MRI Elbow – Generic Standard Protocol

1. Executive Summary

The elbow is a complex hinge-pivot joint composed of three articulations — humeroulnar, humeroradial, and proximal radioulnar — whose stability depends on an intricate balance of osseous constraints, collateral ligaments, and musculotendinous origins. MRI is the primary advanced imaging modality for the elbow, offering simultaneous multiplanar assessment of all soft tissue and osseous components without the radiation exposure of CT and without the operator dependence and osseous limitation of ultrasound.

The generic non-arthrographic elbow MRI protocol described here is designed for the broad population presenting with elbow pain of unclear aetiology, suspected epicondylitis, post-traumatic evaluation, suspected collateral ligament injury, distal biceps or triceps tendon pathology, osteochondral disease, nerve entrapment syndromes, or inflammatory arthropathy. It provides a comprehensive non-invasive survey of all major anatomical compartments.

1.1 Core Strengths

Tendon and ligament assessment: MRI is the primary modality for evaluating the common extensor tendon (lateral epicondylitis), common flexor-pronator tendon (medial epicondylitis), ulnar collateral ligament (UCL), lateral collateral ligament complex (LCLC), distal biceps tendon, and triceps tendon. No other modality provides equivalent simultaneous multiplanar soft tissue resolution without ionising radiation.

Collateral ligament integrity: the anterior bundle of the UCL — the primary stabiliser against valgus stress — is reliably evaluated on coronal MRI sequences. UCL tear diagnosis guides surgical decision-making in overhead athletes and after acute elbow dislocation.

Osteochondral disease: MRI demonstrates osteochondral lesions (osteochondritis dissecans, osteochondral fracture) with superior sensitivity and specificity for assessing subchondral bone and cartilage integrity compared to radiographs and ultrasound.

Nerve entrapment syndromes: ulnar nerve at the cubital tunnel, radial nerve at the radial tunnel and arcade of Frohse, and median nerve at the pronator syndrome level are all assessable on MRI with dedicated nerve-sensitive sequences. MRI demonstrates the nerve morphology, signal, and surrounding anatomy that ultrasonography cannot fully characterise, particularly for radial tunnel and posterior interosseous nerve entrapment.

Bone marrow assessment: occult stress fractures, medial epicondyle apophyseal injury in young athletes, and early avascular necrosis at the capitellum 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.

Superior to ultrasound for: deep structures (posterior interosseous nerve, arcade of Frohse, intra-articular bodies), complete tendon tear characterisation with retraction, bone marrow changes, and complete joint survey.

Superior to CT for: soft tissue pathology (tendons, ligaments, nerves, cartilage, marrow oedema) without radiation.

1.2 Intrinsic Limitations of the Generic Protocol

The generic non-arthrographic elbow MRI protocol is a diagnostic compromise between breadth, acquisition time, and diagnostic specificity. Several important limitations must be stated explicitly.

Intra-articular pathology detection: loose bodies, chondral flaps, and partial UCL tears are significantly better evaluated by direct MR arthrography than non-arthrographic MRI. The ACR Appropriateness Criteria specifically endorse direct MR arthrography for suspected intra-articular loose bodies and UCL partial tears that are operative considerations [1]. When the primary clinical question is intra-articular, the generic protocol is a preliminary study, not the definitive examination.

Partial collateral ligament tears: non-arthrographic MRI sensitivity for partial UCL tears is significantly lower than for complete tears. MR arthrography provides joint distension that opens the UCL laminar structure and demonstrates partial-thickness undersurface tears that are invisible on non-arthrographic sequences.

Cartilage assessment: thin elbow cartilage (1–2 mm) at the trochlea, capitellum, and radial head requires dedicated high-resolution sequences and arthrographic distension for reliable grading of focal chondral defects. The generic protocol provides a cartilage survey but not high-precision cartilage quantification.

Positioning limitation: the elbow is imaged off-isocentre in the standard supine position, degrading B0 homogeneity and fat suppression uniformity — a specific technical challenge for the elbow that does not apply to most other joints.

When dedicated child protocols are required: direct MR arthrography for suspected UCL partial tears, intra-articular loose bodies, and osteochondral lesion stability assessment; FABS position for distal biceps tendon full evaluation; dedicated nerve imaging for suspected Parsonage-Turner syndrome or brachial plexus involvement; post-operative protocol with MARS sequences for metal hardware.

