Soft Tissues Neck MRI — Generic Standard Protocol

Soft Tissues Neck MRI — Generic Standard Protocol

1. Executive Summary

Neck soft tissue MRI provides superior tissue characterisation, multicompartmental assessment, and vascular detail compared with any other imaging modality for the complex anatomy of the cervical region. The neck contains one of the highest concentrations of diagnostically distinct compartments in the body — the suprahyoid and infrahyoid spaces, the retropharyngeal space, the parapharyngeal space, the carotid space, the masticator space, the posterior cervical space, the visceral space, and the perivertebral space — each with characteristic pathological entities and distinct imaging requirements. CT provides superior speed and bone detail; ultrasound provides real-time nodal assessment and guided biopsy; but neither provides the soft tissue contrast, perineural spread detection, bone marrow assessment, or vascular characterisation that MRI delivers.

The generic neck MRI protocol is designed to provide a complete survey of all cervical compartments, identifiable pathology in the soft tissues, muscles, salivary glands, lymph nodes, thyroid gland, and associated vasculature, within a reasonable acquisition time. It is not designed as a dedicated mucosal primary tumour protocol (which requires specific sequences for oral cavity, oropharynx, hypopharynx, and larynx), nor as a dedicated vascular protocol (which requires TOF or phase-contrast MRA sequences). It is the appropriate starting point for the broad category of patients referred with neck mass, neck pain (soft tissue component), suspected deep space infection, lymphadenopathy assessment, salivary gland pathology, and thyroid-adjacent lesions.

The ACR Appropriateness Criteria [1] and the European Society of Head and Neck Radiology (ESHNR) recommendations define the clinical framework within which neck MRI indications are assessed.

1.1 Core Strengths

Multicompartmental soft tissue characterisation: the neck contains multiple fascial-defined spaces whose pathological contents produce characteristic MRI signal patterns. The deep cervical fascia, the middle layer of deep cervical fascia, and the carotid sheath define compartments that CT cannot reliably separate at equivalent contrast resolution. MRI distinguishes neural (intermediate T2, homogeneous enhancement), vascular (flow void or flow-related signal), inflammatory (T2-bright, rim-enhancing), and neoplastic (variable T2, solid enhancement) processes within each compartment.

Perineural tumour spread (PNS): the detection of perineural tumour spread along the cranial nerves — particularly CN V3 via the foramen ovale, CN VII via the stylomastoid foramen, and the hypoglossal nerve — is one of the most clinically critical applications of neck MRI. PNS is visible as nerve enlargement, T2 signal abnormality, enhancement, and replacement of the normal fat within the neural foramina. CT is relatively insensitive for early PNS, making MRI the mandatory staging modality when PNS is clinically suspected.

Deep space infection characterisation: retropharyngeal and parapharyngeal space infections require MRI for: (a) defining the spatial extent across fascial boundaries; (b) distinguishing early cellulitis/phlegmon from frank abscess (rim-enhancing fluid collection requiring drainage); (c) identifying descending cervical mediastinitis; (d) detecting internal jugular vein thrombophlebitis (Lemierre syndrome). CT is faster for the acute setting; MRI is preferred for complex cases requiring precise compartmental mapping.

Salivary gland and parotid pathology: MRI is the primary modality for parotid gland tumour characterisation. Pleomorphic adenoma (T2-very bright, well-defined), Warthin tumour (T2-intermediate, bilateral 20%), mucoepidermoid carcinoma (T2-variable, infiltrative margins), and perineural spread from parotid malignancy along CN VII are all best assessed on MRI.

Lymph node characterisation: MRI DWI provides additional node characterisation beyond size criteria. Malignant nodes show restricted diffusion (low ADC); necrotic nodes show central T2-bright signal and rim enhancement; extracapsular spread (ECS) produces irregular enhancing margin with adjacent fat stranding visible on fat-suppressed T1 post-contrast. DWI is a meaningful complement to size- and morphology-based assessment.

Thyroid and parathyroid: for assessment of thyroid lesions extending below the level of the clavicle (substernal goitre), for parathyroid adenoma localisation when sestamibi is non-contributory, and for characterising thyroid-adjacent neck masses, MRI provides information not available from ultrasound or CT.

