Imaging and classifications of atlantoaxial dislocation: a narrative review

Abstract

• Radiography is of importance in the diagnosis of atlantoaxial dislocation (AAD), and it is the basic imaging technique. However, it should not be the sole diagnostic modality, especially in complex or unclear cases.

• Conventional X-ray includes an open-mouth odontoid view and a cross-table lateral view, and careful study of radiological findings is crucial to give an early diagnosis of AAD. Lateral flexion-extension dynamic views are only used as an additional supplement in some special cases.

• Although X-ray images are enough to diagnose AAD in most cases, some patients suspected with AAD should be evaluated with the readily available and quick CT scan.

• If patients with AAD have symptoms of spinal cord and medullary compression, apart from conventional radiographs, a combination of high-quality CT and MRI of cervical spine are necessary for the diagnosis and choice of treatment.

• For patients with AAD, both the thin slice CT scanning with coronal, sagittal and three-dimensional reconstruction images and MRI of cervical spine are fundamental to surgical planning.

• Clinical classifications of AAD associated with imaging are useful in determining treatment strategies. The present study reviews publications on imaging and clinical classification of AAD to aid the clinician in the evaluation and management of these dislocations.

Introduction

The first cervical vertebra (C1, atlas) and second cervical vertebra (C2, axis) are regarded as two atypical vertebrae, and the anatomy of the atlantoaxial region and mechanical stabilization allows complex movements of the atlantoaxial joints (1, 2). Atlantoaxial dislocation (AAD) refers to a loss of normal articulation and malalignment of C1 and C2, making imaging investigation one of the basic evaluation of patients in whom AAD is suspected (1). The pathologies of AAD can generally be divided into traumatic, congenital, inflammatory, degenerative and tumor factors (3). The tear damage to the transverse atlas ligament by trauma can eventually lead the loss of normal function of the transverse atlas ligament, manifesting as AAD (4). The congenital factors included Down syndrome, Goldenhar syndrome and Morquio syndrome, and those lead to developmental abnormalities of the ligaments and bone structures (5). What is more, inflammatory infiltration of other local tissues including rheumatoid arthritis, recurrent pharyngitis in children and tuberculosis lesions involving the atlantoaxial vertebrae can lead to AAD (6). Looseness and instability of the ligaments around the atlantoaxial joint and normal bone and ligament structure broken by the tumor all lead to AAD (7). Considering the etiology for patients with AAD is important when selecting imaging approaches (8). Certainly, evaluation of imaging includes not only the direction of dislocation and etiology but also the associated skeletal anomalies, such as os odontoideum (OS), C2-3 fusion, Arnold-Chiari malformation and so on (8).

The value of the classifications of AAD is useful in determining treatment strategies for different conditions based on the clinical and radiographic parameters, since varying opinions on indications of different treatment methods are present. In recent years, clinical classifications of AAD, which can help spine surgeons with choosing treatment strategies, are also developed (1, 9, 10, 11, 12, 13). The aim of this study is to provide the reader with a review of the image approach and clinical classification of AAD, and aid the clinician in the evaluation and management of these dislocations.

Image approach

Conventional X-ray or radiographs, computed tomography (CT) and magnetic resonance (MR) imaging of the cervical spine are the common image approaches to diagnose AAD with various etiologies.

Radiographs

The first cervical vertebra (atlas) and the second cervical vertebra (axis) formed bilateral atlantoaxial joints by articular surfaces of C1–C2 lateral masses, and a central atlanto-odontoid joint by an articular surface in the dorsal aspect of the anterior arch of the atlas and odontoid process of the axis. Open mouth and lateral radiographs of cervical spine are the basic imaging techniques required for observing these joints (Figs 1 and 2), making radiography an important imaging technique in the diagnosis of AAD. Open mouth view can show the relationship between the anatomic axis of C1 and odontoid process of C2, the shape of the lateral mass of the atlas and the symmetry of bilateral gap between the odontoid process of C2 and lateral mass of C1. Cervical lateral view can show the alignment of C1–C2, the continuity of the atlantoaxial joint and the atlantodental interval (ADI).

Figure 1.

