Catastrophic Neurologic Disorders in the Emergency Department , 2nd Edition

Chapter 19. Traumatic Brain and Spine Injury

The management of accidental traumatic brain and spine injuries commonly involves patients who have had motor vehicle accidents or have fallen–some from significant heights. On arrival in the emergency department, the medical complexities in these highly unstable patients and in those with nonpenetrating trauma to the central nervous system may be a predicament for most neurologists. The life-threatening potential of injuries to vital organs encompasses most of the activity in the emergency department. On closer examination, it appears in some instances that neurologists and neurosurgeons may have only a peripheral role and be consulted only after many hours or even days of stabilization of vital organ functions. (As the adage goes, many patients are “too sick to operate.”1) Quite frankly, the priority to manage chest and abdominal trauma in multitraumatized patients may undermine die management of head and spine trauma. Also, intoxication often leads to trauma (e.g., bar brawls, intravenous drug use), and its drugging effect on the brain may confound clinical assessment. Some of these patients are found on the street or in the recesses of buildings, often in deplorable clinical condition. This chapter discusses how to determine the severity of the initial injury, accomplish medical management of traumatic brain edema and contusions, and assess the need for acute neurosurgical intervention.

Traumatic Brain Injury

The severity of injury is graded on the basis of die Glasgow coma score (see Chapter 8, Table 8.4, Fig. 8.5) and the presentation computed tomography (CT) scan. Head injury can be classified into several characteristic categories, and each may have different triage options (Table 19.1). Coexisting categories are common in catastrophic trauma.

Clinical Presentation

Traumatic brain injury frequently impairs level of alertness, but the mechanism is diverse. Coma may occur from bihemispheric contusions, extra-cerebral hematoma causing a mass effect on the opposite hemisphere or thalamus and, rarely, an isolated brain stem lesion.

The Glasgow coma score reliably measures the degree of traumatic coma. Eye opening and motor responses remain the most important leads and often closely correspond with other changes in the brain stem reflexes, such as pupillary and oculocephalic responses. In addition, the Glasgow coma score predicts outcome after traumatic brain injury irrespective of the underlying structural lesion. Assessment is most reliable after cardiopulmonary resuscitation and at least 6 hours after injury.2 Coma is closely linked to a Glasgow coma sum score of 8 or less (e.g., ability to open eyes only to pain, E2; incomprehensible sounds, V2; and arm withdrawal to pain, M4).

Pupillary light response and pupil size further localize and measure degree of impact. Unilaterally enlarged pupil is often due to an evolving intracranial mass lesion and may become oval when intracranial pressure (ICP) is increased. The fixed pupil of an extracranial hematoma is on the same side; but occasionally a falsely localizing contralateral pupil is seen, an observation not adequately explained. Mydriasis of one pupil after head injury indicates a swollen temporal lobe causing traction of the third nerve or direct compression. However, when a patient with a fixed pupil is seen early after trauma, the most frequent cause is an ipsilateral epidural or large subdural hematoma.

Table 19.1. Classification of Traumatic Brain Injury

Closed
   Parenchymal
      Hemorrhagic contusion
      Contrecoup contusion
      Shear lesion
      Malignant brain edema
      Diffuse axonal injury
   Extracerebral
      Epidural
      Subdural
Penetrating
   Parenchymal
      Intracerebral hematoma
   Extracerebral
      Subdural hematoma
Skull fracture
   Vault
      Linear or stellate
      Depressed
      Open or closed
   Basilar
      With spinal fluid leak
      With facial nerve palsy
      With carotid artery trauma

Fixed pupils of midposition size (diameter 5–6 mm) may indicate a mesencephalic stage of herniation and may be the first indication of brain death in the emergency department. Brain death in head injury is a result of massive cerebral edema, multiple hemorrhagic contusions, or a rapidly evolving epidural hematoma causing an irreversible shift of the brain stem. It is often observed in subtentorial epidural hematomas, which are not accommodated for in the small compartment of the posterior fossa.

In the midst of multiple trauma to limbs, abdomen, or chest, facial trauma may receive less attention; and neurologists are often the first to point out the injuries when the cranial nerves are examined. Causes of hypotension (e.g., abdominal/thoracic bleeding source) should be aggressively sought. Scalp avulsions may cause significant blood loss and shock and should be repaired immediately. However, hypotension in adults is almost never a direct result of central nervous system trauma. In a few instances, hypotension may result from major external scalp bleeding, spinal shock, and brain death; and in children it may occur with a large epidural hematoma.3 Orbital swelling can be profound and may prevent full examination of fundi and eye movements. If the swelling is associated with ecchymosis of the eyelids (so-called raccoon or panda bear eyes) (see Color Fig. 19.1 in separate color insert), it may indicate a fracture of the orbital roof or, more commonly, a Le Fort III (nasalorbital-ethmoid midface) or zygomatic fracture. The orbital roof fracture may extend through the ethmoid or cribriform-ethmoid junction and result in a cerebrospinal fluid (CSF) fistula. Petrous bone fractures may result in facial paralysis from direct injury to the facial nerve, ecchymosis over the mastoid (Battle's sign), and a CSF leak from the external canal. The Battle and raccoon eye signs, however, take several hours to develop, and specificity for basal skull fracture is low. Abrasions of the chin are clues to possible retroflexion trauma of the spine and should prompt precautionary measures, such as a collar, until a cervical spine radiograph or CT scan has excluded a fracture or dislocation.

Seizures are associated with traumatic brain injury in only up to 10% of patients and are more common in patients with a cortical contusion or traumatic intracerebral hematoma, depressed skull fracture, and dural tear.

Dysautonomia may coexist, but if so, it indicates catastrophic diffuse axonal brain trauma.4 Patients clench both fists, burying the thumbs into the palms; grind teeth; lock the jaw; and bite the endotracheal tube. Other manifestations are profuse sweating (Fig. 19.2) and tachycardia.

Many algorithms and trauma scores have been devised and may be helpful in triage; however, it is more important to weigh factors that would justify rapid transportation to a neurotrauma center or neurologic-neurosurgical intensive care unit (Table 19.2).

