Measurement Techniques for Upper Cervical Spine Injuries: Consensus Statement of the Spine Trauma Study Group

Christopher M. Bono, MD; Alexander R. Vaccaro, MD; Michael Fehlings, MD; Charles Fisher, MD; Marcel Dvorak, MD; Steven Ludwig, MD; James Harrop, MD


Spine. 2007;32(5):593-600. 

In This Article


There are several methods to describe the spatial relationship between the occiput, atlas, and axis.[11] The most popular in the setting of trauma are the basion-dental interval (BDI) and the basion-posterior axial line interval (BAI). Originally described by Harris et al,[9,10] the BDI and BAI have been carefully examined (Figure 1). In the first of 2 companion studies, Harris et al[10] measured the BAI and BDI on lateral radiographs of 400 normal adults. The BAI and BDI did not exceed 12 mm in 98% and 95% of adults, respectively. The BDI and BAI have come to be known as Harris Measurements, and more descriptively as the Rule of Twelve. Deliganis et al[12] recommended use of the BDI and the BAI to help detect occipitocervical dissociations. Likewise, Fisher et al[8] found these to be the most useful radiographic parameters for these injuries.

Measurement technique for the BDI and BAIs.

In their second study, the investigators retrospectively reassessed lateral radiographs of 37 patients in whom occipitoatlantal dissociation had been previously diagnosed at the time of admission.[9] They found the BDI and BAI to be greater than 12 mm in 23 patients (group 1) with frank occipitoatlantal dislocation and 8 (group 2) with occipitoatlantal subluxation/dissociation. Measurements were less than 12 mm for 6 patients who initially had suspected instability but did not have supportive clinical findings. In the same patients, the Power ratio (Figure 2) and Lee X line method (Figure 3) could not be measured in 17 of 37 cases because either the opisthion could not be seen or the posterior C1 arch had a developmental anomaly (i.e., not fused). In the remaining patients, these methods enabled detection of only 60% and 20% of injuries, respectively. While these data strongly suggest that BDI/BAI are superior to the Power ratio and Lee X lines, the interobserver and intraobserver reproducibility of these measurements has not been assessed to the authors' knowledge.

Measurement technique for the Power ratio.

Measurement of Lee's lines.

Other lines and measurements have also been described for the craniocervical junction. The Power ratio (Figure 2) describes the relationship between the occiput and C1 through a ratio of the distance between the basion and C1 posterior arch divided by the distance between the opisthion and anterior C1 arch. Ahuja et al[11] found lower Power ratios in 5 surviving patients compared to a higher ratio seen in 1 fatal case of occipitoatlantal dislocation.

Most other descriptive measurements of the occipitocervical junction were originally developed to assess rheumatoid patients with basilar invagination (cranial settling). Thus, the critical pathologic values are usually seen in the lower, rather than upper limits of normal due to the process of settling rather than distraction. These measurements, although possibly useful for some injuries, were not intended to detect widening or translation of the occipitocervical junction. They include: (1) the Chamberlain line, drawn from the hard palate to the tip of the opisthion (Figure 4); and (2) McCrae's line, drawn from the basion to the opisthion, representing the anteroposterior width of the foramen magnum (Figure 5).

The Chamberlain line.

McCrae's line.

Considering the available data, the current authors' feel that Harris measurements (BDI/BAI or the Rule of Twelve) are probably the most useful, sensitive, and reproducible radiographic parameters for detecting and characterizing occipitocervical dissociation and dislocations.

Midsagittal computed tomography (CT) reconstruction is recommended, although a lateral cervical plain radiograph may be used if it is centered on the occipitocervical junction and is targeted from 6 ft with appropriate magnification correction.

The most inferior and posterior aspect of the basion is marked. The superior tip of the odontoid process is also marked. The distance between these 2 points (BDI) is measured in millimeters. A value greater than 12 mm is highly suggestive of occipitoatlantal dissociation (Figure 1).

A vertical line is drawn along the posterior vertebral border of the C2 body. This line should extend superiorly, just past the level of the foramen magnum. A perpendicular line is drawn from this line to the basion mark. This distance is then measured in millimeters (BAI). A value more than 12 mm should be considered abnormal and indicative of an anterior occipitoatlantal dissociation.