Open mouth radiograph showed the relationship between the anatomic axis of C1 and odontoid process of C2, the shape of the lateral mass of the atlas and the symmetry of bilateral gap between the odontoid process and lateral mass.

Figure 2.

Cervical lateral radiograph showed the alignment of C1–C2, the continuity of the atlantoaxial joint and the atlantodental interval (ADI).

Traditionally, neutral lateral and open mouth radiographs of the cervical spine have been used as the basic means for diagnosing AAD (Figs 1 and 2) (13). In the past three decades, dynamic cervical radiographs are often used as additional measurements to diagnose atlantoaxial instability and AAD; however, the increase of the diagnostic sensitivity is controversial (1). Some investigators reported that additional dynamic cervical radiographs could increase the diagnostic sensitivity, while some others did not think additional dynamic cervical radiographs could provide more information, since muscle spasm of cervical spine was common in symptomatic patients (13, 14, 15, 16). It is a common view that the use of combination of conventional radiographs and MRI or CT for diagnosis could decrease the false-negative diagnostic rate of AAD (2, 11, 13, 14, 15, 16).

When evaluating trauma patients with concerns about occipitocervical region involvement, it is vital to give early diagnosis of atlanto-occipital and atlantoaxial injuries and avoid missing the potential injury (14). For the use of plain radiography, the National Emergency X-Radiography Utilization Study (NEXUS) group recommended the combination of an open-mouth odontoid view, an anterior-posterior (AP) view and a cross-table lateral view were as a standard three-view imaging of the cervical spine for cervical spine injuries (1, 17).

For the use of plain radiography in inflammatory diseases causing AAD, an X-ray evaluation in AP, lateral and open-mouth odontoid views are also recommended, and lateral flexion-extension dynamic projections are also recommended to assess atlanto-occipital and atlantoaxial region involvement, and to give diagnosis of AAD or atlantoaxial instability (18, 19, 20). AAD in RA patients are commonly accompanied with odontoid erosions, instability of craniovertebral junction and subaxial subluxation, which could be seen on such X-ray views (21, 22). The use of plain radiography can provide clear identification of bone alignment, fractures and deformities of cervical spine. However, it has some obvious limitations for diagnosis of AAD; for instance, it is difficult to identify early bony erosions and soft tissue ligaments injuries or neurological structure changes, such as pannus, incomplete ligaments injuries and spinal cord compression (18).

In the presence of radiographic alterations in atlantoaxial region, if symptoms of atlantoaxial region instability with or without neurological deficits are also showed, including neck pain with or without limitation of neck movement, weakness and/or numbness of the upper and lower extremities and neurological symptoms, CT and/or MRI of the cervical spine are strongly recommended for diagnosing or ruling out AAD (8, 14, 23).

Plain radiographs

Yin et al. (13) and other investigators systematically reviewed literature and proposed some principles applying to the interpretation of open mouth and lateral views (1, 2, 18). The principles for the interpretation of open mouth views are as follows (1, 13, 18):

First, the correlation between the anatomic axis of C1 and C2 should be evaluated primarily to determine whether there is an overlap or deviation between them, and an overlap or deviation of the anatomic axis of atlas and axis with more than 3 mm is usually considered as an abnormal situation. Specially, a bigger departure hits at transverse displacement or rotational dislocation or subluxation.

Second, a discontinuous line from the lateral rim of the lateral masses of atlas to axis usually means transverse displacement between C1 and C2. An asymmetry of the lateral mass of atlas might hint at rotational dislocation of C1–C2, but it must be combined with medical records and clinical manifestations, such as neck pain, neck movement restriction, cervical muscle spasm, tenderness and torticollis, before the diagnosis is given.

Third, asymmetry of the lateral gap between mass of C1 and odontoid process of C2 sometimes may indicate a rotational dislocation or transverse displacement, and, likewise, it must be also combined with clinical manifestations, including neck pain, restriction in cervical rotation, torticollis and so on.

Finally, separation of the lateral masses of C1 indicates atlas fracture, and the overhang of atlas by more than 6.9 mm denotes transverse atlantal fracture associated with rupture of the transverse ligament and instability is diagnosed, according to the ‘rule of Spence’ (24, 25).