2. Main Clinical Indications

2.1 Standard Indications

Lateral epicondylitis and common extensor tendon pathology is the most frequent clinical indication for elbow MRI in clinical practice. Lateral epicondylitis (tennis elbow) involves the common extensor tendon origin at the lateral epicondyle, with the extensor carpi radialis brevis as the primary affected structure. MRI demonstrates tendon signal change, partial tearing, peritendinous oedema, and the frequently associated radial collateral ligament and lateral UCL abnormality that correlates with clinical severity [3, 4]. The generic protocol is sufficient for the large majority of cases; MR arthrography is not required unless intra-articular pathology is specifically suspected.

Medial epicondylitis and common flexor-pronator tendon pathology is less common than lateral epicondylitis but involves the common flexor-pronator tendon origin at the medial epicondyle. The clinical question is more complex because the medial compartment also contains the UCL and the ulnar nerve, both of which may be simultaneously injured. MRI evaluates all three structures in a single examination, which is its major advantage over ultrasound for the medial elbow.

UCL assessment — particularly the anterior bundle of the medial UCL — is the primary indication for elbow MRI in overhead athletes (baseball pitchers, javelin throwers, gymnasts) presenting with medial elbow pain. Complete UCL tears are reliably diagnosed on non-arthrographic MRI (coronal PD-FS). Partial tears — particularly partial undersurface tears of the anterior bundle — require MR arthrography for reliable detection.

Distal biceps tendon pathology including complete rupture, partial tear, and tendinopathy is well evaluated on axial and sagittal sequences. Complete rupture is obvious on standard sequences; partial tears and chronic degeneration benefit from dedicated FABS position imaging (a conditional sequence described in Section 4.2) which displays the tendon in its longitudinal axis.

Triceps tendon pathology including partial and complete avulsion tear at the olecranon insertion is assessed on sagittal and coronal sequences. Triceps tear is less common than biceps tear but surgically important; MRI characterises the degree of retraction.

Osteochondral lesions of the capitellum — the most common site for osteochondritis dissecans in the elbow — and of the radial head and trochlea are evaluated on MRI for fragment size, stability, and cartilage integrity. The generic protocol provides the initial assessment; MR arthrography improves characterisation of fragment stability when surgical planning is contemplated.

Cubital tunnel syndrome (ulnar nerve entrapment at the elbow) is the second most common peripheral nerve entrapment in the upper extremity. MRI demonstrates ulnar nerve T2 signal change, nerve enlargement, and the morphology of the cubital tunnel, which determines surgical decompression approach. The generic protocol adequately evaluates the nerve at the cubital tunnel level.

Radial tunnel and posterior interosseous nerve entrapment: the radial nerve and its posterior interosseous branch are evaluated on axial sequences through the radial tunnel, the arcade of Frohse, and the supinator muscle. MRI is the only modality that reliably images deep nerve compression at this level.

Inflammatory arthropathy: rheumatoid arthritis, crystal arthropathy, spondyloarthropathy, and pigmented villonodular synovitis (PVNS) of the elbow joint are assessed on MRI for synovitis, erosions, marrow oedema, and joint effusion. Post-contrast sequences are required for synovitis quantification.

Post-traumatic evaluation: after acute elbow trauma with normal or inconclusive radiographs, MRI detects occult fractures, muscle and tendon injuries, haemarthrosis, and nerve injury.

2.2 Urgent Red Flags Requiring Expedited or Emergency Imaging

The elbow does not generate life-threatening emergencies. However, the following scenarios require expedited imaging or urgent clinical escalation.

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

The elbow off-isocentre problem: the primary anatomy-specific technical challenge for elbow MRI is that the elbow joint cannot be placed at isocentre in the standard supine position — the body occupies isocentre and the arm lies at the side. This creates systematic B0 inhomogeneity and reduced SNR that affects fat suppression uniformity and signal quality. The patient and technologist must understand this limitation and its positioning implications before the examination begins.

Arm position and forearm rotation: the forearm should be in supination (thumb upward, palm upward) with the elbow fully extended. This position:

• Places the UCL anterior bundle under mild tension, maximising its visibility on coronal sequences
• Removes the bicipital tuberosity from the direct axial plane, reducing superimposition
• Is the standard anatomical position for which normal reference values are described

Metal near the elbow region: document any history of prior elbow surgery, metallic anchors, hardware from fracture fixation, total elbow arthroplasty, or retained fragments from trauma. Metal hardware markedly degrades image quality and may require MARS sequences. Prior radiographs should be reviewed before planning to identify hardware composition and location.

