Radiological Evaluation of Cervical Spine Involvement in RA
Radiological Evaluation of Cervical Spine Involvement in RA
The classical diagnostic measurements for AAS are based on plain radiographs. Of note, most of these criteria were published before the advent of modern CT and MRI, which can clearly visualize subluxation of the facet joints and all the bony landmarks of the craniocervical junction. However, the craniocervical relationships proposed in the plain radiography era are still used for the initial evaluation, avoiding the cost of routine CT or MRI. Some of these plain radiographic criteria are presented below.
Anterior Atlantodental Interval. The normal value of the anterior atlantodental interval (ADI)—the distance from the posterior border of the anterior tubercle of the atlas to the dens—is less than 3 mm in healthy adults (Fig. 1). As the ADI increases, the chance of spinal cord compression progressively increases. Some authors report that when the anterior ADI exceeds 8 mm, surgery is recommended, as this value suggests total rupture of the transverse and alar ligaments. However, most no longer use the anterior ADI for evaluating patients with RA, as the posterior ADI has been found to be a better predictor of paralysis and recovery.
(Enlarge Image)
Figure 1.
Lateral plain radiograph showing the anterior atlantodental interval (designated in this image by ADI) and the posterior atlantodental interval (designated in this image by PDI) as well as the Ranawat index, the distance from the center of the C-2 pedicle to the transverse axis of C-2.
Posterior Atlantodental Interval. The posterior ADI—the distance from the posterior border of the dens to the anterior aspect of the posterior arch of C-1—evaluates the maximum amount of space available for the upper cervical spinal cord. This has been found to be a better predictor of the neurological risk and recovery in the setting of atlantoaxial subluxation (AAS) than the ADI (Fig. 1). The posterior ADI represents the anteroposterior diameter of the spinal canal at this level. In the cervical spine, the cord itself occupies 10 mm of the canal diameter. In addition, it requires 1 mm for the dura and 1 mm for the CSF anterior to the cord, and the same posteriorly, for a total of 14 mm. Therefore, if the available space is less than 14 mm, the cord becomes compressed.
The posterior ADI and anterior ADI should both be measured on images obtained in flexion and extension. Boden et al. found that patients with a posterior ADI greater than 14 mm had a higher rate of neurological recovery after fusion and stabilization, whereas a posterior ADI less than 10 mm was associated with worse clinical outcome.
Neither the anterior ADI nor the posterior ADI can evaluate cord compression by soft tissues, such as a pannus formation in the retro-odontoid region. For this reason, spinal cord compression can occur even when the plain radiographic measurements are in the normative range.
Lateral Displacement of the Atlas Over the Axis. The open-mouth view is useful for evaluating lateral AAS. Rotatory AAS should be suspected when there is asymmetry or lateral displacement of the atlas on the axis by more than 2 mm in an open-mouth view. It should also be suspected when there is asymmetrical collapse of the lateral atlas mass. Lateral displacement can also occur with fractures of the dens. A CT scan should be performed to confirm the diagnosis.
Cranial Settling. Cranial settling is also known as basilar impression, atlantoaxial impaction, superior migration of the odontoid, and vertical subluxation, and there are numerous plain radiographic criteria for making the diagnosis in RA patients.
The diagnosis of cranial settling based on plain radiographs is sometimes a challenge for radiologists and physicians, as osseous structures of the cranial base are superimposed upon the landmarks, especially in the upper cervical spine. Furthermore, erosion of the dens can make it difficult, if not impossible, to identify its tip. Finally, although the terms "basilar invagination" and "basilar impression" are used synonymously by many authors, the former term may be better used when referring to a congenital craniocervical junction anomaly, whereas basilar impression is more accepted for the description of secondary causes of cranial settling, as occurs in RA.
Below, we list some of the many plain radiographic criteria to diagnose basilar impression that have been described in the literature and their original dates of publication (see also Figs. 1–6).
Chamberlain line (1939): Findings are considered positive if the apex of the odontoid is 3 mm above a line from the posterior edge of the hard palate to the opisthion.
(Enlarge Image)
Figure 2.
Lateral plain radiograph showing the McRae, Chamberlain, and Wackenhelm lines for evaluation of the relationships between the occiput, C-1, and C-2.
(Enlarge Image)
Figure 3.
Lateral plain radiograph showing the McGregor line and the Redlund-Johnell measurement from the McGregor line to the midpoint of the caudal margin of the C-2 body.
(Enlarge Image)
Figure 4.
Lateral plain radiograph showing the Clark stations. The odontoid process is divided into 3 equal parts or stations. The position of the anterior arch of the atlas is assessed relative to these stations.
