Marc Darrow, MD, JD
An MRI report may describe a bulging disc, arthritis, or narrowing around a nerve. The next question is whether that finding explains the pain, numbness, or weakness that brought you to the office.
Cervical MRI is an important examination for suspected nerve compression in the neck. Recent research also shows why the images, the report, and the clinical examination need to be considered together. Some narrowing is underestimated in reports. Certain compression sites are easier to see with specialized sequences. And even a clearly visible abnormality does not establish that it is responsible for every symptom.
What cervical MRI can tell us
MRI provides detailed views of the discs, spinal cord, spinal canal, and spaces around the nerve roots. The American College of Radiology’s updated guidance, published in 2025, considers MRI without contrast usually appropriate for chronic cervical pain with radiculopathy in adults without trauma or red flags. Radiculopathy means a problem involving a spinal nerve root, often producing radiating arm pain, altered sensation, weakness, or reflex changes. Imaging decisions differ for uncomplicated neck pain, trauma, suspected infection, and other circumstances. (1)
MRI answers a structural question. It does not directly measure electrical nerve function or identify every source of neck pain. In my clinical approach, a useful scan finding should fit the patient’s symptom location and neurological examination before it drives treatment.
Is there a single percentage for cervical MRI accuracy?
No single percentage adequately describes how well MRI diagnoses all cervical nerve problems. Studies may measure detection of narrowing, agreement between readers, agreement with the physical examination, or confirmation of compression at surgery. These are different outcomes.
Sensitivity measures how often a test detects a condition when it is present. Specificity measures how often it is negative when the condition is absent. Neither can be interpreted properly without knowing which condition was studied, which patients were included, and how the diagnosis was confirmed.
A study in patients already selected for surgery, for example, cannot automatically tell us how well MRI performs among everyone with new neck pain.
A 2024 study found that reports sometimes understated foraminal narrowing
The neural foramina are the openings through which spinal nerves leave the spine. Narrowing of these openings is called foraminal stenosis.
A retrospective study examined 283 patients undergoing one- or two-level cervical fusion for radiculopathy. Researchers compared radiology reports with a standardized MRI grading system across 998 spinal levels. Among levels graded as moderately narrowed by that system, 28.9% were described as having no stenosis in the reports. Among severely narrowed levels, 29.7% were reported as mild-to-moderate narrowing or less. Narrowing severity also did not correlate with baseline patient-reported symptom measures. (2)
This was a comparison of interpretations of MRI, rather than a demonstration that the scanner failed to capture the anatomy. Its surgical population also limits generalization. For a patient whose symptoms and report disagree, reviewing the actual images with the treating clinician may be more informative than relying on a brief report alone.
The findings do not mean that more severe narrowing necessarily requires surgery or explains more severe pain.
A 2025 study examined compression missed on routine images
Standard MRI usually acquires two-dimensional image slices. A nerve travels through three-dimensional anatomy, and the angled course of cervical exit openings can make some compression sites difficult to assess.
A 2025 study of 80 patients with cervical spondylotic radiculopathy compared conventional 2D MRI, contrast-enhanced 3D T2-SPACE imaging, and surgical findings. Conventional imaging correctly localized compression in 69 patients; the specialized sequence clarified the remaining 11 cases. Seven of those involved the foraminal region and four involved the lateral region. (3)
These results support considering a more targeted protocol when an important clinical question remains unresolved. They do not establish that every cervical MRI should use contrast or that specialized 3D imaging is necessary for routine neck pain. The study involved selected patients with established radiculopathy and surgical correlation.
What a broader review says about 3D MRI
A 2024 systematic review assessed 30 studies of 3D MRI for spinal narrowing. Overall, 3D techniques provided similar or better detection than conventional 2D imaging, but protocols and results varied. Importantly, the review included different spinal regions. Its reported diagnostic ranges should not be presented as a cervical-only accuracy estimate. (4)
Three-dimensional acquisitions can allow reconstruction along an angled opening or nerve pathway. That can help answer a specific anatomical question. It does not replace clinical judgment about whether the finding matters for the individual patient.