Anderson and Montesano[17] described a classification system for occipital condyle fractures. In essence, it distinguishes injuries that are primarily bony versus those that are primarily ligamentous with only small bony flecks. Tuli et al[18] distinguished fractures based on displacement and on the presence of ligamentous instability. Others have noted that condyle injuries with larger bony fragments have a greater likelihood of stable healing with nonoperative treatment, such as a halo fixator.[19,20] From these findings, it appears that the size of the occipital condyle fragment is an important determinant of treatment and outcome; however, there is currently no standard method to measure the size of occipital condyle fragments or the magnitude of fracture fragment displacement. An analysis of the influence of fragment size and displacement on stability and outcome has not been performed.

Axial CT, parasagittal, and coronal reconstructions.

The occipital condyles are often not well visualized on plain radiographs.[21] Better bony detail is afforded by CT images through the occipitocervical junction.[20,22] While axial CT images are useful in assessing the presence of occipital condyle fractures, they do not usually enable visualization of the entire condyle in 1 image. For these reasons, the authors suggest that a parasagittal and coronal CT reconstruction through the approximate midaspect of the condyle be used. As principles of fracture healing dictate, the area of apposed fracture surfaces can influence the potential for fracture healing. Thus, a similar quantification of the apposed contacting surfaces of occipital condyle fracture fragments is proposed. In regards to displacement, greater amounts imply greater concomitant soft tissue disruption and instability. The ability to reduce the fractured fragments would also influence healing potential.

Because actual detailed quantitative measurements would be exceedingly difficult, the length of the fracture line is measured in millimeters on the midsagittal and coronal condyle images (Figure 6). The coronal and sagittal measurements are added to derive a total amount of bony apposition for each condyle. For shell-type avulsion fractures that leave a rim of subarticular cortical bone juxtaposed to the upper C1 articular process, the length of apposed fracture surface would be essentially complete. This would suggest that the fracture would have a high likelihood of healing. The same would be true for a fracture through the junction of the condyle and occiput, with minimal displacement. In contrast, a primarily ligamentous injury with a small bony fleck would have a short measured distance of apposed bone and a presumed lower rate of healing. Future prospective evaluations using this measurement technique and its relationship to stability and healing rates need to be performed to support or disqualify the current belief that fragment size is a prognostically important factor.

Technique for measuring displacement (gap) and apposition of occipital condyle fracture fragments. It is suggested that injuries with small fracture fragments (flecks) are primarily ligamentous that may have less healing potential than broad fracture surfaces that are well aligned and minimally gapped.

On coronal and parasagittal reconstructed images, the maximal gap between the fractured fragments are measured in millimeters (Figure 6).

Lee et al[23] used an open-mouth view to measure displacement of C1 lateral mass fractures. Most authors seem to agree on this method of assessment. Spence determined that 6.9 mm is the critical amount of total displacement necessary to disrupt the transverse ligament. This measurement was derived by direct cadaveric measurements, however. Heller et al[24] warned of overestimating the amount of displacement based on plain films. They determined that the transverse ligament should be considered intact in patients with less than 8.1-mm total displacement as measured on a plain open-mouth radiograph. This consideration may be obviated using calibrated axial and coronally reconstructed CT images.

Coronal CT-reformatted images through the center of the lateral masses of the atlas.

Vertical lines are drawn along the most lateral aspect of the bone of the C1 and C2 articular processes (Figure 7). The transverse distance between them is then measured in millimeters. These are then added to calculate the total lateral mass displacement.

Measurement technique for lateral mass overhang.

