6.4 Lower cervical spine trauma classification systems

1 Introduction

Classification of injuries should provide a common and understandable language for the scientific community to enhance our ability to aid in treatment decision-making, treatment prognosis, and enable meaningful research.

Cervical spine fractures represent a particularly unsatisfactory example of absence for a widely accepted classification system. Challenges towards describing a comprehensive cervical spine classification system a recompounded by the distinct differences, both in anatomical and biomechanical terms, by the upper and lower cervical spine regions. There have been largely anatomically based very detailed classifications for most varieties of upper cervical spine injuries; this approach for lower cervical spine injuries has been far more incomplete. Approaches towards classification of lower cervical spine injuries have been numerous using primarily anatomic, biomechanical, ra-diographic, and combined criteria. Attempts at trying to adapt the more widely accepted AO thoracolumbar fracture classification system to the lower cervical spine have not been well received for a number of factors. Over time technological advances in neuroimaging have also greatly advanced our ability to detect cervical spine injuries and describe osseous, neural, vascular, and ligamentous structures early on in the assessment pathway. To date, attempts at integrating these changing technical advances into a single model, however, has remained incomplete.

From a neurologic injury perspective the ASIA system, which has evolved from Frankel’s original classification, consists of a seven scale grading system and together with the ASIA Motor Score approach provides a comprehensive normative as well as quantitative scale which has stood the test of time. It remains without serious competitors. Unfortunately, this system continues to be limited to use in larger spinal cord injury units rather than be employed in routine descriptions for all spine trauma. (The ASIA system is described and discussed in greater detail in the section on neurologic injuries) [1]. In contrast to the clarity of descriptors used in the ASIA system, attempts at lower cervical spine fracture classification have remained far less well accepted. From a cluster of classification attempts, of which some are represented in the following chapter, several examples deserve further discussion.

2 Current lower cervical spine classifcation systems

Anatomically based systems

Purely anatomically based descriptions of injured structures have limitations in not addressing injury stability. Several authors have used the concepts of ‘columns’ or ‘pillars’ as relevant to determining stability of the cervical spine. With continued disagreement of the actual number of these lower cervical spine columns or pillars consisting of two, three, or four elements, there is little hope for rapid resolution of the differences for purely descriptive anatomic injury models [2].

Morphologic systems

A typical morphologic system was proposed by Bohlman in 1979. He divided the skeletal lesions into five groups: lesions of the articular processes, fractures of the vertebral body, fractures of a pedicle, fracture of a lamina, and fractures of a spinous process [3].

Mechanistic systems

Mechanistic models, while appealing in so many ways, sadly are difficult to apply consistently in light of frequently complex multifactorial force vectors to which the human neck is exposed during trauma. Attempts at classification using purely mechanistic models, such as the very comprehensive Allen-Ferguson system described in 1982, are further hampered by a number of confounding influences, such as head position at time of impact, osteopenia, pre-existent degeneration or deformity, and structural abnormalities such as ankylosis [4]. This classification asks the observer to deduct the major deforming force from inspection of all available recumbent and statically obtained imaging studies and fit the resultant conclusion into one of the six major groups based on major force vector and a total of 25 subgroups based on injury severity. This very well thought-out and complete system has been limited in acceptance due to its inherent complexity and questions of reproducibility. Undoubtedly, the realities of multidi-mensional spine trauma can result in complex injury patterns, which exceed the unidimensional injury directions proposed by the Allen Ferguson system. Harris in 1986 proposed a modification of the Allen and Ferguson classification, which included rotational vectors inflexion and extension on the ex-pen se of the d istractive forces [5]. A rgenson et al in 2000 t hen attempted to reintegrate distractive forces, yet limit the number of categories by using four categories: compression, flex-ion-distraction, extension-distraction, and rotation with a subdivision of each group into three subgroups according to the severity of the anatomical lesion [6]. Despite these refinements, all of these systems suffer from persistent shortcomings of questionable reproducibility and the possibility of recoil from an initial injury position hampering the observer’s ability to appreciate injury severity based on present recumbent skeletal displacement.

Combined systems

The injury models proposed by Louis in 1985 combined anatomic and biomechanical dimensions by studies on dry European and African skeletons in combination with his clinical experiences [7]. The resultant three pillar model was then juxtaposed with concepts of axial and transverse instability patterns. In application, axial stability would be maintained along a vertical column system. The anterior column is formed by the vertebral bodies and discs, and the two posterior columns by posterior joints. Transverse stability is achieved for each motion segment level by a coupling of bony buttresses and ligamentous brakes. The three-joint motion segment is characterized by a triangular disposition of joints with opposing joint spaces, thus supporting the articular orthogonal triangulation concept. This functional anatomic concept also has clinical relevance in the field of spinal growth and clinical instability.

