Current Concepts in the Emergency Management of Severe Traumatic Brain Injury in Children

Authors: Ronald M. Perkin, MD, MA, FAAP, FCCM, Professor and Chairman, Department of Pediatrics, Brody School of Medicine at East Carolina University; Medical Director, Children’s Hospital, University Health Systems of Eastern Carolina, Greenville, NC; James A. Moynihan, DO, Fellow, Pediatric Emergency Medicine, Department of Emergency Medicine, Loma Linda University Children’s Hospital, Loma Linda, CA; and Michael McLeary, MD, Associate Professor of Radiology, Chief of Pediatric Body MRI, Department of Pediatric Radiology, Loma Linda University Children’s Hospital, Loma Linda, CA.

Peer Reviewer: Robert W. Hickey, MD, Assistant Professor of Pediatrics, Division of Pediatric Emergency Medicine, Children’s Hospital of Pittsburgh, PA.

Traumatic injury remains the leading cause of death and a major cause of disability among children around the world.1,2 Each year in the United States, there are more than 100,000 cases of traumatic brain injury (TBI) in children, of which 10-15% are severe, resulting in permanent neurologic damage.3 In addition to emotional burdens to the family, the economic consequences of providing a lifetime of care for a patient with severe head injury can exceed $2 million.3 TBI, isolated or associated with other extracranial injuries, occurs in up to 85% of pediatric traumas, most due to a blunt mechanism.4,5

Although TBI is the most common cause of death in childhood, many of these deaths may be preventable. Inadequate evaluation, resulting in inappropriate treatment, may contribute to approximately 30% of deaths in children with severe trauma.6 Prompt, accurate assessment of the severity of injury and early initiation of appropriate critical care is of crucial importance in preventing these deaths.

The treatment of TBI in children is a complex problem with a very broad scope. Many of the treatments and treatment strategies for infants and children with severe TBI have been extrapolated from those used in adults. This may be problematic for two reasons. First, there may be age-related differences in the pathobiology and pathophysiology of head injury.7,8 Second, no specific therapy for the treatment of severe TBI in pediatric patients has successfully undergone a multicenter trial to unequivocally prove efficacy.9-11 This issue reviews the anatomy, pathophysiology, and current therapy available for severe TBI in children.

— The Editor

Mechanisms of Injury

Though similar to adults in that head trauma in children (older than 4 years old) is often motor vehicle related, the percentage of motor vehicle-related accidents also increases with increasing age: 20% in children 0-4 years of age, and up to 66% in adolescents.12 Younger children more commonly suffer pedestrian or bicycle-related injuries, whereas adolescents are more often injured in motor vehicle accidents as passengers, similar to adults. Motor vehicle injuries also account for the highest proportion of fatal injuries in all groups. Among infants, toddlers, and young children, the major causes of head trauma are assaults/child abuse and falls; they account for nearly 50% of cases. In older children, falls and assaults/abuse result in less than 20% of TBI. In older children, other etiologies of TBI include sports- or recreation-related injuries and penetrating injuries. In later childhood, including adolescence, the leading causes of trauma are similar to those found in adults and include motor vehicle accidents (as a passenger), sports, and recreational activities.8

Pathophysiology

Pathological and experimental data have determined the type of damage that occurs after a closed head injury. At the time of the event, the forces of the impact cause the primary damage through contact of the intracranial contents with the internal bony prominences of the skull.13,14 Acceleration and deceleration can introduce shearing forces that further injure the brain. Following these injuries, secondary damage may occur through swelling and ischemia.13,14

By definition, primary traumatic injury occurs immediately on impact. This may lead to irreversible damage from direct mechanical disruption, and is dependent on the cause and severity of the inciting injury. Secondary injuries are pathophysiologic events occurring within minutes, hours, or days of the primary injury, and may lead to further damage of nervous tissue, prolonging or contributing to neurologic deficits. (See Figure 1.) Secondary insults likely to be responsible for ischemia and other forms of secondary brain damage may be of intracranial or systemic origin.

The primary injury potentially can elicit a secondary response from the brain as a reaction to the injury.8 This secondary response is a cascade of biochemical and physiologic events within the brain that is believed to contribute to the diffuse cerebral swelling and further tissue damage and loss frequently seen following pediatric TBI. It is important to differentiate between the secondary injury from this cascade of events and secondary injury caused by secondary systemic insults. The secondary response normally observed after TBI includes the loss of cerebral autoregulation, breakdown of the blood-brain barrier, intracellular and extracellular edema (cytotoxic vs vasogenic), diffuse and focal edema, and ischemic brain injury. Multiple factors are believed to contribute to the evolution of the injury, including intracranial hypertension from diffuse swelling, ischemia, and vasospasm.

Early outcome studies of severe TBI in children suggested that the diffuse brain swelling was from vasomotor paralysis, cerebrovascular dilation, and increased cerebral blood volume, but not edema.15 On computed tomography (CT) scans, the brains in these patients had increased density due to hyperemia, which was believed to result in the "malignant" brain swelling observed. More recently, this concept has come into question. Early after TBI, cerebral blood flow is low, with an increase in the arterial-jugular venous difference in oxygen content, suggesting early hypoperfusion or ischemia.16-19 It is during this initial period, following the injury with early flow reduction, that oxygen delivery can be marginal, and can be contributory to adverse outcome.19 One-third of patients with severe TBI have "ischemic" cerebral blood flow (< 18 mL/100g/min).8

The implication of "hyperemia" as the primary cause of brain swelling immediately after TBI in children has influenced early therapy. First, vigorous hyperventilation to bring down the cerebral blood volume (CBV) has been advocated. From data now available, it cannot be concluded that CBV is increased substantially with diffuse cerebral swelling.20 Therefore, the routine use of hyperventilation cannot be supported and, in fact, it may induce ischemic brain damage. Second, patients with TBI may be vulnerable to excessive vasoconstriction during acute hyperventilation. Nearly 20% of adult TBI patients and 73% of children developed wide cerebral arteriovenous content differences during hyperventilation.21,22

With the early findings of regional and global ischemia in pediatric head injury and the association of hypoxic-ischemic episodes to outcome,19 the effects of ischemia on the pathophysiologic cascade are believed to remain an important facet to the ongoing secondary injury. Ischemia contributes to energy failure leading to depolarization, calcium accumulation, lactate accumulation, and acidosis.8 This can lead to further cellular breakdown and damage. A local inflammatory response is also initiated in the brain after TBI. Cerebrospinal fluid levels of cytokines (IL-1, IL-6, and IL-10) are increased.23 IL-6 has been shown to be elevated in the cerebrospinal fluid of children early after TBI, to levels comparable to those seen in blood during sepsis.23 Levels of IL-10 were higher in those patients with a poor outcome and young age. Additionally, endogenous cascades as a result of the secondary response to TBI play a role in the worsening of secondary brain injury. The importance of excitatory amino acids (glutamate and aspartate) causing or enhancing secondary brain injury has been reported. 8,24

These biochemical cascades, in combination with the cerebrovascular response and edema, contribute to diffuse cerebral swelling and uncontrolled intracranial hypertension. Additional studies on cerebrospinal fluid in the pediatric age group will be important for understanding the biochemical response to TBI and will be essential to the development of novel treatment modalities to interrupt the ongoing secondary injury.8

The very young developing brain can be particularly susceptible to extensive damage and, as a result, have a greater likelihood of worse outcome. It is during this developmental period that the immature brain can be anatomically more vulnerable to shearing, or to disturbances of autoregulation of cerebral perfusion after the injury.8,25

In addition to the evolution of the secondary damage in the injured brain, injury can be greatly augmented by secondary extracerebral insults such as hypoxemia or hypotension. One group reported a strong association between diffuse brain swelling and either hypoxemia or early hypotension, and commented that these factors play a particular role in the pathogenesis of diffuse swelling frequently seen in children.26 Children are prone to suffer a hypoxemic or hypotensive episode post trauma and, as a result, are more likely to develop diffuse swelling.