While the atlanto-dens interval (ADI) and posterior atlanto-dens interval (PADI) are widely used to detect traumatic cervical instability, the authors could find no article assessing their reliability in the setting of acute injury. Studying the flexion-extension radiographs of 72 patients with Down syndrome, Wellborn et al[13] found ADI to have statistically significant intraobserver agreement in 2 of 3 observers. Although the interobserver agreement was considered fair, it was statistically better than measurements using the Power ratio. Supportively, Cremers et al[25] found the ADI to be reliable in detecting instability in 279 children with Down syndrome. Intraobserver and interobserver agreement was better using flexion versus the neutral radiograph. The PADI represents the anteroposterior diameter of the spinal canal at this level. It has been shown to be a more useful prognosticator in rheumatoid arthritis patients than the ADI. However, it has not been validated in the same manner for traumatic atlantoaxial instability. With traumatic atlantoaxial instability being relatively uncommon and the aforementioned measurements having undergone significant epidemiological rigor for other pathology, the authors propose the use of the ADI and PADI as described below.

Lateral cervical radiograph or midsagittal CT reconstruction.

The craniocaudal midpoint of the anterior ring of C1 is marked. A line parallel to the ring of C1 is drawn from this point toward the odontoid process. The distance between the C1 mark and intersection with the anterior aspect of the odontoid process is measured in millimeters (Figure 8).

The ADI and PADI.

Rotation between the C1 and C2 rings can occur by itself or in combination with a sagittal translational deformity. Most reports of atlantoaxial rotational instability are in children, commonly associated with nontraumatic upper pharyngeal infections. Most recommendations concerning optimal measurement of atlantoaxial rotation are derived from this body of literature.[26,27,28] Most authors agree that CT is the imaging modality of choice for detection and quantification of rotational deformity,[26,27,29] though its reliability and accuracy for detecting dynamically unstable joints has been questioned.[30] In the authors' experience, dynamic CT is rarely indicated (and can be dangerous) in the adult patient with traumatic instability.

Transaxial CT images through mid body of C2 and mid body of C1.

For optimal measuring, the CT gantry angle should be aligned along the transverse plane of the upper cervical vertebrae. An axial CT slice at the level of the C2 body and C1 ring are then obtained. An anteroposterior line is drawn from the midpoint of the C2 body to the center of the spinous process. A perpendicular to this line is drawn along the posterior C2 vertebral body. On the best axial slice of C1, the midpoints of the transverse (vertebral artery) foramens are marked, and a line is drawn between them. The angulation between these 2 lines represents the degree of static atlantoaxial rotatory deformity (subluxation) (Figure 9). By convention, the side (right or left) toward which the atlas (head) points is considered the side of the rotation.

Measurement of right (R) (A) and left (L) (B) atlantoaxial rotation.

Odontoid fracture displacement and angulation are known to be important prognostic factors of fracture healing. However, in the authors' review of the literature, there are no assessments of the optimal method for measuring these parameters. Carlson et al[15] measured displacement by drawing lines along the anterior aspect of the dens fragment and the intact caudal body of C2. The angle subtended by these lines would be the degree of fracture angulation. The location of the apex of fracture angulation would be described as anterior or posterior.

Lateral cervical radiograph or midsagittal CT reconstruction.

A tangent line is drawn along the anterior aspect of the odontoid fragment and the anterior aspect of the C2 body. At the level of the fracture, a transverse line is drawn connecting these 2 lines. This distance is measured in millimeters and represents sagittal fracture displacement (Figure 10).

Odontoid fracture displacement.

A tangent line is drawn along the posterior aspect of the odontoid fragment and the posterior aspect of the C2 body. The angle subtended by these lines would be the degree of fracture angulation. The location of the fracture apex angulation would be used for the descriptor anterior or posterior (Figure 11).

Odontoid fracture angulation.

Levine and Edwards[14] described a method of measuring both angulation and translation of C2 with traumatic spondylolisthesis (Hangman's fractures).

Lateral cervical radiograph or midsagittal CT reconstruction.

Endplate method. As depicted in Figure 12, lines are drawn perpendicular to the inferior endplate of C2 and C3 to measure angulation.

C2-C3 angulation measurement using the endplate method.

Similarly, posterior vertebral body lines are drawn along the posterior aspect of C2 and C2 to enable measurement of angulation (Figure 13) and displacement (Figure 14). Both of these parameters are important in classification and treatment decision making for this injury.

C2-C3 angulation measurement using the posterior vertebral body line method.

Measurement of C2-C3 translation.