The AO system evolved as an attempt at combining the concepts of simplicity, logical severity progression with impact on treatment decisions and reproducibility as an unprecedented collaboration of international spine specialists with an extensive patient data set. Following its inception in 1986 it underwent further refinements in 1994 with a thoracolumbar version [8, 9]. It is based on a combination of three simple injury forces—“A” as in axial, “B” as in bending, and “C” as in complex/circumferential torsion, which are reasonably easy to understand and differentiate. Further comprehensiveness of this classification is offered by a systematic differentiation into three categories and three subcategories for each major injury category. This system offers a very wide-ranging and reasonably systematic injury description which is well suited for database entries. Despite being conceived with an eye towards improved inter- and intra-examiner reliability this system has been criticized for its complexity, if all subcategories are considered. More recent independent studies indeed have called into question the AO spine classification as showing relatively poor inter- and intra-examiner reliability. Unresolved other issues surrounding inconsistent stability prediction and lack of influence on treatment have led to a continued search for other forms of cervical spine injury assessment even within the AO group.

The search for a graded injury severity scale reflecting various degrees of instability not addressed by traditional systems were the starting points for the efforts of Moore and Anderson as well as the Spine Trauma Study Group starting in 2006 [10]. The initial severity scale was composed of morphologic factors similar to those described by Bohlman, with categories of anterior and lateral column injuries and posterior column injuries combined with a secondary quantitative scale of injury severity. Instability was quantified based on the degree of bony and ligamentous disruption using a visual analog scale from 0 to 5 applied to each column and then expressed as a sum total. Th is system d id not take clinical factors, such as neurologic status or pain, into account.

A further development of the initial classification of the Moore system was then presented in 2007 as SLIC score (Subaxial Cervical Spine Injury Classification) [11]. This scoring system was built upon the axioms of injury morphology and integrity of the discoligamentous complex (anterior and posterior liga-mentous structures and the intervertebral disc), with the presence of neurologic injury as an important additional clinical component. Within each of the three categories, subgroups were identified in grades from least to most severe. The h igher the number of points assigned to one of the components (injury morphology, discoligamentous complex, and neurologic status) the more severe the injury and more likely a surgical procedure would be indicated. The objective of the SLIC score was meant to serve primarily as a checklist for completeness of patient evaluation and expression of injury severity as a research tool, and perhaps also as a decision-making aid as to patient care. In several publications performed by the authoring groups both systems have been shown to have inter- and intra-observer reliability scores beyond the AO and Ferguson Allen systems, yet leave something to be desired in their overall reproducibility.

3 Summary

In principle we have to rely largely on static and recumbent images, such as plain x-rays, CT, or MRI for any attempt at morphological fracture characterization. Fractures involving the vertebral body account for the two problems that raise controversy. The first is related with the individual interpretation of the type of lesion. Although fractures of the vertebral body have been classified according to mechanistic terminology (hyperflexion fracture, hyperextension fracture), terms such as compression fracture are not precise and under individual subjective judgment and interpretation. A tear-drop fracture can be interpreted as compression flexion injury or even facet dislocation by different observers. The second point is the lack of correlation between the fracture morphological description and the quantification of stability.

Any useful classification system should be clinically relevant, easy to use, applicable in the majority of affected cases for the vast majority of situations, teachable, reliable, and valid. Such a system should be readily usable on initial patient contact by offering a degree of intuitive characteristics and also aid in treatment decision-making. Unfortunately, all of the current lower cervical spine systems still fall short of these promises in one way or another. In the absence of an ideal system we therefore persist in using generally accepted injury descriptions and utilize the AO system in large parts of the world for coding purposes, while the evolution of a cervical spine injury severity scale spearheaded by the Spine Trauma Study Group is undergoing further field testing and perhaps refinement.

4 References

1. American Spinal Injury Association (1992) International standards for neurological classification of spinal cord injury. Chicago: American Spinal Injury Association. 2. Mirza SK, Mirza AJ, Chapman JR, et al (2002) Classifications of thoracic and lumbar fractures: rationale and supporting data. J Am Acad Orthop Surg; 10:364–377. 3. Bohlman HH (1979) Acute fractures and dislocations of the cervical spine. An analysis of three hundred hospitalized patients and review of the literature. J Bone Joint Surg Am; 61:1119 –1142. 4. Allen BL Jr, Ferguson RL, Lehmann TR, et al (1982) A mechanistic classification of closed, indirect fractures and dislocations of the lower cervical spine. Spine; 7:1–27. 5. Harris JH Jr, Edeiken-Monroe B, Kopaniky DR (1986) A practical classification of acute cervical spine injuries. Orthop Clin North Am; 17:15–30. 6. Argenson C, de Peretti F, Ghabris A, et al (2000) Traumatic rotatory displacement of the lower cervical spine. Bull Hosp Jt Dis; 59:52–60. 7. Louis R (1985) Spinal stability as defined by the three-column spine concept. Anat Clin; 7:33–42. 8. Magerl F, Aebi M, Gertzbein SD, et al (1994) A comprehensive classification of thoracic and lumbar injuries. Eur Spine J; 3:184–201. 9. Defino HLA (2003) [Classification of fractures of the cervical spine]. Coluna; 2:53–57. Portuguese. 10. Moore TA, Vaccaro AR, Anderson PA (2006) Classification of lower cervical spine injuries. Spine; 31:37–43. 11. Vaccaro AR, Hulbert RJ, Patel AA, et al (2007) The subaxial cervical spine injury classification system: a novel approach to recognize the importance of morphology, neurology, and integrity of the disco-ligamentous complex. Spine; 32:2365–2374.