Compression wraps and splints: any compression bandaging, taping, or temporary splinting over the elbow must be removed before imaging. These items may contain metallic or ferromagnetic components and can compress surface coils, degrading local signal.

Clothing: no metallic fasteners over or near the arm. A gown with the sleeve removed from the affected arm provides optimal coil placement.

Pain management: patient discomfort during the 30–45 minute examination is the primary cause of motion artefact in elbow MRI. For patients with severe acute elbow pain, pre-examination analgesia (as appropriate clinically) significantly improves image quality by enabling comfortable positioning and reducing involuntary motion.

Prior corticosteroid injection: intra-articular or peritendinous corticosteroid injections within 4–6 weeks may alter signal around the injection site. Document date and location.

3.2 Patient Positioning on the MRI System

Two positioning options exist with distinct advantages and disadvantages. The choice depends on patient tolerance, scanner bore geometry, and the primary clinical question.

Option 1 — Supine, arm at side (standard for most clinical departments): The patient lies supine, feet first, with the affected arm extended along the side of the body, forearm supinated. The elbow is positioned as close to the body as possible with the arm in contact with the thigh. This is the most comfortable position and achievable in virtually all patients.

Disadvantage: the elbow is positioned approximately 15–25 cm lateral to isocentre in most scanners. This produces suboptimal B0 homogeneity, reduced SNR (approximately 20–30% loss compared to isocentre), and patchy fat suppression — particularly on the side of the elbow facing the body wall. Dixon fat suppression significantly mitigates this limitation.

Option 2 — Prone, arm extended overhead ("Superman position"): The patient lies prone with the affected arm extended over the head, elbow near isocentre, forearm pronated, thumb downward. The elbow is at or near isocentre, optimising B0 homogeneity, SNR, and fat suppression uniformity.

Disadvantage: this position is uncomfortable, particularly for elderly patients, those with shoulder pathology, or any patient with difficulty lying prone. Duration tolerance is limited and motion artefact from discomfort may paradoxically worsen image quality despite the improved field position. Claustrophobia is more common in the prone head-first position.

Practical recommendation: for most routine clinical departments, the supine arm-at-side position with Dixon fat suppression is the pragmatic standard. The Superman position is preferred when fat suppression quality is critical (suspected oedema in the peritendinous tissues, suspected bone marrow pathology) and when the patient can tolerate it comfortably for the duration of the study.

Coil selection: a dedicated small phased-array surface coil (elbow coil, flex coil, or multi-channel wrap coil) is mandatory. The coil should encircle the elbow joint circumferentially or be positioned to cover the entire joint from the distal humeral metaphysis to the bicipital tuberosity. A general-purpose body coil or spine coil is inadequate for diagnostic elbow imaging.

Centering: isocentre at the centre of the elbow joint (the level of the medial and lateral epicondyles, approximately at the joint line). In the supine position, the laser alignment should be verified at the joint level, not at the distal humerus or the proximal forearm.

Immobilisation: foam padding medial and lateral to the forearm prevents involuntary rotation. A wedge under the hand maintains forearm supination. Instruct the patient explicitly not to move the arm during the entire examination.

Common positioning errors:

• Forearm in pronation instead of supination: the UCL anterior bundle is suboptimally displayed; bicipital tuberosity positioning is altered
• Elbow flexion: any flexion beyond 5–10° changes the coronal plane anatomy and may close the UCL, reducing its visibility
• Coil positioned too distally: the common tendon origins at the epicondyles are excluded from coverage
• No padding, forearm free to rotate: motion degrades the entire examination

4. Standard Protocol Design

The standard elbow MRI protocol is built around three mandatory orthogonal planes — coronal, sagittal, and axial — each with fat-suppressed fluid-sensitive (PD-FS) and non-fat-suppressed T1-weighted components. The three-plane PD-FS orthogonal acquisition and the coronal T1 are the irreducible diagnostic minimum.