(Enlarge Image)
Figure 5.
Lateral cervical flexion (left) and extension (right) radiographs obtained in a patient with RA and cervical pain refractory to nonsurgical treatment. Note the increase (in flexion) of the anterior ADI, confirming an atlantoaxial instability.
(Enlarge Image)
Figure 6.
Preoperative and postoperative images obtained in a 53-year-old patient with severe RA. A: Lateral cervical radiograph showing the dens protruding into the foramen magnum, with basilar impression. B and C: CT images obtained in extension (B) showing the tip of the dens 5.49 mm above the McRae line and in flexion (C) showing the dens 9.43 mm above the McRae line. D: Sagittal T2-weighted MR image showing the dens protruding into the medulla and posterior compression of the upper spinal cord by the posterior arch of the atlas. E: Sagittal reconstruction of postoperative CT scan obtained after occipitocervical decompression and craniocervical fixation. F and G: Sagittal and coronal CT reconstructions showing the autologous bone graft (black arrow) used as a spacer between C-1 and C-2 to reduce the protrusion of the tip of the odontoid process into the foramen magnum.
McGregor line (1948): Findings are considered positive if the apex of the odontoid is > 4.5 mm above a line drawn from the posterior hard palate to the most inferior point on the occipital curve.
Fischgold and Metzger line (1952): Findings are positive if the apex of the odontoid is above the line connecting the tips of the mastoid processes bilaterally in an open-mouth view.
McRae line (1953): Findings are positive if the tip of the odontoid extends above a line drawn from the basion (anterior rim of the foramen magnum) to the opisthion (posterior rim of the foramen magnum).
Wackenheim line (1974): Findings are positive if the odontoid protrudes posterior to a line drawn extending from the superior surface of the clivus through the spinal canal.
Ranawat criterion (1979): A line is drawn from the midpoint of the C-2 pedicle along the center of the odontoid process until it intersects a horizontal line through the atlas. Findings are positive if the length is < 15 mm in males or < 13 mm in females.
Redlund-Johnell criterion (1984): A line is drawn from the midpoint of the caudal surface of the C-2 body to the McGregor line. Findings are considered positive if the length is < 34 mm in males or < 29 mm in females.
Clark station (1989): The odontoid process is divided into 3 equal parts ("stations") from craniad to caudad in the sagittal plane. The results are positive if the anterior arch of the atlas is in the second or third station. This method is the simplest one, as the relationship does not change in flexion, extension, or neutral views.
Riew et al. evaluated the reliability and sensitivity of the diagnosis of basilar invagination in 131 cervical radiographs obtained in patients with RA, according to the criteria proposed by Clark et al., McRae and Barnum, Chamberlain, McGreger, Redlund-Johnell and Pettersson, Ranawat et al., Fischgold and Metzger, and Wackenheim. As a final conclusion, no single plain radiographic criteria had sensitivity and a negative predictive value greater than 90% as well as a reasonable specificity and acceptable positive predictive value. Therefore, they suggested that the results of screening for basilar impression should be considered positive when at least one of 3 following criteria are positive: the Clark station, the Redlund-Johnell criterion, or the Ranawat criterion. The use of the combined criteria improved the sensitivity to 94% and the negative predictive value to 91%. If at least one of the 3 is positive, a CT scan or an MRI should be performed. Figures 1–4 depict the various plain radiographic measurements for assessing the upper cervical spine.
Subaxial subluxation (SAS) commonly occurs in RA patients after degeneration of the ligamentous structures, such as the facet joints, the intervertebral disc, and inter-spinous ligaments. Anterior SAS is much more common than posterior SAS. Subaxial subluxation can be an isolated finding involving one or multiple levels, but not uncommonly, it is associated with antlantoaxial subluxation (AAS). White et al. proposed that biomechanical instability for SAS occurs when there is more than 3.5 mm of horizontal displacement of one vertebra in relation to an adjacent vertebra measured on lateral radiographs. However, some authors report that even 2 mm of anterior subluxation increases the risk of cervical spinal cord injury. As proposed by Yurube et al., the diagnosis of SAS should be considered when an irreducible translation of more than 2 mm is documented, and severe SAS occurs when there is more than 4 mm of translation. Some authors report that the clinical outcome of patients with SAS is worse than those with AAS, generally with late neurological deterioration even after surgery.
Similar to the spinal canal at the atlantoaxial level, the subaxial spinal canal sagittal diameter must have at least 14 mm to avoid cord compression. The normal diameter measured on lateral radiographs from C-3 to C-7 is 14–23 mm, and the diameter of the canal is a better predictor of neurological impairment than the degree of subluxation between the vertebrae.
