The cervical spine is the most mobile area of the spine, and as such, it is prone to the greatest number of injuries. Injuries to the cervical spine and spinal cord are also potentially the most devastating and life altering of all injuries compatible with life. In the United States, approximately 10,000 spinal cord injuries occur each year (about 39 per million). An estimated 80% of the victims are younger than 40 years, with the highest proportion of injuries reported in those between 15 and 35 years of age. Approximately 80% of all people who suffer from spinal column injuries are male. Falls account for 60% of injuries to the vertebral column in patients older than 75 years. In younger patients, 45% of injuries result from motor vehicle accidents, 20% from falls, 15% from sports injuries, 15% from acts of violence, and the remainder from other causes.
With the use of seat belts and air bags in motor vehicles and the advent of trauma centers and improved emergency service awareness of potential cervical injuries, fewer patients with cervical spine injuries are dying secondary to respiratory complications. The approach in treating these patients is early recognition of cervical spine injuries with rapid immobilization to prevent neurologic deterioration while the evaluation and treatment of associated injuries are carried out. After the patient is stabilized, the goals are restoration and maintenance of spinal alignment to provide stable weight bearing and facilitate rehabilitation.
Identification and Stabilization of Life-Threatening Injuries
Eighty-five percent of all neck injuries requiring medical evaluation are a result of a motor vehicle accident. Many of the affected patients are multiple-trauma victims and therefore may have more urgent life-threatening conditions. The ABCs of trauma are followed in order of priority, with airway, breathing (ventilation), and circulation secured before further evaluation proceeds. Throughout the evaluation of other body systems, the cervical spine should be presumed injured and thus immobilized. Approximately 20% of patients with cervical trauma are hypotensive upon presentation. The hypotension is neurogenic in origin in approximately 70% of cases and related to hypovolemia in 30%. Concomitant bradycardia is suggestive of a neurogenic component. Another finding suggestive of cervical spine injury is an altered sensorium secondary to head trauma or lacerations and facial fractures. Appropriate diagnosis and fluid management are critical in the early hours of postinjury management. After all life-threatening injuries are identified and stabilized, the secondary evaluation, including an extremity examination and neurologic examination, can be safely carried out.
History and General Physical Examination
Details of the history of the injury should be obtained. If the patient is conscious, much of the information can be obtained directly. If not, family members or witnesses of the injury should be questioned. In the case of a motor vehicle accident, for example, pertinent questions include the following: Which part of the patient's body was the point of impact? Was the patient thrown from the car? Was there head trauma or a loss of consciousness? Were there any transient signs of paresis? Was the patient able to move any of his or her extremities at any time following the accident and before loss of function? What were the speeds of the involved motor vehicles? Was the patient restrained with a seat belt? Did an air bag deploy?
The history taken from the patient or family members should also include information about preexisting conditions such as epilepsy or seizures and about preexisting injuries. If the patient had any previous radiographic examinations, the radiographs might be useful for comparison.
It is helpful to question patients about what they are experiencing at the time of the examination. Are there areas of numbness, paresthesia, or pain? Can they move their extremities? The examiner should then proceed with the physical evaluation, beginning by observing the face and head of the patient for any areas of potential injury and attempting to determine the potential mechanism of injury. For instance, any lacerations or contusions to the forehead might indicate a hyperextension-type injury because as the head hits the windshield and stops, the body continues to move forward from the momentum of the impact. Observation should next include watching the extremities for any signs of motion. A genital examination should be performed because a sustained penile erection may be indicative of severe spinal cord injury. Then without moving the patient, palpation can be performed. Although palpation can be helpful in identifying potential levels of injury of the spine, it should not be used as the sole screening examination because false-negative results are possible.
A meticulous neurologic examination should be performed following the history and general physical examination.
The neurologic evaluation should start with documentation of the function of the cranial nerves, working proximally to distally. Observation is particularly important in the unconscious patient. Spontaneous motion in an extremity may be a sole source of information regarding spinal cord function. Respiratory efforts made with intrathoracic musculature versus abdominal musculature are also significant. In the conscious patient who is able to follow commands, a motor examination should be fairly straightforward. Rectal and perianal sensations should be documented because these may be the sole signs of intact distal spinal cord function.
An extensive sensory examination should also be performed with careful attention to dermatomal innervation. In the acute setting, it is useful to document sharp and dull sensations as well as proprioception. Sharp and dull sensations are carried via the lateral spinothalamic tract, whereas proprioception is carried through the posterior columns. Sharp and dull sensations are effectively tested with the sharp and blunt ends of a pen, and proprioception is tested by having the patient verify the position of the large toe and other joints as the examiner places them in dorsiflexion and plantarflexion. It proves helpful to make ink markings directly on the patient's skin to show the level of the dermatomal deficit, which decreases the chance for intraobserver or interobserver error over sequential examinations.
Reflexes should be checked bilaterally. In the upper extremity, the biceps reflex at the flexor side of the elbow evaluates the C5 nerve root, and the brachioradialis stretch reflex at the radial aspect of the forearm just proximal to the wrist checks the C6 nerve root. The triceps reflex is innervated by C7. In the lower extremity, the knee jerk reflex is innervated by L4, and the ankle jerk is innervated by S1.
The presence or absence of the four reflexes listed in Table 4–8 should be checked. The Babinski reflex (plantar reflex) is evaluated by firmly stroking the lateral plantar aspect of the foot distally and then medially over the metatarsal heads and then observing the toes. If the toes flex, the response is considered negative (normal). If the toes extend and spread, the response is considered positive (abnormal) and indicative of an upper motor neuron lesion. The bulbocavernosus reflex has its root in the S3 and S4 nerves and is evaluated by squeezing on the glans in a male patient or applying pressure to the clitoris in a female patient. This action should elicit a contraction of the anal sphincter. If a Foley catheter is in place, simply pulling on the Foley catheter can stimulate the anal sphincter contraction. The cremasteric reflex is evaluated by stroking the inner thigh and observing the scrotal sac, which should retract upward secondary to contraction of the cremasteric muscle. This function is innervated by T12 and L1. Finally, the anal wink, innervated by S2, S3, and S4, is elicited by stimulating the skin about the anal sphincter and eliciting a contraction.