Primary Injury

Concussion. Traditionally, concussion refers to a brief loss of consciousness after a closed head injury and is accompanied by a flaccid motor state followed by complete recovery.13 More recently, experts have placed such mild head injury at the lower end of a spectrum of conditions caused by diffuse axonal injury. There is evidence for disruption of axonal functioning after such injury.27 Frequently, a brief period of both retrograde and anterograde amnesia follows this event, and there may also be longer-term sequelae.28

Fractures. Skull fractures frequently accompany closed head injury. Fractures may be a relatively benign event causing no neurological insult, or they may be found in association with severe damage to the brain. Skull fractures are generally divided into two types: linear and depressed.13 Linear fractures are caused by deformation of the skull over a wide area and may be located over the cranial vault or the basilar skull.29 Closed linear fractures are of little consequence themselves, but they may signify underlying pathology. Depressed fractures, on the other hand, are segments of the bone that are driven into the cranium below the surrounding skull surface.

If a skull fracture crosses a dural vessel, such a vessel may be torn, which may lead to an epidural hematoma. Skull fractures occur in up to 25% of pediatric TBI patients and are less commonly associated with epidural hematoma (40%) than in adults (61%).8 Sufficient force to cause a fracture may also damage the intracranial contents. An open linear fracture leaves the bone and the dura exposed and raises the possibility of infection.

The skull base is a frequent site of fracture and accounts for approximately one-fifth of all cases of skull fractures.30 The weaker areas of the skull, including the petrous temporal bone, the foramen magnum, and parts of the sphenoid bone and sinus, are typical sites.13,31 Fractures in these locations may tear the dura and involve air sinuses, which then allow ingress of infecting organisms; a fistula may also form, allowing cerebral spinal fluid (CSF) to leak. The incidence of intracranial infections is approximately 10% in such cases.13,32 In addition, the cranial nerves are at risk for injury as they exit through the base of the skull. Vascular injury, particularly to the internal carotid artery, may also occur. The well known raccoon’s eyes and Battle’s sign may accompany an anterior skull base fracture or a temporal bone fracture, respectively.

Hematoma and Hemorrhage. Following closed head injury, bleeding can occur at a number of sites within the cranium. Outside the dura, epidural hematomas may form; intradural hematomas include subdural and intracerebral hematomas. The pathophysiology of these two types of injury differ significantly. Traumatic subarachnoid hemorrhage also frequently accompanies closed head injury. Focal injuries such as subdural, epidural, and intracerebral hematomas occur with a higher incidence in adults (30-42%) vs. children (15-20%).8

Epidural Hematoma. As a result of closed head injury, laceration of meningeal vessels, typically branches of the middle meningeal artery, can result in the formation of an extradural or epidural hematoma, which is often accompanied by a linear fracture. Epidural hematomas result from brief linear contact forces that commonly occur in unintentional falls.33 The squamous part of the temporal bone is a common site for such injuries. Bleeding from these vessels strips the dura away from the inner table of the skull. (See Figure 2.) As the hematoma grows in size, progressive compression of the brain and elevation in intracranial pressure (ICP) lead to fatality. Only 3-13% of epidural hematomas occur in the occipital region, which is the most common site for a venous epidural hematoma. A wide spectrum of clinical presentations may occur with this lesion, including minimal to severe alterations in consciousness.34 The classical presentation of a lucid interval followed by clinical deterioration is nonspecific for epidural hematoma. Rapid CT scanning and surgical intervention have dramatically reduced the morbidity and mortality due to this lesion.

Subdural Hematoma. In contrast to epidural hematoma, the characteristic lesion in subdural hematomas is accumulation of blood in the subdural space, that is the space between the dura and the arachnoid. These lesions are commonly graded in terms of age: acute, subacute, and chronic. Subdural hematomas form from forceful acceleration and deceleration of the cranium. In the acute phase, a torn bridging vein or, less frequently, a damaged surface cerebral artery, allows egress of blood into the subdural space.13 This accumulation occurs most commonly over the cerebral convexities in the frontal and temporal areas. (See Figure 3.) The blood may cause mass effect on the underlying brain and result in elevated ICP leading to neurological dysfunction and ischemia. In addition, acute subdurals are commonly associated with underlying parenchymal injury which may lead to swelling and shift of the intracranial contents out of proportion to the extent of the hematoma.35 Recent studies point to continued poor outcomes in acute subdural hematoma, despite early surgical intervention.35

If the acute hematoma is present for more than three days, it may appear less dense on imaging studies (i.e., CT) and form a subacute subdural. A remote head injury (older than 2 weeks) may lead to the appearance of a chronic subdural hematoma. This subdural fluid collection has a density only slightly greater than CSF on CT. As the clot of an acute subdural liquifies, it is replaced by the dark fluid that is characteristic of chronic subdural hematomas. This fluid may also be surrounded by vascular membranes. This chronic collection can cause mass effects on the brain and be responsible for seizures, progressive dementia, or alterations in level of consciousness.

Intracerebral Hematomas and Contusions. Moderate and severe closed head injuries are frequently associated with intracerebral hematomas. The stress of shear forces tears small vessels within the brain parenchyma and produces an accumulation of blood, most commonly in the frontal and temporal lobe white matter.13 Higher impact injuries can produce intracerebral hematomas deep in the basal ganglia, corpus callosum, or brain stem. (See Figure 4.)

It is well documented that intracerebral hematomas may arise long after closed head injury, although initial imaging studies may have been negative.36

Points of contact between the brain and the skull may lead to contusions following head injury.13 In these areas, necrotic tissue and ruptured vessels make up the damaged area. The frontal and temporal tips or poles are the most common sites of this type of injury. Contusions below the area of impact on the skull are termed coup contusions, whereas contusions occurring on the side of the brain opposite the site of impact are called contrecoup contusions. Both types of lesions can lead to delayed swelling and hemorrhage in the brain and neurological deterioration.

Subarachnoid Hemorrhage. Trauma frequently leads to subarachnoid hemorrhage and is, indeed, the main cause of this condition.13 Superficial vessels running in the subarachnoid space may be lacerated by the forces from an injury. (See Figure 5.) Subarachnoid hemorrhage following trauma may lead to vasospasm and further ischemic injury to the brain, though vasospasm may also occur in the absence of subarachnoid hemorrhage after trauma.37 Diffuse subarachnoid hemorrhage following trauma is associated with a high mortality rate.13

Diffuse Axonal Injury. The shearing forces that accompany closed head injury can directly damage the axons in the central nervous system (CNS).13 The importance of this mode of injury, called diffuse axonal injury (DAI), has come to be recognized in a wide spectrum of insults. Clinical experience has demonstrated that there are patients who suffer closed head injury and never recover consciousness, but nevertheless lack large macroscopic intracranial lesions such as contusions, hematomas, or diffuse edema. Pathological studies have shown that these patients suffer shearing injuries of axons in the white matter as well as rents in the rostral brainstem and corpus callosum.13 Pioneering work in microscopic examination of these lesions demonstrated that axons under these conditions underwent transection and characteristic formation of retraction balls. In the severest form, lesions in the brainstem and corpus callosum lead to permanent loss of consciousness.13 Less severe forms, which do not involve the brainstem, lead to temporary alterations in consciousness and neurologic deficits. The least severe forms of DAI may correlate with concussion. DAI may also accompany contusions and subdural and intracerebral hematomas. Lesions that occur with DAI are commonly small and noted bilaterally at the gray-white matter interfaces of the frontal lobes and anterior parietal lobes. They are also commonly located in the corpus callosum, basal ganglia, and brainstem. The orientation of the lesions are along the white matter tracts and are, therefore, typically larger in their sagittal and vertical dimensions than in their transverse dimension. DAI lesions are usually nonhemorrhagic and are often not seen on CT. Magnetic resonance imaging (MRI) utilizing T2, GLAIR, or T1 sequences, easily identifies nonhemorrhagic and small hemorrhagic lesions. Therefore, MRI may be indicated in the persistently unresponsive traumatized child with a negative CT.38 (See Figure 6.)

Hypoxic Ischemic Injury (HII). HII is a major cause of morbidity and mortality in the traumatized pediatric patient. CT is usually the first neuroimaging test completed in a traumatized child with HII. The CT scan may be negative if it is completed soon after the traumatic episode. The interval between cerebral insult and sufficient cerebral edema to be detected on CT can be as long as 12 and 24 hours, but has been identified as early as 77 minutes.39 Early CT findings of edema may present as multifocal areas of low density with loss of the gray-white matter junction, which may later become diffuse.

Profoundly injured children may manifest diffuse cerebral edema that predominately involves the cerebral cortex and subcortical white matter, but spares the basal ganglia, thalami, brainstem, and cerebellum. This has been called the "reversal sign" due to low density of the peripheral cerebrum and the relatively high density of the basal ganglia, brainstem, and cerebellum.

