Standard MRI is performed supine — lying flat, inside a closed bore, with the neck fully supported and gravity essentially removed from the equation. It is superb at showing anatomy: soft tissue detail, cord signal, tonsillar position, disc pathology. What it is structurally poor at showing is function — what happens to that same anatomy when the patient sits up, bears the weight of their own head, and moves their neck the way they do hundreds of times a day. For craniocervical instability, where the defining problem is excess or abnormal motion, that gap is not a minor limitation. It is often the reason a genuinely unstable joint gets photographed and still called normal.
reported upright vs. supine
flexion and extension
angle for an adequate study
Gravity Is Part of the Anatomy
When a patient lies down for a standard MRI, two things change simultaneously: the weight of the head is transferred to the table instead of the neck, and cerebrospinal fluid pressure gradients shift with posture. Both changes can genuinely alter measurements that were the entire reason the scan was ordered. A clivo-axial angle, a tonsillar tip position, or a joint space measured with the neck fully unloaded is not necessarily the same number that same structure would show while the patient is upright, symptomatic, and using their neck normally. Multiple published series comparing supine and upright imaging in the same patients have found measurable differences in tonsillar position, cervical alignment, and craniocervical angles between the two positions — differences large enough, in some patients, to move a finding from "normal" to "abnormal" depending purely on posture.
None of this means supine MRI is a poor test; it remains the correct first study for the overwhelming majority of craniovertebral junction complaints, and it is what most of the measurements earlier in this series were validated against. The issue is narrower and more specific: in patients whose symptoms are clearly positional or motion-provoked, and whose supine imaging looks unremarkable despite a strong clinical picture, a normal static scan does not settle the question. It simply means the wrong moment was photographed.
Upright / Weight-Bearing MRI
What upright MRI most reliably changes on the report
The best-documented use of upright MRI in this series' context is re-assessing cerebellar tonsil position. The standard threshold for Chiari malformation Type I — tonsillar herniation of 5 mm or more below McRae's line (the foramen magnum line) — was established using supine imaging. Several comparative studies have found that tonsillar position can descend further when the same patient is scanned upright, occasionally revealing what is sometimes described as positional or occult Chiari: a supine scan below the 5 mm threshold, with the same patient measuring at or above it once upright and gravity-loaded. Clivo-axial angle and overall cervical curvature can also shift between supine and upright positioning, which is why upright studies are frequently paired with flexion and extension views in the same session rather than a single static upright image alone.
Flexion-Extension Imaging
Because flexion-extension views depend entirely on the patient actively moving their own neck to its true end range, pain, fear of pain, muscle guarding, or simple unfamiliarity with the instructions can produce a "flexion-extension" study that never actually reached meaningful flexion or extension. A technically normal-looking dynamic study performed with inadequate motion is not reassuring — it is an inadequate study that happens to look normal, and the distinction matters enormously when deciding whether instability has genuinely been excluded.
Digital Motion X-ray (DMX)
Fig. 1 — The same craniovertebral junction viewed in four positions. Supine imaging (left) provides the best anatomic detail but removes gravity load. Upright, flexion, and extension views each stress the junction differently, and abnormalities that are position-dependent may only appear in one or two of the four.
Chiari Malformation: A Diagnosis That Can Move
The standard radiological threshold for Chiari malformation Type I — cerebellar tonsils extending 5 mm or more below McRae's line — was defined using supine imaging, and it remains the correct baseline reference. What upright and positional imaging has added to the picture is the observation that tonsillar position is not perfectly fixed: it can shift with posture, with CSF pressure dynamics, and with cervical alignment. A patient whose supine MRI shows tonsils sitting just under that 5 mm line, but who has a clinical picture strongly suggestive of Chiari-type symptoms, is a reasonable candidate for upright imaging specifically to see whether gravity loading changes that number meaningfully. This does not mean every headache patient needs an upright scan — it means the tool exists for the specific, narrower situation where supine imaging and the clinical picture disagree.
The idea that tonsillar descent can be posture-dependent, and that a subset of symptomatic patients have a form of Chiari that is occult on supine imaging alone, is supported by a growing but still limited body of literature, and it is not yet uniformly accepted across all neurosurgical practice. Patients considering upright imaging for this reason should understand that findings from it are typically interpreted alongside — not instead of — standard supine MRI and a thorough clinical exam, and that management decisions based on upright-only findings warrant a specialist experienced specifically in this area.
Tethered Cord: A Different Kind of Motion Problem
Tethered cord syndrome, covered in depth in Article 4 of this series, is not typically reassessed with upright or dynamic imaging in the same way the craniocervical junction is. The conus medullaris and filum terminale are evaluated on standard supine MRI, since the core question — is the cord abnormally low-lying and fixed rather than free to move within the canal — does not require gravity loading or motion capture to answer. What dynamic and upright imaging contributes to tethered cord care is indirect but clinically important: because a tethered cord restricts normal caudal mobility of the spinal cord during flexion, and craniocervical instability allows excess motion at the opposite end of the same neuraxis, some patients carry both conditions simultaneously — a combination sometimes described as a tension mismatch along the entire spinal cord. In these patients, flexion-extension and upright imaging of the craniocervical junction is used specifically to characterise the cervical component of a problem that will likely need coordinated, rather than isolated, surgical planning between the two ends.
