Maneesh Bawa and Reginald Fayssoux
Injuries to the spine are common in trauma patients. The morbidity associated with these injuries can be significant and life-changing, and some injuries can be life-threatening. Because polytrauma patients are initially seen in urgent circumstances, many fractures may be overlooked. In addition, the diversity of injury patterns and the potential for neurologic compromise make the evaluation and treatment of spinal trauma complex. Failure to recognize these injuries or to properly manage known injuries can have catastrophic consequences. In this chapter, the epidemiology of injuries to the spinal column, the anatomy, biomechanics, and physiology of the spine and spinal cord, the acute management and evaluation of trauma patients with suspected spinal injury, and the management of specific injuries and situations will be reviewed.
To date, the bulk of the literature on the epidemiology of spinal injury in North America has focused on patients sustaining spinal cord injury (SCI), while the epidemiology concerning patients with spinal column injuries without SCI has been less studied. There is currently only one population-based study that has been conducted on spinal column injuries.1 This 1996 study by Hu et al.1 reviewed spinal injuries within the Canadian province of Manitoba in the early 1980s. More recent studies that have attempted to define the epidemiology of spinal injury have relied on the review of patients with blunt trauma presenting to emergency rooms.2,3 While these are perhaps a less accurate reflection of the true incidence, they provide a useful estimation of the scope of the problem.
In the United States, the incidence of spinal fractures has been estimated to be greater than 50,000 injuries per year. Hu et al.1 reported an annual incidence rate of spinal fractures to be 64 per 100,000 in Manitoba. In the United States, with a population of just over 300 million, this would translate to over 192,000 injuries per year. In patients with blunt trauma, the reported incidence of spinal injury is between 3% and 4% in the cervical spine and approximately 6% in the thoracolumbar spine.2,3 Demographically, young men and elderly women are most commonly involved. Although the incidence of females sustaining spinal injury has increased in recent years, males continue to account for the majority of all patients with injury to the spine (52–70%). The most common mechanism of injury is motor vehicle crashes, followed by falls, acts of violence (gunshots, stab wounds), and sports. In certain urban regions, assaults and gunshot wounds (GSWs) may surpass falls as the principal mechanism of spinal injury (Fig. 23-1).
FIGURE 23-1 Causes of SCI since 2005. (Reproduced with permission from National Spinal Injury Statistical Center (NSCISC).)
Data regarding the epidemiology of SCI are much more robust as a result of the vast amount of time, effort, and research that has been undertaken to improve outcomes with these devastating injuries. The National Spinal Cord Injury Statistical Center (NSCISC), established at the University of Alabama in Birmingham, supervises and directs the collection, management, and analysis of data from a network of 16 federally sponsored regional centers for SCI throughout the United States. Over the past 30 years in North America, the incidence of SCI has remained relatively stable and is currently estimated to be approximately 40 cases per million population (excluding lethal cases).4,5 In the United States, this translates to over 12,000 new disabled patients each year. Currently, approximately 260,000 Americans live with an SCI. The average age at injury is 40.2 years and has been increasing over the past three decades as a result of the increasing proportion of elderly individuals affected. The mode age (i.e., the most common age at injury), however, has remained relatively consistent at 19 years. Males account for 80% of patients with SCI, blacks are at higher risk than whites, and the percentage of cases occurring among blacks has been increasing in recent years. Median hospitalization days in the acute setting has declined from 24 days from 1973 to 1979 to 12 days from 2005 to 2009.5
The etiologies of SCIs are not significantly different from the etiologies of injuries to the spinal column. Motor vehicle crashes are the most common cause of SCIs, comprising approximately 41% of all injuries with rollovers accounting for 70% of these.5 Ejections occurred in 39% of those injured, and only 25% reported using seatbelts. The next most common etiology is falls (27%), followed by violence (15%), sports/recreation (8%), and other causes (9%) (Fig. 23-1).
A complex interplay of social issues and advances in regulatory oversight have influenced trends in spinal injury. Improvements in emergency medical services systems, the development of safer automobiles, legislation requiring safety measures such as seatbelts, more occupational safety standards, and better regulation of contact sports have resulted in more individuals with SCI surviving the prehospitalization phase of injury and having better outcomes in survivors. As evidence of this, 38% of individuals with SCIs in 1970 died before hospitalization. In 2000, this figure had decreased to 15.8%.6
The proportion of injuries that are due to violent acts has varied over time as well, reflecting variation in national crime rates. Violent acts caused 13% of SCIs prior to 1980 and then peaked at 25% from 1990 to 1999 before declining to only 15% from 2005 to 2009.4,5 In Canada, violent acts are less common and caused only approximately 4% of all cases of SCI between 1997 and 2001.6
Falls are responsible for an increasing proportion of injuries due to the aging of the population and continue to be a major public health concern.7 Prevention and appropriate medical treatment of osteoporosis may help mitigate this trend. The assessment of comorbidities is integral to outcome following traumatic injury to the spine in these patients.
The implementation of injury prevention programs developed through an understanding of injury mechanisms in national injury tracking registries and other observational studies has caused a decrease in sport-related SCIs. Previously, diving accounted for a large majority of sport-related SCIs, but the incidence of dive-related injuries has steadily decreased as a result of prevention programs. The majority of these now occur in unsupervised recreational settings.4 American football still accounts for a significant number of catastrophic spinal injuries in the United States, but the incidence of these has also decreased as a result of rule changes banning spear tackling (tackling maneuver using the crown of the head to “spear” an opposing player).8 Extreme sports such as snowboarding and mountain biking now account for an increasing percentage of SCIs.
Incomplete quadriplegia is the most frequent neurologic category of SCI (38%), followed by complete paraplegia (23%), incomplete paraplegia (22%), and complete quadriplegia (17%). Over the past 15 years, the proportion of persons with incomplete paraplegia has increased with a concomitant decrease in the proportion of persons with complete paraplegia and quadriplegia. Improved survival after occipitocervical and upper cervical injuries as a result of improvements in EMS and direct medical care has likely contributed to an increase in ventilator-dependent discharges in the last 30 years (2.3–6.8%).4
The overall impact of an SCI on the individual patient, family, and society remains staggering. Few conditions aside from SCI result so abruptly in such a degree of permanent disability. The young and highly functional individuals that SCI so typically affects (recall the mode age of 19 years) face the severe challenge of reintegrating into society after injury. The patient has suffered a devastating transformation in quality of life and loss of independence, and the injury will have a profound impact on his or her lifestyle, personal goals, economic security, and interpersonal relationships. Data from the NSCISC have shown that only one third of persons with paraplegia and about one fourth of those with quadriplegia were employed at postinjury year 8.5 Among those who were married at the time of injury, as well as those who marry after injury, the likelihood of their marriage remaining intact is much lower when compared to the general population.5 Also, less than 5% of patients with SCI will marry following their injury.
In addition, the economic burden of all persons living with SCIs in the United States has been estimated to approach $10 billion per year.5 The lifetime costs for health care and living expenses vary depending on severity of injury and age at the time of injury. Estimates for these lifetime costs (in 2009 dollars) range from $750,000 for a 50-year-old patient with paraplegia to $3.3 million for a 25-year-old individual with high quadriplegia (C1–C4). Note that these are the direct costs and do not account for the indirect costs associated with lost wages and productivity.
ANATOMY AND BIOMECHANICS OF THE SPINAL COLUMN
The spine functions to allow spinal motion while protecting the enclosed neural elements from injury. The spinal column is composed of 7 cervical, 12 thoracic, 5 lumbar, 5 sacral, and 4–5 coccygeal vertebral segments and forms the foundation of the axial skeleton of the body, extending from the base of the skull to the pelvis with articulations to the rib cage. Each vertebra has an opening that contributes to the anatomy of the spinal central canal as well as multiple processes that serve as lever arms for the ligamentous and muscular attachments (e.g., spinous process, transverse process). The normal spine in the uninjured state positions the head directly over the pelvis in the coronal and sagittal planes (i.e., coronal and sagittal balance). In the coronal plane the spine is straight, while, in the sagittal plane, the cervical and lumbar spines are lordotic and the thoracic spine is kyphotic. These sagittal curvatures in conjunction with the intervertebral discs provide resiliency to applied loads.
The unique, highly specialized anatomy of the upper cervical spine allows weight transfer between the head and neck, facilitates neck motion, and protects the neurovascular elements from injury (Fig. 23-2). The occiput articulates with the atlas through paired synovial joints formed between the convex occipital condyles located at the lateral margins of the foramen magnum and the concave facets of the atlantal lateral masses. The occipitoatlantal articulation is responsible for 50% of the normal flexion–extension arc. The atlas itself is composed of two lateral masses connected by anterior and posterior arches that serve as attachment points for controlling and stabilizing ligamentous and muscular insertions. Grooves on the superior surface of the posterior arch accept the paired vertebral arteries after they pass through the paired transverse foramina. The odontoid process, or dens, extends rostrally from the body of the axis to articulate with the posterior aspect of the anterior arch of the atlas. Embryologically, the odontoid is the former centrum of the atlas that separates during gestation and fuses with the centrum of the axis. These three atlantoaxial synovial articulations allow rotational motion to occur and are responsible for 50% of the normal rotational range of motion (ROM). The spinous process of the axis is generally large and bifid due to the multiple insertions it accommodates.
FIGURE 23-2 (A) Midsagittal section of the upper cervical spine. Note the tectorial membrane as the cranial continuation of the posterior longitudinal ligament (PLL). (B) Posterior view of the cruciate ligament composed of a transverse atlantal ligament (TR) with superior and inferior longitudinal bands. The strong transverse atlantal ligament (TR) is important for preventing atlantoaxial subluxation. Note the apical (AP) and alar (AL) ligaments just anterior to the cruciate ligament. (C) Anterior view of the apical and alar ligamentous attachments. (Reproduced with permission from Jackson RS, Banit DM, Rhyne AL, Darden B V. Upper cervical spine injuries. J Am Acad Orthop Surg. 2002;10:271. Copyright 2002 by the American Academy of Orthopaedic Surgeons.)
The motion afforded to the upper cervical spine by this complex osseous anatomy requires stabilizing ligamentous restraints to prevent damage to the enclosed neural elements. This reliance on ligamentous structures for stability is important, and proper evaluation of the stability of the upper cervical spine requires an assessment of the integrity of these ligamentous structures (i.e., dynamic radiography, magnetic resonance imaging, etc.).
Multiple ligaments stabilize the upper cervical articulations, and these can be divided into intrinsic and extrinsic ligaments. The intrinsic ligaments, located within the spinal canal, are the most important contributors to stability of the upper cervical articulations. They form three layers anterior to the dura and include, from dorsal to ventral, the tectorial membrane, the cruciate ligament, and the odontoid ligaments. The tectorial membrane is the cranial continuation of the posterior longitudinal ligament connecting the posterior body of the axis with the anterior margin of the foramen magnum. The cruciate ligaments lie just ventral to the tectorial membrane where they stabilize the odontoid articulation with the anterior arch of the atlas. The transverse atlantal ligament (TAL) is the strongest component of the cruciate ligament. Injury to it can result in instability of the atlantoaxial articulation. The odontoid ligaments are the furthest ventral and include the apical and paired alar ligaments. The smaller and less structurally important apical ligament connects the tip of the odontoid process with the anterior margin of the foramen magnum. The much stronger paired alar ligaments connect the odontoid to the occipital condyles. Extrinsic stability is provided by the ligamentum nuchae, which extends from the external occipital protuberance to the posterior arch of the atlas and the tips of the cervical spinous processes as well as the paired occipitoatlantal and atlantoaxial joint capsules.
In the upper cervical spine, flexion is limited by the bony anatomy, while extension is limited by the tectorial membrane. Rotation and lateral bending are restricted by the contralateral alar ligaments. The cruciate ligaments restrict potentially dangerous anterior translation during flexion, while still allowing torsion around the dens. Distraction >2 mm is prevented by the tectorial membrane and alar ligaments. Translation is limited by the facet joints when the tectorial membrane and alar ligaments are intact. The apical ligament has a negligible effect on restricting motion between the occiput and C2.
The remaining vertebrae of the subaxial cervical spine (C3–C7) more closely resemble the vertebrae of the thoracic and lumbar spines (Fig. 23-3). Vertebral bodies are separated by intervertebral disks. The posterior elements are composed of paired pedicles, lateral masses, facet joints, laminae, transverse processes, and a single spinous process. The transverse process contains the transverse foramen, through which the vertebral artery (the first major branch of the subclavian artery) passes. The motion segments are stabilized by three structures as follows: (1) the anterior longitudinal ligament running on the ventral aspect of the vertebral bodies from the foramen magnum to the sacrum; (2) the posterior longitudinal ligament running on the dorsal aspect of the vertebral bodies from the foramen magnum to the sacrum (its cranial extent between C2 and the occiput is referred to as the tectorial membrane); (3) and the posterior ligamentous complex (PLC). The anatomical structures of the PLC include the supraspinous ligament, interspinous ligament, ligamentum flavum, and facet joint capsules (Fig. 23-4). The PLC plays a critical role in protecting the spine and spinal cord against excessive flexion, rotation, translation, and distraction. Some have likened it to a posterior tension band that restricts excessive motion. Once disrupted, the ligamentous structures demonstrate poor healing and the need for adjunctive surgical stabilization of the involved vertebrae to prevent progressive kyphotic collapse. The PLC plays an important role in spinal stability in the thoracic and lumbar spines, as well.
FIGURE 23-3 Cervical vertebrae. Front, back, side, and axial views.
FIGURE 23-4 Midsagittal section of the lumbar spine detailing the components of the posterior ligamentous complex (PLC). This important posterior tension band is composed of the facet capsules, ligamentum flavum, and the interspinous and supraspinous ligaments.
Thoracic vertebrae are similar in structure to the cervical, but they are larger, lack transverse foramina, and have larger transverse and spinous processes. There is much more inherent stability of the thoracic spine compared with the cervical and lumbar regions due to the stabilizing effects of the rib cage and sternum. The thoracolumbar junction is a transition zone between the relatively rigid thoracic spine and the more flexible lumbar segments.
The lumbar vertebrae are the most stout as they must carry more body weight than their cervical and thoracic counterparts (Fig. 23-5). An important stabilizing ligament is the iliolumbar ligament, between the L5 transverse process and the ilium, which can be injured with spinal or pelvic trauma. The sacrum forms the base of the spinal column and also functions as the keystone of the pelvic ring. There are no true intervertebral disks within the sacrum though occasionally a rudimentary disk may be noted in the presence of transitional lumbosacral anomalies. These anomalies are important to recognize because they affect the numbering of vertebrae. In these situations, numbering from the sacrum up may not match up with the numbering when counting down from the last thoracic rib. This can result in confusion between caregivers and has been shown to contribute to wrong-level surgery. The sacral nerve roots lie within intraosseous sacral foramina. The S1 and S2 nerve roots are larger and take up more room within their foramina in contrast to the S3 and S4 nerve roots and are more susceptible to injury.
