Consensus Statement on Craniocervical Instability (CCI)


Senior Member
Just want to post the full text of the 2013 Consensus Statement on Craniocervical Instability, for easy reference.

This consensus statement is the basis of current practice in detecting CCI via MRI or CT scan measurements.

As Dr Paolo Bolognese explains in this video at 29:12, there are more than 20 measurements that cover the craniocervical junction. For years nobody was quite sure which were the best ones to use to detect craniocervical instability, so many of these 20 were employed. Later these 20 measurements got reduced to 14 measurements.

Finally, in the CSF Craniocervical Instability Colloquium, San Francisco, October 2013, a consensus statement was arrived at, in which the experts agreed that the primary 4 measurements used for diagnosing CCI should be:
  • Clivo-axial angle (CXA) — pathological if 135º or less
  • Grabb-Oakes measurement — pathological if 9 mm or more
  • Basion-axial interval (BAI) — pathological if 12 mm or more
  • Translational BAI — pathological if greater than 2 mm
More info on these reference ranges for pathology can be found in this post.

Here is the full Consensus Statement on CCI:
The Consensus Statement

Proceedings of the CSF Colloquium 2013: Basilar Impression & Hypermobility at the Craniocervical Junction. San Francisco, October 2013.
(1) Ventral brainstem compression, medullary kinking and deformation of the upper spinal cord and/or brainstem over the odontoid process are potentially deleterious to the brainstem and upper spinal cord.

(2) Deformation of the brainstem may manifest clinically as the cervical medullary syndrome.

(3) The clinical findings of cervical medullary syndrome may include, but are not limited to, the following:

(i) headaches, suboccipital pain and neck pain,​
(ii) Bulbar and related symptoms: altered vision, diplopia, nystagmus, decreased hearing, tinnitus, imbalance, vertigo, dizziness, choking, dysarthria, dysphagia dysautonomia, postural orthostatic tachycardia, pre-syncopal or syncopal episodes disordered sleep architecture, sleep apnea,​
(iii) Symptoms of myelopathy: weakness, clumsiness, spasticity, altered sensation, paresthesias, dysesthesia, change in gait, constipation, urinary urgency and frequency.​

(4) In assessing the potential for craniocervical instability, it is reasonable to measure the angle between the clivus and the spine. This angle has been termed the clivus canal angle, the clivus vertebral angle, the clivus spinal angle, the clivus cervical angle and the clivus-axial angle.

In keeping with the greater part of the literature, we recommend the uniform adoption of the term clivo-axial angle. This angle may be abbreviated CXA.

(5) The clivo-axial angle is the angle between the clivus line and the posterior axial line. The clivus line is drawn along the lower third of the clivus -from the spheno-occipital synchondrosis to the basion, or in the case of basilar invagination, the superior most aspect of the odontoid .

When assessing the CXA with sagittal CT scan or X-ray, the posterior axial line may be drawn along the posterior edge of the odontoid.

When assessing the CXA with MRI, the posterior axial line should be drawn from the posterior edge of the tectorial membrane to the inferior posterior edge of the posterior ligament of the C2 vertebra.

The CT and MRI measurements may differ in the same patient: the CXA determined by CT reflects the more traditional means of measurement; the CXA determined by MRI will necessarily include thickening of the ligament due to pannus.

(6) The literature suggests that a clivo-axial angle of 135 degrees or less is potentially pathological. That is a CXA of 135 degree, may in some circumstances, result in harmful deformative stress upon the brainstem and upper spinal cord and, therefore, warrants consideration for further evaluation and possible treatment.

(7) The CXA can be measured on sagittal CT or MRI, with the patient assuming moderate flexion of the craniocervical junction. If a flexion view is not available, a neutral position will suffice in most circumstances. An upright dynamic MRI may be desirable in some circumstances – but, such is often not available.

(8) In assessing the potential for craniocervical instability, it is reasonable and appropriate to measure the BpC2 line, also known as the Grabb-Oakes measurement or line, or the Grabb Mapstone Oakes Measurement, as one method to approximate the potential presence and magnitude of ventral brainstem compression.