The principles for the interpretation of lateral views are summarized as follows (13):

First, the anterior line from the posterior arch of C1 to the posterior wall of spinal canal of C2 level normally forms a natural curve with a little lordosis, and when AAD is developed, the natural curve is usually lost.

Second, the atlantodental interval (ADI) and the posterior atlantodental interval (PADI) should be measured (Fig. 3). ADI is measured between the posterior aspect of the anterior arch of the atlas and the anterior aspect of the odontoid process or dens of the axis, and its normal value is less than 3 mm in adults and less than 5 mm in children, varying by <1 mm with neck flexion or extension. ADI increased by greater than 3 mm in adults and greater than 5 mm in children suggests anterior dislocation, and ADI greater than 5 mm in adults usually hints at AAD with the transverse ligament loosening or rupture or odontoid malformation.

Figure 3.

Cervical lateral radiograph showed the measurement of ADI and PADI. ADI, anterior atlantodental interval, measured between the posterior aspect of the anterior arch of C1 and the anterior aspect of the odontoid process; PADI, posterior atlantodental interval, measured between the posterior aspect of the odontoid process and the anterior aspect of the posterior arch of C2.

Third, the distance from the middle point of the posterior arch of the atlas to the spinal process of the axis elongates significantly when AAD (anterior dislocation) develops.

Finally, the inclination angle is the angle between the posterior rim of C2 and the odontoid process, and the range of such angle is from 8 to 25°, with the average value of 11.7°. An anteversion of the inclination angle usually indicates odontoid fracture (13).

In addition, PADI is measured between the posterior aspect of the odontoid process and the anterior aspect of the posterior arch of the atlas. The value of PADI is that it represents the space available for the spinal cord, while a significant decrease of PADI may reveal the compression of the upper spinal cord and medullary in patients with AAD.

Research showed that the cervical spinal cord in the atlantoaxial region usually occupies 10 mm of the canal diameter, and approximately 1 mm for the dura and at least 1 mm for the cerebrospinal fluid (CSF) anterior and posterior to the cord are needed, respectively (18). Therefore, as the space available for the spinal cord, a total of at least 14 mm of PADI is needed to prevent the compression of spinal cord and neurological impairment (18, 19). PADI was also reported as a useful predictor of paralysis and recovery (21, 26). After surgical treatment of AAD, patients with 14 mm or more of PADI were found to have a higher rate of neurological recovery than those with less than 14 mm of PADI (21).

Dynamic radiographs

Adding to an open-mouth odontoid view and cross-table lateral view, lateral flexion-extension dynamic views are regarded as useful additional radiographs that can increase both the sensitivity and specificity of the diagnosis of AAD (1). If it is possible and safe for patients, the authors suggest that lateral flexion-extension dynamic views in our institutions, which usually needs physician supervision during the process of movement, and active flexion-extension dynamic views are preferred over passive flexion-extension views (Fig. 4).

Figure 4.

Lateral flexion-extension dynamic views showed AAD.

It should be emphasized that passive flexion-extension radiographs are questionable and might be dangerous in some patients. For instance, patients with AAD, in whom compression of the brainstem or spinal cord or myelopathy is severe (severe numbness and weakness in the upper and/or lower extremities, and the numbness and weakness may culminate in paralysis), are not allowed to have passive flexion-extension views. In addition, active flexion-extension dynamic views can also have false negatives because neck pain with muscle spasms are common symptoms for patients with AAD, and these symptoms make the patients have a limited range of flexion-extension movement.

For traumatic AAD, it is controversial with the use of lateral flexion-extension dynamic views for most patients. In patients with trauma, plain radiographs and CT are reported to be accurate for detecting these injuries in most patients (27). In their retrospective review of 14,577 patients with blunt trauma victims, Chiu et al. reported that 614 (4.2%) patients had C-spine injury, 2,605 (18%) patients were not evaluable and 14 patients had isolated ligamentous injury detected on plain static radiographs or CT (27).

To sum up, careful physical examination and imaging with basic X-ray (an open-mouth odontoid view, cross-table lateral view and lateral flexion-extension dynamic views) usually may be enough to give the diagnosis of AAD in most non-trauma patients. However, imaging with basic X-ray is not to effectively determine the management of AAD, and additional high-quality CT and MR imaging of cervical spine are necessary to give the diagnosis of the exact type of AAD (reducible or irreducible) and the choice of treatment.