1.2 Intrinsic Limitations of the Generic Protocol

Swallowing motion — the primary technical limitation: swallowing produces rapid displacement of all anterior neck structures — the larynx, trachea, thyroid, and hyoid move superiorly 2–3 cm with each swallow. This is unavoidable and produces ghosting artefacts in the phase-encoding direction throughout the neck at every sequence. The artefacts are most severe on T2 sequences (long acquisition time) and are particularly problematic for the hypopharynx, larynx, and thyroid. Instructing the patient not to swallow is ineffective for extended acquisitions; antiperistaltic agents are not applicable to swallowing. Mitigation strategies are discussed in Section 5.

Mucosal disease requires a dedicated protocol: the generic neck protocol is not adequate as the primary protocol for staging mucosal head and neck cancers (oral cavity, oropharynx, hypopharynx, larynx). These require thin-slice (3 mm) axial and coronal T2 TSE sequences through the primary tumour site, specific attention to tumour-specific anatomical boundaries, and dedicated staging sequences for each subsite. The generic protocol may be used as a survey but must be supplemented with the specific oncological protocol. Child pages for each head and neck cancer subsite are planned in MRIninja.

Dental amalgam and dental metalwork: dental metallic restorations (amalgam fillings, dental implants, orthodontic hardware) produce extensive susceptibility artefacts on all sequences in the oral cavity, oropharynx, and lower face regions. At 3T, these artefacts extend further than at 1.5T and may obscure the submandibular space, the floor of the mouth, and the anterior oropharynx. This is particularly limiting for oral cavity tumour assessment. Titanium implants produce less artefact than stainless steel.

Coverage limit — brain and thorax: the generic neck protocol covers from the skull base to the thoracic inlet. Intracranial disease (intracranial extension of skull base tumours, brain parenchymal abnormalities relevant to the clinical context) requires a separate brain MRI protocol. Mediastinal extension of neck disease requires a dedicated chest MRI or CT.

When dedicated child protocols are required: oral cavity/oropharynx cancer staging; laryngeal/hypopharyngeal cancer staging; thyroid carcinoma cervical staging; salivary gland mass detailed assessment; perineural spread detailed mapping; brachial plexus assessment; cervical lymphoma; deep space infection with suspected mediastinitis; vascular pathology (carotid dissection — MRA supplement; vascular malformation); paraganglioma.

2. Main Clinical Indications

2.1 Standard Indications

Cervical lymphadenopathy is the single most common neck MRI indication. When ultrasound identifies abnormal cervical nodes and CT is equivocal or the question requires deeper tissue characterisation, MRI provides: nodal morphology on T2 (central necrosis = bright; lymphomatous = intermediate T2 with homogeneous signal); DWI-ADC for malignant vs reactive discrimination; perineural spread from adjacent primary; extracapsular spread features; and synchronous deep space pathology. The generic protocol is adequate for most lymphadenopathy assessments; a dedicated oncological protocol is required when a primary mucosal tumour is the clinical concern.

Neck mass assessment encompasses a broad range of pathologies: branchial cleft cyst, thyroglossal duct cyst, dermoid/epidermoid cyst, ranula, lymphangioma/venous malformation, lipoma, paraganglioma, schwannoma, neurofibroma, and metastatic adenopathy. MRI characterises the compartment of origin, the tissue composition (T2 signal, internal structure, enhancement pattern), and the relationship to adjacent critical structures (carotid artery, jugular vein, brachial plexus). The generic protocol provides adequate characterisation for most primary neck masses; dedicated paraganglioma protocol (T2*, DSA consideration) may be required for suspected vascular tumours.

Deep space infections — retropharyngeal abscess, parapharyngeal space abscess, Ludwig's angina, peritonsillar abscess, cervicothoracic mediastinitis — require MRI for compartmental mapping and abscess vs phlegmon distinction when CT is equivocal or when the surgeon requires precise anatomical detail before drainage. Post-contrast T1 with fat suppression is mandatory for this indication. The generic protocol with contrast is appropriate.

Salivary gland pathology: parotid masses, parotid duct calculi, submandibular gland lesions, and Sjögren's syndrome assessment are addressed by the generic protocol. Fat-suppressed T2 and post-contrast T1 characterise parotid lesions (pleomorphic adenoma, Warthin tumour, malignant parotid tumour). The proximity of parotid lesions to CN VII requires specific comment in the report about nerve involvement, facial nerve canal, and stylomastoid foramen status.

Thyroid and parathyroid pathology: substernal goitre extent; intrathyroidal vs extrathyroidal tumour extension (for planned thyroid surgery); parathyroid adenoma localisation (when sestamibi and ultrasound are non-localising, MRI adds parathyroid-specific T2 and contrast sequences). Dedicated parathyroid protocols supplement the generic neck protocol.

