Thoracic Spine Fractures and Dislocations

Background

Thoracic spine fractures, especially those resulting from high energy, can be devastating, often resulting in permanent neurologic injury. Neurologic deficit is encountered in 10-25% of all spinal column injuries, irrespective of the level of injury. A deficit occurs in 15-20% of all thoracolumbar injuries.

In the event of a complete neurologic injury, very few patients regain any useful motor function. Concomitant neurologic injury with spine fractures also adversely affects long-term survival. The 10-year survival rate for people younger than 29 years is 86%. This percentage drops precipitously to 50% for patients older than 29 years.

Documented treatment of spine fractures dates back several thousands of years. Closed treatment and manipulation to correct the sustained deformity were typically used. In the early 20th century, most treatment consisted of immobilization in hyperextension.

Treatment of spine fractures did not begin to evolve from universally closed treatments to the surgical modalities that are in place today until the advent of current anesthesia and radiographic techniques. Internal fixation was first seen after World War II. Initially, it was in the form of spinous process plating. Harrington then introduced his posterior spinal instrumentation. From this, modern surgical techniques and instrumentation developed.

Although the spinal stability and alignment established with these newer techniques have dramatically improved, there has been relatively little growth in the ability to improve the neurologic deficits sustained in these injuries over the years of spine fracture management.

For patient education resources, see Vertebral Compression Fracture.

• References

Anatomy

A thorough knowledge of thoracic spine anatomy is essential in the treatment of thoracic spine fractures. Twelve thoracic vertebrae exist. The normal thoracic spine has an inherent kyphotic curve ranging from 18° to 51°. The vertebral bodies are wedge-shaped, being larger posteriorly than anteriorly. The kyphosis of the thoracic spine results in a center of gravity anterior to the apical T7 vertebrae, resulting in compression anteriorly and tension posteriorly in the resting state.

Significantly less flexion capabilities exist in the thoracic spine relative to the cervical and lumbar spine. The C7-T1 articulation flexes approximately 9°, T1-6 flexes 4°, and T6-7 to T12-L1 gradually increases to 5-12°. Less lateral bending occurs within the thoracic spine as well. Lateral bending is approximately 6° per level for T1-10 and approximately 8° at the thoracolumbar junction.

Axial rotation is 8° from T1 to T8. This is largely due to the coronal orientation of the facets in the thoracic spine. The axial rotation of the lower thoracic spine and thoracolumbar junction is reduced to 2° below T10 because of the transition to facets that are more sagittally oriented than those seen in the lumbar spine.

The thoracolumbar junction is relatively susceptible to injury. Injuries in this region constitute 50% of all vertebral body fractures. The decrease in rib restraint is largely responsible for the susceptibility of this area to injury. Other factors include changes in stiffness in flexion and axial rotation and the changes in disk size and shape that occur at the transition between the thoracic and lumbar spine.

The terminal portion of the spinal cord, the conus medullaris, normally begins at the T11 level. It ends at the L1-2 disk space in males and slightly more proximally in females. The cauda equina emanates from this region and extends distally into the lumbosacral spine with each peripheral nerve root exiting at its corresponding neural foramen. The cauda equina is more resistant to injury and has greater potential for recovery than the spinal cord.

The diameter of the spinal canal is also of great significance in thoracic spine fractures. The canal diameter of the thoracic spine is narrower than that of the cervical and lumbar spine. At the T6 level, the long axis of the spinal canal is approximately 16 mm in diameter, whereas in the midcervical and midlumbar spine, the long axis is 23 mm and 26 mm, respectively.

These dimensions have ramifications regarding the smaller amount of space available before cord compression is sustained in the event of a thoracic spine fracture. In addition, the smaller diameter may make fixation techniques such as sublaminar wire fixation more difficult and, thus, a less desirable method of stabilization.

The orientation and shape of the pedicles in the thoracic spine are different from those of their lumbar counterparts and can often preclude pedicle fixation. The pedicle isthmus width is smaller in the thoracic spine than in the lumbar spine. The transverse angle is approximately 27° medial inclination from posterior to anterior in the proximal thoracic spine, decreasing to 1° at T11 and to –4° at T12.

• References

Etiology

The vast majority of spine fractures occur as a result of motor vehicle accidents (45%), falls (20%), sports (15%), acts of violence (15%), and miscellaneous activities (5%). The percentage secondary to acts of violence is higher in urban areas. The male-to-female ratio is roughly 4:1. For mechanisms of injury, see Presentation.

• References

Prognosis

The results are favorable for correction of deformity, maintenance of reduction, healing, and fusion rates. Overall clinical outcome is generally good, depending on the patient's final neurologic function. Return of neurologic function, however, is variable, with little significant recovery seen in complete injuries irrespective of treatment.