FIGURE 23-5 Lumbar vertebrae. Front, back, side, and axial views.
ANATOMY AND PHYSIOLOGY OF THE SPINAL CORD
The spinal cord represents the caudal continuation of the brain and brainstem, extending from the brainstem at the level of the foramen magnum through the spinal canal to T12–L1 where it terminates as the conus medullaris. A collection of lumbosacral nerve roots continues from the conus medullaris forming the cauda equina. At each intervertebral space, the ventral and dorsal roots join to form a nerve root that exits the spinal canal through the neural foramen. The central nervous system is invested by three layers of meninges from superficial to deep including the dura, arachnoid, and pia mater. The spinal cord and intraspinous portions of the nerve roots are contained within dura mater, the thickest of the meningeal layers. Between the arachnoid and the pia mater lies the subarachnoid space. Cerebrospinal fluid within the subarachnoid space surrounds the spinal cord providing a mechanical buffer to injury (i.e., shock absorber) and also allows for homeostatic regulation of the distribution of neuroendocrine factors.
The neural elements within the spinal cord itself are arranged geographically. The long neural tracts extending to and from the brain are arranged peripherally and are composed primarily of white matter. The more central gray matter contains the cell bodies of the lower motor neurons.
The main descending motor pathway is the lateral corticospinal tract. The upper motor neuron originates in the contralateral cerebral cortex, decussates in the midbrain, and descends on the ipsilateral periphery of the spinal cord. The upper motor neuron then synapses with its corresponding lower motor neurons in the anterior horn of the gray matter. The lateral corticospinal tract has traditionally been thought to be arranged with the tracts subserving function of the upper extremity more centrally located and the tracts subserving function of the lower extremity and sacral roots more peripherally located. This has been proposed as the reason for the disproportionately greater motor impairment in upper compared to lower extremities in patients with central cord syndrome; however, whether this lamination truly exists is controversial.
The major ascending sensory pathways include the posterior column tracts (fasciculus gracilis, fasciculus cuneatus) and the more ventrally located lateral spinothalamic tracts. Sensory input from the periphery synapses at the neuronal cell bodies located in the dorsal root ganglion and then enters the posterior horn of the gray matter. Pain and temperature input cross immediately to the opposite side of the spinal cord and ascend in the contralateral lateral spinothalamic tract. Proprioception and vibratory sensation ascend ipsilaterally in the posterior column of the spinal cord and decussate at the level of the brainstem. Similar to the lateral corticospinal tract, the dorsal columns are arranged such that tracts subserving function of the upper extremity are more centrally located and tracts subserving function of the lower extremity and sacral roots are more peripheral.
The major vessels (i.e., aorta and vena cava) lie anterior to the thoracic and lumbar vertebral bodies. Segmental branches arising from the aorta and iliac arteries course around the lateral edge of the vertebral bodies where they enter the vertebral foramina to form the anterior and paired posterior spinal arteries, the main blood supply to the spinal cord (Fig. 23-6).
FIGURE 23-6 Blood supply of the spinal cord. (Reproduced, with permission, from Prasad P, Price RS, Kranick SM, Woo JH, Hurst RW, Galetta S. Clinical reasoning: a 59-year-old woman with acute paraplegia. Neurology. 2007;69:E41–E47.)
Location of SCI
The cervical region and thoracolumbar junction are the most frequent sites of injury in patients with SCI. Cervical spine injuries can occasionally be lethal, especially when the upper cervical cord is involved because it is critical for respiratory drive. Fracture–dislocations and subluxations of the cervical spine are most common at the level of the C5–C6 vertebrae. Thoracic fractures are less common than cervical, and most of these involve the vertebrae of the thoracolumbar junction (T10–12). These injuries are typically caused by crushing or extreme flexion of the spine in motor vehicle crashes or falls. Injuries involving the lumbar spine can damage the conus or the cauda equina depending on the level of injury.
Primary and Secondary Spinal Cord Injury
Primary SCI results from direct mechanical forces such as shear, laceration, distraction, and compression that cause structural disruption of neural and vascular structures with abrupt and indiscriminate cell death. Persistent pressure on the cord by space occupying bone, ligaments, or a disc can potentiate mechanical damage to the cord after the primary injury.
As a response to the initial mechanical insult, hemorrhage, edema, and ischemia rapidly follow, extending to contiguous areas of neural tissue. A subsequent biochemical cascade of events that involves a variety of complex chemical pathways leads to delayed or secondary cell death that evolves over a period of days to weeks. These “secondary injury” mechanisms result in the death of a population of neural cells that otherwise would have survived the initial insult. Thus, except for petechial hemorrhage, the human spinal cord may show no significant macroscopic or histopathologic changes until 6–24 hours after trauma.9 Although the exact mechanism of secondary spinal cord damage is not well understood, various functional hypotheses are proposed.
After the initial hemorrhage, inflammation proceeds in the central gray matter. On a systemic level, hypotension, either from hypovolemia or from autonomic dysfunction with neurogenic shock, contributes to impaired perfusion of the spinal cord and ischemia. Experimental studies in animal models of SCI have shown an increase in products of anoxic metabolism in neural tissue. Multiple other theoretical mechanisms could potentially contribute to the pathophysiology of secondary injury; however, a synergistic effect of several of these mechanisms is most likely responsible.
In the inflammatory theory, increased activity of cyclooxygenase and lipoxygenase results in accumulation of inflammatory mediators (i.e., prostaglandins, leukotrienes, platelet-activating factor, serotonin) that produce secondary neuronal damage.10 The effect of inflammatory mediators seems to be potentiated in anoxic conditions with diminished tissue perfusion.11 The neurotransmitter theory posits that increased levels of excitatory amino acid neurotransmitters such as glutamate and aspartate are released as a result of primary SCI and may cause secondary neuronal injury.12 Evidence to support this theory includes the experimentally induced neurologic dysfunction that occurs when the cord is exposed to excitatory amino acids, as well as the reduction in the extent of functional deficits with pretreatment using amino acid antagonists.13 The free-radical theory suggests that free radicals accumulate in the injured neural tissue and damage nucleic acids within the cell as well as lipids and proteins that comprise the cell membrane. Inability to maintain the integrity of the cell membrane results in neuronal death due to uncontrolled influx of ions and unbalanced osmotic pressure. The calcium ion theory implicates the influx of extracellular calcium ions into nerve cells as the cause of secondary injury since intracellular accumulation of calcium with efflux of potassium has been observed in experimental SCI.14 An excess of calcium ions activates phospholipases, proteases, and phosphatases that in turn lead to interruption of mitochondrial activity and disruption of the cell membrane. Initial neuronal swelling is related to sodium influx, whereas subsequent neuronal disintegration results from calcium influx. Both competitive and noncompetitive calcium channel blockers have been demonstrated experimentally to reduce secondary neurologic injury.15 Another theory postulates the involvement of endogenous opioids such as peptides, dynorphin, endorphin, and enkephalins, because time-dependent injuries can be related to dynorphin. Also, application of opiate antagonists such as naloxone has improved neurologic recovery in experimental models.15
All spinal cord lesions can be classified as neurologically complete or incomplete using the American Spinal Injury Association (ASIA) Scale16 (Fig. 23-7). This distinction is important prognostically since incomplete injuries have a chance at neurologic recovery, whereas motor recovery is achieved in only 3% of patients with complete injury during the first 24 hours and never after 24–48 hours.17,18
FIGURE 23-7 The ASIA classification of neurologic deficit following spine injury. (This form may be copied freely but should not be altered without permission from the American Spinal Injury Association.)
Patients with complete cord injury have no motor or sensory function caudal to the level of the injury. Although the ASIA classification requires a patient to have sacral sparing in order to be classified as incomplete, any sensory or motor function caudal to the level of injury is sufficient to designate a patient as incomplete because this signifies at least partial continuity of the long white-matter tracts (i.e., corticospinal and spinothalamic) from the cerebral cortex to the conus medullaris. During the initial evaluation of a patient with SCI, sacral root sparing may be the only neurologic function present to differentiate incomplete from complete SCI. Evaluation for sacral sparing consists of perianal sensation to light touch and pinprick, rectal tone, and voluntary contraction of the external anal sphincter. Spinal shock can complicate this assessment. It is a temporary state of spinal cord dysfunction associated with complete areflexia that usually resolves 24–48 hours after the time of injury. Until spinal shock has resolved, the completeness of the neurologic injury cannot be determined. Return of the bulbocavernosus reflex heralds the end of spinal shock. This clinical test assesses the integrity of the local S3–S4 reflex arc and is performed by squeezing the glans penis, placing pressure on the clitoris, or tugging on a Foley catheter while performing a rectal exam. An intact reflex will result in contraction of the anal sphincter. If there continues to be no distal sensory or motor recovery at the point the bulbocavernosus reflex has returned, the injury is designated as complete and no further significant neurologic improvement can be expected.
Anterior Cord Syndrome
This incomplete SCI results classically from vascular injury, resulting in anterior spinal artery insufficiency and ischemic injury to the anterior two thirds of the cord, but can also occur after blunt trauma to the anterior spinal cord (Fig. 23-8). Clinically, patients present with loss of motor function and pain and temperature sensation below the level of injury from involvement of the ventrally located lateral corticospinal and spinothalamic tracts. They do, however, retain proprioception and the ability to sense vibration and deep pressure from preservation of the posterior columns. Because ischemic neural tissue has a poor prognosis for recovery, the chance of meaningful clinical recovery in anterior cord syndromes is poor.
FIGURE 23-8 The most common patterns of incomplete spinal cord injury.
Central Cord Syndrome
Classically, central cord syndrome results from a hyperextension injury in an older patient with preexisting cervical spondylosis. It can, however, arise from a variety of different mechanisms. Clinically, the upper extremities are more involved than the lower extremities, due to the more central location of the upper extremity axons within the spinal cord tracts. Patients typically regain the ability to walk, but have more limited return of upper extremity function.
This incomplete cord syndrome can result from hemitransection of the spinal cord with unilateral damage to the corticospinal tract, spinothalamic tract, and dorsal columns. Patients present with loss of ipsilateral light touch sensation, proprioception, and motor function and contralateral loss of pain and temperature sensation. The prognosis is generally good.
Posterior Cord Syndrome
This syndrome is rare and results from involvement of the dorsal columns with subsequent loss of proprioception and vibration and preserved motor function. The prognosis is variable with many patients experiencing difficulty walking due to the deficit in proprioceptive sensation.
Cervical Root Syndrome
This represents an isolated nerve root injury that causes a deficit in sensation and motor function. This injury can be associated with an acute disc herniation or facet fracture, subluxation, or dislocation.
Conus Medullaris Syndrome
The conus medullaris is typically located at the level of the L1–L2 intervertebral space. Injury can produce mixed upper and lower motor neuron findings. Isolated injury to the conus may result in loss of bowel and bladder control (no sacral sparing), and the prognosis for recovery is poor.
Cauda Equina Syndrome
The cauda equina extends distal to the conus and is composed of the lumbar and sacral nerve roots. Injury results in lower motor neuron findings with sensory loss and motor dysfunction. Involvement of the lower sacral roots can result in bladder and bowel dysfunction. Urgent decompression (within 72 hours) optimizes outcomes, and the prognosis for motor recovery is moderate.
Advances in prehospital screening and transport have helped reduce the chance of missing a significant spinal injury.19 Field evaluation of patients with suspected spinal injury begins with the primary and secondary surveys as detailed by the American College of Surgeons Advanced Trauma Life Support (ATLS) course. The primary survey begins with evaluation of the airway, breathing, and circulation, followed by assessment of disability and exposure (ABCDE).
All patients are considered to have a spinal injury until proven otherwise. If lack of significant spinal injury cannot be ruled out at presentation, then immediate institution of spinal precautions is necessary. The cervical spine can be immobilized with a rigid cervical collar, but this is not a substitute for careful handling of the patient. With complete ligamentous disruption, the collar provides minimal stabilization. Manual stabilization of the spine is much more important and effective in restricting motion during patient transfers than any external orthosis.20 The thoracic and lumbar spines can be immobilized with a backboard at the time of injury. Recently, considerable attention has been directed toward the role of immobilization at the scene of injury since uniform application of spinal immobilization to all trauma patients may be unnecessary (e.g., patient with GSW to torso).21
If it becomes necessary to secure an airway, care is required during intubation to prevent hyperextension of the neck that might cause an iatrogenic injury to the cervical spine or spinal cord. Therefore, intubation should be performed with in-line cervical traction, a cervical collar in place, or fiber-optic assistance. Maintenance of oxygenation and hemodynamic stability with supplemental oxygen, blood pressure support, and early use of blood products may minimize the potential for secondary ischemic injury in patients with a suspected SCI. Patients who present with hypotension and shock usually have hypovolemia from hemorrhage and should be aggressively treated with fluid resuscitation and blood products. In patients with an injury to the spinal cord, however, hypotension may result from neurogenic shock, which is due to disruption of sympathetic output to the heart and peripheral vasculature. Neurogenic shock is distinguished by bradycardia, instead of tachycardia, in the presence of hypotension. These patients typically require the use of inotropic and chronotropic support to maintain adequate systolic blood pressures. Aggressive fluid resuscitation in patients with neurogenic shock risks fluid overload, pulmonary edema, and heart failure. The secondary survey consists of a thorough head to toe evaluation of the patient, and a complete neurologic exam should be obtained.
Details of the injury and the past medical history of the patient should be obtained from the patient, family members, and bystanders and relayed to the treating team. Knowledge of the mechanism of injury is important for the caregivers because associated injuries can then be predicted. Any transient neurologic symptoms noted after the traumatic event should be reported to the trauma team in the emergency department because such findings after trauma, even transient ones, suggest spinal instability. Disease states that may predispose patients to spinal injury should be asked about, as well. Included would be diseases that affect the structural integrity of the vertebrae (e.g., osteoporosis, metastatic disease), those that may be associated with instability (e.g., rheumatoid arthritis, trisomy 13, skeletal dysplasias), and those that result in a stiff spinal column (e.g., ankylosing spondylitis [AS], diffuse idiopathic skeletal hyperostosis [DISH], Klippel–Feil syndrome). Also, preexisting stenosis of the spinal canal may predispose to acute SCI.
Urgent transport to centers with the appropriate resources should follow initial stabilization in the field. Ideally, patients with SCI should be transferred directly to a facility experienced in the care of these patients (i.e., regional SCI center). If immediate transport is not possible, provisions should be made for early transfer once the patient is stabilized.
EVALUATION IN THE EMERGENCY ROOM
On arrival in the emergency room, the primary and secondary surveys are repeated. In obvious cases of cervical SCI, the need for ventilatory assistance should be determined in patients with paradoxical abdominal movement with respirations. A patient with an SCI above C5 and complete neurologic lesion is more likely to require intubation. As oxygenation and hemodynamic parameters are maintained, the patient should be examined for signs of injury and a repeat neurologic examination should be performed.
