The yearly incidence of traumatic cervical spine fracture dislocations is over 64 per 100,000 population, one-tenth of which, or nearly 15,000 people per year, suffer concomitant spinal cord injuries (SCIs).1 Subaxial cervical spine motion segments are composed of anterior and posterior elements joined together by discoligamentous complex and facet joints, enabling them limited smooth, effortless, and painless motion without the possibility of nerve root injury or SCI.2 Translation of traumatic kinetic energy into disruption of the anatomical integrity of cervical spine has many morphologies. Vector forces of various magnitudes and from myriad angles can fracture the vertebral body or the posterior arch in conjunction with partial or complete disruption of the viscoelastic discoligamentous complexes, thus exposing the spinal cord to trauma. The morphology of the cervical spine on imaging studies includes numerous features depending on injury severity and the direction of vector forces during a fraction of a second. Clinical investigators since 1949 including Nicoll3 and later Holdsworth4,5 have tried to classify injury severity by the extent of damage to the discoligamentous complex and the degree of crushed vertebral bodies or the posterior arch. In 1962, Allen et al tried to classify injury morphology and, therefore, injury severity by conceptualizing a mechanistic basis of fractures and dislocations.6 In 1986, JH Harris took advantage of multiplanar computed tomography (CT) in addition to plain x-rays and X, Y, and Z coordinates to introduce a radiologist's view of traumatic cervical spine injuries.7
Morphology classifications should be end-user friendly, have high-intra and interrater reliability, be easy to understand, and serve as a tool for the clinical researcher. These classifications must indicate injury severity, guide surgical planning, and predict outcome. The Allen et al and Harris et al classifications have low interrater reliability and are both complex and difficult to remember.8,9 In 2015, Vaccaro et al introduced a simple, conceptual, and easy to recall subaxial cervical spine injury classification with high-intra and interrater reliability scores.8–10 This classification is not only inclusive of almost all injuries culminating in SCI but also effectively covers less severe facet fractures (nondisplaced and displaced and floating lateral mass), includes modifiers such as ankylosing spondylitis, and clearly describes the degree of neurologic injury.9 We hypothesized that the different classes of this morphologic classification may embody different degrees of admission imaging [intramedullary lesion length (IMLL) on magnetic resonance imaging (MRI)] and clinical [American Spinal Injury Association (ASIA) motor score (AMS)] injury severity, may be used as a guide for surgical planning (anterior, posterior, or circumferential decompression and internal fixation), and may be predictors of neurologic outcome [AMS and ASIA impairment scale (AIS) grade conversion].
MATERIAL AND METHODS
Design, Objective, and Rationale
We performed a retrospective analysis of prospectively collected clinical information to examine the clinical applicability of AOSpine subaxial cervical spine injury classification as a measure of injury severity and as a predictor of neurologic outcome.
Morphology and its relation to neurologic status of subaxial cervical spine traumatic injuries in patients with AIS grades A–C.
Consecutive series of 92 patients with traumatic cervical spine and SCI who were admitted to this level 1 trauma center over a 9-year period. Patients were included if they (1) had blunt cervical spine and SCI; (2) were AIS grades A–C; (3) were surgically treated for their SCI; (4) had complete pre and postoperative CT and MRI studies; (5) had evidence of SCI on postoperative MRI; and (6) were followed for at least 6 months. Patients were excluded if they (1) had penetrating cervical spine injuries; (2) had incomplete studies; (3) were treated nonoperatively; (4) died within 6 months; or (5) did not have a complete 6 months of follow-up. This study was approved by the institutional review board.
We studied patients' management at the scene of accident; demographics; mechanism of injury; injury severity as reflected in injury severity score (ISS), AMS, AIS, and IMLL on postoperative MRI; AOSpine subaxial cervical spine injury class; traction; surgical planning and technique; and follow-up AMS, AIS, and AIS grade conversion during a 6-month follow-up.
Emergency Management at the Scene of Accident
Scene of accident resuscitation and rescue was managed by the emergency medical technicians of the Maryland Institute for Emergency Medical Services Systems (MIEMSS). When needed, patients were intubated, ventilated, back boarded, and transferred with forehead and chin straps to the Trauma Resuscitation Unit (TRU).11,12 The time from scene to admission in 54 direct admissions was 1.0 hours. The time from accident to transfer from another hospital in 9 patients was 12.7 hours. Time past injury to admission to the trauma center was unknown in 29 patients.