Figure 19.2 Profuse sweating (arrows) from dysautonomia in a patient with axonal brain injury.

Interpretation of Diagnostic Tests

Computed Tomography and Magnetic Resonance Imaging

CT scan imaging is imperative in any patient with facial lacerations or hematoma, reduced level of consciousness, significant impact to the cranium (particularly from a fall or fist fight), and certainly any evidence of focal neurologic signs or pupillary inequality during transport.5 CT scan of the brain should be part of a complete evaluation in a patient with multitrauma and not be deferred to a later time. Magnetic resonance imaging (MRI) is considered when CT scans do not fully explain the clinical presentation, but it is not readily available and is probably unsafe in mechanically ventilated patients with unstable multiple traumatic lesions to vital organs. Major dissimilarities between CT scan findings after trauma and the patient's state of impaired consciousness should point to confounding factors, including additional insults to the brain, some of which are reversible (Table 19.3).

Table 19.2. Severity Indicators of Traumatic Brain Injury

Inability to remember trauma
Fall, fist fight, car collision
Age >60 years
Tachypnea
Hypotension
Scalp or face injury
Penetrating injury
Pupils fixed to light
Abnormal findings on computed tomographic scan

Table 19.3. Coma in Traumatic Brain Injury but “Normal” Findings on Computed Tomographic Scanning

Drug or alcohol overdose
Postanoxic insult to both hemispheres from asphyxia (vomiting, aspiration, foreign body)
Postictal state after seizures or nonconvulsive status epilepticus
Vertebral artery dissection with basilar artery occlusion rare)

Several CT scan patterns can be recognized, and they are illustrated in the figures of this chapter. The severity of head injury in comatose patients can be further classified. Absence of visualization of the basal cisterns, midline shift, and a mass lesion are strong predictors of increased ICP. Mass effect with absence of a third ventricle and trapping of the temporal horn correlates strongly with increased ICP.

Most parenchymal injuries are a direct effect of a blow to the brain. These contusions are created when brain tissue becomes impacted against the bony protuberances of the base of the skull. They are further subdivided into fracture contusions, contrecoup contusions, and shear lesions.

Contusions are common in the frontal and temporal lobes (Fig. 19.3) and can be seen in association with a fracture of the anterior fossa. This abnormality may be accompanied by an epidural or subdural hematoma, which may be responsible for the clinical symptoms. In patients with acute subdural hematoma, follow-up CT scans show lacerated brain tissue at the same site or this becomes apparent during craniotomy and removal of the extradural hematoma.

Contrecoup contusions (Fig. 19.4A,B) are often two or more lesions diametrically opposite to one another. Shear lesions (Fig. 19.5) are punctate lesions from disruption of small penetrating arteries due to rotational forces at impact. The localization of shear lesions varies, but often they are identified at the gray-white matter junction. Basal ganglia (predominantly putamen) may be involved.

Figure 19.3 A: Frontal lobe contusions (arrows). B: Temporal lobe burst hematoma (arrows).

Hemorrhagic contusions may not be evident on initial CT scanning, are unmasked only after repeat studies, and are not always associated with clearly documented clinical deterioration (Fig. 19.6). Mass effect from cortical contusions is seldom severe initially but may increase from peri-contusional swelling.

Another rather frequent CT scan image is diffuse axonal injury.6 The pathologic damage is caused by acceleration-deceleration forces,7 and the overwhelming evidence of axonal destruction may be noted only microscopically at autopsy.8 CT scan findings may appear initially normal, but often subtle changes are present, such as intraventricular blood (small amounts from corpus callosum tearing), punctate shear lesions, or sulci effacement. Later CT scans may show (ex vacuo) enlargement of the ventricles from reduction of the white matter tissue.

Corpus callosum lesions may be demonstrated on MRI but seldom are found by CT scanning. Corpus callosum lesions are in the splenium posterior body because relative fixation from the posterior falx results in a tensile force rather than release with rotation. Brain stem lesions have been noted, often with multiple hemispheric lesions. A focal hemorrhage is most commonly seen in the dorsolateral aspect of the brain stem from a blow to the edge of the tentorium or in the interpeduncular cistern. Diffuse cerebral swelling (Fig. 19.7) occasionally is seen early after impact and indicates severe axonal damage. Typically, the differentiating features of the white matter and gray matter disappear and the basal cisterns are obliterated, resulting in a “featureless grayout of the brain.” This malignant edema is often fatal and may occur after an asymptomatic interval and “trivial” impact (fall off a horse or bicycle).

Figure 19.4 A: Contrecoup contusions (arrows). B: Epidural hematoma with contrecoup temporal lobe contusion (arrowheads) and basal cistern effacement (arrows).

A subdural hematoma typically is recognized as a hyperdense lesion with a characteristic crescentic (curving with the skull) collection.9 Acute subdural hematomas are hyperdense, but when marked anemia (hemoglobin <8 g/dL) is present, they may approach the density of brain tissue. When hyperdensity is seen within an isodense collection, rebleeding is likely. A fluid-blood interface may suggest rebleeding (Fig. 19.8). Usually, a subdural hematoma isodense to the gray matter is evident 3 weeks after onset. A change to a hypodense collection follows, but estimation of the age of the hematoma on CT images is difficult. In elderly patients, isodense subdural hematomas may only be recognized by loss of sulci and small ventricles (false CT-age mismatch). Small layers of subdural hematoma may go undetected on CT scans because they can hardly be distinguished from bone; MRI visualizes them (Fig. 19.8A). Subdural hematoma can be seen as an inter-hemispheric collection along the falx, and this collection tends to enlarge (Fig. 19.8E).

Figure 19.5 Shear lesions (arrows), including basal ganglia.

Figure 19.6 Delayed abnormalities on computed tomographic scans. Top row, Contusion (arrow) is barely seen. Bottom row, Multiple lobar contusions (arrows) develop later.

Figure 19.7 Axonal injury to the brain and diffuse swelling (arrows), with so-called featureless grayout.

