Suspected Type I Rotatory Atlatoaxial Subluxation in Asymptomatic Patients

In an earlier post on rotatory atlatoaxial subluxation, we discussed the Fielding and Hawkins classification, and its application to symptomatic patients (e.g., those with torticollis). Here, we discuss the challenge of making a diagnosis of rotatory atlatoaxial subluxation in unselected patients.

That is, what do you do if your neck or c-spine CT scan is obtained with the head turned, and you see what looks like a Type I rotatory atlatoaxial subluxation?

The patient below was referred to our institution in a cervical collar for management of atlantoaxial rotatory subluxation.

30-year-old man: Thick reconstructions centered at C1-C2 from a CTA obtained with the head rotated to the left. There is apparent type I atlantoaxial rotatory subluxation.
 
Let's take a look at what we know:

• Normal maximum rotation of the head on body is between 60–80°.
• Normal maximum rotation of C1 on C2 makes up for about half of that: 30–45° (although some sources say it can be up to 50°).
• Cutoff for abnormal C1-C2 rotation: >45° or >56°, depending on source.

Looking at our image above, our rotation of C1 on C2 is about 30 degrees, which puts it in the normal range. But, what if the number was higher? How well do we do as radiologists in correctly identifying this as normal?

One study that was helpful came from the forensic field. Pathologists had noted that CTs of cadavers, which are impossible to position "correctly," were resulting in a lot of over-calling of atlantoaxial rotatory subluxation. When they looked at their data they found:

• 19 cases where the C-spine was stable on autopsy. In those 19, 13 had suspected rotatory subluxation based on on CT (false positives)
• The C1-C2 angle in those 13 false positives were between 16–47°.
• All false positives were type I.
• 2 cases where the C-spine was unstable at autopsy. In those 2 cases, 1 had suspected rotatory subluxation on CT (true positive).
• The C1-C2 angle in the 1 true positive was 42°.
• There was a significant association between the false positives and degree of head rotation.

Generalizing this cadaveric study suggests that we need to be careful in calling type I atlantoaxial rotatory subluxation in asymptomatic patients who simply happen to have their head turned in the scanner. This supports the conclusions from a study in living, asymptomatic patients, where the authors showed that incomplete rotational facet displacement on CT was not sufficient to define subluxation.

References

• Fielding JW, Hawkins RJ. Atlanto-axial rotatory fixation. (Fixed rotatory subluxation of the atlanto-axial joint). J Bone Joint Surg Am. 1977 Jan;59(1):37-44.
• Persson A, Falk J, Berge J, Jackowski C. Atlanto-axial rotatory subluxations in postmortem CT: radiologists be aware of a common pitfall. Forensic Sci Int. 2013 Feb 10;225(1-3):9-14.
• Mönckeberg JE1, Tomé CV, Matías A, Alonso A, Vásquez J, Zubieta JL. CT scan study of atlantoaxial rotatory mobility in asymptomatic adult subjects: a basis for better understanding C1-C2 rotatory fixation and subluxation. Spine (Phila Pa 1976). 2009 May 20;34(12):1292-5.

Endologix Stent-graft

Arterial phase CT shows smoothly marginated contrast extending beyond the confines of the metal struts of the stent

Delayed CT shows the smoothly marginated contrast extending beyond the confines of the metal strut unchanged in size and configuration when compared to arterial phase imaging

Most endostent devices used for endovascular repair of aortic aneurysms have the graft material sutured on the luminal side of the stent. The AFX endostent device (Endologix, Irvine, California) is different from other devices in that the graft material is sutured on the outside of the metal component, attached only at the proximal and distal ends.

By intent, the graft material of the AFX stent-graft can separate from the metal struts where it is not directly apposed to the aortic wall.  During surveillance with contrast-enhanced imaging of patients who have reviewed this device, contrast can be detected outside of the metallic construct. Although this finding can imply a type 1 or 3 endoleak when identified with other endostents, it is a normal finding with the AFX device.

Avoiding misinterpretation as an endoleak can be achieved by confirmation of the type of endostent device used for aneurysm repair. In addition, the contrast extending beyond the metal struts does not change in configuration or size on delayed imaging.

References:

J Vasc Interv Radiol. 2012 Nov;23(11):1544-6.