• References

Physical Examination

Upon initial presentation, an extensive physical examination should be performed and neurologic status documented. Concomitant injuries should be assessed.^[1]  The patient's overall physical condition should be optimized promptly. Once the patient is stabilized hemodynamically and other visceral injuries have been investigated and excluded, definitive treatment of the thoracic spine injury can be contemplated.^{[2, 3, 4]}

Several distinct classification schemes are available to assess spinal stability (see Classification below).^[5]  

• References

Classification

Fracture classification systems

Holdsworth initially proposed a model for assessing spinal stability that divided the vertebra into two columns as follows^[6] :

• Anterior column - Vertebral body, intervertebral disk, anterior longitudinal ligament (ALL), and posterior longitudinal ligament (PLL)
• Posterior column - Facets, neural arch, and interspinous ligaments

Disruption of one or both columns implies instability of the involved segment.

Denis expanded on this model to develop the most common model used for assessing spinal stability,^[7]  in which the vertebra is divided into three columns as follows:

• Anterior column - Anterior two thirds of the vertebral body, ALL, anterior two thirds of the annulus fibrosus
• Middle column - Posterior one third of the vertebral body, posterior one third of the annulus fibrosus, PLL
• Posterior column - Posterior ligamentous complex and posterior bony elements

When two of the three columns are disrupted, the fracture may be unstable.^[7]

Classification schemes generally also encompass mechanisms of injury and their resultant fracture patterns. Several different mechanisms of injury can occur within the thoracic spine. Most commonly, a combination of one or two mechanisms accounts for the injury. These mechanisms include the following:

• Axial compression
• Flexion
• Lateral compression
• Flexion-rotation
• Shear
• Flexion distraction
• Extension

Axial compression results in a purely compressive load. Endplate failure occurs, followed by vertebral body compression. With higher energy, a centripetal displacement occurs, resulting in what is commonly referred to as a burst fracture (see the image below). In severe burst fractures, disks become fragmented and the posterior elements are disrupted. Radiographically, this mechanism can manifest as a widened interpedicular distance.

View Image

Thoracic spine fractures and dislocations. Burst fracture T12. Note the widened interpedicular distance.

Flexion results in compression anteriorly. Disruption of posterior elements with flexion often results in instability of the involved area. If anterior compression exceeds 40-50%, the posterior ligamentous structures are often disrupted. Instability ultimately can result in progressive deformity and neurologic deficit if not appropriately stabilized.

Lateral compression usually results in a stable injury unless disruption of posterior structures or associated axial compression occurs.

With a flexion-rotation injury, posterior ligamentous structures commonly fail. Oblique disruption of the anterior vertebral body and disk failure occur. This type of injury can result in what commonly is known as a slice fracture. With fractures of the facets and concomitant disruption of posterior elements, thoracic spine dislocation can occur.

Shear injuries often result in severe ligamentous disruption and subsequent anterior, posterior, or lateral listhesis. Anterolisthesis is the most common of the three, with complete spinal cord injury (SCI) often being the unfortunate result. However, occasionally, concomitant fractures through the pars interarticularis result in autolaminectomy, with resultant neural sparing.

Flexion distraction injury (see the image below) is more commonly referred to as the seatbelt injury. The axis of flexion is anterior to the vertebral column. Osseous, disk, and ligamentous structures are disrupted, either alone or in combination. Combined osteoligamentous or purely ligamentous injuries can be present, and this injury occurs most commonly at the thoracolumbar junction. Bilateral facet dislocation can occur.

View Image

Thoracic spine fractures and dislocations. Flexion distraction injury with facet dislocation.

Extension places tension on the anterior longitudinal ligament, with compression occurring posteriorly. Facet, laminar, and spinous process fractures often occur. Most of these injuries are stable, provided that significant retrolisthesis does not occur.

In the Denis classification system, significant fractures are divided into the following three groups:

• Primarily axial load injuries, including compression and burst fractures
• Flexion-distraction injuries
• Fracture subluxation and/or dislocation (see the image below)

View Image

Thoracic spine fractures and dislocations. Fracture dislocation T2-T3.

The mechanism of failure of the middle column further differentiates the various types of fractures. The middle column is spared in compression fractures, yielding a stable fracture. It fails in compression with burst fractures, distraction in seatbelt injuries, and shear and/or rotation injuries. Fracture dislocations yield unstable injuries.

The Denis classification system has been criticized on the basis of its occasional inability to be provide an adequate distinction between stable and unstable fractures—for example, the "stable" burst fracture. In addition, biomechanical studies have brought into question the importance of the middle column. Recognizing this issue, McAfee expanded on the Denis classification scheme to further elucidate the distinction between stable and unstable fractures. His classification system emphasized the posterior ligamentous complex as a major factor in fracture stability.

Another shortcoming of structural or mechanistic classifications is that they often fail to take neurologic deficit into account. Significant neurologic injury implies instability irrespective of the fracture pattern, in that the spine has clearly failed to protect the neural elements.

Complete vs incomplete spinal cord injuries

Neurologic deficits in these injuries can range from normal neurologic function to complete SCI.^[8]  Complete SCI is defined as a complete loss of motor and sensory function below the level of the injury in question, whereas with an incomplete injury, there is some level of residual motor or sensory function below the level of the injury.