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 in elbow MRI, comparable in clinical priority to the oblique coronal for the shoulder. The coronal plane displays, in a single acquisition:

• The UCL anterior bundle in its full longitudinal extent from medial epicondyle to sublime tubercle
• The lateral collateral ligament complex from lateral epicondyle across the radial head
• The common extensor tendon origin at the lateral epicondyle
• The common flexor-pronator tendon origin at the medial epicondyle
• The articular cartilage of the trochlea and capitellum at their maximum surface
• The medial and lateral epicondyles for periosteal reaction, stress changes, and calcification
• The brachioradialis and extensor carpi radialis at their origins
• Joint effusion in the medial and lateral recesses

Fat suppression is mandatory: the periarticular fat at the epicondyles, the peritendinous fat surrounding the ligament bundles, and the fat between the joint space planes would all overwhelm the fluid signal without suppression. PD weighting (TE 20–40 ms) is preferred over T2 for the elbow tendons and ligaments for the same physics rationale as the shoulder — magic angle effect in tendon fibres is most prominent at short TE but less clinically problematic in the elbow than the shoulder because the critical zone angle is less consistently at 55° to B0. However, the intermediate TE of PD provides superior SNR compared to T2 at the voxel sizes required for high-resolution elbow imaging.

Axial PD-FS is the primary plane for:

• The ulnar nerve at the cubital tunnel (size, signal, surrounding groove morphology)
• The radial nerve and posterior interosseous nerve at the radial tunnel and supinator
• The common extensor and common flexor-pronator tendon cross-sections
• The distal biceps tendon at its course toward the radial tuberosity
• The annular ligament around the radial head
• The trochlear and capitellar cartilage in cross-section

High in-plane resolution on the axial sequence is critical: the ulnar nerve at the cubital tunnel is typically 3–6 mm in diameter; the posterior interosseous nerve in the radial tunnel is 2–4 mm. At standard 3 mm slice thickness and in-plane resolution ≤ 0.4 × 0.4 mm, these structures are reliably identified and characterised.

Sagittal PD-FS provides the third orthogonal plane and is critical for:

• Distal biceps tendon: its full course from the musculotendinous junction, through the antecubital fossa, to the radial tuberosity insertion
• Triceps tendon at its olecranon insertion
• Anterior and posterior capsule
• Coronoid and olecranon fossae and their fat pads (posterior fat pad displacement = joint effusion marker)
• Brachialis muscle and tendon
• Articular cartilage of the trochlear notch

The sagittal plane is the optimal plane for assessing the distal biceps tendon because the tendon is oriented approximately in the sagittal plane as it courses anteriorly from the musculotendinous junction to the radial tuberosity. Any retraction of a ruptured distal biceps is measured on the sagittal sequence.

Coronal T1 without fat suppression provides:

• Anatomical landmark identification without signal loss from fat suppression failure
• T1 characterisation of signal (distinguishing fat from fluid, blood, and calcification)
• Bone cortex and trabecular definition
• Goutallier-equivalent assessment of any muscle fatty replacement in the forearm
• Baseline for comparison with post-contrast sequences when gadolinium is used

STIR provides B0-independent bone marrow and soft tissue oedema sensitivity. It is the most reliable sequence for medial epicondyle stress reaction, apophyseal injury in paediatric patients, early osteochondritis dissecans at the capitellum, and peritendinous inflammatory oedema when spectral fat suppression is compromised by the off-isocentre elbow position. STIR must always be acquired before gadolinium injection.

4.4 Sequence Matching and Cross-Sequence Consistency

The three orthogonal planes must be truly orthogonal — prescribed from the same coronal reference. If the coronal and sagittal are not prescribed to be perpendicular to each other, cross-referencing between planes is unreliable and tear size measurement is inaccurate.

Pre- and post-contrast sequences must use identical prescription (FOV, angulation, slice positions) for meaningful enhancement comparison. When post-contrast T1-FS is acquired, the pre-contrast T1 should be acquired first and the sequence saved as a reference. Any rotation or FOV difference between the two T1 acquisitions produces interpretive difficulty in identifying enhancement versus pre-existing T1-bright structures (calcification, fat, proteinaceous fluid).

For serial follow-up studies, the coronal prescription angle must be reproducible. Document the angle used on all examinations where serial comparison is required.

