C0 (Occipital Condyle) Fractures

Traditionally, occipital condyle fractures are categorized by the system of Anderson and Montesano.² This grades occipital condyle fractures according to occipital fracture (type 1), large condyle fracture (type 2), and avulsion condyle fracture (type 3). The inference from this study is that small condyle fractures represent disruption of the alar ligament. The disruption of the alar ligament has been demonstrated to increase the mobility of the C0-1 joint.³

Tuli et al. defined fractures according to evidence of ligamentous instability.⁴ Type 1 represents large bony fractures, condensing Anderson and Montesano types 1 and 2 into one group. Type 2 is both small bony fractures of the condyle. The fractures are further subdivided into 2a (stable) and 2b (unstable). Instability is characterized by MRI evidence of alar ligament disruption or CT/radiographic criteria. However, use of MRI to assess disruption of the alar ligament remains controversial.⁵

To simplify the issue, Maserati et al. focused on the C0-1 joint.⁶ Determination of instability is made using the elongation of the distance of the C0-1 joint described by Pang.⁷ This method is also used to determine AOD and will be more completely described in the next section.

Unstable condyle fractures are a form of AOD and need to be treated as such.⁶ However, once instability has been determined, treatment is also not standardized. Fractures without apparent ligamentous disruption can be treated conservatively with a cervical collar or halo vest. Immobilization may be performed if the fracture fragment is large enough and aligned enough to allow bony fusion.¹ If the bony fragment appears small or there is an apparent alar ligament disruption, it may be necessary to perform an occipital cervical fusion because purely ligamentous injury is unlikely to heal by immobilization.⁶

C0-1 Fractures or Atlanto-Occipital Dislocation

In the diagnosis of AOD,vigilant clinical suspicion is most important. The deformity may reduce spontaneously because of recoil of the elastic ligamentous structures. Suspicion should be raised based on the mechanism of injury (e.g., high-velocity crash) or findings on neurologic examination (severe neurologic injury, brainstem or C1-2 level deficits), lateral cervical spine radiograph (obvious separation of the condyle-C1 joint or C1-2 prevertebral swelling), or head CT (subarachnoid hemorrhage around the brainstem or upper cervical spinal cord, or epidural/subdural blood at C1-2).^7–9

AOD is determined by measurements made from normal plain radiographs. These techniques are the Powers ratio,¹⁰ basion-axial interval (Harris),¹¹ and the Wholey dens-basion interval (Fig. 92-1). These measurements essentially infer dislocation based on measurement of structures remote from the occipital condyle–atlas joint,^10–12 which can lead to false-negative examinations and lack of interobserver reliability. It has been found that the diagnostic sensitivities for the common tests range from 25% to 50%, with false-negative rates of 50% to 75%. However, the diagnostic sensitivity of the nonstandard indicators (perimedullary blood, tectorial membrane damage, C1-2 extra-axial blood) is 63% to 75%.⁷

FIGURE 92-1 Classic determination of the atlanto-occipital dislocation. A, Power ratio is calculated by measuring distance between the basion (B) and the midpoint of the dorsal arch of C1 (C) as well as the distance between the opisthion (O) and the ventral arch of C1 (A). The normal BC/OA ratio is less than 0.9, borderline is 0.9 to 1.0, and greater than 1.0 is abnormal. Surprisingly, this measure is not sensitive to dislocations in the longitudinal or dorsal direction. B, The basion axial interval (BAI) is the distance between the basion and the rostral extension of the dorsal cortical line of the axis. In adults, the basion is between −4 mm and +12 mm. In children, it is between 0 and 12 mm. C, The basion dental interval (BDI) is the distance between the basion and the superior cortex of the dens and is less than 12 mm in normal subjects.