Computed tomography (CT) scan

High-quality CT scans of cervical spine with multiplanar reconstruction are strongly recommended for patients with AAD, which usually include continuous thin slice CT scans with coronal, sagittal and three-dimensional reconstruction images (Fig. 5).

Figure 5.

A 79-year-old female patient with AAD. Top row left: coronal CT image; top row right: midsagittal CT scan; low row left: parasagittal CT image; low row right: three-dimensional reconstruction image.

The utility of high-quality CT scans of cervical spine for patients with AAD are listed as follows:

First, compared with the conventional X-ray imaging, a high-quality CT scan of cervical spine with multiplanar reconstruction is a more reliable method for the diagnosis of AAD due to probable poor visualization on conventional X-rays in some patients. Therefore, grossly dislocation of C1–C2 could be usually recognized by conventional radiographs. Nowadays, with the widespread availability of CT scanners in most hospitals, CT scans are preferable in the diagnosis and management of AAD. Thin slice CT scans with multiplanar reconstruction certainly help in defining the pattern of displacement, morphology of articular facet and even possible mechanism of AAD, which can help to determine the appropriate treatment or surgical strategy (1, 8, 16, 18, 28).

Second, if operative treatment is needed, high-quality CT scans of cervical spine can help physicians choose the proper surgical technique for AAD to avoid various operative complications. The authors usually use the posterior C1–C2 pedicle screw fixation and fusion for most patients with AAD in our institutions, when the technique is feasible (Fig. 6).

Figure 6.

A 53-year-old male patient with the diagnosis of AAD. Top row left: preoperative X-ray lateral view showed that old type II odontoid fracture caused AAD; top row right: postoperative X-ray lateral view showed that posterior C1–C2 pedicle screw fixation and fusion was performed, and dislocation was reduction satisfactory; low row left and right: axial CT scans showed that the normal size of C1 and C2 pedicle screws were safely placed on bilateral sides.

It should be kept in mind that, however, C1–C2 pedicle screw placement itself is technically demanding and some patients with hypoplasia of pedicle of the atlas and axis, high riding of vertebral artery and bony anomaly are not rare in patients (29). In these situations, the technique of posterior C1–C2 pedicle screw fixation and fusion is not suitable, and alternative techniques, such as lateral mass pedicle fixation, lamina pedicle and occipitocervical fusion with C3 rather than C2 as caudal anchors, could be used.

Thin slice CT scans with multiplanar reconstruction images allow assessment of the presence of bony structures anomaly, small size of C2 pedicle and anatomy variations of vertebral artery in patients with AAD and help physicians choose the alternative technique, which is of paramount importance to choose proper surgical strategy and decrease the rate of surgical risk (Figs 7, 8, 9) (12, 29, 30, 31, 32).

Figure 7.

A 79-year-old female patient with AAD and assimilation of atlas. Left: sagittal CT scan; right: axial CT scan showed small size of C2 pedicle on the two sides.

Figure 8.

The same patients of Fig. 7. Left: posterior occipitocervical fixation and fusion with C2 lamina screws as caudal anchors. Right: axial CT scan showed that C2 lamina screws were placed.

Figure 9.

A 66-year-old female patient with AAD and cranial setting. Top row left: sagittal CT scan showed AAD with cranial settling and C2-C3 congenital fusion; top row right: axial CT scan showed small size of C2 pedicle on the left side; low row left: axial CT scan showed normal size of C3 pedicle on bilateral sides, which could safely place pedicle screws; low row right: X-ray showed posterior occipitocervical fixation and fusion with C3 pedicle screws as caudal anchors.

Third, CT scans with multiplanar reconstruction enable the computed tomography angiography (CTA) as a routine modality (Fig. 10). Thin slice CT scans with multiplanar reconstruction images and CTA can detect anatomy variations of vertebral artery, vessel anomalies and dominance artery on one side and so on. These potential vessel variations may cause disastrous consequences if surgical treatment is performed without identification, making them the mandatory modality before operation (2, 29, 32, 33, 34, 35, 36). So, for ADD patients who need surgical treatment especially including C1 screws, a CTA is mandatory (37, 38).

