MRI Hand – Generic Standard Protocol

1. Executive Summary

The hand — defined in this protocol as the anatomical unit from the metacarpal shafts through the metacarpophalangeal (MCP) joints and the carpometacarpal (CMC) joints, including the thumb at all levels — presents a distinct set of imaging challenges compared to both the wrist and the individual fingers. It is a region of densely packed small joints, multiple overlapping tendon compartments, and thin capsuloligamentous structures, all contained within a structure no more than 2–3 cm in thickness, that must be imaged with spatial resolution approaching that of the wrist while covering a substantially larger field of view.

MRI of the hand is the primary advanced imaging modality when the clinical question involves soft tissue characterisation beyond the capability of radiographs and ultrasound — most importantly in inflammatory arthropathy at the MCP and CMC joints, post-traumatic assessment of the thumb CMC joint, soft tissue masses of the palm and metacarpal spaces, nerve and vessel pathology of the palm, and bone marrow abnormality of the metacarpals. It fills the gap between the wrist protocol (proximally) and the finger protocol (distally) without duplicating either.

The generic hand MRI protocol covers the metacarpals, all five MCP joints, the five CMC joints, the thenar and hypothenar musculature, the palmar spaces (thenar, hypothenar, mid-palmar), the digital flexor and extensor tendons from their musculotendinous junctions to the proximal phalanx level, and the palmar neurovascular bundles.

1.1 Core Strengths

Inflammatory arthropathy at the MCP joints is the most common clinical indication for hand MRI. Rheumatoid arthritis (RA) preferentially targets the MCP joints — particularly the second and third MCP joints bilaterally and symmetrically — producing synovitis, tenosynovitis, periarticular bone marrow oedema (osteitis), and bone erosions that are detectable by MRI months to years before radiographic change [3, 4]. The RAMRIS (Rheumatoid Arthritis MRI Scoring System) — the validated OMERACT outcome measure for RA clinical trials — specifically targets the MCP joints and wrist as the primary MRI scoring sites [5]. No other imaging modality (radiographs, ultrasound) matches MRI for the detection of subcortical erosions and bone marrow oedema in early RA.

Thumb CMC joint assessment: the first CMC joint (trapeziometacarpal joint) is the most common site of osteoarthritis in women over 50, and the thumb MCP joint is the primary site for gamekeeper's/skier's thumb (ulnar collateral ligament injury). Both are well characterised on hand MRI. The UCL of the thumb MCP joint — particularly distinguishing a Stener lesion (avulsed ligament entrapped by the adductor aponeurosis) from a simple partial tear — is the surgical decision point, and MRI is the definitive investigation when the clinical examination is equivocal [7].

Soft tissue masses of the palm and metacarpal spaces: giant cell tumour of the tendon sheath (GCTTS), ganglia of the CMC or MCP joints, lipomas, fibromata, glomus tumour of the palmar digits, and other soft tissue masses are characterised on MRI for anatomical relationships and tissue composition. MRI determines surgical approach and prognosis.

Bone marrow pathology: metacarpal stress fractures in athletes, metacarpal bone contusions, and metastatic disease involving the metacarpals are demonstrated on STIR and fat-suppressed sequences before radiographic abnormality. For sequence-level protocol optimisation, vendor terminology and artefact management, see the dedicated MRIninja page STIR Sequence.

Palmar space infections and compartment assessment: deep space infections (thenar space, mid-palmar space, hypothenar space) require MRI for defining anatomical extent, guiding surgical drainage, and identifying flexor tendon sheath involvement. This is a time-sensitive clinical application.

1.2 Intrinsic Limitations of the Generic Protocol

Resolution limits for MCP joint cartilage: the articular cartilage of the MCP joints is approximately 1–1.5 mm thick. At the spatial resolution of the generic hand MRI protocol (0.3–0.5 mm in-plane at 3T), thin focal chondral defects at the MCP joint head may be visible but cannot be reliably graded. A dedicated cartilage protocol with 3D isotropic sequences and thin sections is required for quantitative cartilage assessment.