Figure 19.8 Different types of subdural hematoma. A: Subdural hematoma (arrows) on magnetic resonance imaging (fluid-attenuated inversion recovery) in a patient with transient ischemic attacks on presentation who had normal computed tomographic findings but abnormal findings on magnetic resonance imaging. B,C: Typical subdural hematoma with no shift but compression of the white matter (white arrows in B). D: Rebleeding subdural hematoma with fluid interface (arrow) between recent and older collections of blood. E: Falx subdural hematoma (arrow).

Epidural hematomas (Fig. 19.9) are associated with fracture in more than 95% of cases. The hematoma, which is due to a tear in a meningeal artery, strips the dura away from the inner table of the skull, producing a biconvex, or lens-shaped, configuration. Anterior-posterior extension is usually limited by the skull sutures. Its mass effect is significant in both the supratentorial and the infratentorial spaces and may rapidly lead to brain herniation syndromes. An ominous CT scan feature is a hyperlucent area, which should be recognized on CT images (Fig. 19.9A). It indicates active bleeding because a completed epidural hematoma is uniformly dense. An epidural hematoma in the posterior fossa is caused by a torn dural sinus (Fig. 19.9D); it may become large enough to cause full effacement of the brain stem cisterns. Vertex epidural hematomas, which are very rare, are caused by rupture of the superior sagittal sinus. Unlike the rapid evolution of arterial epidural hematomas, involvement of the dural sinus may not become clinically noticeable for several hours.10

Figure 19.9: A-C: Supratentorial epidural hematoma (arrows, see hyperlucent areas). D: Subtentorial epidural hematoma.

Traumatic subarachnoid collections in sulci and fissures should not involve any of the suprasellar cisterns.11 Occasionally, some sediment is seen in the ambient cistern (see Chapter 13). Blood from traumatic subarachnoid hemorrhage may have collected in the sylvian fissure alone. The distinction between a ruptured middle cerebral artery aneurysm, with blood deposited in the sylvian fissure, that causes the patient to fall and subarachnoid hemorrhage from trauma alone is impossible if no clear history of a thunderclap headache is volunteered by the patient. Cerebral angiography is always needed. Traumatic intraventricular hemorrhages are commonly associated with other contusional lesions, which may become apparent on follow-up CT scan.12,13,14 Coma and traumatic intraventricular hemorrhage are associated with poor outcome, but isolated intraventricular blood in an alert patient is associated with good outcome and more often seen with use of warfarin. The source of bleeding is typically the choroid plexus. Primary intraventricular hemorrhage may be associated with a fall and additional traumatic lesion, further obfuscating the etiology. Thus, in young patients with frank intraventricular hemorrhage, a cerebral angiogram is warranted.

Bone Window Computed Tomography

Bone windows on CT scans are important to show linear fractures in the skull and are equivalent to routine skull radiographs. In many instances, they indicate the site of the blow and are next to the contused area of the brain. In one study, skull fracture was present in 77% of patients with contusion, 87% with an extradural hematoma, 72% with a subdural hematoma, and 66% with an intra-cerebral hematoma.15 The degree of depression of skull fracture can be easily visualized. Linear fractures are more commonly associated with epidural or subdural hematomas than are depressed skull fractures (Fig. 19.10).15 Of specific note is the presence of blood in the sphenoid, which is typical of a foramen lacerum fracture and may indicate damage to the carotid artery (Fig. 19.10).

Serum

Next to a routine laboratory survey, obtaining a serum (and urinary) toxicologic screen is of value. Blood alcohol level is imperative not only for medicolegal reasons but also for judging its influence on level of consciousness. Increased serum osmolality may also indicate alcohol intoxication (a more detailed discussion is in Chapter 8). Laboratory support of disseminated intravascular coagulation needs to be sought and includes prolonged prothrombin time, thrombocytopenia (<60,000 platelets), increase in fibrin degradation products, red cell fragments in smears, and increased D-dimer. Early appearance of disseminated intravascular coagulation indicates massive destruction of brain tissue, releasing thrombogenic substances such as thromboplastin into vascular space.

Miscellaneous

Routine X-ray imaging in a head-injured patient should include the cervical spine with lateral and odontoid views and, when relevant, plain films of the abdomen and pelvis. Diagnostic peritoneal lavage or CT scan of the abdomen or chest is indicated in patients with fluctuating blood pressure after adequate fluid replacement.

First Priority in Management

The main principle of management of traumatic head injury is immediate treatment of increased ICP, which may involve removal of an extracranial hematoma or contusion with mass effect.16,17

Immediate stabilization is summarized in Table 19.4. Rapid triage to CT scanning is essential because it determines the cause of impaired consciousness in most instances. Scalp lacerations should be temporarily repaired in the emergency department unless they are contaminated. It is important to secure the airway, provide fluids with two large-bore catheters, and exclude abdominal trauma with peritoneal lavage if blood pressure remains marginal despite fluid loading with hypertonic saline or dextran.18,19 Some experts argue for a flat body position to maximize cerebral perfusion pressure (CPP). Early fluid resuscitation in an attempt to improve organ and brain perfusion may seem beneficial but could also worsen bleeding due to increase in blood flow. Others have argued that fluid administration may reduce blood viscosity, dilute clotting factors, or interfere with hemostasis (particularly, use of starch). Administration of poorly warmed fluids may also contribute to a coagulopathy. A reasonable consensus is to use aliquots of 250 ml normal saline in traumatized patients.20

Figure 19.10 Major skull fractures. A: Nondepressed. B: Depressed. C: Petrous. D: Blood in the sphenoid sinus is typical of a fracture of the foramen lacerum.