SCIs are typically classified according to the American Spinal Injury Association (ASIA) classification, which is a grading system used to help quantify the degree of neurologic impairment in patients with acute spinal trauma. The ASIA classifications includes the following five grades of SCI:

• ASIA A – Complete loss of motor and sensory function
• ASIA B – Complete loss of motor function with intact or partially intact sensory function
• ASIA C – Motor function preserved below the neurologic level, with more than half of the muscles below having a muscle grade less than 3
• ASIA D – Motor function preserved below the neurologic level, with at least half of the muscles having a muscle grade of 3 or higher
• ASIA E – Normal

Definitive classification of complete SCIs can only be done in a patient who is not in spinal shock.^[9]  Spinal shock is defined as a state of flaccid paralysis from disturbance of spinal cord function below the level of an SCI. This state typically lasts for approximately 48 hours, and its conclusion is defined as return of the bulbocavernosus reflex.

As mentioned earlier, complete SCIs are those in which motor and sensory function are completely absent, whereas incomplete injuries are those in which some level of motor or sensory function remains. There are a few classic patterns in which incomplete SCIs present, as follows.

Central cord syndrome

Central cord syndrome, the most common incomplete SCI, typically results from a hyperextension injury to the cervical spine in a patient with a stenotic spinal canal.

The injury occurs in the center of the spinal cord, damaging the lateral corticospinal tract. Because it is a more centrally based injury, it damages the axonal tract supplying the upper extremity (which is more central) before damaging the axonal tract supplying the lower extremity. Therefore, individuals with this type of injury typically present with more upper-extremity neurologic dysfunction than lower-extremity dysfunction.

The overall prognosis is good, especially in younger patients, with regard to the ability to ambulate. However, patients may have permanent upper-extremity dysfunction.^[10]

Brown-Sequard syndrome

Patients with Brown-Sequard syndrome, which is typically seen in the setting of penetrating trauma, exhibit impaired ipsilateral motor function with contralateral pain and temperature sensation dysfunction below the level of injury. Anatomically, this type of injury is due to the decussation of motor fibers occurring within the brainstem, while pain and temperature fibers decussate within the spinal cord itself in the spinothalamic tracts innervating the second level below the region in which the fibers cross.

Of patients with incomplete SCIs, these individuals typically have the best prognosis, more specifically with regard to motor functional outcomes.^[11]

Anterior cord syndrome

Anterior cord syndrome results from injury to the anterior two thirds of the spinal cord, which includes the corticospinal and the spinothalamic tracts. These two tracts contain nerve fibers that deal with motor function and with pain and temperature function, respectively. The mechanism of injury is typically direct compression from thoracic spine fractures or injury to the anterior spinal artery. Patients typically have motor dysfunction below the injury level with preserved deep pressure and proprioception.

The prognosis for these injuries is poor.^[12]

Posterior cord syndrome

Posterior cord syndrome, the least common incomplete SCI, involves injury to the posterior columns of the spinal cord. It typically results from a posteriorly based injury (eg, impact from a lamina fracture).^[12]

Thoracolumbar Injury Classification and Severity Score

Several classification systems have been devised for the evaluation of thoracolumbar trauma; however, the most commonly used scheme is the Thoracolumbar Injury Classification and Severity (TLICS) score (also sometimes known as the Thoracolumbar Injury Severity Score [TISS]). The TLICS was developed with the goals of improving the care of patients with these injuries and providing a system that could be applied to everyday practice. It has subsequently been shown to be valid and clinically useful for determining the final treatment of these injuries.

Briefly, the TLICS takes three injury characteristics—injury morphology, neurologic status, and posterior ligamentous complex integrity—and assigns points to each characteristic on the basis of severity. These points are summed to yield a total score, which is then used to guide treatment. An injury with a score higher than 4 is typically managed surgically; an injury with a score lower than 4 is typically managed without surgery; and an injury with a score of 4 may be managed with or without surgery, at the discretion of the surgeon.^{[13, 14]}

• References

Imaging Studies

Radiography

Anteroposterior (AP) and lateral radiographs of the thoracic spine can be obtained initially as part of the evaluation when a spine fracture is suspected; however, most centers proceed with computed tomography (CT), if available, when a spine fracture is suspected. Radiographic evidence of a fracture at any level of the spine warrants radiographic analysis of the entire spine, particularly in high-energy injuries, to evaluate for noncontiguous injury.^[15]

Computed tomography

CT of the thoracic spine with sagittal reformatted images provides information about the extent of injury to osseous structures and posterior elements.^[16] Smith et al found that nonreconstructed abdomen and pelvis CT detected thoracolumbar spine fractures more accurately than plain radiography did and recommended CT for diagnosis of such fractures in acute trauma patients with altered mental status.^[17] Although many centers will obtain dedicated spine films, Imran et al showed that CT of the chest, abdomen, and pelvis did not miss clinically significant fractures.^[18]