4.5 Fat Suppression — Region-Specific Technical Considerations

Fat suppression is mandatory for all fluid-sensitive sequences in elbow MRI for the same reason as the shoulder: the periarticular fat would overwhelm the fluid signal without it. However, the elbow presents a more challenging fat suppression problem than most joints because of its off-isocentre position in the standard supine positioning.

B0 inhomogeneity at the off-isocentre elbow: when the elbow is positioned 15–25 cm lateral to isocentre (standard supine position), the B0 field is progressively less homogeneous. This produces incomplete spectral fat saturation — characteristically worse on the medial side of the elbow (facing the body) than the lateral side (facing away). The result is patchy fat suppression that can simulate peritendinous oedema or obscure true signal abnormality.

Dixon fat suppression is strongly recommended for elbow MRI in the supine position because its multi-echo acquisition is B0-independent and produces consistent fat suppression regardless of off-isocentre position. A 2-point or 3-point Dixon acquisition produces simultaneous fat-only and water-only images that are diagnostically useful beyond fat suppression alone (fatty replacement quantification, simultaneous T1 and fluid-sensitive information from a single acquisition).

SPAIR (spectral adiabatic inversion recovery) provides improved B0 robustness over standard CHESS fat saturation and is an acceptable alternative when Dixon is not available. It performs better than CHESS at the off-isocentre elbow, particularly at 3T.

STIR is the mandatory backup when spectral fat suppression is unreliable, and should always be included in the standard protocol as the bone marrow screening sequence. STIR performance is not degraded by off-isocentre position.

STIR contraindicated post-gadolinium — identical rule as in all other MRIninja protocols.

At 3T specifically: B0 inhomogeneity effects are amplified at 3T compared to 1.5T. For the off-isocentre supine elbow, this means spectral fat saturation is even less reliable at 3T than at 1.5T. Dixon is more strongly recommended at 3T. Alternatively, the Superman position is preferred at 3T when fat suppression quality is the primary technical concern.

5. Optimisation Strategy

5.1 Artifact Reduction by Source

Off-isocentre B0 inhomogeneity and fat suppression failure is the most important and elbow-specific artefact problem. As described in Section 4.5, the medial aspect of the elbow in the standard supine position consistently shows worse fat suppression than the lateral aspect. This produces apparent peritendinous or periligamentous signal increase on the medial side that can simulate peritendinous oedema around the UCL or medial epicondyle. Recognising the asymmetric pattern — signal abnormality on the medial side without corresponding abnormality on T1, and without clinical correlation — is key to avoiding this pitfall. STIR provides the diagnostic verification: STIR is not affected by B0 inhomogeneity and shows true oedema if present. If STIR is normal where the PD-FS shows apparent signal, fat suppression failure is the explanation.

Motion artefact: the most common cause of non-diagnostic elbow MRI in clinical practice. The elbow is an uncomfortable position to maintain motionless for 30–45 minutes, particularly in patients with acute pain. Motion is most prominent on sequences with long TR (STIR is most vulnerable). Practical measures: adequate analgesia, clear patient instruction, foam padding for immobilisation, and prioritising the most critical sequences early in the examination.

Chemical shift artefact: at bone-fat interfaces (epicondyles, radial head, olecranon), chemical shift produces a signal line at the fat side of the interface and a void on the opposite side. This is visible on PD-FS and can simulate subchondral changes. Wider receiver bandwidth reduces the displacement; at 1.5T 130–200 Hz/px; at 3T 200–400 Hz/px.

Magic angle artefact in the UCL: the anterior bundle of the UCL runs in a semicircular arc from the medial epicondyle to the sublime tubercle. At various points along this arc, the ligament fibres pass through 55° relative to B0, producing apparent T1 and PD signal increase that can simulate partial tearing. The rule is the same as for all tendons: signal on PD that disappears on T2-weighted sequences (TE > 60 ms) is likely magic angle. True ligament tears show signal on all fluid-sensitive sequences and on the T1.

Susceptibility artefact from metal: previous elbow surgery frequently leaves metallic anchors, hardware, or retained fragments. MARS sequences (wider bandwidth, STIR, reduced TE) mitigate but do not eliminate susceptibility effects. Titanium hardware causes significantly less artefact than stainless steel. The reporting radiologist should be informed of hardware presence before scanning.