Individual finger exclusion: this protocol does not provide the spatial resolution or plane prescription required for individual finger diagnosis — particularly the collateral ligaments, volar plate, and flexor pulley system of the lesser fingers. These require the dedicated MRI Finger protocol, which uses a smaller FOV centred on the individual finger.

Thumb UCL partial tear assessment: while the generic protocol demonstrates complete UCL tears and Stener lesions reliably, partial UCL tears — particularly of the accessory collateral ligament — may require dedicated sequences or MR arthrography in clinical settings where complete characterisation is operatively important.

Off-isocentre challenges: hand MRI shares the same B0 inhomogeneity limitations as the wrist and elbow in the supine arm-at-side position. Dixon fat suppression mitigates but does not fully eliminate this limitation for a full-hand FOV (14–18 cm).

Inflammatory arthropathy scoring: the RAMRIS system is designed for a combined wrist + MCP joint protocol. When the hand MRI is performed separately from the wrist, the full RAMRIS scoring requires adding the wrist examination. This is discussed in Section 7.

When dedicated child protocols are required: MR arthrography for thumb MCP UCL characterisation; individual finger protocols for PIP and DIP joint assessment; dedicated palmar nerve protocol for suspected nerve tumour or peripheral nerve sheath tumour; post-operative protocol with MARS sequences.

2. Main Clinical Indications

2.1 Standard Indications

Inflammatory arthropathy — MCP joints is the primary indication for hand MRI in clinical rheumatology. Rheumatoid arthritis, psoriatic arthritis, reactive arthritis, and gout all produce characteristic patterns of MCP joint involvement that are distinguishable on MRI when combined with clinical and serological data. The early MRI findings — bone marrow oedema preceding erosion formation by months — are the key argument for MRI over radiography in early disease. For RA disease activity monitoring in clinical trials, the combined hand + wrist MRI RAMRIS protocol is the standard [5]; in clinical practice, the hand MRI protocol alone (covering MCP joints and CMC joints) provides the primary MCP joint assessment.

Thumb CMC osteoarthritis and instability: the trapeziometacarpal joint is the second most common site of osteoarthritis in the hand (after the distal interphalangeal joints of the fingers). When clinical and radiographic assessment is inconclusive, MRI characterises the degree of cartilage loss, ligamentous laxity, synovitis, and periarticular changes. For surgical planning (trapeziectomy, ligament reconstruction), MRI provides the pre-operative anatomical map.

Gamekeeper's/Skier's thumb — UCL injury of the thumb MCP joint — is the most important acute indication for hand MRI. The critical clinical and surgical distinction is between a Stener lesion (complete UCL avulsion with the torn ligament lying superficial to the adductor aponeurosis, requiring operative repair) and an incomplete or partial tear (managed conservatively). MRI sensitivity for Stener lesion detection approaches 96% at 3T with dedicated sequences [7]. The generic protocol is sufficient for this indication in most clinical settings.

Palmar soft tissue masses: ganglia (dorsal and volar), GCTTS (tenosynovial giant cell tumour), lipoma, fibroma, glomus tumour, and epidermoid inclusion cysts are the most common palmar soft tissue masses. MRI characterises their tissue composition, anatomical relationships to tendons and nerves, and extent in the palmar spaces, informing the surgical approach. The generic non-contrast protocol is sufficient for diagnosis in most cases; post-contrast sequences are required for vascular malformations and suspected solid neoplasms.

Palmar space infection: deep palmar space infections — thenar space, mid-palmar space, hypothenar space — are surgical emergencies that require urgent MRI to map the extent of the abscess, identify flexor tendon sheath involvement, and plan drainage approach. Post-contrast sequences are mandatory for this indication.

Metacarpal stress fractures and bone marrow pathology: stress fractures of the metacarpal shafts (particularly in racquet sport athletes), metacarpal bone contusions, and inflammatory bone marrow changes are detected on STIR and fat-suppressed sequences. The generic protocol is adequate.

Post-traumatic assessment without specific localisation: for patients with diffuse hand pain after trauma and normal or inconclusive radiographs, the generic hand MRI provides a comprehensive survey of all metacarpals, MCP joints, and CMC joints, enabling detection of occult fractures, ligamentous injuries, and soft tissue damage.