Magnetic resonance imaging

Magnetic resonance imaging (MRI) is useful for evaluating soft-tissue injury to the ligaments, disks, and epidural spaces.^[19] MRI is most useful in cases where traumatic disk herniation, epidural hematoma, or spinal cord injury (SCI) is suspected. In addition, MRI is used when CT and radiographic analysis do not adequately explain the patient's symptoms or neurologic findings.^[20]

• References

Approach Considerations

The primary goals of treatment for thoracic spine fractures include protecting the neural elements and preventing deformity and instability. Surgery often facilitates achieving these goals and often hastens the patient's rehabilitation. Hospital stays are often shorter with surgery. Surgery is particularly often beneficial in patients with multiple traumatic injuries. Ultimately, the decision whether to operate is based on many factors, including fracture morphology, and the choice is often complex (see Presentation).

In general, stable fracture patterns in a neurologically intact patient can be treated nonoperatively. Indications for surgery can vary and include significant neurologic deficit and fracture subluxations. Excessive deformity is also an indication, though defining this is difficult, and the effect of kyphosis on long-term results is uncertain. Kyphosis greater than 30º may be associated with poorer long-term results, and kyphosis greater than 25º is often mentioned as a relative indication for surgery.

The presence of other injuries also may affect the choice between operative and nonoperative treatment. The most predictable benefit of surgery is more rapid mobilization, which can be an important consideration in the patient who has experienced multiple traumatic injuries.

Relatively few contraindications exist for operative stabilization of unstable thoracic spine fractures. Patients who are unstable medically with thoracic spine fractures requiring operative intervention should not undergo surgical stabilization. Once the patient is in optimal medical condition, surgery should be undertaken. Operative intervention for thoracic spine fractures is also contraindicated in the presence of an active infection.

Guidelines on management of thoracolumbar burst fractures are available from the Congress of Neurological Surgeons,^[21] and guidelines on management of spinal cord injury (SCI) are available from AOSpine^{[22, 23, 24, 25, 26, 27]}  (see Guidelines).

Timing of surgical treatment

The literature regarding the timing of surgical intervention for thoracic and lumbar fractures with an acute spinal cord injury (SCI) is scarce. However, several studies have looked at the timing of surgery in cervical spine trauma.

In particular, the Surgical Treatment of Acute Spinal Cord Injury Study (STASCIS), a prospective multicenter trial, examined the results of early vs late decompression surgery in cervical spine trauma.^[28] Early decompression was defined as less than 24 hours, and late was defined as after 24 hours. A total of 313 patients were evaluated on the basis of cervical spinal cord compression identified on advanced imaging. The primary outcome studied was change in the American Spinal Injury Association (ASIA) Impairment Scale (AIS) from presentation to 6-month follow-up.

In this study, a higher rate of significant neurologic recovery was identified in the early decompression group.^[28] However, these findings must be considered within the overall clinical context for each patient, in that these patients can have multisystem injuries that may preclude immediate operative intervention. A multidisciplinary approach should be followed in treating these patients and in determining the most appropriate time to operate.

• References

Nonoperative Therapy

Nonoperative treatment begins with pain management and attention to concomitant injuries. Mobilization with bracing can then begin if nonsurgical treatment is chosen. Use of the three-column rule can be helpful in determining brace types.

Single-column injuries (eg, compression fractures involving only the anterior column) are generally stable and can be treated with a simple extension orthosis to limit flexion. If contiguous fractures are encountered, the cumulative compression and angular deformities are considered in choosing between operative and nonoperative treatment. Isolated posterior-element fractures are usually stable, and conservative treatment with mobilization is appropriate. Light bracing can be used with these injuries for comfort and to hasten mobilization.^{[29, 30, 31, 32, 33]}

More severe injuries with two-column involvement require more rigid immobilization. Standard thoracolumbosacral orthoses (TLSOs), such as the Boston brace, provide good immobilization, but only of the lower thoracic spine. The usefulness of TLSOs is limited to injuries from about T7 distally.

Extension of the brace to the cervical spine (cervical thoracolumbosacral orthoses [CTLSOs]) can allow immobilization of upper thoracic segments; however, these braces are very poorly tolerated by patients. Upper thoracic spine injuries are more difficult to treat with bracing; therefore, fracture instability is a relative indication for surgical stabilization.

Soliman et al evaluated 695 isolated compression fractures to the thoracic or lumbar spine, of which 195 derived from auto accidents (age range, 19-82 years; male-to-female ratio, 60:40).^[34] No patient with compression of less than 40% underwent surgery unless he or she presented with a neurologic deficit. Patients without neurologic deficit underwent surgery only if they had compression of 40% or higher. The investigators found that the use of a TLSO in patients with less than 40% compression was of no value in terms of outcomes; the with-brace and without-brace groups had similar outcomes.