We use the term Grabb-Oakes measurement herein. The Grabb-Oakes measurement is the distance in millimeters from the dura to the line drawn from the basion to the posterior inferior edge of the C2 vertebra. A Grabb Oakes measurement of 9 mm represents the diagnostic threshold for ventral brainstem deformity. Some clinicians may choose 8 mm as the diagnostic threshold at which there may is potential ventral brainstem deformity.

(9) The Harris measurement, also known as the basion axial interval (BAI = distance from tip of basion to posterior axial line), when drawn horizontally, should be less than 12 mm.

The basion to dens interval (BDI = distance from basion to tip of odontoid) drawn vertically, should be less than 12 mm. The posterior axial line should be drawn along the posterior ligamentous surface of the C2 vertebra. In keeping with the literature, a Harris measurement exceeding 12 mm is considered potentially pathological, and reflects craniocervical instability.

(10) In the presence of known ligamentous instability, such as a hereditary hypermobility connective tissue disorder, the BAI (the Harris measurement) may be measured with the cervical spine in the flexion and extension positions. This will assess and quantify translation of the basion with respect to the dens (odontoid process). In keeping with the literature, any translation [translational BAI] noted on dynamic imaging that exceeds 2 mm (the delta BAI 2 mm), will be considered abnormal and potentially pathological.

(11) Craniocervical hypermobility is common, and defined by the presence of hyper-extensibility of the connective tissue, and in particular, hyper extensibility of the joints. While hypermobile joints occur frequently in healthy children, such can also be severely disabling in others. Ehlers-Danlos syndrome, cleidocranial dysostosis, Down syndrome, Marfan syndrome, Morquio syndrome and several other less well known connective tissue disorders are associated with ligamentous laxity. A pathological Lax Ligament Syndrome may result in craniovertebral instability, kyphosis of the clivo-axial angle and ventral brainstem compression.

The growing body of knowledge regarding the prevalence of hypermobility connective tissue disorders should lead to more widespread recognition of the impact of ligamentous laxity on the health of sufferers of hypermobility syndromes.

Last edited:


Senior Member
in this 2019 article, they talk about ADI (atlanto-dens interval), and they say that X ray is better to determine bones measures:

"The atlantodens interval (ADI) is an additional method of determining atlantoaxial instability. The ADI is determined by measuring the distance between the posterior edge of the anterior arch of C1 and the anterior edge of the dens.
A value less than three mm is considered normal in adults and a value less than 4.5 mm is considered normal in children. Numbers greater than the maximal normal values listed above suggest disruption of the transverse ligament. There are numerous proposed measurements for the assessment of atlantoaxial instability specific to patients with RA."

(In another article found ADI less that 2.5 mm is normal in adulte female)

They also say:


In individuals without any predisposing factors, atlantoaxial instability is extremely rare.
Radiographic atlantoaxial instability is seen in up to 30% of patients with Down syndrome (DS), but only 1% of patients with DS have symptomatic atlantoaxial instability.
Patients with rheumatoid arthritis (RA) are also susceptible to cervical spinal instability and atlantoaxial instability. Ranges of 25% to 80% have been seen, and newer disease-modifying RA drugs may change the course of the disease and decrease the incidence of atlantoaxial instability.
Atlantoaxial instability can occur equally in both genders and can present at any age.
A high risk of atlantoaxial instability is seen in Down syndrome and older patients with rheumatoid arthritis"


Senior Member
Clivo-axial angle is used by Henderson (normal 150-180°), but he doesn't seem to use the strict 135° pathological limit….
"The preoperative mean CXA was 135.8° in the neutral view (range 131°–140°). While the pathological threshold for the CXA in this report was considered 135°, one subject who had been followed for 9 months was admitted to the study with a CXA of 140°, because of the progressive pain, the compelling neurological findings, and the response to the neck brace."


Fig. 1
a Normal craniocervical junction in the neutral position. The CXA varies from 150° to 165°. There is minimal or zero deformative strain in the neutral state.
b Normal craniocervical junction in flexion. The neuraxis stretches by approximately 10% of its total length with flexion of the craniocervical junction creating a strain ε = 0.1.
c Pathological craniocervical junction with an abnormal CXA in flexion. Upon full flexion at the craniocervical junction, the increase in the tangent arc creates a deformative strain approaching ε = 0.2 (i.e., 20% stretch). In vivo and in vitro models demonstrate decreased or loss of neurological function with strains of 0.2
Last edited: