Browse the corpus

1696 SECTION 21: Trauma Spine Trauma Steven Go INTRODUCTION AND EPIDEMIOLOGY Trauma to the spine can cause a vertebral column injury, a spinal cord injury, or both. A few studies have tried to estimate the annual incidence of specific types of vertebral column injuries in the general population with results ranging from 15 to 65 cases per 100,000, 1,2 but no current figures are available for the U.S. population. In contrast, the estimated annual incidence of spinal cord injury in the United States is 54 cases per million or 17,500 new cases per year, with 81% male victims, a mean age of 42 years, and a 63% non-Hispanic white predominance. 3 Since 2010, the leading causes of spinal cord injury are vehicular (38%), falls (31%), and violence (13%). Lifetime costs for spinal cord injury victims vary according to age at time of injury, severity of injury, and socioeconomic status; however, estimates range in millions of dollars per patient. FUNCTIONAL ANATOMY  VERTEBRAL COLUMN The vertebral column is composed of 33 vertebrae: 7 cervical, 12 tho racic, 5 lumbar, 5 fused sacral, and 4 (usually fused) coccygeal. The axial vertebrae (C1 and C2) are anatomically unique in that they are designed for rotary motion. The odontoid (dens) of the axis (C2) is held against the atlas (C1) by the strong transverse ligament. The remaining vertebrae share some common anatomic features (Figure 258-1). A typical subaxial vertebra is composed of an anterior body and a posterior CHAPTER Lamina Ligamentum flavum Spinous process Transverse processBody Supraspinous ligament Intervertebral disk Cut surface of pedicle Posterior longitudinal ligament Interspinous ligament Anterior longitudinal ligament FIGURE 258-1. Vertebral anatomy. Each vertebra consists of a vertebral body and posterior element. Vertebrae are stabilized by an anterior longitudinal ligament, posterior ligament, and interspinous ligament. vertebral arch. The vertebral arch is comprised of two pedicles, two laminae, and seven processes (one spinous, two transverse, and four articular). These articulations enable the spine to engage in flexion, extension, lateral flexion, rotation, or circumduction (combination of all movements). The orientation of these articular facet joints changes at different levels of the spine and accounts for variations in motion of specific regions of the vertebral column. Due to its inherent flexibility, the cervical spine is overall the most commonly injured region of the spinal column, with most injuries occurring at the C2 level and from C5 to C7. 4 The second most common region of injury is in the thoracolumbar transition zone (T11 to L2). However, these population tendencies can be influenced by mechanism of injury. For example, in one study, 5 cervical spine fractures were more common in traffic collisions, whereas falls more commonly resulted in lumbar fractures. Even the type of motor vehicle collision can skew injury patterns, as motorcycle drivers tend to have cervical spine injuries much more commonly than other types of motor vehicle collision victims. A series of ligaments serves to maintain alignment of the spinal column. The anterior and posterior longitudinal ligaments run along the vertebral bodies. Surrounding the vertebral arch are the ligamentum flavum and the supraspinous, interspinous, intertransverse, and capsu lar ligaments.

tor vehicle collision victims. A series of ligaments serves to maintain alignment of the spinal column. The anterior and posterior longitudinal ligaments run along the vertebral bodies. Surrounding the vertebral arch are the ligamentum flavum and the supraspinous, interspinous, intertransverse, and capsu lar ligaments. Between adjacent vertebral bodies are the intervertebral disks, consisting of a peripheral annulus fibrosus and a central nucleus pulposus. The intervertebral disks act as shock absorbers to distribute axial load. When compressive forces exceed the absorptive capacity of the disk, the annulus fibrosus ruptures. This allows the nucleus pulposus to protrude into the vertebral canal, and this may result in spinal nerve or spinal cord compression.  SPINAL CORD The spinal cord is a cylindrical structure that begins at the foramen magnum, where it is continuous with the medulla oblongata of the brain and extends down the spinal canal to the first and second lumbar vertebrae. The spinal cord gives rise to 31 pairs of spinal nerves: 8 cervical, 12 thoracic, Tintinalli_Sec21_p1669-1766.indd 1696 8/1/19 12:20 PM

structure that begins at the foramen magnum, where it is continuous with the medulla oblongata of the brain and extends down the spinal canal to the first and second lumbar vertebrae. The spinal cord gives rise to 31 pairs of spinal nerves: 8 cervical, 12 thoracic, Tintinalli_Sec21_p1669-1766.indd 1696 8/1/19 12:20 PM CHAPTER 258:  Spine Trauma      1697 5 lumbar, 5 sacral, and 1 coccygeal. Each spinal nerve emerges through the intervertebral foramen corresponding to the appropriate spinal cord level. The lower nerve roots form an array of nerves called the cauda equina (“horse’s tail”). PATHOPHYSIOLOGY  SPINAL COLUMN INJURIES Given their multiple axes of motion, the bony vertebrae can be injured via several mechanisms and present with a number of different injury patterns (Table 258-1). 7-12 The variable anatomic qualities of the regions of the spinal column cause characteristic injury patterns in each region. The cervical spine (C1-C7) is particularly vulnerable to injury because it is the most exposed, flexible, and mobile portion of the spinal column. The cervicothoracic junction (C7-T1) is one of the transitional zones of the spinal column, which are locations where the vertebral morphology changes. This designation is important because transitional zones sustain the greatest amount of stress during motion and are most vul nerable to injury. In contrast to the cervical spine, the thoracic spine (T1-T10) is a rigid segment, with its stiffness enhanced by articula tion with the rib cage. Therefore, not only is injury to the thoracic spine less common than in other regions, but this also means that the presence of a thoracic vertebral injury indicates the patient was subjected to severe traumatic forces and is at high risk for intratho racic injuries. Moreover, the spinal canal in the thoracic region is also narrower than in other regions. This increases the risk of cord injury, which is often complete when it occurs. The thoracolumbar junction (T11-L2) is a transitional zone between the highly fixed thoracic and relatively mobile lumbar spine. In addition to this change in bone anatomy, the thoracolumbar junction serves as the level of transition from the end of the spinal cord (about L1) to the nerve roots of the cauda equina. Relative to the thoracic spine, the width of the spinal canal in the thoracolumbar region is greater. Therefore, despite a large number of vertebral injuries at the thoracolumbar junction, most do not have neurologic deficits, or, if present, they are partial or incom plete. Relative to the thoracic and thoracolumbar regions, the lower lumbar spine (L3-L5) is more mobile. Because of the width of the spinal canal in the lumbar region and the ending of the spinal cord at the L1 level, isolated fractures of the lower lumbar spine rarely injure the spinal cord or result in neurologic injury. The sacrum and coccyx form the lower portion of the spinal column. The vertebral foramina of the sacrum together form the sacral canal that contains the nerve roots of the lumbar, sacral, and coccygeal spinal nerves and the filum terminale. The coccyx, which articulates with the sacrum, consists of four vertebrae fused together. When neurologic injuries occur, they are usually complete cauda equina lesions or isolated nerve root deficits. Sacral fractures that involve the central sacral canal can produce bowel or bladder dysfunction.  FRACTURE STABILITY Much has been written about determining whether or not a particular injury is “stable. ” Spinal stability is defined as the ability of the spine to limit patterns of displacement under physiologic loads so as not to damage or irritate the spinal cord or nerve roots.

ce bowel or bladder dysfunction.  FRACTURE STABILITY Much has been written about determining whether or not a particular injury is “stable. ” Spinal stability is defined as the ability of the spine to limit patterns of displacement under physiologic loads so as not to damage or irritate the spinal cord or nerve roots. Several paradigms have been created, including the Denis column system, which splits the spinal column into anterior, middle, and posterior elements. 13 A spine injury is considered unstable if at least two columns of a particular region are involved. Although this schema and other instability scoring systems have been published, 14-17 determining spinal stability after an acute injury in the ED is particularly difficult. This is because these injuries often occur in the setting of polytrauma, altered mental status, and severe pain, which may result in suboptimal initial imaging. In addition, many EDs lack quick access to emergent MRI to evaluate the spinal ligaments. Therefore, assume any spine fracture is unstable, and maintain appropriate precautions until expert consultation can be obtained from a spine surgeon.  SPINAL CORD INJURIES Damage to the spinal cord is the result of two types of injury. First is the primary injury from mechanical forces that directly traumatize the spinal cord and vasculature. This insult sets into motion a series of vascular and chemical processes that lead to secondary injury. 18 The initial phase is characterized by hemorrhage into the cord and forma tion of edema at the injured site and surrounding region. Local spinal cord ischemia ensues secondary to vasospasm and thrombosis of the small arterioles within the gray and white matter. Extension of edema may further compromise blood flow and increase ischemia. A later tis sue degeneration phase begins within hours of injury. This is associated with neural membrane dysfunction, driven by a pathologic excitation of sodium ion channels, an influx of calcium ions, and the release of glutamine. 19 Cell death ensues from a combination of mechanisms including electrolyte imbalances, cell edema, and the formation and release of oxidative substances.  SPINAL CORD LESIONS The severity of spinal cord injury determines the prognosis for recov ery of function, so it is important to distinguish between complete and incomplete spinal cord injuries. The American Spinal Injury Associa tion defines a complete neurologic lesion as the absence of sensory and motor function below the level of injury . This includes loss of function to the level of the lowest sacral segment. In contrast, a lesion is incomplete if sensory, motor, or both functions are partially present below the neurologic level of injury. This may consist only of sacral sensation at the anal mucocutaneous junction or voluntary contrac tion of the external anal sphincter upon digital examination. Complete lesions have a minimal chance of functional motor recovery. Patients with incomplete lesions are expected to have at least some degree of recovery. The differentiation between complete and incomplete spinal cord damage may be complicated by the presence of spinal shock. Patients in spinal shock lose all reflex activities below the area of injury, and lesions cannot be deemed truly complete until spinal shock has resolved. A number of descending and ascending tracts have been identified in the spinal cord (Figure 258-2). The three most important of these in terms of neuroanatomic localization of cord lesions are the corticospinal tracts, spinothalamic tracts, and dorsal (posterior) columns. The corticospinal tract is a descending motor pathway. Its fibers originate from the cerebral cortex through the internal capsule and the middle of the crus cerebri.

ortant of these in terms of neuroanatomic localization of cord lesions are the corticospinal tracts, spinothalamic tracts, and dorsal (posterior) columns. The corticospinal tract is a descending motor pathway. Its fibers originate from the cerebral cortex through the internal capsule and the middle of the crus cerebri. The tract then breaks up into bundles in the pons and finally collects into a discrete bundle, forming the pyramid of the medulla. In the lower medulla, approximately 90% of the fibers cross to the side opposite that of their origin and descend through the spinal cord as the lateral corticospinal tract. These fibers synapse on lower motor neurons in the spinal cord. The 10% of corticospinal fibers that do not cross in the medulla descend in the anterior funiculus of the cervical and upper thoracic cord levels as the ventral corticospinal tract. Damage to the corticospinal tract neurons (upper motor neu rons) in the spinal cord results in ipsilateral clinical findings such as muscle weakness, spasticity, increased deep tendon reflexes, and a Babinski’s sign. The two major ascending pathways that transmit sensory infor mation are the spinothalamic tracts and the dorsal columns. The spinothalamic tracts transmit pain and temperature sensation. As the axons of the first neurons enter the spinal cord, most ascend one or two levels before entering the dorsal gray matter of the spinal cord, where they synapse with the second neuron of the spinothalamic tract. The second neuron immediately crosses the midline in the ante rior commissure of the spinal cord and ascends in the anterolateral funiculus as the lateral spinothalamic tract. Therefore, when the spi nothalamic tract is damaged, the patient experiences loss of pain and temperature sensation in the contralateral half of the body. The (pain and temperature) sensory loss begins one or two segments below the level of the damage. The dorsal columns transmit vibration and proprioceptive information. Neurons enter the spinal cord proximal to pain and temperature Tintinalli_Sec21_p1669-1766.indd 1697 8/1/19 12:20 PM

the contralateral half of the body. The (pain and temperature) sensory loss begins one or two segments below the level of the damage. The dorsal columns transmit vibration and proprioceptive information. Neurons enter the spinal cord proximal to pain and temperature Tintinalli_Sec21_p1669-1766.indd 1697 8/1/19 12:20 PM 1698 SECTION 21: Trauma TABLE 258-1 Major Spinal Column Injuries Mechanism of Injury Injury Spinal Column Regions Typically Affected Image Notes Flexion Anterior subluxation (hyperflexion sprain) (usually stable, but depends on the integrity of posterior ligaments) Cervical [Photo contributors: Mark Silverberg, MD/Steven Pulitzer, MD. (Reproduced with permission from Shah BR, Lucchesi M, Amodio J (Eds): Atlas of Pediatric Emergency Medicine, 2ed, © 2013, McGraw-Hill Education, New York, NY. Figure 20-57.] Anterior subluxation produces ligamentous failure and may have no associated fractures. Plain films can be normal. However, significant ligamentous injury can display anterior soft tissue swelling, a widening of the spinous processes at the level of injury (“fanning”), posterior widening of the intervertebral space, and cervical disk space alignment ≥11 degrees between adjacent spaces. Atlantoaxial dislocation (unstable) Cervical [Used with permission of Jake Block, MD.] Transverse ligament rupture without an associated fracture can occur in older patients from a direct blow to the occiput. Radiographic diagnosis relies on measuring the predental space, which is the space between the posterior aspect of the anterior arch of C1 and the anterior border of the odontoid. A predental space of >3 mm on a lateral radiograph (2 mm for CT images) implies damage to the transverse ligament; >5 mm implies rupture of the transverse ligament. Bilateral interfacetal dislocation (unstable) Cervical Bilateral interfacetal dislocation (locked facets) occurs when the articular masses of one vertebra dislocate anteriorly and superiorly from the articular surfaces of the adjacent vertebra below. Disruption of all ligamentous structures occurs. On radiographs, the vertebral body is dislocated anteriorly ≥50% of its width. These injuries usually present with neurologic deficits due to compromise of the intervertebral foramen, unless the dislocation is only partial (perched facets). (Continued) Tintinalli_Sec21_p1669-1766.indd 1698 8/1/19 12:20 PM

tures occurs. On radiographs, the vertebral body is dislocated anteriorly ≥50% of its width. These injuries usually present with neurologic deficits due to compromise of the intervertebral foramen, unless the dislocation is only partial (perched facets). (Continued) Tintinalli_Sec21_p1669-1766.indd 1698 8/1/19 12:20 PM CHAPTER 258:  Spine Trauma      1699 TABLE 258-1 Major Spinal Column Injuries Mechanism of Injury Injury Spinal Column Regions Typically Affected Image Notes

tures occurs. On radiographs, the vertebral body is dislocated anteriorly ≥50% of its width. These injuries usually present with neurologic deficits due to compromise of the intervertebral foramen, unless the dislocation is only partial (perched facets). (Continued) Tintinalli_Sec21_p1669-1766.indd 1698 8/1/19 12:20 PM CHAPTER 258:  Spine Trauma      1699 TABLE 258-1 Major Spinal Column Injuries Mechanism of Injury Injury Spinal Column Regions Typically Affected Image Notes Simple wedge (compression) fracture (usually stable) Cervical; TL [Reproduced with permission from Block J, Jordanov MI, Stack LB, Thurman RJ (Eds): The Atlas of Emergency Radiology. McGraw-Hill, Inc., 2013. Fig. 11-22 Part A.] Most common thoracic fracture (52%).5 A vertebral wedge fracture typically involves a fracture of the superior end plate of the vertebral body while sparing the inferior end plate. An isolated simple wedge fracture is stable, but the presence of significant posterior ligamentous disruption can make the injury unstable. A simple wedge fracture is differentiated from a burst fracture by the absence of a vertical fracture of the vertebral body and lack of bulging of the posterior vertebral border. Spinous process avulsion fracture (stable) Cervical [Reproduced with permission from Block J, Jordanov MI, Stack LB, Thurman RJ (Eds): The Atlas of Emergency Radiology. McGraw-Hill, Inc., 2013. Fig. 11-10 Part A.] This is a spinous process avulsion most commonly seen in the lower cervical and upper thoracic spine. When a single avulsion is present in this region, it is often called a “clay-shoveler” fracture. It is thought to be caused by strong muscle contractions pulling on the bone via the ligamentous complex. In isolation, it is generally not associated with neurologic compromise. Flexion teardrop fracture (highly unstable) Cervical [Reproduced with permission from Block J, Jordanov MI, Stack LB, Thurman RJ (Eds): The Atlas of Emergency Radiology. McGraw-Hill, Inc., 2013. Fig. 11-1.] Extreme hyperflexion causes complete disruption of the spinal ligaments at the level of injury. The “teardrop” is the anteroinferior portion of the vertebral body that is separated and displaced from the vertebral body by the anterior spinal ligament. “Fanning” of the spinous processes may be present, with or without fracture. A sagittal fracture through the vertebral body may be seen on CT. Anterior spinal cord syndrome is associated with this injury (Continued) (Continued) Tintinalli_Sec21_p1669-1766.indd 1699 8/1/19 12:20 PM

body by the anterior spinal ligament. “Fanning” of the spinous processes may be present, with or without fracture. A sagittal fracture through the vertebral body may be seen on CT. Anterior spinal cord syndrome is associated with this injury (Continued) (Continued) Tintinalli_Sec21_p1669-1766.indd 1699 8/1/19 12:20 PM 1700 SECTION 21: Trauma TABLE 258-1 Major Spinal Column Injuries Mechanism of Injury Injury Spinal Column Regions Typically Affected Image Notes Flexionrotation Unilateral facet dislocation (stable unless associated with an articular mass fracture) Cervical [Reproduced with permission from Simon RR, Sherman Scott C (eds): Emergency Orthopedics, 6th ed. McGraw-Hill, Inc., 2011. Fig 9-21B.] A unilateral facet dislocation occurs when the articular mass and inferior facet on one side of the vertebra are anteriorly dislocated. On a lateral radiograph, the involved vertebral body will be displaced <50% of its width. On the anterior view, the spinous process at the level of the rotation will be pointing toward the side that is dislocated. Fracture of lateral mass (can be unstable) Cervical Comminuted fracture of the lateral mass of C4 extending into the right lamina. Typically presents with severe neck pain and sometimes radicular symptoms. May be associated with Brown-Séquard syndrome or vertebral artery injury; therefore, some experts feel that magnetic resonance angiography should be done in all patients with this lesion.6 A pillar fracture is a type of lateral mass fracture that consists of an isolated vertical or oblique fracture through the lateral mass. The adjacent lamina and pedicle remain intact. The fractured articular mass is displaced posteriorly and may be visible as a double outline on the lateral radiograph. Flexiondistraction Anterior compression with associated transverse fracture through vertebral body (unstable) Arrow points to splaying of the posterior elements.   [Reproduced with permission from Block J, Jordanov MI, Stack LB, Thurman RJ (eds): The Atlas of Emergency Radiology . McGraw-Hill,  Inc., 2013. Fig 11-27C.] These injuries are associated with seatbelt injuries, especially when lap belts alone are used. Radiographic findings include posterior vertebral wall fracture, increased height of the posterior vertebra, and “fanning” of the spinous processes. The Chance fracture variant presents with minor anterior vertebral compression and significant distraction of the middle and posterior ligamentous structure. It often occurs from T11 to L2 (TL transition zone). These injuries are often misdiagnosed as an anterior compression fracture. They may require CT to visualize and are often associated with intra-abdominal injuries. (Continued) (Continued) Tintinalli_Sec21_p1669-1766.indd 1700 8/1/19 12:20 PM

or ligamentous structure. It often occurs from T11 to L2 (TL transition zone). These injuries are often misdiagnosed as an anterior compression fracture. They may require CT to visualize and are often associated with intra-abdominal injuries. (Continued) (Continued) Tintinalli_Sec21_p1669-1766.indd 1700 8/1/19 12:20 PM CHAPTER 258:  Spine Trauma      1701 TABLE 258-1 Major Spinal Column Injuries Mechanism of Injury Injury Spinal Column Regions Typically Affected Image Notes Vertical compression Jefferson burst fracture of atlas (potentially unstable) Cervical [Reproduced with permission from Block J, Jordanov MI, Stack LB, Thurman RJ (eds): The Atlas of Emergency Radiology. McGraw-Hill, Inc., 2013. Fig 11-13.] Vertical compression forces the occipital condyles downward and produces a burst fracture by driving the lateral masses of C1 apart. This is best seen as outward displacement of the lateral masses on the open-mouth odontoid radiograph or on CT. If displacement of both lateral masses (measured as offset from the superior corner of the C2 vertebral body on each side) is >7 mm when added together, rupture of the transverse ligament is likely, and the spine is unstable. Burst fracture (unstable) Cervical; TL A burst fracture occurs when a vertebra is crushed by an axial load, causing fragments to displace in all directions. The lateral radiograph may show an obvious fracture of the end plates, but sometimes all that is seen is a bowing or disruption of the posterior cortex of the affected vertebra. The anterior radiographic view may show a vertical fracture through the vertebral body and widening of the interpedicular distance. The burst fracture is usually obvious on CT. The spinal cord may be injured if a retropulsed fragment enters the spinal canal. (Continued) (Continued) Tintinalli_Sec21_p1669-1766.indd 1701 8/1/19 12:21 PM

