CHAPTER 77 Injuries of the Lower Cervical Spine
Cervical spine injuries occur in 3% to 4% of all trauma patients.1 Spinal cord injury is much rarer occurring in only about 12,000 people per year in the United States. Because of the association with traumatic brain injury, the risk of spinal injury can be stratified by level of consciousness, occurring in only 2.8% of alert compared with 7.7% of nonevaluable patients.2 It is estimated that about 40% of these injuries will be unstable with potential for neurologic injury.
The importance of the evaluation and protection of the cervical spine is underscored by continued reports of neurologic deterioration that occurs after patients are admitted to hospitals. Levi and colleagues3 retrospectively reviewed 24 cases of patients who deteriorated after admission to level I trauma centers. Neurologic deterioration occurred overall in 0.026% of all trauma patients and 0.21% of patients with known spinal injuries. The common cause was inadequate radiographs, misreading, and poor-quality studies.
Positive changes documenting improvement of the overall care of spinal cord–injured patients have occurred. The time of admission to rehabilitation centers decreased from 23 to 7 days.4 A fivefold decrease in rehabilitation duration has occurred due to better overall management and greater financial constraints. Functional impairment at discharge has seen improvement for three consecutive decades. This is also consistent with decreasing admission of compete quadriplegia compared with increased incomplete quadriplegics. Overall mortality (occurring within the first year postinjury) decreased over 65% from the 1970s to 1990s.5
Many of these improvements resulted from better initial accident care, earlier diagnosis, improved resuscitation methods, and modern surgical techniques. The emergency care of the spinal cord–injured patient is discussed in Chapter 81. This chapter reviews the evaluation of the cervical spine in a trauma patient, new classification systems of lower cervical spine injuries, and their application to surgical decision making.
The anatomy of the cervical spine is reviewed in detail in Chapter 2. These remarks are aimed at the anatomy specific to cervical injuries and their treatment. The subaxial cervical spine is highly mobile but protective of its soft tissue contents, namely the spinal cord, nerve roots, and vertebral arteries. Each vertebra from C3 to 7 is progressively larger and is connected both above and below by three articulations, the disc complex and the paired lateral pillars. These form three columns that are essential for the weight-bearing function of the spine. The anterior column is connected to each lateral pillar by pedicles, and the lateral pillars are connected by the lamina. Thus the lamina and pedicle act as tie-rod connectors. The large spinous processes are levers for attachment of the paraspinal muscles and ligaments. These combined with the nuchal ligaments form the posterior ligamentous complex. which is essential to maintaining stability against flexion and anterior shear forces.
The vertebral bodies are relatively large weight-bearing cuboid structures that are connected to each other by the intervertebral disc. The uncinate process or uncus is a cranial extension at the posterolateral corner of the vertebral body. The uncus is responsible for resistance to lateral bending, lateral listhesis, and rotation. With aging or degenerative processes the uncus can hypertrophy, narrowing the neuroforamina, which are located directly behind. Preexisting foraminal stenosis may predispose to neural injury, especially in hyperextension injuries. Short transverse processes extend lateral from the body about midlevel in anteroposterior direction and are confluent with vestigial ribs forming the foramen transversarium. Within this structure lies the vertebral artery and venous plexus except at C7, which is void.
The pedicles extend posteriorly and outward from the cranial aspect of the posterior wall of the body to the lateral masses. These short tubular structures may be cannulated with screws but, because of the close proximity of the vertebral arteries, this technique is often considered dangerous and supplanted by the use of the lateral masses as screw anchorage sites. The lateral masses or pillars when viewed laterally are parallelogram in shape and appear square when viewed ventrally or dorsally. They have cranial and caudal projections, the superior and inferior articular facets, respectively. These have cartilage surfaces that form the facet articulations. The superior facet is located behind the neuroforamen and, when fractured, may cause root injury. The laminae are thin plates of bone that extend from the lateral mass obliquely posterior and meet each other at the base of the spinous process. The spinous processes project posteriorly, angling downward, and are progressively larger from cranial to caudal direction. The spinous process of C3 to C5 is always bifid, whereas C6 may be and C7 is never bifid.
The ligaments of the cervical spine are essential to maintain both stability and range of motion. The anterior and posterior longitudinal ligaments lie on the respective surfaces of the vertebral body. The latter extends laterally at each disc and narrows behind the body. The annulus joins the corresponding vertebral bodies and is the primary restraint to all directions of movements. All cases of traumatic subluxation or dislocation indicate injury to this structure.