Table 4–8. Evaluation of Reflexes in Patients with Injuries of the Cervical Spine. ||Download (.pdf)
Table 4–8. Evaluation of Reflexes in Patients with Injuries of the Cervical Spine.
Upper motor neurons
Extension and spread of toes
Upper motor neuron lesion is present
S3 and S4
Contraction of anal sphincter
Spinal shock is over
T12 and L1
Retraction of scrotal sac
Spinal shock is over
S2, S3, and S4
Contraction of anal sphincter
Spinal shock is over
The presence of spinal shock causes the absence of all reflexes below the level of injury and typically lasts up to 24 hours after the injury. The bulbocavernosus reflex is the reflex that returns first (see Table 4–8), thus marking the end of spinal shock. This point has prognostic importance because recovery from a complete neurologic deficit that is still present at the end of spinal shock is extremely unlikely. A complete neurologic examination should be repeated over time as the patient is manipulated and treated.
The ability to interpret the results of a patient's neurologic examination appropriately depends on a thorough knowledge of the anatomy of the spinal cord and peripheral nerves.
Peripheral nerves are a combination of afferent fibers, which carry information from the periphery to the central nervous system, and efferent fibers, which carry information away from the central nervous system. As the peripheral nerve approaches the spinal cord, it becomes known as the spinal nerve. Prior to entering the spinal cord, the fiber splits, with the afferent fibers becoming the dorsal root or sensory root and the efferent fibers becoming the ventral root. The afferent fibers are often regrouped in various plexuses that are located between the spinal cord and the periphery. This regrouping takes place before the fibers enter the dorsal root, therefore leading to significant overlap between the dorsal root and the respective dermatomes. The implications of this anatomic fact should be kept in mind by the clinician when performing a sensory examination. For example, a sectioned peripheral nerve is demonstrated by a highly specific sensory loss in that particular area served by that nerve, whereas the clinical findings are more variable for a sectioned dorsal root.
The spinal cord is a caudal continuation of the brain, extending in an organized fashion from the foramen magnum at the base of the skull down to the proximal lumbar spine. The spinal cord has three primary functions: It provides a relay point for sensory information; it serves as a conduit for ascending sensory information and descending motor information; and it mediates body and limb movements because it contains both interneurons and motor neurons. Headed from caudal to rostral, the spinal cord is highly organized with a central butterfly-shaped area of gray matter and surrounding white matter.
The overall diameter of the spinal cord varies as a relative percentage of the spinal canal. The cord fills approximately 35% of the canal at the level of the atlas but increases to approximately 50% of the canal in the lower cervical spine. This variation results from the relative increasing and decreasing size of the spinal gray matter and spinal white matter. As the spinal roots become larger, as occurs at the base of the cervical spine, the size of the gray matter increases relative to the white matter, whereas the size of the white matter decreases linearly from cephalad to caudal.
The gray matter, so called because it appears gray on unstained cross sections, is divided into three zones: the dorsal horn, the intermediate zone, and the ventral horn. Made up predominantly of lower motor neurons, it is prominent in the cervical swellings and lumbar swellings, where axons concentrate before exiting to innervate the upper extremities and lower extremities, respectively.
The white matter derives its name from the fact that the axons in this area are myelinated, casting a white hue on unstained sections. White matter is functionally and anatomically divided into three bilaterally paired columns: the ventral columns, the lateral columns, and the dorsal columns.
The two major ascending systems that relay somatic sensory information are the dorsal columns and the anterolateral system. The ascending axon has its cell body located in the dorsal root ganglion before proceeding without synapsing through the dorsal horn at that level and then ascending along the dorsal column before synapsing at the approximate level of the medulla and crossing over to the contralateral side before proceeding to the cerebral cortex. The topography of the dorsal column is such that the sacrum and lower extremities are medial, with the trunk and cervical region being lateral. The anterolateral system carries pain and temperature sensorium. The afferent fibers have a cell body in the dorsal root ganglion and then synapse at that given level in the dorsal horn before crossing directly to the contralateral side and traveling up the spinothalamic tract.
Motor pathways originate in the cerebral cortex and travel distally to the contralateral side approximately at the level of the medulla and travel down the lateral corticospinal tract before synapsing with the lower motor neuron in the ventral horn of the gray matter. The topography of the corticospinal tract is such that the sacrum and legs lie lateral to the trunk and cervical axons. Thus, at the level of the cervical spine, the spinal cord contains both lower motor neurons traversing to the upper extremities and upper motor neurons being transmitted to the lower extremities. Therefore, injury in this area can give both upper and lower motor findings.
The anatomy of the reflex arc and especially its relationship to spinal shock should be kept in mind. The basic reflex circuitry is an afferent nerve coming from a stretch receptor through the dorsal horn of gray matter before synapsing with the lower motor neuron in the ventral horn of the gray matter, which sends a positive signal to the same muscle via an alpha motor neuron. This simple arc, however, is modulated by input from higher centers. If all descending influence is interrupted, such as would occur in a traumatic transection of the spinal cord, all reflexes are lost. This is also seen during spinal shock. If the local circuitry of the reflex arc is not disturbed, reflexes return at the end of spinal shock. The earliest reflex to return is the bulbocavernosus reflex, which typically returns within 24 hours of injury. Peripheral reflexes may take several months before they return.
Risk of Neurologic Damage
As mentioned earlier, the spinal cord varies in its diameter from cephalad to caudad. In the upper cervical spine, it occupies approximately a third of the spinal canal. In the lower cervical spine, it occupies approximately half of the canal. As inferred from this anatomy, the risk of neurologic damage from injury is greater in the lower cervical spine.
Cord compromise extends from two causes: mechanical destruction resulting directly from the trauma and vascular insufficiency. With vascular insufficiency, hypoxia and edema follow and result in further tissue damage. By approximately 6 hours after the trauma, axonal transport ends, and by 24 hours, cord necrosis begins.
Classification of Neurologic Status
Approximately 60% of injuries to the cervical spine result in no neurologic sequelae. In most of these cases, the injuries are in the upper cervical spine, where the ratio of the spinal cord to the spinal canal is smaller. It is obviously critical to identify unstable injuries of the cervical spine in the intact patient because the evolution of neurologic deficits is both potentially catastrophic and preventable.
