patients with severe head trauma. J Neurosurg 1991;75:S59YS66. 19. Stein DM, Hu PF, Brenner M, et al. Brief episodes of intracranial hypertension and cerebral hypoperfusion are associated with poor functional outcome after severe traumatic brain injury. J Trauma 2011;71(12): 364Y373; discussion 373Y374. 20. Spiotta A, Stiefel MF, Gracias VH, et al. Brain tissue oxygen-directed management and outcome in patients with severe traumatic brain injury. J Neurosurg 2010; 113(3):571Y580. 21. Timofeev I, Carpenter KLH, Nortje J, et al. Cerebral extracellular chemistry and outcome following traumatic brain injury: a microdialysis study of 223 patients. Brain 2011;134(pt 2):484Y494.
www.aan.com/continuum

REFERENCES
1. Centers for Disease Control and Prevention. Web-based injury statistics query and reporting system (WISQARS). www.cdc.gov/ injury/wisqars. Accessed December 2, 2011. 2. Faul M, Xu L, Wald MM, Coronado VG. 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, 2010. 3. Menon DK, Schwab K, Wright DW, et al; Demographics and Clinical Assessment Working Group of the International and Interagency Initiative toward Common Data Elements for Research on Traumatic Brain Injury and Psychological Health. Position statement: definition of traumatic brain injury. Arch Phys Med Rehabil 2010;91(11): 1637Y1640. 4. Finkelstein EA, Corso PS, Miller TR. The incidence and economic burden of injuries in the United States. New York: Oxford University Press, 2006. 5. Brunicardi FC, Andersen DK, Billiar TR, et al, editors. Schwartzs principles of surgery. 9th ed. New York, NY: McGraw-Hill Professional, 2009. 6. Rosenthal G, Sanchez-Mejia RO, Phan N, et al. Incorporating a parenchymal thermal diffusion cerebral blood flow probe in bedside assessment of cerebral autoregulation and vasoreactivity in patients with severe traumatic brain injury. J Neurosurg 2011;114(1):62Y70. 7. Coles JP, Fryer TD, Smielewski P, et al. Defining ischemic burden after traumatic brain injury using 15O PET imaging of cerebral physiology. J Cereb Blood Flow Metab 2004;24(2):191Y201. 8. Hiler M, Czosnyka M, Hutchinson P, et al. Predictive value of initial computerized tomography scan, intracranial pressure, and state of autoregulation in patients with traumatic brain injury. J Neurosurg 2006; 104(5):731Y737. 9. Ronne-Engstrom E, Winkler T. Continuous EEG monitoring in patients with traumatic brain injury reveals a high incidence of epileptiform activity. Acta Neurol Scand 2006;114(1):47Y53. 10. Hartings JA, Watanabe T, Bullock MR, et al.
Continuum Lifelong Learning Neurol 2012;18(3):532546