The posterior ligamentous complex consists of the nuchal ligaments, ligamentum flava, facet joint capsules, and all bony attachments. The nuchal ligaments include the ligamentum nuchae, which is a thick band strongly attached to the spinous processes of C2 and C7 and which extends fibers downward attaching to the tips of the C3-C6 spinous processes. The ligamentum nuchae blends with the supraspinous ligament, which bands together each spinous process at their tips. Between the spinous processes are the interspinous ligaments. The ligamentum flava attaches to the underside of the cranial lamina and top-most aspect of the caudal vertebrae. These elastic structures afford flexion and, when disrupted, indicate a significant failure from that direction. Each facet articulation has relatively redundant facet capsules that offer only a small amount of stability but when disrupted, like the ligamentum flava, indicate significant injury.
The neural anatomy is important to explain neurologic injury and to plan treatment. Advanced imaging affords an excellent means of visualizing the pertinent neuroanatomy and must be judiciously analyzed before placement of any fixation devices. The spinal cord is an elastic structure that lies inside the spinal canal. In adults its anteroposterior diameter measures about 8 mm and cross-section is slightly oval. At each disc level ventral and dorsal roots sprout and join laterally, forming the spinal nerve in the neuroforamina. The neuroforamen is a truncated cone angling outward. It is bordered anteriorly by the posterior lateral corner of the disc and uncus, above and below by pedicles and posteriorly by the superior articular facet and lateral mass. After exiting from the neuroforamen, the spinal nerve divides into ventral and dorsal primary rami.
The vertebral artery ascends from the subclavian artery to pass within the foramen transversarium at C6, although in 1% of cases this can occur aberrantly at C7 or higher levels. It exits the foramen transversarium of C2 turning anteriorly and medially in C2 and then again laterally into C1. The vertebral artery being enclosed within the spinal column may be injured due to intervertebral distraction, fractures about the foramen transversarium, and translational injuries such as facet dislocations. Further, the vertebral arteries may be damaged intraoperatively by lateral mass or pedicle screw placement.
The subaxial cervical spine has complex motions that are controlled by the disc annulus complex, various ligaments, and bony projections including the facets and uncus. The motions are usually coupled combining angulations along more than one axis or angulations with translation. For example, during flexion-extension, 1 to 2 mm of obligatory translation occurs. Similarly, in rotation, both lateral bending and rotation occurs simultaneously.
Kinematically each segment of the subaxial spine normally has approximately 11, 5, and 5 degrees of movement in flexion-extension, lateral bending, and rotation.6 Small amounts (1 to 2 mm) of translation occur in both anterior-posterior and lateral directions. Normal passive motion requires little application of load, which is measured by the neutral zone. Injury will increase the dimension of the neutral zone and when this exceeds certain thresholds such that the neural structures are at risk, the spine is deemed unstable.
Understanding anatomy is essential for the avoidance of adverse events during surgery. Anomalies are not uncommon in the cervical spine and should be carefully interpreted. An important axiom is that congenital bony malformations are associated with vascular malformations. Therefore when anomalies are present, a preoperative computed tomography (CT) angiogram or magnetic resonance angiography should be considered.
Anterior decompression is well described using the standard Smith-Robinson approach. If required, corpectomies can be performed after removal of the disc. If possible, the posterior longitudinal ligament should be left intact. Distraction either by pins placed into adjacent vertebral bodies or with cranial tongs can restore axial length. Decompression should be oriented to the midline and be at least 16 mm wide depending on the transverse distance between the foramen transversarium. Common errors are biasing the decompression usually to the side opposite that from which the surgeons stands or too narrow a decompression. A plate with unicortical screws may be placed after insertion of a graft or a cage. The plate should be oriented so that screws have purchased into the body and not adjacent discs or laterally into the foramen transversarium.
The posterior cervical spine is dissected carefully as disruption of ligamentum flava exposes the spinal canal to risk of iatrogenic damage. Further, laminar fractures may be displaced downward onto the spinal cord. The joint capsules should be preserved to avoid any further destabilizing injury. In younger patients, simple exposure of noninvolved levels may result in spontaneous fusion.
Screw fixation into the lateral masses requires identification of a proper starting point and screw direction. The lateral mass is first examined and its borders defined. The cranial and caudal borders are the corresponding facet joints, the lateral border is the edge of the lateral mass, and the medial border is the valley at the junction of the lamina and lateral mass. Several starting points have been described.7–9 However, with modern fixation most authors use a modified technique where the starting point is 1 to 2 mm medial to center of the lateral mass. The author marks this point with a small burr so that when a pilot hole is started it is located at the selected point. The pilot hole is oriented upward attempting to be parallel to the facet articulations and outward about 15 to 30 degrees. This outward orientation is essential to avoid injury to the vertebral artery, which lies directly anterior to the starting point as seen on axial images. The outward angulation is often limited by the next caudal spinous process.
All blunt trauma patients should be assumed to have a cervical spine injury until proven otherwise. The process to exclude injury when present is termed cervical spine clearance. The goal is to identify patients who do not have injury so that restrictions such as collar immobilization can be removed. It is not designed to identify all injuries, many of which are trivial, but to properly identify those with significant injury so that proper imaging can be obtained and treatment rendered. The systematic process is performed by consideration of the mechanism of injury, clinical examination, and when necessary, imaging. Many recent protocols to accomplish this have been proposed and tested for reliability and validity.
The clinical examination is an essential component of the evaluation process. It depends on cooperation between examiner and patient, requiring normal mentation. Sensitivity is reduced by the patient’s inability to report symptoms due to conditions such as distracting pain, mild head injuries, or intoxication.10 The clinical examination is performed by reviewing the history of injury and symptoms of cervical pain, numbness, weakness, or paresthesias. The examination assesses spinal tenderness, which is best achieved by carefully logrolling the patient and palpating from the occiput to sacrum. Tenderness, swelling, hematoma, and gaps between spinous processes are indications for further evaluation.
The neurologic examination is performed using standards established by the American Spinal Injury Association (Fig. 77–1).11 These should be recorded in the medical record in a timely manner. The neurologic examination includes testing of cranial nerves, motor and sensory function, and perineal function. Motor function of key muscle groups in the arms and legs are assessed from grades 0 to 5 (Table 77–1). Several sensory functions including pinprick, light touch, and proprioception or vibratory function are tested in both the upper and lower extremities. Sensory function is graded 0, 1, or 2 (absent, hypoesthesia, and normal, respectively). Deep tendon and pathologic reflexes are determined. Hyperreflexia and findings of clonus, Babinski signs, or Hoffmann signs are indications of spinal cord compression, although in cases of acute spinal cord injury areflexia is initially common. The level of spinal cord injury is defined as the lowest-functioning root level with at least grade 3 motor function. Complete spinal cord injuries are defined as the absence of any motor or sensory function including sacral roots distal to the zone of injury. Incomplete cord syndromes have some retained motor or sensory function and have a significantly better prognosis than complete lesions. The overall impairment is graded by the modified Frankel (ASIA) scale: A—motor and sensory complete; B—motor complete with sensory sparing; C—motor and sensory sparing but not functional strength; D—motor and sensory sparing with functional strength; E—normal motor and sensory.11
|Extensor hallucis longus||L5|
Examination of the perineum is an important component of the clinical examination. The perianal skin is tested for pinprick sensation. Digital rectal examination is performed assessing tone and the presence of voluntary contraction. Finally, sacral reflexes are assessed. The anal wink is obtained by gently scratching the perianal skin and observing a contracture. The bulbocavernosis reflex is tested during digital rectal examination by compression of the glans penis or clitoris or alternatively by pulling on a Foley catheter. During this maneuver an anal contracture should occur.
Interpretation of the perineal examination depends on the level of injury and time from injury. In cervical cord injuries the perineal function may be initially absent, but the bulbocavernosis reflex may return in several days, indicating resolution of spinal shock, which is an initial depolarization of axonal tissue after injury. If the patient has not made any distal recovery at this time, the prognosis is grave. In some cases only sacral function is spared, changing the patient from complete to incomplete with a dramatic change in prognosis. Injuries at the conus or cauda equina that have retained perineal function or bulbocavernosis reflex have excellent prognosis for bladder and bowel function.
After the clinical examination, a decision is made whether radiologic studies are indicated. To determine the proper course, patients can be divided into one of four groups: asymptomatic; temporarily nonevaluable but asymptomatic; symptomatic; and obtunded. Asymptomatic patients are those who have no pain, no cervical tenderness, neurologic signs and symptoms and who are awake, alert, and nonintoxicated.1 Additionally, they do not have any distracting injuries that may preclude pain assessment of a potential cervical spine injury. Distracting injuries include long bone fractures, burns, visceral injury, dislocations, and craniofacial or thoracic trauma. The temporarily nonevaluable patient is otherwise asymptomatic but is either intoxicated where sobriety is expected within 24 hours or will have resolution of other painful injuries through reduction or fixation. It is hypothesized that these patients can be evaluated similar to the asymptomatic patients in a delayed fashion.12 Symptomatic patients have cervical pain, tenderness, or neurologic symptoms such as paresthesias, weakness, or numbness. The obtunded patient is one who will not be able to fully participate in the clearance process and at the same time is at highest risk for injury. Included in this group are disabled and infant patients.
Multiple large multicenter prospective trials, institutional prospective studies, systematic reviews, and meta-analyses demonstrate conclusively that the asymptomatic patient can be cleared after clinical assessment and does not require radiographic evaluation.13–17 Designation as asymptomatic will clear 99.8% of patients without radiographs and use of such protocol only rarely misses a significant injury.13 In more than 62,000 cases, no neurologic complication occurred after clearance of asymptomatic patients on clinical grounds.
This group has not been studied and therefore no strong recommendations can be made. However, the author’s recommendation is to maintain collar immobilization for 24 to 48 hours and to avoid initial imaging. Unless contraindicated by other injuries, the patients may be upright to 30 degrees in bed. After sobriety or treatment of other injuries, the patient is re-examined and if asymptomatic, the spine is cleared. Alternatively, if the status of the spine is required immediately (such as for treatment of cranio-cervical injury), the patient is evaluated according to the obtunded protocol.
Symptomatic patients are at significant risk of cervical injury and all authorities agree that imaging is required.14,15 In cases of obvious cervical cord injury, this may be by means of a lateral radiograph so that traction can be instituted immediately if a fracture dislocation is present. More commonly, a multidetector CT (MDCT) is performed. MDCT allows acquisition of thin-slice thicknesses (1.25 to 1.5 mm) and affords high-resolution reconstruction in all three planes. Compared with plain radiographs, it has significantly increased sensitivity and is cost effective if CT imaging of other body parts are required.18 Magnetic resonance imaging (MRI) is indicated after CT or plain radiographs when there are unexplained neurologic defects, progressive deterioration, preoperatively, and in cases of facet dislocation of neurologically intact patients. Further, MRI with fat suppression is useful to identify ligamentous disruption, which may aid surgical decision making.
No agreement has been reached as to the best method to evaluate the cervical spine of cognitively impaired patients. However, all stakeholders agree that radiologic evaluation is warranted in this high-risk group.15 Current recommendation is divided into two groups: MDCT alone19,20 and MDCT with MRI.21 MDCT is easy to perform and likely identifies all osseous injuries. However, ligamentous injury may be missed if the spine is reduced in normal alignment. If restrictions are released, then displacement may occur when the patient is no longer immobilized. Given the high resolution of MDCT compared with older CT studies, proponents of this modality believe the risk of this is small. The latter technique takes advantage of both MCDT, which identifies osseous injury, and MRI, which is sensitive for ligamentous injury. A recent meta-analysis shows that MRI has a 100% negative predictive value, indicating that a negative study excludes any significant injury.1 However, there is a high false-positive rate, meaning that many patients will be treated unnecessarily. In the absence of proof of the superiority of either approach, the author recommends that each institution develop and use a protocol to evaluate these patients.
Classification of spinal injuries is difficult due to the complex anatomy of vertebrae, the presence of a three-joint complex, and the many ligamentous structures responsible for stability. Classification systems can be broadly categorized as morphologic, mechanistic, and by stability. Morphologic systems use common names that describe the pathoanatomy such as burst fracture or facet dislocation. These are useful but do not help to define treatment or imply prognosis. Mechanistic systems attempt to identify force vectors that cause injury.22 Using these, potential treatment strategies can be induced, usually by reversing the vector direction. Stability is hard to define but implies the ability of the spine to withstand physiologic loads. This is well grounded in sound biomechanical principles but is more difficult to use clinically. All of these classification systems have many limitations including not being comprehensive, not aiding treatment decision making, not using modern imaging, and lacking reliability and validity.
Stability and neurologic status are the most important factors when treating patients with cervical spine trauma. Stability was initially described by Nicoll, who identified patterns of injury that were associated with poor functional outcomes (i.e., inability to return to work as a miner).23 He found that posterior ligamentous injuries and dislocations were “unstable patterns.” Stability has been defined by White and Panjabi6 as “ability of the spine under physiologic loads to maintain a relationship between vertebral segments in such a way that there is neither damage nor subsequent irritation of the spinal cord or nerve roots, and, in addition, there is no development of incapacitating deformity or pain due to structural changes.” Stability as thus defined is dichotomous when in fact stability is a continuous variable with many shades of gray. In an effort to quantify clinical stability and thus make it useful for traditional decision making, White and Panjabi6 developed a checklist. Points are assigned for injury to anterior or posterior columns, translation, excessive angulation, distraction, and neurologic injury. Higher scores indicate greater degrees of instability. This concept is well founded with biomechanical studies but has not proven useful as a clinical tool or even been tested for validity.
White and colleagues24 performed elegant biomechanical studies on cadaveric spines with serial ligamentous sectioning. After each ligament was divided, loads were applied and motion measured. They found that the cervical spine developed excessive motions when all elements of one column (anterior or posterior) and one other ligament were sectioned. The amount of displacement at this time was 3.5 mm of translation and 11 degrees of greater angulation than adjacent levels. These parameters were exceeded only after injury and are accepted as rendering the spine unstable clinically. Practically, however, the cases most difficult to assess for stability do not have displacements or angulation exceeding these limits.
Two new systems have been developed on the basis of modern imaging. They allow quantification over a range rather than being only dichotomous (stable or unstable) and have been tested for reliability and validity. Further, these systems can be used to aid in decision making.
The Cervical Spine Injury Severity Score (CSISS) is based on independent analysis of four columns (anterior, posterior, right column, and left lateral column) (Table 77–2).25 The anterior column includes the body, disc including the annulus, anterior and posterior longitudinal ligaments, and transverse processes. Each lateral column is scored separately and includes the facet projections, lateral mass, pedicles, and facet joint capsules. The posterior column includes the lamina, spinous process, ligamentum flavum, and nuchal ligaments (Fig. 77–2A).
|CSISS Analog Score|
FIGURE 77–2 A, The Cervical Spine Injury Severity Score (CSISS) is based on analysis of four anatomic columns: anterior, each lateral column, and posterior. B, Each column is scored independently using the analog scale from 0 to 5. A nondisplaced fracture is valued at “1,” while increasing scores are given proportionally to the amount of displacement. A “5” is given for the worst injury to a given column that is possible. Fractional values may be used. The CSISS is the sum from each column ranging from 0 to 20.
Each column is scored using a 0-5 analog scale (Fig. 77–2B). Fractional scores can be used. Scores increase proportional to either displacement of fracture fragments or separation as a result of soft tissue injury. For example, a nondisplaced fracture is scored 1 while the worst injury possible for that column (e.g., facet fracture dislocation with 10-mm displacement) is a 5. Each column is scored independently and summed, giving the CSISS ranging 0-20.
Anderson and colleagues25 assessed reliability in 34 cases with 15 examiners. Both intraobserver and interobserver reliability were excellent. Construct validity was also good as all patients with scores equal to 7 had surgery, whereas only 15% less than 5 had surgery (Fig. 77–3). Caput and colleagues26 retrospectively correlated CSISS score to surgical approach in 70 patients. They found a significant correlation with high CSISS scores (>11) to a posterior or combined anteroposterior approach. The anterior approach had lower scores with a mean 6.3, indicating that anterior surgery was used for less severe injuries, in accordance with well-known biomechanical studies.
FIGURE 77–3 A, Sagittal reconstruction showing C5 flexion-axial loading injury. There is a small teardrop fragment and the body is retropulsed into the spinal canal. The patient was a Frankel C quadriplegia. B, The next sagittal image shows wide separation of the spinous processes, indicating disruption of the posterior osseous-ligamentous complex. The CSISS anterior and posterior columns are scored 4 and 5, respectively. C and D, Subluxation of the left and right sagittal computed tomography (CT) in the plane of the lateral masses is present. The CSISS is 2 for each lateral column. E, Axial CT. Note the displaced laminar fracture. F, Traction lateral radiograph after applying 50 pounds. Alignment is significantly improved and retropulsion of C5 is reduced. G, The CSISS score totaled 13, indicating a highly unstable fracture and the need for surgery. The patient was treated by anterior corpectomy and fusion with fibular allograft and translational plate. Neurologically the patient improved to Frankel D and healed the fusion.
The Subaxial Cervical Spine Injury Classification system (SLIC) evaluates fracture morphology, the discoligamentous complex, and neurologic function, creating a comprehensive system to aid treatment decision making (Table 77–3).27 The system assigns points for each domain and if the score exceeds a threshold, surgery would be indicated.
Compression injuries are assigned one point, with an additional point being assigned for burst fractures or when greater than 10 degrees of scoliosis is created by lateral compression injuries. Distraction injuries involve ligamentous structures such as the disc-anulus or posterior ligamentous complex. These are more severe, have a poor prognosis, and are therefore assigned three points. Distraction injuries can include hyperflexion injuries with posterior ligamentous injuries or more commonly disc distraction from hyperextension. Rotation and translational injuries such as from facet dislocations are given four points.
The discoligamentous structures are critical to the stability of the spine. Anteriorly these include the disc annulus complex and the anterior/posterior longitudinal ligaments. Posteriorly these include the nuchal ligaments (supraspinal, interspinal ligaments, and ligamentum nuchae), as well as the ligamentum flavum and facet capsules. Injury to any of these structures is assessed by vertebral displacement where widening or translation exceeds physiologic limits, indicating ligamentous disruption and by MRI. The criterion for ligamentous disruption is high-signal intensity passing transversely through a known ligamentous structure. The discoligamentous complex is scored 0 for intact, 1 for indeterminate, and 2 for disrupted. If a potential discoligamentous injury is present or the structure of this complex cannot be determined, an intermediate score of 1 is assigned.
The SLIC system includes the neurologic status, which is the most important when considering surgical treatment. Root injuries have a good prognosis and are scored 1 while complete cord injuries are given 2 points. Three points are assigned for incomplete, which is a higher score than for complete because the former are more likely to benefit from surgery. Finally, an additional point is added when residual neural tissue compression is present in patients with neurologic deficits.
The SLIC score is the sum of the individual components. A patient with an SLIC score equal to 3 is treated nonoperatively, whereas surgery is recommended in a patient with a score equal to 5 (Fig. 77–4). Scores of 4 can be treated either operatively or nonoperatively.
FIGURE 77–4 A, Sagittal computed tomography (CT) showing C7 burst fracture. The patient was neurologically intact. The discoligamentous complex is intact. The SLIC score is 2 (2 for burst fracture, 0 for DLC, and 0 for neurologic). The Cervical Spine Injury Severity Score is 6 (4 for anterior, 1 for each facet diastasis, and 0 for posterior). The patient was treated successfully nonoperatively in halo vest. B, Diastasis (arrow) of both facet joints (left shown) is present. C, Retropulsion of bone from C7 burst fracture is seen on axial CT. D, Traction of 25 pounds reduced the fracture. E, The patient was treated in a halo vest. The lateral radiograph at 12 months shows excellent alignment and healing of the burst fracture.
Vaccaro and colleagues27 tested the reliability of the SLIC system. Interobserver reliability was only moderate, and both injury morphology and discoligamentous complex scores were only fair. Intraobserver reliability was excellent for overall and individual components. Similar to the CSISS, the system was valid when tested against treatment decisions. The SLIC system predicted accurately what the individual observers would recommend in 91% of cases.
These quantitative systems do not “name” or describe the injuries. Both authors of the aforementioned studies recommend using common colloquial terms that have a long history of use. Examples of these are facet dislocations, lateral mass fractures, and flexion/axial loading injuries. Unfortunately, injuries may be combined or similar so that distinguishing between any two fractures may be difficult.
Fractures are divided into categories on the basis of the most severe (or obvious) spinal column injured. This is done because treatment is often similar among different fracture types in each category. However, in many cases the injury extends through all columns and treatment must be individualized. In each case, the severity of injury must be determined to make correct judgments as to treatment.
Anterior column injuries include compression fractures, burst fractures, flexion axial loading injury, and disc distraction injuries (Table 77–4). Transverse process fractures are included in this group, although they have no effect on spinal stability but may be associated with vertebral artery injury. Most anterior column injuries are easily recognizable on plain radiographs or CT. When present, they suggest a hyperflexion mechanism and a search for a concomitant injury to the posterior ligamentous complex should ensue. The notable exception is the disc-distraction injury resulting from opposite forces (i.e., extension and posterior shear). These injuries can be subtle, especially in patients with preexisting spondylosis.
Anterior compression fractures, like burst fractures, occur from hyperflexion and/or axial loading forces. During this loading, the disc is pressurized, resulting in failure of the endplate and creating wedging of the vertebral body. This usually occurs along the superior endplate. During hyperflexion the posterior ligaments may be strained beyond physiologic limits, causing disruption. This injury pattern has been termed “hidden flexion injury” and often fails nonoperative treatment.28
Radiographically, wedging of the anterior body and fractures of the superior endplate is seen. The posterior wall is intact. Alignment may be neutral to kyphotic. In the latter case, posterior ligamentous rupture should be suspected. Widening of the space between spinous processes or perching of the facets is pathognomonic of this. In some cases associated fractures of the lamina or spinous processes are present.
Isolated compression fractures without posterior or lateral mass involvement almost always score low CSISS (0-2) and SLIC (1) scores (see Table 77–4). When associated with posterior ligamentous injury or facet perching, CSISS and SLIC scores range 7 to 10 and 5 to 6, respectively, indicating a moderate degree of instability.
Burst fractures result from a rapid increase in intradiscal pressure resulting in failure of the superior endplate, which is driven along with the disc into the vertebral body. The rapid increased intravertebral pressure creates hoop stresses and eventual failure of the body with radial displacement of bone fragments. In addition, the pedicles are pushed outward, causing vertical fractures in the posterior elements. A single fragment from the body displaces posteriorly into the spinal canal and can be incarcerated when the pedicles reapproximate. Often flexion is present, causing posterior ligamentous complex disruption and/or facet subluxation, fracture, or dislocation.
Burst fractures occur most commonly at C6 and C7 rather than the mid or upper cervical spine (see Fig. 77–4). On lateral views, both anterior and posterior vertebral body heights are shortened and a small fragment from the posterior superior body is rotated into the spinal canal. Interspinous widening or facet perching or subluxation, when present, indicate posterior ligamentous injury.
Burst fractures isolated to the anterior column have low CSISS (2-3) and SLIC (2) scores when not associated with neurologic injury, which is the usual case. Those associated with posterior ligamentous complex disruption have significantly higher CSISS (8-13) and SLIC (5-7) scores.
Flexion/axial loading injuries, also called tear drop fractures, are devastating injuries due to the propensity for neurologic injury and often are the result of diving or other sports-related activities. The injury occurs from a compression force obliquely applied in a downward and posterior direction. Forces are concentrated in the anterior inferior corner of the vertebral body, which is sheared off, giving the injury its name (see Fig. 77–3). The remaining part of the vertebral body shears through the disc space and rotates posteriorly into the spinal canal, crushing the spinal cord. Varying amounts of flexion strain occur in the posterior ligamentous complex, which accounts for a wide presentation of stability with this injury.
Flexion/axial loading injuries are often confused with burst fractures. In the former, the vertebral body is displaced posteriorly and not a fracture fragment as in the latter. Displacement of the main fragment in the flexion/axial loading injury occurs by shearing through the posterior half of the disc space. Further, a small triangular fragment (teardrop) is located in its normal position at the anterior inferior vertebral body margin. Injury to the posterior elements and facets is common including interspinous widening, bilateral lamina fractures, and lateral mass fractures. In more severe cases, they can be associated with facet fractures, subluxations, or dislocations.
In cases without posterior ligamentous disruption, CSISS scores range from 3 to 5 because of anterior displacement with minimal injury to lateral masses or posterior ligamentous complex. The SLIC score is 3 for a distraction injury with increasing scores being based on neurologic involvement. When associated with posterior ligamentous complex injury or facet fractures/subluxation, high scores on both CSISS and SLIC are present (see Fig. 77–3).
Transverse process fractures are quite common, occurring in isolation or in association with other more severe injuries. These fractures are not involved with spinal stability and therefore are not significant in treatment decision making for the spine. However, transverse process fractures at C6 and above may warn of possible vertebral artery injury.29 No consensus has been reached regarding whether routine vascular evaluation and subsequent treatment with anticoagulants are required in patients with these injuries. This topic is discussed in detail in Chapter 82.
Impacting the head or face in a fall or forward striking a windshield creates hyperextension, compression, and posterior shear. The anterior longitudinal ligament and disc annulus can fail from tension. Excessive compressive loading of the lateral masses and impaction of the spinous process may cause fracture. Displacement is mild, although up to 50% posterior translation can be seen. Tension in the anterior longitudinal ligament and annulus may avulse a small fragment, which can be confused with the teardrop fragment generated in flexion axial loading injury. The disc distraction injury occurs often in spondylotic spines. The forced hyperextension with compression can transiently narrow the spinal canal, causing spinal cord injury, typical of the central cord type.
A variety of fracture patterns are encompassed with this moniker. Typically the disc space is wider than normal or is hyperlordotic. This may be difficult to discern, especially in spondylotic spines. Retrolisthesis is usually present, although its magnitude, usually around 1 to 2 mm, is often underestimated in terms of its significance. Large degrees of displacement, up to 50% translation, can occur but are rare. Because of the injury to the anterior longitudinal ligament, soft tissue swelling will usually be present. Small avulsions of the anterior endplate can be confused with fractured osteophytes. Impaction of the spinous process can cause their fracture, also involving the lamina, which may be displaced into the spinal canal. Similarly, lateral mass fractures from impaction can occur.
MRI is useful to identify questionable cases. Classic findings are increased signal in anterior soft tissues both rostral and caudal to the injured disc. Increased signal intensity across the disc space transversing the anterior annulus and anterior longitudinal ligament is pathognomonic (Fig. 77–5). Rarely the posterior longitudinal ligament will be involved. Disc herniation and spinal cord compression from stenosis or malalignment may be present as well. Posterior increased signal in the facet joints and ligamentum flavum are indicative of more extensive injury and greater instability.