Clinical Assessment of Acute Coma in Children

Persistent CNS damage remains one of the main causes of mortality and long-term morbidity in previously healthy children who sustain acute TBI. In emergencies, intensive resuscitative measures, surgery, and specific therapeutic intervention are employed to limit the risk of severe cortical damage.

In children, coma is the most prominent clinical presentation of acute brain injury which, if precipitated by trauma, may set in after a longer lucid interval than in adults. In supratentorial lesions, progressive rostral-caudal deterioration (central syndrome), uncal syndrome, or combined central and uncal herniation can be identified by careful clinical assessment of consciousness; respiratory pattern; pupillary, corneal, and motor responses; and oculovestibular reflexes. The Glasgow Coma Scale (GCS), devised for the clinical assessment of the progression of the central syndrome, is used primarily for bedside monitoring of the degree of impairment of consciousness.40 Scores are assigned for the severity of impairment, with specific focus on three cortically determined functions: motor responses, verbal responses, and eye opening. In the initial phase of the injury, such "cortical" aggregates of gross neurological function are very helpful as reliable measures of change when used by experienced and highly trained observers.41

The GCS requires an adult level of development. Modifications of the GCS or other coma scales have been introduced and found to be valid for assessing children whose neurodevelopmental function is below 10 years.42 An assessment of the merits of six different coma scales in pediatric practice showed that interobserver agreement was highest for the pediatric coma scale, which includes age-adjusted verbal and motor-response scores.43-45

Infants with severe TBI frequently require endotracheal intubation and mechanical ventilation as part of stabilization. The presence of an endotracheal tube prevents any assessment of verbal response. Furthermore, sedatives, analgesics, muscle relaxants, and anticonvulsants also confound clinical assessment. Both of these commonly encountered issues are of particular relevance in clinical neurological assessment since these children may be the very individuals in whom baseline assessment is critical for determining further investigations or interventions. In this context, a recent study of intubated children has suggested that a grimace score might overcome the difficulty of assessing verbal response in these patients.46 The rationale for the score is that, since facial expression, which is an important part of non-verbal communication, is not totally independent of verbal language skills, a grimace can be used as a surrogate for this aspect of cortical assessment. There is moderate to good interobserver reliability for the grimace score.

In post-traumatic head injury in adults, a GCS of 8 or less signifies a high likelihood of raised intracranial pressure and the need for a strict regimen of special care.47 The main reason for monitoring patients by use of a score is to identify pathological changes at a stage when they are evolving and potentially reversible either by emergency surgery or other measures. However, there is no information on whether a pediatric GCS of 8 in children has the same connotation as in adults. Furthermore, there is no information on whether a score of 8 based on a verbal response carries the same implications as the same score based on a grimace.

The pediatric trauma literature is replete with a perplexing variety of scales and scoring systems, often with similar names or abbreviations.48 Scoring systems are designed to enhance effective prehospital triage of trauma patients, organize and improve trauma system resource planning, allow accurate comparison of different trauma populations, and serve as quality assurance filters in trauma patient care. This aspect of pediatric trauma care is in evolution. Only when there is international consensus on what scoring system is ideal will it be possible to assess the effectiveness of many of the therapeutic interventions used in traumatic coma.42 Then it may also be possible to deal with the issue of assessment of prognosis.

Initial Assessment and Management

In the initial stabilization of all trauma patients, including head-injured children, control of the airway and adequate ventilation ("breathing") are the first priorities. In TBI, it is particularly important to maintain oxygenation and prevent hypoxemia, since even moderate reductions in PaO2 can contribute to secondary neural injury in the injured brain.8 The traumatically injured brain is particularly susceptible to secondary insults such as hypoxia-ischemia. In addition, even moderate hypoxia (PaO2 < 40-50 mmHg), which might not reach a level that affects cerebral viability, is a potent vasodilator and may contribute to "cerebral swelling." It is also important to maintain normocarbia, since even moderate hypercarbia can cause arteriolar vasodilation and increased cerebral blood volume, which could further contribute to increased intracranial pressure and precipitate herniation. Hypercarbia (and hypoxemia) may have several causes in head-injured patients, especially in the field, where poor airway control and respiratory failure are common. Hypercarbia has been found in 15-20% of head-injured patients and can be prevented by intubation and ventilation.12

Emergent Airway Control

The intubation sequence for head-injured patients should preserve cerebral oxygenation, prevent aspiration of gastric contents, avert hypercarbia, maintain systemic blood pressure, minimize increases in intracranial pressure (ICP), and avoid aggravation of cervical spine injury.49 Orotracheal intubation of hemodynamically stable, sedated, paralyzed patients is most likely to provide such conditions if no anatomic airway abnormalities or facial fractures are present. Because of the possibility of cervical spine injury accompanying TBI, it is desirable for an assistant to apply manual in-line axial stabilization during laryngoscopy.

Both thiopental and etomidate dose dependently reduce CMRO2, CBF, and ICP.49-51 Etomidate is less likely to precipitate hypotension in hypovolemic patients.49 Supplementation with lidocaine, 1.5 mg/kg IV, will blunt sympathetic responses to intubation and limit associated increases in ICP but may contribute to hypotension.49 Midazolam decreases CBF, does not increase ICP, and provides satisfactory hemodynamic stability when given in small doses (< 0.15 mg/kg over 15 secs).50 Propofol reduces ICP and CBF in head-injured humans.52 However, propofol-induced hypotension may be poorly tolerated, especially if hypovolemia has not been completely corrected. Muscle relaxants should be chosen primarily to facilitate rapid, effective laryngeal exposure.

Once intubated, patients will need sedation to remain calm, relieve pain, and prevent rises in intracranial pressure; fentanyl, morphine, or propofol are frequently used for this purpose.50,53

Controlled Ventilation

The use of controlled ventilation has long been a modality for the treatment of intracranial hypertension and is based on the known cerebrovascular response to changes in PaCO2. The relative change in CBF during variations of PaCO2 depends on several factors, including baseline CBF, cerebral perfusion pressure (CPP), and anesthetic drugs.54 However, in a wide variety of subjects and conditions, most studies report a change in global CBF of 1-2 mL/100g/min for each 1 mmHg change in PaCO2. One group suggested that intracranial hypertension in children should be effectively treated almost exclusively with vigorous hyperventilation.15 Random and "blind" hyperventilation recently has come into question as a therapeutic intervention since it was found to worsen outcome in adults with severe TBI.21 Hyperventilation therapy is not recommended in the first 24 hours after head injury, especially as a prophylactic therapy.55 This position is supported by the results of reviews and studies in adults and children that, together, indicate that hypocapnia in the setting of acute head injury may cause harm by inducing cerebral ischemia.55 In addition, experimental models have demonstrated that the effect of hyperventilation on arteriolar diameter is short lived, with its therapeutic effect lasting only 20 hours.8 In contrast, the loss of cerebrospinal fluid bicarbonate buffer that occurs from sustained hyperventilation makes the cerebral circulation more sensitive to abrupt changes in PaCO2. This limited therapeutic effect, in conjunction with a concern of decreasing perfusion in the early period after injury, has prompted the use of moderate hyperventilation (PaCO2 of 35 mmHg) that can aid intracranial pressure management without inducing ischemia.21,22 Current recommendations for PaCO2 management after TBI discourage the use of prophylactic hyperventilation and suggest that hyperventilation should be used only when increased ICP is refractory to other methods of control.54 (See Table 1.)