At the TRU, the patients had a quick primary and an extensive secondary survey by 1 of 3 teams of trauma surgeons and then neurosurgical consultation was followed by a complete ASIA motor, sensory, and impairment scale determination.13 When completely stable and resuscitated, the patients were ready for imaging studies.
Preoperative Diagnostic Imaging
Preoperative CT and MRI of the cervical spine were performed a mean of 2.9 and 7.5 hours after trauma, respectively. In addition, each patient had a postoperative MRI 46 hours after injury.
Morphology classification was modeled on the AOSpine subaxial cervical spine injury classification system introduced in 2015 by Vaccaro and colleagues (Figs. 1–6).14 This classification has 3 broad categories: class A, class B, and class C. Fractures primarily involving the vertebral body are designated class A: A0 indicates no injury to the vertebral body; A1 indicates depression of the upper or lower end plates; A2 indicates pincer type fractures; A3 indicates fractures involving 1 end plate with protrusion of bone fragments into the vertebral canal; and A4 indicates fracture of both end plates with protrusion of bone fragments into the spinal canal (Fig. 1). Class B injuries are tension band injuries in either hyperflexion (B1/B2) or hyperextension (B3) along the Y coordinate with no translation in any plane (Figs. 2–4). If hyperflexion results in a fracture line through the posterior element and the vertebral body, it is called B1 (Fig. 2). If hyperflexion disrupts the posterior osseoligamentous complex, it is called B2 (Fig. 3). And if hyperextension ruptures anterior longitudinal ligament, annulus fibrosus, and disc space with intact posterior elements, it is called B3 (Fig. 4). Fracture dislocations associated with significant degrees of translation in any plane (X–Y–Z) are designated class C (Fig. 6). In concept, class A1, A2, B1, B2, and B3 injuries are less severe than class A4 and C fractures. The A1, A2, B1, B2, and B3 injuries do not necessarily transgress the spinal canal. As suggested by the classification system, B-type injuries can be associated with A3 or A4 fractures, in which case there is violation of the spinal canal. We encountered a hybrid combination of B2 and A4 in which the morphology of the body was A4 with hyperflexion of the cervical spine exactly like B2 on CT and MRI (Fig. 5). These injuries also turned out to be relatively severe.
MRI and IMLL
IMLL was the rostrocaudal length of high-signal intensity from the injury epicenter, measured in millimeters. IMLL was measured on T2W images or short T1 inversion recovery sequence. IMLL was measured in postoperative MRI images taken approximately 46 hours after trauma.12,15
Closed reduction was performed after imaging studies, and in a minority of patients open reduction was performed at the time of surgical intervention.
All patients had surgical intervention for decompression and internal fixation. The operating neurosurgeon decided the specific surgical procedure to be used.16
The patients were kept in the intensive care unit and subsequently in the intermediate care unit until they were off the ventilator and ready to be discharged to rehab.
The patients were re-examined for determination of their AMS and AIS grade at 6 weeks, 3 months, and 6 months or longer in the clinics of the Shock Trauma Center.
To assess the association between class of injury and any categorical variable, we used an appropriate χ2 test. For any continuous variable we first performed an analysis of variance (ANOVA). Then, if the variance ratio for the ANOVA was statistically significant, we used a Tukey-honest significant difference test to find out which classes of injury were significantly different from each other. The significance level for all statistical tests was set at 0.05.
Table 1 shows the patients' baseline demographic and clinical characteristics.
Demographics and Injury Mechanisms
Comparing the mean age distribution between the 5 AOSpine subaxial cervical spine injury classes, we found that patients in classes A4, B2A4, and C were significantly younger than patients in classes B2 and B3 (P < 0.0001). Classes A4, B2A4, and C are thought to reflect more severe injuries than classes B2 and B3 which are defined as with intact anterior element or posterior element and no evidence of translation (Table 1).