Post-treatment assessment: after head and neck surgery or radiotherapy, MRI assesses residual or recurrent disease, post-treatment oedema vs recurrence, and hardware-related complications. Post-radiation changes produce diffuse T2 hyperintensity and enhancement of the irradiated soft tissues that must be distinguished from recurrent tumour — a task requiring comparison with prior imaging and clinical correlation.

2.2 Urgent Red Flags Requiring Expedited or Emergency Imaging

3. Preparation Reference

Universal MRI safety screening belongs to the general MRI preparation page and is not repeated here.

3.1 Anatomy-Specific Preparation Items

Removal of all items near the neck and lower face: earrings, necklaces, neck chains, neckties, collars with metallic fasteners, hearing aids, removable dental prostheses, and neck brace hardware (unless firmly attached — check with the clinical team). Metallic items in the neck region produce susceptibility artefacts that can obscure the carotid bifurcation, the parotid gland, and the submandibular space. Ferromagnetic hearing aids must be removed; cochlear implants require specific MRI compatibility assessment.

Dental metalwork: dental amalgam restorations, crowns, bridges, and metallic implants in the lower jaw and upper jaw produce susceptibility artefacts on all sequences, particularly severe on fat-suppressed T1 post-contrast and on gradient echo sequences. This cannot be removed but must be documented. The extent of artefact is scanner- and field-strength-dependent; 3T produces larger susceptibility voids than 1.5T. For suspected oral cavity or oropharyngeal pathology with extensive dental metalwork, discuss with the referring clinician whether CT provides more reliable information for the specific clinical question.

Cardiac implanted electronic devices (CIEDs) and stimulators: vagal nerve stimulators, deep brain stimulators, and spinal cord stimulators near the neck require specific MRI compatibility check and device management protocol before any neck MRI. This is addressed in the general preparation page.

Swallowing preparation: the single most important patient-specific preparation for neck MRI. Instruct the patient to: (a) avoid eating or drinking for at least 2 hours before the examination (reduces swallowing frequency triggered by saliva); (b) not swallow during each sequence acquisition if instructed by the technologist; (c) breathe normally through the nose rather than the mouth. These measures reduce but do not eliminate swallowing artefacts.

Claustrophobia considerations: the neck MRI requires the patient to be fully inside the bore, often with a surface coil placed directly on the neck. The proximity of the coil to the face and neck is more claustrophobic than standard torso examinations. If claustrophobia is anticipated, pre-examination anxiolytic management should be arranged.

Prior surgery: a history of prior neck surgery (thyroidectomy, parathyroidectomy, neck dissection, carotid endarterectomy, tracheostomy) fundamentally alters the anatomical landmarks. Document surgical history and request operative notes when available. Titanium surgical clips (used in neck dissection) produce less artefact than stainless steel but are still visible as small susceptibility foci. Tracheostomy tubes (if in situ) produce susceptibility artefact in the tracheal region; the tube type must be documented and MRI compatibility confirmed.

3.2 Patient Positioning on the MRI System

Position: supine, head-first. The neck is maintained in a neutral position — neither hyperextended nor flexed. Hyperextension straightens the cervical curve and positions the anterior structures optimally, but may be uncomfortable for elderly patients or those with cervical spondylosis. A small pillow or foam wedge under the head achieves comfortable neutral positioning.

Coil selection: a dedicated neck or head-and-neck phased-array coil is mandatory. Options:

• Head-and-neck coil (combined head + neck channels): provides coverage from the skull base to the thoracic inlet in a single coil, using the posterior head channels and the anterior neck channels simultaneously. Optimal for neck pathology that requires simultaneous brain coverage (intracranial extension, skull base involvement).
• Dedicated neck coil (anterior cervical spine coil or bilateral neck coil): placed directly on the anterior neck, used in combination with the integrated table spine coil. Provides excellent SNR for the anterior neck compartments and lymph node chains. Better for pure neck pathology without skull base involvement.
• Flexible loop coil: for specific localised pathology (parotid, submandibular gland, isolated node), a small flexible surface coil placed directly over the region provides the highest SNR for focused assessment.

Centring: isocentre at the level of the hyoid bone — the anatomical midpoint of the neck from skull base to thoracic inlet. This centring optimises B0 homogeneity across the full neck FOV and positions the most diagnostically critical level (the hyoid level separates supra- and infrahyoid compartments) at isocentre. Verify on the three-plane localiser that: the skull base is within the superior coverage; the thoracic inlet (level of the sternal manubrium) is within the inferior coverage.