iographic view may show a vertical fracture through the vertebral body and widening of the interpedicular distance. The burst fracture is usually obvious on CT. The spinal cord may be injured if a retropulsed fragment enters the spinal canal. (Continued) (Continued) Tintinalli_Sec21_p1669-1766.indd 1701 8/1/19 12:21 PM 1702 SECTION 21: Trauma TABLE 258-1 Major Spinal Column Injuries Mechanism of Injury Injury Spinal Column Regions Typically Affected Image Notes Extension Hyperextension dislocation (unstable) Cervical [Reproduced with permission from Schwartz DT (ed): Emergency Radiology: Case Studies. McGraw-Hill, Inc., 2008. Sect V: Cervical Spine Radiology; Fig 6.] Extreme hyperextension can cause a complete tear of the anterior longitudinal ligament and intervertebral disk, with disruption of the posterior ligamentous complex. On the lateral radiographic view, the vertebrae may appear normal if the dislocation spontaneously reduces or if the injury is masked by a cervical immobilization collar. Prevertebral soft tissue swelling may be the only radio graphic finding present. Anterior disk space widening or fracture of the anteroinferior end plate of the vertebral body may occur. Patients usually present with a central cord syndrome. Hyperextension teardrop fracture or extension corner avulsion fracture (unstable in extension) Cervical Hyperextension may cause the anterior longitudinal ligament to avulse a fragment off the anteroinferior corner of the vertebral body. The height of the avulsed fragment usually exceeds its width. This fracture is more common in older patients with osteoporosis. Fracture of posterior arch of atlas (stable) Cervical Fracture of posterior arch of C1. [Reproduced  with permission from Galli, et al: Emergency Orthopedics: The Spine. New York, NY: McGraw-Hill, 1989.] Fracture occurs from wedging of the posterior arch between the occipital bone and the C2 vertebra. A CT is indicated to rule out an associated Jefferson fracture or a dens fracture. (Continued) (Continued) Tintinalli_Sec21_p1669-1766.indd 1702 8/1/19 12:21 PM CHAPTER 258:  Spine Trauma      1703 TABLE 258-1 Major Spinal Column Injuries Mechanism of Injury Injury Spinal Column Regions Typically Affected Image Notes

1702 SECTION 21: Trauma TABLE 258-1 Major Spinal Column Injuries Mechanism of Injury Injury Spinal Column Regions Typically Affected Image Notes Extension Hyperextension dislocation (unstable) Cervical [Reproduced with permission from Schwartz DT (ed): Emergency Radiology: Case Studies. McGraw-Hill, Inc., 2008. Sect V: Cervical Spine Radiology; Fig 6.] Extreme hyperextension can cause a complete tear of the anterior longitudinal ligament and intervertebral disk, with disruption of the posterior ligamentous complex. On the lateral radiographic view, the vertebrae may appear normal if the dislocation spontaneously reduces or if the injury is masked by a cervical immobilization collar. Prevertebral soft tissue swelling may be the only radio graphic finding present. Anterior disk space widening or fracture of the anteroinferior end plate of the vertebral body may occur. Patients usually present with a central cord syndrome. Hyperextension teardrop fracture or extension corner avulsion fracture (unstable in extension) Cervical Hyperextension may cause the anterior longitudinal ligament to avulse a fragment off the anteroinferior corner of the vertebral body. The height of the avulsed fragment usually exceeds its width. This fracture is more common in older patients with osteoporosis. Fracture of posterior arch of atlas (stable) Cervical Fracture of posterior arch of C1. [Reproduced  with permission from Galli, et al: Emergency Orthopedics: The Spine. New York, NY: McGraw-Hill, 1989.] Fracture occurs from wedging of the posterior arch between the occipital bone and the C2 vertebra. A CT is indicated to rule out an associated Jefferson fracture or a dens fracture. (Continued) (Continued) Tintinalli_Sec21_p1669-1766.indd 1702 8/1/19 12:21 PM CHAPTER 258:  Spine Trauma      1703 TABLE 258-1 Major Spinal Column Injuries Mechanism of Injury Injury Spinal Column Regions Typically Affected Image Notes Laminar fracture (usually stable) Cervical [Reproduced with permission from Simon RR and Sherman SC (EDS). Emergency Orthopedics, 6 ed. McGraw-Hill, Inc., 2011. Fig. 9-26B.] Laminar fractures may be associated with spinous process fractures. They may not be evident on plain radiographs and usually require CT for diagnosis. Traumatic spondylolisthesis (hangman’s fracture) (unstable) Cervical The hangman’s fracture is a fracture of both pedicles of C2, with the anterior displacement of C2 on C3. This was associated with the neck hyperextension from judicial hangings, where the noose knot is placed under the subject’s chin and snaps the head backward. Suicidal hangings do not usually cause extreme hyperextension and are not associated with the hangman’s fracture. Because the spinal canal at the level of C2 is large, a hangman’s fracture does not cause neurologic injury. Injuries caused by a combination of mechanisms or poorly understood mechanisms Occipital condyle fractures (usually stable) Cervical [Reproduced with permission from Block J, Jordanov MI, Stack LB, Thurman RJ (eds): The Atlas of Emergency Radiology. McGraw-Hill, Inc., 2013. Fig. 11-11B.] Occipital condyle fractures are rarely visible on plain radiographs and usually require CT imaging for detection. Presentation is rather variable due to proximity of multiple neurovascular structures.7 Neurologic impairment is common and usually involves lower cranial nerve deficits and/or limb weakness. (Continued) (Continued) Tintinalli_Sec21_p1669-1766.indd 1703 8/1/19 12:21 PM

aphs and usually require CT imaging for detection. Presentation is rather variable due to proximity of multiple neurovascular structures.7 Neurologic impairment is common and usually involves lower cranial nerve deficits and/or limb weakness. (Continued) (Continued) Tintinalli_Sec21_p1669-1766.indd 1703 8/1/19 12:21 PM 1704 SECTION 21: Trauma TABLE 258-1 Major Spinal Column Injuries Mechanism of Injury Injury Spinal Column Regions Typically Affected Image Notes

aphs and usually require CT imaging for detection. Presentation is rather variable due to proximity of multiple neurovascular structures.7 Neurologic impairment is common and usually involves lower cranial nerve deficits and/or limb weakness. (Continued) (Continued) Tintinalli_Sec21_p1669-1766.indd 1703 8/1/19 12:21 PM 1704 SECTION 21: Trauma TABLE 258-1 Major Spinal Column Injuries Mechanism of Injury Injury Spinal Column Regions Typically Affected Image Notes Atlanto-occipital dissociation (AOD) (highly unstable) Cervical [Photo contributors: Konstantinos Agoritsas, MD/Steven Pulitzer, MD. Reproduced with permission from Shah BR, Lucchesi M, Amodio J (eds): Atlas of Pediatric Emergency Medicine, 2nd ed. © 2013, McGraw-Hill Education, New York, NY. Fig 20-52.] Secondary to high-energy impact. Historically, strongly associated with mortality8; however, modern patients may survive due to better prehospital care/transport. The classic presentation is paralysis of upper extremities with lack of lower extremity paralysis or weakness (cruciate paralysis).9 However, presentation can be variable with a common presentation being lower cranial nerve deficits. CT may be required for detection. In radiographs in the normal patient, the distance between the basion and the superior cortex of the dens (basion-dental interval [BDI]) should be ≤10 mm in adults (≤8.5 mm on CT). In addition, the distance from the basion to the posterior border of the body of C2 (basion-atlantal interval [BAI]) should be ≤12 mm anterior displacement or ≤4 mm posterior displacement on a lateral radiograph. If there are abnormalities in both the BDI and BAI, this strongly suggests the existence of AOD.10 Odontoid (dens) fractures (type II and III are unstable) Cervical Type 1 odontoid fracture. [Reproduced with permission from Block J, Jordanov MI, Stack LB, Thurman RJ (eds): The Atlas of Emergency Radiology. McGraw-Hill, Inc., 2013. Fig 11-18.] Frequently involves other injuries to the cervical spine and multisystem trauma. Conscious patients will usually describe immediate and severe high cervical pain with muscle spasm. The pain may radiate to the occiput. Neurologic injury is present in 18% to 25% of cases with odontoid fractures, ranging from minimal sensory or motor loss to quadriplegia. Odontoid fractures are classified according to the level of injury. CT can miss odontoid fractures if the fracture line is aligned with the cut of the CT (en face). Type 2 odontoid  fracture.  [Reproduced  with permission  from Block  J, Jordanov  MI, Stack LB, Thurman  RJ (eds):  The Atlas of Emergency Radiology . McGraw-Hill,  Inc.,  2013.  Fig. 11-19C.] (Continued) (Continued) Tintinalli_Sec21_p1669-1766.indd 1704 8/1/19 12:21 PM

ine is aligned with the cut of the CT (en face). Type 2 odontoid  fracture.  [Reproduced  with permission  from Block  J, Jordanov  MI, Stack LB, Thurman  RJ (eds):  The Atlas of Emergency Radiology . McGraw-Hill,  Inc.,  2013.  Fig. 11-19C.] (Continued) (Continued) Tintinalli_Sec21_p1669-1766.indd 1704 8/1/19 12:21 PM CHAPTER 258:  Spine Trauma      1705 TABLE 258-1 Major Spinal Column Injuries Mechanism of Injury Injury Spinal Column Regions Typically Affected Image Notes Type 3 odontoid fracture. [Reproduced with permission from Block J, Jordanov MI, Stack LB, Thurman RJ (eds): The Atlas of Emergency Radiology. McGraw-Hill, Inc., 2013. Fig 11-20C.] Translational fracturedislocation (unstable) T10-T11 fracture-dislocation. This is a high-energy disruption of all three columns of spine and is readily apparent both on radiographs and CT. Patients commonly present with severe neurologic findings. These fractures are most often unstable; however, in the absence of destabilizing rib cage fractures, lesions above T7 can be stable. (Continued) (Continued) Tintinalli_Sec21_p1669-1766.indd 1705 8/1/19 12:21 PM