Injury mechanism was distinctly different in the 5 classes of the AOSpine subaxial cervical spine injury classification system. Twenty-seven (64%) of 44 patients with class C injuries sustained their injury in a motor vehicle collision. This is in sharp contrast to the patients with class B3 injuries, 73.7% of which resulted after a fall. One half of B2/A4 injuries were due to sports injuries, primarily shallow dives.
Four parameters were used to assess the degree of injury severity: (1) ISS indicated that class C injuries had the highest score (30.4) and class B3 injuries had the lowest score (13.8); (2) admission AMS did not differ statistically among all categories; however, 72.8% of patients in class C had complete injuries (AIS grade A) and 89.5% of patients in class B3 had incomplete injuries (AIS grades B and C); and (3) comparing the IMLL in all classes of injury, patients with class C, B2A4, and A4 injuries had statistically longer IMLL than patients with class B2 and B3 injuries (P < 0.007).
Morphology of Fracture Dislocations
Among the 92 patients, 8 were in class A4, 21 were in class B2 with 16 of them having associated A4 fractures (B2A4), 19 were in class B3, and 44 were in class C.
A class C injury was the strongest indicator for skeletal traction. None of the patients with class A4 or B3 injuries underwent skeletal traction. Among patients who needed traction, 20% had class B2 injuries, 25% had class B2/A4, and 72.7% had class C. The class C injuries were primarily unilateral or bilateral locked facets or tear-drop fractures (Allen's flexion distraction phylogeny stages 2–5 and compressive flexion phylogeny stages 3–5)6 with complex fractures and translation in need of skeletal traction and anatomical realignment.
Surgical intervention in this group of patients was aimed at adequate spinal cord decompression and anatomical alignment and internal fixation. Sixty-one of 92 (66.3%) patients had either anterior cervical discectomy and fusion (ACDF) or anterior cervical corpectomy and fusion (ACCF), with 32 of these 61 patients also having posterior spinal fusion. Twenty-four patients had ACDF and ACCF and laminectomy and posterior spinal fusion (PSF). Six patients had laminectomy, of whom 5 also had PSF. Only 1 patient primarily had PSF. Table 1 shows the distribution of different surgical procedures. Circumferential surgery was performed in 70.4% of patients with complex class C injuries and 26.3% of patients with class B3 injuries.
AIS Grade Conversion
Forty-five of 92 (49%) patients had an improvement of at least 1 grade over their original AIS grade at the time of admission. AIS grade conversion was seen in 73.7% of patients in class B3 and 38.6% of patients in class C. Grade conversion was conspicuously more visible in patients whose injuries were the simple classes of B2 and B3 versus the complex classes of B2A4 and C. AIS grade conversion also reflects the severity of SCI as indicated by IMLL and admission AIS grade (Table 1).
In this study, injury severity and prognosis among the different classes of the AOSpine subaxial injury severity classification system had an ordinal ranking, the clarity of which is important to rescue and resuscitation, diagnostic work up, in-service academic communication, nonoperative or operative management, surgical technique, clinical research, and therapeutic trials.
Evidence from extensive clinical, radiographic, and postmortem investigations performed by Nicoll3 and Holdsworth4,5,17 indicated that posterior osseoligamentous complexes played a major role in segmental stability. These studies laid the foundation for future morphologic classifications of cervical and thoracolumbar fractures and fracture–dislocations.6,7,18,19 Many of the attempted injury morphology classifications, including the mechanistic classification of Allen et al,6 the 3 column classification of Denis,18 the practical classification of Harris et al,7 and the comprehensive classification of Magerl et al,19 are all complex, extensive, cumbersome, and hard to remember and communicate to other spine surgeons. In addition, low validity and reliability scores render these classifications unfit for prospective multicenter comparative investigations.9,10,20
The AOSpine subaxial cervical spine injury classification system is simple, ordinal in rank, easy to remember, and it facilitates communication with other spine surgeons and has a high degree of validity and reliability. Importantly, it not only classifies large injury but also describes minor facet injuries (Fig. 7) or facet injuries that might be within other subclasses. In addition, the classification includes the degree of neurologic injury (N0–N4) and modifiers such as sequestrated disc, injury to the discoligamentous complex, or ankylosing spondylitis (Fig. 2).9,10,14
Evidence from numerous preclinical studies of animal models of SCI suggests a relationship between injury severity and IMLL and functional outcome.21,22 We used 4 parameters representing injury severity in our study: (1) admission ISS; (2) admission AMS (Fig. 8); (3) admission AIS grade (Fig. 9); and (4) admission IMLL (Fig. 10). ISS and AMS at admission did not differ significantly in all 5 classes of the present classification (Table 1). Low numbers in the categorical variables in AIS grades precluded us from definitely ascertaining whether AIS grade was significant in the different classes of the AOSpine classification system. However, the trend indicated that classes B2A4 and C had the highest number of patients with AIS grade C (Fig. 9). Since 1989 when Schaefer et al23 reported on the relationship between injury severity and IMLL, numerous reports have corroborated those investigators' findings.24–29 In this study, IMLL was longer in classes A4, B2A4, and C than classes B2 and B3 (Fig. 10, P < 0.007).