Eight cervical nerve roots correspond to the seven cervical vertebral bodies. Each of the first seven nerve roots exits above its respective body (the C1 nerve exiting above the C1 vertebral body, the C2 nerve exiting above the C2 body, and so forth), whereas the C8 nerve root exits through the foramen between the C7 and T1 vertebral bodies. Nerve root injuries can happen either in isolation or in conjunction with more severe spinal cord injuries. Injury to the nerve root alone may result from a compression or fracture of the lateral bone mass and thus impingement on the neural foramen. The clinical findings of a root injury would be those of a lower motor neuron lesion. If the nerve root is still intact and the ongoing pressure to the root is removed, the prognosis for recovery of nerve root function is good.
Incomplete versus Complete Neurologic Injury
In the acute setting, any evidence of neurologic function distal to the level of injury is significant and defines the lesion as being incomplete rather than complete. As Lucas and Ducker reported in a prospective study published in 1979, “The less the injury, the greater the recovery,” and “partial lesions partially recover, whereas complete lesions do not.”
The motor and sensory examination outlined by the American Spinal Injury Association (ASIA) is the most widely accepted and utilized system to assess the impact of spinal cord injury on the patient. It involves the use of a grading system to evaluate the remaining sensory and motor function. The system allows the patients to be assessed through scales of impairment and functional independence.
The sensory level is determined by the patient's ability to perceive pinprick (using a disposable needle or safety pin) and light touch (using a cotton ball). Testing of a key point in each of the 28 dermatomes on the right and the left sides of the body and evaluation of perianal sensation are required. The variability in sensation for each individual stimulus is graded on a three-point scale:
- 0 = absent
- 1 = impaired
- 2 = normal
- NT = not testable
In the cervical spine, the C3 and C4 nerve roots supply sensation to the entire upper neck and chest in a cape-like distribution from the tip of the acromion to just above the nipple line. The next adjacent sensory level is the T2 dermatome. The brachial plexus, C5-T1, supplies the upper extremities.
ASIA also recommends testing of pain and deep pressure sensation in the same dermatomes as well as an evaluation of proprioception by testing the position sense of the index fingers and great toes on each side.
The motor level is determined by manual testing of a key muscle in the 10 paired myotomes from rostral to caudal. The strength of each muscle is graded on a six-point scale:
- 0 = total paralysis
- 1 = palpable or visible contraction
- 2 = full range of motion of the joint powered by the muscle with gravity eliminated
- 3 = full range of motion of the joint powered by the muscle against gravity
- 4 = active movement with full range of motion against moderate resistance
- 5 = normal strength
- NT = not testable
For myotomes that are not clinically testable by manual muscle evaluation, the motor level is presumed to be the same as the sensory level (C1-C4, T2-L1, S2-S5).
ASIA also recommends evaluation of diaphragmatic function (via fluoroscopy, C4 level) and the abdominal musculature (via the Beevor sign, which is the upward migration of the umbilicus from upper abdominal contraction in the absence of lower abdominal contraction due to paralysis at the T10 level). Evaluation of medial hamstring and hip adductor strength is also recommended but not required.
Clinical Features of Spinal Cord Syndromes
Combining the findings on examination with knowledge of the cross-sectional anatomy of the spinal cord allows the examiner to identify specific injury patterns (Figure 4–28).
Diagrams illustrating cross-sectional views of the normal and injured spinal cord. The diagram of the normal spinal column shows the segmental arrangement (C, cervical; L, lumbar; S, sacral; T, thoracic) and the area of flexors and extensors (FLEX and EXT). Central cord syndrome, anterior cord syndrome, Brown-Séquard syndrome, and posterior cord syndrome are incomplete injuries, with affected areas shaded. In complete spinal cord injury, all areas are affected.
The most common of the incomplete cord syndromes is the central cord syndrome, which occurs most frequently in elderly (>65 years) people with underlying degenerative spondylosis but can also be found in younger people who have had a severe hyperextension injury with or without evidence of a fracture, known as spinal cord injury without radiographic abnormality (SCIWORA). Central cord syndrome is defined by ASIA as a clinical presentation characterized by “dissociation in degree of motor weakness with lower limbs stronger than upper limbs and sacral sensory sparing.” The syndrome typically occurs following a hyperextension injury and is thought to be caused by an expanding hematoma or edema forming in the central aspect of the spinal cord. Central cord syndrome can be quite variable in presentation and in recovery. A mild presentation may consist of a slight burning sensation in the upper extremities, whereas a severe central cord syndrome includes motor impairment in both the upper and lower extremities, bladder dysfunction, and a variable sensory deficit below the level of injury. The pattern of clinical presentation is directly related to the cross-sectional anatomy of the spinal cord. Because the lower extremity and sacral tracts of the spinothalamic and corticospinal tracts are lateral, these areas are often spared in central cord syndrome. In cases in which they are involved, they are the areas whose function returns first. The upper extremity deficit is caused by a lesion in the gray matter, and the damage here is largely irreversible.
From 50 to 75% of patients with central cord lesions show some neurologic improvement, but the amount of improvement varies considerably among patients. The usual order in which motor function recovery occurs is as follows: return of lower extremity strength, return of bladder function, return of upper extremity strength, and return of intrinsic function of the hand.
The patient with an anterior cord syndrome typically presents with immediate paralysis and loss of pain and temperature sensation. Both the spinothalamic and corticospinal tracts are located in the anterior aspect of the spinal cord and are therefore involved. With the dorsal columns preserved, the patient still has intact proprioception and vibratory sense as well as intact sensation to deep pressure. This clinical presentation is the most common in the younger (<35 years) trauma victim. The mechanism of injury is typically a flexion injury to the cervical spine. It is usually associated with an identifiable lesion of the cervical spine, most commonly a vertebral body burst fracture or a herniated disk. Return of useful motor function is reported in only 10–16% of patients with anterior cord syndrome. The prognosis is slightly improved, however, if evidence of spinothalamic tract function is present early.
