Cone-beam computed tomography does one thing extremely well. It renders the bony anatomy of the upper cervical complex in three dimensions at high resolution: the skull base, the atlas, the axis, the joint surfaces between them. Not the ligaments. Not the cerebrospinal fluid. Not blood flow. Bone, in detail, in space.
That is a narrow contribution and it is also the piece most often missing. The rest of this series on the eight imaging types we use to evaluate mild traumatic brain injury is largely devoted to soft tissue and flow — studies that can tell you a ligament is compromised without telling you what the skeleton underneath is doing.
How Is a Cone-Beam Scan Taken?
Conventional CT builds the volume one layer at a time from thin slices taken with a fan-shaped beam. Cone-beam CT uses a cone-shaped beam that rotates around the head and neck, capturing the whole volume in a single pass and reconstructing it in software — a high-resolution three-dimensional data set of the C0–C2 bony anatomy at a comparatively low dose.
That volume can then be sliced in any plane after the fact — including obliquely, along a particular joint's true plane — without re-scanning the patient. The joints at the top of the neck are not oriented along the body's natural planes, and cutting along the angle a joint sits at is the difference between seeing a surface and inferring one.
What Does It Show at C0–C2?
Three findings are the practical yield.
**Atlantoaxial misalignment.** The atlas-axis articulation is the most mobile in the spine and has no disc constraining it. Displacement there is measurable on a 3D data set in a way it is not on a single projection.
**Condylar asymmetry.** The occipital condyles are the two surfaces the skull rests on. When they differ in shape, depth, or orientation, the head sits on an asymmetric foundation and everything below compensates.
**Small fractures.** Fractures of the dens or the lateral masses of the atlas can be subtle enough to be missed on plain films, and a review of craniocervical junction imaging in blunt trauma describes both the complexity of the region and the difficulty of assessing it after injury (Offiah & Day, 2017).
Why Do Direction and Angle Matter?
There is a difference between knowing something is misaligned and knowing how. "The atlas is off" is an impression. It does not say which way, by how much, or around what axis, and two patients fitting it can need corrections in opposite directions.
A volumetric measurement converts that impression into vectors — a direction and an angle of displacement, expressed numerically. It is what allows a correction to be calculated rather than estimated, and what makes the work checkable: the imaging can be repeated to verify the structure moved as predicted.
I offer that emphasis as a clinical position rather than an established evidence base. What I can say plainly is that measurement makes a procedure auditable and estimation does not. CBCT is also not the only route to it — standard craniocervical X-rays taken in three dimensions serve a similar purpose and remain widely used.
How Much Radiation Is Involved?
Here the evidence needs care, because the source closest to my own field is the weakest in this article. A 2022 literature review in *Dose-Response* examined craniocervical junction visualization and radiation dose for upper cervical chiropractic application. Its figures: an ultra-low-dose large skull protocol on one manufacturer's unit at "41 μSv or .041 mSv"; a general range for CBCT of the head and cervical spine of "50-250 μSv (.05–.25 mSv)"; a conventional upper cervical chiropractic five-image radiographic series at "200 μSv or .20 mSv"; and, for comparison, US annual background radiation of "3100 μSv (3.1 mSv)" (DeNunzio et al., 2022).
The caveats are not optional. That paper is a literature review, not original dosimetry. Its authors work within, and write for, chiropractic application — I share that alignment, and you should weigh it accordingly. They also concede the literature on CBCT at the craniocervical junction is limited and that they extrapolated heavily from dental studies.
The independent check is a systematic review of radiation dose in non-dental CBCT applications, which concluded that "CBCT of extremities, cervical spine, ears and paranasal sinuses was found to be a low-dose volumetric imaging technique" (Nardi et al., 2018). That is genuine support. But the same conclusion carries a caution worth reproducing rather than paraphrasing: "Effective doses varied significantly because of different exposure settings of CBCT-units and different dosimetry systems used to estimate dose." Only fifteen studies met inclusion criteria, and only one examined the cervical spine.
So: CBCT of the cervical spine appears to be low-dose, and dose is machine- and protocol-dependent enough that a figure from one unit does not transfer to another. There is also no verified head-to-head dose comparison between cone-beam CT and conventional CT specific to C0–C2. Any such number is extrapolated rather than measured, so I will not give you one.
What Can Cone-Beam CT Not Do?
It images bone. That is the whole boundary, and a hard one. The alar ligament, transverse ligament, and tectorial membrane — the structures doing most of the stabilizing work at that joint — are invisible to it, and [assessing them requires a dedicated MRI protocol](post-what-an-mri-of-the-craniocervical-junction-shows.html). The cord, muscles, and vessels are not shown in useful detail either, and no flow is measured.
More importantly, a single static acquisition does not by itself establish instability. Instability is a property of motion under load — how far a joint travels when the head turns or flexes, and whether the structures meant to stop it do. A still image of well-positioned bone, taken while the patient lies motionless, carries no load and reports no travel. Work distinguishing structural from functional instability at the craniocervical junction draws exactly this line (Godek & Ruciński, 2024).
So [craniocervical instability](condition-craniocervical-instability.html) is not a diagnosis a bone scan hands you. CBCT shows the substrate; the demonstration comes from motion studies and examination.
I regard CBCT as the gold standard for bone detail at C0–C2, and I offer that as my characterization from inside this field rather than external consensus. It is a claim about one tissue type in one region, not a claim that this is the most important study in the workup.
Have You Been Told the Bones Look Fine?
If your [neck pain](condition-neck-pain.html) or post-injury symptoms were evaluated on a single projection film and nothing was found, that is worth knowing — but a two-dimensional image of a three-dimensional asymmetry is an incomplete question.
Imaging the upper cervical complex well enough to calculate a correction, then re-imaging to confirm the structure responded, is the work we do at Cerebral. If you'd like a real evaluation, we're here.
References
- DeNunzio G, Evans T, Beebe ME, et al. Craniocervical junction visualization and radiation dose consideration utilizing cone beam computed tomography for upper cervical chiropractic clinical application: a literature review. *Dose-Response*. 2022;20(2):15593258221107515. https://pubmed.ncbi.nlm.nih.gov/35719850/
- Nardi C, Salerno S, Molteni R, et al. Radiation dose in non-dental cone beam CT applications: a systematic review. *La Radiologia Medica*. 2018;123(10):765–777. https://pubmed.ncbi.nlm.nih.gov/29869227/
- Offiah CE, Day E. The craniocervical junction: embryology, anatomy, biomechanics and imaging in blunt trauma. *Insights into Imaging*. 2017;8(1):29–47. https://pubmed.ncbi.nlm.nih.gov/27815845/
- Godek P, Ruciński W. Differentiating the structural and functional instability of the craniocervical junction. *Healthcare (Basel)*. 2024;12(19):2003. https://pubmed.ncbi.nlm.nih.gov/39408183/