During inspection of the face and trunk, it is important to keep in mind that certain injuries can be associated with significant visceral and axial skeletal injuries. Facial trauma should alert the examining physician to the possibility of an injury to the cervical spine. An abrasion under the strap of a restraint can be associated with significant injuries to the cervical spine and cervicothoracic junction. Lap belt contusions should heighten suspicion for flexion–distraction injuries to the thoracolumbar spine. These can be associated with visceral injury, as well. Calcaneal fractures from significant decelerations (e.g., falls, motor vehicle crashes) are associated with fractures of the thoracolumbar and lumbar spines.
The unstable spine is at risk for injury from careless manipulation. Therefore, strict logrolling is the preferred method for evaluation of the back of the patient with a suspected spinal injury. The paraspinal soft tissues should be inspected for evidence of swelling, malalignment, or bruising. Systematic palpation of the spinous processes of the entire spinal column can help to identify and localize a spinal injury as significant gapping between processes can occur from flexion–distraction and fracture–dislocation mechanisms.
Following a systematic inspection, a complete neurologic examination that includes assessment of light touch and pinprick, graded motor examination, and reflexes is performed. In the appropriate settings, examination of sacral root function can be critically important though spinal shock can complicate this assessment. In patients with SCI, the ASIA Impariment Scale is useful for the characterization of residual function below the level of the SCI.
The goal of the secondary assessment is to identify and provide initial treatment of potentially unstable spinal fractures from both a mechanical and a neurologic basis. All clinical examinations of the spine should follow a consistent and repeated pattern. This pattern allows for comparison of neurologic status on a longitudinal basis, thus avoiding potential confusion about a progressive neurologic deficit.
Clinicians should have a low threshold for obtaining appropriate x-rays in the polytrauma patient because missed spinal injuries complicated by a progressive neurologic deficit most commonly result from insufficient imaging studies. Two factors associated with missed injuries in at least one study were traumatic brain injuries and AS.22
Plain x-rays are useful screening tools and are good for assessing overall alignment, though they have largely been replaced by computed tomographic (CT) imaging when evaluating the cervical spine. This is because a cervical CT is quicker to perform, more accurate, and cost-effective. The major difficulty with plain x-rays is obtaining technically adequate studies that are orthogonal and visualize the cervicothoracic junction. The lateral view of the cervical spine is still commonly obtained in the setting of an unstable polytrauma victim, however, because it can provide sufficient information to the surgical team to allow the patient to proceed to the operating room (Fig. 23-9). While plain radiographs of the thoracic and lumbar spines are useful for screening, it is difficult to obtain technically adequate films, especially in bedridden patients.
FIGURE 23-9 Analyzing a lateral cervical spine x-ray. The anterior vertebral line, posterior vertebral line, and the spinolaminar line should be smooth, collinear curves. An adequate study should visualize the entire cervical spine to the C7–T1 junction. The retropharyngeal soft tissue shadow can suggest the presence of injury when its thickness exceeds 6 mm at C2 and 2 cm (i.e., 20 mm) at C6 (“6 at 2 and 2 at 6”).
CT allows for better visualization of bony detail and is especially useful for visualizing the occipitocervical and cervicothoracic junctions. During preoperative planning, it can assist the surgeon in appreciating fracture planes and the degree of compromise of the spinal canal. Subtle translations evident on CT may suggest soft tissue disruptions that can be further evaluated with MRI. In cases with known spinal injury, it is a rapid means of screening for noncontiguous injury that can be present in 15–20% of patients.23 It bears repeating that, in patients with an identified spinal injury, an aggressive search for noncontiguous injury should be undertaken as this is a common cause for iatrogenic morbidity.
Magnetic resonance imaging allows for a detailed assessment of the soft tissues and is especially helpful for identifying pathology of the neural elements, intervertebral disks, and ligaments. It is not routinely used in the evaluation of the polytrauma patient because of the time required to perform a technically adequate scan. MRI is useful in the following patients: (1) when radiographic imaging is inconsistent with a patient’s neurologic presentation; (2) in the determination of ligamentous disruption when evaluating for spinal instability; and (3) prior to reduction maneuvers to exclude the presence of extruded intervertebral disks that could potentially be displaced dorsally into the thecal sac.
Patients with AS or other conditions that result in long fused spinal segments (e.g., DISH, extensive degenerative disease) deserve special mention in terms of their evaluation. These patients are particularly susceptible to fracture, even with low-energy mechanisms, because their spine functions as a long bone with no intervertebral motion to absorb energy. Patients with AS may present with relatively benign complaints. During their evaluation, careful attention should be paid to the position of their cervical spine. Because these patients typically have significant thoracic kyphosis, allowing their head to rest on the bed may result in excessive hyperextension, potentially through an unstable fracture (Fig. 23-10). Their head should be propped up so that the cervical spine assumes its normal configuration with the thorax. It is imperative that the entire spine is thoroughly imaged as noncontiguous injuries are common. Also, these patients have an increased incidence of epidural hematoma, so close serial neurologic examinations are mandatory.
Clearing the Cervical Spine
Historically, nearly one third of patients with injuries to the cervical spine had a delay in diagnosis or treatment due to inappropriate assessment.17 Of these patients with “missed injuries,” up to 5% may experience neurologic deterioration. Thus, early recognition of these injuries may prevent or limit neurologic compromise. While making sure injuries are not missed is essential, equally important is timely clearance in the absence of significant injury. This is because protracted evaluations may prolong immobilization, inhibit or restrict the thorough assessment of other organ systems, and complicate or delay recovery. Unfortunately, issues surrounding access and cost containment do not allow for the indiscriminant use of medical imaging on every trauma patient. For these reasons algorithms have been developed to identify patients who would benefit most from selected imaging studies. Within the context of these algorithms, imaging of the cervical spine in patients with minor trauma and in obtunded patients generates the most controversy. As noted above, CT of the cervical spine is the current imaging of choice.24,25
Anderson et al.26 have described a useful algorithm for clearance of the cervical spine whereby patients are classified into four groups as follows: asymptomatic, temporarily nonassessable secondary to distracting injuries or intoxication, symptomatic, and obtunded. Asymptomatic patients can be cleared on clinical grounds without imaging. Current ATLS recommendations advocate immediate removal of a cervical collar in the awake, alert, sober, and neurologically normal patient who has no tenderness to palpation in the cervical spine and who exhibits full, pain-free ROM. Two relevant algorithms include the NEXUS Low-Risk Criteria and the Canadian C-Spine Rules (Figs. 23-10 and 23-11), and the latter has better sensitivity and specificity.27 Temporarily nonassessable patients can be reassessed within 24–48 hours after return of mentation or following treatment of painful injuries. In urgent situations, the evaluation is the same as that of the obtunded patient. Symptomatic patients require advanced imaging studies including adequate cervical x-rays and CT. The clearance of obtunded patients is controversial and, unfortunately, no clear standard has emerged despite extensive recent research. These patients should be evaluated in an expeditious manner to minimize the restrictions and sequelae of continued immobilization. Some major trauma centers have started clearing the cervical spine in obtunded patients if there is a normal CT scan. Advocates of the use of CT as a single modality argue that MRI may detect additional abnormalities (20–30%), but most of these are false positives and require no further treatment.28,29Another option is to obtain an MRI if the CT is normal. Studies in favor of the use of MRI after a normal CT point to the high incidence of new abnormalities and an occasional unstable injury detected.30,31Currently, both options can be supported in the literature.
FIGURE 23-10 NEXUS Low-Risk Criteria for clearance of the cervical spine. (Reproduced with permission from Stiell IG, Clement CM, McKnight RD, et al. The Canadian C-spine rule versus the NEXUS low-risk criteria in patients with trauma. N Engl J Med. 2003;349:2510. Copyright © 2003 Massachusetts Medical Society. All rights reserved.)
FIGURE 23-11 Canadian Cervical Spine Rules for clearance of the cervical spine. (Reproduced with permission from Anderson PA, Gugala Z, Lindsey RW, et al. Clearing the cervical spine in the blunt trauma patient. J Am Acad Orthop Surg. 2010;18:149. © 2010 by the American Academy of Orthopaedic Surgeons.)
PROPHYLAXIS AGAINST DEEP VEIN THROMBOSIS
Prophylaxis against deep vein thrombosis (DVT) and venous thromboembolism (VTE) is an important consideration in patients with trauma to the spine. The incidence of VTE in trauma patients (with and without spinal trauma) varies from 0.36% to 30% within the literature.32,33 In patients with spinal trauma with no or a minimal neurologic deficit, studies suggest that the rate is low (0–2.1%).34 In contrast, one clinical study of patients with SCI using venography showed rates of DVT in the calf approaching 80% when no prophylaxis was used.35 Symptomatic VTE has been reported to occur in 4–10% of patients with SCI.36 Risk factors for DVT and VTE in trauma patients include the following: ventilator dependency >3 days, age >40 years, fracture in the lower extremity, major traumatic brain injury, venous injury, major surgical procedure, blood transfusion, and SCI.33 General patient risk factors include the following: older age, male patients, tobacco usage, diabetes mellitus, cancer, and obesity. Specific risk factors associated with spine surgery include the following: prolonged procedures, prone positioning, and anterior exposures to the lumbar spine (because of retraction of the great vessels).
All patients sustaining spinal trauma, especially those with an associated SCI, should have mechanical prophylaxis instituted as soon as possible with graduated compression stockings, sequential compression devices, or both. Pharmacologic prophylaxis (e.g., unfractionated heparin [UH] or low-molecular-weight heparin [LMWH]) is an additional consideration, but the benefit obtained in terms of reduction of DVT and VTE must be weighed against the risk of bleeding complications (e.g., spinal epidural hematoma). LMWH appears to be more effective for the prevention of DVT with fewer bleeding complications than UH.37 Patients at low risk (e.g., ambulatory patient with an isolated stable thoracolumbar compression fracture) do not need pharmacologic prophylaxis. Patients falling into the large “gray zone” between these two extremes must be considered on a case-by-case basis. In high-risk patients with spinal injuries amenable to nonoperative management, it is prudent to delay the start of pharmacologic DVT prophylaxis 24–48 hours to allow time for organization of any associated hematomas. In patients undergoing spinal surgery, pharmacologic DVT prophylaxis should be delayed until 24–48 hours after surgery to minimize the risk of wound problems and epidural hematoma. If pharmacologic prophylaxis is used in a patient who is scheduled for spinal surgery, it should be stopped an appropriate amount of time prior to surgery to allow the drug to clear the system as this minimizes the chances of bleeding complications (e.g., 12 hours prior to the procedure for LMWH). In patients who are poor candidates for pharmacologic prophylaxis (i.e., at high risk for bleeding complications), placement of a removable filter is another treatment option to be considered. The duration of pharmacologic prophylaxis should be based on the patient’s mobility in spinal trauma without SCI, while patients with SCI should be treated for at least 6 weeks.
THERAPEUTIC INTERVENTION WHEN AN SCI IS PRESENT
Once the diagnosis of an SCI has been established, prevention of decubiti, maintenance of adequate oxygenation and hemodynamic parameters, the role of pharmacologic and cellular interventions, and timing of surgical decompression must be considered (Table 23-1).
TABLE 23-1 SCI Protocol for Initial Evaluation and Treatment
Prevention of Decubitus Ulcer
In the insensate patient, it is important to limit the time on a spine board (ideally, less than 2 hours) to avoid the development of ischemia of soft tissue. In patients with SCI, 8 hours on a backboard has been associated with a 95% likelihood that a decubitus ulcer will form.38 Although the damage is done in the acute setting, the decubitus may not become evident for several days. Patients who cannot be mobilized immediately can be temorized in a rotating bed. These beds provide continuous mobilization (rotation of the bed along its longitudinal axis), improve drainage of lung secretions and ventilation, and reduce the risk of DVT and VTE.
Maintenance of Oxygenation and Hemodynamic Parameters
Maintenance of adequate arterial oxygenation and blood pressure is critical as ischemia, whether as a result of hypoxia or decreased perfusion, may potentiate secondary injury of the spinal cord. Patients should be placed in intensive care during the acute phase of care to ensure these parameters are optimized. Central venous and indwelling arterial catheters can be used to monitor hemodynamic parameters and responses to therapy, and a Swan–Ganz catheter may be necessary. Current treatment guidelines for patients with SCI recommend maintenance of systolic blood pressure >90 mm Hg and mean arterial blood pressure between 85 and 90 mm Hg for the first week after injury.39Although there are limited data, volume resuscitation supplemented by inotropic or chronotropic support as needed to maintain these parameters has been shown to improve neurologic outcomes.
Pharmacologic and Cellular Interventions
Neuroprotective interventions aim to attenuate the effects of secondary injury. How much secondary injury mechanisms contribute to overall neurologic deficit in patients with acute SCI is unknown. Atomic absorption spectroscopy suggests that secondary injury mechanisms may only account for 10% of the total pathology after SCI; however, the relative benefit of this small amount of preserved neural tissue may be significant.
The administration of high-dose methylprednisolone (30 mg/kg bolus followed by a 5.4 mg/kg infusion), in accordance with the findings of the second and third National Acute Spinal Cord Injury Studies (NASCIS), had been the standard of care at most North American institutions for over a decade.40 Steroids effectively limit the cellular and molecular events of the inflammatory cascade and are hypothesized to decrease the extent of secondary injury. The improvement in motor scores in the NASCIS trials, however, was minimal and, many have argued, not clinically significant. Recent criticism of the methodology (i.e., post hoc analysis) and interpretation of data from these trials has resulted in changing practice patterns, and currently, the use of methylprednisolone therapy is controversial.41–44
This is because the routine use of methylprednisolone has significant adverse side effects. In the NASCIS III trial, severe pneumonia affected twice as many patients and severe sepsis four times as many patients in the 48-hour steroid group compared with the 24-hour group. Six times as many patients died from respiratory complications in the 48-hour steroid group.40
Overall, the literature only, but not the ATLS course, supports the use of methylprednisolone within an acute time frame after injury (≤8 hours) and in adult patients with nonpenetrating injuries. Whether or not to administer steroids, however, should be individualized for each patient. The risk of infection in patients with certain comorbidities (e.g., diabetes mellitus, HIV infection) probably outweighs any potential improvement in the SCI. Additionally, patients with thoracic injuries are unlikely to improve with steroids. The ideal patient for steroid administration is young and healthy and has an incomplete injury to the cervical spinal cord.
While technically not a “drug,” systemic hypothermia is relatively noninvasive, systemically applied proposed treatment for patients with SCI. Though its use in this application was studied in the early 1990s with little success, renewed interest has resulted from its well-publicized application in Kevin Everett, an NFL football player who sustained a cervical SCI in 2007.45 He had immediate immobilization, steroids, hypothermia, and urgent surgery, though it is impossible to say which one of these contributed to his neurologic improvement. Systemic hypothermia is hypothesized to reduce the effects of secondary injury mechanisms through attenuation of the inflammatory cascade, but animal studies have had mixed results. Human clinical trials have not shown a consistent benefit, and, therefore, systemic hypothermia cannot be considered the standard of care.
