MRI Cervical Spine – Generic Standard Protocol

MRIninja Knowledge Base | Master / General Page Version 1.0 — April 2026 | Evidence review through April 2026 Audience: Radiologists · Neuroradiologists · MRI Technologists · Advanced Students

1. Executive Summary

MRI is the gold-standard modality for the evaluation of cervical spine pathology. Its ability to directly visualise the spinal cord, nerve roots, intervertebral discs, ligaments, and bone marrow simultaneously — without ionising radiation — makes it the investigation of choice for the vast majority of non-traumatic cervical spine indications [1, 2].

Compared with CT, cervical spine MRI provides decisive superiority for soft tissue and neural structure assessment: cord signal characterisation, disc-thecal sac relationships, foraminal nerve root compression, ligamentous integrity, and bone marrow infiltration detection. CT retains advantages in the acute trauma setting (cortical fracture detail, subluxation geometry with dynamic imaging), for complex posterior element anatomy, and for ossified posterior longitudinal ligament (OPLL) extent assessment. CT myelography remains a valid alternative in patients with contraindications to MRI and provides excellent thecal sac and nerve root definition at higher spatial resolution.

Radiography provides limited structural information — adequate for alignment, spondylolisthesis, and gross deformity assessment — but cannot assess neural structures, cord, or soft tissue disease. Ultrasound has no diagnostic role in cervical spine imaging.

Cervical spine MRI differs fundamentally from lumbar spine MRI in two critical respects that define the entire protocol design philosophy: first, the spinal cord is present throughout the cervical region — cord signal, morphology, and canal diameter are primary diagnostic targets in addition to the disc-nerve root assessment that dominates lumbar imaging; second, motion artefact sources are qualitatively different and more numerous, including cardiac pulsation transmitted to the cord, CSF pulsation in the thecal sac, swallowing from the pharynx/larynx immediately anterior to the spine, and physiological neck motion. These factors must be actively managed through protocol design.

1.1 Core Strengths

• Spinal cord characterisation: Direct visualisation of cord signal intensity, diameter, morphology, and atrophy across all cervical levels — the fundamental advantage over all other modalities for myelopathy evaluation.
• Disc-neural interface: High-resolution assessment of disc-thecal sac and disc-nerve root relationships for radiculopathy and myelopathy localisation.
• Bone marrow and ligamentous assessment: Simultaneous bone marrow signal characterisation and ligamentous integrity evaluation in a single examination.
• No ionising radiation: Essential for a region requiring serial monitoring and in younger patients.
• Multiparametric: Each sequence interrogates different tissue properties — cord signal (T2, STIR), disc hydration (T2), bone marrow (T1), and oedema (STIR) — simultaneously in one examination.

1.2 Intrinsic Limitations of the Generic Protocol

The generic standard cervical spine protocol is a broad-sensitivity survey not optimised for any single entity. These limitations define when child protocols are required.

Motion artefact vulnerability: The cervical spine is the most motion-affected region of the spinal column. Cardiac pulsation transmitted cranially, CSF flow, swallowing, and physiological neck micromotion all generate phase-encoding ghosting. Unlike the lumbar spine, where motion arises primarily from the anterior abdomen, cervical spine motion sources are in direct contact with or immediately adjacent to the diagnostic region of interest.

Spatial resolution constraints: The spinal cord at the cervical level is 8–10 mm in transverse diameter. Subtle cord signal changes, small demyelinating plaques, and early myelopathic cord damage may be at the boundary of reliable detection with standard protocol parameters. Dedicated thin-slice high-resolution protocols are required for these applications.

Small FOV challenge: The cervical canal is small. Axial images require smaller FOV than lumbar (160–200 mm vs 180–220 mm) to achieve adequate in-plane resolution for disc-canal-root relationships, increasing susceptibility to phase-wrap artefact.

Cervicothoracic junction fat saturation failure: The fat suppression failure rate is highest at the cervicothoracic junction (C7–T1) due to susceptibility variations from the shoulder girdle. This is the most technically vulnerable point of the examination and is consistently the most common cause of non-diagnostic STIR or fat-saturated sequences in this region.

Field strength limitations: The cervical cord at 1.5T may approach the threshold for reliable intramedullary lesion detection; 3T is preferred when cord pathology is the primary clinical question, but introduces higher artefact burden from cardiac motion, susceptibility, and chemical shift.

When a dedicated child protocol is required: Multiple sclerosis (thin-slice STIR/PSIR, standardised MAGNIMS protocol), suspected cord infarction, post-operative cervical spine, suspected infection (STIR + contrast), suspected neoplasm (DWI + contrast), cervical cord atrophy assessment (3D T2 volumetry), rheumatoid arthritis craniocervical assessment, and cardiac-gated cord imaging. For sequence-level protocol optimisation, vendor terminology and artefact management, see the dedicated MRIninja page Diffusion-Weighted Imaging (DWI) Sequence.

2. Main Clinical Indications

2.1 Standard Indications

Cervical radiculopathy is the dominant indication for cervical spine MRI in clinical practice. MRI is the preferred modality for assessment of new or increasing radiculopathy due to its superior nerve root definition [1]. The standard protocol reliably detects disc herniation, foraminal stenosis, and neural compression at all cervical levels. ACR Appropriateness Criteria (2024 update) designate MRI without contrast as "usually appropriate" for new or increasing cervical radiculopathy without red flags in adults [1]. The generic protocol is generally sufficient for the initial diagnostic workup.