2.2 Urgent Red Flags Requiring Expedited or Emergency Imaging

The hand does not generate life-threatening emergencies, but the following require expedited imaging.

3. Preparation Reference

Universal MRI safety screening, contraindication assessment, and IV access documentation are covered in the general MRI preparation page and are not repeated here.

3.1 Anatomy-Specific Preparation Items

Metacarpal inclination and hand position: unlike the wrist (where neutral deviation is mandatory), the hand has a natural palmar arch that affects the coronal plane in the true anatomical position. The hand should be positioned flat — palm down or palm up — to minimise this arch and produce a genuinely flat coronal plane for the metacarpals. A flat foam pad under the dorsum or palm helps maintain this position.

Jewellery and rings: all metallic rings, bracelets, and hand jewellery must be removed. Metal rings produce susceptibility artefact that distorts the adjacent MCP joint and metacarpal head assessment.

Nails with metallic polish or gel coatings: some metallic nail polishes contain ferromagnetic pigments. These should be removed before scanning when the hands are being imaged, as they can produce focal susceptibility artefacts at the fingertip level that propagate into the adjacent phalanges.

Wound dressings, splints, and casts: any metallic component in wound dressings, dynamic splints, or casts over the hand must be removed or verified as MR-compatible. Plaster of Paris casts are generally compatible; dynamic splints often contain metallic springs.

Prior surgery: document history of metacarpal plating, K-wire fixation, MCP joint arthroplasty, CMC joint arthroplasty, or soft tissue reconstruction. Hardware type determines artefact degree and whether MARS sequences are needed.

Contrast-related preparation: for suspected infections, soft tissue tumours, and inflammatory arthropathy requiring synovitis quantification, IV access should be established before positioning.

3.2 Patient Positioning on the MRI System

Standard position: the same positioning options as the wrist apply here with the same considerations — the hand MRI is acquired in either the supine arm-at-side position or the Superman (prone, arm overhead) position.

Supine, arm at side (preferred for comfort and most clinical departments): the patient lies supine, feet first, with the affected hand extended along the side of the body, palm down (pronation), fingers extended and flat on the coil. This is more comfortable than Superman and achieves acceptable image quality at 3T with Dixon fat suppression. The hand is positioned approximately 20–30 cm lateral to isocentre, with the same B0 inhomogeneity limitations as the wrist.

Superman position (prone, arm overhead): produces near-isocentre positioning and optimal B0 homogeneity. Preferred when fat suppression quality is critical and the patient can tolerate the position for 25–35 minutes. Comfort is the main limiting factor.

Thumb position: the thumb naturally extends perpendicular to the palm in pronation (thumb pointing upward). This places the thumb in a different plane from the palm — critical for planning thumb-specific sequences. The dedicated thumb MCP plane prescription must account for the individual thumb orientation (see Section 4.6).

Coil selection: a dedicated hand/wrist phased array coil or a multichannel wrist coil (8-channel minimum) is mandatory. The coil must cover the full hand from the metacarpal bases to the MCP joint line — usually requiring a 14–18 cm FOV. A general-purpose flex coil is acceptable when a dedicated hand coil is not available, but provides lower SNR. Body coil or spine coil is inadequate for diagnostic hand MRI.

Centering: isocentre at the third MCP joint — the central joint of the hand — or at the midpoint of the metacarpal shafts for a full-hand acquisition. Adjust proximally if the primary question involves the CMC joints.

Immobilisation: foam padding medially and laterally between the fingers prevents involuntary splaying. A flat rigid foam pad under the dorsum maintains the hand flat. Instruct the patient to avoid finger movement during the entire examination.

Common positioning errors: - Fingers not flat: the natural metacarpal arch creates apparent obliquity in the coronal plane; foam pad under the dorsum corrects this - Thumb allowed to flex: the thumb MCP joint moves out of the coronal plane; maintain thumb extension throughout - Insufficient proximal coverage: the CMC joints (particularly the thumb CMC) are excluded when the FOV is centred too distally - Rings not removed: susceptibility artefact at the MCP joint level

4. Standard Protocol Design

The standard hand MRI protocol uses the same three-plane orthogonal structure as the wrist and elbow. The coronal plane is the primary diagnostic plane for the MCP joints, the metacarpal shafts, and the CMC joints. The axial plane covers the tendons, neurovascular bundles, and palmar spaces. The sagittal plane provides longitudinal tendon and joint assessment.