Patients with this syndrome have a motor weakness on the ipsilateral side of the lesion and a sensory deficit on the contralateral side caused by a functional hemisection of the spinal cord. For example, a cervical lesion on the right side of the spinal cord disrupts the ipsilateral corticospinal tract, which is the tract that carries motor function to the right side of the body distal to the level of the lesion. The right spinothalamic tract is also disrupted. This tract carries pain and temperature fibers from the contralateral side of the body distal to the level of injury. Position sense and vibratory sense, which are carried in the posterior column, have not yet crossed the midline; therefore, these sensory functions are disrupted on the ipsilateral side of the injury.
Brown-Séquard syndrome may result from a closed rotational injury such as a fracture-dislocation or may result from a penetrating trauma such as a stab wound or from iatrogenic injury while placing surgical instruments within the spinal canal. The prognosis in cases resulting from a closed injury is quite favorable, with 90% of patients regaining function of the bowel and bladder as well as the ability to walk.
The posterior cord syndrome is the least common of the incomplete syndromes and typically a result of an extension-type injury. Its clinical presentation is one of loss of position and vibratory sense below the level of injury secondary to disruption of the dorsal columns. With these deficits as isolated findings, the prognosis for recovery of ambulation and function of the bowel and bladder is excellent.
Complete Spinal Cord Injury
A complete neurologic deficit is characterized by a total absence of sensation and voluntary motor function caudal to the level of spinal cord injury in the absence of spinal shock. Initial evaluation must rule out any evidence of sacral sparing and the presence of a bulbocavernosus reflex. In the absence of sacral sparing and with the return of the bulbocavernosus reflex, which typically occurs within 24 hours, the spinal cord injury is termed complete and there is virtually no likelihood of functional spinal cord recovery. Affected patients may gain some root function about the level of the injury—a phenomenon called root escape because this damage to nerve roots is a peripheral nerve injury. Although the presence of root escape should not be taken as a potential return of spinal cord function, it can significantly improve the patient's rehabilitation efforts because vital function of the upper extremities may be regained.
A lateral radiograph of the cervical spine may be the only screening tool obtained upon initial radiographic evaluation of the multiple-trauma patient. This radiograph must be carefully reviewed. Should a patient present with a complete neurologic injury or a densely affected incomplete neurologic injury indicating a traumatically malaligned cervical spine, closed reduction of the cervical spine should be urgently attempted with axial traction through Gardner-Wells tongs. Once the patient is fully evaluated and life-threatening injuries are stabilized, secondary diagnostic studies can then be undertaken. If the patient is fully alert, has full pain-free rotational range of motion, no palpable tenderness, and no other injuries, the cervical spine can be cleared on clinical grounds.
Subsequent Plain Radiographs
Full radiographic evaluation of the cervical spine with plain radiographs includes lateral, AP, open-mouth (odontoid), right oblique, and left oblique views. The lateral radiograph, if adequate, visualizes approximately 85% of significant cervical spine injuries. It must display the base of the skull with all seven cervical vertebrae, as well as the proximal half of the T1 vertebral body. If the C7-T1 junction is not visualized, a repeat radiograph should be done with axial traction on the upper extremities caudally to attempt to visualize the C7-T1 junction. If this is unsuccessful, a swimmer's view, which is a transthoracic lateral with the patient's arm fully abducted, should be taken. If this plain radiograph is not satisfactory and if suspicion of injury is still high, a CT scan must be obtained.
When evaluating a lateral cervical spine radiograph, the clinician should first evaluate the bony anatomy. Four lines or curves should be kept in mind (Figure 4–29). The anterior spinal line and the posterior spinal line are imaginary lines drawn from the anterior cortex and posterior cortex, respectively, of the cervical vertebral body from C2 all the way down to T1. The spinal laminar curve is an imaginary line drawn from the posterior aspect of the foramen magnum connecting the anterior cortex of each successive spinous process. These three lines (labeled A, B, and C in Figure 4–29) should have a gentle, continuous lordotic curve with no areas of acute angulation. The fourth line (labeled D in Figure 4–29) is known as the basilar line of Wackenheim, and it is drawn along the posterior surface of the clivus and should thus be tangent to the posterior cortex of the tip of the odontoid process. After the clinician examines the radiograph in terms of these four lines or curves, he or she should look at the individual vertebral bodies to see if there is loss of height of any of them or if a rotational deformity is present with alterations in the alignment of the facets.
Diagram illustrating normal lines and curves in the bony anatomy of the cervical spine. The anterior spinal line (line A), the posterior spinal line (line B), and the spinal laminar curve (line C) should have a gentle, continuous lordotic curve. The basilar line of Wackenheim (line D) is drawn along the posterior surface of the clivus and should thus be tangent to the posterior cortex of the tip of the odontoid process. (Reproduced, with permission, from El-Khoury GY, Kathol MH: Radiographic evaluation of cervical spine trauma. Semin Spine Surg 1991;3:3.)
The evaluation of soft tissues can also prove valuable diagnostically. Prevertebral soft tissues have an upper limit of normal width beyond which a prevertebral hematoma indicative of vertebral injury can be suspected. The upper limits of normal are 11 mm at C1, 6 mm at C2, 7 mm at C3, and 8 mm at C4. The measurements below C4 become more variable and therefore less reliable clinically.
The AP view of the cervical spine is at first a confusing projection to those who are unfamiliar with cervical anatomy, yet careful attention to bony detail in the AP view can be of significant diagnostic aid in picking up subtle injuries. The bony and soft-tissue anatomy seen on the AP projection should be symmetric. The spinous processes should be equally spaced because a single level of increased intraspinous process distance suggests posterior instability. Abrupt malalignment of the spinous processes suggests a rotatory injury such as a unilateral facet dislocation. After checking for these problems, the clinician should inspect the lateral masses. The facet joints are typically angled away from the vertical and therefore not clearly seen on the AP projection. If, however, the facet joint can be seen at a particular level, this is indicative of a fracture through the lateral masses and a rotational malalignment of the facet.
The open-mouth (odontoid) view is the projection most useful for looking at C1-C2 anatomy. It permits visualization of both the dens in the AP plane and the lateral masses of C1 on C2.
The right and left oblique views can be taken of the cervical spine with the patient in the supine position. These views are useful as confirmatory studies in ruling in or out lateral mass injuries.