We examined how the selection of surgical technique differed among the different classes of injury in the present system. We found a greater preponderance of circumferential fusion in patients with class C and B2A4 injuries. Our results agree with the findings reported in Dvorak et al's16 systematic review of different surgical techniques used in patients whose injury severity was based on the subaxial cervical spine ISS classification (Fig. 11).9
Allen et al's6 mechanistic classification is, to some extent, predictive of outcome. Those investigators gave an ordinal rank to the stages of their 5 phylogenies. For example, flexion distraction stage 1 (facet subluxation) had a relatively better prognosis than flexion distraction stage 5 (spondyloptosis).30 AOSpine cervical classification not only has an ordinal rank of injury severity but also is predictive of AIS grade conversion. AIS grade conversion was much more impressive in B2 and B3 injuries than B2A4 and C injuries (Fig. 12). Almost 60%–74% of AIS grade B2 and B3 injuries converted to a better AIS grade over at least 6 months of follow-up as compared with 37%–39% in B2A4 and C injuries.
Cervical spine fracture dislocation classifications should be simple, easy to remember, have intrarater and interrater reliably among surgeons, indicate injury severity, guide surgery, and predict outcome. Compared with other morphologic classifications, the AOSpine subaxial cervical spine injury classification meets all of those objectives and might be effectively applied as a research tool in future prospective comparative, observational, or randomized trials. Specifically, the present classification clearly reflects injury severity and predicts AIS grade conversion at the time of admission. Both of which seem to be an expression of IMLL.
1. Torretti JA, Sengupta DK. Cervical spine trauma. Indian J Orthop. 2007;41:255–267.
2. Aarabi B, Walters BC, Dhall SS, et al. Subaxial cervical spine injury classification systems. Neurosurgery. 2013;72(suppl 2):170–186.
3. Nicoll EA. Fractures of the dorso-lumbar spine. J Bone Joint Surg Br. 1949;52:376–394.
4. Holdsworth FW. Fractures, common dislocations, fractures-dislocations of the spine. J Bone Joint Surg. 1963;45:6–26.
5. Holdsworth FW. Fractures, dislocations and fracture-dislocations of the spine. J Bone Joint Surg. 1970;52A:1534–1551.
6. Allen BL Jr, Ferguson RL, Lehmann TR, et al. A mechanistic classification of closed, indirect fractures and dislocations of the lower cervical spine. Spine (Phila Pa 1976). 1982;7:1–27.
7. Harris JH Jr, Edeiken-Monroe B, Kopaniky DR. A practical classification of acute cervical spine injuries. Orthop Clin North Am. 1986;1:15–30.
8. Patel AA, Hurlbert RJ, Bono CM, et al. Classification and surgical decision making in acute subaxial cervical spine trauma. Spine (Phila Pa 1976). 2010;25(21 suppl):S228–S234.
9. Vaccaro AR, Hulbert RJ, Patel AA, et al. 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 (Phila Pa 1976). 2007;32:2365–2374.
10. Schroeder GD, Kepler CK, Koerner JD, et al. A worldwide analysis of the reliability and perceived importance of an injury to the posterior ligamentous complex in AO type a fractures. Glob Spine J. 2015;5:378–382.