Cervical spondylotic myelopathy (CSM) is the most common non-traumatic cause of spinal cord dysfunction in adults over 55 years. MRI is the diagnostic standard, providing direct visualisation of cord compression, canal diameter, cord signal change (T2 hyperintensity indicating myelomalacia or oedema), cord atrophy, and multi-level disease extent. The standard protocol is typically sufficient for initial assessment; dedicated thin-slice axial acquisitions may be needed for surgical level planning.

Non-specific neck pain with clinical concern for structural pathology: The ACR Appropriateness Criteria 2024 do not support MRI as the initial investigation for acute or increasing neck pain without radiculopathy in the absence of red flags, noting that spondylotic changes are identified in asymptomatic individuals and may generate false-positive interpretation [1]. When clinical features justify imaging, the standard protocol provides a structural survey.

Post-traumatic assessment (subacute to chronic): In the subacute and chronic post-traumatic setting, MRI assesses disc injury, ligamentous integrity, cord signal change (cord contusion evolution), epidural haematoma resolution, and vertebral marrow oedema. Acute cervical trauma is covered under emergency imaging (Section 2.2).

Pre-operative assessment for cervical spine surgery: MRI informs surgical planning regarding level selection, canal diameter, disc morphology, cord signal, and adjacent segment status. CT is complementary for osseous anatomy, and CTA or MRA for vascular assessment if relevant.

Cervical spine involvement in rheumatoid arthritis and systemic inflammatory disease: Atlantoaxial instability, pannus formation, and subaxial subluxation require dedicated craniocervical coverage with sagittal sequences extending from the posterior fossa to the mid-thoracic spine. The standard cervical protocol requires extension of coverage superiorly to include the occipitoatlantal junction.

Suspected or known demyelinating disease (initial assessment): The standard protocol provides initial cervical cord assessment for demyelinating lesions. Dedicated MS protocols with thin-slice STIR and axial PD/T2 are required for definitive lesion counting and monitoring (child page).

Cervical cord compression monitoring in known stenosis: Serial MRI monitors disease progression, canal diameter changes, and cord signal evolution in patients with known cervical stenosis managed conservatively.

2.2 Urgent Red Flags Requiring Expedited or Emergency Imaging

3. Preparation Reference

Universal MRI preparation is centralised in the MRIninja Patient Preparation master page. The following covers only items specific to cervical spine MRI.

3.1 Anatomy-Specific Preparation Items

Metallic implants and prior surgery: Anterior cervical discectomy and fusion (ACDF) hardware (titanium cages, plates, screws) and posterior cervical fixation constructs must be identified before imaging. Modern titanium cervical hardware is generally MR-compatible and does not require conditional clearance at 1.5T or 3T, but generates susceptibility artefact that degrades adjacent tissue assessment. Stainless steel hardware produces more extensive artefact than titanium. Cervical spinal cord stimulators require device-specific MR conditional clearance per manufacturer. Vertebral haemangioma embolisation coils require individual evaluation.

Jewellery, necklaces, and neck accessories: All necklaces, pendants, chains, and chokers must be removed. These are in the direct field of view of the cervical spine and are a common missed source of susceptibility artefact. Neck piercings — particularly near the midline — must be removed where physically possible.

Clothing: Metal-containing clasps, underwire bras, or zip fasteners at the shoulder and upper thorax level generate susceptibility artefact that propagates into the lower cervical spine. Patients should be provided with a gown.

Pain management and positioning considerations: Cervical spine pain — particularly acute disc herniation or acute myelopathy — may prevent the patient from maintaining neutral head position. Slight neck extension or rotation adopted to relieve pain produces diagnostic-quality images in non-standard anatomical planes that complicate disc assessment. If clinically appropriate, oral or IV analgesia before positioning improves both patient tolerance and image quality.

History of dysphagia or excessive salivation: Patients with Parkinson's disease, neurological conditions affecting swallowing, or recent posterior pharyngeal surgery produce increased swallowing artefact. The technologist must be aware and may need to acquire short-duration sequences or repeat sequences affected by swallowing.

Patient history modifying the protocol:

• Prior cervical spine surgery → consider contrast; assess for post-operative change
• Known or suspected inflammatory arthritis (RA, AS) → extend coverage to craniocervical junction
• Known or suspected MS → dedicated thin-slice STIR/PSIR protocol (child page)
• Known malignancy → DWI + contrast; full spine coverage
• Acute trauma → emergency modified protocol (CT first in most pathways)
• Respiratory disease or frequent coughing → short acquisition sequences; cardiac gating may be needed

3.2 Patient Positioning on the MRI System

Patient position: Supine, head-first entry. This is the universal standard for cervical spine MRI. The patient's head is placed within a dedicated head-neck coil or cervical spine coil array.

Coil selection: A dedicated multichannel head-neck coil array is the optimal choice for cervical spine MRI. These coils provide combined coverage of the brain, cervical spine, and upper thoracic region within a single coil configuration, avoiding coil repositioning. If a dedicated cervical spine coil is used without the head component, ensure that the upper cervical spine (C1–C2) is not at the periphery of the coil's sensitive region. The lower cervical spine (C6–C7) and cervicothoracic junction (C7–T1) are frequently at the inferior edge of coil sensitivity — this is the region of greatest SNR vulnerability and must be actively monitored.

Centering: The coil isocentre should be positioned at the level of C4–C5, representing the mid-cervical spine. This ensures that the C2–C3 to C7–T1 range — the clinically critical levels for disc disease and radiculopathy — falls within the region of maximum coil sensitivity.

Head and neck alignment: The patient's head must be in the neutral position — no flexion, extension, or rotation. Slight residual lordosis of the cervical spine is normal and expected. The patient's head axis should be aligned with the scanner bore axis. Even mild rotation introduces asymmetry in axial slice assessment and apparent foraminal dimension asymmetry.

Immobilisation: Foam pads on each side of the head within the coil reduce lateral micromotion. A small support under the neck maintains the physiological cervical lordosis. Excessive padding under the neck that forces abnormal extension degrades the geometry of the disc-canal assessment. Patients should be instructed to swallow before each sequence begins and to minimise swallowing and neck movement during acquisition. This verbal instruction is the single most effective strategy for swallowing artefact reduction in clinical practice.

Shoulder depression: In patients with significant trapezius muscle bulk or elevated shoulders, the shoulder girdle may encroach on the lower cervical spine field of view and contribute to B0 inhomogeneity at C7–T1. Gentle downward shoulder positioning with shoulder straps or patient self-effort to depress the shoulders during the examination reduces this effect. Advise the patient to "keep the shoulders relaxed and low, away from the ears."

Pre-scan technologist checks:

1. Confirm correct coil activation on console; verify all elements of the head-neck coil array are active.
2. Verify laser centring at the mid-cervical level.
3. Ensure no metallic jewellery at the neck.
4. Confirm patient can maintain the neutral head position without discomfort.
5. Instruct patient regarding swallowing suppression.
6. Acquire the three-plane localiser and verify cervical spine from C1 to T2 is within the field of view before committing to full protocol.
7. Verify anterior saturation band position on the localiser before starting diagnostic sequences.

4. Standard Protocol Design

4.1 Mandatory Core Sequences

4.2 Conditional Sequences

4.3 Rationale Summary Per Sequence

Sagittal T2 TSE — the primary diagnostic sequence for the cervical spine. Bright nucleus pulposus in hydrated discs; bright CSF; dark cord (intermediate signal); dark cortical bone and ligaments. The spinal cord is the primary diagnostic target unique to the cervical spine: T2 hyperintensity within the cord indicates myelomalacia, acute oedema, or demyelination — the most clinically critical finding in cervical MRI. The sagittal T2 also provides overview of canal diameter, disc degeneration and herniation, vertebral body alignment, and overall spinal anatomy from the posterior fossa to the upper thoracic region.

What it detects well: Cord signal (T2 hyperintensity in myelopathy, oedema, demyelination), disc herniation morphology, canal compromise, vertebral alignment, CSF spaces, posterior fossa overview.

What it misses at standard parameters: Subtle cord signal changes may be below reliable detection at 1.5T. Small demyelinating plaques require thin-slice STIR or PSIR (at 3T). Foraminal assessment is suboptimal in the sagittal plane alone.

Critical limitation: Unlike the lumbar spine, the cervical cord T2 is susceptible to cardiac pulsation artefact producing apparent T2 signal change or cord border obscuration on individual slices. Sagittal T2 without an anterior saturation band will invariably show CSF and vascular pulsation ghosting that propagates through the cord in the phase-encoding direction.

Technologist note: An anterior saturation band placed over the prevertebral soft tissues and pharynx is mandatory for all sagittal sequences. Without it, swallowing-related and vascular pulsation ghosting will degrade image quality in the phase direction regardless of other optimisations.

Sagittal T1 TSE — bone marrow characterisation. Same role as in the lumbar spine: bright fatty marrow is the baseline for T1 signal loss interpretation (metastasis, infection, oedema). Additionally useful for: subacute cord haemorrhage (T1-bright methemoglobin within the cord), prevertebral soft tissue assessment, and morphological survey of vertebral body alignment and disc height.

What it detects well: Bone marrow T1 signal (T1 dark = marrow replacement, oedema, or infiltration), subacute haemorrhage within disc or cord, prevertebral soft tissue swelling (epidural haematoma, retropharyngeal abscess), ligament assessment.

Limitation: Cord internal signal changes are less conspicuous on T1 than T2.

Sagittal STIR — bone marrow oedema sentinel and cord lesion-enhancing sequence. Same principle as in lumbar spine: TI nulls fat, revealing oedema and inflammation by additive T1+T2 contrast. In the cervical spine, STIR is particularly valuable for: vertebral marrow oedema (fracture, Modic, metastasis), ligamentous oedema (post-traumatic injury), cord oedema (acute cord contusion), and paraspinal soft tissue inflammatory change.

At 1.5T, STIR has been demonstrated to significantly improve cervical spinal cord lesion detection compared to standard T2 alone [3, 4]. At 3T, PSIR achieves superior sensitivity over STIR in the cervical cord. Full sequence comparison data are in Section 10.3. The standard protocol includes STIR as the primary fat-suppressed sequence for cervical MRI.

What it detects well: Marrow oedema, cord oedema, ligament tears (increased water content), paraspinal soft tissue oedema, post-traumatic cord change.

Critical pitfall: STIR must never be acquired after gadolinium — see Section 4.5 for the full technical rationale.

Technologist-specific challenge: At the cervicothoracic junction (C7–T1), B0 inhomogeneity from the shoulder girdle fat-air interface causes STIR TI null-point shift, resulting in incomplete fat suppression regionally. This is the most common STIR quality failure in cervical spine MRI [5]. The shim volume should be restricted to the vertebral bodies, canal, and spinous processes, excluding the shoulder soft tissues.

Axial T2 TSE (multi-level) — the nerve root compression and canal morphology sequence. Provides cross-sectional anatomy at each disc level: disc-thecal sac relationship, lateral recess and foraminal dimensions, spinal cord shape and signal at each level. In the cervical spine, axial T2 is also the primary sequence for detecting cord flattening, lateral cord displacement, and central cord compromise by disc or osteophyte complex.