Brachial artery pulsation artefact: the brachial artery courses through the antecubital fossa anterior to the elbow joint. Its pulsation produces phase-direction ghosting that can overlap with the antecubital fossa structures (distal biceps tendon, median nerve, brachial artery bifurcation) on axial sequences. An anterior presaturation band over the brachial vessels and a posterior saturation band over the olecranon region effectively suppresses this artefact.

Noise amplification in small FOV imaging: the elbow requires a small FOV (100–160 mm) to achieve the in-plane resolution required for ligament and nerve assessment. Small FOV with parallel imaging at high R factors reduces acquisition time but amplifies noise. A minimum coil channel count of 8 is required for R > 2 at small FOV without diagnostic SNR penalty.

5.2 Protocol Efficiency and Throughput

A full diagnostic elbow MRI protocol at 3T can be completed in 20–30 minutes. At 1.5T, 30–40 minutes.

Short protocol for limited cooperation or pain: Coronal PD-FS + Coronal T1 + Axial PD-FS provides the minimum clinically useful elbow MRI in approximately 12–15 minutes at 3T. The primary UCL, common tendon, and nerve pathology targets are covered. Sagittal coverage of the distal biceps and triceps is sacrificed.

3D isotropic sequences: a single 3D oblique coronal PD-FS acquisition provides coronal, sagittal, and axial reformats from a single 5–8 minute acquisition. For standard epicondylitis and UCL assessment, 3D isotropic sequences perform comparably to 2D at 3T. For nerve assessment, 2D axial with high in-plane resolution is generally preferred because the thin nerve cross-sections require guaranteed in-plane resolution in the axial plane.

5.3 Field Strength Considerations

3T advantages: higher SNR allows smaller voxels at the same acquisition time — critical for the elbow where the UCL anterior bundle is typically 2–4 mm thick and the ulnar nerve 3–6 mm in diameter. At 3T, these structures are consistently resolved; at 1.5T, partial volume averaging may reduce diagnostic confidence for partial tears.

3T disadvantages: B0 inhomogeneity is worse at 3T than 1.5T for the off-isocentre supine elbow, making Dixon fat suppression more important at 3T. SAR limitations restrict flip angle choices for TSE sequences. Chemical shift artefact is doubled compared to 1.5T for equivalent bandwidth.

1.5T: clinically adequate for the majority of elbow indications. Full-thickness UCL tears, complete tendon ruptures, and gross bone marrow changes are reliably diagnosed at 1.5T. Where 1.5T falls short is in partial UCL tear characterisation and subtle nerve enlargement, where the lower SNR margin reduces diagnostic confidence.

6. Contrast Use Principles Specific to Elbow MRI

6.1 Non-Contrast Standard Protocol — Sufficient For

Non-contrast elbow MRI is sufficient for:

• Lateral epicondylitis assessment (common extensor tendon)
• Medial epicondylitis assessment (common flexor-pronator tendon)
• Complete UCL tear diagnosis
• Distal biceps tendon complete rupture
• Triceps tendon tear
• Cubital tunnel syndrome nerve assessment
• Osteochondral lesion detection (not stability assessment)
• Bone marrow screening (stress reaction, occult fracture, avascular necrosis)
• Post-traumatic soft tissue evaluation
• Most inflammatory arthropathy survey (non-contrast adequate for detection)

6.2 Gadolinium Indicated — Region-Specific Contexts

Direct MR arthrography (intra-articular gadolinium) is a separate child protocol and is the preferred technique for:

• Suspected partial UCL tears (the most validated indication for elbow MRA)
• Osteochondral lesion stability assessment and intra-articular loose bodies
• SLAP-equivalent superior capitellum lesions

Intravenous GBCA post-contrast T1-FS is specifically indicated for:

• Inflammatory synovitis quantification (rheumatoid, crystal, PVNS)
• Suspected septic arthritis (synovial enhancement pattern)
• Suspected soft tissue or bone tumour
• Nerve enhancement in suspected Parsonage-Turner syndrome, neuritis, or infiltrative nerve pathology
• Post-operative assessment for repair integrity and peritendinous scar characterisation

6.3 Post-Contrast Acquisition Timing

For inflammatory synovitis assessment, imaging should begin 3–5 minutes after injection to allow adequate synovial enhancement. Earlier imaging may show incomplete synovial enhancement; later imaging (> 10 minutes) allows gadolinium diffusion into joint fluid, reducing synovium-fluid contrast.

STIR is absolutely contraindicated post-gadolinium — applies identically to elbow MRI.