Two techniques are used in obtaining cervical stress radiographs. The first is to apply axial distraction to the cervical spine through a halo or traction device and obtain a lateral radiograph. This technique should be carefully performed in the presence of a physician and only after gross instabilities of the cervical spine are ruled out. Serial lateral radiographs are taken as weight is sequentially added, reaching an amount equivalent to approximately a third of body weight or 30 kg, depending on the level of suspected injury. Occult instability can be inferred by noting an interspace angulation of at least 11 degrees or an interspace separation of at least 1.7 mm (Figure 4–30).
(A) Diagram illustrating an increase of the C2-C3 interdisk space in a patient with type IIA traumatic spondylolisthesis. (B) Radiograph demonstrating an increased space. (Reproduced, with permission, from Levine AM, Rhyne AL: Traumatic spondylolisthesis of the axis. Semin Spine Surg 1991;3:47.)
The second technique, which should only be performed in a fully alert and cooperative patient, is used to obtain flexion-extension lateral radiographs that are helpful in the diagnosis of late instability. The technique is to have the patient flex the head forward as far as possible while a lateral radiograph is taken and then to have the patient put the head in full extension while another radiograph is taken. Findings presumptive of instability are facet subluxation, forward subluxation of 3.5 cm of one vertebral body on the next, and interbody angulation of greater than 11 degrees.
CT scanning is the most useful means for definitive delineation of bony fracture anatomy. Its advantages are its ready availability and its ability to be performed with a minimal amount of patient manipulation. CT scans provide excellent axial detail, and if thin enough sections are taken, the computer can reconstruct images in sagittal, coronal, or oblique planes. CT scans can now even be reformatted into a three-dimensional construct for excellent visualization of the bony anatomy.
Magnetic Resonance Imaging
MRI is the most effective way to evaluate the soft-tissue component of cervical trauma. The major advantage of MRI is that it can visualize occult disk herniation, hematoma, or edema about the spinal cord, as well as ligamentous injury. Current disadvantages are that MRI is disrupted by metallic objects, so these should be removed from the area of examination, and it also requires a prolonged amount of time to perform, therefore making close monitoring of the acutely ill patient difficult.
Diagnostic Checklist of Spinal Instability
The concept of spinal stability is central to the understanding and treatment of cervical spine injuries. In a broad sense, patients with injuries that are deemed unstable require surgical intervention, whereas those deemed to have stable injury patterns can be treated nonoperatively. Spinal injuries, however, are not readily divided into unstable and stable injuries, and in actuality, they fall along a spectrum of spinal instability.
White and Panjabi's diagnostic checklist of spinal instability (Table 4–9) has nine categories, each of which is assigned a point value. If a total of 5 points is present in a given patient, the injury is deemed unstable.
Table 4–9. White and Panjabi's Diagnostic Checklist of Spinal Instability. ||Download (.pdf)
Table 4–9. White and Panjabi's Diagnostic Checklist of Spinal Instability.
Disruption of the anterior elements, with >25% loss of height
Disruption of the posterior elements
Sagittal plane translation of >3.5 mm or >20% of the anteroposterior diameter of the vertebral body
Intervertebral sagittal rotation of >11 degrees
Intervertebral distance of >1.7 mm on a stretch test
Evidence of cord damage
Evidence of root damage
Acute intervertebral disk space narrowing
Anticipated abnormally large stress
Holdsworth's two-column theory of spine stability, as well as Denis's three-column theory, proposed for application to the thoracolumbar spine, are also applied to the cervical spine in an attempt to better predict stability in the neck.
General Principles of Managing Acute Injuries of the Cervical Spine
Management of acute cervical spine injury is predicated on two principles: protection of the uninjured spinal cord and prevention of further damage to the injured spinal cord. This is accomplished by following spine precaution principles from the very onset of medical care, starting at the accident scene. The cervical spine should be considered injured until proven otherwise and securely immobilized before the patient is transported to a medical center. The equipment for initial immobilization should not be removed until the definitive means of immobilization can be put in place or the cervical spine is cleared of injury. Use of a spinal board, with the patient's head taped to the board and held between two sandbags, is the most secure form of immobilization readily available in the field. This technique can be supplemented by a Philadelphia collar. When the medical center is reached, if a definitive cervical spine injury is identified and deemed unstable, skeletal traction for immobilization, reduction, or both may be applied. Gardner-Wells traction is easily applied and adequate for axial traction. Halo traction affords the added advantage of four-point fixation and thus controlled traction in three planes. Prior to application of traction, it is important to make sure that the patient does not have an occipitocervical dislocation. In these cases, application of traction can lead to worsening of the dislocation and neurologic injury. These specific cases should be treated with immediate application of a halo vest. Halo traction can also be easily converted at a later time to halo-vest immobilization.
Among the various agents that show potential benefits in laboratory studies of models of spinal cord injury are corticosteroids, opiate receptor antagonists (such as naloxone or thyrotropin-releasing hormone), and diuretics (such as mannitol). The National Acute Spinal Cord Injury Studies (NASCIS) II and III reported neurologic improvement with steroid treatment given within 8 hours of injury. Those treated within 3 hours did best; those treated between hours 3 and 8 only did better by extending to 48 hours of treatment. Criticism of the NASCIS studies called to question the validity of the conclusions, and many professional organizations downgraded their enthusiasm for the use of methylprednisolone in the patient with the acutely injured spinal cord. However, many hospitals still use the protocol in blunt trauma cord injuries if the medicine can be administered within 3 hours of the injury. The recommended dosage of methylprednisolone in an acute setting is 30 mg/kg given as a bolus and followed by 5.4 mg/kg/h for 24 hours. However, some thought should be given to its use because, for example, the Congress of Neurological Surgeons stated that steroid therapy “should only be undertaken with the knowledge that the evidence suggesting harmful side effects is more consistent than any suggestion of clinical benefit.”