Table 19.4. Immediate Priorities in Head Injury

Secure airway
Remove foreign body
Endotracheal intubation
Immobilize spine
Inspect for scalp laceration and depressed fracture
Peritoneal lavage with hypotension
Secure venous access with two catheters
Normal saline resuscitation in hypotension
Obtain cervical spine X-ray
Chest radiography
Computed tomographic scan of brain
Computed tomographic scan of abdomen (if appropriate)

The focus of management involves not only reduction of ICP but also maintenance of interrelated CPP (Box 19.1). It is prudent to hyperventilate the patient (frequency of more than 20 breaths per minute or squeezing the anesthesia [Ambu] bag every 3 seconds) and to give a single loading dose of mannitol (20%), 1 g/kg over 10 minutes, but only if blood pressure has remained stable throughout. Mannitol in multiple doses may have the opposite effect by accumulating within the brain and thus reversing the osmotic gradient between edematous brain and plasma.22 However, en route to the operating room, mannitol in a dose of 1.4 g/kg was significantly better than lower doses when infused in patients with contusional expanding temporal lobe hematoma.23

A restless patient may require intubation and sedation with propofol (infusion of 0.5 mg/kg hourly)24 or, less attractive, morphine (infusion of 1 mg hourly) if the endotracheal tube and mechanical ventilator are not tolerated. Because propofol also reduces increased ICP, it is a useful drug in agitated patients with early brain swelling.24 There is no rationale for use of corticosteroids or barbiturates, all of which may seriously complicate management and possibly inversely affect outcome through adverse effects of hypotension, hyperglycemia, and infectious complications.25,26,27,28,29 Simple measures also reduce ICP, such as preventing head rotation to one side (jugular vein compression); suctioning without stimulation of the soft palate or posterior pharyngeal wall, which elicits a gag and cough reflex; and suctioning through an endotracheal tube with one passage only. Intravenous administration of lidocaine30,31 or increasing the dose of propofol may blunt these ICP responses. Several drugs may increase ICP through an increase in cerebral blood flow by vasodilation (Table 19.5).

Hypoxemia should be aggressively managed, and after intubation, positive end-expiratory pressure (PEEP) is needed to improve gas exchange.32 PEEP may increase intrapleural pressure and superior vena cava pressure and reduce cerebral venous outflow. ICP may become seriously elevated if PEEP values higher than 10 cm H2O are needed, but its increase can be countered with mannitol and head elevation. PEEP may increase partial pressure of arterial CO2 (PaCO2) because of increased physiologic dead space; this effect should be anticipated and managed by increasing the minute volume of the ventilator. Conversely, the effects of hyperoxia are unclear and, when studied with microdialysis catheters, did not result in improved glucose oxidation.33

Systemic hypothermia (32°C-33°C) within 6 hours is not likely to be beneficial in severe head injury. Its benefit and potential complications (cardiac arrhythmias, pneumonia, coagulation problems, and increase in prothrombin time) have been investigated. There is some early indication that 35°C-35.5°C may be more optimal.34 Certainly, fever should be aggressively treated with cooling blankets, alcohol rubbing, or, as a last resort, ice gastric lavage. These measures are certainly needed in patients with a severe sympathetic outburst, who usually have tachycardia, tachypnea, and profuse sweating, with temperature increases up to 40°C. Propofol can effectively mute shivering most of the time.

Box 19.1. Management of Cerebral Perfusion in Traumatic Brain Injury

Cerebral blood flow is held constant within the range of mean arterial pressure from 80 to 160 mm Hg. Outside this homeostatic range, cerebral blood flow is linearly coupled with pressure. Below the lower threshold, a decrease in CPP results in a decrease in blood flow and ischemia. Theoretically, above the upper threshold, an increase in CPP results in breakdown of the blood-brain barrier and edema, but studies suggest that high perfusion pressures are tolerated for brief periods.

Management of CPP has been advocated, but it assumes intact autoregulation, which may not be present in up to 50% of patients with severe traumatic head injury. Cerebral perfusion is usually aimed at 70–80 mm Hg and can be increased by increasing systolic arterial blood pressure (SABP), churning CSF pressure, and using mannitol (CPP = MAP - ICP).

These vasodilatory and vasoconstriction cascades (Fig. 19.11), have led to management of CPP irrespective of ICP.21 Blood pressure can be increased with adrenergic receptor drugs, and normovolemia is maintained with albumin. Early results suggest good to superior outcome, but this approach (endorsed by the Brain Trauma Foundation–http://www.braintrauma.org) remains unproven.

The use of prophylactic antiepileptic agents to prevent seizures has been established. Current reasonable recommendations are intravenous loading with phenytoin in patients with contusions on CT scan or depressed skull fractures.35 Maintaining therapeutic phenytoin levels for at least 1 month, a period when seizures are most common, seems reasonable. Surgical management of extracerebral hematomas is urgently indicated. Usually, time can be allowed to evacuate the hematoma in the operating room, but in rapidly deteriorating patients without a nearby CT scanner, emergency drilling in the emergency department has been lifesaving.36 Medical management and observation in extracerebral hematomas are considered only for patients with a maximal Glasgow coma score. Epidural hematomas with a diameter less than 1.5 cm, no midline shift, and, as alluded to earlier, no lucent area inside the hematoma suggesting recent bleeding could be managed with observation.37,38 For subdural hematomas, medical management is considered if the thickness of the hematoma is similar to the thickness of the skull. Large subdural hematomas without any shift (caused by atrophy of the brain n elderly patients or chronic alcoholics) can be surgically managed in a delayed fashion when the clinical signs are minimal. Burr hole placement after the hematoma has liquefied is preferred above a large craniotomy.

Table 19.5. Medications to be Avoided with Increased Intracranial Pressure

Hydralazine
Sodium nitroprusside
Halogenated inhalation anesthetics (halothane, isoflurane)
Ketamine
Calcium channel blockers (nicardipine, nimodipine)

Figure 19.11 Vasodilatation and vasoconstriction cascades. CPP, cerebral perfusion pressure; ICP, intracranial pressure; CBV, cerebral blood volume.

Acute chest or abdominal exploration for injury often has a priority over traumatic brain injury management due to its immediate life-threatening condition. Complicated decisions pertain to timing of fixation or instrumentation of long bone fractures. When the patient is stable for several hours, bone fixation avoids traction and its complications of prolonged immobilization but also reduces pain and possibly fat emboli.1

Predictors of Outcome

It is impossible to make an accurate prediction of outcome in the emergency department. Early predictors commonly have been inaccurate, proposed cut-off points in certain scales have been disproved, and successful resuscitation of a patient for systemic injuries may greatly improve the outcome. Prognosis estimates at least in the 24 hours after injury are thus unreliable.39

Age remains an important factor in prognosis, with 80% mortality from diffuse axonal injury in patients older than 55 years.40 Prognosis is worse with the following factors: shock, subdural hematoma,41 and coma in the elderly; early diffuse edema and increased ICP;42 and failure to decrease ICP with conventional methods. In a study of MRI in 80 severely injured patients, corpus callosum lesions and dorsolateral brain stem lesions predicted an unfavorable outcome, including vegetative state.43

Triage

·     Evacuation of any epidural or subdural hematoma if the patient has a decrease in consciousness, hemiparesis, or speech deficit.