A scan taken lying flat and motionless answers the question "what does this look like at rest." Many of the patients this imaging is built for have never once described their symptoms as happening at rest.
The Practical Protocol
Clinical justification first: Because access is limited and cost or insurance coverage is often a genuine barrier, most centres require a documented, strongly positional or motion-provoked symptom pattern, and often a supine study that has already been reviewed, before ordering upright or dynamic imaging.
Neutral baseline: A neutral position image is obtained first — upright neutral for open MRI, or a resting lateral view for flexion-extension films — to serve as the reference point for comparison.
Active, patient-driven motion: For flexion-extension views, the patient is coached to actively flex and extend the neck to their own genuine maximum, under supervision, rather than being passively positioned by the technologist. Adequate effort — commonly at least 30 degrees of motion in each direction — is essential; the study is only interpretable if true end-range was reached.
Multiple positions, one session where possible: Upright MRI protocols frequently capture neutral, flexion, and extension positions within the same appointment, allowing direct comparison of tonsillar position and craniocervical angles across all three without needing separate visits.
DMX as a targeted add-on: When ordered, DMX is typically performed as a focused continuous-motion study of the specific segment in question — most often the occiput-C1-C2 complex — rather than the entire cervical spine, to manage radiation exposure.
Where Dynamic & Upright Imaging Goes Wrong
Dynamic & Upright Imaging — Complete Reference
| Modality | What it adds | Radiation | Access | Best used for | Status |
|---|---|---|---|---|---|
| Supine MRI | Baseline anatomy, best resolution | None | Wide | First-line study, nearly all patients | Standard of care |
| Upright MRI | Gravity-loaded position, tonsil/alignment shift | None | Limited | Suspected positional Chiari, posture-linked symptoms | Lower resolution trade-off |
| Flexion-Extension | End-range ADI, BAI, CXA, Grabb-Oakes comparison | Low–moderate | Moderate | Suspected ligamentous instability, positional exam findings | Well-validated |
| Digital Motion X-ray | Continuous motion arc, transient subluxation | Higher | Very limited | Intermittent instability missed by static endpoints | Emerging evidence |
When a Static Scan Is Not Enough
No single symptom proves that dynamic or upright imaging is needed, but a consistent pattern across several is a reasonable prompt to raise the question with a treating clinician.
Symptoms clearly worse upright than lying flat, or vice versa: Headaches, pressure, or dizziness that reliably improve with lying down and worsen with prolonged sitting or standing point toward a gravity-dependent mechanism that static supine imaging, by definition, cannot capture.
Symptoms triggered by specific neck positions: Consistent provocation with flexion (looking down), extension (looking up), or rotation is the clinical signature that flexion-extension or dynamic imaging is specifically designed to investigate.
A strong clinical picture with unremarkable supine imaging: When exam findings, symptom pattern, and risk factors (such as hEDS) are all consistent with craniocervical instability but the standard MRI report is read as normal, this mismatch is the classic indication for adding dynamic or upright studies rather than concluding the workup is complete.
Transient or episodic neurological symptoms: Brief drop attacks, momentary vision changes, or fleeting weakness that resolve completely between episodes can reflect intermittent, motion-triggered compression that only continuous or end-range imaging is positioned to catch.
What to Ask For — Practical Steps
Start with standard imaging and a thorough exam. Dynamic and upright studies are additive, second-line tools for a specific clinical mismatch — not a shortcut around a proper baseline work-up, and most centres will expect that baseline before considering the more specialised studies.
Describe your actual triggers in detail. "Worse when I look down," "worse standing for more than ten minutes," or "comes and goes with certain sleep positions" gives the ordering clinician and imaging centre something concrete to try to reproduce, rather than a generic request for "more imaging."
Ask what position was actually achieved. For flexion-extension studies in particular, ask whether the report documents the actual degrees of flexion and extension reached — a study performed with minimal true motion is not equivalent to one performed with genuine end-range effort, even if both are labelled "flexion-extension views."
Treat upright and DMX findings as part of a bigger picture. Given the still-developing standardisation of these newer modalities, findings from them are best reviewed by a specialist experienced specifically in craniocervical instability, alongside standard imaging and exam findings — not as a stand-alone verdict.
Chiari, Tethered Cord
& EDS: The Overlap
That Ties It Together.
Article 7 closes the series by bringing every measurement together — how Chiari malformation, tethered cord syndrome, and Ehlers-Danlos syndrome intersect with craniocervical instability, and what that overlap means for diagnosis and treatment planning.
Every threshold in this series was measured off a still image, because still images are what medicine has mostly had. Upright and dynamic imaging do not replace that body of work — they extend it, into the position and motion the patient actually lives in. The scan that matters most is not always the one taken lying still in a quiet room. Sometimes it's the one taken while the joint is finally asked to do the thing that hurts.