Summary of Key Points
- •
Systems of describing spinal column injuries have historically focused on fracture classification and morphology. With advancements in our understanding of spinal biomechanics, modern classification systems have arisen that emphasize injury mechanisms as well as clinical evaluations.
- •
The Thoracolumbar Injury Classification and Severity Score (TLICS) factors the presence of a neurologic deficit, fracture morphology, and integrity of the posterior ligamentous complex to assess the need for surgical stabilization of a thoracolumbar injury.
- •
The new AOSpine Thoracolumbar Spine Injury Classification System is based on evaluating three basic parameters: morphologic classification of the fracture, neurologic status, and patient-specific modifiers. This classification adds clinical aspects, which can better guide fracture management when combined with a severity score in the future.
- •
The subaxial injury classification, also known as the subaxial cervical spine injury classification system, is much like the TLICS based on injury morphology, discoligamentous complex injury, and the neurologic status of the patient.
Spinal cord injury (SCI) frequently leads to devastating neurologic deficits and disability with an estimated United States prevalence in 2013 of approximately 273,000 persons, with some studies estimating the prevalence to be as high as 332,000 persons. Hence, there has been great interest and effort to define and optimize classifications of spine injuries to facilitate communication, achieve optimal treatment, and maximize neurologic recovery.
The spine community has had a long inherent tradition in the use of classification systems. This chapter discusses spinal cord injury classification systems including the thoracolumbar and cervical spine injury classification schemes.
Spinal Cord Injury Classification Systems
In 1969 in the journal Paraplegia, Frankel and colleagues attempted to define spinal cord injuries ( Table 126-1 ). In this review that took place over a 19-year period, Frankel examined and grouped 682 spinal cord injury patients into five categories. Patients were categorized from the most severe injury or complete spinal cord/cauda injury (Frankel A) to the least severe injury or patients with only a brief, transient deficit (neurologically intact patient, Frankel E). This was the basis for future SCI classifications.
| Type A | Complete spinal cord/cauda injury |
| Type B | Only sensation present |
| Type C | Motor present, but useless |
| Type D | Motor useful |
| Type E | Neurologically intact |
The need to further classify and define injury types in patients became more apparent with the initiation of treatments to improve neurologic recovery with spinal cord injury trials. In 1978, Bracken and colleagues used a modified Frankel scale in an in-depth review of spinal cord injury. They used the Bracken scale with a five-scale motor examination and seven-scale sensory examination. This classification system did not gain widespread use, as there was the elimination of a bowel and bladder examination.
ASIA Classification
In 1982, the American Spinal Injury Association (ASIA) expanded on the Frankel scale with the implementation of a 0-5 motor scale of 10 predefined motor groups, which represented specific motor root distributions. This ASIA scoring system was in use for roughly a decade until 1992, when ASIA in association with the International Medical Society of Paraplegia (now the international Spinal Cord Society) further included the use of the functional impaired measurement (FIM) scale. Thus, in the updated 1996 version, the modified ASIA classification included the ASIA impairment scale, ASIA motor scale, ASIA sensory score, and FIM outcome scale. Further revisions of the standards were completed in 2000 and 2010. Figure 126-1 shows the revised 2013 edition.
Spinal Injury Classification Systems
To be helpful for clinicians, patients, and society, classification systems need to have a uniform method of description in such that they allow for direction of treatment. Moreover, they should be reproducible, usable, accurate, and comprehensive.
Thoracolumbar Classification
Of all the classification systems developed for the spine, thoracolumbar fractures have been the most studied. Original systems arose out of the ability to describe fracture morphology. Physicians at this time were struggling to understand what constituted to instability, thus requiring surgery. As such, Watson-Jones in 1943 emphasized the concept of “instability” and presented the first morphologic classification of thoracolumbar injuries. In this manuscript, he reviewed 252 patient radiographs. He identified seven discrete types of fractures, which are the fracture types we commonly describe today such as wedge fractures (compression fractures), comminuted fractures (burst fractures), and fractures dislocations. However, a setback of pure morphologic classification schemes is that they do not take into account the clinical presentation.