Table 1. Current Concepts in the Management of Severe TBI
Protect airway, Control Ventilation
      • Intubation with appropriate medications and precautions
       • Head position (elevate if not hypotensive)
       • Sedation
       • Selective use of muscle relaxants
Avoid Prophylactic Hyperventilation (PaCO2 < 35 mmHg)
       • Particularly true in first 24 hours
       • In cases of hyperemia, use hyperventilation titrated to measured adequacy of CBF
       • Hyperventilation should always be used in the setting of impending herniation
Avoid Hypotension
       • Aggressive hemodynamic support
       • Maintain CPP (MAP-ICP)
       • > 70 mmHg in adolescents
       • > 60 mmHg in children
       • > 45 mmHg in infants
       • No hypotonic fluids
       • Avoid glucose containing fluids; monitor glucose level
       • Hypertonic saline solutions may be beneficial
Avoid Hyperthermia
       • Mild hypothermia may be beneficial (32-33°C)
       • Careful/controlled rewarming
Treat Seizures

In addition to assessing and controlling PaCO2, the maintenance of good cerebral oxygenation is also necessary for the prevention of secondary injury. It has been shown that hypoxemia (defined as apnea or cyanosis in the field) or a PaO2measurement of less than 60 mmHg by arterial blood gas analysis, early after injury, contributes importantly to poor outcome.12

Mean Arterial Blood Pressure

Cerebral vascular mechanisms contributing to secondary cerebral ischemia include cerebral vasoconstriction (especially in the first few hours after TBI) and impaired pressure autoregulation. Chesnut, when reviewing the outcome of 493 patients entered into the Traumatic Coma Data Bank, found that hypotension (systolic blood pressure of < 90 mmHg) at the time of admission or during intensive care was associated with dramatic deterioration in the likelihood of favorable outcome.56

The influence of hypotension on outcome in head-injured children is at least as great as in head-injured adults. One group reported a mortality rate of 66% in children in whom systolic blood pressure was less than 90 mmHg on admission to the emergency room, in comparison to 22% in children who were normotensive on admission.57 The authors compared these data with mortality after head injury in the National Pediatric Trauma Registry and found a similar effect of hypotension on mortality. Another group identified hypotension as one of four variables, along with the GCS score, respiratory rate, and multisystem anatomic injury, that predicted outcome after head injury.58 In younger children (ages 0-4 years), hypotension is more common than in older children and adults; the mortality rate is also higher in these younger children.12,59-61 Diffuse brain swelling, an ominous prognostic feature in both children and adults, is strongly associated with early hypoxia and hypotension.26

Therefore, to minimize mortality and morbidity, severely head-injured patients require effective and prompt resuscitation, especially prevention and treatment of hypotension.56 Arterial blood pressure management after TBI is one of the most commonly targeted areas for intervention. Cerebral perfusion pressure (CPP), the pressure decrease across the vascular bed of the brain, is the mean arterial pressure minus ICP or central venous pressure, whichever is higher. After TBI, short-term neuronal function and long-term outcome are related to maintenance of a minimal CPP. Contemporary management of the infant or child with severe TBI includes maintenance of an adequate perfusion that can usually be achieved with normotension or mild systemic hypertension.8 Although induced hypertension could eventually be used as an intervention to increase cerebral perfusion pressure, it has not yet been proven efficacious in TBI.62,63

Since cerebral injury can be complicated by either partial or complete loss of pressure autoregulation, significant systemic hypertension would contribute to further hyperemia and cerebral swelling. In some cases, however, a higher systemic pressure is necessary to maintain adequate perfusion to compromised brain regions. Careful monitoring of intracranial pressure and cerebral blood flow is necessary to determine the impact of blood pressure manipulation. Induced hypertension could have particular effectiveness in patients in whom pressure autoregulation of cerebral blood flow is intact. If the autoregulatory mechanisms for blood pressure are intact, and cerebral perfusion pressure is reduced with hypotension, cerebral vasodilation occurs to maintain cerebral blood flow, increasing cerebral blood volume and intracranial pressure.64 In such cases, increasing cerebral perfusion pressure by increasing mean arterial pressure would lead to cerebrovascular constriction, decreased CBV, and intracranial pressure, but preserved CBF. It is possible that mean arterial pressure could be optimized to maximize adequate CBF and CBV and reduce ICP.65

Because of the uniqueness of children and age and size variances for the optimal cerebral perfusion pressure, the issue of optimal mean arterial pressure in children is complex. It is becoming clear that, in the future, management of the patient with TBI will need to be tailored and titrated individually so as not to iatrogenically worsen an already compromised brain. Based on available data, hypotension should be aggressively managed with fluids and, if necessary, using pressor support with dopamine, neosynephrine, or other agents as indicated to optimize and maintain adequate cerebral perfusion pressure. (See Table 1.)

Fluid Resuscitation

Frequently, head-injured patients have lost a significant amount of intravascular volume, and resuscitation with intravenous fluid can dramatically improve neurological status. Isotonic solutions appear to be the best for resuscitation.13 Hypo-osmolar solutions such as dextrose (5%) and water cause a large amount of cerebral edema in experimental models.66 Isotonic saline (0.9%), on the other hand, expands the intravascular volume well and is inexpensive. Development of brain edema from movement of the solution into the brain’s extracellular space has been a concern. However, clinical studies have shown that large amounts of isotonic saline can be given without fear of elevating ICP as long as hypo-osmolality of the serum does not develop.67

For clinical purposes, it is important to note that lactated Ringer’s solution is slightly hypotonic, that 0.9% saline is slightly hypertonic, and that markedly hypertonic saline solutions may be used to reduce ICP.68 Although the effects on brain water or ICP of lactated Ringer’s solution or 0.9% saline may be undetectable when administered slowly in small volumes, rapid administration in large volumes may be sufficient to upset osmotic equilibrium and alter brain water in areas where the blood-brain barrier is intact.

Because the blood-brain barrier enhances the influence of brain water on changes in serum sodium, hypotonic solutions, including lactated Ringer’s solution, are more likely to increase brain water content than 0.9% saline or colloids.69 However, after TBI, the blood-brain barrier may be damaged.

Colloid in the form of albumin or hetastarch has appealing properties because these solutions remain in the intravascular space longer. However, clinical studies have not shown any benefit from the use of these for resuscitation from trauma, and their cost is prohibitive.70,71 In addition, crystalloid resuscitation has been associated with a lower mortality rate in trauma patients.70

Hyponatremia

Hyponatremia has long been recognized as a potentially serious metabolic consequence of CNS insult in adults and children.72,73 Hyponatremia is often seen in head-injured patients and is often attributed to inappropriate secretion of antidiuretic hormone. Recent studies have shown that hyponatremia in many patients with intracranial injury may actually be caused by cerebral salt wasting, in which a renal loss of sodium leads to hyponatremia and a decrease in extracellular fluid volume. The appropriate treatment of cerebral salt wasting and fluid and salt replacement, is opposite from the usual treatment of hyponatremia caused by inappropriate secretion of antidiuretic hormone. Severe hyponatremia, or a rapidly falling serum sodium level, can lead to cerebral swelling and ischemia, altered neurologic status, and seizures; all of which can result in significant secondary injury. Early diagnosis and effective treatment of hyponatremia is critical in TBI patients.

Glucose

Glucose-containing solutions should be avoided.74 Hyperglycemia causes dose-dependent increases in the severity of ischemic and traumatic neurologic injury. Clinical studies demonstrate a strong association between plasma glucose and neurologic outcome after TBI, although it is difficult to separate the hyperglycemic effects of more severe trauma from the injury-aggravating effect of hyperglycemia. In head-injured patients receiving intravenous glucose during general anesthesia, a significant association was also noted between serum glucose and outcome.74 Of those head-injured patients in whom serum glucose exceeded 150 mg/dL on the first postoperative day, 23 of 55 patients had a favorable outcome. In contrast, of those patients with a serum glucose of less than 150 mg/dL, 27 of 32 had a favorable outcome. The inescapable clinical implication is that glucose should be given during resuscitation of head-injured patients only if they are at risk for hypoglycemia. Unfortunately, data are insufficient to provide guidance regarding the advisability of attempting to tightly control hyperglycemia with insulin during intensive care.

Hypertonic Saline Solutions

In experimental animals, resuscitation with hypertonic saline solutions (3.0% to 7.5%) is associated with a lower ICP than resuscitation with isotonic or slightly hypotonic fluids.75,76 Prehospital resuscitation of hypotensive head-injured patients with 7.5% saline, with or without added dextran, was associated with improved outcome compared with conventional fluid resuscitation.77

The ability of hypertonic saline to reduce ICP has been appreciated since the early 1900s. This has been demonstrated in pediatric patients with TBI. In head-injured children, one group reported that intravenous administration of 3.0% saline promptly reduced ICP in most patients.78 In contrast, administration of comparable volumes of 0.9% saline produced no consistent change in ICP. One group further addressed the issue of fluid composition used in fluid resuscitation after severe TBI in children. These authors designed a prospective, randomized, controlled, unblinded study comparing fluid management, using lactated Ringer’s solution vs. hypertonic saline (598 mOsm/L) for the first three days after head injury in 35 children.79 Treatment of severe head injury with hypertonic saline was superior to that treatment with lactated Ringer’s solution. An increase in serum sodium concentration significantly correlated with lower ICP and high CPP. Importantly, these investigators reported no adverse effects associated with the use of hypertonic saline.