d is readily apparent both on radiographs and CT. Patients commonly present with severe neurologic findings. These fractures are most often unstable; however, in the absence of destabilizing rib cage fractures, lesions above T7 can be stable. (Continued) (Continued) Tintinalli_Sec21_p1669-1766.indd 1705 8/1/19 12:21 PM 1706 SECTION 21: Trauma TABLE 258-1 Major Spinal Column Injuries Mechanism of Injury Injury Spinal Column Regions Typically Affected Image Notes Sacrum and coccyx fractures Sacral fracture [Reproduced with permission from Block J, Jordanov MI, Stack LB, Thurman RJ (eds): The Atlas of Emergency Radiology. McGraw-Hill, Inc., 2013. Fig. 8-16A.] Usually associated with pelvic fracture(s). Transverse fractures through the body can injure the cauda equina. Longitudinal fractures can cause radiculopathies. Central sacral fracture can present with bowel/bladder incontinence. Coccyx fracture [Reproduced with permission from Block J, Jordanov MI, Stack LB, Thurman RJ (eds): The Atlas of Emergency Radiology. McGraw-Hill, Inc., 2013. Fig 8-24.] Coccygeal injuries are usually associated with a direct fall onto the buttocks, with resultant coccyx pain exacerbated by sitting or straining. Localized tenderness can be elicited with coccyx palpation during a rectal exam, but this is not required for diagnosis. Imaging is not needed to diagnose coccygeal fractures. Treatment is symptomatic with analgesics and use of a rubber doughnut pillow. Abbreviations: C1 = first cervical vertebra; C2 = second cervical vertebra; C3 = third cervical vertebra; C5 = fifth cervical vertebra; C7 = seventh cervical vertebra; T1 = first thoracic vertebra; T7 = seventh thoracic vertebra; T12 = 12th thoracic vertebra; TL = thoracolumbar. (Continued) neurons. They differ from pain and temperature neurons in that they do not immediately synapse in the spinal cord. Instead, these axons enter the ipsilateral dorsal column and do not synapse until they reach the gracile or cuneate nuclei of the medulla. From these nuclei, fibers cross the midline and ascend in the medial lemniscus to the thalamus. Injury to one side of the dorsal columns will result in ipsilateral loss of vibration and position sense. The sensory loss begins at the level of the lesion. Light touch is transmitted through both the spinothalamic tracts and the dorsal columns. Therefore, light touch is not completely lost unless there is damage to both the spinothalamic tracts and the dorsal columns. Each spinal nerve is named for its adjacent vertebral body (Figure 258-3). In the cervical region, there is an additional pair of spinal nerve roots compared to the number of vertebral bodies. The first seven spinal nerves are named for the first seven cervical vertebrae, each exiting through the intervertebral foramen above its corresponding vertebral body. The spinal nerve exiting below C7, however, is referred to as the C8 spinal nerve, although no eighth cervical vertebra exists. All subse quent nerve roots, beginning with T1, exit below the vertebral body for which they are named. During fetal development, the downward growth of the vertebral column is greater than that of the spinal cord. Because the adult spinal cord ends as the conus medullaris at the level of the lower border of the first lumbar vertebra, the lumbar and sacral nerve roots must continue inferiorly below the termination of the spinal cord to exit from their respective intervertebral foramina. These nerve roots form the cauda equina. A potential consequence of this arrangement is that injury to Tintinalli_Sec21_p1669-1766.indd 1706 8/1/19 12:21 PM

ertebra, the lumbar and sacral nerve roots must continue inferiorly below the termination of the spinal cord to exit from their respective intervertebral foramina. These nerve roots form the cauda equina. A potential consequence of this arrangement is that injury to Tintinalli_Sec21_p1669-1766.indd 1706 8/1/19 12:21 PM CHAPTER 258:  Spine Trauma      1707 a single lower vertebra can involve multiple nerve roots in the cauda equina. For example, an injury at the L3 vertebra can involve the L3 nerve root as well as the lower nerve roots that are progressing to a level caudal to the L3 vertebra. PREHOSPITAL CARE The prehospital treatment of patients with spinal injury involves recog nition of patients at risk, appropriate immobilization, and triage to an appropriate facility (see Chapter 1, “Emergency Medical Services” and Chapter 2, “Prehospital Equipment”). Presume that patients with an appropriate traumatic mechanism who have complaints of neck or back pain, tenderness on prehospital exam, neurologic complaints, significant injury above the clavicles, or altered sensorium that precludes accurate evaluation of the spine to have a spinal cord injury, and take appropriate spinal precautions. Transport of these patients to a center that is capable of rapid diagnostics and therapeutics is important to optimize outcome following spinal injury. Prehospital care for spinal injuries traditionally involves immobi lization of the entire spine at the scene with a rigid cervical collar (or similar devices) plus a long backboard. However, there is little evidence that cervical collars and/or long spine boards reduce neurologic injury, spinal instability, or mortality. 20,21 In fact, cervical collars and long backboards can induce complications such as pressure sores,22,23 patient discomfort,24 and respiratory compromise.25 In light of these data, some experts have recommended transporting the patient on a gurney with a scoop stretcher 26 or other soft, padded devices 27 to avoid the rigid spine board. In addition, based on data suggesting extrication cervical collars are associated with pressure ulcers, 28 indentation marks, pain, and increased intracranial pressure,29 some authors have proposed early replacement of the extrication cervical collar, 30 or even abandoning the routine use of cervical collars. 31 The most recent position statement by C1 C1 C1 T1 L1 C2 C3 C4 C5 C6 C7 C8 T1 T2 T3 T4 T5 T6 T7 T8 T9 T10 T11 T12 L1 L1 L2 L3 L4 L5 T1 S1 S2 S3 S4 Voluntary rectal toneS2,S3,S4 - Bladder Anal sphincter Ankle plantar flexionS1,S2 Gastrocnemius Great toe extensionL5,S1 - Extensor hallucis longus Ankle dorsiflexionL4,L5 Tibialis anterior Knee flexionL4,L5,S1,S2 Hamstrings Knee extensionL2,L3,L4 - Quadriceps L1,L2,L3 - Iliopsoas Hip flexion T9–T12 - Abdominal muscles T2–T7 - Chest muscles Finger abduction Hand grasp C8,T1 - Hand intrinsics Flexor digitorum profundus Elbow extensionC7 ,C8 - Triceps C6,C7 - Extensor carpi radialis Wrist extension C5,C6 - Deltoid Biceps Arm abduction Elbow flexion S5 FIGURE 258-3. Spinal cord level. The spinal cord level of injury can be delineated by physical examination, including a detailed neurologic examination. Dorsal column (position, vibration, light touch) Sacral Lumber Posterior horn Lateral corticospinal tract (motor) Lateral spinothalamic tract (pain, temperature) Anterior spinal artery Arm Trunk Leg Arm Trunk Leg Thoracic Cervical FIGURE 258-2. The anatomy of a cross section of cervical spinal cord. [Reproduced with permission from Simon RR, Sherman SC (eds):  Emergency Orthopedics, 6th ed. McGraw-Hill, Inc., 2011. Fig 9-5.] Tintinalli_Sec21_p1669-1766.indd 1707 8/1/19 12:21 PM

Anterior spinal artery Arm Trunk Leg Arm Trunk Leg Thoracic Cervical FIGURE 258-2. The anatomy of a cross section of cervical spinal cord. [Reproduced with permission from Simon RR, Sherman SC (eds):  Emergency Orthopedics, 6th ed. McGraw-Hill, Inc., 2011. Fig 9-5.] Tintinalli_Sec21_p1669-1766.indd 1707 8/1/19 12:21 PM 1708 SECTION 21: Trauma the National Association of EMS Physicians and the American College of Surgeons Committee on Trauma suggested utilization of backboards should be “judicious” and used in such a way that “benefits outweigh the risks”; however, they continued to recommend the use of rigid cervical collars. 32 A consensus document by the Faculty of Pre-hospital Care went further, stating that the long board should only be used as an extrication device and not during transport. 33 Furthermore, they strongly argued for “selective immobilization” based on clinical factors (e.g., level of consciousness) and that manual in-line stabilization may be more appropriate than cervical collars in the setting of airway com promise, increased intracranial pressure, or patient combativeness and in children. 33 However, some authors have concluded that these position statements do not go far enough given that there is little evidence to show benefit compared with ample data that demonstrate harm. 34 Thus, this area remains controversial, with all sides calling for more random ized controlled data to settle these issues. In contrast, expert consensus is that spinal immobilization is NOT recommended for fully conscious, neurologically intact patients with isolated penetrating neck injury because collars can delay resuscitation and obscure neck injuries. 32,33 INITIAL ED STABILIZATION  AIRWAY ED evaluation of the patient with potential spinal injury should not differ substantially from that of any patient with multiple injuries, with the first priority being the airway. The higher the level of spinal injury, the more likely is the need for early airway intervention. For example, unstable spine lesions above C3 can cause immediate respiratory arrest, and lesions affecting C3-C5 can affect the phrenic nerve and diaphragm function. For this reason, some experts recommend that any patient with an injury at C5 or above should have the airway secured by endotracheal intubation. Delayed respiratory compromise can occur if spinal cord edema from more caudal lesions progresses rostrally to cause phrenic nerve paralysis. Many patients can initially support ventilatory function using intercostal muscles or abdominal breathing, but they eventually tire and subsequently develop respiratory failure. Therefore, be vigilant for respiratory compromise in patients with high cervical injuries. If safety allows, perform a brief focused neurologic assessment before sedation and intubation. Given the unclear benefit of cervical collars, do not hesitate to substitute manual in-line stabilization if the collar is compromising the airway. Maintain manual in-line spinal stabilization while intubating, because human cadaver studies demonstrate less cervical motion and better glottis visualization with in-line stabilization than with cervical collars in place, and movement of an unstable cervical spine may produce or worsen spinal cord injury. Videolaryngoscopes improve visualization in cervical immobilized patients and may reduce failed intubations, 36,37 but manual in-line stabilization is still recommended to minimize cervical motion.38  HYPOTENSION Hypotension in patients with spinal cord injuries may be due to neuro genic shock, blood loss, cardiac injury, tension pneumothorax, or other injuries.