Other agents under investigation such as GM1 ganglioside, naloxone, thyrotropin-releasing hormone, nimodipine, and tirilazad mesylate have proven promising in animal studies, also, but have not demonstrated sufficient efficacy in clinical trails.
Minocycline, erythropoietin, neurotrophic growth factors, and cellular therapies are promising neuroprotective agents and are being investigated. Minocycline, a tetracycline derivative, exhibits its neuroprotective properties by inhibiting matrix metalloproteinases, microglial activation (both are present during neuroinflammation), and preventing cell apoptosis.46 The administration of minocycline shortly after an experimentally induced SCI increased axonal sparing, reduced the apoptotic demise of oligodendrocytes, diminished axonal death, and culminated in improved locomotor and behavioral outcomes in animals.47 Erythropoietin, a hormone produced primarily by the kidney in response to hypoxia, has proven to be especially capable of minimizing SCI in ischemic models based on aortic occlusion.48 It has prevented motor neuron apoptosis and promoted motor functional recovery in animal models of SCI. Interestingly, erythropoietin reduced lipid peroxidation at the site of injury to a greater extent than methylprednisolone at the doses recommended in the Second National Acute Spinal Cord Injury Study (NASCIS II).
Recently, there has been a renewed interest in applying neurotrophic factors, growth factors, cytokines, and various forms of cell therapies in the treatment of SCI. Neurite outgrowth at the site of injury can be inhibited by myelin, myelin-associated protein (MAG), and Nogo protein.49 The application of specific Nogo receptor blockers facilitates axonal sprouting and enhanced functional recovery in rats. Both brain-derived neurotrophic factor (BDNF) and glial-derived neurotrophic factors (GDNF) increase the levels of cyclic adenosine monophosphate (cAMP) in neurons and promote axonal regeneration over the long distances relevant for functional recovery of the spinal cord.50 In addition, several types of bone morphogenetic proteins (BMPs) and interleukin-6 (IL-6) are being actively investigated to elucidate their roles in triggering reparative cascades in the injured spinal cord.51
Cellular therapies aim to deliver committed or uncommitted cells locally to the injury site in an effort to restore a functionally competent cellular environment to the injured cord. The primary cell types used in this approach include Schwann cells, olfactory ensheathing cells (OECs), and uncommitted stems cells. Schwann cells have been recognized as the key cellular constituent for peripheral nerve regeneration. Several animal studies have demonstrated that transected spinal cords can be bridged with Schwann cells delivered to the site of injury where they function as chaperones, guiding the sprouting axons.52 OECs are distinct glial cells that guide the growing axons and play a crucial role in the renewal of sensory neurons within the olfactory epithelium. Unlike Schwann cells, OECs have demonstrated the unique ability to extend across glial scar within the transected cord.53 In animal models, the transplantation of OECs into a completely transected spinal cord facilitated the long-distance regeneration of corticospinal, noradrenergic, and serotonergic fibers culminating in significant functional recovery.54 Prompted by these results, many SCI centers have initiated human trials that focus on the application of putative OECs.
Uncommitted mesenchymal and hematopoietic cells in the bone marrow are particularly promising for spinal cord repair due to their apparent ability to transdifferentiate into neurons and glia without cell fusion.55 These cells have great appeal because they can be easily procured, expanded in culture, and delivered intravenously. Preclinical studies have supported the feasibility of this approach and have confirmed the ability of intravenously administered mesenchymal stem cells to target regions of intraspinal cavitation.56 Significant concerns, however, exist for the potential of developing cancer from uncontrolled differentiation of these stem cell populations in vivo.
Experimental animal models of SCI have generated a number of promising experimental neuroprotective interventions, but have also exposed the overwhelming complexity of the neurobiological challenges. A greater understanding of this technology will be necessary for the further development of the optimal therapeutic approaches to the injured spinal cord.
TIMING OF SURGICAL DECOMPRESSION IN SCI
There are currently no standards regarding the role and timing of surgical decompression in acute SCI. Animal studies have shown that neurologic recovery is enhanced by early decompression, but the time frame for intervention in these studies is on the order of minutes to several hours.57,58 While several human studies have found a modest neurologic benefit to decompression within several hours, none of these studies were prospective, randomized, or controlled.59,60 Vaccaro et al.61 have performed the only prospective, randomized controlled study looking at functional outcomes in patients undergoing early versus late surgery after an SCI. This study showed no significant difference in outcomes between early and late surgery when the cutoff was 72 hours.61 Thus, the neurologic benefit of emergent (less than 4–8 hours) surgical decompression remains unclear. A prospective study, the Surgical Timing in Acute Spinal Cord Injury Study (STASCIS), is attempting to better define the benefits of urgent decompression and stabilization.
Studies have suggested that patients with SCI who undergo “early surgery” have significant benefits in terms of more rapid rehabilitation, decreased incidence of pneumonia, decreased length of stay in the intensive care unit, and decreased hospital costs.62
The most recent recommendations suggest that there is good evidence that early (>24 hours) surgical decompression can be performed safely and that urgent decompression (<24 hours) is recommended following isolated acute cervical SCI provided hemodynamic stability is maintained.63 In addition, urgent closed reduction of unilateral or bilateral facet dislocations in patients with incomplete quadriplegia and urgent decompression in a patient with neurologic deterioration following an SCI are advocated in the literature.
MANAGEMENT OF SPINAL INJURIES
Traditionally, injuries of the spine have been treated nonoperatively but, over the past century, surgical indications have been continuously redefined as surgical techniques have improved and an evidentiary basis for intervention has been developed. Optimal management must weigh the risk and benefit of interventions against the goals of early mobilization, timely treatment of neurologic deficits, and maintenance of acceptable spinal alignment and stability in the context of the polytraumatized patient. To this end, communication between the trauma team and the spine surgeon is essential.
One of the most important tasks for the treating spine surgeon is the determination of spinal stability. Most spine surgeons agree with the general definition that spinal stability refers to the ability of the spine to maintain its alignment and protect the neural structures during normal physiologic loads. Biomechanically, and in the laboratory, spinal instability refers to an abnormal response to applied loads and can be characterized by motion in spinal segments beyond the normal constraints. Clinically, this situation risks progressive deformity, neurologic compromise, or both and, therefore, is an indication for operative stabilization. In clinical practice, however, the quantitative assessment of spinal stability is challenging.
Most classification systems of spinal trauma provide a descriptive analysis of various fracture patterns, and the extent to which these injuries reflect spinal stability remains controversial. One of the early classification systems proposed to aid in the assessment of spinal stability was region-specific parameter checklists, developed by White and Panjabi64 (Tables 23-2 and 23-3). This system assigns points based on a combination of radiographic findings, neurologic examination, and the anticipated biomechanical demands of the patient. Scores above a certain cutoff (e.g., >5) suggest an unstable spine. Neurologic findings were considered important because these implied the spinal column had at some point failed in its ability to protect its enclosed neural elements and could, therefore, be considered unstable. In practice, most spine surgeons would agree that spine fractures associated with significant neurologic injury are unstable, except for those injuries secondary to penetrating trauma (i.e., gunshot, stab). While these systems have been in use for years, some of the parameters are difficult to define clinically. For example, it is difficult to determine whether anterior or posterior elements are “destroyed or unable to function” and what the definition of “dangerous loading anticipated” should be.
TABLE 23-2 Checklist for the Diagnosis of Clinical Instability of the Thoracic and Thoracolumbar Spine (Total ≥5 Points = Unstable)
TABLE 23-3 Checklist for the Diagnosis of Clinical Instability of the Lumbosacral Spine (Total ≥5 Points = Unstable)
Mechanistically, cervical spine fractures and dislocations have been most comprehensively classified by Allen et al.65 (Fig. 23-12). In this system, injury patterns are described according to the position of the neck at the time of the injury and the direction of the injuring force. The six most common patterns of cervical injuries are as follows: (1) flexion–compression; (2) vertical compression; (3) compression–extension; (4) flexion–distraction; (5) extension–distraction; and (6) lateral flexion injuries. Each injury pattern is graded in terms of the degree of injury to the involved motion segment and a higher stage denotes a more complex injury.
FIGURE 23-12 Patterns of subaxial cervical spine injuries based on the injury mechanism. (Reproduced with permission from McAfee P. Cervical spine trauma. In: Frymoyer JW, ed. Adult Spine. New York: Raven Press; 1991:1080.)
More recently, evidence-based guidelines have been developed for the classification of subaxial spinal injuries. The Subaxial Injury Classification (SLIC) Scale proposes a scoring system using three major injury characteristics to help direct the management of subaxial injuries66 (Table 23-4). In addition to the morphologic fracture pattern considered with the earlier systems, the neurologic status of the patient and the integrity of the discoligamentous complex (DLC) that stabilizes the spine (i.e., the intervertebral disk, the facet joint capsule, the ligamentum flavum, the interspinous and supraspinous ligaments) are taken into consideration. The SLIC scoring system can be used to direct treatment into the broad categories of either surgical or nonsurgical by summing the points in each of the three categories. Injuries that score 5 or more points are treated surgically, whereas those scoring 3 or less are treated nonsurgically. A score of 4 is considered equivocal.
TABLE 23-4 Subaxial Injury Classification (SLIC) Scale
In the SLIC, injury morphology is divided into the following three categories based on the relationship of the vertebral bodies to each other (anterior support structures): (1) compression; (2) distraction; and (3) translation or rotation. The DLC is graded as disrupted, intact, or indeterminate. The integrity of these soft tissue constraints directly relates to spinal stability. Neurologic status is graded as intact, root injury, complete injury, and incomplete injury in increasing order of point value. An additional point is given if there is continuous neural compression in the setting of a neurologic deficit. In this system, the presence of an incomplete neurologic injury, particularly in the presence of ongoing root or cord compression, leads to the highest point score as these would tend to benefit most from operative stabilization.
The management of thoracolumbar fractures has improved due to imaging techniques such as CT and MRI that better delineate fracture patterns, associated injury to soft tissue, and compromise of the neural canal allowing a more complete understanding of these injuries. A number of classification systems have been developed to determine stability. Holdsworth67 introduced the concept of dividing the thoracolumbar spine into two structural columns. The anterior column consisted of the anterior longitudinal ligament, vertebral body, intervertebral disc, and the posterior longitudinal ligament. The posterior column consisted of the remaining osteoligamentous structures. According to Holdsworth’s model, instability occurs when both columns have been compromised. Denis68 developed the three-column classification system, separating Holdsworth’s anterior column into the anterior and middle columns. Compression fractures involved the anterior column, burst fractures the anterior and middle columns, flexion–distraction injuries all three columns, and fracture–dislocation injuries all three columns, as well. According to this model, involvement of at least two of the three columns results in spinal instability. Panjabi et al.69 tested an ex vivo cadaveric burst fracture model and validated the three-column theory. In this study the middle column proved to be the principal determinant of stability of the thoracolumbar spine. Magerl et al.70 described a comprehensive classification of thoracolumbar injuries that defined compression, distraction, and torsion as the three mechanisms. These 3 categories were divided into 53 specific subtypes altogether.70 Unfortunately, all these systems have deficiencies as they are cumbersome to use and they do not always predict the best clinical practice. For example, the Denis system defines burst fractures as unstable (i.e., require surgical stabilization) because of involvement of the anterior and middle columns. More recent studies have shown that many burst fractures are stable and amenable to nonoperative management.
Recently, the Thoracolumbar Injury Classification and Severity Score (TLICS) has been introduced based on an extensive review of the literature as well as consensus opinion from an international group of spinal trauma surgeons. The goal is to simplify injury classification and facilitate decision making for treatment.71 Similar to the SLIC, the three major injury characteristics are defined as injury morphology, neurologic status, and integrity of the PLC. Point values are assigned to each major category based on injury severity (Table 23-5). The sum of these points represents the TLICS severity score, which may be used to guide treatment. Injuries with a score >4 are treated surgically because of significant instability, whereas injuries with a score <4 are treated nonsurgically. A score of 4 is considered equivocal. In the setting of multiple fractures, management is determined based on the injury with the greatest TLICS severity score. For noncontiguous fractures, the severity score of each injury may be used.
TABLE 23-5 Thoracolumbar Injury Classification and Severity Score (TLICS)
Of note, there have been no systems used to define instability in the upper cervical spine where the integrity of the ligamentous restraints is the most important. Disruption of the facet capsules, tectorial membrane, alar ligament, and TAL have all been associated with significant clinical instability of the upper cervical spine.
NONOPERATIVE MANAGEMENT OF SPINAL INJURIES
The nonoperative management of spine fractures consists of spinal immobilization with an orthotic brace that restricts motion to an extent that allows healing of the injury while permitting mobilization of the patient. The success of nonoperative treatment is dependent on proper patient selection and on the physician’s understanding of the spinal injury. Most stable injuries without a neurologic deficit are managed nonoperatively with the orthotic brace used to minimize the risk of progressive collapse. Most bony injuries have the potential to heal with this approach, while healing of ligamentous injuries is less predictable and commonly requires spinal fusion.
For patients in whom nonoperative management of their spinal injuries is chosen, continual reassessment of neurologic status and spinal stability is essential. Serial neurologic examinations are essential to monitor for progressive neurologic dysfunction. When the patient is able to be upright, repeat x-rays with the patient standing should be obtained. This will ensure that the initial assessment of spinal stability was accurate and that the spine is not progressively deforming with applied load due to collapse of the anterior column or insufficiency of the posterior ligaments.
Spinal orthoses are external devices that can restrict the motion of the spine by acting indirectly through the intervening soft tissue. Despite the heterogeneity of designs, the theoretical functions of all spinal braces are similar and include restriction of spinal movement, maintenance of spinal alignment, reduction of pain, and support of the trunk musculature. Also, spinal braces act psychologically as proprioceptive reminders for the patient to restrict spinal motion. The immobilizing effectiveness of spine braces is dependent on the mobility and anatomical features of the spine to be stabilized, injury biomechanics, length and rigidity of the orthosis, thickness of the intervening soft tissues, and patient’s compliance with the orthosis. Although spinal braces are generally applied to stabilize a specific motion segment, their immobilization properties typically affect the entire spinal region (i.e., cervical, thoracic, or lumbar).
Cervical spine bracing is particularly challenging due to the wide range of normal spinal motion. Due to the inability of the vital structures in the neck to withstand prolonged compression, cervical braces utilize the cranium and thorax as fixation points. Cervical orthoses can be used as a definitive treatment for some spinal injuries or as temporary immobilizers for postinjury transport or during the early hospital diagnostic process. These braces can generally be divided into soft collars, short cervicothoracic orthoses (CTOs), and long CTOs.