In 1949, Nicoll expanded the thoracolumbar classification system in an attempt to find a “stable” versus “unstable” thoracolumbar injury. In this review of 152 coalminers’ radiographs for which Nicoll provided primary care, he initially described the importance of the posterior longitudinal ligament such that he was able to define fractures as either stable or unstable based on the fact that the posterior ligamentous complex was intact. Intact posterior ligamentous complexes were deemed to be stable fractures.
In the 1960s through the 1980s, further modification occurred to thoracolumbar classification based on arbitrary compartmentalization of the spine. In 1970, Holdsworth divided the thoracolumbar junction as “a two-column structure”: the anterior column consisting of the vertebral body and intervertebral disc, and the posterior column consisting of facet joints and the posterior ligamentous complex. The stability was based on the ability of these columns to retain loads.
However, there was a failure in this modeling, and numerous fractures that were defined as unstable were, in fact, stable fractures. For example, unstable burst fractures were falsely categorized as stable. Therefore, numerous authors devised a “three-column” theory and a number of classification schemes. Of these classification schemes, the Denis classification became the most widely adopted due to the rise in use of computed tomography (CT). Denis astutely recognized modern technology and adapted the CT scanner to define fracture morphology and aid his classification scheme. In the Denis classification, the spine is subdivided into an anterior, middle, and posterior column. The anterior column was defined as the anterior vertebral body, anterior annulus fibrosus, and anterior longitudinal ligament. The middle column consists of the posterior wall of the vertebral body, posterior annulus fibrosus, and posterior longitudinal ligament (PLL). The posterior column consists of everything posterior to the PLL, including the pedicles, lamina, facet joints, and spinous process. Denis categorized four distinct fracture types: compression fractures, burst fractures, fracture-dislocations, and seatbelt injuries. The Denis classification scheme recognized the importance of the middle column, in that fractures with violation of the posterior vertebral body wall would be defined as burst type fractures.
Unfortunately, there has been much misunderstanding about this classification system and its interpretation in clinical practice. For example, the Denis classification does not state that a fracture is unstable with a three-column violation requiring surgical treatment. Several prospective randomized studies have shown that these fractures are stable and can be treated quite well with nonoperative therapy.
In 1994, McCormack and colleagues published an important paper documenting the significance of “anterior column support.” In this classification scheme, they determined fractures on a 1- to 9-point scale known also as the “Load Sharing Score.” Fractures were defined as more severely injured if they had more comminuted fractures and a wider dispersion of the fractures’ displacement as well as the ability to correct kyphosis. Thus the patients had an unstable spine if they had not anterior column support as shown by the severely dispersed fracture fragments as well as the spine could be easily manipulated surgically (showing no support structures).
In 1994, Magerl and colleagues, through the AOSpine (Arbeitsgemeinschaft für Osteosynthesefragen Spine) Society, developed the thoracolumbar mechanism classification. This comprehensive classification was developed after a review of 1445 cases. It is based on the mechanism and the direction of internal forces. There are three broad classification schemes: type A refers to compression type injuries, type B refers to distraction type injuries, and type C refers to rotational injuries. These are all based on increasing severity. The unfortunate issue with this classification scheme is that there are 53 separate subtype injury patterns, making the overall system confusing and the interrated variability less than ideal. In fact, Oner and coworkers, in an interview of 53 patients using CT and magnetic resonance imaging (MRI), noted that the interobserver reliability for the AO classification was 0.28 with MRIs and 0.31 with CT scans, whereas the Denis classification was 0.6 for CT scans and 0.52 for MRIs.