Plain Radiographs
Screening for AAS and Cranial Settling
The classical diagnostic measurements for AAS are based on plain radiographs. Of note, most of these criteria were published before the advent of modern CT and MRI, which can clearly visualize subluxation of the facet joints and all the bony landmarks of the craniocervical junction. However, the craniocervical relationships proposed in the plain radiography era are still used for the initial evaluation, avoiding the cost of routine CT or MRI. Some of these plain radiographic criteria are presented below.
Anterior Atlantodental Interval. The normal value of the anterior atlantodental interval (ADI)—the distance from the posterior border of the anterior tubercle of the atlas to the dens—is less than 3 mm in healthy adults (Fig. 1). As the ADI increases, the chance of spinal cord compression progressively increases. Some authors report that when the anterior ADI exceeds 8 mm, surgery is recommended, as this value suggests total rupture of the transverse and alar ligaments. However, most no longer use the anterior ADI for evaluating patients with RA, as the posterior ADI has been found to be a better predictor of paralysis and recovery.
(Enlarge Image)
Figure 1.
Lateral plain radiograph showing the anterior atlantodental interval (designated in this image by ADI) and the posterior atlantodental interval (designated in this image by PDI) as well as the Ranawat index, the distance from the center of the C-2 pedicle to the transverse axis of C-2.
Posterior Atlantodental Interval. The posterior ADI—the distance from the posterior border of the dens to the anterior aspect of the posterior arch of C-1—evaluates the maximum amount of space available for the upper cervical spinal cord. This has been found to be a better predictor of the neurological risk and recovery in the setting of atlantoaxial subluxation (AAS) than the ADI (Fig. 1). The posterior ADI represents the anteroposterior diameter of the spinal canal at this level. In the cervical spine, the cord itself occupies 10 mm of the canal diameter. In addition, it requires 1 mm for the dura and 1 mm for the CSF anterior to the cord, and the same posteriorly, for a total of 14 mm. Therefore, if the available space is less than 14 mm, the cord becomes compressed.
The posterior ADI and anterior ADI should both be measured on images obtained in flexion and extension. Boden et al. found that patients with a posterior ADI greater than 14 mm had a higher rate of neurological recovery after fusion and stabilization, whereas a posterior ADI less than 10 mm was associated with worse clinical outcome.
Neither the anterior ADI nor the posterior ADI can evaluate cord compression by soft tissues, such as a pannus formation in the retro-odontoid region. For this reason, spinal cord compression can occur even when the plain radiographic measurements are in the normative range.
Lateral Displacement of the Atlas Over the Axis. The open-mouth view is useful for evaluating lateral AAS. Rotatory AAS should be suspected when there is asymmetry or lateral displacement of the atlas on the axis by more than 2 mm in an open-mouth view. It should also be suspected when there is asymmetrical collapse of the lateral atlas mass. Lateral displacement can also occur with fractures of the dens. A CT scan should be performed to confirm the diagnosis.
Cranial Settling. Cranial settling is also known as basilar impression, atlantoaxial impaction, superior migration of the odontoid, and vertical subluxation, and there are numerous plain radiographic criteria for making the diagnosis in RA patients.
The diagnosis of cranial settling based on plain radiographs is sometimes a challenge for radiologists and physicians, as osseous structures of the cranial base are superimposed upon the landmarks, especially in the upper cervical spine. Furthermore, erosion of the dens can make it difficult, if not impossible, to identify its tip. Finally, although the terms "basilar invagination" and "basilar impression" are used synonymously by many authors, the former term may be better used when referring to a congenital craniocervical junction anomaly, whereas basilar impression is more accepted for the description of secondary causes of cranial settling, as occurs in RA.
Below, we list some of the many plain radiographic criteria to diagnose basilar impression that have been described in the literature and their original dates of publication (see also Figs. 1–6).
Chamberlain line (1939): Findings are considered positive if the apex of the odontoid is 3 mm above a line from the posterior edge of the hard palate to the opisthion.
(Enlarge Image)
Figure 2.
Lateral plain radiograph showing the McRae, Chamberlain, and Wackenhelm lines for evaluation of the relationships between the occiput, C-1, and C-2.
(Enlarge Image)
Figure 3.
Lateral plain radiograph showing the McGregor line and the Redlund-Johnell measurement from the McGregor line to the midpoint of the caudal margin of the C-2 body.
(Enlarge Image)
Figure 4.
Lateral plain radiograph showing the Clark stations. The odontoid process is divided into 3 equal parts or stations. The position of the anterior arch of the atlas is assessed relative to these stations.
(Enlarge Image)
Figure 5.