The treatment of burst fractures of the thoracic spine and the thoracolumbar junction is an area of debate. Surgical advocates believe that surgery allows earlier mobilization and return to function, more pain relief, and better correction of any kyphotic deformity that exists. Studies have failed to show a significant difference in results in patients without neurologic injury as long as significant posterior-column injury is not present.^[35]

Significant remodeling of the spinal canal has been shown to occur within the first year in burst fractures treated nonoperatively. Residual kyphosis is also seen, but the degree of kyphosis present does not correlate with the patient's pain or functional abilities.^{[30, 31, 32, 36, 37, 38, 39, 40]}

Additional studies have reported similar or even more beneficial results with nonoperative as opposed to operative treatment of thoracic spine fractures, both with and without neurologic deficit. No correlation has been shown between neurologic deficit and the extent of canal compromise or, more important, between the resolution of the deficit and surgical decompression. In addition, nonoperative treatment eliminates the risk of postoperative infection, which ranges from 7% to 15% in various studies.

If neurologic deficit (spinal cord) is present and less than 8 hours has elapsed from the time of injury, treatment with high-dose methylprednisolone (5.4 mg/kg bolus followed by 30 mg/kg/hr infusion for 23 hours) is an option (see below for more details). Operative versus nonoperative treatment can be entertained, depending on the clinical status of the patient and radiographic appearance of the fracture. The stability of the fracture, its location, and the underlying mechanism of injury all can play major roles in the decision whether to operate or treat conservatively.

If immobilization with prolonged bed rest is chosen as the method of treatment, strict deep venous thrombosis (DVT) prophylaxis, the use of a kinetic bed, vigilant inspection for decubitus ulcers, and aggressive respiratory therapy must be implemented to prevent the complications that can arise with bed rest.

Flexion-distraction injuries involving significant disruption of the supporting ligamentous structures are generally unstable and are managed surgically.

Steroid administration in trauma setting

Administration of methylprednisolone has been the topic of debate in the spine literature for quite some time; whether its efficacy and potential benefit outweigh its adverse effects has been a source of controversy in the treatment of SCI.^{[41, 42, 43, 44, 45, 46]}

Methylprednisolone is a synthetic corticosteroid that acts to inhibit proinflammatory cytokine production. In the spine, its proposed effects include inhibition of lipid peroxidation, prevention of progressive ischemia development, reversal of intracellular calcium accumulation, and various other effects.^[47]  The use of methylprednisolone in acute SCI has been extensively studied, most notably in National Acute Spinal Cord Injury Study (NASCIS), which sought to evaluate treatment strategies for reducing the effects of SCI. There were three different NASCIS trials, as follows:

• NASCIS 1 studied the effects of high-dose versus low-dose methylprednisolone for the treatment of acute SCI^[41] ; no difference between the two groups was found with respect to neurologic recovery of motor function or light touch sensation at 6 weeks or 6 months post injury, but wound complications and infections were found in the high-dose group  
• NASCIS 2 examined the effects of an initial bolus of methylprednisolone followed by infusions consisting either methylprednisolone, naloxone, or placebo^[42] ; initial analysis failed to identify any significant neurologic benefit in the steroid group, but subgroup analysis of individuals receiving methylprednisolone within 8 hours of injury demonstrated statistically significant motor and sensory improvement in short- and long-term follow-up
• NASCIS 3 sought to investigate the efficacy of high-dose methylprednisolone for 24 versus 48 hours^[44] ; significant results were found when methylprednisolone was given within 3-8 hours and continued for 48 hours

The results of these studies have been criticized by the medical community. For instance, it was noted that NASCIS 1 had no control group and that it reported a significant amount of side effects. Other criticisms focused on the methods of statistical analysis used. It was pointed out that in NASCIS 2, it took a secondary analysis of a patient subgroup to obtain a positive effect of methylprednisolone. It was also noted that the placebo group treated before 8 hours did poorly, in contrast to the placebo group treated after 8 hours. These findings raised questions about the validity of the subgroup analysis. Similar statistical anomalies were found in NASCIS 3, further fueling the criticism of these studies.

In summary, methylprednisolone is a treatment option in the setting of acute SCI; however, it is not the current standard of care, nor is it routinely used by all spine surgeons. Some medical centers have abandoned the use of methylprednisolone for this application altogether, whereas others use it intermittently. The standard protocol consists of administration within 8 hours of SCI in the form of an initial bolus of 30 mg/kg given over 15 minutes followed by infusion at a rate of 5.4 mg/kg/hr for 23 hours post injury.^[48]

• References

Surgical Therapy

Surgical options

If surgical management is chosen, the next step is to determine the most appropriate approach: anterior, posterior, or both.^{[49, 50, 51, 52, 53]} Many factors, including fracture morphology and neurologic status, can play a role in this decision. Patients with complete neurologic deficit who are no longer in spinal shock have very little chance of significant neurologic recovery. The primary goals of surgery in this group are realignment and stabilization, typically via a posterior approach.^{[49, 50, 51, 52]}

When partial neurologic deficit is present, improving residual canal compromise is also a goal of surgery. This situation most typically occurs with burst fractures (see the image below). If performed early enough (generally within 72 hours), posterior instrumentation allows distraction and correction of sagittal alignment and successful indirect decompression of the spinal canal. Laminectomy with transpedicular decompression also can improve the canal clearance achieved through a posterior approach.