ervical immobilized patients and may reduce failed intubations, 36,37 but manual in-line stabilization is still recommended to minimize cervical motion.38  HYPOTENSION Hypotension in patients with spinal cord injuries may be due to neuro genic shock, blood loss, cardiac injury, tension pneumothorax, or other injuries. Although hypotension and relative bradycardia are classic signs of neurogenic shock, bradycardia can also be associated with intraperitoneal bleeding or prior medication with calcium channel blockers or β-blockers. In one study, 39 74% of hypotensive patients with penetrating spinal cord injury had major blood loss causing hypotension. Therefore, presume blood loss as the cause of hypotension in spinal injury patients until proven otherwise. Hypotension is initially treated with IV crystalloid or blood as appropriate. (See Chapter 254, “Trauma in Adults. ”)  SPINE PRECAUTIONS IN THE ED Long spine boards are associated with pressure sores, so remove them as soon as possible if the patient arrives with them. Logrolling is the tradi tional method for board removal. In cadaver studies, logrolling has been shown to generate significantly more spine motion at injured vertebrae than alternative maneuver and transfer techniques 40; however, there is a paucity of published data demonstrating negative clinical outcomes from logrolling. 41 Nevertheless, some experts recommend the “6+ lift and slide maneuver” because it produces less spine motion than logrolling. 42 The 6+ maneuver consists first of unstrapping the patient from the board. Next, one person maintains in-line stabilization at the head, while six others positioned at the chest, pelvis, and lower extremity levels lift the patient as a unit 10 to 20 cm above the board. Another person slides the board out from under the patient, and the patient is then lowered to the bed, maintaining spinal alignment. Disadvantages are the need for many staff members to perform this maneuver and inability to visualize the patient’s back. Hard cervical collars are associated with patient discomfort and pressure sores of the neck. 44 Therefore, promptly clear the cervical spine if possible (see “Clinical Decision Rules in Cervical Spine Imaging” and “Cervical Spine Imaging” below). Do not overtighten the cervical collar on head-injured patients, because jugular venous compression can raise intracranial pressure, 29,45 although Stifneck® and Miami J® collars may be better than other rigid collars in this regard. 46 Avoid physically fighting with an agitated patient in a forcible attempt to place a cervical collar, as overly aggressive attempts to restrain a patient may possibly cause or exacerbate a cervical injury. CLINICAL FEATURES  HISTORY If the patient is able to give a history, focus on key historical points as they pertain to spine injury. Specifically, seek the presence or absence of the historical elements included in imaging decision rules (see Tables 258 4-7 47–50). Obtain an accurate history regarding the mechanism of injury. Evaluate for symptoms of midline spine pain, painful distracting injury, paresthesias, loss of function, change in mental status (including loss of consciousness), or other neurologic symptoms (especially pria pism or urinary or fecal incontinence). Pay particular attention to any symptoms indicating present or impending respiratory compromise, including dyspnea, palpitations, abdominal breathing, and anxiety, which may indicate a high cervical spine injury .  PHYSICAL EXAMINATION Once the patient is stabilized and other life-threatening injuries have been excluded or treated, perform a detailed neurologic assessment. An appropriately detailed initial neurologic examination is important to allow for comparison later should the patient deteriorate.

h cervical spine injury .  PHYSICAL EXAMINATION Once the patient is stabilized and other life-threatening injuries have been excluded or treated, perform a detailed neurologic assessment. An appropriately detailed initial neurologic examination is important to allow for comparison later should the patient deteriorate. Assess the patient’s mental status and note any clinical evidence of intoxication. Focus the physical examination on delineating the level of the spinal cord injury (Figure 258-3). Document the presence or absence of mid line neck or back tenderness. Test motor function for muscle groups (Table 258-2). Determine the level of sensory loss ( Figure 258-4), and investigate proprioception or vibratory function to examine posterior column function. Test for “saddle anesthesia, ” which is a sensory deficit in the region of the buttocks, perineum, and inner aspect of the thighs. Test deep tendon reflexes along with anogenital reflexes because “sacral sparing” with preservation of anogenital reflexes denotes an incomplete spinal cord level, even if the patient has complete sensory and motor loss. To test the bulbocavernosus reflex, squeeze the penis to determine whether the anal sphincter simultaneously contracts. Document anal sphincter tone and sensation around the anus. An “anal TABLE 258-2 Motor Grading System Grade Movement 0 No active contraction 1 Trace visible or palpable contraction 2 Movement with gravity eliminated 3 Movement against gravity 4 Movement against gravity plus resistance 5 Normal power Tintinalli_Sec21_p1669-1766.indd 1708 8/1/19 12:21 PM

ensation around the anus. An “anal TABLE 258-2 Motor Grading System Grade Movement 0 No active contraction 1 Trace visible or palpable contraction 2 Movement with gravity eliminated 3 Movement against gravity 4 Movement against gravity plus resistance 5 Normal power Tintinalli_Sec21_p1669-1766.indd 1708 8/1/19 12:21 PM CHAPTER 258:  Spine Trauma      1709 wink reflex” (contraction of the anal musculature when the perianal region is stimulated with a pin) indicates some sacral sparing. Test the cremasteric reflex by stroking the medial thigh with a blunt instru ment. If the scrotum rises, some spinal cord integrity exists. Conversely, priapism implies a complete spinal cord injury. In 2015, the American Spinal Injury Association published a revised version of the Interna tional Standards for Neurological Classification of Spinal Cord Injury. This scoring system is used by spine surgeons to document their initial examination and has prognostic value 52; however, the scale is rather lengthy and is not practical for ED initial assessment.  INCOMPLETE SPINAL CORD SYNDROMES There are three major incomplete spinal cord syndromes identified by predictable physical examination findings, although overlap in findings may occur (Table 258-3).  ANTERIOR CORD SYNDROME The anterior cord syndrome results from damage to the corticospinal and spinothalamic pathways, with preservation of posterior column function. This is manifested by loss of motor function and pain and temperature sensation distal to the lesion. Only vibration, position, and tactile sensation are preserved. This syndrome may occur fol lowing direct injury to the anterior spinal cord. Flexion of the cervical spine may result in cord contusion or bone injury with secondary cord injury. Alternatively, thrombosis of the anterior spinal artery can cause ischemic injury to the anterior cord. Anterior cord injury can also be produced by an extrinsic mass that is amenable to surgical decompres sion. The overall prognosis for recovery of function is poor.  CENTRAL CORD SYNDROME The central cord syndrome is usually seen in older patients with pre existing cervical spondylosis who sustain a hyperextension injury. As named, this injury preferentially involves the central portion of the cord more than the peripheral. The centrally located fibers of the corticospi nal and spinothalamic tracts are affected. The neural tracts providing function to the upper extremities are most medial in position compared with the thoracic, lower extremity, and sacral fibers that have a more lateral distribution. Clinically, patients with a central cord syndrome present with decreased strength and, to a lesser degree, decreased pain and temperature sensation, more in the upper than the lower extremities. Vibration and position sensation are usually preserved. TABLE 258-3 Three Major Incomplete Spinal Cord Syndromes Syndrome Mechanisms Symptoms General Prognosis * Anterior cord Direct anterior cord compression Complete paralysis below the lesion with loss of pain and temperature sensation Poor Flexion of cervical spine Thrombosis of anterior spinal artery Preservation of proprioception and vibratory function Central cord Hyperextension injuries Quadriparesis—greater in the upper extremities than the lower extremities. Some loss of pain and temperature sensation, also greater in the upper extremities Good Disruption of blood flow to the spinal cord Cervical spinal stenosis Brown- Séquard Transverse hemisection of the spinal cord Ipsilateral spastic paresis, loss of proprioception and vibratory sensation, and contralateral loss of pain and temperature sensation Good Unilateral cord compression *Outcome improves when the effects of secondary injury are prevented or reversed.

pinal stenosis Brown- Séquard Transverse hemisection of the spinal cord Ipsilateral spastic paresis, loss of proprioception and vibratory sensation, and contralateral loss of pain and temperature sensation Good Unilateral cord compression *Outcome improves when the effects of secondary injury are prevented or reversed. C2 C3 C4C5 C5C6 C6 T1 T1C8 C8 C7 C7T2 T3 T4 T5 T6 T7 T8 T9 T10 T12 T1 1 L1 L2 L2L3L5 L5 L4 L4L3 L3S1 S1 S2 S2 S3 S3 S4 S5 S6 Coc C2 C3 C4 C4C5 C5C6 C6 T1 T1C8 C8T2 T3 T4 T5 T6 T7 T8 T9 T10 T12 T1 1 L1L2 L2L5 L5 L4 L4 L3 L3 S1 S1 S3 S3 C7 C7 FIGURE 258-4. Dermatomes for sensory examination. Tintinalli_Sec21_p1669-1766.indd 1709 8/1/19 12:21 PM