Figure 10.

Anterior view (left) of the three-dimensioned reconstruction of cervical spine and CTA showed vertebral arteries on bilateral sides; left side view (right) of the three-dimensioned reconstruction and CTA showed vertebral artery on the left side.

For trauma patients, high-quality CT scans surpass conventional radiographs in detecting AAD and other cervical spine injuries, especially in patients with AAD caused by atlas fractures and the transverse ligaments rupture (14). Early published researches supported that CT scans outperform plain radiographs in patients with cervical trauma, and a CT scan of the cervical spine could be the first screening imaging for some patients (39, 40, 41, 42). In a series of 1,199 patients, Griffen et al. (39) evaluated using plain radiographs and CT scans. The investigators found that 41 (3.2%) patients with cervical spine injuries were missed by plain radiographs, which could be detected only by the CT, and there were no missed cervical spine injuries among the patients evaluated by CT scans (39). Recently published researches have confirmed the importance of CT scans (43, 44).

In a series of 5,676 blunt trauma patients evaluated by CT scans, Vanguri et al. found that the incidence of any injury to the cervical spine was 7.4%, and the incidence of fracture was 7.2%, respectively; the incidence of ligamentous injury was 0.92% (52 of 5,676), and 20 of the 52 were suspected with ligamentous injuries by CT scans (28). The interesting finding of Vanguri et al.’s study was that, for the remaining patients who were not suspected with ligamentous injuries, all had associated fractures, which were identified by CT, and investigators concluded that CT scan of the cervical spine had a 100% sensitivity and specificity in ruling out cervical spine injuries, which was an excellent tool to rule out cervical spine injuries without the need for further imaging (28). In a retrospective study, Gargas et al. (45) evaluated 173 trauma patients who were younger than 18 years, and found that high-quality CT scans with sagittal and coronal reconstructions might be comparable with MRI for the detection of unstable cervical spine injuries in the pediatric trauma population. In summary, compared with other imaging modalities, the higher sensitivity and specificity of CT examination makes CT scans of cervical spine the modality of choice for patients with suspected cervical spine injuries in emergency department of most hospitals or many trauma centers.

Although a CT scan is sensitive, it should not be the first-line screening tool routinely for all patients with AAD who require radiographic evaluation, because high-quality CT scans with multiplanar reconstruction cause much more radiation exposure compared to plain radiographs, especially in children. Therefore, the initial assessment tool for AAD should be plain radiographs, and MRI should be the next consideration after plain X-rays that do not diagnose children with AAD, and a CT scan should be reserved for patients in whom more diagnostic certainty is required or when patients require further treatment. CT scans with multiplanar reconstruction are reported to represent the so-called ‘gold-standard’ radiological assessment for bone changes in patients with AAD; however, the limitations of this method are that it provides inferior visualization of soft tissue and neural elements compared with MR imaging (16, 23).

Magnetic resonance imaging

Although the assessment using plain radiographs and a CT scan is enough to give accurate diagnosis for most patients with AAD, more diagnostic certainty for the compression of spinal cord is required or when patients require further treatment. For patients with AAD, if the sign of neurological deficits is present or suspected, magnetic resonance imaging (MRI) of the cervical spine is suggested for detecting the compression of spinal cord, which might enable much more accurate diagnosis and treatment planning. Simultaneously, if the X-ray does not diagnose children with AAD, MRI should be the next choice that decreases the risk of radiation exposure. The main role of MRI in evaluating patients with AAD is to detect probable spinal cord injury showing abnormal signals of spinal cord in T2, and the location and extent of the spinal cord if it is present. MRI can also clear the suspected disco-ligamentous injury, soft-tissue injury and associated diseases or injuries of subaxial cervical spine.

In our institutions, MRI of cervical spine is generally recommended for patients with AAD who have an abnormal neurologic examination or require special investigation of the involvement of spinal cord compression. Atlantoaxial dislocation can be accompanied by spinal cord injury in subaxial cervical spine, when congenital anomaly or lesion (congenital C2-3 fusion, developmental cervical spinal stenosis) is present. Sometimes, the use of MRI can alter the treatment protocols (Fig. 11), and spine surgeons should keep it in mind.