Hypertonic saline exerts its effect primarily through establishment of an osmotic gradient between the intravascular space and the cerebral tissue. Water diffuses passively from the cerebral intracellular and interstitial space into capillaries, thereby reducing cerebral water content and ICP. In addition, hypertonic saline is an effective plasma volume expander, providing hemodynamic support in trauma victims, and it may have beneficial effects on cerebrovascular regulation.80

Some concerns exist about the adverse effects of hypertonic resuscitation, especially adverse neurologic sequelae.1,80 However, no adverse effects of supraphysiologic hyperosmolarity such as renal failure, pulmonary edema, or central pontine demyelination were noted in a recently published study of 68 pediatric patients with severe TBI treated with 3% NaCl via continuous infusion.81

Osmotic Therapy

Mannitol is commonly used for the treatment of intracranial hypertension and has replaced other osmotic diuretics over the last 20 years. There are probably two distinct mechanisms by which mannitol exerts its beneficial effects. By increasing the osmotic pressure of the plasma, osmotic diuretics extract water out of the brain. The effect on the ICP is delayed and the maximum effect may not be seen for 20-40 minutes or more.82 Second, immediate plasma expansion reduces blood viscosity, increases cerebral blood flow, and increases oxygen delivery. In patients with intact cerebral autoregulation, this leads to vasoconstriction and reduced intracerebral blood volume. The net effect is reduced ICP and increased or unchanged CBF.83 This mechanism explains the immediate ICP response to mannitol and its decreased effect when the CPP is already more than 70 mmHg.83

The beneficial effect of mannitol on ICP, CPP, and CBF is generally accepted, although there has never been a controlled study testing mannitol against placebo. Doses known to be useful for these purposes range from 0.25 to 1.5 g/kg body weight.82

Mannitol is excreted entirely by the kidneys. There is a significant risk of acute renal failure with its administration in large doses, particularly if other nephrotoxic drugs are administered, or in the presence of pre-existing renal failure.81-83 Serum osmolarity should be kept below 320 mOsm when there is concern for renal failure.84 It is important to maintain an adequate volume status, as hypotension increases the risk of ischemic brain damage and can produce intracranial hypertension.84,85 With prolonged administration, mannitol can accumulate in the brain. This may cause a reverse osmotic effect followed by a rise in ICP.86

Mannitol vs. Hypertonic Saline

Hypertonic saline may have advantages over mannitol bolus infusions in the management of TBI. The blood-brain barrier is more able to exclude sodium chloride than mannitol, making hypertonic saline the ideal agent to reduce cerebral edema.80 Mannitol will move water out of the brain for a shorter duration of time than hypertonic saline.77,81 Depending upon the tonicity used, hypertonic saline requires a smaller volume of infusion compared with mannitol.77 Hypertonic saline will not result in hypotension, and unlike mannitol, will not result in acute renal complications due to osmotic diuresis.75

Given the amount of evidence in its favor, hypertonic saline is seeing surprisingly little clinical use in the resuscitation and management of head-injured patients.84,87,88 However, resuscitation of patients with TBI and shock remains one of the most challenging aspects of trauma care.87,89 Controversy persists regarding the most appropriate type, timing, and volume of resuscitation of trauma victims. Further study and consideration is required because hypertonic saline appears to be a logical and beneficial choice for the management of TBI.

Temperature and Metabolism

Pyrexia is found in 85% of head-injured patients during their ICU stay and has been found to correlate with a bad outcome.14,90 It has been well-recognized that whole body oxygen consumption is proportional to body temperature, with a 10-12% increase per 1°C core temperature increase.91 In animals, a strictly comparable injury produces more severe functional and histologic injury when body temperature is raised by only 1°C.92 Moreover, in brain-injured patients, brain temperature is usually above rectal, jugular vein, or bladder temperature by 0.3-1.9°C.93,94 There is now sufficient evidence that hyperthermia is deleterious to the injured brain. Aggressive treatment of pyrexia using drugs, neuromuscular blockers, and external or body cavities cooling is effective in reducing this type of secondary brain injury.

Attempts at using hypothermia decades ago did not generate ongoing interest. Recently, however, several factors have renewed interest in the therapeutic role of hypothermia. First, experimental models demonstrated that mild hypothermia (32°-33°C) could reduce CNS damage in head injury, even when instituted after the insult. Second, it was recognized that the severe systemic complications that were seen with more profound hypothermia were virtually absent when mild hypothermia was employed. Third, the critical care services available for these patients are now able to provide the systemic support necessary to safely maintain a hypothermic state for 1-3 days.

Two randomized trials of moderate hypothermia (32°C for 24 or 48 hrs) showed beneficial effects in adults after severe TBI.95,96 Seizure incidence was lower in the hypothermic patients, and ICP was decreased during the hypothermic interval. In the most recent study by Marion and colleagues, neurologic recovery was hastened in the subgroup of patients with an initial GCS score of 5 to 7.97 However, a multicenter trial of therapeutic hypothermia in severely traumatic brain-injured adults has recently been discontinued due to apparent lack of effect.3

In conclusion, hyperthermia must be aggressively treated in brain-injured patients, but the use of moderate hypothermia to prevent secondary insult in injured brains cannot be routinely supported, although it shows some promise. Patients who are moderately hypothermic in the ED should not be aggressively rewarmed.

Anticonvulsant Therapy

Numerous studies have examined the occurrence of post-traumatic seizures following head trauma.13 Seizures can lead to cerebral hypermetabolism, elevated intracranial hypertension, and hypoxia. Thus, a seizure may precipitate further injury to an already compromised CNS.

Several risk factors for post-traumatic seizures have been identified: intracerebral, subdural, and epidural hematomas; cortical contusions; depressed skull fractures; penetrating head injuries; and seizures within 24 hours of injury.98,99 Several studies have shown that anticonvulsants, particularly phenytoin or tegretol, are effective in preventing early post-traumatic seizures, but they have no advantage in the late post-traumatic period.99,100 Anticonvulsant therapy, therefore, is suggested during the first week after moderate to severe head injury, particularly in the presence of one of the lesions mentioned. Typically, patients with severe head injury are loaded with intravenous phenytoin (15-20 mg/kg). The loading dose is given slowly (< 50 mg/min) because of the possibility that arrhythmias or hypotension may occur.

Role of Barbiturates

High-dose barbiturates have, for a long time, been known to lower the ICP.82 Intracerebral vasoconstriction and reduction of CBF and intracranial blood volume are responsible for the reduction of the ICP in response to barbiturates.101 In addition, barbiturates reduce the cerebral metabolic rate, which may be protective against cerebral ischemia.

Based on the available data, the Brain Trauma Foundation recommends high-dose barbiturate therapy in hemodynamically stable patients with severe head injury and ICP refractory to maximal medical and surgical treatment.102 Of all barbiturates, pentobarbital has been used most frequently, and various dosing regimens have been used. EEG monitoring may be a more reliable way of monitoring barbiturate administration than following plasma levels.101

When barbiturates are used, it is essential that hypotension is avoided, and patients should be carefully monitored for reduction in cardiac output or inadequate systemic perfusion. Pulmonary, renal, and hepatic complications are also common with the administration of high-dose barbiturates in patients with head injury.82

Glucocorticoids

Glucocorticoids can improve altered vascular permeability, reduce cerebrospinal fluid production, and attenuate free radical production. Steroids are commonly used in the treatment of brain edema. The administration of glucocorticoids to patients with brain tumors often leads to marked clinical improvement. One early study found a favorable dose-related effect on mortality with the administration of steroids to patients with head injury, but the rate of severely disabled and vegetative patients was much higher in the treatment group and, therefore, the overall outcome was not improved.82 In the following years, five large prospective double-blind studies found no benefit in the administration of glucocorticoids to patients with severe head injury.103-107 Based on the currently available data, there is no role for glucocorticoids in the treatment of patients with head injury.

Child Abuse

Trauma is the most common cause of death in childhood, and inflicted head injury is the most common cause of traumatic death in infancy.108-112 On average, among children hospitalized for blunt trauma, those injured by abuse sustain more severe injuries, use more medical services, and have worse survival and functional outcome than children with unintentional injuries.111 This is despite a plethora of interventions developed over the last 30 years, including legislatively mandated reporting and the establishment in 1974 of the National Center on Child Abuse and Neglect as a mechanism to increase knowledge of the problem and identify steps to prevent it.111

The major issue plaguing the description of abuse-related injuries to young children has been and continues to be accurate diagnosis.110,113 The dire consequences of either false-positive or false-negative diagnosis intensifies the need to establish accurate diagnostic criteria.