·     Evacuation of hemorrhagic contusion if mass effect occurs.

·     Placement of ICP monitor and monitoring in neurologic-neurosurgical intensive care unit.

Spinal Cord Injury

Spinal cord injury may have been caused by traffic accidents in more than 50% of patients arriving at emergency departments. The demographic profile consists of men in their thirties seen most often during weekends in the summer months. Complete cord lesions have decreased in prevalence, which is tentatively explained by improved care in the field, increased use of seat belts, and, possibly, surgical care of unstable spine trauma. It is prudent to assume that spinal cord injury may have occurred in patients who have had multiple trauma, motor vehicle or sports accidents, or a documented spine fracture.44,45

This section is not intended to comprehensively discuss all spinal traumatic lesions and surgical management, which can be properly addressed only by experienced spine surgeons or neurosurgeons with special qualifications in the management of spine trauma. Guidelines for surgical management are outlined (Box 19.2). The field has evolved into a subspecialty in which the role of a neurologist or emergency physician is important but limited to accurate clinical description of the damage, appropriate stabilization, recognition of unstable spine fractures or dislocations, and, when necessary, early specific medical management. An overview of spine fractures can be found elsewhere.48

Clinical Presentation

Traumatic spinal cord injury involves the cervical cord in approximately 50% of cases, the thoracic segment in 35%, and the lumbar segment or conus in the others. Subtle signs of cervical spine injury are physical signs of an injury above the clavicle, neck pain, and tilting of the head to one side.

Tetraplegia or paraplegia is evident from the onset, but most clinical challenges involve the management of its commonly associated dysautonomia and urogenital manifestations.

In the spinal shock phase, a generalized state of hypoexcitability occurs. The marked reduction of sympathetic outflow results in peripheral vasodilatation, decreased cardiac output, bradycardia (from cardiac chronotropic activity), and venous pooling, all factors that reduce blood pressure. Blood pressure typically is less than 100 mm Hg and depends on position and volume; reduction in the sitting or upright position may result in syncope. The lower extremities may show a bluish discoloration from vasodilatation and venous pooling. Typically, pain does not produce an increase in heart rate or blood pressure. Passive engorgement of the penis (priapism) occurs as a consequence of sympathetic loss and always indicates an extensive spinal cord lesion.

Box 19.2. Surgical Management of Spinal Cord Injury

Surgical management of spine injury in a patient with spinal cord injury is pursued to prevent further injury in incomplete lesions, to ensure stability, and to prevent deformity. Unstable cervical lesions should be expected if anterior or posterior elements are destroyed, sagittal diameter of the spinal canal is less than 13 mm, or sagittal displacement is more than 3.5 mm or 20%. A stable lesion allows for earlier mobilization and transfers. Operative stabilization has not been shown to improve recovery in complete or incomplete lesions, although one survey called for a trial based on suggestive data review.46

Deformity requires instrumentation and posterior fusion in many instances.47

Temperature may be unregulated. Shivering cannot occur below the lesion because of loss of sympathetic tone, and lack of increase in metabolism may result in hypothermia. Paradoxically, core hypothermia may be present in patients with otherwise warm extremities. The bladder is completely paralyzed, causing urinary retention and overflow. Detrusor muscle contraction only later results in spontaneous or external stimuli-induced voiding.

It is important to determine the sensory level by cutaneous innervation of the dermatomes. Pinprick sensation should be evaluated serially. It is important to memorize and document clinical markers (nipple, T4; navel, T10; midway from arm to chest, C4-T2 border). It may be difficult to determine whether the level is cervical or thoracic because the C4 and T2 levels abut each other, but examination of motor function and reflexes of the arm further helps localization.

Reflexes such as the anal wink (puckering of the anus with stimulus to the perianal region) and the bulbocavernosus reflex (traction on a Foley catheter or digital pressure on the clitoris or penis while anal sphincter contraction is monitored with a gloved finger in the rectum) should be examined. Neurologic examination should be carefully documented by use of the American Spinal Injury Association neurologic classification of spinal cord injury (Appendix 19.1). The American Spinal Injury Association Impairment Scale can also be used to monitor progress.

Interpretation of Diagnostic Tests

Neuroimaging of the spine and spinal cord has a high priority. The extent of imaging is determined by neurologic findings at presentation. Careful clinical delineation of level of involvement may tailor orientation and selection of the studies.

Neuroradiologists should have access to clinical information that may further determine certain MRI sequences. The priority in the emergency department is to diagnose unstable cervical or thoracic spine fractures or spine compression.

Spine Plain Films

Screening cervical spine radiographs for alert patients with no neck tenderness and no neurologic abnormalities have a very low yield.49 Combined lateral, anteroposterior, and odontoid views have a high diagnostic yield and should recognize more than 90% of the lesions, although a false-negative rate of 26% was found in a study of 70 patients. The lateral cervical spine radiograph and odontoid views (Fig. 19.12) should be viewed systematically.50,51 A common pitfall is focusing on a single fracture or misalignment while overlooking other abnormalities. The essentials of cervical spine plain film viewing are shown in Table 19.6. Evaluation is very difficult. Only the trained eye of an experienced physician can identify fractures, but even then a CT scan is often needed for confirmation. The threshold for ordering a CT scan of the cervical spine must be very low, particularly when plain films of the cervical spine give dubious information or are of marginal quality.

First, the cervical vertebral bodies should be identified, and particular attention should be paid to the lower cervical spine. Hand traction should be used to pull both arms and shoulders of the patient down. Inadequate films should prompt CT scanning of the spine. When a lateral spine film is evaluated, four lordotic curves and alignments are assessed to look for displacement (Fig. 19.12A).