Immobilisation: a foam wedge or neck brace insert positions the head symmetrically and reduces voluntary head movement. For post-radiation patients with limited neck mobility, accept the available position without forcing extension.

Phase-encoding direction verification: the phase-encoding direction for axial sequences must be A-P (anterior-posterior) to displace swallowing artefacts anteroposteriorly rather than through the cervical lymph node chains laterally. This is the single most important phase direction choice for neck MRI.

4. Standard Protocol Design

4.1 Mandatory Core Sequences

4.2 Conditional Sequences

4.3 Rationale Summary Per Sequence

Axial T2 TSE (non-fat-suppressed, large FOV) is the primary anatomical survey sequence. The non-fat-suppressed T2 provides: spatial relationships between all compartments; the fascial planes as thin hypointense lines separating fat-containing spaces; the internal structure of neck masses; vascular flow voids (carotid and jugular as dark tubular structures); nerve structures (CN X, brachial plexus) as intermediate T2 tubular structures within the carotid sheath and posterior triangle. The non-fat-suppressed T2 is the sequence on which compartment assignment is made — is the mass in the parapharyngeal space (medial to the parotid, displacing the parapharyngeal fat medially)? Or in the masticator space (containing the pterygoid muscles and the mandibular ramus)? Or in the carotid space (the jugular vein and carotid are its components)?

Fat as an anatomical discriminator: the parapharyngeal fat — the T2-bright fat-containing space medial to the parotid and lateral to the pharyngeal mucosa — is the central anatomical landmark of suprahyoid neck space identification. The direction of displacement of parapharyngeal fat by a mass indicates its compartment of origin. This analysis is only possible on the non-fat-suppressed T2, not on fat-suppressed sequences.

Axial T2 fat-suppressed (STIR or SPAIR) reveals pathological signal within soft tissues, muscles, nodes, and salivary glands by suppressing normal background fat. On STIR/SPAIR-T2: reactive and malignant lymph nodes appear T2-bright against dark fat; abscess and phlegmon produce T2-bright signal change in the spaces; muscle oedema (from infection, trauma, or denervation) appears T2-bright; intralesional fluid (cysts, necrotic tumour) appears very bright. This fat-suppressed T2 is the most sensitive sequence for pathological change in the neck soft tissues.

Coronal T2 fat-suppressed provides the longitudinal view of the cervical nodal chains, allowing assessment of the full craniocaudal distribution of nodal disease in a single plane. The level II–V nodal chain from the skull base to the clavicle is visible simultaneously, enabling pattern recognition (e.g., bilateral II nodes = reactive/lymphoma; unilateral levels II–IV = head and neck squamous cell carcinoma metastasis pattern; supraclavicular IV–Vb = metastatic thyroid or lung). The coronal plane also provides the best view of the relationship between the thyroid gland, parathyroid, and the adjacent strap muscles.

Axial T1 non-fat-suppressed provides the anatomical T1 map and characterises T1-bright lesions: fat-containing structures (lipoma, dermoid, macroscopic fat in a branchial cleft cyst or thyroglossal cyst wall); haemorrhage (T1-bright in the subacute phase); proteinaceous cysts; melanin-containing lesions; and the normal T1-bright fatty marrow of cervical vertebrae. The T1 is the essential complement to the T2 for lesion characterisation: T2-bright + T1-bright = fat or haemorrhage; T2-bright + T1-dark = simple cyst or oedema; T2-intermediate + T1-intermediate = solid tumour.

DWI and ADC for neck soft tissues: lymph node characterisation is the primary DWI application. Malignant nodes (metastatic, lymphomatous) restrict diffusion; benign reactive nodes have higher ADC. Published ADC thresholds for cervical nodes vary widely across studies (malignant: ADC < 0.8–1.1 × 10⁻³ mm²/s; benign: > 1.1 × 10⁻³ mm²/s at b=1000), and considerable overlap exists, particularly for low-grade lymphoma vs reactive nodes. DWI is used for: (a) identifying the most restricted node as the optimal biopsy target; (b) characterising neck masses (abscess: peripheral restriction; solid tumour: variable; necrotic node: heterogeneous); (c) perineural spread in nerve trunk tissue (restricted diffusion in nerve replaced by tumour).