Lateral cervical flexion (left) and extension (right) radiographs obtained in a patient with RA and cervical pain refractory to nonsurgical treatment. Note the increase (in flexion) of the anterior ADI, confirming an atlantoaxial instability.
(Enlarge Image)
Figure 6.
Preoperative and postoperative images obtained in a 53-year-old patient with severe RA. A: Lateral cervical radiograph showing the dens protruding into the foramen magnum, with basilar impression. B and C: CT images obtained in extension (B) showing the tip of the dens 5.49 mm above the McRae line and in flexion (C) showing the dens 9.43 mm above the McRae line. D: Sagittal T2-weighted MR image showing the dens protruding into the medulla and posterior compression of the upper spinal cord by the posterior arch of the atlas. E: Sagittal reconstruction of postoperative CT scan obtained after occipitocervical decompression and craniocervical fixation. F and G: Sagittal and coronal CT reconstructions showing the autologous bone graft (black arrow) used as a spacer between C-1 and C-2 to reduce the protrusion of the tip of the odontoid process into the foramen magnum.
McGregor line (1948): Findings are considered positive if the apex of the odontoid is > 4.5 mm above a line drawn from the posterior hard palate to the most inferior point on the occipital curve.
Fischgold and Metzger line (1952): Findings are positive if the apex of the odontoid is above the line connecting the tips of the mastoid processes bilaterally in an open-mouth view.
McRae line (1953): Findings are positive if the tip of the odontoid extends above a line drawn from the basion (anterior rim of the foramen magnum) to the opisthion (posterior rim of the foramen magnum).
Wackenheim line (1974): Findings are positive if the odontoid protrudes posterior to a line drawn extending from the superior surface of the clivus through the spinal canal.
Ranawat criterion (1979): A line is drawn from the midpoint of the C-2 pedicle along the center of the odontoid process until it intersects a horizontal line through the atlas. Findings are positive if the length is < 15 mm in males or < 13 mm in females.
Redlund-Johnell criterion (1984): A line is drawn from the midpoint of the caudal surface of the C-2 body to the McGregor line. Findings are considered positive if the length is < 34 mm in males or < 29 mm in females.
Clark station (1989): The odontoid process is divided into 3 equal parts ("stations") from craniad to caudad in the sagittal plane. The results are positive if the anterior arch of the atlas is in the second or third station. This method is the simplest one, as the relationship does not change in flexion, extension, or neutral views.
Riew et al. evaluated the reliability and sensitivity of the diagnosis of basilar invagination in 131 cervical radiographs obtained in patients with RA, according to the criteria proposed by Clark et al., McRae and Barnum, Chamberlain, McGreger, Redlund-Johnell and Pettersson, Ranawat et al., Fischgold and Metzger, and Wackenheim. As a final conclusion, no single plain radiographic criteria had sensitivity and a negative predictive value greater than 90% as well as a reasonable specificity and acceptable positive predictive value. Therefore, they suggested that the results of screening for basilar impression should be considered positive when at least one of 3 following criteria are positive: the Clark station, the Redlund-Johnell criterion, or the Ranawat criterion. The use of the combined criteria improved the sensitivity to 94% and the negative predictive value to 91%. If at least one of the 3 is positive, a CT scan or an MRI should be performed. Figures 1–4 depict the various plain radiographic measurements for assessing the upper cervical spine.
Screening for SAS
Subaxial subluxation (SAS) commonly occurs in RA patients after degeneration of the ligamentous structures, such as the facet joints, the intervertebral disc, and inter-spinous ligaments. Anterior SAS is much more common than posterior SAS. Subaxial subluxation can be an isolated finding involving one or multiple levels, but not uncommonly, it is associated with antlantoaxial subluxation (AAS). White et al. proposed that biomechanical instability for SAS occurs when there is more than 3.5 mm of horizontal displacement of one vertebra in relation to an adjacent vertebra measured on lateral radiographs. However, some authors report that even 2 mm of anterior subluxation increases the risk of cervical spinal cord injury. As proposed by Yurube et al., the diagnosis of SAS should be considered when an irreducible translation of more than 2 mm is documented, and severe SAS occurs when there is more than 4 mm of translation. Some authors report that the clinical outcome of patients with SAS is worse than those with AAS, generally with late neurological deterioration even after surgery.
Similar to the spinal canal at the atlantoaxial level, the subaxial spinal canal sagittal diameter must have at least 14 mm to avoid cord compression. The normal diameter measured on lateral radiographs from C-3 to C-7 is 14–23 mm, and the diameter of the canal is a better predictor of neurological impairment than the degree of subluxation between the vertebrae.
Source...