Planning requirement: Each disc level must be individually angulated perpendicular to the cord at that level — see Section 4.6. A single block angled to the C4–C5 disc will produce oblique axial sections at C2–C3 and C6–C7 due to the cervical lordosis and the convergence of disc spaces.

Axial T1 TSE (multi-level) — bright epidural and foraminal fat provides natural contrast for nerve root assessment. In the cervical foramina, fat obliteration indicates foraminal compromise. Also valuable for bone marrow signal assessment at the pedicle level.

4.4 Sequence Matching and Cross-Sequence Consistency

Sagittal T1, T2, and STIR must share identical or closely matched slice geometry (thickness, gap, FOV, centering) so that direct level-by-level comparison is possible. The combination of sagittal T1 and STIR signal at each vertebral body defines the acute vs. chronic nature of any bone marrow change (T1 dark + STIR bright = acute; T1 bright = chronic fat replacement).

Axial T2 and T1 must share identical per-level angulation and coverage to enable direct cross-modality comparison at each disc level.

For contrast examinations, pre-contrast fat-suppressed T1 must exactly match post-contrast acquisition geometry. Pre-contrast baseline is mandatory before any gadolinium injection.

Serial follow-up examinations require identical positioning, coil configuration, and sequence geometry. The cervical cord signal must be evaluated at comparable spinal levels across studies; angulation discrepancy between studies may produce apparent signal change artefactually.

4.5 Fat Suppression in Cervical Spine MRI

STIR: The preferred fat suppression technique for sagittal bone marrow oedema and cord assessment. B0-independent; more robust than spectral fat saturation across the large cervical FOV and at the cervicothoracic junction. TI must be recalibrated for 3T (≈200–230 ms vs ≈160–175 ms at 1.5T). Cannot be used post-gadolinium.

Spectral fat saturation (SPIR/SPAIR/ChemSat): Used for post-contrast T1 fat-saturated sequences. More vulnerable to B0 inhomogeneity than STIR, particularly at the cervicothoracic junction. Shim volume restriction to the vertebral canal region (excluding shoulder fat) significantly improves suppression quality at this level [5]. If spectral fat saturation completely fails and produces water saturation instead, the sequence must be repeated without fat saturation — a water-saturated non-diagnostic image is worse than a non-fat-suppressed image.

Dixon technique: Increasingly available and preferred at 3T for its B0-independent fat-water separation. Particularly valuable for the cervicothoracic junction where spectral fat saturation most frequently fails. Provides fat-only and water-only images from a single acquisition — additional characterisation capability without extra scan time.

Fat suppression is not applied to: standard sagittal T1 (bright marrow fat is the diagnostic signal); standard axial T2 (foraminal fat provides natural contrast); standard axial T1 (same reason).

Critical warning: At the cervicothoracic junction, spectral fat saturation failure can produce complete water saturation — the image appears to show uniformly abnormal signal that can falsely suggest extensive marrow oedema or ligamentous injury. This is the single most common serious quality failure in cervical spine MRI. The technologist must verify fat suppression by confirming that subcutaneous fat is dark before accepting the STIR or fat-saturated T1 sequence.

5. Optimisation Strategy

5.1 Artifact Reduction by Source

Swallowing artefact is the dominant and most cervical spine-specific artefact problem. It is not present in lumbar spine MRI in the same way. Swallowing displaces the pharynx and larynx anteriorly and superiorly during deglutition, generating a sharp motion event that corrupts k-space data in the phase-encoding direction. On sagittal sequences, this manifests as ghost images of the pharynx and larynx superimposed on the cord and disc spaces. The artefact is characteristically repetitive, semi-regular, and propagates across the full image in the S-I direction. It can directly simulate disc disease, cord signal change, or epidural pathology.

Reduction strategies:

• Verbal instruction before each sequence: "Please swallow now, then try not to swallow during the sequence"
• Anterior saturation band on all sagittal and axial sequences (mandatory)
• S-I phase encoding for sagittal (displace ghost pattern superiorly/inferiorly)
• Shorter TR and total acquisition time — fewer swallowing opportunities per sequence
• PROPELLER/BLADE k-space trajectories (radial filling) reduce motion sensitivity compared to rectilinear TSE — available as an option on some vendor platforms for cervical spine [Peh, 2001]

CSF and vascular pulsation artefact is more severe in the cervical spine than any other spinal region. Cardiac pulsation generates cranially directed systolic CSF flow pulses in the cervical canal, producing phase errors in synchrony with the cardiac cycle. These manifest as ghost images of the thecal sac and cord displaced in the S-I (phase) direction, periodically overlying the cord at displaced positions. They can simulate cord signal heterogeneity, intradural lesions, or cord displacement.

Reduction strategies:

• Phase encoding S-I direction displaces ghost pattern toward brain and thoracic spine rather than directly through the cord region of interest
• Flow compensation (gradient moment nulling): reduces velocity-dependent phase errors from CSF flow; increases minimum TE by ~15–20% but significantly reduces pulsation ghosting on sagittal T2
• Cardiac gating: eliminates pulsation artefact completely but increases acquisition time 50–100% due to variable TR (trigger delay-dependent); reserved for dedicated cord imaging protocols
• Anterior saturation band suppresses signal from vascular pulsation in the prevertebral region
• Shorter TE: at 3T, shorter TE reduces the phase accumulation time for CSF pulsation effects

Chemical shift artefact at cervicothoracic junction: Fat saturation failure at C7–T1 is the most common fat suppression failure point in the entire spine. The shoulder girdle fat-air interface creates B0 field perturbations that shift the fat resonance frequency locally, causing the fat saturation pulse to miss fat in this region while potentially saturating water-containing structures (water saturation artefact). On STIR, this appears as a region of incomplete fat nulling with residual bright fat signal, or alternatively as erroneous signal suppression in the disc/marrow. On SPAIR/CHESS fat-saturated T1, it appears as a bright region of unsuppressed fat.

Reduction strategies:

• Restrict shim volume to the cervical canal and vertebral bodies, excluding shoulder fat
• Use STIR (B0-independent) rather than spectral fat saturation for sagittal oedema sequences
• Use Dixon technique for the most robust fat suppression at the cervicothoracic junction
• Reducing the shimming volume: on Siemens, use the "SmartExam" or manual shim volume; on GE, restrict the ASSET/ARC calibration region

Cardiac motion transmission: The aortic arch and heart transmit periodic motion to the craniocervical and cervical regions, particularly at 3T where the reduced T1 of myocardium relative to cervical cord reduces natural CSF suppression efficiency. Combined CSF pulsation and cardiac ghosting creates complex artefact patterns at the lower cervical cord level.

Reduction strategies:

• As above (cardiac gating, flow compensation)
• Increasing NSA/NEX: coherent motion ghosts do not average out proportionally, but the signal-to-noise ratio of the cord signal improves, making distinction from artefact easier
• Cardiac gating with trigger delay set to diastole (minimum cord motion period)

Metallic susceptibility from cervical hardware: Post-operative cervical spine with titanium ACDF cages or posterior fixation constructs produces susceptibility blooming on all sequences, most severely on GRE-based sequences. TSE sequences are intrinsically less susceptible than GRE. Metal artefact is worst at 3T. For sequence-level protocol optimisation, vendor terminology and artefact management, see the dedicated MRIninja page Gradient Echo (GRE/FLASH) Sequence.

Reduction strategies:

• Use TSE (spin echo) sequences; avoid GRE in post-operative cervical spine
• SEMAC/WARP sequences for dedicated metal artefact reduction
• Reduce TE; increase bandwidth
• 1.5T preferred over 3T in the presence of extensive metallic hardware

Gibbs ringing at cord surface: Truncation artefact from finite k-space sampling produces oscillating signal bands at the cord-CSF interface. These can simulate an intramedullary lesion or syrinx. Adequate matrix size (minimum 256 in the phase direction) and avoidance of aggressive partial Fourier reduce this artefact.

5.2 Protocol Efficiency and Throughput

Routine full protocol: Sagittal T2 + T1 + STIR + axial T2 + axial T1 = approximately 25–35 minutes at 1.5T; 20–28 minutes at 3T with parallel imaging.

Abbreviated protocol for radiculopathy: Sagittal T2 + STIR + axial T2 (15–20 minutes) is adequate for disc herniation, foraminal stenosis, and cord compression assessment. Some departments omit sagittal T1 and axial T1 in low-complexity radiculopathy without bone marrow concern, accepting reduced marrow characterisation.

When 3D TSE is worth the time: Isotropic 3D T2 TSE enables curved planar reformatting along the cord axis, true coronal reconstruction for bilateral foraminal comparison, and sub-millimetre voxel size for nerve root and foramen detail. Evidence supports its value for foraminal nerve root assessment [cord 3D ref]. However, cord signal characterisation is generally inferior to dedicated 2D STIR, and 3D acquisitions are more vulnerable to bulk motion.

When 2D is more robust: Standard 2D TSE is more motion-robust per slice, provides more reliable cord signal characterisation with STIR, and allows individual level angulation. For routine clinical cervical MRI, 2D remains the reference standard.

5.3 Field Strength Considerations

Clinical recommendation: 3T is preferred when cord signal assessment is the primary question (myelopathy, demyelination). 1.5T remains adequate for standard radiculopathy evaluation, post-operative hardware cases, and patients with implants approved only at 1.5T.

6. Contrast Use Principles Specific to Cervical Spine MRI

6.1 Non-Contrast Standard Protocol — Sufficient For

The non-contrast standard protocol is adequate for: cervical radiculopathy and disc herniation assessment; degenerative spondylosis with myelopathy evaluation; non-specific neck pain assessment; pre-operative planning in degenerative disease; most acute trauma assessments; vertebral fracture evaluation; monitoring of known degenerative stenosis; standard spondyloarthropathy screening.

Most ACDF post-operative assessments in straightforward clinical contexts do not require contrast, as anterior cervical surgery produces minimal epidural scar (unlike posterior lumbar discectomy). The ACR 2024 update notes that contrast-enhanced MRI is not routinely required after ACDF unless infection, neoplasm, or recurrent radiculopathy is suspected [1].

6.2 Gadolinium Indicated — Cervical Spine-Specific Contexts

Suspected cervical epidural abscess or spondylodiscitis: GBCA is required to characterise the extent of infection, differentiate infectious from degenerative endplate changes, and delineate epidural extension. Full spine coverage is indicated in confirmed infection (see child page).

Known or suspected neoplasm: Enhancement characterises lesion biology, differentiates benign from aggressive lesions, and delineates epidural and perineural extent. Post-contrast fat-suppressed T1 is the primary post-contrast sequence.

Suspected intradural extramedullary pathology: Nerve sheath tumours, meningiomas, and leptomeningeal disease require post-contrast enhancement for diagnosis and extent characterisation.

Post-operative cervical spine with new neurological symptom: Unlike post-operative lumbar spine, contrast is not routinely required after ACDF (minimal epidural scar). However, new neurological symptoms in any post-operative context require contrast if infection, haematoma, or recurrent or new disc/tumour is considered.

Suspected inflammatory myelopathy or cord lesion: Post-contrast T1 detects active enhancement of demyelinating lesions (active blood-brain barrier disruption). Used when MS activity assessment will change management. The addition of a post-contrast sequence should follow MS protocol child page design.

Vascular malformations: Dural arteriovenous fistulas, cord AVMs, and cavernomas may benefit from post-contrast enhancement for lesion detection and extent assessment. Standard T2 and SWI/T2* sequences detect most cavernomas; contrast MRA may be required for AVF. For sequence-level protocol optimisation, vendor terminology and artefact management, see the dedicated MRIninja page Susceptibility Weighted Imaging (SWI) Sequence.

6.3 Post-Contrast Acquisition Timing

Standard timing: 3–5 minutes after GBCA injection for most cervical spine indications. Allows adequate tissue accumulation while intravascular concentration has equilibrated.

For leptomeningeal or epidural involvement, delayed imaging at 10–15 minutes increases sensitivity. Document injection time in PACS. Pre-contrast T1 fat-suppressed must be acquired before injection.

7. Reporting Essentials

7.1 Interpretation Framework

Cervical spine MRI report interpretation follows the same diagnostic axes as the lumbar spine, with the critical addition of the spinal cord as a primary reporting element:

Cord assessment must be addressed first: Any cord signal change, cord atrophy, or canal compromise threatening cord perfusion must be the first priority in every cervical MRI report, regardless of the clinical question. Cord signal change may be asymptomatic or may explain unrecognised neurological signs.

Degenerative vs. non-degenerative: Cervical spondylotic changes are ubiquitous in adults; up to 90% of asymptomatic individuals over 60 years have MRI evidence of disc degeneration at some cervical level. The interpreter must explicitly link structural findings to the clinical presentation and avoid over-reporting incidental age-related change.

Acute vs. chronic: Bone marrow signal pattern (T1 + STIR combination) distinguishes acute oedema from chronic fat replacement at each level.

7.2 Mandatory Reporting Checklist

Spinal cord (primary target in cervical spine):

• Cord signal: T2 hyperintensity present or absent, location (central/lateral/dorsal), level, extent
• Cord morphology: atrophy, flattening, diameter at each level
• Cord-canal relationship at all levels
• Craniocervical junction and cervicomedullary cord

Vertebral bodies (each level C1–T1):

• Alignment and height
• Bone marrow signal (T1 + STIR combination)
• Fracture deformity or cortical irregularity
• Modic endplate changes (if present)

Intervertebral discs (C2–C3 to C7–T1):

• Disc height
• T2 signal / degeneration
• Disc herniation (type, direction, level)
• Disc-cord and disc-nerve root relationships

Spinal canal:

• Canal anteroposterior diameter at each level (normal: ≥13 mm at C3–C5; stenosis: <10 mm)
• Thecal sac compression
• Central cord compression (T-shaped or crescent deformity of cord)

Neural foramina:

• Foraminal stenosis each level each side
• Nerve root visibility and morphology

Posterior elements:

• Facet joints (degeneration, effusion)
• Ligamentum flavum (hypertrophy)
• Posterior longitudinal ligament (OPLL if present)
• Ligamentum nuchae, interspinous ligaments (STIR signal in trauma)

Paravertebral and prevertebral soft tissues:

• Prevertebral soft tissue swelling (trauma, infection, mass)
• Paraspinal musculature

Technical items:

• Motion artefact impact on diagnostic confidence
• Fat suppression quality at cervicothoracic junction
• Coverage adequacy (posterior fossa to T2 minimum)
• Comparison with prior studies

7.3 Structured Reporting

Reports should follow the standard structure: Indication → Technique (field strength, sequences, coil, saturation bands, contrast) → Comparison → Findings (cord first, then systematic by level) → Impression (concise, answering the clinical question) → Limitations → Critical communication if required.

Critical communication: Unexpected acute cord compression, acute cord oedema, epidural abscess, or cord infarction requires direct verbal communication to the referring clinician, documented in the report with time and recipient.

7.4 Incidental Findings — Clinical Decision Framework

Usually benign, no action required: Age-appropriate disc desiccation and height loss; mild multi-level spondylosis without cord compromise; small Schmorl's nodes; mild facet degeneration; developmental bone variant (os odontoideum if clinically stable and known); Chiari I malformation with tonsils <5 mm below foramen magnum in asymptomatic patient.

Requires documentation and follow-up: Unruptured dural arteriovenous malformation or suspected vascular anomaly; incidental thyroid nodule visible in the lower cervical FOV; Chiari I with tonsils ≥5 mm in symptomatic patient; indeterminate vertebral lesion.

Urgent/clinically important: Unexpected cord compression with or without cord signal change — communicate directly; unexpected atlantoaxial instability (atlantodental interval >3 mm in adults); incidental epidural abscess or mass; cord infarction pattern not consistent with clinical history.

8. MRI Technologist Pearls

8.1 Sequence Order Logic

Recommended standard order:

1. Three-plane localiser — verify coverage C1 to T2 minimum; check position of anterior saturation band
2. Sagittal T2 TSE — first diagnostic sequence; planning reference for all others; anterior saturation band applied
3. Sagittal STIR — copied from sagittal T2 geometry; anterior saturation band applied
4. Sagittal T1 TSE — copied from sagittal T2 geometry
5. Axial T2 TSE (multi-level) — planned individually per level from the acquired sagittal T2
6. Axial T1 TSE (multi-level) — copied from axial T2 geometry per level

Rationale: If the patient cannot complete the full protocol, sagittal T2 and STIR provide cord signal, disc, canal, and marrow information. The axial series requires active re-planning from the sagittal T2 and is acquired after the sagittal series is confirmed adequate.

8.2 Positioning Tricks

• Shoulder depression instruction: Before starting, ask the patient to "press the shoulders gently downward, away from the ears, and relax the trapezius." Repeat this before the STIR and fat-suppressed sequences, as shoulder position drift during the examination is common.
• Swallowing instruction before each sequence: "Please swallow now. Try not to swallow during the next few minutes while the scanner is making noise."
• Head neutral position check: Before starting, confirm the nose is pointing directly to the ceiling with no lateral rotation. Even 10–15° of head rotation produces visible foraminal dimension asymmetry on axial images.
• Anterior saturation band check: Verify saturation band position on the planning images before every sequence. A band drifting posteriorly onto the vertebral bodies will suppress cord and disc signal.
• Lower cervical spine coverage check: After the sagittal T2, verify that C7–T1 is included and that the STIR fat suppression quality is adequate at this level before proceeding. If not, readjust shim volume and repeat STIR.
• Intercom between sequences: Regular communication reassures the patient and allows early detection of increased motion before long sequences (STIR) begin.

8.3 Fast Salvage Protocol

Core minimum (emergency or severely compromised patient): Sagittal T2 + STIR = 7–10 minutes; provides cord signal, disc morphology, and acute marrow change at all levels.

8.4 Common Avoidable Errors

9. Quality Control Checklist

Coverage:

• Sagittal series includes pons/cervicomedullary junction superiorly
• Sagittal series includes T2–T3 inferiorly
• Both foramina visible on lateral sagittal slices (right and left transverse processes)
• Axial series covers all six disc levels (C2–C3 through C7–T1) individually
• Each axial block covers from mid-pedicle to mid-pedicle at each level

Sequence completeness:

• Sagittal T2: acquired, reviewed, no major motion degradation of cord signal
• Sagittal STIR: acquired, fat suppression visually uniform including at C7–T1 (subcutaneous fat is dark)
• Sagittal T1: acquired, no major motion degradation
• Axial T2: all six levels planned individually from sagittal T2, slices correctly angulated per level
• Axial T1: acquired if indicated per local protocol

Artefact assessment:

• No diagnostic-grade swallowing ghosting obscuring cord on sagittal sequences
• No severe CSF pulsation artefact masking cord signal at any level
• Fat suppression quality acceptable at C7–T1 (confirm STIR fat nulling)
• No phase wrap overlying the cord or canal
• Gibbs ringing at cord surface noted if present (does not simulate syrinx or lesion)

Contrast (if used):

• Pre-contrast T1 fat-suppressed acquired before injection
• Injection time documented in PACS and report
• Post-contrast acquisition timing documented
• STIR was NOT acquired post-contrast

Labelling and orientation:

• Patient identifiers correct on all series
• Left-right orientation verified on axial images
• Series correctly labelled (including anterior saturation band documentation)
• Slice orientation and angulation correct per level

Critical finding communication:

• Any unexpected cord signal change, cord compression, epidural abscess, or acute fracture flagged for immediate radiologist review and direct clinical communication

11. Evidence Gaps and Ongoing Debate

Optimal T2 sequence for cervical cord lesion detection at 3T: Standard FSE T2 has demonstrated sensitivity as low as 50.9% for cervical cord MS lesions at 3T compared to STIR (89.6%) and PSIR (96.2%) [3]. The clinical implication — whether STIR should be mandatory in all cervical protocols rather than just MS protocols — has not been evaluated in a broad indication population study.

PSIR vs. STIR for standard cervical spine protocol: PSIR outperforms STIR for cord lesion specificity at 3T in MS; whether this advantage justifies inclusion in all generic cervical protocols (not just MS-specific) versus selective use remains an open practical question. PSIR is not yet universally available across vendors.

2D vs. 3D TSE for cervical spine: Evidence for foraminal assessment advantage of 3D TSE is accumulating; equivalence for cord signal assessment has not been fully established. Motion sensitivity of 3D acquisitions at cervical level (cardiac pulsation over long acquisition time) is a genuine limitation without cardiac gating.

Cardiac gating necessity in routine cervical protocol: Cardiac gating significantly reduces CSF pulsation artefact and cord signal ghosting but increases acquisition time by 50–100%. No prospective study has established whether the diagnostic benefit of cardiac gating outweighs the time cost in routine (non-MS) cervical MRI.

Optimal axial angulation — perpendicular to cord vs. parallel to disc: The clinical superiority of one approach over the other for routine cervical radiculopathy evaluation has not been evaluated in a controlled prospective study. Current practice follows expert consensus and institutional preference.

Role of DWI in routine cervical protocols: DWI is not standard but has demonstrated utility in cord infarction and infection. Reduced FOV DWI has improved technical feasibility; whether DWI should be added to the standard protocol for all cervical myelopathy workups is not established by evidence.

Field strength choice for cervical myelopathy: While 3T is generally preferred for cord signal assessment, no randomised controlled study has compared clinical diagnostic accuracy (not just lesion detection) between 1.5T and 3T for cervical myelopathy outcomes. The recommendation for 3T is based on physical SNR advantages and studies using lesion detection as a surrogate endpoint.

AI reconstruction for cervical spine: Deep learning reconstruction enables protocol acceleration or resolution enhancement at the cervical level. Validation of diagnostic equivalence across the range of cervical spine pathologies is at early stages.

12. Evidence-Based References

A. Guidelines / Consensus / Society Recommendations

[1] Eldaya RW, Parsons MS, Hutchins TA, et al; Expert Panel on Neurological Imaging. ACR Appropriateness Criteria® Cervical Pain or Cervical Radiculopathy: 2024 Update. J Am Coll Radiol. 2025;22(5S):S136–S162. DOI: 10.1016/j.jacr.2025.02.035. PMID: 40409873. (Evidence Level: High — Guideline) Relevance: Primary 2024 ACR guidance for cervical spine imaging across all clinical variants; defines MRI appropriateness.

[2] Expert Panel on Neurological Imaging; Hutchins TA, et al. ACR Appropriateness Criteria® Cervical Neck Pain or Cervical Radiculopathy. J Am Coll Radiol. 2019;16(5S):S57–S68. PMID: 31054759. (Evidence Level: High — Guideline) Relevance: Prior ACR guideline version; establishes non-contrast MRI appropriateness for cervical radiculopathy.

[7] ACR Appropriateness Criteria® Myelopathy. American College of Radiology. Updated 2022. (Evidence Level: High — Guideline) Relevance: Defines MRI appropriateness for myelopathy evaluation including cord compression and signal change.

[8] ACR Appropriateness Criteria® Acute Spinal Trauma. J Am Coll Radiol. 2025;22(5S). PMID: 40409895. (Evidence Level: High — Guideline) Relevance: Defines role of MRI vs CT in acute cervical spine trauma.

B. Systematic Reviews / Meta-analyses

[9] Brinjikji W, Luetmer PH, Comstock B, et al. Systematic literature review of imaging features of spinal degeneration in asymptomatic populations. AJNR Am J Neuroradiol. 2015;36(4):811–816. PMID: 25430861. (Evidence Level: High — Systematic review) Relevance: Establishes that cervical spondylotic changes are ubiquitous in asymptomatic individuals; informs cautious over-reporting guidance.

C. Important Original Studies

[3] Bucher SF, Seelos KC, Reiser MF, et al. Comparison of Sagittal FSE T2, STIR, and T1-Weighted Phase-Sensitive Inversion Recovery in the Detection of Spinal Cord Lesions in MS at 3T. AJNR Am J Neuroradiol. 2016;37(5):970–977. PMID: 26797141. DOI: 10.3174/ajnr.A4655. PMC: 7960295. (Evidence Level: Moderate — Original prospective study) Relevance: Demonstrates PSIR (96.2%) and STIR (89.6%) are significantly superior to FSE T2 (50.9%) for cervical cord demyelinating lesion detection at 3T; directly defines cervical spine STIR protocol recommendation.

[10] Matsumoto M, Fujimura Y, Suzuki N, et al. MRI of cervical intervertebral discs in asymptomatic subjects. J Bone Joint Surg Br. 1998;80(1):19–24. PMID: 9460947. (Evidence Level: Moderate — Original study) Relevance: Foundational study on prevalence of cervical disc abnormalities in asymptomatic subjects; informs cautious reporting of incidental disc pathology.

[11] Ross JS, Modic MT. Current assessment of spinal degenerative disease. Semin Ultrasound CT MR. 1992;13(6):429–437. PMID: 1359282. (Evidence Level: Moderate — Original review) Relevance: Historical landmark for MRI characterisation of cervical disc disease; establishes the disc degeneration-signal change relationship.

D. Technical MRI Papers

[12] Peh WCG, Chan JHM. Artifacts in musculoskeletal and spinal MRI. Skeletal Radiol. 2001;30(4):179–191. PMID: 11398948. (Evidence Level: Technical / Foundational) Relevance: Comprehensive illustrated reference for cervical spine MRI artefacts including swallowing, pulsation, and fat saturation failure.

[13] Noda C, Venkatesh BA, Wagner JD, et al. Primer on Commonly Occurring MRI Artifacts and How to Overcome Them. RadioGraphics. 2022;42(3). DOI: 10.1148/rg.210176. (Evidence Level: Technical) Relevance: Recent practical artefact guide applicable to cervical spine protocol design.

[14] Stroman PW, Wheeler-Kingshott C, Bacon M, et al. The current state-of-the-art of spinal cord imaging: methods. NeuroImage. 2014;84:1070–1081. PMID: 24018307. DOI: 10.1016/j.neuroimage.2013.04.124. (Evidence Level: Technical) Relevance: Comprehensive methodology review for spinal cord MRI including positioning, cardiac gating, DWI, and motion management.

[15] Del Grande F, Santini F, Herzka DA, et al. Fat-suppression techniques for 3-T MR imaging of the musculoskeletal system. RadioGraphics. 2014;34(1):217–233. PMID: 24428290. DOI: 10.1148/rg.341135130. (Evidence Level: Technical) Relevance: Fat suppression comparison including cervicothoracic junction challenges; supports STIR and Dixon preference over CHESS at large FOV.

E. Landmark Historical References

[16] Modic MT, Masaryk TJ, Ross JS, Carter JR. Imaging of degenerative disk disease. Radiology. 1988;168(1):177–186. PMID: 3289089. DOI: 10.1148/radiology.168.1.3289089. (Evidence Level: High — Landmark) Relevance: Original description of disc disease MRI characterisation; directly applicable to cervical disc assessment.

[17] Bydder GM, Young IR. MR imaging: clinical use of the inversion recovery sequence. J Comput Assist Tomogr. 1985;9(4):659–675. PMID: 3998939. (Evidence Level: Foundational) Relevance: Original STIR sequence description; foundational reference for STIR clinical application.

End of document — MRI CERVICAL SPINE Generic Standard Protocol — MRIninja Master Page v1.0 — April 2026

This document is designed to serve as the reference base for all child pages dedicated to specific cervical spine pathologies, clinical indications and dedicated protocols.