Why symptoms and MRI may point to different levels
In a prospective study of 83 patients with neck-related arm pain, clinical evaluation and MRI agreed on the same side and nerve-root level in 31%. An additional 28% had agreement at an adjacent level. The clinical examiners themselves showed only fair agreement about the affected root. (5)
It would be incorrect to describe this as “MRI is only 31% accurate.” The study measured agreement between two approaches; it did not establish that either was an independent, error-free reference standard.
The practical lesson is that pain patterns and neurological findings can overlap. When several levels look abnormal, the largest disc bulge may not automatically be the treatment target. A careful reassessment can also identify a peripheral nerve problem, shoulder disorder, or another explanation requiring a different approach.
Can a routine MRI miss a position-dependent problem?
A conventional cervical MRI usually examines the neck while the patient is lying down in a limited position. It may not reproduce the posture that provokes symptoms.
A 2025 report described two patients in whom flexion-and-extension MRI revealed findings not adequately explained by standard static imaging. This illustrates a possible role for dynamic imaging in selected unresolved cases, but two cases cannot establish its general sensitivity, specificity, or effect on outcomes. (6)
Persistent symptoms alone do not establish a need for dynamic MRI. A specialist should decide whether another position, sequence, or imaging method is likely to change management and whether positioning is appropriate for that patient.
How ultrasound compares with cervical MRI
Ultrasound can examine accessible nerve segments and nearby soft tissues in real time. Bone restricts its view into the spinal canal and neural foramina. Consequently, a normal external nerve ultrasound cannot exclude a disc-related root compression or spinal cord disorder.
A 2017 pilot study compared 102 patients with cervical radiculopathy and 219 healthy volunteers. Selected nerve-root size measurements produced promising diagnostic performance. Because the comparison involved defined cases and healthy controls, those findings do not demonstrate equivalent performance in an unselected clinic population with competing causes of arm symptoms. (7)
A 2023 study of 51 patients with several degenerative cervical conditions and 29 healthy volunteers found no significant correlation between ultrasound nerve-root size and MRI foraminal diameter or measured muscle-response amplitude. The examinations provided complementary information. (8)
| Clinical question | Examination that may help |
|---|---|
| Is a disc, exit opening, or spinal canal compressing neural structures? | Cervical MRI, with protocol tailored when necessary |
| Is an accessible peripheral nerve enlarged or affected by nearby tissue? | Targeted high-resolution ultrasound |
| Is there evidence of motor nerve-root dysfunction or another nerve disorder? | EMG and nerve-conduction studies |
| Does a specific posture reveal additional compression? | Selected specialist-directed positional imaging |
Ultrasound guidance for an injection is a separate application. Accurate needle placement does not prove that the target causes the symptoms or establish that a particular injection will help.
When EMG and nerve-conduction studies add information
Electromyography examines electrical activity in muscles. Nerve-conduction studies assess responses in tested peripheral nerves. They can help distinguish a cervical root disorder from a problem such as nerve entrapment at the wrist or elbow.
A 2025 review described needle EMG sensitivity of approximately 50–71% and specificity approaching 100% for cervical radiculopathy. A properly interpreted positive result can therefore be useful, while a normal study does not exclude every root problem. These figures concern EMG and should not be used as MRI accuracy figures. (9)
The decision to order electrodiagnostic testing should follow an unanswered clinical question. It may be particularly useful when MRI shows several potential abnormalities or when symptoms suggest a peripheral nerve disorder.
What newer examination research adds
A 2026 systematic review evaluated physical tests used for painful cervical radiculopathy. Certain maneuvers and combinations of tests could assist clinical reasoning, but the authors emphasized sparse evidence and low or very low certainty. No single maneuver provides a dependable answer in every patient. (10)
For patients, this means that the examination remains important even when MRI is available. The location of numbness, changes in strength and reflexes, and the circumstances that provoke symptoms help determine what additional testing is useful.