1 Aebi Classification of Cervical Spine Injuries

Aebi M, Nazarian S (1987) [Classification of injuries of the cervical spine]. Orthopade; 16:27–36. German.

SCALE DESCRIPTION

Injuries classified as to those occurring in the upper and those occurring in the lower cervical spine. Each of these groups are further divided into subgroups.

In the lower cervical spine injuries are classified as to whether they affect the anterior or posterior spine more, or both to the same degree:

Injuries are subdivided with reference to whether they affect primarily bone, bone and ligament equally, or primarily ligament.

Further subdivision by severity and implications for treatment allows more detailed differentiation.

Interpretation:

Descriptive of anatomy of injury. One type not necessarily more severe than the next.

SCALE ILLUSTRATION

**Fig 6.4-1a–n** Atlas fractures: a Posterior arch fracture. b Anterior arch fracture. c Posterior + anterior arch fracture (lateral separation of the lateral masses without rupture of the transverse ligament) (lateral dislocation < 7 mm). d Unilateral comminuted fracture of the lateral mass. e Bilateral comminuted fracture of the lateral mass. Axis fractures: f No or minor dislocation (anterolisthesis of C 2 < 3.5 mm, angulation C2/C3 < 11°) (Effendi I). g Moderate dislocation [angulation C2/C3 > 11°, listhesis > 3.5mm, < 50% of vertebral body] (Effendi II). h Severe dislocation inflexion ( > 50% of vertebral body) C2/C3 joint dislocation (Effendi III). i Fracture of the tip of the dens—avulsion j Fracture of the base resp. the body of the dens k Fracture of neck of the dens l Fracture of the isthmus+base and body of the dens. m Fracture of the isthmus+neck of the dens. n Fracture of the dens with impaction+fragments in the body of C2.

**Fig 6.4-2a–q** Combination of atlas and axis fractures: a Posterior arch fracture of the atlas + isthmus fracture of C2 b Posterior arch fracture of the atlas + base/body fracture of C2 c Posterior arch fracture of the atlas + neck fracture of C2 C1.1 injury: d Base fracture of the dens + luxation of C1/C2(greater than width of dens—11mm). e Neck fracture of the dens + luxation of C1/C2(greater than width of dens—11mm). f Atlas fracture with separation of the lateral masses > 7 mm laterally (Jefferson fracture). g Severe sprain: anterior luxation of C1/C2. h Rotation-subluxation (4 types according to Fieldig). i Complete posterior subluxation of C1/C2. Type A subaxial fractures: j Symmetrical compression k Superior wedge fracture without visible ligament lesion. l Wedge fracture without visible ligament lesion (angulation < 11°). m Vertebral body fracture, multiple fragments, one endplate affected (1 disc injured). n Vertebral body fracture, multiple fragments (2 discs injured). o Comminuted fracture, posterior wall dislocated < 3 mm, posterior elements not visibly injured. p Disrupture of the anterior longitudinal ligament and the disc (caused by hyperextension). q Traumatic disc herniation.

**Fig 6.4-3a–h** Type B subaxial fractures: a Isolated fracture of the posterior elements. b Fractures of the articular processes without dislocation (compression or longitudinal fracture). c Combination fracture of the articular processes and fracture of the posterior elements without dislocation. d Fracture of the posterior elements with subluxation. e Shearing facet fracture and subluxation of the adjacent facets. f Fracture separation of the lateral mass (fracture through pedicle and arch). g Rupture of the posterior ligament complex with subluxation of the articular processes (bilateral). h Rupture of the posterior ligament complex with asymmetrical subluxation of the articular processes (unilateral).

**Fig 6.4-4a–p** Type C subaxial fractures: i Burst fracture of the vertebral body combined with burst fracture of the posterior elements (arch, spinous process). j Horizontal fracture through the vertebral body with burst fracture of the posterior elements (arch, spinous process). k Complete luxation fracture wih fracture of the posterior elements. l Wedge fracture of the vertebral body (< 11°) and disruption of the posterior ligament complex. m Fracture of the vertebral body (split in the anterior superior part + dislocation of the posterior fragment > 3 mm into the spinal canal) [genuine “Tear drop fracture] n Pure luxation with unilateral facet interlocking (disrupture of the disc and the posterior ligament complex) o Pure luxation with bilateral facet interlocking (disrupture of the disc and the posterior ligament complex) p Dorsal luxation with disrupture of the disc and the posterior ligament complex.

METHODOLOGY

No predictive validity or reliability studies were identified.

Predictive validity


Population tested in	Outcome	Predictive validity
Not tested

Reliability


Population tested in	Interobserver reliability	Intraobserver reliability
Not tested

CONTENT

RATING

2 Allen and Ferguson Classification of Lower Cervical Spine Fractures

Allen BL Jr, Ferguson RL, Lehmann TR, et al (1982) A mechanistic classification of closed, indirect fractures and dislocations of the lower cervical spine. Spine; 7:1–27.