4.1 Mandatory Core Sequences

4.2 Conditional Sequences

4.3 Rationale Summary Per Sequence

Coronal PD-FS is the single most diagnostically important sequence for the hand, directly equivalent to the coronal PD-FS in the wrist and shoulder protocols. In the coronal plane, the hand MRI displays: - All five MCP joints simultaneously — joint space width, synovitis signal, cartilage interface, periarticular soft tissues - The heads of all five metacarpals — for bone marrow oedema (osteitis in RA) and subchondral erosions - The bases of all five metacarpals and the CMC joints — particularly the first CMC (trapeziometacarpal) - The interosseous muscles in their longitudinal extent - The palmar aponeurosis and the hypothenar fat pad

Fat suppression is mandatory for the same reason as in all MSK protocols: the periarticular fat at the MCP joint margins and the intraosseous fat within the metacarpal shafts would overwhelm the fluid signal of joint effusion and synovitis without suppression. PD weighting (TE 20–40 ms) is preferred over T2 for the MCP joint structures because the ligamentous structures around the MCP joints contain collagen fibres that are subject to magic angle effect at short TE, but the intermediate TE of PD provides the best SNR-to-artefact compromise.

Coronal T1 provides the essential complement for: (i) bone erosion characterisation — cortical break and subchondral trabecular disruption are best seen on T1; (ii) T1 signal characterisation — distinguishing fat and haemorrhage from fluid; (iii) anatomical landmark definition. In RA specifically, the T1 coronal is required for the RAMRIS erosion score: erosions are defined as cortical breaks of ≥ 1 mm visible on at least two orthogonal planes on T1-weighted images.

Axial PD-FS is the primary plane for: - The flexor tendon sheaths of all four fingers and the thumb (cross-sectional assessment for tenosynovitis — fluid thickness exceeding the tendon diameter is the diagnostic threshold) - The lumbrical muscles (anomalous lumbrical as a cause of carpal tunnel syndrome; lumbrical muscle mass) - The palmar and dorsal interosseous muscles - The superficial palmar arch and the digital neurovascular bundles - Palmar space anatomy (thenar, mid-palmar, hypothenar spaces) — critical for deep space infection assessment - Ganglion and soft tissue mass anatomical relationship to tendons and nerves

Sagittal PD-FS provides longitudinal profile of: - The flexor tendons at the MCP level — A1 pulley and digital flexor sheaths at the proximal entrance - The palmar plate of the MCP joints - The metacarpal shafts for cortical integrity - Extensor tendons on the dorsal surface

STIR is the mandatory B0-independent bone marrow screening sequence. For the hand, it is critical for: metacarpal stress fractures, RA bone marrow oedema (osteitis) as a RAMRIS feature, gout-related bone changes, and periosteal reaction assessment. As in all other MRIninja protocols, STIR must always be acquired before gadolinium injection.

4.4 Sequence Matching and Cross-Sequence Consistency

The three orthogonal planes must be truly orthogonal to the hand anatomy — prescribed from the metacarpal bone axes, not from the body axes. Pre- and post-contrast T1-FS sequences must use identical prescription for meaningful enhancement comparison, which is particularly important in RA RAMRIS scoring where the enhancement pattern is compared to the pre-contrast baseline. For serial RAMRIS studies, exact reproduction of the coronal FOV position, angulation, and slice thickness is required for reproducible scoring; these parameters should be saved as a protocol preset and verified at each examination.

4.5 Fat Suppression — Region-Specific Technical Considerations

Fat suppression at the hand shares the same off-isocentre B0 challenges as the wrist and elbow (see those protocol pages for the full physical explanation). For the hand, the challenge is slightly less severe than the wrist because the FOV is larger (14–18 cm vs. 8–12 cm), which partially averages the B0 inhomogeneity over the larger volume.

Dixon fat suppression is the recommended choice for all hand MRI PD-FS sequences. Its B0-independent mechanism produces consistent fat suppression regardless of off-isocentre position, and it provides simultaneous fat-only and water-only images that enable fat-fraction quantification of the thenar and hypothenar muscles. The 2025 study of symptomatic-free subjects using mDixon hand MRI confirmed that Dixon-based fat suppression provides reliable RAMRIS-compatible images without contrast in the standard supine position [6].

SPAIR: acceptable alternative when Dixon is unavailable. Performs better than CHESS at off-isocentre positions.

STIR: mandatory backup; always included as the primary bone marrow screening sequence. At the hand, STIR is particularly valuable because the periarticular fat at the MCP joints provides a high-contrast background against which subtle periarticular bone marrow oedema (RAMRIS osteitis) is detected.

STIR contraindicated post-gadolinium — the same absolute rule applies here as in all other protocols.

Post-contrast T1-FS: when gadolinium is used for synovitis assessment (RAMRIS), the post-contrast sequence requires fat suppression to distinguish enhancing synovium from periarticular fat. Dixon-based fat suppression is the first choice; SPAIR is acceptable. The post-contrast T1 must be parameter-matched to the pre-contrast T1 for subtraction and comparison.

5. Optimisation Strategy

5.1 Artifact Reduction by Source

Off-isocentre B0 inhomogeneity and fat suppression failure: the same off-isocentre B0 problem as the wrist applies here, amplified by the larger FOV of the hand protocol. Spectral fat saturation (CHESS/ChemSat) shows asymmetric failure at the radial and ulnar edges of the full-hand FOV in the supine position. Dixon fat suppression is the appropriate solution. In the OMERACT RA hand MRI study using mDixon at 1.5T and 3T in the supine position, Dixon provided adequate fat suppression across the full hand FOV in virtually all subjects [6].

Finger motion and hand tremor: the most common cause of degraded hand MRI. Minor involuntary finger movements during acquisition — particularly during STIR and post-contrast sequences (the longest acquisitions) — produce motion artefacts that degrade the entire examination. Practical measures: explicit verbal instruction to hold the fingers motionless; foam padding between fingers; acquiring the most critical sequence (coronal PD-FS) first; keeping long-TR sequences as short as possible.

Chemical shift artefact at metacarpal-fat interfaces: the same principle as at the wrist. At the metacarpal head–periarticular fat interface on coronal PD-FS, chemical shift displacement can simulate a subchondral or cortical abnormality. Wider receiver bandwidth (300–400 Hz/px at 3T) reduces this displacement.

Susceptibility artefact from metallic implants: metacarpal plate fixation, K-wires, and MCP joint arthroplasty all produce susceptibility artefact. The degree depends on material and proximity to the joint of interest. MARS sequences (STIR rather than spectral FS, wider bandwidth, reduced TE) reduce but do not eliminate susceptibility effects. K-wires and titanium plates produce less artefact than stainless steel.

Magic angle at the flexor tendons: the flexor tendons within the palmar sheaths cross at approximately 55° to B0 at certain points in the carpal tunnel region, producing PD signal increase that can simulate tendinopathy. Verify on T1 and T2-weighted sequences — magic angle disappears at TE > 60 ms.

5.2 Protocol Efficiency and Throughput

A full diagnostic hand MRI at 3T can be completed in 25–35 minutes. At 1.5T, 35–50 minutes.

Combined hand + wrist protocol for RAMRIS: clinical RA trials require the combined MCP joint + wrist RAMRIS score. When both the hand and wrist are to be assessed, the most efficient approach is a single examination covering both regions, using a single wider coronal FOV (18–22 cm) that encompasses the wrist and MCP joints simultaneously. This is the standard approach in RA clinical trials and should be considered when the clinical question requires both. The specific OMERACT RAMRIS technical specifications define the combined protocol.

Short protocol for targeted assessment: when the clinical question is limited (e.g., thumb UCL only; single MCP joint mass), an abbreviated three-sequence protocol (coronal PD-FS + T1 + targeted axial PD-FS) provides sufficient coverage in approximately 12–15 minutes.

3D isotropic sequences: a single 3D PD-FS acquisition covering the full hand at 0.5 mm isotropic provides all three planes without the need for separate acquisitions. At 3T, this is achievable in 8–12 minutes. The trade-off is slightly lower SNR per slice compared to 2D optimised acquisitions — relevant for RAMRIS erosion detection where 1 mm cortical breaks must be visible.

5.3 Field Strength Considerations

3T advantages: at 3T with a dedicated hand/wrist coil and Dixon fat suppression, hand MRI provides in-plane resolution of 0.3–0.4 mm — sufficient for the 1 mm erosion threshold of RAMRIS and for 1–2 mm UCL fibres. SNR is approximately 60–70% higher than at 1.5T, enabling smaller voxels at equivalent acquisition time.

1.5T: clinically adequate for soft tissue masses, bone marrow screening, and post-traumatic assessment. For RAMRIS erosion scoring and small UCL tear characterisation, 1.5T is diagnostically limited compared to 3T. Low-field (0.2–0.4T) extremity MRI systems have been validated for RAMRIS scoring in clinical trials but produce significantly lower resolution and are not discussed further here.

SAR considerations: no specific SAR limitation issues for standard hand TSE sequences at 3T. The small FOV and dedicated surface coil mean SAR is well within limits.

6. Contrast Use Principles Specific to Hand MRI

6.1 Non-Contrast Standard Protocol — Sufficient For

• Metacarpal bone marrow screening (stress fracture, contusion, avascular changes)
• Complete UCL tear and Stener lesion identification at 3T
• Soft tissue mass characterisation (ganglion, GCTTS, lipoma)
• Bone erosion detection in established RA (RAMRIS erosion and osteitis scoring with modern Dixon-based non-contrast sequences)
• Post-traumatic hand assessment without specific synovitis question
• Metacarpal fracture characterisation (staging, alignment)
• Trigger finger / A1 pulley assessment

6.2 Gadolinium Indicated — Region-Specific Contexts

RAMRIS synovitis scoring specifically requires post-contrast T1-FS sequences. The OMERACT RAMRIS definition of synovitis is based on the volume of enhancing tissue in the synovial compartment. Non-contrast Dixon T2/STIR sequences detect osteitis and can identify fluid in the joint, but active synovitis quantification by RAMRIS requires contrast enhancement. The 2025 study of mDixon hand MRI without contrast showed that inflammatory features in the general population are rare, supporting the use of non-contrast MRI as a screening tool — but confirmed that synovitis scoring by RAMRIS requires contrast for validation [6].

Deep palmar space infection: post-contrast sequences are mandatory to distinguish abscess cavities (non-enhancing central fluid with peripheral enhancement) from cellulitis (diffuse enhancement) and to identify the extent of flexor tendon sheath enhancement.

Vascular malformations: the enhancement kinetics (arteriovenous malformation, venous malformation, lymphatic malformation) require post-contrast sequences for complete characterisation.

Suspected neoplasm: any hand mass with aggressive features (cortical destruction, growth, pain) requires post-contrast assessment.

Early RA disease activity monitoring: in patients where the clinical question is treatment response, post-contrast RAMRIS synovitis quantification is the validated outcome measure [5].

6.3 Post-Contrast Acquisition Timing

For RAMRIS synovitis scoring: imaging should begin within 5 minutes of injection and be completed within 10 minutes. The OMERACT RAMRIS protocol specifies post-contrast imaging 4.5–7 minutes after injection for the wrist and MCP joints.

For deep space infection assessment: immediate post-injection axial and coronal sequences; no delay required.

STIR absolutely contraindicated post-gadolinium — the same absolute rule as all other MRIninja protocols.

7. Reporting Essentials

7.1 Interpretation Framework

The hand MRI report should assess three anatomical domains: the MCP and CMC joints (bone marrow, cartilage, synovium, capsule, and collateral ligaments); the tendons and tendon sheaths (flexor and extensor, from the CMC level to the A1 pulley at the MCP joint); and the palmar spaces and neurovascular structures (palmar arches, digital nerves, thenar and hypothenar spaces, interosseous spaces).

Clinical context is essential: the same finding — MCP joint effusion with periarticular bone marrow oedema — carries different clinical significance in a 30-year-old with symmetric polyarthritis (probable early RA) versus a 70-year-old with unifocal third MCP pain after a single trauma (contusion or gout). The reporting framework must integrate the MRI findings with the clinical indication.

RAMRIS scoring context: when the indication is RA monitoring, the report should follow the RAMRIS framework explicitly, scoring synovitis, osteitis, and erosions at each MCP joint site. If only the hand (not the wrist) is included, this should be stated as a partial RAMRIS assessment.

7.2 Mandatory Reporting Checklist

MCP joints (each of the five joints individually): - Joint space: normal / narrowed / fused - Synovitis (fluid / enhancing tissue if contrast used): absent / mild / moderate / severe - Bone marrow oedema (osteitis): absent / present (metacarpal head / proximal phalanx base) - Bone erosions: absent / present (size, location, number) - Collateral ligaments: intact / partial tear / complete tear (for clinically relevant joints)

CMC joints: - First CMC (trapeziometacarpal): cartilage, joint space, synovitis, ligamentous integrity - Second to fifth CMC: synovitis, erosions if present

Thumb-specific (when relevant): - Thumb UCL at MCP: intact / partial tear / complete tear / Stener lesion - Thumb CMC: osteoarthritis grade, cartilage, ligamentous laxity

Flexor tendon sheaths: - Digital flexor sheaths: normal / tenosynovitis (fluid volume relative to tendon diameter) - A1 pulley region: normal / nodular thickening (trigger finger)

Palmar spaces: - Thenar space: normal / fluid / mass - Mid-palmar space: normal / fluid / mass - Hypothenar space: normal / fluid / mass

Soft tissue masses: location, size, tissue characteristics (T1 signal, T2 signal, enhancement)

Metacarpal bones: cortex, marrow signal, stress fracture, periosteal reaction

Technical limitations: fat suppression quality, hardware artefact, motion

7.3 Structured Reporting

Reports: Indication (clinical question, laterality); Technique (field strength, position, sequences, contrast use and timing); Comparison (prior MRI, prior radiographs); Findings (joint-by-joint per checklist); Impression (clinically relevant summary); Limitations; Critical communication if required.

7.4 Incidental Findings — Clinical Decision Framework

Usually benign: small MCP joint effusion without synovitis (physiological); minor second or third MCP joint arthrosis in patients over 50; small palmar ganglion; mild tenosynovial fluid in a single flexor sheath in elderly patients.

May require clinical correlation: bone marrow oedema (RAMRIS osteitis) in a patient referred for trauma — even if asymmetric and clinically unreported, this finding may indicate early inflammatory arthritis; isolated fourth or fifth MCP erosion in an asymptomatic patient (may indicate early RA or gout).

Require explicit communication: aggressive bone lesion features at the metacarpal; deep space abscess requiring urgent surgical drainage; complete UCL tear with Stener lesion not mentioned in the clinical indication (changes management urgency); unexpected significant erosions in a patient referred for soft tissue mass (indicates active inflammatory disease).

8. MRI Technologist Pearls

8.1 Sequence Order Logic

Recommended sequence order for standard non-contrast hand MRI:

1. Three-plane localiser
2. Coronal PD-FS ← most diagnostically critical; acquire first
3. Coronal T1 ← same prescription; fast acquisition immediately after
4. STIR coronal ← bone marrow and osteitis screening
5. Axial PD-FS
6. Sagittal PD-FS

If contrast is used: STIR before injection; coronal T1 non-FS before injection; post-contrast coronal T1-FS at 4–7 minutes post-injection (per OMERACT RAMRIS protocol); axial T1-FS post-contrast.

8.2 Positioning Tricks

• Verify the hand is flat before scanning: view the coronal localiser — the five metacarpal heads should be approximately equidistant in the dorsal-palmar direction. If one metacarpal head is more dorsal than the others, the hand is not flat and must be repositioned.
• For the thumb: extend and abduct the thumb to the maximum comfortable position before scanning. Allowing the thumb to flex into the palm during the examination will move it out of the prescribed oblique axial plane.
• For rheumatoid hands with deformity: the hand may not lie flat due to MCP joint subluxations or ulnar deviation deformity. Document the position and use foam padding to stabilise the hand in the best achievable flat position. State the deformity in the report as a technical limitation.
• For the Superman position: pad the shoulder and provide wrist support. For hand-specific examinations, the patient may be more comfortable if the elbow is slightly flexed rather than fully extended, reducing forearm pronation strain.
• Bilateral hand imaging for RAMRIS: position both hands simultaneously in a wider coil if available (some centres use a dedicated bilateral hand coil), or image sequentially with repositioning. Sequential bilateral imaging doubles the examination time; note this when scheduling.

8.3 Fast Salvage Protocol

Three sequences in approximately 10 minutes provide the minimum diagnostically interpretable hand MRI.

8.4 Common Avoidable Errors

9. Quality Control Checklist

• Hand flat: metacarpal heads equidistant on coronal localiser
• CMC joints included proximally in coronal and axial coverage
• All five MCP joint heads included in distal extent of coronal
• Palmar spaces covered axially (CMC to MCP level)
• Fat suppression uniform across full FOV on PD-FS
• STIR acquired before gadolinium (if contrast used)
• Post-contrast sequences acquired 4–7 minutes after injection (RAMRIS protocol, if applicable)
• No significant motion artefact on coronal PD-FS (priority sequence)
• Thumb oblique sequences completed if UCL tear is the clinical question
• Metal artefact noted and MARS applied if hardware present
• Correct laterality documented in sequence labels
• Patient position (pronation/supination) documented
• All five mandatory sequences completed

11. Evidence Gaps and Ongoing Debate

RAMRIS without contrast (non-enhanced scoring): the OMERACT RAMRIS protocol is defined with post-contrast T1-FS sequences for synovitis scoring. However, emerging data using Dixon-based non-contrast hand MRI suggest that osteitis and fluid-based synovitis features are detectable without contrast with adequate specificity [6]. Whether non-contrast scoring can fully replace contrast-based RAMRIS for clinical trial endpoints remains an active debate.

Optimal field strength for RAMRIS in clinical practice: validated RAMRIS scoring thresholds were established at 1.5T. Whether 3T RAMRIS produces systematically different scores (due to higher SNR enabling detection of smaller lesions) and whether existing 1.5T-based response criteria are directly transferable to 3T remains unresolved. The EULAR recommendations acknowledge field-strength variability in scoring.

3D isotropic sequences for erosion detection: 3D isotropic TSE sequences at 3T detect bone erosions at comparable or superior sensitivity to 2D sequences in the wrist, but dedicated prospective studies at the MCP joint level are limited.

Non-contrast MRI for Stener lesion diagnosis: the sensitivity of non-contrast 3T MRI for Stener lesion detection at the thumb MCP UCL is approximately 96% in some series [7], but the comparison with direct MR arthrography has not been systematically evaluated. For equivocal cases, MRA may provide superior partial tear characterisation.

Bilateral hand MRI in RA: whether sequential bilateral hand MRI (left then right) with repositioning produces reliably comparable RAMRIS scores on both sides within a single session remains undetermined. Simultaneous bilateral coil systems provide the most reproducible bilateral scoring.

12. Evidence-Based References

A. Guidelines / Consensus / Society Recommendations

B. Systematic Reviews / Meta-analyses

(No systematic reviews specifically addressing hand MRI technical protocols were identified. The evidence base rests primarily on cohort studies, OMERACT/EULAR guidelines, and technical papers.)

C. Important Prospective / Original Studies

D. Technical MRI Papers

E. Landmark Historical References

End of document — MRI Hand Generic Standard Protocol — MRIninja v1.0 — May 2026 This master page is the reference for all future hand MRI child pages including: RA hand RAMRIS protocol, thumb UCL injury (Gamekeeper's/Skier's thumb), first CMC osteoarthritis, deep palmar space infection, metacarpal fractures, GCTTS and soft tissue masses, and combined hand-wrist RA protocol.