Soft collars are basically foam cylinders that encircle the neck. The mechanical function of soft collars is negligible, and their effectiveness relates more to the psychological effect of wearing an orthosis. Soft collars are indicated for mild cervical sprains or to provide postoperative comfort following stable internal fixation.
Short CTOs have molded occipital–mandibular supports that extend not lower than the level of the sternal notch anteriorly and the T3 spinous process posteriorly. These collars limit cervical motion to a much greater degree than soft collars, but still restrict motion only 50–80%.72 They control motion best in the sagittal plane. While certain collars restrict motion best among the aforementioned examples, they exert relatively high skin pressures and should be used only on a short-term basis.
Long CTOs attach to the cranium at the occiput and mandible and extend to the lower thorax below the sternal notch and T3 spinous process. All of these braces provide better fixation to the head and trunk than short CTOs and, therefore, constitute the most effective of all cervical braces; however, they are cumbersome and the least well tolerated by patients. The major difference between short and long CTOs is the ability of the latter to provide better control of spinal rotation and sagittal motion in the middle and lower cervical spine.
The thoracic spine (T1–10) is unique among spinal regions in terms of its inherent rigidity and location between highly mobile adjacent cervical and lumber segments. Stable injuries above T6 are not typically braced because of the stabilizing effects of the rib cage, sternum, and shoulder girdle and the poor compliance and unclear benefit of long cervicothoracic bracing. In addition, traditional thoracolumbosacral orthoses (TLSOs) do not effectively stabilize this region even with proximal extension of the brace to the armpits. Injuries below T7 may be amenable to bracing with a standard TLSO, but achieving significant restriction of movement in the unstable thoracic spine can be difficult. This is because rotation, the principal motion of the thoracic spine, is much more difficult to control than flexion and extension.
The thoracolumbar region is the region best controlled by traditional TLSO bracing. The goal of thoracolumbar bracing is to support the spine by limiting overall motion of the trunk, decreasing muscular activity, increasing the intra-abdominal pressure, reducing spinal loads, and limiting spinal motion. Interestingly, interim analysis of a recent multicenter randomized trial found that thoracolumbar burst fractures without an associated injury to the PLC were successfully treated without a brace due to the inherent stability of fractures in this region.73
The lower lumbar spine (L4–L5) is difficult to brace due to the limited caudal fixation points and its physiologic hypermobility. Typically, adequate stabilization requires that the brace extend as much as four or five vertebral levels proximal and distal to the unstable segment. Even when the brace includes a hip spica component, hip flexion is not adequately controlled, resulting in inadequate lumbar protection. Current available orthoses include lumbosacral corsets, braces, and full-contact custom-molded orthoses.
The halo fixator consists of a ring attached to the skull via pins that in turn is connected via upright longitudinal struts to either a detachable vest or thoracic cast that can be applied with local or general anesthesia. The ring is positioned below the cranial brim similar to tongs, and the pins are inserted with approximately 6–8 in lb of torque. Anteriorly, pins should be placed 1 cm above the lateral one third of the eyebrow to avoid the temporalis fossa and the supraorbital and supratrochlear nerves medially (Fig. 23-13). The cast or vest must be well fitted to the iliac crest to prevent vertical toggle of the apparatus. Despite adequate fit, the halo vest does not guarantee complete spinal stabilization, especially when the patient is upright.
FIGURE 23-13 (A and B) Halo pin placement.
The halo may be used to temporarily stabilize injuries prior to surgery or to definitively treat certain injuries. In comparison to other options for bracing of the cervical spine, halos are the best means of restricting upper cervical motion and, in addition, are superior to other options in the ability to restrict motion in all planes (Fig. 23-14). In a study of normal subjects, the halo fixator allowed only 4% of cervical sagittal motion compared with 13% for a rigid cervical collar and 74% for a soft collar and only 1% of lateral rotation and 4% of lateral bending.74 Despite the fact that halos restrict cervical motion to a significant degree, they do not necessarily prevent progression of unstable injuries. Therefore, the decision to use a halo for definitive management should be carefully made. Halos are poorly tolerated in elderly patients and potentially increase the risk of injury due to falls as the device shifts the center of gravity upward and makes patients “top-heavy.”75 In patients with SCI, halo fixation can be problematic. Loss of protective sensation over the trunk can lead to formation of a decubitus ulcer. Restriction of expansion of the chest wall can lead to pulmonary complications in the patient with already compromised respiratory function. Furthermore, the bulk of the halo can hinder nursing care and mobilization.
FIGURE 23-14 (A and B) Halo ring apparatus.
Cervical traction is often indicated in the management of injury to the cervical spine to reduce bony deformity, indirectly decompress neural elements, and provide stability. The utilization of traction is helpful in the management of injury to the cervical spine because it is simple to apply and has low morbidity when applied properly. Use of this technique requires an awake and alert patient (i.e., cannot be intoxicated). Cervical traction can be applied with tongs or a halo ring, and the use of CT- and MRI-compatible materials is preferred.
Traction tongs achieve fixation into the bony skull through two pins oriented 180° from each other. These pins have pointed tips that abruptly flare out to allow for fixation into the outer cortical table of the skull while preventing inner cortical penetration. These pins are placed in line with the external auditory meatus below the cranial brim or the widest diameter of the skull. Pins must be positioned posterior to the temporalis muscle since placement here can become symptomatic due to the thin bone in this region or from muscle irritation during mastication. The pins are tightened to 6–8 in lb torque and require repeat tightening within 24 hours to maintain the applied load. If the pins loosen again, they can be retightened and, if loosening continues to recur, the pins should be moved. Traction tongs control motion in a single plane through the application of longitudinal traction and are associated with a greater incidence of loosening than halo rings. They are indicated when the need for longitudinal traction is temporary or when the patient is bedridden. Use of a halo ring achieves better skeletal fixation due to its multipin circumferential application and is able to withstand higher loads for a longer period of time.
Once a set of tongs or a halo ring has been applied, cervical traction can begin with approximately 10–15 lb. This is followed by immediate evaluation by x-rays to assess for overdistraction and to rule out occipitoatlantal injuries. The traction weight can then be increased by 5- to 10-lb increments with serial neurologic examinations and lateral x-rays obtained at approximately 10- to 15-minute intervals after each weight increase. The patient must be completely relaxed, and analgesia or muscle relaxants are often required to relieve muscle spasms or tension. The head of the bed should be elevated to provide bodyweight resistance to the traction, and the shoulder portion should be depressed to optimize visualization of the cervicothoracic region. The maximum amount of weight that can be safely applied for closed traction reduction of the cervical spine is controversial. Some authors recommend a maximum of 5 lb per level of cervical injury, beginning with 10 lb for the head. This would limit the weight applied to a C5–C6 facet dislocation to 35 lb. Other authors have supported a more rapid incremental increase in load, applying weights up to 150 lb without any adverse effects. Typically, the maximum weight tolerated is limited by the skeletal fixation utilized, and, for cranial tongs, this limit is up to 100 lb. Most injuries to the cervical spine can be reduced with only longitudinal traction, but small changes in the vector of traction (i.e., slightly more flexion or more extension) are necessary in some cases.
The urgency of the reduction is based on animal studies of SCI that suggest a window of 6–8 hours during which decompression may reverse neurologic deficits.57,58 Pretraction MRI may be obtained to identify the presence of herniation of an intervertebral disc, but this is controversial.76,77 A recent survey of fellowship-trained spine surgeons found the timing and utilization of MRI for patients with a traumatic dislocation of a cervical facet to be variable.77 Early closed reduction in awake and alert patients presenting with significant motor deficits (e.g., complete SCI) without prior MRI appears reasonable because the benefit of early reduction appears to outweigh the risk of neurologic deterioration. In patients without a neurologic deficit or those unable to participate in a neurologic examination (e.g., obtunded patients), the risk of iatrogenic neurologic injury probably outweighs the benefit of early reduction.
OPERATIVE MANAGEMENT OF SPINAL INJURY
Operative management requires a thorough understanding of the anatomy, biomechanics, and physiology of the injury as well as knowledge of the surgical strategies to achieve adequate spinal alignment, spinal stability, and neurologic decompression.
A decision to proceed with surgery should include consideration of perioperative strategies to minimize the chances of untoward events. In general, patients should be moved with strict spinal precautions that include some form of immobilization if the cervical spine has not yet been cleared, logrolling of the patient for turns, and use of a backboard for transfers. It is important to remember that manual immobilization of the cervical spine restricts motion better than any orthosis. If the cervical spine has not yet been cleared, the neck must be carefully manipulated during intubation. Intubation with in-line cervical traction may be acceptable in more stable injuries. Consideration can be given to fiber-optic intubation, nasal intubation, or awake intubations in patients with less stable injuries. Oxygenation and blood pressure (mean arterial pressure >85–90 mm) should be supported during the procedure, especially in patients with unstable injuries or frank SCI. Intraoperative neurophysiologic monitoring may add a measure of protection against iatrogenic neurologic injury.
Goals of Operative Management
Only spinal injuries that are unstable with or without neurologic involvement require surgical treatment. Surgical objectives include the correction of spinal alignment, the restoration and maintenance of spinal stability with instrumentation, and decompression of compromised neural elements to permit maximal functional recovery. Modern anterior and posterior surgical fixation devices provide stabilization while limiting the number of motion segments included in the construct in comparison to older devices. Finally, the injury type and the time to surgical intervention determine the most appropriate type of decompression. Neural decompression can be performed anteriorly, posteriorly, and/or indirectly (via restoration of spinal alignment). The literature suggests that the results of anterior direct versus posterior indirect spinal canal decompression are similar for patients with incomplete neurologic deficits.78 The absolute indications for anterior decompression would include the neurologically incomplete patient with greater than 50% canal compromise, greater than 72 hours postinjury, a failed attempt at posterior reduction, or significant loss of anterior and middle column (vertebral body) support despite posterior reduction.
Timing of Stabilization of Spinal Fractures
The optimal timing of stabilization following spinal injury is controversial. Some insist that surgery be performed as soon as possible as stabilization and mobilization of patients with SCI has been found to reduce the incidence of complications such as adult respiratory distress syndrome and DVT.79,80 Others advocate a delay of surgery to allow for cardiopulmonary optimization and to minimize the risk of bleeding. The indications for immediate surgery are progressive neurologic deterioration and fracture–dislocations associated with incomplete or no neurologic deficit. In the absence of a neurologic deficit, it is reasonable to delay surgery to facilitate surgical planning and to allow for edema of the spinal cord and nerve roots to resolve. Furthermore, organization of a hematoma occurs at about 48 hours after the injury and decreases intraoperative blood loss. An excessive delay to surgery may adversely affect the clinician’s ability to reduce the fracture and achieve clearance of the canal. Reports have shown that optimum clearance of the spinal canal is most effective if surgery is ideally performed within 4 days and certainly no later than 7–10 days from the time of injury.81,82 Management of the polytrauma patient with an associated spinal injury is a particularly difficult problem. In several studies examining the effects of early versus late stabilization of spinal fractures in such patients, surgery performed within 72 hours on patients with Injury Severity Scores (ISS) greater than 18 consistently and significantly decreased morbidity and length of stay without significant differences in the rate of perioperative complications.83
Recently, with the development of less invasive spinal interventions, there has been growing interest in the application of “damage control” principles to spinal trauma. Because the morbidity of spinal surgery may not be tolerated in the acutely injured patients, less invasive techniques may find a role in the timely stabilization of injuries to allow for early mobilization of the patient. While the theoretical benefit of applying these techniques in the trauma population is sound, there are currently insufficient data from which to draw firm conclusions as to whether these techniques are associated with superior outcomes.
SPECIFIC MANAGEMENT OF INJURIES OF THE UPPER CERVICAL SPINE
Occipital Condyle Injuries
Fractures of the occipital condyles typically occur through an axial loading mechanism. They can be associated with significant occipitocervical instability and should be evaluated carefully. The integrity of the restraining ligaments is key in the assessment of stability. Anderson and Montesano84 described a classification system that evaluates the potential for instability based on CT patterns of bony injury and the presumed associated ligamentous injuries (Fig. 23-15). Type I fractures are comminuted (impaction fractures of the condyles) and are generally stable as the ligaments are typically intact. Type II fractures of the condyles are associated with a basilar skull fracture. These are stable unless the entire condyle is separated from the occiput (i.e., “floating” condyle). Both type I and type II fractures can be treated with a cervical orthosis. Type III fractures are avulsions of the insertion of the alar ligament into the occipital condyle. These have the greatest potential for instability and have been found to occur in 30–50% of patients with occipitocervical dislocations. For type III fractures, stable fracture patterns can typically be treated in a rigid collar for 6–8 weeks. Less stable (e.g., displaced) variants may benefit from halo immobilization for 8–12 weeks. Unstable patterns (e.g., significant translation, joint incongruity, or diastasis) typically require occiput to C2 fusion.
FIGURE 23-15 Classification of occipital condyle injuries. (Reproduced with permission from Anderson PA. Injuries to the occipital cervical articulation. In: Clark CR, ed. The Cervical Spine. 3rd ed. Philadelphia: Lippincott Williams & Wilkins; 1998:391.)
Acute Occipitocervical Instability
Occipitocervical injuries are rare and usually fatal, but may be increasing because of better automobile restraints, improved prehospital care, and more thorough imaging techniques. These injuries consist of subluxations or dislocations and can include fractures of the occipital condyles. The most popular methods to describe the relationships between the occiput, atlas, and axis in the setting of trauma are the basion–dental interval (BDI) and the basion–posterior axial line interval (BAI).85 As values greater than 12 mm are associated with occipitoatlantal subluxation or dislocation, these measurements have been referred to as the “Rule of 12’s” (Fig. 23-16). The Power’s ratio can be used, as well; however, this is only useful for detecting anterior occipitocervical subluxations/dislocations (Fig. 23-17). Injuries are classified as type I (anterior displacement of the occiput on the atlas), type II (longitudinal facet distraction injury with diastasis of the occiput from the atlas), and type III (posterior displacement of the occiput on the atlas).86 Traction should be avoided in these injuries in favor of gentle manipulation, and reduction is achieved by either extension or flexion of the occiput. Occipitocervical subluxations/dislocations are extremely unstable and require operative stabilization and fusion. Halo immobilization is used for temporary urgent stabilization prior to definitive treatment.
FIGURE 23-16 The Rule of 12’s. The basion–atlantal interval (BAI) and the basion–dens interval (BDI) measure less than 12 mm in the uninjured state.
FIGURE 23-17 Power’s ratio.