View Image

Thoracic spine fractures and dislocations. Preoperative axial CT image of burst fracture with partial neurologic deficit.

Laminectomy should never be performed alone in the treatment of thoracic burst fractures. Another option is anterior decompression and fusion with instrumentation. Surgeon preference often plays a role, as does fracture morphology. Concomitant lamina fractures with posterior canal compromise generally necessitate beginning with a posterior approach because of possible neural entrapment and dural tears.^[54]

Flexion-distraction injuries result in disruption of the posterior and middle columns in tension. Very often, the anterior column remains intact, acting as a hinge; however, it may fail in compression. Surgical intervention for these fractures typically involves a posterior approach. To preserve the intact anterior column, anterior approaches are not routinely used in these injuries.

Fracture-dislocation injuries result in disruption of all three columns and, as a result, carry a high incidence of complete SCI. Therefore, the main objective of surgical intervention is solely to provide posterior stabilization facilitating early mobilization and rehabilitation. Anterior decompression and stabilization are performed after posterior surgical realignment of the fracture in rare cases where partial neurologic deficit exists in the presence of significant anterior neural compression.

Various methods exist for surgical stabilization, as do many opinions and accounts in the literature supporting the numerous techniques. Harrington rods had been used for many years to stabilize the spine with unstable fractures. The main disadvantage of Harrington rod instrumentation is that it requires two to three levels above and below the injured segment for stability. Additionally, it performs relatively poorly in three-column injuries because of the predisposition to overdistraction and the relatively high incidence of rod breakage and hook cutout.^[55]

Hybrid constructs consisting of spinous process and sublaminar or Luque wires provide segmental fixation with improved results. A disadvantage of this mode of fixation is the risk of neurologic injury with sublaminar wire passage and wire migration. Because of these potential complications, many surgeons do not routinely use sublaminar wires in patients with incomplete neurologic injuries or normal neurologic status.

Harrington instrumentation has been supplanted by segmental instrumentation systems initially developed for scoliosis. These systems use multiple fixed anchors along the fixation rod. Multiple forces can be applied at different points, resulting in a relatively low incidence of fixation failure. Compression, distraction, and translation are all possible within the same construct. Initially, these systems used hooks (sublaminar, pedicle, and transverse process) for fixation, and they now allow for pedicle screw fixation as well.

Pedicle screw fixation allows instrumentation of vertebrae with fractured or absent laminae. In addition, it allows rigid bony purchase through all three columns. Because of this increased rigidity, fewer segments may be needed for stable fixation, allowing preservation of more motion segments. Preserving motion segments is less important in the thoracic spine; little motion is lost in comparison with the cervical and lumbar segments. However, limiting instrumentation of distal segments is important with thoracolumbar injuries.^{[36, 56, 57, 58]}

Thoracic pedicle screw placement can be challenging because of the smaller dimensions of the thoracic pedicle as compared with the lumbar pedicle.^[59]  At some institutions, cortical disruptions have been reported to occur as often as 50% of the time when standard fluoroscopic techniques are used. Computer image guidance is useful in dealing with difficult anatomy, as in the placement of thoracic pedicle screws and in rotational deformities. However, a clear role in spine trauma management has not been established.

Fischer et al evaluated the feasibility and accuracy of minimally invasive transpedicular screw placement in 35 cervicothoracic fractures (28 traumatic, three pathologic, three infectious, and one osteoporotic) with the help of computed tomography (CT)-controlled guide wires (293 guide wires were inserted).^[60]  They found that treating vertebral fractures with a guide wire–based pedicle screw insertion technique under CT imaging yielded very high accuracy and a low complication rate.

The osseous structures are generally fused concomitantly with posterior instrumentation. Some surgeons fuse only the injured vertebral segments and subsequently perform staged removal of hardware as opposed to fusing the entire length of the instrumentation. With modern segmental fixation, fewer segments must be instrumented to provide stability. In the main thoracic spine, where motion preservation is less critical, it is common for the entire instrumented region to be fused.^{[61, 62]}

Minimally invasive surgical techniques with percutaneous pedicle screw placement has gained popularity over the past few years. Such techniques aim to minimize soft-tissue injury and perioperative morbidity. Early studies demonstrated shorter operating times, shorter hospital stays, reductions in intraoperative blood loss, and lower infection rates.^[63]  The development of navigation systems has reduced insertion time and radiation exposure while maintaining accurate screw placement.^[64]

Individual anatomic factors and posterior element fracture morphology can affect the surgeon's choice of anchors. In the thoracic spine, it is not uncommon for pedicles to be too small to allow screw placement. In these situations, hooks or bands may be considered. Generally, depending on the injury, two or three segments of fixation above and below the level of injury are required if hooks or bands alone are used. With pedicle screws, fixation often can be limited to one or two segments. The image below shows a burst fracture after stabilization.