Head injury in infants and toddlers can be difficult to diagnose because symptoms are often nonspecific. Vomiting, fever, irritability, and lethargy are common symptoms of a variety of conditions seen in children, including head trauma.114 When caretakers do not give a history of injury and the victim is preverbal, an abusive head injury can be mistakenly diagnosed as a less-serious condition. When physicians fail to recognize that the child’s symptoms are secondary to nonaccidental head injury, the child is frequently re-injured or has serious complications of the unrecognized, untreated head injury.114

CT scanning is a mainstay of the diagnosis of nonaccidental head injury.115,116 (See Figure 7.) Subdural or subarachnoid hemorrhage can nearly always be detected on CT scans, although the more subtle findings may be missed by less experienced observers.108,115,116 MRI is useful in detecting and characterizing small extraaxial hemorrhages in infants with equivocal CT findings.108,117

Several selected points culled from an extensive literature review may prove useful in recognizing nonaccidental trauma in infants and children.112,118-125 (See Table 2.)

Table 2. Abusive Head Trauma: Hints to Facilitate an Early Diagnosis
Falls from heights: general comments
  • Severe head injury and central nervous system damage or injury of any type are extremely rare when children, ages younger than 5 years, fall out of bed.119
  • Simple falls from heights rarely, if ever, result in significant primary brain injury.120
  • Skull fractures are infrequent consequences of short falls, serious accident injuries are rare in day care centers, and most children who fall from considerable heights (>10 feet) do not die.121
  • Falls from less than 4 feet result in the following possible injuries:
     Soft tissue injury
     Ping-pong fracture
     Concussion
     Epidural hematoma
     Linear skull fracture120
  • Falls from greater heights can have these additional findings:
     Depressed fracture
     Subarachnoid hemorrhage
     Basilar fracture
     Contusion120
  • Subdural hematomas and retinal hemorrhages are virtually never seen in short falls.110
Babies with bruises
  • With advancing age and motor skills, bruises occur more commonly. Bruises in mobile infants are normally located on the front of the body, typically on the shins and forehead.122,123
  • Bruises in infants younger than 9 months who are not yet beginning to ambulate should lead to consideration of abuse or illness.122
  • Bruises in toddlers that are located in atypical areas, such as the trunk, hands, or buttocks, should be considered abuse.122
Stairway-related injuries124
  • Fall down stairs Superficial injury (usually)
Involves only one body area (very few extremity or truncal injuries)
  • Significant injuries:
      Concussion
      Skull fracture
      Cervical spine fracture
  • Infants who fall with caretaker:
      Fall from height plus fall down stairs
      Higher incidence of severe head injury
      Significant injuries
      Skull fracture (46%)
      Contusion
      Subdural hematoma124,125

Conclusion

Head trauma is a leading cause of death and disability and can be particularly devasting for the family of the pediatric patient. Although the primary TBI is not amenable to treatment, conventional management is focused on lessening the effects of the secondary physiologic events and the secondary insults that occur after the injury. Much of the treatment modalities used for the child after severe TBI have been extrapolated from the adult data, since there is little literature that primarily involves children. Children as a group have better outcome than adults, but many factors influence prognosis in the pediatric population. The mechanism of injury, injury severity, multiple trauma, secondary insults, or the extent of secondary injury can all affect the final outcome and are directly and indirectly related. It is clear that many of the poor outcomes observed are best treated by preventing either the initial impact or the secondary insults that typically occur following TBI.

Better understanding of the mechanisms of secondary neural injury following head trauma eventually may open avenues of intervention that will effectively and reliably improve the still-disappointing outcomes of severe head injury. Future research, both in the laboratory and at the bedside, and improved diagnostic and monitoring technology will yield more scientifically based protocols for prehospital and ED management of head-injured patients.

Finally, several promising treatment strategies are being investigated: 1) more selective, less toxic anti-excitotoxic drugs; 2) antioxidants with better brain and cellular penetration; 3) pharmacologic and molecular therapies that inhibit endogenous cell death effectors and augment endogenous neuroprotective genes; and 4) utilization of trophic factors such as nerve growth factor. Hopefully, novel specific therapies on the horizon will take the treatment of acute brain injury to the next level.

References

1. Orliaquet G, Lejay M. Pediatric trauma patients. Curr Opin Crit Care 1999;5:506-510.

2. Rodriquez JG. Childhood injuries in the United States. Am J Dis Child 1990;144:625-626.

3. Clarke RSB, Statler KD, Ruppel RA, et al. Neuroprotective strategies for the treatment of severe traumatic brain injury: Past, present, and future. In: Fineman JR, Giroir BP, eds. Current Concepts in Pediatric Critical Care – 2000. Anaheim: Society of Critical Care Medicine 2000 69-83.

4. Knapp JF. Practical issues in the care of pediatric trauma patients. Curr Probl Pediatr 1998;28:305-328.

5. Orliaquet GA, Meyer PG, Blanot S, et al. Predictive factors of outcome in severely traumatized children. Anesth Analg 1998;87:537-542.

6. Sharples PM, Storey A, Aynsley-Green A, et al. Avoidable factors contributing to death of children with head injury. Br Med J 1990;300:87-91.

7. Kochanek PM, Bell MJ, Forbes ML, et al. Pathophysiology. In: Marion DW, ed. Traumatic Brain Injury. . New York: Thieme Medical Publishers 1999 233-256.

8. Adelson PD, Kochanek PM. Head injury in children. J Child Neurol 1998;13:2-15.

9. Doppenberg EMR, Bullock R. Clinical neuro-protection in severe traumatic brain injury: Lessons from previous studies. J Neurotrauma 1997;14:71-74.

10. Young A, Willatts S. Controversies in management of acute brain trauma. Lancet 1998;352:164-165.

11. Allen CH, Ward JD. An evidence-based approach to management of increased intracranial pressure. Crit Care Clinics 1998;14:485-495.

12. Levin HS, Aldrich EF, Saydjari C, et al. Severe head injury in children: Experience of the traumatic coma data bank. Neurosurgery 1992;31:435-444.

13. Benardete EA, Jallo GI. Closed head injury: Pathophysiology, diagnosis, and management. Trauma Quarterly 1998;13:397-430.

14. Ravussin P, Braco D, Moesehler O. Prevention and treatment of secondary brain injury. Curr Opin Crit Care 1999;5:511-516.

15. Bruce DA, Alavi A, Bilaniuk LT, et al. Diffuse cerebral swelling following head injuries in children: The syndrome of "malignant brain edema." J Neurosurg 1981;54:170-178.

16. Muizelaar JP, Marmarou A, DeSalles AAF, et al. Cerebral blood flow and metabolism in severely head-injured children. Part 1: Relationship with GCS score, outcome, ICP, and PVI. J Neurosurg 1989;71:63-71.

17. Adelson PD, Clyde B, Kochanek PM, et al. Cerebral blood flow and CO2 vasoresponsivity following severe traumatic brain injury in children. J Neurosurg 1996;84:357A.

18. Sharples PM, Stuart AG, Matthews SDF, et al. Cerebral blood flow and metabolism in children with severe head injury. Part 1: Relation to age, Glasgow coma score, outcome, intracranial pressure, and time after injury. J Neurol Neurosurg Psychiatry 1995;58:145-152.

19. Robertson CS, Constant CF, Narayan RK, et al: Cerebral blood flow, AVDO2, and neurologic outcome in head-injured patients. J Neurotrauma 1992;9:S349-S358.

20. Zwienenberg M, Muizelaar JP. Severe pediatric head injury: The role of hyperemia revisited. J Neurotrauma 1999;16:937-943.

21. Muizelaar JP, Marmarou A, Ward JD, et al. Adverse effects of prolonged hyperventilation in patients with severe head injury: A randomized clinical trial. J Neurosurg 1991;75:731-739.

22. Skippen P, Secar M, Poskill K, et al. Effect of hyperventilation on regional cerebral blood flow in head-injured children. Crit Care Med 1997;25:1402-1409.

23. Bell M, Adelson PD, Doughty L, et al. Interleukin-6 (IL-6), and interleukin–10 (IL-10) in the cerebrospinal fluid (CSF) following severe traumatic brain injury in children. J Neurotrauma 1996;13:596-599.