Figure 19.12 Plain spine radiographs with examination techniques to uncover fractures and dislocations. A: Normal alignment lines: I, normal alignment along the anterior (A) vertebral margins; II, posterior (P) vertebral margin alignment line; III, spinal laminar (L) line; IV, relation of the dorsal spinous processes. B: Indicators of normal odontoid interspace (arrows). C: Incomplete cervical spine examination (C7 is missing, arrow).

Common findings are compression of the vertebral bodies (vertebral body often several millimeters less anteriorly than posteriorly), displacement in the lateral view (more than 3 mm between adjacent vertebral bodies), and displacement of the odontoid bone (odontoid tip should be aligned with tip of clivus).

Indications of instability are displacement of a vertebral body, widening of the interspinous or interlaminar distance, widening of the facet joints or spinal canal, disruption of the posterior spinal line, and anterolisthesis or retrolisthesis with flexion or extension (Table 19.7).52 The odontoid bone typically lies within 13 mm of the posterior cortex of the anterior Cl arch.

Table 19.6. Essentials of Disciplined Cervical Spine Viewing

Count the number of cervical vertebral bodies; seven cervical spine bodies and the superior end plate of T1 should be visible
Trace contour of every body to detect fractures
Evaluate the alignment lines (see Figs. 19.12, 19.13A)

1. Anterior spinal line along the anterior longitudinal ligament

2. Posterior spinal line along the anterior longitudinal ligament

3. Spinolaminar line along the base of the spinous processes

4. Line of the spinous processes

Assess soft tissue
Assess facet lines
Assess craniocervical junction
Assess space between the dens axis and lateral masses of C1

Hyperflexion injuries of the cervical spine–direct trauma to the head, spine in the flexed position–can be seen in various degrees, from mild widening of the posterior intervertebral space to subluxation of the vertebra; and the inferior articular facet may become lodged on the superior facet of the vertebra below (so-called perched facet). A cord lesion is common or easily induced with further manipulation, making this a highly unstable condition.

Unilateral facet distortion can be recognized by an alteration in the laminar space, namely, the distance between the spinolaminar line and the posterior margin of the articular mass.

Table 19.7. Spine Instability

Cervical

·  Widened interspinous space or facet joints

·  Anterior listhesis >3.5 mm

·  Narrowed or widened disk space

·  Focal angulation >11 degrees

·  Vertebral compression >25%

Thoracic

·  Fracture dislocation

·  Posttraumatic kyphosis >40 degrees

·  Spine fractures associated with sternal fractures

·  Concomitant rib fracture or costovertebral dislocation

Source: Imhof and Fuchsjäger.52 By permission of Springer-Verlag.

Hyperextension injuries (sudden deceleration impact) produce fairly characteristic features, such as a hyperextension teardrop fracture (avulsion of the site of attachment of the anterior longitudinal ligament), hangman's fracture with bilateral fracture through the pars interarticularis of C2, Jefferson's fracture (fracture of the ring of C1), and odontoid fractures (tip is type I, base is type II, and extension into the body of C2 is type III). Figure 19.13 illustrates the most common unstable cervical fractures. Facial injury is more common as a result of direct impact.

Computed Tomography and Magnetic Resonance Imaging

CT scans added to a plain cervical spine film are unsurpassed in diagnostic value for demonstration of fractures. Myelography combined with CT can more clearly demonstrate the cord and nerve roots and determine whether they are compressed by the misalignment or fracture. In most instances, specific areas are scanned with axial slices 1.5–3 mm thick for the cervical spine and 3–5 mm thick for the thoracic and lumbar spines. CT scan reconstructions are very useful in imaging loose bone fragments and facet dislocation.

Intrathecal administration of contrast medium (myelography, CT scanning) is usually reserved for patients who cannot undergo MRI (presence of a pacemaker, aneurysm clips, cochlear implants, bullet fragments, and morbid obesity). It has become the second-choice imaging modality in spine injury because it is time-consuming and requires patient movement.

MRI in spine injury should first obtain sagittal T1-weighted images, with axial images through abnormal areas.53 A T1-weighted image is important to rule out major abnormalities and can be followed by T2 or gradient echo sequences (short time to acquire and sensitive for early hemorrhages in the spine). On T1-weighted images (short TR, 300–1000 msec; TE, 10–30 msec), subacute hemorrhage is bright and CSF is dark. On T2-weighted images (long TR, 1500–3000 msec; TE, 60–120 msec), CSF is bright, cord edema is bright, and acute hemorrhage is dark.54 An example of cord trauma and swelling is shown in Figure 19.14.

First Priority in Management

Immediate cervical spine immobilization and endotracheal intubation are needed.

Figure 19.13 A,B: Two examples of hangman fracture (C2 bilateral fracture through pars articularis), common with windshield injuries. C: Odontoid fracture. The spinal laminar line is disrupted. Dens is outlined (arrow). D: Locked facet dislocation (arrows, hyperflexion injury). E: Jefferson C1 fracture (arrows).

Hypotension from unopposed parasympathetic tone is common, particularly with change in position or in the first minutes after connection to the mechanical ventilator. Volume resuscitation or an α-adrenergic receptor agent, such as phenylephrine, is needed. In occasional patients, autonomic dysregulation is manifested as marked hypertensive surges, which should be treated with labetalol. Administration of methylprednisolone has resulted in better recovery 1 year after injury (Third National Acute Spinal Cord Injury Randomized Controlled Trial).55 Maintenance therapy depends on when therapy has been started (Table 19.8).

Table 19.8. Emergency Room Management of Traumatic Spine Injury

Intubation and mechanical ventilation with lesion at C3 or higher
Intubate if aspiration
Volume loading with albumin or Ringer's lactated solution
Epinephrine drip
Body warming with blanket, warming intravenous fluids
Subcutaneous heparin, 5000 U
Codeine for pain
Proton pump inhibitor (e.g., Protonix) to prevent gastrointestinal bleeding
Interval 0–3 hours methylprednisolone 30 mg/kg
   Infusion of methylprednisolone 5.4 mg/kg hourly for 24 hours
Interval 3–8 hours methylprednisolone 30 mg/kg
   Infusion of methylprednisolone 5.4 mg/kg hourly for 48 Hours

Catheter placement, gastric ulcer prophylaxis, correction of core hypothermia, and deep venous thrombosis prophylaxis are important before triage.