Cripps RA, Lee BB, Wing P, et al: A global map for traumatic spinal cord injury epidemiology: towards a living data repository for injury prevention. Spinal Cord
Denis F: The three-column spine and its significance in the classification of acute thoracolumbar spinal injuries. Spine (Phila Pa 1976)
Ito Y, Sugimoto Y, Tomioka M, et al: Does high dose methylprednisolone
sodium succinate really improve neurological status in patient with acute cervical cord injury? A prospective study about neurological recovery and early complications. Spine (Phila Pa 1976)
White AA III, Panjabi MM: Update on the evaluation of instability of the lower cervical spine. Instr Course Lect
Injuries of the Upper Cervical Spine
With the exception of occipitoatlantal dissociation, traumatic injuries to the upper cervical spine are less frequently associated with significant neurologic injury than are traumatic injuries to the lower cervical spine. This is secondary to the fact that the spinal cord occupies only a third of the upper spinal canal versus a half of the lower spinal canal.
Occipitoatlantal dissociation is a disruption of the cranial vertebral junction, and it implies a subluxation or complete dislocation of the occipitoatlantal facets. This injury is typically fatal, yet the clinician must be aware of it because unrecognized occipitoatlantal dissociation may have catastrophic results. The mechanism of dissociation is poorly understood, but it most likely results from either a severe flexion or distraction type of injury. Anterior translation of the skull on the vertebral column is a common presentation and most likely a hyperflexion injury. Bucholz, however, presented the pathologic anatomic findings of fatal occipitoatlantal dissociation and proposed a mechanism of hyperextension with resultant distractive force applied across the craniovertebral junction.
When the dissociation is a frank dislocation, the findings are clear on a lateral radiograph. When the dissociation is a subluxation, however, findings may be more subtle. In normal individuals, the distance between the tip of the dens and the basion (the anterior aspect of the foramen magnum) should be no greater than 1.0 cm, and the previously described Wackenheim line should run from the base of the basion tangentially to the tip of the dens. If the dens penetrates this line, anterior translation of the cranium is implied. Calculation of the Powers ratio can also be helpful in securing the diagnosis. Powers and his colleagues described a ratio of two lines (Figure 4–31), the first of which runs from the tip of the basion to the midpoint of the posterior lamina of the atlas (line BC) and the second of which runs from the anterior arch of C1 to the opisthion (line AO). When the ratio of BC to AO is greater than 1:1, anterior occipitoatlantal dissociation is present. Other radiographic signs include marked soft-tissue swelling and the presence of avulsion fractures at the occipitovertebral junction.
Diagram showing lines used in the calculation of the Powers ratio, which is helpful in diagnosing occipitoatlantal dissociation. The distance between the basion (point B) and the posterior arch (point C) is divided by the distance between the anterior arch of C1 (point A) and the opisthion (point O). The normal ratio of BC to AO is 1:1. A ratio of greater than 1 suggests the head is dislocated anteriorly on the spine.
Early recognition and surgical stabilization are the mainstays of treatment in cases of occipitoatlantal dissociation.
Fractures of Vertebra C1 (Atlas Fractures)
The mechanism of injury in the fracture of the atlas is most typically axial compression with or without extension force, and the anatomic findings of the fracture are indicative of the specifics of the force and the position of the head at the time of impact. In 1920, Jefferson presented his classic description of the four-part fracture of the atlas following an axial injury. This fracture is a burst type that occurs secondary to the occipital condyles being driven into the interior portions of the ring of the atlas and driving the lateral masses outward, resulting in a two-part fracture of the anterior ring of the atlas as well as a two-part fracture of the posterior ring. More common than the classic four-part atlas fracture, however, are the two-part and three-part fractures. Isolated anterior arch fractures are the least common, and they are typically associated with fractures of the dens, whereas the more common posterior arch fracture is typically the result of a hyperextension injury.
A fracture of the atlas is typically diagnosed on plain radiographs. Findings may be subtle on the lateral cervical spine radiograph. The open-mouth (odontoid) view may show asymmetry of the lateral masses of C1 on C2 with overhang (Figure 4–32). A bilateral overhang totaling more than 6.9 mm is presumptive evidence of a disruption to the transverse ligament and suggests potential late instability. Presumptive evidence for transverse ligament disruption can also be seen on the lateral radiograph if the ADI is greater than 4 mm.
Open-mouth (odontoid) radiographic view demonstrating asymmetry of the lateral masses of C1 on C2 with overhang in a patient with a Jefferson fracture. (Reproduced, with permission, from El-Khoury GY, Kathol MH: Radiographic evaluation of cervical spine trauma. Semin Spine Surg 1991:3:3.)
The treatment for fractures of the atlas as isolated injuries is typically nonoperative (Figure 4–33). If there are signs of transverse ligament disruption, halo traction is indicated with later transfer to halo-vest immobilization for a total of 3–4 months. Immediate halo-vest application is indicated in cases involving a moderately displaced fracture with lateral mass overhang up to 5 mm, although collar immobilization is preferred in cases involving a minimally displaced fracture of the atlas. At completion of bony union, flexion-extension views should be obtained to rule out any evidence of late instability. If late instability is present and the bony elements were allowed to heal, a limited C1-C2 fusion can address the instability. If a nonunion is present or if the posterior arch remains disrupted, an occiput to C2 fusion is necessary to control the late instability.
Imaging studies in a patient who was in a motor vehicle accident and sustained a distractive extension injury to his cervical spine and a three-part fracture of his atlas (a Jefferson fracture). (A) Lateral radiographic view showing a fracture of the posterior arch. (B) Axial section of a CT scan further delineating the fracture anatomy. This injury was deemed stable and treated nonoperatively in a halo vest.
Dislocations and Subluxations of Vertebrae C1 and C2
Atlantoaxial Rotatory Subluxation
Atlantoaxial rotatory subluxation is most common in children and may be associated with minimal trauma or even occur spontaneously. Although some patients are asymptomatic, others present with neck pain or torticollis (a position in which the head is tilted toward one side and rotated toward the other). Inasmuch as the mechanism of injury is often unclear, the propensity for the C1-C2 location is based on anatomic factors. In approximately 50% of cases, cervical spine rotation occurs at the C1-C2 junction, where the facet joints are more horizontal and less inherently stable in rotation.