7. Reporting Essentials

7.1 Interpretation Framework

The elbow MRI report should systematically assess four anatomical compartments: medial (UCL, common flexor-pronator tendon, ulnar nerve), lateral (LCLC, common extensor tendon, radial nerve), anterior (distal biceps tendon, brachialis, median nerve), and posterior (triceps tendon, olecranon, posterior capsule). Additionally, the intra-articular compartment (cartilage, loose bodies, joint effusion, synovium) and the osseous structures (bone marrow, epicondyles, capitellum, radial head) require systematic assessment.

The clinical context determines where to focus: a young overhead athlete with medial pain requires prioritised UCL and ulnar nerve assessment; an older patient with lateral pain after repetitive work requires prioritised common extensor tendon assessment; a patient with acute trauma requires prioritised assessment for complete tendon and ligament disruptions.

7.2 Mandatory Reporting Checklist

Medial compartment:

• UCL anterior bundle: intact / signal change / partial tear (articular surface, bursal surface, midsubstance) / complete tear; if tear: location, retraction
• Common flexor-pronator tendon: intact / tendinopathy / partial tear / complete tear; medial epicondyle changes
• Ulnar nerve at cubital tunnel: calibre, T2 signal, surrounding groove morphology, accessory anconeus

Lateral compartment:

• Common extensor tendon origin: intact / tendinopathy / partial tear / complete tear; epicondyle bone changes
• Lateral UCL (LUCL): intact / signal change / tear
• Radial collateral ligament: intact / partial / complete
• Radial nerve branches: in radial tunnel and at supinator level if clinically indicated

Anterior compartment:

• Distal biceps tendon: intact / tendinopathy / partial tear (anterior, posterior surface) / complete tear; if ruptured: retraction distance
• Brachialis tendon: intact / tear
• Bicipital tuberosity: cortical and marrow signal

Posterior compartment:

• Triceps tendon at olecranon: intact / tendinopathy / partial / complete tear
• Olecranon: cortical, marrow signal; posterior impingement
• Posterior fat pad: position (normal / elevated = effusion sign)
• Anconeus muscle: signal, integrity

Intra-articular:

• Joint effusion: absent / minimal / moderate / large
• Loose bodies: absent / present (number, location if visible)
• Osteochondral lesions: location, size, stability signs
• Cartilage: trochlea, capitellum, radial head, trochlear notch

Osseous:

• Medial and lateral epicondyle: cortex, marrow signal
• Capitellum: marrow oedema, osteochondral changes
• Radial head: fracture, impaction, marrow oedema

Technical limitations: fat suppression quality, motion artefact, hardware artefact if present.

7.3 Structured Reporting

Reports should include: Indication (clinical question); Technique (field strength, position, sequences, contrast); Comparison (prior studies); Findings (compartment-by-compartment); Impression (clinically relevant summary); Limitations; Critical communication if required.

7.4 Incidental Findings — Clinical Decision Framework

Usually benign, no follow-up required: small joint effusion without synovitis; mild common extensor tendinopathy in the absence of symptoms referrable to the lateral elbow; minor subcortical cysts at the epicondyles (degenerative); small fatty lipoma in the subcutaneous tissue remote from the joint.

May require clinical correlation: os supratrochleare (uncommon accessory ossicle — document but usually benign); medial epicondyle calcification in a patient without history of epicondylitis (calcific tendinopathy or prior avulsion — clinical correlation needed); moderate joint effusion without obvious cause in an asymptomatic contralateral elbow.

Require explicit communication: aggressive bone lesion features at the distal humerus, proximal ulna, or radial head; nerve signal change outside the cubital tunnel suggesting a more proximal or systemic nerve process; unexpected significant UCL tear in a patient referred for epicondylitis only (changes surgical planning); suspected septic arthritis.

8. MRI Technologist Pearls

8.1 Sequence Order Logic

Recommended acquisition order for standard non-contrast elbow MRI:

1. Three-plane localiser

2. Coronal PD-FS ← most diagnostically critical; acquire first while patient is freshest

3. Coronal T1 ← same prescription; fast acquisition immediately after

4. Axial PD-FS

5. Sagittal PD-FS

6. STIR (coronal or axial)

Rationale: coronal PD-FS is the primary diagnostic sequence. Motion is least in the early part of the examination. Acquiring the coronal first ensures optimal quality for the most important structure — the UCL and common tendon origins.