C8 C8 C7 C7T2 T3 T4 T5 T6 T7 T8 T9 T10 T12 T1 1 L1 L2 L2L3L5 L5 L4 L4L3 L3S1 S1 S2 S2 S3 S3 S4 S5 S6 Coc C2 C3 C4 C4C5 C5C6 C6 T1 T1C8 C8T2 T3 T4 T5 T6 T7 T8 T9 T10 T12 T1 1 L1L2 L2L5 L5 L4 L4 L3 L3 S1 S1 S3 S3 C7 C7 FIGURE 258-4. Dermatomes for sensory examination. Tintinalli_Sec21_p1669-1766.indd 1709 8/1/19 12:21 PM 1710 SECTION 21: Trauma Spastic paraparesis or spastic quadriparesis can also be seen. The majority will have bowel and bladder control, although this may be impaired in the more severe cases.  BROWN-SÉQUARD SYNDROME The Brown-Séquard syndrome results from hemisection of the cord. It is manifested by ipsilateral loss of motor function, proprioception, and vibratory sensation, and contralateral loss of pain and temperature sensation. The most common cause of this syndrome is penetrating injury. 53,54 It can also be caused by lateral cord compression secondary to disk protrusion, hematomas, spine fractures, infections, spinal cord infarctions, multiple sclerosis, or tumors.  CAUDA EQUINA SYNDROME Cauda equina syndrome is not a true spinal cord syndrome because the cauda equina is composed entirely of lumbar, sacral, and coccygeal nerve roots; therefore, injuries to this region produce peripheral nerve injuries. Symptoms and signs may include bowel and/or bladder dys function, decreased anal sphincter tone, “saddle anesthesia” (sensory deficit over the perineum, buttocks, and inner thighs), variable motor and sensory loss in the lower extremities, decreased lower extremity reflexes, and sciatica. Bowel or bladder incontinence is not a universal finding. Anal sphincter tone can be spared , 55 and the accuracy of a digital rectal exam to detect decreased anal sphincter tone has been questioned. 56 In addition, if the patient presents acutely, the patient’s bladder may not yet be full enough to cause overflow urinary incon tinence. Careful history and physical examination, including identification of saddle anesthesia, 57 are helpful to suggest the diagnosis, but no one symptom or sign has 100% predictive value for this entity .58,59 Therefore, perform an emergent MRI of the lumbosacral spinal cord if clinical suspicion warrants. See the section “Epidural Compression Syndrome” in Chapter 279, “Neck and Back Pain, ” for further discussion of cauda equina syndrome.  NEUROGENIC SHOCK Neurogenic shock is a type of distributive shock that can occur with CNS or spinal cord injury that probably occurs in less than 20% of spinal cord–injured patients. 60,61 Loss of peripheral sympathetic innervation results in extreme vasodilatation secondary to loss of sympathetic arterial tone. This causes blood pooling in the distal circulation with resul tant hypotension. If the T1 through T4 cord levels are compromised, loss of sympathetic innervation to the heart leaves unopposed vagal parasympathetic cardiac innervation. This results in bradycardia or an absence of reflex tachycardia. In general, patients with neurogenic shock are warm, peripherally vasodilated, and hypotensive with a relative bradycardia. Patients tend to tolerate hypotension relatively well, because peripheral oxygen delivery is presumably normal. Loss of sympathetic tone and subsequent inability to redirect blood from the periphery to the core may cause excessive heat loss and hypothermia. The diagnosis of neurogenic shock is one of exclusion. Certain clues, such as bradycardia and warm, dry skin, may be evident, but hypotension in the trauma patient can never be presumed to be caused by neurogenic shock until other possible sources of hypotension are excluded.  SPINAL SHOCK Spinal shock is not neurogenic shock; the two terms have very different meanings and are not interchangeable.

dycardia and warm, dry skin, may be evident, but hypotension in the trauma patient can never be presumed to be caused by neurogenic shock until other possible sources of hypotension are excluded.  SPINAL SHOCK Spinal shock is not neurogenic shock; the two terms have very different meanings and are not interchangeable. Spinal shock is the tempo rary loss or depression of spinal reflex activity that occurs below a complete or incomplete spinal cord injury. The typical presentation involves flaccidity, loss of reflexes, and loss of voluntary movement. The lower the level of the spinal cord injury, the more likely it is that all distal reflexes will be absent. Loss of neurologic function that occurs with spinal shock can cause an incomplete spinal cord injury to mimic a complete cord injury. Therefore, cord lesions cannot be called complete until spinal shock has resolved. The delayed plantar and bulbocavernosus reflexes are among the first to return as spinal shock resolves. 63 The duration of spinal shock is variable; it generally lasts for days to weeks but can persist for up to 6 months.64 DIAGNOSIS Although spinal column and spinal cord injuries can sometimes be diagnosed clinically, diagnostic imaging is necessary to confirm the diagnosis and direct definitive care. However, judicious use of imaging is desirable to avoid unnecessary costs and ionizing radiation exposure to patients. Therefore, the challenge is identifying the appropriate patients to image and selecting the appropriate imaging modality.  CLINICAL DECISION RULES IN CERVICAL SPINE IMAGING In some cases, it is obvious who needs cervical spine imaging. For example, patients with head or neck trauma who are not fully alert (Glasgow Coma Scale score of <15) should undergo imaging of their cervical spine because the frequency of cervical spine injury in association with trau matic brain injury ranges from 1.7% to 8%. 65 However, in less obvious cases, the decision to perform imaging is not quite so clear cut. An unstructured clinical exam is not adequately sensitive for the detection of cervical spine injuries,66 so guidelines can assist clinical judgment in deciding whom to image. In alert, stable, adult trauma patients who have no neurologic deficits (i.e., low-risk trauma patients), two major clinical decision rules have been defined to avoid unnecessary radiography. The first decision rule was derived by the National Emergency X-Radiography Utilization Study (NEXUS) , which determined that plain cervical spine imaging can be safely avoided in patients who have all five clinical criteria (Table 258-4) (See Video: NEXUS Criteria). 47 In the study population of 34,069 patients, the NEXUS criteria were 99.6% sensitive (95% confidence interval [CI], 98.6% to 100%) for detecting prospectively defined clinically significant cervical spine injuries, but only 12.9% specific (95% CI, 12.8% to 13.0%), with a negative predictive value of 99.9% (95% CI, 99.8% to 100%). The original NEXUS trial excluded patients >60 years old, but a post hoc analysis of the NEXUS data set showed the criteria to be 100% sensitive (95% CI, 97.1% to 100%) and 14.7% specific (95%  CI, 14.6% to 14.7%) for clinically significant injuries in 2943 patients ≥65  years  of  age. 67 In a retrospective Australian trial (n = 4035) that investigated NEXUS’s performance in patients ≥65 years of age, NEXUS was 94.8% sensitive (95.5% on a one-way sensitivity analysis) for cervi cal spine injuries detected on a 64-slice multidetector CT. 68 However, this trial contained several sources of potential bias (i.e., single center, undocumented NEXUS data in 14.1% of those with cervical spine frac tures). This study also did not clarify whether the fractures detected on CT were clinically significant or if any intervention was required.

lice multidetector CT. 68 However, this trial contained several sources of potential bias (i.e., single center, undocumented NEXUS data in 14.1% of those with cervical spine frac tures). This study also did not clarify whether the fractures detected on CT were clinically significant or if any intervention was required. Two previous prospective studies have also challenged the use of NEXUS in elders, 69,70 but significant selection biases hamper external validity. Other studies71,72 have suggested using modified NEXUS criteria (more specific definitions of “normal alertness” and “painful distracting injuries”) in elders in an attempt to improve performance characteristics. However, these studies sacrifice a strength of the original NEXUS study: the defi nitions in the original study were based on the individual clinician’s judgment, yet the decision rule performed well in the NEXUS cohort TABLE 258-4 NEXUS Criteria47 •   Absence of midline cervical tenderness •   Absence of focal neurologic deficit •   Normal level of alertness and consciousness* •   No evidence of intoxication •   Absence of painful distracting injury† *For example: Glasgow Coma Scale score <15; disorientation to person, place, time, or events; inability to remember three objects at 5 minutes; delayed or inappropriate response to external stimuli. †”a clinically apparent, painful injury that might distract . . . [the patient] . . . from the pain of a cervicalspine injury.” Tintinalli_Sec21_p1669-1766.indd 1710 8/1/19 12:21 PM

o person, place, time, or events; inability to remember three objects at 5 minutes; delayed or inappropriate response to external stimuli. †”a clinically apparent, painful injury that might distract . . . [the patient] . . . from the pain of a cervicalspine injury.” Tintinalli_Sec21_p1669-1766.indd 1710 8/1/19 12:21 PM CHAPTER 258:  Spine Trauma      1711 of 34,000 patients 47,73 and in the subgroup of elders in this cohort. 67 Therefore, more prospective, externally validated data are needed to determine the best way to use NEXUS in the elderly. The Canadian Cervical Spine Rule for Radiography (CCR) was developed for alert, stable trauma patients to reduce practice variation and inefficiency in the ED use of plain cervical spine radiography. The Canadian rule consists of three assessments, which are asked in sequential order (Table 258-5). 48 To proceed to the next assessment, the answer to the previous assessment must be “Y es. ” If the answer to any assessments is “No, ” then imaging is immediately performed. In the original study sample of 8924 patients, the CCR was 100% sensitive (95% CI, 98% to 100%) and 42.5% specific (95% CI, 40% to 44%) for identifying patients with “clinically important” cervical spine injuries. The CCR has also been validated in both larger hospital-based studies 74 and prehospital studies.75 However, the CCR has been criticized for its complexity relative to NEXUS, 76,77 although this may now be less of an issue with the availability of CCR calculators that can be accessed on a smartphone. There is one published direct prospective comparison of NEXUS and CCR (n = 8283) that reported that CCR was more accurate for detect ing cervical spine injury compared to NEXUS, with superior sensitivity (99% vs. 91%), specificity (45% vs. 37%), positive likelihood ratio (1.81 vs. 1.44), and negative likelihood ratio (0.01 vs. 0.25). 79 However, some have questioned the methodology of this comparison as being biased in favor of CCR. 80,81 A meta-analysis of 15 studies (79,526 patients) concluded that the CCR appeared to have better diagnostic accuracy than NEXUS 82; however, the quality of methods of the included studies was termed “modest, ” and further, more rigorous studies were suggested to be done. In both rules, the more subjective parts (“absence of pain ful distracting injury” and “no evidence of intoxication” for NEXUS; “dangerous mechanism of injury” and assessment of range of motion for CCR) are the most common misinterpretation of the rules, which obvi ously affects their performance. Both NEXUS and CCR were developed in an era prior to the routine use of CT as a primary tool to evaluate the cervical spine in blunt trauma patients. Consequently, studies have been done to compare both decision rules using CT scan as the gold standard. In a 2011 study of 2606 blunt trauma patients, NEXUS was found to only be 82.8% sensitive and 45.7% specific for spine injury. Of the 26 missed injuries, 19 patients required further intervention, including two who went to the operating room and one who needed a halo. 83 The same group compared CCR to CT scan (3201 blunt trauma patients), finding excellent sensitivity of 100% but only 60% specificity. 84 Nevertheless, the use of NEXUS has been recom mended in several national guidelines and trauma societies.85,86 In summary, many experts feel that because both NEXUS and CCR have been widely validated and have demonstrated adequate sensitivity, either rule may be used to determine which low-risk patients should undergo cervical spine imaging .