The diagnosis of atlantoaxial rotatory subluxation is typically suspected on the basis of radiographs taken in several views. The odontoid view may show displacement of the lateral masses with respect to the dens; a lateral view may show an increased ADI; and the AP view may show a lateral shift of the spinous process of C1 on C2. CT scanning can be used to confirm the diagnosis, and a dynamic CT scan with full attempted right and left rotation can demonstrate a fixed deformity.
There are four types of atlantoaxial rotatory subluxations. In type I, the ADI is less than 3 mm, which suggests the transverse ligament is still intact. In type II, the interval is 3–5 mm, which suggests the transverse ligament is not structurally intact. In type III, the interval exceeds 5 mm, which is indicative of disruption of the transverse ligament as well as secondary stabilization of the alar ligament. In type IV, there is a complete posterior dislocation of the atlas on the axis, a finding typically associated with a hypoplastic odontoid process such as that seen in several forms of mucopolysaccharidosis (eg, the Morquio syndrome).
Treatment of atlantoaxial subluxation is typically conservative, consisting of traction followed by immobilization. Approximately 90% of patients respond to this treatment regimen. There is a high incidence of recurrence, however. For patients who do not respond to conservative measures and for patients with recurrent problems, C1-C2 arthrodesis may be required to control the deformity.
Disruption of the Transverse Ligament
The transverse ligament and secondarily the alar ligament are the main constraints to anterior displacement of C1 on C2. It was previously presumed that because anterior subluxation of C1 on C2 typically involves a fracture through the dens, the transverse ligament is in fact stronger than the bony elements of the dens. Fielding and his colleagues, however, showed that experimentally this was not the case, yet clinically the higher association of anterior dislocation of dens fractures still holds true.
The mechanism of disruption is typically a flexion injury, and the diagnosis is made on lateral radiographs. The ADI should not exceed 3 mm in the adult. If the interval is 4 mm or larger and the dens is intact, a rupture of the transverse ligament is presumed.
High-resolution CT scan can be used to categorize this injury into two types. Type 1 is a disruption in the substance of the transverse ligament, whereas type 2 involves an avulsion fracture of the insertion of the transverse ligament on the lateral mass of C1. Type 1 injuries predictably fail conservative treatment and should be managed with a C1-C2 arthrodesis. A trial of nonoperative care in type 2 injuries using a rigid cervical orthosis may be a reasonable alternative. A 74% success rate can be anticipated, with surgery reserved for patients who fail nonoperative care, showing persistent instability after 12 weeks in mobilization.
Fracture of the Odontoid Process
Fracture of the odontoid process is typically associated with high-velocity trauma, and the mechanism of injury is flexion in most cases. Depending on the fracture pattern, extension may be the predominant force in a smaller subset of cases. Associated injuries, particularly fractures of the ring of the atlas, should be ruled out. Neurologic involvement is relatively rare with odontoid fractures. In a study of 60 patients with acute fractures of the odontoid process, Anderson and D'Alonzo reported that 15 had some neurologic deficit on presentation, but only five of the 15 had major neurologic involvement, and only two of this group of five remained quadriparetic at follow-up.
Odontoid fractures may be suspected on the basis of clinical presentation and confirmed on plain radiographs, although spasm and overlying shadows can obscure the diagnosis. CT scan with sagittal and coronal reconstruction is the most sensitive study to diagnose these injuries. CT scan with axial sectioning alone may miss the horizontal fracture line typical of these injuries; thus, the reconstructions are necessary.
Both the risk of nonunion with delayed instability and the method of treating odontoid fracture depend on the classification of the fracture. Reported rates of nonunion range from 20 to 63%. According to the classification system proposed in 1974 by Anderson and D'Alonzo, there are three types of fracture of the odontoid process (Figure 4–34).
Diagram showing the three types of fractures of the odontoid process.
Type I is a fracture through the tip of the odontoid process. In this configuration, the blood supply is maintained through the base of the odontoid process and through the attachment of the alar transverse ligaments. The mechanical stability of this fracture pattern is left intact. Symptomatic care and immobilization are the treatment of choice.
Type II, the most common type, is a fracture through the base of the odontoid process at its junction with the body of the axis. In this configuration, soft-tissue attachments to the fracture fragment cause distraction at the fracture site. Because the amount of cancellous bone available for opposition is limited, a high nonunion rate is expected, particularly if displacement is significant or the patient is older (>60 years). In this case, primary surgical treatment may be indicated. Anterior screw fixation of the odontoid process is now the treatment of choice for most type II odontoid fractures. Although it is technically demanding, it does allow for the maintenance of motion at C1-C2 (Figure 4–35).
Imaging studies in a patient with a type II odontoid fracture nonunion. (A) Open-mouth radiographic view showing the fracture line at the base of the odontoid process. (B) Sagittal reconstruction using CT scanning to better delineate the fracture anatomy. (C) Radiograph taken after the patient underwent anterior placement of two odontoid screws under fluoroscopic control using a cannulated screw system.
Type III is a fracture through the body of the axis. The blood supply is maintained through soft-tissue attachments, and abundant cancellous bone opposition at the fracture site facilitates a high rate of union. The treatment, therefore, is conservative, consisting of halo traction or halo-vest immobilization until bony union occurs. Although the rate of union is acceptable, there is a relatively high rate of malunion that may limit the patient's cervical rotation.
Hangman's Fracture (Traumatic Spondylolisthesis of Vertebra C2)
Hangman's fracture occurs when a fracture line passes through the neural arch of the axis. The anatomy of the axis is such that the superior facets are anterior and the inferior facets are posterior, thus concentrating stress through the neural arch. Because of the high ratio of spinal canal size to spinal cord size at this level, neurologic damage associated with hangman's fracture should be unusual. However, in his postmortem studies, Bucholz reported that traumatic spondylolisthesis was second only to occipitoatlantal dislocations in cervical injuries leading to fatalities.
According to the scheme proposed by Levine and Rhyne, hangman's fractures can be classified on the basis of anatomic factors and the presumed mechanism of injury. Treatment depends on the type of fracture. Imaging studies in a patient with hangman's fracture are shown in Figure 4–36.
Imaging studies in a patient who was in a motor vehicle accident and sustained a hangman's fracture, or traumatic spondylolisthesis of C2. (A) Lateral radiographic view, which is largely unremarkable. (B) Sagittal reconstruction using CT scanning to better delineate the fracture site at the base of the posterior elements. The patient was treated nonoperatively.