View Image

Thoracic spine fractures and dislocations. Burst fracture with partial neurologic deficit after stabilization with medial resection of right pedicle t....

View Image

Thoracic spine fractures and dislocations. Pedicle screw fixation of a T12 burst fracture.

The condition of the anterior column also can affect instrumentation choices. If severe comminution or kyphosis is present anteriorly, extending the length of the posterior instrumentation or improving anterior support should be considered. This is often an issue with burst fractures. Historically, transpedicular bone grafting also was performed in an attempt to improve the anterior column. Studies have shown little difference with this technique in regard to hardware failure and final vertebral height. Thus, in unstable fracture patterns with anterior-column involvement, dorsal stabilization with concomitant or staged anterior interbody fusion provides a more stable construct, with improved maintenance of reduction.

Anterior instrumentation systems also have been developed for the treatment of spinal fractures. Their use often requires reconstruction of the anterior column with strut grafting, cages, or both. Historically, anterior instrumentation also required the use of posterior instrumentation because of the lack of stability of the older fixation systems. Newer anterior systems, however, have been developed that provide enough structural stability to allow them to be used alone. They are extremely rigid, and some have been shown to provide greater torsional stiffness than the intact spine. Biomechanical studies have shown that this type of fixation can be equal in strength to a two-above and two-below pedicle screw construct.

Advantages of the anterior approach include direct neural decompression and preservation of motion segments at the thoracolumbar junction. However, because of the morbidity of this approach and advances in posterolateral approaches to the spine, posterior fixation techniques are much more commonly used than the direct anterior approach.

The timing of surgery is also an important issue in the treatment of thoracic spine fractures. Progressive neurologic deficit in the presence of continued canal compromise is an accepted indication for immediate decompression and stabilization. Quite often, patients with thoracic spine fractures have concomitant injuries, making the timing of spinal stabilization difficult to plan.

Some studies suggest that patients with thoracic spine fractures treated within 72 hours, irrespective of concomitant injuries, do much better physiologically after the operation than those in whom stabilization is delayed. Early fixation results in less time in the intensive care unit, reduced need for ventilator support, fewer pulmonary complications, and a shorter overall hospital stay.^[65]

Preparation for surgery

Upon initial presentation, an extensive physical examination should be performed and the patient's neurologic status documented. Concomitant injuries should be assessed, and the patient's overall physical condition should be optimized promptly. Next, a thorough evaluation of the fracture pattern with appropriate radiologic studies is necessary to select the appropriate type of instrumentation to be used.

Operative details

Care must be taken in positioning patients for surgery after induction of anesthesia. Intraoperative radiographs should be obtained to assess hardware placement and adequacy of reduction. In patients without neurologic deficit or with a partial deficit, neurologic function should be monitored during surgery with intraoperative evoked potentials (motor and sensory) as the patient's condition allows.

Determining the adequacy of decompression can be difficult if a posterior approach is chosen. Plain films can be helpful, and pedicle resection can allow anterior access without cord manipulation. Intraoperative spinal sonography (IOSS) and CT can also be used to evaluate for residual compression.

• References

Postoperative Care

Early mobilization and rehabilitation are essential to decrease postoperative complications and to achieve the highest level of functional status attainable. Serial neurologic examinations are performed in the acute postoperative setting to assess for changes in neurologic status. Adequate stabilization is often achieved with instrumentation alone, though postoperative bracing sometimes may be required. If a partial neurologic deficit persists, follow-up CT may be performed to evaluate the adequacy of the decompression.^[66]

• References

Complications

Even with careful preoperative planning and meticulous surgical technique, complications can occur during surgical treatment of a thoracic spine fracture. DVT, pulmonary embolism, urinary tract infections, and even death can occur with any surgical procedure, and measures should be taken to prevent such complications.

Neurologic injury can occur during spine surgery; the incidence is variable with reports ranging from 0.46% to 17%.^[67]  Injury can occur as a result of overdistraction, overcompression, or insertion of various forms of instrumentation.

Dural tears can occur during exposure, instrumentation, or decortication, and they may also be a result of the traumatic injury. In the case of dislocation and traumatic laceration, realignment of the spine often results in stopping the cerebrospinal fluid (CSF) leak, with no need for further specific treatment of the dura.^[68]  If repair is deemed necessary, the full extent of the tear should be completely exposed, and primary repair should be attempted if possible. Muscle or fascial grafts can be used for large tears that are not amenable to primary repair. Lumbar transdural drains can be placed to decrease pressure across the tear and facilitate healing.^[68]

Infection can occur as a result of surgical treatment. Infections superficial to the fascia can be treated with debridement with packing or closure over a drain. Infections deep to the fascia require prompt surgical debridement with retention of bone graft and instrumentation. The wound can be serially debrided or closed over deep drains or over an inflow-outflow system providing constant irrigation of the wound. A 6-week course of intravenous antibiotics followed by a course of oral antibiotics is routinely administered in conjunction with these measures.