11. MIEMSS. Maryland Institute for Emergency Medical Services Systems; 2015. Available at: https://www.miemss.org
12. Aarabi B, Akhtar-Danesh N, Diaz C, et al. Intramedullary lesion length on postoperative magnetic resonance imaging is a strong predictor of AIS grade conversion following decompressive surgery in cervical spinal cord injury. Neurosurgery. 2016;80:610–620.
13. Ditunno JF Jr. New spinal cord injury standards, 1992. Paraplegia. 1992;30:90–91.
14. Vaccaro AR, Koerner JD, Radcliff KE, et al. AOSpine subaxial cervical spine injury classification system. Eur Spine J. 2016;25:2173–2184.
15. Aarabi B, Sansur CA, Ibrahimi DM, et al. Intramedullary lesion length on postoperative magnetic resonance imaging is a strong predictor of ASIA impairment scale grade conversion following decompressive surgery in cervical spinal cord injury. Neurosurgery. 2017;80:610–620.
16. Dvorak MF, Fisher CG, Fehlings MG, et al. The surgical approach to subaxial cervical spine injuries: an evidence-based algorithm based on the SLIC classification system. Spine (Phila Pa 1976). 2007;32:2620–2629.
17. Holdsworth FW. Neurological diagnosis and the indications for treatment of paraplegia and tetraplegia associated with fractures of the spine. Manit Med Rev. 1968;48:16–18.
18. Denis F. The three column spine and its significance in the classification of acute thoracolumbar spinal injuries. Spine (Phila Pa 1976). 1983;8:817–831.
19. Magerl F, Aebi M, Gertzbein SD, et al. A comprehensive classification of thoracic and lumbar injuries. Eur Spine J. 1994;3:184–201.
20. Patel AA, Dailey A, Brodke DS, et al. Subaxial cervical spine trauma classification: the subaxial injury classification system and case examples. Neurosurg Focus. 2008;25:E8.
21. Hackney DB, Asato R, Joseph PM, et al. Hemorrhage and edema in acute spinal cord compression: demonstration by MR imaging. Radiology. 1986;161:387–390.
22. Fujii H, Yone K, Sakou T. Magnetic resonance imaging study of experimental acute spinal cord injury. Spine. 1993;18:2030–2034.
23. Schaefer DM, Flanders A, Northrup BE, et al. Magnetic resonance imaging of acute cervical spine trauma. Correlation with severity of neurologic injury. Spine (Phila Pa 1976). 1989;14:1090–1095.
24. Miyanji F, Furlan JC, Aarabi B, et al. Acute cervical traumatic spinal cord injury: MR imaging findings correlated with neurologic outcome–prospective study with 100 consecutive patients. Radiology. 2007;243:820–827.
25. Boldin C, Raith J, Fankhauser F, et al. Predicting neurologic recovery in cervical spinal cord injury with postoperative MR imaging. Spine (Phila Pa 1976). 2006;31:554–559.
26. Aarabi B, Simard JM, Kufera JA, et al. Intramedullary lesion expansion on magnetic resonance imaging in patients with motor complete cervical spinal cord injury. J Neurosurg Spine. 2012;17:243–250.
27. Andreoli C, Colaiacomo MC, Rojas Beccaglia M, et al. MRI
in the acute phase of spinal cord traumatic lesions: relationship between MRI
findings and neurological outcome. Radiol Med. 2005;110:636–645.
28. Hayashi K, Yone K, Ito H, et al. MRI
findings in patients with a cervical spinal cord injury who do not show radiographic evidence of a fracture or dislocation. Paraplegia. 1995;33:212–215.
29. Talbott JF, Whetstone WD, Readdy WJ, et al. The brain and spinal injury center score: a novel, simple, and reproducible method for assessing the severity of acute cervical spinal cord injury with axial T2-weighted MRI
findings. J Neurosurg Spine. 2015;23:1–10.
30. Wilson JR, Vaccaro A, Harrop JS, et al. The impact of facet dislocation on clinical outcomes after cervical spinal cord injury: results of a multicenter North American prospective cohort study. Spine (Phila Pa 1976). 2013;38:97–103.