Figure 11.

A 52-year-old female patient with AAD (also termed as atlantoaxial rotatory subluxation or fixation) and C2-C3 congenital fusion. Left: coronal CT showed atlantoaxial rotatory subluxation and left-right asymmetry of atlanto-odontoid space; middle: MRI of cervical spine showed high signal of spinal cord in T2; right: sagittal CT scan after laminoplasty performed.

For patients with traumatic AAD, MRI can be used to detect spinal cord injury, locate the level of neurologic deficits, exclude severe soft-tissue injury and help with determining the treatment planning. MRI was reported to fail to depict all major ligamentous disruptions; however, it particularly helped with the better detection of spinal cord injuries and defining the type of spinal cord compression (46). Although early studies reported that MRI provided a low sensitivity for ligamentous injuries, increased magnetic field strength and new imaging sequences high-resolution (e.g. MRI with 3.0-T images rather than 1.5-T images, fast spin-echo T2-weighted and gradient-echo pulse sequences are commonly used now) have significantly improved the sensitivity and specificity for detecting ligamentous lesions (14, 47, 48). For patients with altered level of consciousness who could not be clinically evaluated within 48 h of injury, if patients with suspected spinal cord compression, MRI of cervical spine is indicated (14).

Whenever a neurologic deficit with or without radiographic findings is present, MRI is useful to assess the extent of injury and ligamentous involvement. MRI is helpful in patients with a suspicious diagnosis of traumatic AAD, which is spontaneously reduced, where plain films are equivocal or even normal.

It is widely accepted that MRI is the most sensitive method for the evaluation of cervical spine involvement in RA, especially because it allows better visualization of soft tissue and neural elements. So, it should be performed in all patients with suspected or confirmed radiographic signs of cervical spine involvement and in patients complaining of neurological symptoms.

The limitations of MRI for AAD are as follows: i) it is difficult to transfer comatose or intensive care patients who are ventilated; ii) the time needed to obtain the evaluation of MRI is longer than that of conventional radiographs, lateral flexion-extension dynamic views or CT; and iii) the cost-effectiveness of MRI in patients with AAD need to be evaluated.

MRI of the cervical spine can reveal the precise location and extent of compression, whether it affects the medulla, spinal cord or both, which is helpful when selecting an operation strategy and assessing prognosis (13). The assessment of AAD with the combination of conventional radiographs, lateral flexion-extension dynamic views, thin slice CT with multiplanar reconstruction and MRI is recommended by many spine surgeons (1, 13).

Classification of AAD

Apart from the classification of AAD, which are based on various characteristics, including etiologies (trauma, inflammatory, congenital abnormalities and iatrogenic condition), direction of displacement (anterior, posterior, lateral, central and rotational) or interval between the onset of AAD and a definite diagnosis (fresh and old, which is termed when the diagnosis is made after the onset of AAD within or over 3 weeks, respectively) and clinical classification of AAD, which can help spine surgeons with choosing treatment strategies, are also developed (1, 9, 10, 11, 12, 13). Several commonly used clinical classifications of AAD are discussed as follows.

In 1968, Greenberg initially focused on the classification of AADs and classified AAD into two subcategories – reducible and irreducible AAD – and investigator also provided treatment recommendations based on this classification (49). Greenberg is believed to make a landmark study for AAD by many investigators, and his work still helps many spine surgeons with treatment for AAD (1, 49).

In 1977, Fielding & Hawkins subsequently developed a classification for AAD, which has been widely accepted for atlantoaxial rotatory subluxation or fixation (50). However, the limitation of Fielding & Hawkins’ classification system is that it has limited clinical significance in grading severity of lesions or choosing treatment methods (1, 13).

In 2003, Yin et al. developed a classification for AAD, in which AADs were classified into three clinical types as follows: i) reducible type, which could be easily reduced by traction; ii) irreducible type, which could be only reduced by the combination of transoral release and traction; and iii) non-reducible type, which usually were associated with solid fusion of the lateral atlantoaxial joints, could not be reduced even by the combination of transoral release and traction (51).