Due to the lack of cohesiveness, prognostic ability, and comprehensiveness of the aforementioned classifications of the thoracolumbar spine, Vaccaro and colleagues, in association with the Spine Trauma Study Group, developed the Thoracolumbar Injury Classification and Severity Score (TLICS). This classification system was developed on the numerous merits of previous classifications, specifically the importance of the injury morphology, neurologic status, and the integrity of the posterior ligamentous complex ( Table 126-2 ). Based on these three separate areas, each fracture is given and rated points to determine whether it is operative or nonoperative ( Table 126-3 ). However, there is a predefined “gray zone,” such as severity score 4, which permits surgeons to use individual clinical judgment to determine surgical options. Another major drawback is that the TLICS is based on MRI for evaluating the integrity of the posterior ligamentous complex, suggesting that its specificity can be as low as 50%.
| M orphology | |
| No abnormality | 0 |
| Compression | 1 |
| + Burst fracture | 2 |
| Rotation/translation | 3 |
| Distraction | 4 |
| P osterior L igamentous C omplex | |
| Intact | 0 |
| Suspected/indeterminate | 2 |
| Injured | 3 |
| N eurologic S tatus | |
| Intact | 0 |
| Root injury | 2 |
| Complete cord/conus medullaris injury | 2 |
| Incomplete cord/conus medullaris injury | 3 |
| Cauda equina | 3 |
| Management | Score |
|---|---|
| Nonoperative | < 4 |
| Consider for operative or nonoperative intervention | = 4 |
| Operative | > 4 |
In 2013, Vaccaro and associates, in cooperation with the AOSpine Spinal Cord Injury and Trauma Knowledge Forum, further expanded the TLICS system and proposed a new thoracolumbar classification system, known as the AOSpine Thoracolumbar Spine Injury Classification System ( Table 126-4 ). This classification system was validated using a diverse international consensus process by independently classifying it twice by group members. The new AOSpine classification system is based on the evaluation of three basic parameters: morphologic classification of the fracture, neurologic status, and clinical modifiers. The morphologic classification is based on three main injury patterns: type A, compression ( Fig. 126-2 ); type B, tension band disruption ( Fig. 126-3 ); and type C, displacement/translation ( Fig. 126-4 ) injuries. Nine subtypes were proposed (five in the A group, three in the B group, and one in the C group). In addition, clinical modifiers address indeterminate injuries and patient-specific comorbidities such as ankylosing spondylitis and diffuse idiopathic skeletal hyperostosis.
| Parameter | Example/Explanation |
|---|---|
| 1. M orphologic C lassification of the F racture | |
| Type A (Compression) | |
| A0: No injury/process fracture | No vertebral fracture or insignificant fractures of the spinous/transverse processes |
| A1: Wedge/impaction | Wedge compression or impaction fracture; involves a single endplate without involvement of the posterior vertebral wall |
| A2: Split/pincer type | Split- or pincer-type fracture; involves both end plates without involvement of the posterior vertebral wall |
| A3: Incomplete burst | Involves single end plate as well as the posterior vertebral wall |
| A4: Complete burst | Involves both end plates as well as the posterior vertebral wall |
| Type B (Tension Band Disruption) | |
| B1: Posterior transosseous disruption | Transosseous disruption of the posterior tension “chance fracture,” which affects a single vertebral level |
| B2: Posterior ligamentous disruption | Ligamentous disruption of the posterior tension band +/− osseous involvement; affects an intervertebral level |
| B3: Anterior ligamentous disruption | Disruption of the anterior longitudinal ligament, causing hyperextension injury; intact posterior tension band |
| Type C (Displacement/Translation) | |
| C | Translation beyond the physiologic range in any plane |
| 2. N eurologic S tatus | |
| N0 | Neurologically intact |
| N1 | Transient neurologic deficit, resolved |
| N2 | Symptoms or signs of radiculopathy |
| N3 | Incomplete spinal cord injury or cauda equina injury |
| N4 | Complete spinal cord injury |
| NX | Cannot be examined (e.g., head injury) |
| 3. C linical M odifiers | |
| M1: Indeterminate injury to the tension band | MRI cannot identify ligamentous disruption |
| M2: Patient-specific comorbidity | Ankylosing spondylitis, diffuse idiopathic skeletal hyperostosis, osteopenia/porosis, and other conditions |
Related posts:
Anatomy of Nerve Root Compression, Nerve Root Tethering, and Spinal Instability
Ventral and Ventrolateral Subaxial Decompression
Diffuse Idiopathic Skeletal Hyperostosis
Spinal Dural and Extradural Vascular Malformations
Spinal Wound Closure
Explant Analysis of Wear, Degradation, and Fatigue in Motion Preserving Spinal Implants
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