Post-contrast T1 fat-suppressed (3D axial) is mandatory when contrast is administered. It reveals: enhancement of tumours (solid, rim, septal); enhancement of infected spaces (phlegmon: diffuse enhancement; abscess: ring enhancement with non-enhancing centre); vascular structures (carotid, jugular, vertebral) as brightly enhancing tubular structures distinguishable from adjacent non-enhancing tissue; nodal enhancement (malignant nodes: heterogeneous; reactive: homogeneous; necrotic: rim only); CN V3, VII, XII perineural spread (asymmetric nerve enhancement and enlargement).

Post-contrast T1 fat-suppressed coronal provides the longitudinal complementary view, particularly for lymph node chains, thyroid/parathyroid enhancement, and any enhancement extending from the neck to the skull base or into the mediastinum.

4.4 Sequence Matching and Cross-Sequence Consistency

Axial T2, T1, and post-contrast T1 must use the same slice geometry (identical level positions, thickness, FOV) to enable direct lesion comparison across sequences. The axial sequences are planned together from the sagittal localiser, ensuring each level is represented in all three sequences. If the post-contrast T1 3D is isotropic, its reformatted axial slices must correspond to the 2D T2 axial slice positions for accurate comparison.

For serial neck MRI (treatment monitoring, surveillance), the axial slice geometry and coverage must be reproduced at each follow-up. The hyoid bone level is the most reliable anatomical reference for matching — it is visible on the sagittal localiser and does not change position relative to the cervical vertebrae.

4.5 Fat Suppression in Neck MRI

STIR vs SPAIR for neck T2: STIR provides B0-independent fat suppression and is the more reliable technique for the neck, where B0 inhomogeneity from the air-tissue interface of the trachea and the adjacent lungs produces spectral fat saturation failure in the lower neck and anterior throat. At 1.5T, STIR is the standard T2 fat suppression technique for neck MRI. At 3T, SPAIR provides higher SNR than STIR when shimming is adequate; STIR is the fallback when SPAIR fails.

Dixon for post-contrast T1: Dixon fat-water separation for post-contrast T1 provides B0-independent fat suppression for the neck post-contrast sequence, addressing the same B0 challenge as STIR for T2. Dixon post-contrast T1 is preferred at 3T for neck MRI. The water-only post-contrast image provides the clean fat-suppressed background for enhancement detection.

Non-fat-suppressed T2 is required: as discussed in Section 4.3, the non-fat-suppressed T2 is mandatory for neck space analysis based on fat displacement patterns. It must not be replaced by the fat-suppressed T2 alone.

STIR contraindication post-gadolinium: absolute, as throughout the MRIninja protocol series.

5. Optimisation Strategy

5.1 Artifact Reduction by Source

Swallowing motion is the dominant artefact source in neck MRI and cannot be fully eliminated. The larynx, hyoid, and thyroid move 2–3 cm superiorly with each swallow. Ghost artefacts from swallowing appear in the phase-encoding direction, producing bright bands that traverse the neck at the level of the larynx and hypopharynx. Mitigation: (a) A-P phase encoding (moves ghosts out of the lateral nodal chain path); (b) patient instruction — "do not swallow, breathe through your nose"; (c) antiperistaltic agents do not affect swallowing; (d) shorter individual sequence times reduce the probability of multiple swallowing events during a single acquisition; (e) dedicated neck coils with tighter dimensions reduce the FOV, reducing the ghost amplitude from structures outside the FOV.

Dental amalgam susceptibility: amalgam fillings produce susceptibility artefacts extending 1–3 cm in all directions at 1.5T and 2–5 cm at 3T, obscuring the oral cavity, floor of mouth, and anterior oropharynx. This cannot be corrected. Document in the report; consider 1.5T for oral cavity-specific assessment; consider CT for these regions if MRI is obscured.

Vascular pulsation artefact: the carotid arteries and internal jugular veins produce pulsatile flow artefacts in the phase-encoding direction. These appear as bright bands crossing the neck at the vessel level. Mitigation: (a) saturation bands superior and inferior to the imaging volume suppress inflowing blood signal; (b) flow compensation (gradient moment nulling) on T2 reduces vascular pulsation artefacts; (c) A-P phase encoding reduces the severity in the lateral neck.

Susceptibility from tracheal air-tissue interface: the trachea produces local B0 disturbance from the air column. This causes fat suppression failure (SPAIR/CHESS) in the anterior neck at the tracheal level. Mitigation: STIR instead of SPAIR for the T2-FS sequence (B0-independent); Dixon for T1-FS post-contrast.

Susceptibility near thyroid gland: the thyroid gland sits adjacent to the trachea with significant air-tissue interfaces nearby. Fat suppression failure in the thyroid region is common with spectral methods. STIR or Dixon are preferred.