An increase in the measurement of the joint distance between the occipital condyle and C1 can be used to determine AOD. This is called the condylar distance. Thin-slice axial CT scanning allowed Pang et al. to calculate that the distance should be less than 4 mm in pediatric patients (Fig. 92-2). This test has been shown in the pediatric population to have a diagnostic sensitivity of 100%.⁷ Dziurzynski et al. showed that in adult patients a condylar distance greater than 2 mm was diagnostic of AOD. This has a sensitivity of 92% and specificity of 95%.¹³

FIGURE 92-2 Condyle–C1 interval. Two images are selected from the midpoint of the joint space in sagittal (A) and coronal (B) CT reconstructions. Measurements of the joint space are selected at four equidistant points. The average of the measurements in each view should be less than 4 mm in children and 2 mm in adults.

If the patient survives the initial injury, he or she should be immediately immobilized. The use of a halo vest to immobilize the patient has been shown to be a safe and effective treatment method to prevent delayed neurologic deterioration while the patient is stabilized and prepared for definitive treatment.¹⁴ Depending on the severity of injury, operative fixation can be performed on an elective basis.¹⁴ The instability of AOD is primarily a ligamentous injury, and therefore internal fixation and fusion is recommended for definitive treatment. If reduction of the AOD is necessary, it should be done with gentle manual manipulation under fluoroscopic guidance. If the patient has a neurologic examination to follow, the reduction can be performed with the patient under mild sedation. In the anesthetized patient, somatosensory evoked responses may provide some help in determining if reduction is affecting the patient neurologically.

C1 Fractures

Fractures of the atlas (C1) can manifest in multiple ways: isolated ventral or dorsal arch, burst, and lateral mass fractures. Isolated arch fractures are a controversial diagnosis because it is unlikely that a ring can have a fracture in one place without fracturing in another, although they have been described.¹⁵ An axially directed force that translates into C1 through the wedge-shaped occipital condyles causes burst fractures of the atlas. These fractures were first described by Geoffrey Jefferson in 1920.¹⁶ These fractures are detected with an open-mouth odontoid radiograph demonstrating spread of the lateral masses of C1 beyond the lateral borders of the C2 lateral masses. Assessment of the integrity of the transverse ligament is critical in determination of the treatment of C1 burst fractures. Initial assessment of the competence of the ligament was made by a cadaveric study performed by Spence et al.¹⁷ in 1970. Spence showed that the transverse ligament typically failed if the spread between lateral masses was 6.9 mm or more. When corrected for the magnification of the radiographs, this distance should be increased to 8.1 mm.¹⁸ This allows for indirect determination of rupture of the ligament based on plain radiographs. Again, the advent and widespread use of CT and MRI have allowed for direct visualization of ligament integrity. Dickman et al. used MRI to evaluate the transverse ligament and found an abnormal atlantodental interval of 3 mm or more implies the incompetence of the transverse ligament.¹⁹ A ruptured transverse ligament was found in cadaver studies to produce hypermobility at C1-2, increasing flexion-extension (42%), lateral bending (24%), and axial rotation (5%).^20–22

There is not enough evidence to provide standardized treatment guidelines, but there are recommendations for this treatment of C1 fractures.²³ Isolated ventral or dorsal ring fractures may be treated with cervical immobilization (collar or halo) for 8 to 12 weeks with good results. C1 burst fractures without ligamentous injury can be treated with collar or halo immobilization for 12 weeks. C1 burst fractures with rupture of the transverse ligament may be treated with halo immobilization for 12 weeks or with internal fixation of C1 to C2 with fusion.