• References

Long-Term Monitoring

With surgically corrected thoracic spine fractures, early follow-up examination to assess wound healing is necessary within the first few postoperative weeks. Subsequent clinical examinations to assess functional status and neurologic function, as well as radiographic examinations, should take place frequently over the first year, followed by annual examinations thereafter if necessary. Significant loss of correction, change in neurologic function, or increase in pain level warrants further workup.

Nonoperative treatment of thoracic spine injuries requires close clinical and radiographic follow-up. Two-column injuries generally require 3 months of bracing, at which point weaning can begin. Activities are often restricted (eg, no lifting of weights exceeding 20 lb [~9 kg] and no impact activities) for 5-6 months.

• References

Congress of Neurological Surgeons Guidelines on Thoracolumbar Burst Fractures

In October 2018, the Congress of Neurological Surgeons (CNS), with the Section on Disorders of the Spine and Peripheral Nerves in collaboration with the Section on Neurotrauma and Critical Care, issued clinical practice guidelines regarding (1) the role of fusion in instrumented fixation in the treatment of patients with thoracolumbar burst fractures and (2) the role of percutaneous instrumentation in the treatment of these patients.^[21] Recommendations were as follows:

• In the surgical treatment of patients with thoracolumbar burst fractures, surgeons should understand that adding arthrodesis to instrumented stabilization (a) has not been shown to impact clinical or radiologic outcomes and (b) is associated with increased blood loss and longer operating times.
• In the treatment of thoracolumbar burst fractures, stabilization using both open and percutaneous pedicle screws may be considered; the evidence suggests that the two approaches have equivalent clinical outcomes.

• References

AOSpine Guidelines for Spinal Cord Injuries

In 2017, AOSpine (a part of AO [Arbeitsgemeinschaft für Osteosynthesefragen]) published a series of systematic review articles in the Global Spine Journal aimed at providing evidence-based guidelines for the management of spinal cord injuries (SCIs). These articles addressed several controversial topics, including optimal surgical timing, the use of corticosteroids, the type and timing of anticoagulation prophylaxis, the role of magnetic resonance imaging (MRI), and the type and timing of rehabilitation.^[22]

Optimal timing of surgical decompression

The guidelines recommended that early surgery (defined as ≤ 24 hr after injury) be considered for adult patients with traumatic central cord syndrome and other adult acute SCI patients.^[23] The level of evidence for these recommendations was low. Patients who underwent early decompression were more likely to demonstrate a clinically significant improvement in neurologic status (≥ 2 grade improvement on the American Spinal Injury Association [ASIA] Impairment Scale [AIS]) than those decompressed later (> 24 hr after injury); however, this relation was not statistically significant.

Corticosteroid use

The guidelines recommended against the use of high-dose methylprednisolone sodium succinate in adult patients who present more than 8 hours after an acute SCI; however, they recommended offering a 24-hour infusion of high-dose methylprednisolone sodium succinate to patients presenting within 8 hours after the injury.^[24] A small long-term neurologic benefit was found with the use of steroids in early treatment. The level of evidence was moderate for both of these recommendations. Additionally, the guidelines recommended against offering a 48-hour infusion of high-dose methylprednisolone sodium succinate, but no literature was cited to support this recommendation.

Anticoagulation prophylaxis

The guidelines recommended routinely offering chemical anticoagulation for thromboprophylaxis to reduce the risk of thromboembolic events in the acute period after SCI treated with or without surgery; either subcutaneous low-molecular-weight heparin (LMWH) or fixed or low-dose unfractionated heparin (UFH) was suggested, but dose-adjusted heparin was not recommended, because of the potential for increased bleeding.^[25] Such chemical thromboprophylaxis should be initiated within 72 hours post injury.

These recommendations were given a grade of low evidence.^[25] Of the studies used in the systematic review, none were able to reach statistical significance for the use of chemical thromboprophylaxis to reduce the likelihood of deep vein thrombosis (DVT), pulmonary embolism (PE), mortality, or increased risk of bleeding; however, the authors concluded that chemical thromboprophylactic therapy is superior to no prophylactic treatment for DVT.

Magnetic resonance imaging

The guidelines recommended performing MRI in adult patients with acute SCIs before surgical intervention when possible to aid in clinical decision-making.^[26] They also recommended that MRI be performed before or after surgery to improve prediction of the neurologic prognosis. The level of evidence for both of these recommendations was low.

Rehabilitation

The guidelines recommended that rehabilitation be offered to patients with acute SCI upon medical stabilization and subject to patients' tolerance of rehabilitation protocols.^[27] No specific studies were cited to support this recommendation, but low-level evidence was available to support several other recommendations. The guidelines recommended offering body weight–supported treadmill training if the resources and expertise are available, as well as electrical therapy to improve hand and upper-extremity function. Offering additional training in unsupported sitting beyond what is currently incorporated in standard rehabilitation programs was not recommended.

• References