Motion from breathing: respiratory motion is less severe in the neck than in the thorax or abdomen, but visible in the lower neck (below the thoracic inlet). If the clinical question involves the lower neck/thoracic inlet, respiratory compensation or breath-hold acquisitions may be considered.

5.2 Protocol Efficiency and Throughput

A complete neck soft tissue MRI with contrast — T2 non-FS + T2-FS axial + T2-FS coronal + T1 axial + DWI + post-contrast T1-FS axial + coronal — requires approximately 35–45 minutes at 3T.

For a focused neck mass assessment without full nodal staging, a 20–25 minute protocol — T2 non-FS + T2-FS + T1 + DWI + post-contrast T1-FS — is adequate for most compartmental assessments.

3D isotropic T2 (SPACE/CUBE) can replace 2D axial + coronal in a single 6–8 minute acquisition, providing MPR in all planes. This is useful for complex mass characterisation requiring multiplanar views, but requires good patient cooperation (longer ETL → more motion sensitivity than 2D).

5.3 Field Strength Considerations

3T is preferred for neck MRI for: (a) superior DWI quality for lymph node characterisation; (b) higher spatial resolution for small nerve assessment (perineural spread on small cranial nerves); (c) faster acquisitions enabling more complete protocols; (d) better fat-suppressed T2 contrast when shimming is adequate.

1.5T is preferred or equivalent for: (a) patients with extensive dental metalwork (less susceptibility artefact); (b) patients with metallic surgical clips in the neck (less artefact radius); (c) patients who had prior radiotherapy with skin applicators (potential heating near metal); (d) when B0 homogeneity issues at 3T produce unacceptable fat suppression failure that cannot be corrected with Dixon.

6. Contrast Use Principles Specific to Neck Soft Tissue MRI

6.1 Non-Contrast Standard Protocol — Sufficient For

Non-contrast neck MRI (T2 non-FS + T2-FS + T1 + DWI) is diagnostically adequate for:

• Simple cyst characterisation (branchial cleft cyst, thyroglossal duct cyst — cysts do not enhance; T2-bright + T1-dark + no DWI restriction is diagnostic)
• Lipoma and dermoid cyst (fat signal on T1 + T2; no enhancement expected)
• Vascular malformation initial characterisation (phleboliths on SWI; T2-bright; no enhancement mandatory for diagnosis)
• Lymphangioma/macrocystic lymphatic malformation (T2-bright multiloculated; no contrast required for diagnosis)
• Nerve sheath tumour screening (schwannoma: T2-bright, T1-intermediate; neurofibroma: target sign on T2)
• Post-treatment surveillance when the clinical question is size change rather than enhancement pattern

6.2 Gadolinium Indicated — Region-Specific Contexts

Gadolinium-enhanced sequences are required or strongly useful for:

• Deep space infection assessment: distinguishing phlegmon (diffuse enhancement) from abscess (rim enhancement with non-enhancing centre) is the most critical clinical application; without contrast this distinction is unreliable
• Malignant node characterisation: enhancement pattern (solid vs ring vs heterogeneous); extracapsular spread (irregular enhancing margin); internal necrosis vs non-necrosis
• Neck mass with uncertain tissue origin: enhancement of solid components; vascular architecture of paraganglioma (intense early enhancement); nerve sheath tumour enhancement
• Perineural spread assessment: asymmetric nerve enhancement along CN V, VII, XII; foraminal enhancement; cavernous sinus involvement
• Salivary gland tumours: enhancement pattern characterises pleomorphic adenoma (delayed), Warthin tumour (minimal), mucoepidermoid carcinoma (solid)
• Post-treatment surveillance with suspected recurrence: enhancing recurrent tumour vs non-enhancing post-radiation fibrosis
• Thyroid/parathyroid: parathyroid adenoma (early arterial enhancement); thyroid carcinoma vascular invasion

6.3 Post-Contrast Acquisition Timing

Standard post-contrast T1-FS (axial + coronal) at 3–5 minutes after injection provides equilibrium phase enhancement characterisation for neck soft tissue pathology. No specific arterial phase acquisition is required for the generic neck protocol; arterial phase may be added for suspected paraganglioma (intense arterial phase enhancement) or for vascular malformation assessment. Delayed phase (10–15 min) may add information for suspected perineural spread (continued enhancement on delayed phase) and for parathyroid adenoma.