24. Baker AJ, Moultan RJ, MacMillan VH, et al. Excitatory amino acids in cerebrospinal fluid following traumatic brain injury in humans. J Neurosurg. 1993;79:369-372.

25. Graham DI, Ford I, Adams J, et al. Fatal head injury in children. J Clin Pathol 1988;42:18-22.

26. Aldrich EF, Eisenberg HM, Saydijari C, et al. Diffuse brain swelling in severely head-injured children. A report from the NIH Traumatic Coma Data Bank. J Neurosurgery 1992;76:450-454.

27. Mittl RL, Grossman RI, Hiehle JF, et al. Prevalence of MR evidence of diffuse axonal injury in patients with mild traumatic brain injury and normal head CT findings. Am J Neuroradiol 1994;15:1583-1589.

28. Esselman PC, Uomoto JM. Classification of the spectrum of mild traumatic brain injury. Brain Inj 1995;9:417-424.

29. Yuoganandan N, Pintar FA, Sances A, et al. Biomechanics of skull fracture. J Neurotrauma 1995;12:659-668.

30. Koonsman M, Dunn E, Hughes K, et al. How much monitoring is needed for basilar skull fractures? Am J Surg 1992;164:487-490.

31. McElhaney JH, Hopper RH, Nightingale RW, et al. Mechanisms of basilar skull fracture. J Neurotrauma 1995;12:669-678.

32. Choi D, Spann R. Traumatic cerebrospinal fluid leakage: Risk factors and the use of prophylactic antibiotics. Br J Neurosurg 1996;10:571-575.

33. Shugerman RP, Paez A, Grossman DC, et al. Epidural hemorrhage: Is it abuse? Pediatrics 1996;97:664-668.

34. Beni-Adani L, Flores I, Spektor S, et al. Epidural hematoma in infants: A different entity? J Trauma 1999;46:306-311.

35. Zumkeller M, Behrman R, Egmont-Heissler H, et al. Computed tomographic criteria and survival rate for patients with acute subdural hematoma. Neurosurgery 1996;39:708-712.

36. Mertol T, Guner M, Acar U, et al. Delayed traumatic intracerebral hematoma. Br J Neurosurg 1991;5:491-498.

37. Chan KH, Dearden NM, Miller JD. The significance of post-traumatic increase in cerebral blood flow velocity. Neurosurgery 1992;30:697-700.

38. Gilman S. Imaging the brain. N Engl J Med 1998;338:889-896.

39. Klienman P and Barnes P: Head Trauma. In:Klienman P, ed. Diagnostic Imaging of Child Abuse. St. Louis: Mosby; 1998:319-325.

40. Teasdale G, Jennett B. Assessment of coma and impaired consciousness: A practical scale. Lancet 1974;2:81-84.

41. Rowley G, Fielding K. Reliability and accuracy of the Glasgow coma scale with experienced and inexperienced users. Lancet 1991;337:535-538.

42. Gemke RJBJ, Tasker RL. Clinical assessment of acute coma in children. Lancet 1998;351:926-927.

43. Morray JP, Tyler DC, Jones TK et al. Coma scale for use in brain injured children. Crit Care Med 1984;12:1018-1020.

44. Simpson DA, Cockington RA, Hanich A, et al. Head injuries in infants and young children: The volume of the Pediatric Coma Scale. Review of the literature and report on a study. Childs Nerv Syst 1991;7:183-190.

45. Yager JY, Johnston B, Seshia SS. Coma scales in pediatric practice. Am J Dis Child 1990;144:1088-1091.

46. Tatman A, Warren A, Williams A, et al. Development of a modified pediatric coma scale in intensive care clinical practice. Arch Dis Child 1997;77:519-521.

47. Kirkpatrick PJ. On the guidelines for the management of the severe head injury. J Neurol Neurosurg Psychiatr 1997;62:109-111.

48. Furnival RA, Sehunk JE. ABCs of scoring systems for pediatric trauma. Pediatr Emerg Care 1999;15:215-223.

49. Gerardi MJ, Sacchetti AD, Cantor RM, et al. Rapid-sequence intubation for the pediatric patient. Ann Emerg Med 1996;28:55-74.

50. Mirski MA, Muffelman B, Ulatowski JA, et al. Sedation for the critically ill neurologic patient. Crit Care Med 1995;23:2038-2053.

51. Movius AJ, Martin LD. Sedation, analgesia, and neuromuscular blockade during pediatric mechanical ventilation. Respir Care Clin N Am 1996;2:509-543.

52. Sanchez-Izquierdo-Riera JA, Caballero-Cubedo RE, Perez-Vela JL, et al. Propofol versus Midazolam: Safety and efficacy for sedating the severe trauma patients. Anesth Analg 1998;86:1219-1224.

53. Joseph MH, Brill J, Zeltzer LK. Pediatric pain relief in trauma. Pediatr Rev 1999;20:75-83.

54. Brian JE Jr. Carbon dioxide and the cerebral circulation. Anesthesiology 1998;88:1365-1386.

55. Tasker RC. Neurologic critical care. Curr Opin Pediatr 2000;12:222-226.

56. Chesnut RM. Avoidance of hypotension: Conditio sine qua non of successful severe head-injury management. J Trauma 1997;42:S4-S9.

57. Pigula FA, Wald SL, Shackford SR, et al. The effect of hypotension and hypoxia on children with severe head injuries. J Pediatr Surg 1993;28:310-314.

58. Gotschall CS, Papero PH, Synder HM, et al. Comparison of three measures of injury severity in children with traumatic brain injury. J Neurotrauma 1995;12:611-615.

59. Luerssen TG, Klauber MR, Marshall LF. Outcome from head injury related to patient’s age. A longitudinal prospective study of adult and pediatric head injury. J Neurosurgery 1988;68:409-416.

60. Berger MS, Pitts LH, Lovely M, et al. Outcome form severe head injury in children and adolescents. J Neurosurg 1985;62:194-199.

61. Kokoska ER, Smith GS, Pittman T, et al. Early hypotension worsens neurological outcome in pediatric patients with moderately severe head trauma. J Pediatr Surg 1998;33:333-338.

62. Bedell EA, Prough DS. Should induced hypertension be beneficial after traumatic brain injury? Anesth Analg 1998;87:751-753.

63. Robertson CS, Valadka AB, Hannay J, et al. Prevention of secondary ischemic insults after severe head injury. Crit Care Med 1999;27:2086-2095.

64. Rosener MJ, Daughton S. Cerebral perfusion pressure management in head injury. J Trauma 1990;8:933-941.

65. Bouma GJ, Muizelaar JP, Stringer WA, et al. Ultra-early evaluation of regional cerebral blood flow in severely head-injured patients using xenon-enhanced computerized tomography. J Neurosurg 1992;77:360-368.

66. Shapira Y, Artru AA, Qassam N, et al. Brian edema and neurologic status with rapid infusion of 0.9% saline or 5% dextrose after head trauma. J Neurosurg Anesthesiol 1995;7:17-25.

67. Scalea TM, Maltz S, Yelon J, et al. Resuscitation of multiple trauma and head injury: Role of crystalloids fluids and inotropes. Crit Care Med 1994;22:1610-1615.

68. Prough DS. Resuscitation of head-injured patients. In: Fineman JR, Giroir BP, eds. Current Concepts in Pediatric Critical Care – 2000. Anaheim: Society of Critical Care Medicine; 2000:47-55.

69. Zornow MH, Todd M, Moore SS. The acute cerebral effects of changes in plasma osmolality and oncotic pressure. Anesthesiology 1987;67:936-941.

70. Choi PT, Yip G, Quinonez LG, et al. Crystalloids versus colloids in fluid resuscitation: A systemic review. Crit Care Med 1999;27:200-210.

71. Schierhout G, Roberts I. Fluid resuscitation with colloid or crystalloid solutions in critically ill patients: A systematic review of randomized trials. BMJ 1998;316:961-964.

72. Kappy MS, Ganony CA. Cerebral salt wasting in children: The role of atrial natriuretic hormone. Adv Pediatr 1996;43:271-308.

73. Harrigan MR. Cerebral salt wasting syndrome: A review. Neurosurgery 1996;38:152-160.

74. Lam AM, Winn HR, Cullen BF, et al. Hyperglycemia and neurologic outcome in patients with head injury. J Neurosurg 1991;75:545-551.