If contrast is used: STIR before injection; post-contrast T1-FS at 3–5 minutes after injection.

8.2 Positioning Tricks

• For the supine arm-at-side position, place a rolled towel between the arm and body to maintain neutral rotation and prevent involuntary supination-to-pronation drift during the examination.
• For patients who cannot fully extend the elbow due to pain: image at the available extension angle. Document the elbow flexion angle in the report. Do not force extension — motion from pain is worse than mild flexion.
• For the Superman position: ensure the patient is truly comfortable before starting. A pillow under the abdomen and a foam wedge under the extended forearm improve tolerance. Stop the Superman position immediately if the patient reports arm numbness or shoulder pain.
• For bilateral elbow comparison: reposition the coil and replanning images are required between sides. Allow 3–5 minutes for coil repositioning.
• For large patients: the flex coil wrapped around the elbow provides better coverage than a rigid elbow coil for large joint diameters.

8.3 Fast Salvage Protocol

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

8.4 Common Avoidable Errors

9. Quality Control Checklist

• Coronal slices parallel to bicondylar line (verified on axial scout)
• Sagittal slices perpendicular to coronal (verified on axial scout)
• Medial and lateral epicondyles fully included in coronal coverage
• Bicipital tuberosity included in distal extent of coronal and sagittal
• Cubital tunnel fully covered axially (proximal approach to distal exit)
• Radial tunnel and proximal supinator included in axial coverage
• Fat suppression uniform — no large regions of failure on PD-FS
• STIR acquired before gadolinium (if contrast used)
• Post-contrast sequences acquired ≥ 3 minutes after injection (if contrast used)
• No significant motion artefact on coronal PD-FS
• No significant brachial artery pulsation artefact on axial (presaturation applied)
• Metal artefact documented if hardware present; MARS applied if needed
• Correct laterality documented in sequence labels
• Patient forearm position documented if non-standard (elbow flexed, forearm pronated)
• All five mandatory sequences completed

11. Evidence Gaps and Ongoing Debate

Partial UCL tear detection on non-arthrographic MRI: the sensitivity of conventional MRI for partial undersurface UCL tears remains poorly characterised in prospective studies. Published estimates range widely (50–80%) reflecting methodological heterogeneity. Whether optimised 3T non-arthrographic MRI with dedicated high-resolution coronal sequences has sufficiently closed the gap with MR arthrography for clinical decision-making in the non-athlete population is unresolved.

Optimal positioning — supine vs. Superman: no prospective randomised study has compared diagnostic accuracy between supine arm-at-side and Superman positioning at 3T with modern Dixon fat suppression. The practical assumption that Superman positioning provides meaningful diagnostic benefit over Dixon-optimised supine imaging has not been prospectively validated.

3D isotropic vs. 2D sequences at the elbow: comparative accuracy data for 3D isotropic PD-FS versus standard 2D at the elbow are limited. Unlike the shoulder, where several prospective studies exist, elbow 3D studies are predominantly based on expert opinion and small retrospective series.

AI reconstruction and deep learning acceleration: early results for AI-accelerated 3D elbow MRI show promise for reducing acquisition time while maintaining diagnostic quality, but prospective clinical validation for elbow-specific pathology (UCL, common tendons, nerves) is ongoing.

Minimum contrast threshold for MRA: the minimum volume of intra-articular gadolinium needed for reliable partial UCL tear detection and loose body characterisation has not been formally studied.

Nerve MRI quantitative thresholds: cutoff values for ulnar nerve cross-sectional area, T2 signal-to-background ratio, and flattening ratio at the cubital tunnel have been proposed in multiple small series without multicentre validation or standardised reference values.

12. Evidence-Based References

A. Guidelines / Consensus / Society Recommendations

B. Systematic Reviews / Meta-analyses

C. Important Prospective / Original Studies

D. Technical MRI Papers

E. Landmark Historical References

End of document — MRI Elbow Generic Standard Protocol — MRIninja v1.0 — May 2026 This master page is the reference for all future elbow MRI child pages including: MR arthrography, UCL tear specific protocols, lateral and medial epicondylitis deep dive, distal biceps tendon protocol, cubital tunnel syndrome, osteochondral lesions, and post-operative elbow.