Predictors of Outcome

The degree of cord injury and the presence of head injury at presentation determine initial outcome and mortality.56,57,58 Patients with complete cervical transection and apnea usually do not recover. Patients are tetraplegic, depend on a mechanical ventilator, and can only operate devices for communication and locomotion (speech may be possible through a special tracheostomy). Patients with transection above the C3 level are not weaned. One study claimed weaning with an average of 80% with lower-level lesions.56

Figure 19.14 Magnetic resonance images of traumatic cord swelling (arrows).

Complete cord lesion very rarely changes to an incomplete lesion and vice versa. Patients with incomplete lesions but no motor function and only sensory function have a 10%–30% chance of regaining some motor function. Patients with incomplete lesions but retained motor function have a 50% chance of improvement. This improvement (defined as Medical Research Council muscle grade 3 or more57) is then noted in half of the common muscles.

MRI abnormalities with intramedullary hematoma or contusion involving more than one segment predict a worse outcome. Central low signal intensity on T2images may represent central cord contusion, with poor prospects. Central high signal intensity without areas of low signal intensity with normal T1 may represent cord edema or ischemia but no infarction. Mixtures of patterns are possible and make the use of MRI for prognosis indeterminate.

Triage

·     Neurologic-neurosurgical intensive care unit for management of dysautonomia; bladder, skin, and bowel care; and planning for stabilizing spinal surgery.

·     Spinal rehabilitation center if surgery is not indicated and dysautonomia is absent.

References

1. Giannoudis PV, Veysi VT, Pape HC, et al.: When should we operate on major fractures in patients with severe head injuries? Am J Surg 183:261, 2002.

2. Marion DW, Carlier PM: Problems with initial Glasgow Coma Scale assessment caused by prehospital treatment of patients with head injuries: results of a national survey. J Trauma 36:89, 1994.

3. Partrick DA, Bernard DD, Janik JS, et al.: Is hypotension a reliable indicator of blood loss from traumatic injury in children? Am J Surg 184:555, 2002.

4. Baguley IJ, Nicholls JL, Felmingham KL, et al.: Dysautonomia after traumatic brain injury: a forgotten syndrome? J Neurol Neurosurg Psychiatry 67:39, 1999.

5. Thornbury JR, Masters SJ, Campbell JA: Imaging recommendations for head trauma: a new comprehensive strategy. AJR Am J Roentgenol 149:781, 1987.

6. Adams JH, Graham DI, Murray LS, et al.: Diffuse axonal injury due to nonmissile head injury in humans: an analysis of 45 cases. Ann Neurol 12:557, 1982.

7. Gennarelli TA, Thibault LE, Adams JH, et al.: Diffuse axonal injury and traumatic coma in the primate. Ann Neurol 12:564, 1982.

8. Blumbergs PC, Jones NR, North JB: Diffuse axonal injury in head trauma. J Neurol Neurosurg Psychiatry 52:838, 1989.

9. Piek J, Chesnut RM, Marshall LF, et al.: Extracranial complications of severe head injury. J Neurosurg 77:901, 1992.

10. Di Rocco A, Ellis SJ, Landes C: Delayed epidural hematoma. Neuroradiology 33:253, 1991.

11. Servadei F, Murray GD, Teasdale GM, et al.: Traumatic subarachnoid hemorrhage: demographic and clinical study of 750 patients from the European Brain Injury Consortium survey of head injuries. Neurosurgery 50:261, 2002.

12. Abraszko RA, Zurynski YA, Dorsch NW: The significance of traumatic intraventricular haemorrhage in severe head injury. Br J Neurosurg 9:769, 1995.

13. Lee JP, Lui TN, Chang CN: Acute post-traumatic intraventricular hemorrhage: analysis of 25 patients with emphasis on final outcome. Acta Neurol Scand 84:85, 1991.

14. Fujitsu K, Kuwabara T, Muramotoa M: Traumatic intraventricular hemorrhage: a report of twenty-six cases and consideration of the pathogenic mechanism. Neurosurgery 23:423, 1988.

15. MacPherson M, MacPherson P, Jennett B: CT evidence of intracranial contusion and haematoma in relation to the presence, site and type of skull fracture. Clin Radiol 42: 321, 1990.

16. Gibson RM, Stephenson GC: Aggressive management of severe closed head trauma: time for reappraisal. Lancet 2:369, 1989.

17. Miller JD, Dearden NM, Piper IR, et al.: Control of intracranial pressure in patients with severe head injury. J Neurotrauma 9(Suppl 1):S317, 1992.

18. Bickell WH, Wall MJ Jr, Pepe PE, et al.: Immediate versus delayed fluid resuscitation for hypotensive patients with penetrating torso injuries. N Engl J Med 331:1105, 1994.

19. Mattox KL, Maningas PA, Moore EE, et al.: Prehospital hypertonic saline/dextran infusion for post-traumatic hypotension. The U.S.A. Multicenter Trial. Ann Surg 213: 482, 1991.

20. Revell M, Porter K, Greaves I: Fluid resuscitation in prehospital trauma care: a consensus view. Emerg Med J 19: 494, 2002.

21. Rosner MJ, Daughton S: Cerebral perfusion pressure management in head injury. J Trauma 30:933, 1990.

22. Kaufmann AM, Cardoso ER: Aggravation of vasogenic cerebral edema by multiple-dose mannitol. J Neurosurg 77:584, 1992.

23. Cruz J, Minoja G, Okuchi K: Major clinical and physiological benefits of early high doses of mannitol for intra-parenchymal temporal lobe hemorrhages with abnormal pupillary widening: a randomized trial. Neurosurgery 51:628, 2002.

24. Kelly DF, Goodale DB, Williams J, et al.: Propofol in the treatment of moderate and severe head injury: a randomized, prospective double-blinded pilot trial. J Neurosurg 90:1042, 1999.