C2 Fractures

C2 fractures can be broadly divided into odontoid, C2 body, and pedicle/pars fractures. Odontoid fractures are classified by the system of Anderson and D’Alonzo (Fig. 92-3).²⁴ Type 1 fractures are rare and are at the distal tip of the odontoid process. Type 2 fractures occur at the base of the odontoid where it meets the body of the axis. Type 3 fractures occur through the body of the axis. The management options for odontoid fractures depend on the type of fracture, the degree of subluxation of the cranial fragment, and the status of the transverse ligament. Type 1 and type 3 fractures are often managed by external immobilization alone, collar or halo. Type 2 fractures can be managed by immobilization or operative intervention depending on patient factors and the degree of subluxation. An increased rate of nonunion has been associated with patient age younger than 60 years and/or subluxation greater than 4 to 6 mm.^25–27 Nonunion rates can be as high as 28%. Type 2a fractures, characterized by comminution of the C2 body, are associated with lower healing rates without surgery.^28,29 C2 pars and pedicle fractures may require surgical intervention, depending on the degree of angulation and distraction between the fragments (see subsequent discussion).^27,29

FIGURE 92-3 Odontoid fracture types and management. Type 1 fractures are fractures of the tip of the dens and can be managed in a cervical collar. Type 2 fractures are fractures of the base of the dens, and treatment varies depending on age and distance of displacement. Patients with fractures with greater than 4 to 6 mm of displacement and those older than 55 years of age are less likely to heal with external immobilization and may benefit from surgical stabilization. Type 3 fractures are fractures that extend through the dens and the body of the axis and can be managed in a cervical collar.

Os odontoideum is defined as an ossicle of cortical bone in the position of the odontoid process often attached to the C2 body by a cartilaginous segment (Fig. 92-4). The cause of this remains unclear. There is some evidence to suggest that this is a consequence of old trauma, often at an early age.³⁰ It is unlikely that this is a failure of fusion during development, because the normal somite pattern of development of the axis does not normally have a site of fusion where the axis meets the body.³¹ However, os odontoideum is associated with congenital disorders, such as Down and Morquio syndromes, and spondyloepiphyseal dysplasia. Patients who have neurologic compromise are offered surgical decompression and fusion. Patients with gross instability or narrow canal diameter are also offered surgery. The treatment of incidentally found os odontoideum is controversial. Most authors recommend close follow-up, with surgery reserved for the development of symptoms or radiographic evidence of instability or progressive deformity.³¹

FIGURE 92-4 Os odontoideum is defined as an ossicle of cortical bone in the position of the odontoid process often attached to the C2 body by a cartilaginous segment. The T2-weighted MRI demonstrates hypointensity at the base of the odontoid. The dynamic radiographs demonstrate movement in flexion and extension.

Fractures of the C2 pars interarticularis are called hangman’s factures because of the similarity to those seen in judicial hangings.³² These fractures are also called C2 traumatic spondylolisthesis fractures. These fractures have been classified into three types by Effendi et al. (Fig. 92-5).^33,34 Type 1 fractures are displaced less than 2 mm and minimally angulated, and the C2-3 disc space remains intact. Type 2 fractures have a displaced and angulated body of the axis and a disrupted C2-3 disc space. Type 3 fractures are like type 2 fractures with locked C2 and C3 facets, and the body of the axis is ventrally displaced.

FIGURE 92-5 Effendi classification of C2 traumatic spondylolisthesis. Type 1 fractures are displaced less than 2 mm and minimally angulated. The C2-3 disc space remains intact. Type 2 fractures have a displaced and angulated body of the axis and a disrupted C2-3 disc space. Type 3 fractures are like type 2 fractures with locked C2 and C3 facets.

Decisions of treatment of C2 pars fractures are primarily guided by degree of subluxation of C2 on C3. A type 1 fracture without significant ligamentous injury can be treated with immobilization. A halo ring can be used to achieve reduction by extension and capital flexion, reversing the mechanism of fracture. When significant ligamentous injury exists, care must be taken with the use of traction to avoid iatrogenic separation of C2 and C3. In type 2 or 3 fractures, if there is displacement greater than 3 mm, operative intervention may be indicated for reduction and fixation.^25,34,35

C2 transverse process fractures do not cause instability, but potential injury to the vertebral artery is an area of concern. It is unclear whether aggressive imaging or treatment of these injuries affects patient outcomes, and decisions should be individualized depending on patient symptoms and anatomy.³⁶