NEXUS has been recom mended in several national guidelines and trauma societies.85,86 In summary, many experts feel that because both NEXUS and CCR have been widely validated and have demonstrated adequate sensitivity, either rule may be used to determine which low-risk patients should undergo cervical spine imaging .  CERVICAL SPINE IMAGING Plain Radiography In the past, cervical spine radiography was the gold standard for the initial evaluation to detect a cervical spine fracture; however, it has largely been replaced by CT in EDs with access to that modality (see below). Plain radiography for the identification of bony cervical injury includes three views of the cervical spine: lateral, anteriorposterior, and odontoid. A single lateral cervical spine film will identify only about 90% of injuries to bone and ligaments. 31 The anterior-posterior and open-mouth odontoid views will identify many of the remaining abnormalities. It is important to image all seven cervical vertebrae, along with the superior border of the first thoracic vertebra , given the propensity for injuries at the cervical-thoracic junction. Therefore, a “swimmer’s view” may be necessary to visualize this junction clearly, but this often requires an assistant to pull down the shoulders during the radiograph. The main advantages of plain radiography are that it can be quickly done at the bedside and exposes the patient to only small amounts of ionizing radiation. The main disadvantage of plain radiography is that it misses injuries when compared with CT 87 and is especially poor for imaging C1 and C2. In addition, visualization of the entire cervical spine by plain films is often problematic in obese, elderly, or extremely muscular patients, especially with a cervical collar in place. Cervical Spine CT CT is the initial imaging modality of choice to evaluate the traumatized cervical spine. Multidetector CT is more sensitive and specific than plain radiography for evaluating the cervical spine in trauma patients and can be performed quickly. 88,89 CT can be used to visualize the entire cervical spine and is particularly useful at the craniocervical and cervicothoracic regions, where the sensitivity of plain films is most limited. In addition, a 3-year retrospective review found that plain radiography did not add any clinically useful informa tion to a cervical spine CT. 90 Furthermore, cost analyses have shown CT to be cost-effective to screen for cervical spine injuries in moderate- to high-risk trauma patients. 91,92 Accordingly, the American Association of Neurological Surgery and the Congress of Neurological Surgeons, 86 the Eastern Association for the Surgery of Trauma (EAST), 85 and the National Institute for Health and Care Excellence (NICE) 93 all recom mend CT as the primary initial diagnostic tool for suspected cervical spine injury. In addition, if plain radiography is chosen as the primary imaging modality, a CT should be ordered if the initial plain radiograph is inadequate or if an injury is detected or still suspected after the initial radiography. Despite this general consensus, some have opined that plain radiography may be preferable to CT for patients who are at “low to moderate risk” 87; however, there are no controlled studies that have defined “moderate risk” and no published data to suggest it is acceptable to miss injuries that would have been detected on CT scanning. Cervical Spine MRI MRI is the imaging modality of choice if a liga mentous or spinal cord injury is strongly suspected because MRI has excellent sensitivity for soft tissue and spinal cord injuries.

te risk” and no published data to suggest it is acceptable to miss injuries that would have been detected on CT scanning. Cervical Spine MRI MRI is the imaging modality of choice if a liga mentous or spinal cord injury is strongly suspected because MRI has excellent sensitivity for soft tissue and spinal cord injuries. 94,95 However, there are practical limitations on its use for emergent presentations, including the requirement for the patient to be stable, availability, cost, and patient tolerance for the procedure. A prospective, multicenter, observational study (n = 10,276) found that CT was 98.5% sensitive (negative predictive value 99.97%) for ruling out clinically significant spinal column injuries in patients who failed the NEXUS criteria. 96 The three patients (0.03%) who had clinically significant injuries missed by CT all had focal neurologic symptoms (which were consistent with central cord syndrome in all three cases, and the diagnoses were subsequently confirmed on MRI). These results are largely consistent with previous smaller studies. Therefore, MRI is appropriately used for symptomatic patients with a negative CT who have persistent neurologic deficits that could be attributed to a spinal lesion and for patients with a positive CT in order to evaluate the spinal cord. Imaging for Cervical Ligamentous Injury In patients with pure ligamentous injuries, the ligaments are disrupted, but the spine spon taneously reduces to a normal position. The resulting instability risks subsequent neurologic injury if the spine moves. Signs and symptoms TABLE 258-5 Canadian Cervical Spine Rule for Radiography: Cervical Spine Imaging Unnecessary in Patients Meeting These Three Criteria 48 Assessment Definitions Assessment #1: There are no high-risk factors that mandate radiography. High-risk factors include: Age 65 years or older A dangerous mechanism of injury* The presence of paresthesias in the extremities Assessment #2: There are low-risk factors that allow a safe assessment of range of motion. Low-risk factors include: Simple rear-end motor vehicle crashes Patient able to sit up in the ED Patient ambulatory at any time Delayed onset of neck pain Absence of midline cervical tenderness Assessment #3: The patient is able to actively rotate his/her neck (regardless of pain). Can rotate neck 45 degrees to the left and to the right *Defined  as fall from a height of >3 feet; an axial loading  injury;  high-speed  motor vehicle  crash, rollover,  or ejection;  motorized  recreational  vehicle  or bicycle  collision. Tintinalli_Sec21_p1669-1766.indd 1711 8/1/19 12:21 PM

k (regardless of pain). Can rotate neck 45 degrees to the left and to the right *Defined  as fall from a height of >3 feet; an axial loading  injury;  high-speed  motor vehicle  crash, rollover,  or ejection;  motorized  recreational  vehicle  or bicycle  collision. Tintinalli_Sec21_p1669-1766.indd 1711 8/1/19 12:21 PM 1712 SECTION 21: Trauma TABLE 258-6 Eastern Association for the Surgery of Trauma Guidelines for Thoracic and Lumbar Imaging after Trauma49 Level I (convincingly justifiable based on scientific evidence) When imaging is deemed necessary, CT scans with axial collimation should be used to screen for and diagnose injury, because CT scans are superior to plain films in identifying thoracolumbar spine fractures. Level II (reasonably justifiable based on scientific evidence and expert opinion) Patients with back pain, thoracolumbar spine tenderness on examination, neurologic deficits referable to the thoracolumbar spine, altered mental status, intoxication, distracting injuries, or known or suspected high-energy mechanisms should be screened for thoracolumbar spine injury with CT scan. In blunt trauma patients with a known or suspected injury to the cervical spine, or any other region of the spine, thorough evaluation of the entire spine by CT scan should be strongly considered due to a high incidence of spinal injury at multiple levels within this population. Patients without complaints of thoracolumbar spine pain who have normal mental status, as well as normal neurologic and physical examinations, may be excluded from thoracolumbar spine injury by clinical examination alone, without radiographic imaging, provided that there is no suspicion of high-energy mechanism or intoxication with alcohol or drugs. Level III (supported by available data, but scientific evidence lacking) MRI should be considered in consultation with the spine service for CT findings suggestive of neurologic involvement and of gross neurologic deficits. Abbreviation: CT = multidetector CT. TABLE 258-7 American Association for the Surgery of Trauma Thoracolumbar Multicenter Study Group (AAST TL-Spine) Clinical Decision Rule for Thoracic and Lumbar Imaging After Blunt Trauma *50 Imaging of the thoracic and lumbar spine is indicated in a multisystem blunt trauma patient when any of the following criteria are met: •   Patient is ≥60 years old or •   Patient is not alert and evaluable† or •   Patient has pain, tenderness to palpation, deformity, or neurologic deficit or •   Mechanism of injury was “high risk” (fall [not including ground-level fall or fall from ≤5 steps], crush injury, motor vehicle collision with rollover or ejection, unenclosed vehicle crash, or automobile versus pedestrian) *Rule may not be applied to patients who are <15 years old, have preexisting paraplegia/tetraplegia, have a concurrent cervical spine injury causing neurologic deficit, or who present >24 hours after injury. †Evaluable = Glasgow Coma Scale score of 15, nonclinically intoxicated, and without painful, distracting injury. include persistent neck pain/midline tenderness, extremity paresthesias, or focal neurologic findings despite normal plain radiographs and/or CT. Although flexion and extension radiographs have been traditionally used to try to detect ligamentous instability, numerous studies have demonstrated their lack of sensitivity and inefficiency (30% to 80% of flexion and extension radiographs are inadequate), and they provide no further information beyond a CT. 97-101 Therefore, flexion and extension radio graphs should not be ordered when more advanced imaging is available. MRI is the test of choice for these patients and should be obtained if available.

ncy (30% to 80% of flexion and extension radiographs are inadequate), and they provide no further information beyond a CT. 97-101 Therefore, flexion and extension radio graphs should not be ordered when more advanced imaging is available. MRI is the test of choice for these patients and should be obtained if available. If MRI is not available and the patient’s only symptom is neck pain, there is a paucity of published data to direct next steps, and local expert opinion may vary. 96 Therefore, in these patients, consult a spine surgeon to help make the decision between spine immobilization until MRI is available versus emergent transfer for MRI. Imaging for Suspected Carotid/Vertebral Dissection Cervical spine fractures and head injury are risk factors for carotid or vertebral artery dissection. Many patients are initially asymptomatic and diagnosis can be delayed for days until neurologic symptoms become evident. See Chapter 260, “Neck Injuries” , and Table 260-7 for further discussion.  CLINICAL DECISION RULES IN THORACIC AND LUMBAR SPINE IMAGING It has been well-established in the literature that clinical examination is insufficiently sensitive to rule out thoracolumbar injuries. A prospective observational study (n = 884) found that a clinical examination was only 48.2% sensitive for thoracolumbar spine fractures (78.6% sensitive for clinically significant fractures), with 52% of thoracolumbar spine frac tures having a negative clinical examination. 102 A prospective study of patients with distracting injuries (n = 950) found the clinical examina tion was only 75% and 89% sensitive for all thoracolumbar fractures and clinically significant fractures, respectively. 103 Since the clinical exams in both of these studies were standardized and structured, it seems more likely than not that these sensitivities would be even lower in actual practice with unstructured exams. However, in contrast to the cervical spine, there are no commonly accepted, prospectively validated clinical decision rules as to which patients require imaging. This led the East ern Association for the Surgery of Trauma to release general guidelines based on expert opinion ( Table 258-6). 49 A prospective, multicenter study (n = 3065) derived a decision rule that was 98.9% sensitive (nega tive predictive value 99.6%) for clinically significant injuries and 100% sensitive (negative predictive value 100%) for those requiring surgery (Table 258-7). 50 To date, a prospective validation of this rule has yet to be published; however, in light of these data, it seems reasonable to be aggressive in seeking thoracolumbar injuries and to not rely solely on clinical examination to defer imaging.  THORACIC AND LUMBAR SPINE IMAGING As with the cervical spine, CT is more sensitive for injuries than plain radiography in screening for thoracolumbar injuries in patients with significant trauma (97.2% and 33.3%, respectively, for unstable fractures), 104 and plain radiography does not add any important diag nostic information when CT is used. 105 CT can reveal the anatomy of an osseous injury, grade the extent of spinal canal impingement by bone fragments, and assess the stability of an injury. Rather than obtaining dedicated CT images, the thoracic and abdominal CT scans typically obtained to evaluate the multiple trauma patient can be used to recon struct images of the thoracic and lumbar spine, thereby sparing the patient additional ionizing radiation exposure while detecting spinal column injuries not seen on the unreconstructed scans alone. 106 As in the case of cervical spine injuries, if an associated spinal cord or nerve root injury is suspected, MRI is the imaging study of choice.