7. Reporting Essentials

7.1 Interpretation Framework

Neck MRI reporting follows a systematic compartmental approach. The report should identify: (1) which neck space the pathology is in; (2) the intrinsic characteristics of the lesion; (3) the local extent (crossing fascial boundaries, vascular encasement, neural involvement); (4) the nodal status; (5) any distant or contiguous extension.

Compartmental analysis: identify the primary compartment using the non-fat-suppressed T2. The parapharyngeal fat displacement direction localises the primary compartment:

• Fat displaced medially → parotid space lesion
• Fat displaced laterally → parapharyngeal space lesion
• Fat displaced posteriorly → masticator space lesion

Broad diagnostic axes:

• Cystic vs solid (T2 signal, lack of enhancement)
• Benign vs malignant (margin, restricted diffusion, necrosis, bone invasion)
• Inflammatory vs neoplastic (clinical context, fever, adjacent fat stranding)
• Unilocular vs multilocular (lymphangioma vs abscess vs cystic tumour)
• Enhancing vs non-enhancing components
• Nodal vs primary mass (node: reniform shape, hilum if benign, cortical thickening)

7.2 Mandatory Reporting Checklist

Technical quality:

• [ ] Field strength; coil
• [ ] Swallowing artefacts: none / mild / moderate (affected region)
• [ ] Fat suppression: uniform / failure in (region)
• [ ] Dental metalwork artefact: absent / present (affected region)

Primary lesion (if identified):

• [ ] Compartment / space localisation (parapharyngeal, carotid, masticator, parotid, submandibular, visceral, retropharyngeal, perivertebral, posterior cervical)
• [ ] Size (three dimensions)
• [ ] T2 signal: bright (cystic/oedema) / intermediate (solid) / dark (fibrosis/calcification)
• [ ] T1 signal: bright (fat/haemorrhage) / dark (fluid/tumour)
• [ ] Enhancement pattern: none / homogeneous / rim / heterogeneous
• [ ] DWI: restricted / not restricted; ADC value if measured

Lymph nodes (bilateral systematic by level I–VII):

• [ ] Level I (submental, submandibular)
• [ ] Level II (upper jugular)
• [ ] Level III (mid-jugular)
• [ ] Level IV (lower jugular)
• [ ] Level V (posterior triangle)
• [ ] Level VI (prelaryngeal, pretracheal)
• [ ] Level VII (superior mediastinal — if coverage permits)
• [ ] Each abnormal node: size, necrosis, ECS features, DWI restriction

Critical structures:

• [ ] Carotid artery: encased / displaced / normal; flow void intact
• [ ] Internal jugular vein: patent / thrombosed / compressed
• [ ] Vertebral arteries: symmetric / asymmetric
• [ ] Cervical spine/vertebral bone marrow: normal / signal change

Specific anatomical structures (as clinically relevant):

• [ ] Thyroid gland: normal / enlarged / mass
• [ ] Parotid glands: symmetric / asymmetric / lesion
• [ ] Submandibular glands
• [ ] Cranial nerve VII, XII, V3 courses (for parotid/skull base lesions)
• [ ] Retropharyngeal space: normal / fluid / mass

7.3 Structured Reporting

Reports must include: Indication (mass, lymphadenopathy, infection, oncological staging); Technique (field strength, coil, sequences, contrast: agent, dose, timing); Comparison (prior neck MRI or CT); Findings (compartmental systematic review); Impression (primary differential diagnosis, BI-RADS equivalent classification if applicable, staging information if oncological); Recommendations (biopsy target, follow-up interval, additional imaging); Limitations (swallowing artefacts, dental metalwork, coverage limits).

7.4 Incidental Findings — Clinical Decision Framework

Usually benign: thyroid cysts < 1 cm; mucous retention cysts in the paranasal sinuses; neck lipoma < 3 cm; enlarged but morphologically benign lymph nodes (fatty hilum, short-axis < 10 mm); thyroglossal duct cyst remnant in the midline.

May require follow-up: thyroid nodule ≥ 1 cm with solid components (recommend neck ultrasound with TI-RADS assessment); indeterminate neck mass with stable size and benign MRI features (follow-up in 6 months); single borderline node (short-axis 10–15 mm, no necrosis, no ECS) in a patient without known malignancy.

Urgent communication: unsuspected deep space infection with abscess and airway compromise; unsuspected carotid occlusion or vertebral artery dissection; unsuspected malignant lymphadenopathy in a patient referred for benign indication; cord compression from vertebral body involvement visible on the inferior coverage; new cranial nerve palsy findings on imaging.