Type I is typically caused by hyperextension with or without additional axial load. There is no angulation of the deformity, and the fracture fragments are separated by less than 3 mm. Treatment should consist of immobilization in a cervical collar or halo vest until union occurs, which is typically 12 weeks.
Type II is thought to be caused by hyperextension and axial load with a secondary flexion component leading to displacement of the fracture. Reduction of the anterior angulation in this type of fracture is necessary and typically obtained by traction therapy and then followed by placement of a halo vest until union occurs. An atypical type II hangman's fracture is described. This fracture occurs through the posterior aspect of the vertebral body, potentially resulting in cord compromise as the anterior aspect of the vertebral body flexes forward. A higher likelihood of neurologic injury with this atypical pattern is seen, and halo-vest immobilization is recommended.
Type IIA has the same fracture pattern as type II but with a component of distraction that also occurred at the time of injury and led to disruption of the C2-C3 disk space, rendering this injury inherently unstable. Traction should be avoided in cases of type IIA fracture because it exacerbates the injury. Treatment should consist of immediate halo-vest application, with the patient's head positioned in slight extension to afford a reduction.
Type III includes a fracture through the neural arch, a facet dislocation, and a disruption of the C2-C3 disk space that renders the injury highly unstable. Treatment generally consists of early closed reduction of the facet dislocation and application of a halo vest to maintain the reduction. If the reduction cannot be obtained in a closed fashion or cannot be maintained conservatively, treatment with open reduction of the dislocation and anterior or posterior fusion is indicated.
Anderson LD, D'Alonzo RT: Fractures of the odontoid process of the axis. J Bone Joint Surg Am
Hsu WK, Anderson PA: Odontoid fractures: update on management. J Am Acad Orthop Surg
Huybregts JG, Jacobs WC, Vleggeert-Lankamp CL: The optimal treatment of type II and III odontoid fractures in the elderly: a systematic review. Eur Spine J
Ramieri A, Domenicucci M, Landi A, et al: Conservative treatment of neural arch fractures of the axis: computed tomography scan and x-ray study on consolidation time. World Neurosurg
Injuries of the Lower Cervical Spine
As stated earlier, fractures and dislocations of the lower cervical spine have a greater frequency of catastrophic neurologic involvement because of the decreased ratio of spinal canal to spinal cord in the lower levels. Treatment of affected patients again relies on early recognition of the injury, recognition of inherent stability or instability of the injury pattern, and institution of appropriate definitive care.
In 1982, Allen and colleagues developed a classification system for closed indirect fractures and dislocations of the lower cervical spine. After reviewing numerous cases previously described by other authors as well as 165 of their own cases, they grouped the injuries into six categories, based on the position of the cervical spine at the time of impact and on the dominant mode of failure. The six categories were compressive flexion, vertical compression, distractive flexion, compressive extension, distractive extension, and lateral flexion. Of these, the distractive flexion injuries were the most common, followed by the compressive extension injuries and the compressive flexion injuries. Some of the categories were further divided into stages, as described next.
Compressive Flexion Injury
There are five stages of compressive flexion injuries, which are labeled compression flexion stage (CFS) I through V (Figure 4–37). CFS I shows a slight blunting and rounding to the anterior superior vertebral margin, without any evidence of posterior ligamentous damage. CFS II shows some additional loss of height of the anterior vertebral body, again sparing the posterior elements. CFS III has an additional fracture line passing from the anterior surface of the vertebral body through to the inferior subchondral plate, with minimal displacement. CFS IV has less than 3 mm of displacement of the inferior posterior vertebral fragment into the neural canal. CFS V has severe displacement of the inferior posterior fragment into the canal, with widening of the spinous processes posteriorly, indicative of three-column disruption.
Radiographs showing the five stages of compressive flexion injury. A shows compression flexion stage (CFS) I. B shows CFS II. C shows CFS III. D shows CFS IV. E shows CFS V. (Reproduced, with permission, from Allen BL, Ferguson RL, Lehmann TR, et al: A mechanistic classification of closed, indirect fractures and dislocations of the lower cervical spine. Spine (Phila Pa 1976) 1982;7:1.)
Within the compressive flexion category are two types of fractures, more commonly referred to as the compression fracture and the teardrop fracture. Most compression fractures without disruption of the posterior elements are thought to be stable, so no surgical intervention is required. The more severe compression fracture injuries, however, can result in displacement of bone into the spinal canal, and if a neurologic injury is present, these require anterior decompression and stabilization. All patients should be carefully checked with flexion-extension views at the completion of their treatment to rule out any evidence of late instability.
Vertical Compression Injury
Vertical compression spinal (VCS) injuries occur secondary to axial loading and are divided into three stages. VCS I consists of an endplate central fracture with no evidence of ligamentous failure. VCS II is a fracture of both vertebral endplates, again with only minimal displacement. VCS III is the more commonly termed burst fracture with a spectrum of fragmentation of the vertebral body, with or without posterior element disruption.
The treatment for VCS injuries is typically nonoperative. Traction is applied to obtain and maintain alignment, and bony union is generally complete after 3 months of halo-vest immobilization. Flexion-extension views should be obtained at the completion of healing because a posterior ligamentous injury can result in late instability.
Distractive Flexion Injury
The category of distractive flexion spinal (DFS) injury was the most common injury category reported by Allen and colleagues, and it includes both unilateral and bilateral facet subluxation and dislocation. There are four stages of DFS injury. DFS I, termed a flexion sprain, is characterized by subluxation of the facet joint, with possible interspinous process widening. This injury has subtle radiographic findings and may easily be missed during initial evaluation and therefore result in late symptomatic instability (Figure 4–38). DFS II is a unilateral facet dislocation, the diagnosis of which can be confirmed on plain radiographs. The lateral radiograph would reveal an anterior subluxation of one vertebra of approximately 25% of vertebral body width at the affected level. The facet itself may be perched or fully dislocated. DFS III is a bilateral facet dislocation with approximately 50% anterior dislocation at the affected level. DFS IV, which is also termed a floating vertebra, is a bilateral facet dislocation with displacement of a full vertebral width.