acic and lumbar spine, thereby sparing the patient additional ionizing radiation exposure while detecting spinal column injuries not seen on the unreconstructed scans alone. 106 As in the case of cervical spine injuries, if an associated spinal cord or nerve root injury is suspected, MRI is the imaging study of choice. It is less clear how to screen for thoracolumbar injuries in patients who have less severe mechanisms of injuries. Although it has been well established that CT is more sensitive for thoracolumbar injuries in severely injured patients, 49 there has been no prospective controlled comparison between plain radiography and CT in more mildly injured patients. Nevertheless, CT should still be considered the standard screening modality for thoracolumbar injuries. Imaging for Thoracolumbar Spinal Cord and Ligamentous Injuries As with cervical imaging, MRI is the diagnostic test of choice for describing the anatomy of ligamentous or neural tissue injury in the thoracolumbar region. MRI is indicated in patients with neurologic findings with no clear explanation after plain films and/or CT scanning. If the patient is stable and MRI is unavailable, transfer to a facility with MRI capabilities is appropriate.  CONCURRENT IMAGING OF THE ENTIRE SPINE The determination of a spinal column injury at one level should prompt imaging of the entire remainder of the spine with CT because approxi mately 20% of patients with a spine fracture in one segment will have a noncontiguous second fracture at another segment. 107-109  SPINE IMAGING IN OBTUNDED PATIENTS While experts recommend that all obtunded patients with significant blunt trauma should have their entire spine imaged, consensus does not Tintinalli_Sec21_p1669-1766.indd 1712 8/1/19 12:21 PM

cture in one segment will have a noncontiguous second fracture at another segment. 107-109  SPINE IMAGING IN OBTUNDED PATIENTS While experts recommend that all obtunded patients with significant blunt trauma should have their entire spine imaged, consensus does not Tintinalli_Sec21_p1669-1766.indd 1712 8/1/19 12:21 PM CHAPTER 258:  Spine Trauma      1713 yet exist on what imaging is necessary to clear the spine in obtunded patients. Specifically, it remains controversial whether a negative CT of the spine is adequate or if a subsequent MRI needs to be done .96,110 The literature on this topic consists largely of small, retrospective, single-center studies. Consequently, in 2015, two systematic reviews of the literature could only “probably” 111 and “conditionally”112 conclude that a “high-quality” CT scan could be used to clear the cervical spine in obtunded patients. However, a more recent prospective multicenter trial (n = 10,276) found an extremely high sensitivity of CT for cervical spine injuries (98.5%), yet three patients had central cord syndrome that would have been missed by CT alone. 96 Therefore, given the absence of definitive data at this time, maintain spinal precautions in the obtunded trauma patient in the ED, and defer any subsequent spine clearance to local expert consultants. TREATMENT AND DISPOSITION OF SPINAL COLUMN INJURIES The goals of treatment are to prevent secondary injury, alleviate cord compression, and establish spinal stability. Maintain spinal immobilization and keep movement to a minimum. Obtain emergent consultation with a spine surgeon (neurosurgeon or orthopedic surgeon depending on the particular facility) on all spinal column fractures or liga mentous injuries, regardless of neurologic compromise.  CERVICAL SPINE FRACTURES The majority of cervical spinal fractures will require admission for definitive treatment or for the care of associated injuries. Until transfer of care to a surgeon, spine precautions should be maintained, associated injuries stabilized, and the patient carefully monitored for respiratory or neurologic deterioration.  THORACIC AND LUMBAR SPINE FRACTURES Thoracolumbar fractures are also high risk for associated spinal cord or other traumatic injuries, such as aortic, intrathoracic, or intra-abdominal visceral injuries. 113 Although many of these injuries will require admission, there are two types of thoracolumbar fractures that may be amenable to outpatient therapy. Compression fractures, also known as “wedge” or “anterior” com pression fractures, have been reported to compose approximately 33.6% 113 to 52% 10 of thoracolumbar fractures. These fractures occur as a result of a hyperflexion during an axial load that crushes the anterior portion of the vertebra. If the percentage of loss of vertebral height is <40%, the patient may be a candidate for outpatient therapy, and this should be discussed on a case-by-case basis by the spine surgeon. How ever, if the loss of vertebral height is ≥50% or if the angle between the damaged vertebra and the rest of the spinal column is >25% to 30%, the compression fracture is generally considered unstable. In addition, make certain that an apparent compression fracture seen on plain radiographs is not a burst fracture, which is a compressiontype fracture that involves the posterior half of the vertebrae and is reported to be 39.5% of thoracolumbar fractures. 113 Burst fractures may result in retropulsed fragments that can impinge on the spinal canal and cause neurologic injury. In two studies, the incidence of misdiagnosis of burst fractures on plain radiographs ranged from 20% to 23%. 114,115 Another fracture that is sometimes misdiagnosed as a wedge compres sion fracture on plain radiograph is the Chance fracture.

d fragments that can impinge on the spinal canal and cause neurologic injury. In two studies, the incidence of misdiagnosis of burst fractures on plain radiographs ranged from 20% to 23%. 114,115 Another fracture that is sometimes misdiagnosed as a wedge compres sion fracture on plain radiograph is the Chance fracture. This fracture occurs via a flexion-distraction mechanism and involves minor anterior vertebral compression and significant distraction of the middle and posterior ligamentous structures. Typical radiographic findings reveal a transverse fracture lucency in the vertebral body, an increased height of the posterior vertebral body, fracture of the posterior wall of the vertebral body, and posterior opening of the disk space. Finally, minor to moderate trauma can cause pathologic fractures secondary to pre existing neoplastic, infectious, or osteoporotic processes in the spine. Because the above-mentioned fractures can be easily misdiagnosed with plain radiography alone, further evaluate compression fractures of the thoracolumbar spine on plain radiographs with CT . If, after a thorough evaluation, a stable wedge compression fracture with no neurologic compromise is diagnosed, the patient may be treated as an outpatient with analgesia, heat, massage, rest, and appropriate follow-up for consideration of physical therapy.  SACRUM AND COCCYX FRACTURES Injuries of the sacral spine and nerve roots are very unusual. When they occur, they are frequently associated with fractures of the pelvis. In general, transverse fractures through the body are most significant in that they cause injury to part or all of the cauda equina. Longitudinal fractures may cause radiculopathy. Sacral fractures that involve the central sacral canal can produce bowel or bladder dysfunction . For these reasons, obtain emergent consultation for sacral fractures. One notable exception to the need for emergent consultation is an isolated coccyx fracture. Coccygeal injuries are usually associated with a direct fall onto the buttocks, with resultant coccyx pain exacerbated by sitting or straining. Imaging is not needed to diagnose coccygeal fractures. Treatment is symptomatic with analgesics and use of a rubber doughnut pillow. SPECIAL CONSIDERATIONS  CORTICOSTEROIDS High-dose methylprednisolone remains a controversial treatment in acute blunt spinal cord injury and should not be given routinely. The major neuroprotective mechanism by which high-dose methylprednisolone is believed to work is in its inhibition of free radical–induced lipid peroxidation. Other proposed beneficial actions include its ability to increase levels of spinal cord blood flow, increase extracellular calcium, and prevent loss of potassium from injured cord tissue. Methylprednisolone is advocated in preference to other steroids because it crosses cell membranes more rapidly and completely. In the 1990s, the National Acute Spinal Cord Injury Study (NASCIS) group published three prospective, double-blind studies to evaluate the efficacy of methylprednisolone in blunt spinal cord injury: NASCIS I, II, and III. 117-119 NASCIS I compared high-dose methylprednisolone and a lower-dose methylprednisolone regimen (n = 330). NASCIS I showed no evidence in recovery of function between the groups. NASCIS II compared a higher dose of methylprednisolone (Table 258-8), naloxone, and placebo (n = 427). This trial was also negative, but based on post hoc subgroup analysis, NASCIS II showed modest improvements in motor function when steroids were administered within 8  hours of injury. NASCIS III compared high-dose methylprednisolone for 24 hours, high-dose methylprednisolone for 48 hours, and tirilazad mesylate for 24 hours (n = 499).

so negative, but based on post hoc subgroup analysis, NASCIS II showed modest improvements in motor function when steroids were administered within 8  hours of injury. NASCIS III compared high-dose methylprednisolone for 24 hours, high-dose methylprednisolone for 48 hours, and tirilazad mesylate for 24 hours (n = 499). NASCIS III was also a negative trial, but post hoc analysis found that patients who received the 48-hour methylprednisolone regimen within 3 to 8 hours of their injury showed motor improvement. In all three trials, patients who received high-dose methylprednisolone and longer duration protocols were more likely to develop complications such as severe sepsis, severe pneumonia, wound infection and delayed healing, pulmonary embolism and deep vein thrombosis, GI bleeding, and death. In 2012, a Cochrane systematic review (written by the lead author of the NASCIS trials) essentially confirmed the conclusions of NASCIS II and III that there was a small but statistically significant benefit to high-dose methylprednisolone when administered within 8 hours of injury, but these patients were also more TABLE 258-8 The National Acute Spinal Cord Injury Study II High-Dose Methylprednisolone Protocol Indications Blunt trauma Neurologic deficit referable to the spinal cord Treatment must be started within 8 h of injury Treatment Methylprednisolone, 30 milligrams/kg IV bolus over 15 min Followed by a 45-min pause Methylprednisolone, 5.4 milligrams/kg/h IV is then infused for 23 h Tintinalli_Sec21_p1669-1766.indd 1713 8/1/19 12:21 PM