25. Bouma GJ, Muizelaar JP, Bandoh K, et al.: Blood pressure and intracranial pressure-volume dynamics in severe head injury: relationship with cerebral blood flow. Neurosurgery 77:15, 1992.

26. 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 58:326, 1983.

27. Cottrell JE, Patel K, Turndorf H, et al.: Intracranial pressure changes induced by sodium nitroprusside in patients with intracranial mass lesions. J Neurosurg 48:329, 1978.

28. Rea GL, Rockswold GL: Barbiturate therapy in uncontrolled intracranial hypertension. Neurosurgery 12:401, 1983.

29. Ward JD, Becker DP, Miller JD, et al.: Failure of prophylactic barbiturate coma in the treatment of severe head injury. J Neurosurg 62:383, 1985.

30. Yano M, Nishiyama H, Yokota H, et al.: Effect of lidocaine on ICP response to endotracheal suctioning. Anesthesiology 64:651, 1986.

31. Evans DE, Kobrine Al: Reduction of experimental intracranial hypertension by lidocaine. Neurosurgery 20:542, 1987.

32. Chesnut RM, Marshall LF, Klauber MR, et al.: The role of secondary brain injury in determining outcome from severe head injury. J Trauma 34:216, 1993.

33. Magnoni S, Ghisoni L, Locatelli M, et al.: Lack of improvement in cerebral metabolism after hyperoxia in severe head injury: a microdialysis study. J Neurosurg 98:952, 2003.

34. Tokutomi T, Morimoto K, Miyagi T: Optimal temperature for the management of severe traumatic brain injury: effect of hypothermia on intracranial pressure, systemic and intracranial hemodynamics, and metabolism. Neurosurgery 52:102, 2003.

35. Chang BS, Lowenstein DH: Practice parameter: anti-epileptic drug prophylaxis in severe traumatic brain injury. Neurology 60:10, 2003.

36. Mahoney BD, Rockswold GL, Ruiz E, et al.: Emergency twist drill trephination. Neurosurgery 8:551, 1981.

37. Hamilton M, Wallace C: Nonoperative management of acute epidural hematoma diagnosed by CT: the neuroradiologist's role. AJNR Am J Neuroradiol 13:853, 1992.

38. Sagher O, Ribas GC, Jane JA: Nonoperative management of acute epidural hematoma diagnosed by CT: the neuroradiologist's role. AJNR Am J Neuroradiol 13:860, 1992.

39. Lang EW, Pitts LH, Damron SL, et al.: Outcome after severe head injury: an analysis of prediction based upon comparison of neural network versus logistic regression analysis. Neurol Res 19:274, 1997.

40. Choi SC, Narayan RK, Anderson RL, et al.: Enhanced specificity of prognosis in severe head injury. J Neurosurg 69:381, 1988.

41. Wilberger JE Jr, Harris M, Diamond DL: Acute subdural hematoma: morbidity and mortality related to timing of operative intervention. J Trauma 30:733, 1990.

42. Signorini DF, Andrews PJ, Jones PA, et al.: Adding insult to injury: the prognostic value of early secondary insults for survival after traumatic brain injury. J Neurol Neurosurg Psychiatry 66:26, 1999.

43. Kampfl A, Schmutzhard E, Franz G, et al.: Prediction of recovery from post-traumatic vegetative state with cerebral magnetic-resonance imaging. Lancet 351:1763, 1998.

44. Hills MW, Deane SA: Head injury and facial injury: is there an increased risk of cervical spine injury? J Trauma 34:549, 1993.

45. Michael DB, Guyot DR, Darmody WR: Coincidence of head and cervical spine injury. J Neurotrauma 6:177,1989.

46. Fehlings MG, Tatar CH: An evidence-based review of de-compressive surgery in acute spinal cord injury: rationale, indications, and timing based on experimental and clinical studies. J Neurosurg 91:1, 1999.

47. Donovan WH: Operative and nonoperative management of spinal cord injury. A review. Paraplegia 32:375, 1994.

48. Wijdicks EFM: Neurologic complications of multisystem trauma. In Neurologic Complications of Critical Illness. 2nd ed. New York: Oxford University Press, 2002, pp. 302–336.

49. Bachulis BL, Long WB, Hynes GD, et al.: Clinical indications for cervical spine radiographs in the traumatized patient. Am J Surg 153:473, 1987.

50. Kreipke DL, Gillespie KR, McCarthy MC, et al.: Reliability of indications for cervical spine films in trauma patients. J Trauma 29:1438, 1989.

51. Mirvis SE, Diaconis JN, Chirico PA, et al.: Protocol-driven radiologic evaluation of suspected cervical spine injury: efficacy study. Radiology 170:831, 1989.

52. Imhof H, Fuchsjäger M: Traumatic injuries: imaging of spinal injuries. Eur Radiol 12:1262, 2002.

53. Kalfas I, Wilberger J, Goldberg A, et al.: Magnetic resonance imaging in acute spinal cord trauma. Neurosurgery 23:295, 1988.

54. Wittenberg RH, Boetel U, Beyer HK: Magnetic resonance imaging and computer tomography of acute spinal cord trauma. Clin Orthop 260:176, 1990.

55. Bracken MB, Shepard MJ, Holford TR, et al.: Administration of methylprednisolone for 24 or 48 hours or tirilazad mesylate for 48 hours in the treatment of acute spinal Cord injury. Results of the Third National Acute Spinal Cord Injury Randomized Controlled Trial. JAMA 277: 1597, 1997.

56. Wicks AB, Menter RR: Long-term outlook in quadriplegic patients with initial ventilator dependency. Chest 90: 406, 1986.

57. Ditunno JF Jr, Formal CS: Spinal cord injury. In JM Gilchrist (ed), Prognosis in Neurology. Boston: Butterworth-Heinemann, 1998, p. 287.

58. Akmal M, Trivedi R, Sutcliffe J: Functional outcome in trauma patients with spinal injury. Spine 28:180, 2003.

Appendices

Appendix 19.1. American Spinal Injury Association (ASIA) Impairment Scale

Figure. No Caption Avalable.

Figure. No Caption Avalable.