What to do when the MRI does not explain the symptoms
I would first revisit the diagnosis and compare the images with the examination. Useful questions include whether the suspected level and side fit, whether the report and actual images tell the same story, and whether the symptoms might arise outside the cervical spine.
Depending on the findings, the next step may involve targeted rehabilitation, image review, EMG and nerve-conduction studies, peripheral nerve ultrasound, or a more specific imaging protocol. Not every patient needs every test.
An MRI abnormality alone also does not establish candidacy for PRP, bone marrow concentrate, an epidural injection, or surgery. Treatment requires its own assessment of the diagnosis, expected benefit, alternatives, and risks.
Progressive weakness, new difficulty walking, loss of hand coordination, or new bladder or bowel dysfunction with neurological symptoms warrants prompt assessment. Those findings can change the urgency of evaluating the spinal cord.
Cervical MRI remains a valuable structural examination. Its greatest clinical value comes from asking a precise question, selecting the right technique, and matching the findings to the person being examined.
References
- Expert Panel on Neurological Imaging; Eldaya RW, Parsons MS, et al. ACR Appropriateness Criteria® Cervical Pain or Cervical Radiculopathy: 2024 Update. Journal of the American College of Radiology. 2025;22(5S):S136–S162. doi:10.1016/j.jacr.2025.02.035.
- Lee Y, Issa TZ, Mazmudar AS, et al. Radiology Reports Do Not Accurately Portray the Severity of Cervical Neural Foraminal Stenosis. Clinical Spine Surgery. 2024;37(8):351–356. doi:10.1097/BSD.0000000000001603.
- Wang Y, Chen X, Li XY, Wang D, Rong D, Sun Z, Lu J, Lu S. Magnetic resonance imaging in the evaluation of different compression locations in patients with cervical spondylotic radiculopathy. Journal of Orthopaedic Surgery and Research. 2025;20(1):611. doi:10.1186/s13018-025-05864-3.
- Nevalainen MT, Vähä J, Räsänen L, Bode MK. Diagnostic utility of 3D MRI sequences in the assessment of central, recess and foraminal stenoses of the spine: a systematic review. Skeletal Radiology. 2024;53(12):2575–2584. doi:10.1007/s00256-024-04689-1.
- Redebrandt HN, Brandt C, Hawran S, Bendix T. Clinical evaluation versus magnetic resonance imaging findings in patients with radicular arm pain—A pragmatic study. Health Science Reports. 2022;5(3):e589. doi:10.1002/hsr2.589.
- Gupta B, Kramer D, Barrett CC, Xu C, Yu A. The Value of Dynamic MRI in Cervical Radiculopathy: A Report of Two Cases. Cureus. 2025;17(5):e83321. doi:10.7759/cureus.83321.
- Takeuchi M, Wakao N, Hirasawa A, Murotani K, Kamiya M, Osuka K, Takayasu M. Ultrasonography has a diagnostic value in the assessment of cervical radiculopathy: A prospective pilot study. European Radiology. 2017;27(8):3467–3473. doi:10.1007/s00330-016-4704-9.
- Kozuki H, Tadokoro N, Aoyama N, Kiyasu K, Takemasa R, Ikeuchi M. Additional benefit of ultrasonography to evaluate nerve root condition of degenerative cervical spine disease. Spinal Cord. 2023;61(1):69–75. doi:10.1038/s41393-022-00865-z.
- Gonçalves LI, Oliveira Junior PH, Baima JPS. Cervical radiculopathy for neurologists: the role of electrodiagnosis. Arquivos de Neuro-Psiquiatria. 2025;83(10):s00451812893. doi:10.1055/s-0045-1812893.
- Thoomes EJ, Arvanitidis M, van Geest S, et al. Diagnostic accuracy of physical examination tests for painful cervical radiculopathy: update of a systematic review and meta-analysis. BMC Musculoskeletal Disorders. 2026;27(1):338. doi:10.1186/s12891-026-09551-0.