Fractures of the Atlas
Fractures of the atlas rarely cause a neurologic deficit, but are painful and compromise mobility of the neck. Fractures can consist of bone disruption without instability or bony and ligamentous injuries with subsequent displacement of the articulation of the lateral mass. The three common types of atlas fractures are as follows: (1) posterior arch fractures, in which the lateral atlantal masses do not spread; (2) burst or Jefferson fractures, in which the lateral masses will spread and displace laterally; and (3) lateral mass fractures, in which lateral displacement of the lateral mass will occur only on the fractured side (Fig. 23-18). Stable fractures with an intact transverse ligament and simple posterior arch fracture can be treated with a brace. For Jefferson fractures, combined overhang of the lateral mass greater than 7 mm (rule of Spence) on an open mouth odontoid x-ray can be associated with disruption of the TAL and instability.87 Unstable, displaced fractures in which the transverse ligament is disrupted require traction initially to reduce the displaced lateral masses and then can be placed in a halo vest. Fractures of the atlas with significant displacement can require traction for up to 6–8 weeks before a halo vest can be applied. Surgery is usually not required acutely, but is reserved for those patients who develop a symptomatic nonunion or late instability.
FIGURE 23-18 Atlas fractures. (Reproduced with permission from Jackson RS, Banit DM, Rhyne AL, Darden BV. Upper cervical spine injuries. J Am Acad Orthop Surg. 2002;10:271. Copyright 2002 by the American Academy of Orthopaedic Surgeons.)
Acute Atlantoaxial Instability
Atlantoaxial subluxation occurs when TAL insufficiency results in abnormal translation or rotation of the atlas with respect to the odontoid. TAL insufficiency can result from isolated rupture or avulsions of the TAL or, more commonly, from burst fractures of C1. On lateral cervical imaging, subluxation can be quantified by an increase in the anterior atlantodens interval (Fig. 23-19). CT is necessary to appropriately characterize these injuries. MRI is useful to evaluate the integrity of the soft tissue restraints, including the TAL. Because these injuries are extremely unstable, there is a high risk for neurologic deficit. Cervical traction with tongs is appropriate initially to achieve reduction. Definitive treatment consists of a posterior cervical fusion of C1–C2. Rotary subluxations of C1–C2 clinically present with torticollis. Subluxation will usually reduce with traction and requires only brace support or application of a halo vest. Irreducible subluxations require open reduction and C1–C2 fusion, but may occasionally require occiput to C2 fusion depending on the extent of injury.
FIGURE 23-19 Atlantoaxial subluxation. Note the increased anterior atlantodens interval (normal <4 mm).
Fractures of the Odontoid
Odontoid fractures have been classified by Anderson and D’Alonzo88 according to the anatomical level of the fracture (Fig. 23-20). Type I fractures consist of avulsion injuries of the apical ligament that are essentially stable and require limited, if any, external support. Type II odontoid injuries occur at the waist of the odontoid, a watershed area for fracture healing, and have the potential for poor healing. Type III fractures extend below the waist of the odontoid into the body of C2 and usually result in uneventful bony union when treated with halo immobilization. Odontoid fractures can be visualized on lateral and open mouth x-rays, but are best visualized on CT. Nonoperative treatment of type II injuries is associated with an up to 15% incidence of nonunion, although not all of these are symptomatic.
FIGURE 23-20 Classification of odontoid fractures. (Reproduced with permission from Jackson RS, Banit DM, Rhyne AL, Darden BV. Upper cervical spine injuries. J Am Acad Orthop Surg. 2002;10:271. Copyright 2002 by the American Academy of Orthopaedic Surgeons. Adapted with permission from Anderson LD, D’Alonzo RT. Fractures of the odontoid process of the axis. J Bone Joint Surg Am. 1974;56:1663.)
Type II odontoid fractures are the most challenging since these require reduction of both translation and angulation and maintenance of fracture stability. Risk factors for nonunion include osteoporosis, rheumatoid arthritis, and diabetes mellitus. Significant displacement (>10° of angulation, >5 mm displacement) and posterior displacement are radiologic risk factors for nonunion or malunion. Odontoid fractures should be reduced in traction prior to definitive management. After an adequate reduction has been achieved, treatment options include immobilization in a halo vest, anterior screw osteosynthesis, and posterior C1–C2 fusion.
Traumatic Spondylolisthesis of the Axis
C2 pedicle or pars interarticularis fracture (Hangman’s fracture) results from hyperextension with an axial load. These fractures are rarely associated with a neurologic injury as separation of the fracture fragments decompresses the spinal canal. They can be detected on lateral x-rays, but are best visualized on CT. MRI is useful to assess the integrity of the associated DLC. These injuries have been classified by Levine and Edwards89 based on the extent of fracture displacement and angulation (Fig. 23-21). The treatment is dependent on the extent of associated injury to the disc and ligament. In the minimally displaced (<3 mm) type I fracture, a rigid collar is sufficient treatment. In type II fractures, the C2–C3 disk and posterior longitudinal ligament are disrupted resulting in >3 mm of translation and significant angulation. These can be treated with gentle traction and reduction with extension followed by immobilization in a halo device. Type IIA fractures are a less common variant with minimal translation, but significant angulation. These have a more oblique fracture line and are thought to occur from a flexion–distraction-type mechanism. It is imperative to recognize these injuries as traction will worsen the deformity and potentially cause an SCI. Such injuries require reduction in extension with axial load followed by immobilization in a halo device. C2–C3 fusion is indicated for II or IIA injuries if reduction cannot be maintained. It is important to recognize an “atypical” Hangman’s fracture because these are associated with an increased risk of a neurologic injury.90 In this pattern, the fracture line extends further anteriorly into the body of the axis. Separation of the major fragments in this instance can result in spinal cord compression by the posterior part of the body. C2 pars fractures with facet dislocation (type III) are uncommon. Nonoperative management is ineffective due to the disrupted continuity between the C2 body and the posterior elements, and open reduction and fusion are required.
FIGURE 23-21 Classification of traumatic spondylolisthesis of the atlas (Hangman’s fracture). (Reproduced with permission from Jackson RS, Banit DM, Rhyne AL, Darden BV. Upper cervical spine injuries. J Am Acad Orthop Surg. 2002;10:271. Copyright 2002 by the American Academy of Orthopaedic Surgeons. Adapted with permission from Levine AM, Edwards CC. The management of traumatic spondylolisthesis of the axis. J Bone Joint Surg Am. 1985;67:217.)
Injuries of the Subaxial Cervical Spine (C3–C7)
Subaxial cervical injuries can consist of fractures, dislocations, subluxation, or a combination of these injuries. The assessment of instability is essential to designing appropriate treatment strategies. Recently the SLIC score has been validated to assist in this determination.
Burst fractures occur as a result of pure axial load applied to the injured vertebrae and are characterized by disruption of the anterior and middle columns of the spine with some degree of compromise of the spinal canal from retropulsed bony fragments. As in all of these spinal fractures, CT is helpful to delineate the bony injury. MRI is useful to assess the integrity of the soft tissue restraints and the degree of neurologic compromise. Appropriate treatment is determined by the presence of a neurologic deficit, degree of malalignment, and the extent of instability. Neurologic compromise is a clear indication for anterior decompression and instrumented fusion to provide stability. In the absence of a neurologic defect, the degree of instability becomes the most important factor guiding treatment. More stable variants may be treated with immobilization in a halo vest. Less stable fracture patterns (e.g., significant translation, kyphosis, loss of vertebral height) benefit from stabilization to prevent a neurologic injury and progressive kyphosis.
These occur from a combination of axial load and forward flexion resulting in compressive failure of the anterior vertebral body and tensile failure of the PLC. A teardrop-shaped avulsion of the anteroinferior tip of the more cranial vertebra is common and responsible for the commonly associated eponym, “teardrop fracture.” These injuries are often complicated by significant neurologic compromise. The algorithm for treatment is similar to that of burst fractures as these fractures are typically very unstable and require stabilization.
Facet Fracture and Dislocation
These injuries have a wide spectrum of presentation from isolated nondisplaced facet fractures to bilaterally jumped facets with SCI. They occur from a flexion–distraction mechanism often with some element of rotation. Facets may be fractured, subluxed, or dislocated, and this may be unilateral or bilateral (Fig. 23-22). These injuries typically require advanced imaging.
FIGURE 23-22 Flexion–distraction subaxial cervical spine injury (i.e., jumped facets).
As unilateral or bilateral facet injuries progress from subluxation to perched facets and then to dislocation, the extent of malalignment of the cervical spine provides a good indication of the degree of disruption of the facet capsule and posterior ligaments. An associated neurologic deficit can present as a radiculopathy with unilateral facet injuries, while compression of the spinal cord frequently occurs with bilateral injuries. Both unilateral and bilateral facet disruptions should be initially treated with closed traction. If the facet is locked in a dislocated position, open reduction and fusion is indicated as it is an unstable injury. Subluxations and dislocations may benefit from traction with a cranial tong to reduce the deformity followed by operative stabilization. In irreducible situations, posterior approaches to the cervical spine allow access to the facets where direct reduction maneuvers can be performed. There is controversy over whether an MRI is indicated prior to attempts at reduction to determine whether a herniated disk that could potentially retropulse into the canal during reduction is present.76,77 In awake and alert patients able to participate in a neurologic exam, reduction can be attempted immediately without the delay in obtaining an MRI. Rapid intervention without MRI may be of some benefit in patients with SCI as more rapid decompression is associated with better outcomes in animal studies. In intact patients, the delay in getting an MRI is reasonable. Patients who are unable to participate in an exam require an MRI prior to reduction attempts.
Minimally displaced facet fractures often result from a rotational mechanism of injury with variable involvement of the rest of the DLC. Neurologic injury is rare and, when present, is typically in the form of a root level injury. Associated pedicle and laminar fractures are common and can complicate the assessment of stability. Injury of the intervening intervertebral disk is possible and would lead to increased instability. Sometimes the instability is rotational that is difficult to identify on imaging studies, so a careful review of CT and MRI imaging is essential. Nondisplaced facet fractures can be treated in a rigid collar, but must be followed closely for rotational displacement. Displaced fractures require an overall assessment of stability to determine treatment.
These injuries occur from a hyperextension mechanism and are particularly common in the elderly after a fall with impact on their forehead. In the setting of spondylosis this mechanism can result in a central cord syndrome without bony and ligamentous instability. Typically, spinal instability is less of a concern in these injury patterns, and the decision centers on the timing of surgical decompression for patients with neurologic compromise.
Fractures of cervical spinous processes are not always benign and can be associated with significant ligamentous injury. Fracture lines that extend toward the ligamentum flavum and those that occur in association with other fractures should be carefully evaluated with CT and MRI. These studies would rule out potentially unstable injuries that could displace, causing delayed neurologic deficits.
SPECIFIC MANAGEMENT OF INJURIES OF THE THORACIC SPINE (T1–T9)
The thoracic spine is inherently stable due to the rigid structural configuration of the sternum and rib cage. Injuries from a low-energy mechanism (e.g., compression fractures) can occur in patients with osteoporosis, but more commonly occur in the thoracolumbar spine. Compression fractures are uncommon in the upper thoracic spine (i.e., above T6), so their occurrence should raise the concern for a pathologic fracture from metastatic disease. Significant injuries to the thoracic spine usually result from a high-energy mechanism. Although stable injuries rarely involve neurologic compromise, profound neurologic deficits typically occur in the more unstable patterns. This high risk for neurologic injury is due to the small size of the neural canal, the tenuous arterial blood supply to the thoracic spinal cord, and the high energy required to inflict injury. Associated injuries occur in approximately 75% of all patients with injuries to the thoracic spine and can include rib fractures, pulmonary contusions, pneumothoraces, cardiac contusions, or vascular injuries. Associated sternal fractures (especially transverse patterns at the same level as the associated thoracic fracture) cause the entire chest cavity to be unstable. Stable fractures of the thoracic spine with an intact neurologic status can usually be treated nonoperatively. These fractures often consist of compression or burst fractures without significant flexion, rotation, or translation. A brace is rarely needed as the thoracic rib cage functions as an internal brace. Unstable fractures of the thoracic spine typically need operative stabilization since bracing is ineffective in this region. Instrumentation with a posterior pedicle screw is the most commonly employed method to regain stability. Fractures associated with significant ventral compression of the spinal cord may need decompression through an anterior thoracic exposure.
SPECIFIC MANAGEMENT OF INJURIES OF THE THORACOLUMBAR SPINE
Fractures of the thoracolumbar spine are second only to injuries of the cervical spine in frequency. Because this region is the transition between the stiff thoracic spine and the mobile lumbar spine, it is highly susceptible to injury. Injuries in this location include compression fractures, burst fractures, flexion–distraction injuries, and fracture–dislocations. As in the cervical spine, the assessment of instability is essential for designing appropriate treatment strategies. Recently the TLICS score has been validated to assist in this determination.71
Compression fractures involve the anterior column and typically occur as a result of an axial loading mechanism. They are the most frequent of all fractures of the thoracolumbar spine and are especially common in elderly patients with osteoporosis. These fractures are typically stable and are rarely associated with a neurologic deficit. Stable fractures can be treated in a brace, but recent interim analysis from a multicenter, randomized clinical trial suggests that treatment without bracing may have equivalent outcomes. Fractures with greater than 50% loss of anterior body height or more than 30° of angulation are considered unstable and may require surgical stabilization. Compression fractures that displace or become kyphotic may need surgical stabilization, as well.
Burst fractures involve the anterior and middle columns and typically occur through an axial loading mechanism (Fig. 23-23). There can be a component of flexion with the rotation centered on the PLL (the junction between the middle and posterior columns) and, therefore, the PLC can fail in tension. The involved middle column characteristically retropulses bony fragments into the canal that can result in a neurologic deficit. On imaging, burst fractures display loss of height of both the anterior and middle columns. They may have an associated vertical laminar fracture from splaying of the posterior elements, and this fracture can be associated with a dural tear and/or entrapment of a nerve root. Retropulsed bony fragments from the middle column may not be evident on plain x-rays, so CT imaging is necessary for full evaluation. Radiographic criteria for a stable burst fracture include less than 50% decrease in body height, less than 25° of angulation, and less than 50% compromise of the spinal canal. Burst fractures with greater than 50% decrease of body height, greater than 25° of angulation, and greater than 50% compromise of the spinal canal and those associated with neurologic deficit are potentially unstable.
FIGURE 23-23 Burst fracture of the thoracolumbar spine.
The most important criterion in determining if a burst fracture requires surgery is the neurologic status of the patient. Stable fractures that occur in neurologically intact patients can be treated nonoperatively even in the presence of significant compromise of the canal because the retropulsed fragments resorb over time.91 Historically, treatment consisted of immobilization with a TLSO with close follow-up to ensure maintenance of spinal alignment. Bailey et al.73 recently found that in adults less than 60 years with no neurologic deficit and kyphosis <35°, thoracolumbar burst fractures could be successfully treated without bracing. With unstable burst fractures or in the presence of significant neurologic deficits, surgery (instrumented posterior spinal fusion) is indicated. If the neural elements are not able to be adequately decompressed with indirect reduction methods (i.e., fracture reduction), direct decompression is necessary. The approach to directly decompressing the neurologic structures depends on the level of the fracture. At the level of the spinal cord and conus medullaris (T10–L1 or L2), retraction of the neural elements is risky. Therefore, an anterolateral approach is preferred for direct decompression of the retropulsed fragments to relieve compression of the ventral thecal sac. Below the level of the conus, the thecal sac can be retracted allowing the protruding bone fragments to be tamped ventrally through a posterior transpedicular approach.
Flexion–distraction injuries, also known as Chance injuries and seatbelt injuries, involve the middle and posterior columns. The center of rotation for these injuries is at the anterior longitudinal ligament or anterior to the vertebral body. Therefore, the posterior elements (i.e., posterior column) fail in tension. Consequently, care must be taken to avoid traction/distraction maneuvers during the initial stabilization of this injury. Because of the mechanism of injury, abdominal viscera may be injured. On imaging, these fractures can look similar to burst fractures, but they are distinguished by a middle column that has failed in tension resulting in a gain in height of the posterior aspect of the vertebral body (i.e., middle column). Associated laminar fractures are horizontal in contrast to the vertical fracture orientation in burst fractures. These spinal column injuries can fail through bone, through the soft tissues (intervertebral disk and the PLC), or through a combination of the two. Consequently, CT and MRI are necessary to evaluate the degree of involvement of osseous and soft tissue. Closed reduction and bracing is appropriate in those injuries involving bone with less than 30° angulation because fractures tend to heal with appropriate bony apposition. Surgical reduction and fixation is indicated for flexion–distraction injuries that are primarily ligamentous (i.e., with poor healing potential) or have more than 30° angulation.
Fracture–dislocations are extremely unstable injuries that involve all three columns of the spine (Fig. 23-23 and Fig. 23-24A, B). These occur through some combination of rotation, flexion, and translation and usually result in a neurologic deficit, of which half are complete lesions. Nonoperative treatment is never appropriate. Incomplete injuries benefit from early surgical reduction, decompression, and stabilization. Complete injuries or intact patients will require surgical stabilization to facilitate early mobilization and rehabilitation.
FIGURE 23-24 (A and B) Fracture–dislocation of the thoracolumbar spine.
Isolated injuries to posterior bony elements (e.g., lamina fractures, avulsions of spinous process tips, transverse process fractures) are typically stable injuries restricted to the posterior column of the spine with a few exceptions. Fractures of the L5 transverse processes are typically nonoperative spinal injuries; however, these can be associated with pelvic ring disruptions and warrant dedicated pelvic imaging.
SPECIFIC MANAGEMENT OF INJURIES OF THE SACRUM
Sacral trauma comprises a complex constellation of injuries that includes associated disruptions of the pelvic ring and injuries to the nerve roots, cauda equina, and the spinal segments. These can be divided into injuries resulting from disruption of the pelvic ring, injuries intrinsic to the sacrum, and injuries involving the lumbosacral articulation. In general, the same principles previously described with regard to management of spinal instability and neurologic deficits apply to the management of sacral injuries. Restoration of spinal alignment and stability and neurologic decompression remain the treatment goals. There are, however, several additional considerations. Associated hemodynamic instability merits consideration for angiographic embolization of bleeding vessels in the pelvis. Also, it is important to rule out open fractures, especially in significantly displaced patterns, by rectal and vaginal examination. Finally, evaluation of pelvic ring stability is necessary, though this is typically done in conjunction with a consulting orthopedic surgeon familiar with the evaluation and treatment of pelvic ring disruptions.
Intrinsic sacral fractures rarely compromise the stability of the sacrum or the lumbosacral articulations. They can be associated with neurologic deficits and occasionally require surgical decompression. The Denis classification system classifies intrinsic sacral fractures based on the relationship of the fracture line to the sacral neuroforamina and is useful for predicting neurologic deficits.92 Zone I fractures are lateral to the foramina and have a 6% chance of root involvement. Zone II fractures enter the foramina and have a 28% chance of root involvement. Because the upper sacral roots are larger, there is less room in their respective foramina and they are more prone to injury than the lower sacral roots. Type III fractures enter the central canal and are associated with a risk approaching 60%. Management of these injuries in the absence of an associated disruption of the pelvic ring is typically conservative and includes some degree of limitations on weight bearing. Associated disruptions of the pelvic ring require evaluation and treatment under the purview of a consulting orthopedic surgeon.
Certain patterns of sacral injury can lead to instability. Vertically oriented transforaminal sacral fractures that extend proximally through or medial to the articular process of S1 can result in instability of the lumbosacral junction.93Therefore, displaced transforaminal sacral fractures warrant close scrutiny of the lumbosacral junction with appropriate CT imaging. These injuries can be surgically stabilized (i.e., L5–S1 fusion) to minimize the sequelae of residual incongruence of the facets. Lumbosacral and lumbopelvic dissociation are rare and often fatal injuries that involve dissociation of the spine from the pelvis94(Fig. 23-25). This can result from bilateral L5–S1 facet fracture/dislocation or a fracture line crossing the sacrum (e.g., U-type sacral fracture). These injuries can be missed as the patient is often in extremis. Surgical options include percutaneous screw techniques and lumbopelvic fixation. The latter is a stronger construct, but is a more invasive procedure.
FIGURE 23-25 CT of lumbosacral dissociation injury.
GUNSHOT WOUNDS TO THE SPINE
The incidence of civilian GSWs to the spine has been steadily increasing and is the second leading cause of SCI in many urban areas. As the wounding capacity of a GSW relates to its kinetic energy , the muzzle velocity has an exponential effect on the energy of the injury (see Chapter 1).
Low-energy bullets travel less than 1,000–2,000 ft/s (e.g., handguns), while high-energy bullets travel at speeds greater than 2,000 ft/s (e.g., military assault rifles). Damage is created by several mechanisms, including the actual passage of the missile through tissue, a secondary shock wave, and cavitation. At impact, the bullet creates a temporary cavity at the entry site due to stretching forces and the vacuum created by its passing. The volume of this cavity and, consequently, the extent of tissue injury are proportional to the energy transferred by the missile. Neurologic injury can occur not only from direct contact but also from close passage of the bullet to the neurologic elements. Because of the low kinetic energy involved, civilian GSWs to the spine typically must traverse or remain inside the spinal canal to cause a neurologic deficit.
All GSWs to the spine should be evaluated according to a routine management algorithm. A patient’s general medical condition should be addressed and stabilized prior to focusing on the spine. A detailed neurologic examination should be performed. Gunshot injuries to the neck may mandate routine angiography and panendoscopy to identify vascular and visceral injuries. Gunshot injuries to the thoracic spine may be associated with pulmonary or cardiac injuries, and those to the lumbar spine may involve abdominal viscera, genitourinary structures, or a major vessel. In particular, colonic perforations that occur before the bullet passes through the spine must be recognized because they are associated with an increased risk of infection if not appropriately treated with a course of broad-spectrum antibiotics. The spine should be routinely imaged with plain x-rays and CT to localize the level of injury and establish the extent of bony disruption. MRI is important to rule out an epidural hematoma as a cause for the neurologic deficit. Because most bullets are not ferromagnetic, MRI is rarely contraindicated except when the bullet is in the canal. Dynamic flexion–extension stress x-rays are rarely indicated in the acute setting, but can be considered in awake and alert patients or in a patient with suspected instability.
Tetanus prophylaxis should be a consideration in all patients with gunshot injuries, and 24 hours of antibiotic coverage is prudent with the duration dictated by the extent of injury to soft tissue. Spinal injuries from transabdominal bullets that perforate a viscus prior to entering the spine have been associated with rates of spinal infection as high as 88% in the absence of appropriate antibiotic coverage.95 These patients should receive between 7 and 14 days of broad-spectrum antibiotic coverage, and surgical debridement has not been shown to be of benefit.
The potential for spinal instability following low-energy GSWs is modest, and these patients rarely require bracing. The majority of reported cases of instability are iatrogenic resulting from ill-advised decompression.96 Isiklar and Lindsey97 reported on late spinal instability among patients with gunshot injuries treated nonoperatively. Of the instability cases identified, 75% occurred in the cervical spine. Therefore, the potential for spinal instability exists with a GSW and should always be considered, especially in the cervical spine.
Neurologic recovery is usually limited after a GSW to the spine resulting in an SCI. In these situations, there is no role for steroids. Surgical decompression of intracanal bullets may result in root level return in the cervical or lumbar spine, but there is no benefit in the thoracic spine.98 The increased risk of infection and cerebral spinal fluid fistula must be weighed against the potential benefit.96 Probably the strongest argument for removal of an intracanal bullet in these regions would be in the setting of an incomplete injury in the cervical spinal cord where the possibility of recovering a neurologic level may justify the increased risks of the procedure.
The most devastating complication in patients with spinal trauma is neurologic deterioration. There are a variety of reasons a patient may decompensate neurologically. Evolution of a cranial bleed is an important consideration in patients with traumatic brain injuries. Inappropriate manipulation of the unstable spine during transfers can result in an iatrogenic injury, also. Intraoperatively, dural tears can expose sensitive neural tissues to inadvertent injury, and placement of hardware risks injury to the nerves and spinal cord. Epidural hematomas in preoperative and postoperative patients can expand, especially when pharmacologic anticoagulation is being used. Serial neurologic examinations in the preoperative and postoperative patients and neurophysiologic monitoring in the intraoperative setting may allow for the timely identification of neurologic deterioration and allow expedited intervention. Patients experiencing neurologic deterioration should be given supplemental oxygen to ensure adequate oxygenation and should have their hemodynamic parameters optimized. Prompt repeat imaging (i.e., MRI, CT, or CT myelogram) is then indicated to determine the etiology. Operative intervention may be indicated in the presence of a compressive lesion.
Dural tears and leakage of cerebrospinal fluid are common complications of spinal trauma. Tears may occur at the time of injury and are frequently associated with laminar fractures. Alternatively, they can result during surgery. Significant leakage of cerebrospinal fluid in the patient with no plans for surgery in the near term may benefit from placement of a subarachnoid drain. In patients undergoing surgery, an attempt should be made to obtain a watertight seal of the durotomy through direct suture repair. If this is not possible, the dural defect may be patched with local fat, fascia, or synthetic patches with adjunctive use of dural sealants. For large, irreparable tears, a subarachnoid drain is required.
Postoperative surgical site infections (SSIs) are unfortunately common in patients undergoing spinal surgery after trauma with an incidence of approximately 8%. This is higher than quoted rates for elective spinal surgery (1–2%) because of a variety of factors. These include injury to other organ systems, traumatized soft tissues, higher incidence of dural tears with leakage of cerebrospinal fluid, poor nutrition, and the lower socioeconomic strata of affected patients.99 In patients undergoing elective spinal surgery, Olsen et al.100 found diabetes mellitus was associated with the highest independent risk of SSI. Preoperative or postoperative hyperglycemia and obesity were other independently associated risk factors. Administration of prophylactic antibiotics within 1 hour before the skin incision and adjustment in antibiotic dosages for obesity are thought to be important strategies to decrease the risk of SSI.
1. Hu R, Mustard CA, Burns C. Epidemiology of incident spinal fracture in a complete population. Spine (Phila Pa 1976). 1996;21(4):492–499.
2. Holmes J, Miller P, Panacek E, et al. Epidemiology of thoracolumbar spine injury in blunt trauma. Acad Emerg Med. 2001;8:866.
3. Lowery DW, Wald MM, Browne BJ, et al. Epidemiology of cervical spine injury victims. Ann Emerg Med. 2001;38:12.
4. Jackson AB, Dijkers M, Devivo MJ, et al. A demographic profile of new traumatic spinal cord injuries: change and stability over 30 years. Arch Phys Med Rehabil. 2004;85:1740–1748.
5. National Spinal Cord Injury Statistical Center. Spinal Cord Injury Facts and Figures at a Glance. Available at: http://www.spinalcord.uab.edu. Accessed March 9, 2010.
6. Dryden DM, Saunders LD, Rowe BH, et al. The epidemiology of traumatic spinal cord injury in Alberta, Canada. Can J Neurol Sci. 2003;30:113–121.
7. Kannus P, Niemi S, Palvanen M, et al. Continuously increasing number and incidence of fall-induced, fracture-associated, spinal cord injuries in elderly persons. Arch Intern Med. 2000;160:2145–2149.
8. Torg JS, Guille JT, Jaffe S. Injuries to the cervical spine in American football players. J Bone Joint Surg Am. 2002;84-A(1):112–122.
9. Jellinger K. Pathology of spinal cord trauma. In: Errico TJ, Bauer RD, Waugh T, eds. Spinal Trauma. Baltimore, MD: Lippincott; 1990:455.
10. Anderson DK, Hall ED. Pathophysiology of spinal cord trauma. Ann Emerg Med. 1993;22:987.
11. Tempel GE, Martin HF. The beneficial effects of a thromboxane receptor antagonist on spinal cord perfusion following experimental cord injury. J Neurol Sci. 1992;109:162.
12. Liu D, Thangnipon W, McAdoo DJ. Excitatory amino acids rise to toxic levels upon impact injury to the rat spinal cord. Brain Res. 1991;547:344.
13. Wrathall JR, Teng YD, Choiniere D, et al. Evidence that local non-NMDA receptors contribute to functional deficits in contusive spinal cord injury. Brain Res. 1992;586:140.
14. Young W, Koren I. Potassium and calcium changes in injured spinal cords. Brain Res. 1986;365:42.
15. Gentile NT, McIntosh TK. Antagonists of excitatory amino acids and endogenous opioid peptides in the treatment of experimental central nervous system injury. Ann Emerg Med. 1993;22:1028.
16. American Spinal Injury Association. Standard Neurological Classification of Spinal Cord Injury Worksheet. Atlanta, GA: American Spinal Injury Association; 2006.
17. Bohlman HH. Acute fractures and dislocations of the cervical spine. An analysis of three hundred hospitalized patients and review of the literature. J Bone Joint Surg Am. 1979;61(8):1119–1142.
18. Stauffer ES. Neurologic recovery following injuries to the cervical spinal cord and nerve roots. Spine (Phila Pa 1976). 1984;9:532–534.
19. Toscano J. Prevention of neurological deterioration before admission to a spinal cord injury unit. Paraplegia. 1988;26:143–150.
20. Rechtine GR, Del Rossi G, Conrad BP, Horodyski M. Motion generated in the unstable spine during hospital bed transfers. J Trauma. 2004;57: 609–611.
21. Hauswald M, Braude D. Spinal immobilization in trauma patients: is it really necessary? Curr Opin Crit Care. 2002;8:566–570.
22. Levi AD, Hurlbert J, Anderson PA, et al. Neurologic deterioration secondary to unrecognized spinal instability following trauma—a multicenter study. Spine. 2006;31:451–458.
23. Vaccaro AR, An HS, Lin S, et al. Noncontiguous injuries of the spine. J Spinal Disord. 1992;5:320–329.
24. Brown CV, Antevil JL, Sise MJ, et al. Spiral computed tomography for the diagnosis of cervical, thoracic, and lumbar spine fractures: its time has come. J Trauma. 2005;58:890–895.
25. McCulloch PT, France J, Jones DL, et al. Helical computed tomography alone compared with plain radiographs with adjunct computed tomography to evaluate the cervical spine after high-energy trauma. J Bone Joint Surg Am. 2005;87:2388–2394.
26. Anderson PA, Gugala Z, Lindsey RW, et al. Clearing the cervical spine in the blunt trauma patient. J Am Acad Orthop Surg. 2010;18(3):149–159.
27. Stiell IG, Clement CM, McKnight RD, et al. The Canadian C-spine rule versus the NEXUS low-risk criteria in patients with trauma. N Engl J Med. 2003;349:2510–2518.
28. Harris TJ, Blackmore CC, Mirza SK, et al. Clearing the cervical spine in obtunded patients. Spine (Phila Pa 1976). 2008;33:1547–1553.
29. Tomycz ND, Chew BG, Chang YF, et al. MRI is unnecessary to clear the cervical spine in obtunded/comatose trauma patients: the four-year experience of a level I trauma center. J Trauma. 2008;64:1258–1263.
30. Stassen NA, Williams VA, Gestring ML, et al. Magnetic resonance imaging in combination with helical computed tomography provides a safe and efficient method of cervical spine clearance in the obtunded trauma patient. J Trauma. 2006;60:171–177.
31. Muchow RD, Resnick DK, Abdel MP, et al. Magnetic resonance imaging (MRI) in the clearance of the cervical spine in blunt trauma: a meta-analysis. J Trauma. 2008;64:179–189.
32. Meissner MH, Chandler WL, Elliott JS. Venous thromboembolism in trauma: a local manifestation of systemic hypercoagulability? J Trauma. 2003;54(2):224–231.
33. Knudson MM, Ikossi DG, Khaw L, et al. Thromboembolism after trauma: an analysis of 1602 episodes from the American College of Surgeons National Trauma Data Bank. Ann Surg. 2004;240(3):490–496 [discussion 496–498].
34. Dai LY, Yao WF, Cui YM, et al. Thoracolumbar fractures in patients with multiple injuries: diagnosis and treatment—a review of 147 cases. J Trauma. 2004;56(2):348–355.
35. Geerts WH, Code KI, Jay RM, et al. A prospective study of venous thromboembolism after major trauma. N Engl J Med. 1994;331(24): 1601–1606.
36. Jones T, Ugalde V, Franks P, et al. Venous thromboembolism after spinal cord injury: incidence, time course, and associated risk factors in 16,240 adults and children. Arch Phys Med Rehabil. 2005;86(12):2240–2247.
37. Ploumis A, Ponnappan RK, Maltenfort MG, et al. Thromboprophylaxis in patients with acute spinal injuries: an evidence-based analysis. J Bone Joint Surg Am. 2009;91(11):2568–2576.
38. Curry K, Casady L. The relationship between extended periods of immobility and decubitus ulcer formation in the acutely spinal cord-injured individual. J Neurosci Nurs. 1992;24:185–189.
39. Hadley MN, Walters BC, Grabb PA, et al. Blood pressure management after acute spinal cord injury. Neurosurgery. 2002;50(suppl):58–62.
40. 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. JAMA. 1997;277:1597–1604.
41. Coleman WP, Benzel E, Cahill DW, et al. A critical appraisal of the reporting of the National Acute Spinal Cord Injury Studies (II and III) of methylprednisolone in acute spinal cord injury. J Spinal Disord. 2000;13:185–199.
42. Hurlbert RJ. Methylprednisolone for acute spinal cord injury: an inappropriate standard of care. J Neurosurg. 2000;93:1–7.
43. Short DJ, El Masry WS, Jones PW. High dose methylprednisolone in the management of acute spinal cord injury: a systematic review from a clinical perspective. Spinal Cord. 2000;38:273–286.
44. Hugenholtz H, Cass DE, Dvorak MF, et al. High-dose methylprednisolone for acute closed spinal cord injury—only a treatment option. Can J Neurol Sci. 2002;29(3):227–235.
45. Cappuccino A, Bisson LJ, Carpenter B, et al. The use of systemic hypothermia for the treatment of an acute cervical spinal cord injury in a professional football player. Spine (Phila Pa 1976). 2010;35(2): E57–E62.
46. Zemke D, Majid A. The potential of minocycline for neuroprotection in human neurologic disease. Clin Neuropharmacol. 2004;27:293.
47. Stirling DP, Khodarahmi K, Liu J, et al. Minocycline treatment reduces delayed oligodendrocyte death, attenuates axonal dieback, and improves functional outcome after spinal cord injury. J Neurosci. 2004;24:2182.
48. Gorio A, Gokmen N, Erbayraktar S, et al. Recombinant human erythropoietin counteracts secondary injury and markedly enhances neurological recovery from experimental spinal cord trauma. Proc Natl Acad Sci U S A. 2002;99:9450.
49. Li S, Liu BP, Budel S, et al. Blockade of Nogo-66, myelin-associated glycoprotein, and oligodendrocyte myelin glycoprotein by soluble Nogo-66 receptor promotes axonal sprouting and recovery after spinal injury. J Neurosci. 2004;24:10511.
50. Qiu J, Cai D, Dai H, et al. Spinal axon regeneration induced by elevation of cyclic AMP. Neuron. 2002;34:895.
51. Kwon BK, Fisher CG, Dvorak MF, et al. Strategies to promote neural repair and regeneration after spinal cord injury. Spine. 2005;30:S3.
52. Cheng H, Cao Y, Olson L. Spinal cord repair in adult paraplegic rats: partial restoration of hind limb function. Science. 1996;273:510.
53. Ruitenberg MJ, Vukovic J, Sarich J, et al. Olfactory ensheathing cells: characteristics, genetic engineering, and therapeutic potential. J Neurotrauma. 2006;23:468.
54. Barnett SC, Riddell JS. Olfactory ensheathing cells (OECs) and the treatment of CNS injury: advantages and possible caveats. J Anat. 2004;204:57.
55. Camargo FD, Chambers SM, Goodell MA. Stem cell plasticity from transdifferentiation to macrophage fusion. Cell Prolif. 2004;37:55.
56. Sykova E, Jendelova P, Glogarova K, et al. Bone marrow stromal cells—a promising tool for therapy of brain and spinal cord injury. Exp Neurol. 2004;187:220.
57. Delamarter RB, Sherman J, Carr JB. Pathophysiology of spinal cord injury: recovery after immediate and delayed compression. J Bone Joint Surg Am. 1995;77:1042–1049.
58. Carlson GD, Gorden CD, Oliff HS, et al. Sustained spinal cord compression: part I: time-dependent effect on long-term pathophysiology. J Bone Joint Surg Am. 2003;85:86–94.
59. La Rosa G, Conti A, Cardali S, et al. Does early decompression improve neurological outcome of spinal cord injured patients? Appraisal of the literature using a meta-analytical approach. Spinal Cord. 2004;42: 503–512.
60. Papadopoulos SM, Selden NR, Quint DJ, et al. Immediate spinal cord decompression for cervical spinal cord injury: feasibility and outcome. J Trauma. 2002;52:323–332.
61. Vaccaro AR, Daugherty RJ, Sheehan TP, et al. Neurologic outcome of early versus late surgery for cervical spinal cord injury. Spine (Phila Pa 1976). 1997;22(22):2609–2613.
62. McKinley W, Meade MA, Kirshblum S, et al. Outcomes of early surgical management versus late or no surgical intervention after acute spinal cord injury. Arch Phys Med Rehabil. 2004;85:1818–1825.
63. Fehlings MG, Perrin RG. The timing of surgical intervention in the treatment of spinal cord injury: a systematic review of recent clinical evidence. Spine (Phila Pa 1976). 2006;31(11 suppl):S28–S35 [discussion S36].
64. White AA, Panjabi MM. The problem of clinical instability in the human spine: a systematic approach. In: White AA, Panjabi MM, eds. Clinical Biomechanics of the Spine. 2nd ed. Philadelphia: Lippincott Williams & Wilkins; 1990:277–378.
65. Allen BL, Ferguson RL, Lehmann TR, et al. A mechanistic classification of closed, indirect fractures and dislocations of the lower cervical spine. Spine. 1982;7:1.
66. Vaccaro AR, Hulbert RJ, Patel AA, et al. Spine Trauma Study Group. The subaxial cervical spine injury classification system: a novel approach to recognize the importance of morphology, neurology, and integrity of the disco-ligamentous complex. Spine (Phila Pa 1976). 2007;32(21): 2365–2374.
67. Holdsworth F. Fractures, dislocations, and fracture–dislocations of the spine. J Bone Joint Surg. 1970;52A:1534.
68. Denis F. The three column spine and its significance in the classification of acute thoracolumbar spinal injuries. Spine (Phila Pa 1976). 1983;8(8):817–831.
69. Panjabi MM, Oxland TR, Kifune M, et al. Validity of the three-column theory of thoracolumbar fractures: a biomechanic investigation. Spine. 1995;20:1122.
70. Magerl F, Aebi M, Gertzbein SD, et al. A comprehensive classification of thoracic and lumbar injuries. Eur Spine J. 1994;3:184.
71. Vaccaro AR, Lehman RA Jr, Hurlbert RJ, et al. A new classification of thoracolumbar injuries: the importance of injury morphology, the integrity of the posterior ligamentous complex, and neurologic status. Spine (Phila Pa 1976). 2005;30(20):2325–2333.
72. Anderson DG, Vaccaro AR, Gavin KF. Cervical orthoses and cranioskeletal traction. In: Clark CR, ed. The Cervical Spine. 4th ed. Philadelphia: Lippincott Williams & Wilkins; 2005:110–121 [Table 7.1].
73. Bailey CS, Dvorak MF, Thomas KC, et al. Comparison of thoracolumbosacral orthosis and no orthosis for the treatment of thoracolumbar burst fractures: interim analysis of a multicenter randomized clinical equivalence trial. J Neurosurg Spine. 2009;11(3):295–303.
74. Johnson RM, Hart DL, Simmons EF, et al. Cervical orthoses: a study comparing their effectiveness in restricting cervical motion in normal subjects. J Bone Joint Surg Am. 1977;59:332–339.
75. Majercik S, Tashijan RZ, Biffl WL, et al. Halo vest immobilization in the elderly: a death sentence? J Trauma. 2005;59:350–356.
76. Eismont FJ, Arena MJ, Green BA. Extrusion of an intervertebral disc associated with traumatic subluxation or dislocation of cervical facets. Case report. J Bone Joint Surg. 1991;73A:1555.
77. Grauer JN, Vaccaro AR, Lee JY, et al. The timing and influence of MRI on the management of patients with cervical facet dislocations remains highly variable: a survey of members of the Spine Trauma Study Group. J Spinal Disord Tech. 2009;22(2):96–99.
78. Hu SS, Capen DA, Rimoldi RL, et al. The effect of surgical decompression on neurologic outcome after lumbar fracture. Clin Orthop. 1993;288:166.
79. Fellrath RF, Hanley EN. Multitrauma and thoracolumbar fractures. Semin Spine Surg. 1995;7:103.
80. Ahn JH, Ragnarsson KT, Gordon WA, et al. Current trends in stabilizing high thoracic and thoracolumbar spinal fractures. Arch Phys Med Rehabil. 1984;65:366–369.
81. Willen J, Lindahl S, Irstam L, et al. Unstable thoracolumbar fractures: a study by CT and conventional roentgenology of the reduction effect of Harrington instrumentation. Spine. 1984;9:215.
82. Yazici M, Gulman B, Sen S, et al. Sagittal contour restoration and canal clearance in burst fractures of the thoracolumbar junction (T12–L1): the efficacy of timing of the surgery. J Orthop Trauma. 1995;9(6):491–498.
83. Campagnolo DI, Esquieres RE, Kopacz KJ. Effect of stabilization on length of stay and medical complications following spinal cord injury. J Spinal Cord Med. 1997;20:331.
84. Anderson PA, Montesano PX. Morphology and treatment of occipital condyle fractures. Spine. 1988;13:731–736.
85. Harris JH Jr, Carson GC, Wagner LK, et al. Radiologic diagnosis of traumatic occipitovertebral dissociation: 2. Comparison of three methods of detecting occipitovertebral relationships on lateral radiographs of supine subjects. AJR Am J Roentgenol. 1994;162:887–892.
86. Traynelis VC, Marano GD, Dunker RO, et al. Traumatic atlantooccipital dislocation: case report. J Neurosurg. 1986;65:863–870.
87. Spence KF Jr, Decker S, Sell KW. Bursting atlantal fracture associated with rupture of the transverse ligament. J Bone Joint Surg Am. 1970;52:543–549.
88. Anderson LD, D’Alonzo RT. Fractures of the odontoid process of the axis. J Bone Joint Surg Am. 1974;56:1663–1674.
89. Levine AM, Edwards CC. The management of traumatic spondylolisthesis of the axis. J Bone Joint Surg Am. 1985;67(2):217–226.
90. Starr JK, Eismont FJ. Atypical Hangman’s fractures. Spine (Phila Pa 1976). 1993;18(14):1954–1957.
91. Wood K, Buttermann G, Mehbod A, et al. Operative compared with nonoperative treatment of a thoracolumbar burst fracture without neurological deficit. A prospective, randomized study. J Bone Joint Surg Am. 2003;85-A(5):773–781 [Erratum in J Bone Joint Surg Am. 2004; 86-A(6):1283].
92. Denis F, Davis S, Comfort T. Sacral fractures: an important problem. Retrospective analysis of 236 cases. Clin Orthop Relat Res. 1988;227: 67–81.
93. Isler B. Lumbosacral lesions associated with pelvic ring injuries. J Orthop Trauma. 1990;4:1–6.
94. Vialle R, Wolff S, Pauthier F, et al. Traumatic lumbosacral dislocation: four cases and review of literature. Clin Orthop Relat Res. 2004;419: 91–97.
95. Romanick P, Smith T, Kopaniky D, et al. Infection about the spine associated with low-velocity missile injury to the abdomen. J Bone Joint Surg. 1985;67:1195–1201.
96. Stauffer ES, Wood RW, Kelly EG. Gunshot wounds of the spine: the effects of laminectomy. J Bone Joint Surg. 1979;61A:389.
97. Isiklar ZU, Lindsey RW. Low velocity civilian gunshot wounds of the spine. Orthopedics. 1997;20:967.
98. Waters R, Adkins R. The effects of removal of bullet fragment retained in the spinal canal. A collaborative study by the National Spinal Cord Injury Model Systems. Spine. 1991;16:934–939.
99. Lim MR, Lee JY, Vaccaro AR. Surgical infections in the traumatized spine. Clin Orthop Relat Res. 2006;444:114–119.
100. Olsen MA, Nepple JJ, Riew KD, et al. Risk factors for surgical site infection following orthopaedic spinal operations. J Bone Joint Surg Am. 2008;90(1):62–69.