75. Buckman RF, Badellino MM. Current status of hypertonic saline resuscitation. Trauma Quarterly 1997;13:281-296.

76. Schmoker JD, Zhuang J, Shackford SR. Hemorrhagic hypotension after brain injury causes an early and sustained reduction in cerebral oxygen delivery despite normalization of systemic oxygen delivery. J Trauma 1992;32:714-720.

77. Vassar MJ, Perry CA, Holcroft JW. Prehospital resuscitation of hypotensive trauma patients with 7.5% NaC1 versus 5% NaC1 with added dextran: A controlled trial. J Trauma 1993;34:622-632.

78. Fisher B, Thomas D, Peterson B. Hypertonic saline lowers raised intracranial pressure in children after head trauma. J Neurosurg Anesth 1992;4:4-10.

79. Simma B, Burger R, Falk M, et al. A prospective, randomized, and controlled study of fluid management in children with severe head injury: Lactated Ringer’s solution versus hypertonic saline. Crit Care Med 1998;26:1265-1270.

80. Khanna S, Davis D, Peterson B, et al. Use of hypertonic saline in the treatment of severe refractory posttraumatic intracranial hypertension in pediatric traumatic brain injury. Crit Care Med 2000;28:1144-1151.

81. Peterson B, Khanna S, Fisher B, et al. Prolonged hypernatremia controls elevated intracranial pressure in head-injured pediatric patients. Crit Care Med 2000;28:1136-1143.

82. Paczynski RP. Osmotherapy-basic concepts and controversies. Critical Care Clin 1997;13:105-129.

83. Sauthoff H, Doerfler ME. Critical care of the patient with severe head injury. Trauma Quarterly 1998;13:373-395.

84. Chesnut RM. The management of severe traumatic brain injury. Emerg Med Clin North Am 1997;15:581-604.

85. Gruen P, Lin C. Current trends in the management of head injury. Emerg Med Clin of North Am 1998;16:63-83.

86. Kaufmann AM, Cardosa ER. Aggravation of vasogenic cerebral edema by multiple-dose mannitol. J Neurosurg 1992;77:584-589.

87. Dolich MO, Cohn SM. Solutions for volume resuscitation in trauma patients. Curr Opin Crit Care 1999;5:523-528.

88. Prough DS, Lang J. Therapy for patients with head injuries: Key parameters for management. J Trauma 1997;42:S10-S18.

89. Clark RSB, Kochanck PM. Pass the salt? Crit Care Med 1998;26:1161-1162.

90. Jones PA, Andrews PJD, Midley S, et al. Measuring the burden of secondary insults in head-injured patients during intensive care. J Neurosurg Anesth 1994;6:4-14.

91. Bruder N, Raynal M, Pellisier D, et al. Influence of body temperature, with or without sedation, on energy expenditure in severe head-injured patients. Crit Care Med 1998;26:568-572.

92. Wass CT, Lanier WL, Hofer RE, et al. Temperature changes of > 1°C alter functional neurologic outcome and histopathology in a canine model of complete cerebral ischemia. Anesthesiology 1995;83:325-335.

93. Rumana CS, Gopinath SP, Uzura M, et al. Brain temperature exceeds systemic temperature in head-injured patients. Crit Care Med 1998;36:562-567.

94. Henker RA, Brown SD, Marion DW. Comparison of brain temperature and rectal temperatures in adults with severe head injury. Neurosurgery 1998;42:1071-1075.

95. Shiozaki T, Sugimoto H, Taneda M, et al. Effect of mild hypothermia on uncontrolled intracranial hypertension after severe head injury. Neurosurgery 1993;79:363-368.

96. Clifton GL, Allen S, Barrodale P, et al. A phase II study of moderate hypothermia in severe brain injury. J Neurotrauma 1993;10:263-271.

97. Marion DW, Penrod LE, Kelsey SF, et al. Treatment of traumatic brain injury with moderate hypothermia. N Engl J Med 1997;336:540-546.

98. Annegers JF, Hauser WA, Coan SP, et al. A population based study of seizures after traumatic brain injuries. N Engl J Med 1998;338:20-24.

99. Chadwick D. Seizures and epilepsy after traumatic brain injury. Lancet 2000;355:334-336.

100. Temkin NR, Dikmen SS, Wilensky AJ, et al. A randomized, double-blind study of phenytoin for the prevention of post traumatic seizures. N Engl J Med 1990;323:497-502.

101. Kassel NF, Hitchon PW, Gerk MK, et al. Alterations in cerebral blood flow, oxygen metabolism, and electrical activity produced by high dose sodium thiopental. Neurosurgery 1980;7:598-603.

102. Brain Trauma Foundation. The use of barbiturates in the control of intracranial hypertension. J Neurotrauma 1996;13:711-714.

103. Braakman R, Schouten HJ, Blaauw-van Dishoeck M, et al. Megadose steroids in severe head injury. Results of a prospective double-blind clinical trial. J Neurosurg 1983;58:326-330.

104. Cooper PR, Moody S, Clark WK, et al. Dexamethasone and severe head injury: A prospective double-blind study. J Neurosurg 1979;51:307-316 .

105. Dearden NM, Gibson JS, McDowall DG, et al. Effect of high-dose dexamethasone on outcome from severe head injury. J Neurosurg 1986;64: 81-88.

106. Giannotta SL, Weiss MH, Apuzzo ML, et al. High dose glucocorticoids in the management of severe head injury. Neurosurgery 1984;15:497-501.

107. Saul TG, Ducker TB, Saleman M, et al. Steroids in severe head injury: A prospective randomized clinical trial. J Neurosurg 1981;54: 596-600.

108. Duhaime A, Christian CW, Rorke LB, et al. Nonaccidental head injury in infants: The "shaken-baby syndrome." N Engl J Med 1998; 338:1822-1829.

109. Overpeck MD, Brenner RA, Trumble AC, et al. Risk factors for infant homicide in the United States. N Engl J Med 1998;339:1211-1216.

110. Reece RM, Sege R. Childhood head injuries: Accidental or inflicted? Arch Pediatr Adolesc Med 2000;154:11-15.

111. DiScala C, Sege R, Li G, et al. Child abuse and unintentional injuries. Arch Pediatr Adolesc Med 2000;154:16-22.

112. Block RW. Child Abuse: Controversies and imposters. Curr Probl Pediatr 1999;29:253-272.

113. Herman-Giddens ME, Brown G, Verbiest S, et al. Underascertainment of child abuse mortality in the United States. JAMA 1999;282:463-467.

114. Jenny C, Hymel KP, Ritzen A, et al. Analysis of missed cases of abusive head trauma. JAMA 1999;281:621-626.

115. Jayawant S, Rawlinson A, Gibbon F, et al. Subdural hemorrhages in infants: Population based study. BMJ 1998;317:1558-1561.

116. American Academy of Pediatrics Committee on Child Abuse and Neglect. Shaken baby syndrome: Inflicted cerebral trauma. Pediatrics 1993;92:872-875.

117. Barlow KM, Gibson RJ, McPhillips M, et al. Magnetic resonance imaging in acute non-accidental head injury. Acta Pediatr 1999;88:734-740.

118. Woestman R, Perkin R, Serna T, et al. Mild head injury in children: Identification, clinical evaluation, neuroimaging, and disposition. J Pediatr Health Care 1998;12:288-298.

119. Helfer RE, Sloris TL, Black M. Injuries resulting when small children fall out of bed. Pediatrics 1977;60:533-535.

120. Duhaime AL, Alario AJ, Lewander WJ, et al. Head injury in very young children: Mechanisms, injury types, and opthalmologic findings in 100 hospitalized patients younger than 2 years of age. Pediatrics 1992;90:179-185.

121. Hymel KP, Bandak FA, Portington MD, et al. Abusive head trauma? A biomechanics-bases approach. Child Maltreat 1998;3:116-128.

122. Sugar NF, Taylor JA, Feldman KW, et al. Bruises in infants and toddlers: Those who don’t cruise rarely bruise. Arch Pediatr Adolesc Med 1999;153:399-403.

123. Carpenter RF. The prevalence and distribution of bruising in babies. Arch Dis Child 1999;80:363-366.

124. Chiavello CT, Christoph RA, Bond GR. Stairway related injuries in children. Pediatric 1994;94:679-681.

125. Jofee M, Ludwig S. Stairway related injuries in children. Pediatrics 1988;82:457-461.