545

Copyright American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

Traumatic Brain Injury

22. Timofeev I, Czosnyka M, Carpenter KLH, et al. Interaction between brain chemistry and physiology after traumatic brain injury: impact of autoregulation and microdialysis catheter location. J Neurotrauma 2011; 28(6):849Y860. 23. Bellander BM, Cantais E, Enblad P, et al. Consensus meeting on microdialysis in neurointensive care. Intensive Care Med 2004;30(12):2166Y2169. 24. Farahvar A, Gerber LM, Chiu Y-L, et al. Response to intracranial hypertension treatment as a predictor of death in patients with severe traumatic brain injury. J Neurosurg 2011;114(5):1471Y1478. 25. Koenig MA, Bryan M, Lewin JL, et al. Reversal of transtentorial herniation with hypertonic saline. Neurology 2008;70(13):1023Y1029. 26. Oddo M, Levine JM, Frangos S, et al. Effect of mannitol and hypertonic saline on cerebral oxygenation in patients with severe traumatic brain injury and refractory intracranial hypertension. J Neurol Neurosurg Psychiatry 2009;80(8):916Y920. 27. Marshall GT, James RF, Landman MP, et al. Pentobarbital coma for refractory intra-cranial hypertension after severe traumatic brain injury: mortality predictions and one-year outcomes in 55 patients. J Trauma 2010;69(2): 275Y283. 28. Coles JP, Minhas PS, Fryer TD, et al. Effect of hyperventilation on cerebral blood flow in traumatic head injury: clinical relevance and monitoring correlates. Crit Care Med 2002; 30(9):1950Y1959. 29. Clifton GL, Valadka A, Zygun D, et al. Very early hypothermia induction in patients with severe brain injury (the National Acute Brain Injury Study: Hypothermia II): a randomised trial. Lancet Neurol 2011;10(2):131Y139. 30. Jiang JY, Xu W, Li WP, et al. Effect of long-term mild hypothermia or short-term mild hypothermia on outcome of patients with severe traumatic brain injury. J Cereb Blood Flow Metab 2006;26(6):771Y776. 31. Weiner GM, Lacey MR, Mackenzie L, et al. Decompressive craniectomy for elevated intracranial pressure and its effect on the cumulative ischemic burden and therapeutic intensity levels after severe traumatic brain injury. Neurosurgery 2010;66(6):1111Y1118; discussion 1118Y1119. 32. Williams RF, Magnotti LJ, Croce MA, et al. Impact of decompressive craniectomy on

functional outcome after severe traumatic brain injury. J Trauma 2009;66(6):1570Y1574; discussion 1574Y1576. 33. Cooper DJ, Rosenfeld JV, Murray L, et al. Decompressive craniectomy in diffuse traumatic brain injury. N Engl J Med 2011; 364(16):1493Y1502. 34. Steiner LA, Johnston AJ, Czosnyka M, et al. Direct comparison of cerebrovascular effects of norepinephrine and dopamine in head-injured patients. Crit Care Med 2004; 32(4):1049Y1054. 35. Johnston AJ, Steiner LA, Chatfield DA, et al. Effect of cerebral perfusion pressure augmentation with dopamine and norepinephrine on global and focal brain oxygenation after traumatic brain injury. Intensive Care Med 2004;30(5):791Y797. 36. Steiner LA, Czosnyka M, Piechnik SK, et al. Continuous monitoring of cerebrovascular pressure reactivity allows determination of optimal cerebral perfusion pressure in patients with traumatic brain injury. Crit Care Med 2002;30(4):733Y738. 37. Zygun DA, Nortje J, Hutchinson PJ, et al. The effect of red blood cell transfusion on cerebral oxygenation and metabolism after severe traumatic brain injury. Crit Care Med 2009;37(3):1074Y1078. 38. Vespa PM, McArthur DL, Xu Y, et al. Nonconvulsive seizures after traumatic brain injury are associated with hippocampal atrophy. Neurology 2010;75(9):792Y798. 39. Coester A, Neumann CR, Schmidt MI. Intensive insulin therapy in severe traumatic brain injury: a randomized trial. J Trauma 2010;68(4):904Y911. 40. Oddo M, Schmidt JM, Carrera E, et al. Impact of tight glycemic control on cerebral glucose metabolism after severe brain injury: a microdialysis study. Crit Care Med 2008;36(12):3233Y3238. 41. MRC CRASH Trial Collaborators; Perel P, Arango M, Clayton T, et al. Predicting outcome after traumatic brain injury: practical prognostic models based on large cohort of international patients. BMJ 2008;336(7641):425Y429. 42. Fakhry SM, Trask AL, Waller MA, et al. Management of brain-injured patients by an evidence-based medicine protocol improves outcomes and decreases hospital charges. J Trauma 2004;56(3):492Y499; discussion 499Y500.

546

www.aan.com/continuum

June 2012

Copyright American Academy of Neurology. Unauthorized reproduction of this article is prohibited.