8. MRI Technologist Pearls

8.1 Sequence Order Logic

1. Three-plane localiser
2. Axial T2 non-FS ← primary compartmental survey; anatomical map for all subsequent planning
3. Axial T2-FS (STIR) ← pathological signal; before contrast
4. Coronal T2-FS ← nodal chain overview
5. Axial T1 non-FS ← T1 characterisation; before contrast
6. DWI ← before contrast; node diffusion characterisation
7. Contrast injection
8. Axial T1-FS post-contrast (3D) ← primary post-contrast assessment
9. Coronal T1-FS post-contrast

The T2 non-FS is acquired first — it is the most diagnostically critical for compartment analysis and is motion-robust. DWI is acquired before contrast to avoid T1 enhancement contaminating DWI sequences with intravascular gadolinium enhancement effects on perfusion-sensitive low-b images.

8.2 Positioning Tricks

The chin position critically affects the retropharyngeal and retrostyloid regions. Mild chin tuck (slight neck flexion) straightens the pharyngeal wall and opens the retropharyngeal space — useful for retropharyngeal pathology assessment. Conversely, neutral head position provides the best view of the thyroid and lower neck. For most generic neck MRI, neutral position is optimal.

For parotid gland assessment, a small foam pad behind the ear ipsilateral to the lesion slightly rotates the head away from the affected side, bringing the deep lobe of the parotid further from the mandibular ramus and slightly reducing dental metalwork artefact impact.

For patients with tracheostomy in situ: position and centre the coil high enough to clear the tracheostomy tube; the tube should be outside the primary FOV if possible.

8.3 Fast Salvage Protocol

Approximately 14 minutes — covers the minimum diagnostic requirements for compartmental analysis, lesion characterisation, and enhancement assessment.

8.4 Common Avoidable Errors

9. Quality Control Checklist

• [ ] Skull base included in superior coverage
• [ ] Supraclavicular (level IV/V) nodes within inferior coverage
• [ ] Axial phase encoding direction: A-P (verified on first sequence)
• [ ] T2 non-fat-suppressed: acquired and interpretable
• [ ] STIR: fat suppression quality assessed — uniform / failure at (level)
• [ ] STIR acquired before gadolinium injection
• [ ] Coronal T2-FS covers bilateral nodal chains from skull base to thoracic inlet
• [ ] DWI: ADC map generated and available
• [ ] Post-contrast T1-FS: both axial and coronal acquired at appropriate timing
• [ ] Swallowing artefacts: acceptable / moderate (document) / severe (note limitation)
• [ ] Dental metalwork artefact: absent / present (document affected region)
• [ ] Bilateral carotid arteries and jugular veins visible on axial sequences
• [ ] Correct laterality labelling (left and right nodes, parotid glands)
• [ ] Comparison with prior imaging performed

11. Evidence Gaps and Ongoing Debate

DWI ADC threshold for malignant vs benign cervical nodes: despite numerous published studies, no universally accepted ADC cutoff has been validated across multiple institutions, scanners, and b-value combinations. The optimal b-value combination (b=0 or b=50 as the reference; b=800 or b=1000 as the diagnostic value) has not been formally standardised for cervical nodal DWI. The ESHNR and European Radiology community have not produced a cervical node DWI consensus comparable to the EUSOBI breast DWI consensus.

3T vs 1.5T for perineural spread: 3T provides nominally superior spatial resolution for small nerve assessment (CN VII in the parotid, CN XII in the carotid space), but no prospective study has demonstrated improved diagnostic accuracy for perineural spread detection at 3T versus optimised 1.5T acquisition.

Role of ultrafast DCE for neck vascular tumours: paragangliomas and highly vascular neck tumours show characteristic early arterial enhancement. Whether ultrafast DCE (< 10 s/phase) provides clinically significant additional information over standard post-contrast T1 for the differential diagnosis of neck vascular lesions has not been formally evaluated.

AI-assisted lymph node characterisation: machine learning models for cervical node malignancy classification from MRI and DWI have been described in single-institution retrospective series. No prospective validation or FDA-cleared clinical tool exists at the time of writing.

Abbreviated protocol for neck surveillance: for post-treatment surveillance after head and neck cancer, the full protocol is rarely needed at every follow-up. Whether an abbreviated protocol (T2-FS + T1 post-contrast only, without DWI and non-FS T2) is diagnostically equivalent for recurrence detection has not been prospectively validated.

Related salivary gland master: Parotid Glands MRI — Generic Standard Protocol.