FIGURE 77–5 A, Lateral sagittal computed tomography (CT) of 57-year-old neurologically intact male with an ankylosed spine from diffuse idiopathic spinal hyperostosis who sustained a hyperextension disc distraction injury at C7-T1. The disc is hyperlordotic and laminar fractures are visualized. Widening between spinous processes indicates probable injury of the posterior ligamentous complex. The patient is neurologically intact. The Cervical Spine Injury Severity Score is 11 (anterior—4, left—2, right—3, posterior—2). The Subaxial Cervical Spine Injury Classification is 5 (distraction—3, DLC—2, neurologic—0). B, Left sagittal reconstruction. Comminuted facet fracture is seen. C, Right sagittal reconstruction demonstrates subluxation of the C6-7 facet and C7 pedicle fracture (arrow). D, Axial CT shows displaced laminar fracture of C7 (arrow). E, Postoperative lateral radiograph following posterior instrumentation of C4-T3. F, Postoperative anteroposterior radiograph.
Disc-distraction injuries are difficult to diagnose and are easily overlooked even with CT. MRI is essential if these lesions are suspected. They cause similar patterns of spinal cord injury recognized as central cord syndrome that are classically assumed not to have skeletal injury. In reality the majority of these probably have some component of disc distraction. Isolated disc distraction injuries of the anterior column usually have small amounts of subluxation or hyperlordosis, thus scoring low (1 to 3) on the anterior domain of the CSISS. Fractures or diastasis in the lateral masses or posterior elements are also minimally displaced or absent, thereby also scoring low, with an overall range in CSISS of 2 to 8, even in worst cases. The SLIC score reflects a more severe injury ranging from 3 to 6 with higher scores when spinal cord injury has occurred.
Anterior column injuries with low SLIC or CSISS scores can be treated non-operatively (Table 77–5). These include compression injuries, stable forms of burst fractures and flexion/distraction injuries, and most disc-distraction injuries (see Fig. 77–4). Less significant injuries can be treated with a cervical collar, while burst fractures and flexion/axial loading injuries can be treated with a cervical thoracic orthosis (CTO) or halo vest.
|Injury Type||Subtype||Treatment Options|
|Without posterior ligamentous injury||COLLAR, CTO|
|With posterior ligamentous injury||Collar, CTO|
|Without posterior ligamentous injury||Collar, CTO|
|With posterior ligamentous injury||Collar, CTO|
|Without posterior ligamentous injury||Collar, CTO|
|With posterior ligamentous injury||Collar, CTO|
|With facet fracture/ subluxation||Collar, CTO|
|Disc Distraction Injury|
|Without posterior ligamentous injury||COLLAR, CTO|
|With posterior ligamentous injury||Collar, CTO|
|Transverse Process Fracture|
|No vertebral artery injury||OBSERVATION|
|Vertebral artery injury||Observation|
CTO, cervicothoracic orthosis.
Authors’ recommendations are in bold capitals.
Surgical indications for anterior column injuries are those with evidence of disruption of the posterior ligamentous complex as demonstrated by high CSISS and SLIC scores (see Table 77–5). In the case of compression fractures when surgery is warranted, the author recommends posterior fusion because a single-level anterior fusion may fail in the face of a vertebral fracture at the location of caudal screw fixation. Burst fractures and flexion/axial loading injuries can be reduced or significantly improved with tong traction. Although both anterior and posterior approaches may be used, the authors recommend addressing pathology at its major location (i.e., anteriorly) (see Fig. 77–3). After strut grafting or insertion of a cage, an anterior plate is applied. Although a rigid locked plate has theoretical advantages over a translational plate, the author has not observed any clinical differences in outcomes between the two plates. In rare cases of extensive comminution, associated facet dislocations, or those having residual displacement of lamina fractures, combined anterior and posterior approaches are indicated.
Surgical indications for disc distraction injuries are cases with any significant displacement, ongoing neurologic symptoms, or chronic pain. Anterior discectomy and fusion are indicated in neurologically intact patients and those with radiculopathy. Patients with cord injuries require individualized care. If stenosis is limited to the injured level, then anterior cervical discectomy and fusion are recommended. Not uncommonly, patients will have developmental narrowing of the spinal canal and multilevel stenosis. Several options are available including single-level anterior cervical discectomy and fusion combined with posterior decompression (laminoplasty) or posterior decompression (laminoplasty or laminectomy) with posterior fusion.