In 2007, Tan et al. developed a new classification for AAD, which is effective in offering therapeutic regimen with strong guidance to the clinic practice (9). Tan et al. classified AAD into three types, which included: type T AAD meant that the dislocation could be reduction by traction type, and type T was further divided into T1 and T2 subtypes based on different cause and course of disease; type O AAD meant that the dislocation could be reduction by operation type; and type I AAD meant that the dislocation was irreducible type (9). Tan’s classification system was also named as TOI classification. Treatment protocols recommended by Tan et al. are as following: for type T1 AAD, applying traction or orthosis for 8–10 weeks is used; for type T2, reduced by traction followed by posterior C1–C2 fixation and fusion is usually performed; for type O, a transoral anterior release followed by posterior fixation is performed; and for type I, decompression and fixation in situ is used. TOI classification of AAD is a comprehensive clinical classification, which can give proper guidance in the treatment for different types of AAD (2, 11, 52).

In 2012, Wang et al. also proposed a classification system, which aimed to improve the management of AAD (12). Wang classification system primarily classified AADs as reducible or irreducible ones, which were divided into two subtypes, respectively. As a consequence, it included four types after evaluation using dynamic views, CT and multi-planar reconstruction and skeletal traction test. In Wang classification, type I AAD meant instability, and the dislocation was reducible in dynamic X-rays; type II was reducible dislocation, and the dislocation was reducible with skeletal traction under general anesthesia; type III was irreducible dislocation, and the dislocation was not reducible with skeletal traction under general anesthesia; and type IV was bony dislocations, and the dislocations with bony anomalies were not reducible with any traction. Type I and II AADs are recommended to be treated with posterior fusion procedure, and type III dislocations are treated with transoral released anteriorly before posterior fusion. Type IV AAD, which is a bony dislocation and is not reduction with any traction, is recommended to be treated with transoral odontoidectomy (1). Just like TOI classification of AAD devised by Tan et al., Wang classification system also helps in the diagnosis and treatment protocol for managing patients with AAD. In fact, both classifications have developed the basic principle treatment for AAD.

The value of these abovementioned classifications of AAD is useful in determining treatment strategies for different conditions since highly varying opinions on indications of different treatment methods are present (1, 53, 54, 55, 56). Based on the TOI classification of AAD, a proper treatment is recommended for AAD to avoid the development of severe neurological deficit and decrease the rates of morbidity and mortality associated with AAD (9, 10). Essentially, the treatment recommendations based on the Wang classification system are the same as the TOI classification of AAD devised by Tan et al., and both the classification systems can provide the clinical and radiographic parameters for the clinicians (9, 11, 12).

Conclusion

Conventional X-ray views are the basic imaging technique for the diagnosis of AAD, including an open-mouth odontoid view and a cross-table lateral view, and lateral flexion-extension dynamic views are only used as an additional supplement in some special cases. Conventional X-ray sometimes is insufficient for the diagnosis of AAD, and an accurate diagnosis can be made based on the combination of the conventional radiographs and CT scan. MRI is generally recommended for patients with AAD who have an abnormal neurologic examination or require special investigation for the involvement of spinal cord compression. Sometimes, the use of MRI can alter the treatment protocols.

For patients with AAD who need surgical treatment, before any technique is used, extensive preoperative imaging, including plain radiographs, thin slice CT scanning with multiplanar reconstruction, CTA (especially including C1 screws) and MRI, is essential for operative planning. Compared to MRI , CTA can detect anatomy variations of vertebral artery, vessel anomalies and dominance artery on one side intuitively in the visual effect.

The value of the classifications of AAD is useful in determining treatment strategies for different conditions based on the clinical and radiographic parameters since highly varying opinions on indications of different treatment methods are present. The TOI classification of AAD is a comprehensive clinical classification, which can provide the clinical and radiographical parameters for the clinicians and give proper guidance in the treatment for different types of AAD.

ICMJE Statement of Interest

The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of this work.

Funding Statement

This work did not receive any specific grant from any funding agency in the public, commercial or not-for-profit sector.

Author contribution statement

GZL helped in investigation, data curation, formal analysis, writing of the original draft, reviewing and editing and supervision. HZ helped with investigation, data curation, formal analysis and writing of the original draft. QW helped in investigation, data curation and reviewing.