MRI Knee – Generic Standard Protocol

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

1. Executive Summary

Knee MRI is one of the most frequently performed musculoskeletal MRI examinations worldwide. Its clinical role is well-defined and evidence-based: it is the gold standard for non-invasive evaluation of internal derangement of the knee, providing unmatched visualisation of the menisci, ligaments, articular cartilage, bone marrow, and periarticular soft tissues — without ionising radiation and without the risks of diagnostic arthroscopy [1, 2].

The standard knee MRI protocol is designed around a core set of multiplanar fat-suppressed fluid-sensitive and non-fat-suppressed anatomical sequences that collectively provide the sensitivity and specificity required for the clinical decisions driving the majority of referrals: meniscal tear characterisation, ligament injury assessment, cartilage evaluation, and bone marrow oedema detection. This protocol is fundamentally different from spinal protocols in its design philosophy: the knee is a peripheral joint with no vital structures, a dedicated small coil is standard, there is no cardiac or respiratory motion concern, and the dominant technical challenges are high-resolution spatial optimisation, fat suppression homogeneity in a small field of view, and metal artefact from surgical hardware.

Compared with radiography, MRI provides soft-tissue characterisation that radiographs cannot. Compared with ultrasound, MRI is superior for intraarticular structures (menisci, ACL, articular cartilage, subchondral bone), bone marrow pathology, and deep structures not accessible by ultrasound. Compared with CT, MRI is superior for soft tissue, cartilage, bone marrow, and ligamentous assessment without radiation; CT retains an advantage for cortical fracture detail, complex fracture pattern mapping, and assessment of metallic hardware artefact.

1.1 Core Strengths

• Meniscal assessment: Direct visualisation of meniscal signal, morphology, tear pattern, and tear type — the most common indication for knee MRI.
• Ligament integrity: ACL, PCL, MCL, LCL, and posterior corner structures assessed non-invasively in all planes.
• Articular cartilage: Direct visualisation of cartilage thickness, signal, and surface integrity — superior to all other non-invasive modalities.
• Bone marrow: Detection of bone marrow oedema (post-traumatic, stress fractures, osteonecrosis, infection) with sensitivity superior to CT and radiography.
• Effusion and synovium: Quantification of joint effusion and assessment of synovial proliferation, pigmented villonodular synovitis, and inflammatory joint disease.
• No ionising radiation: Critical for the predominantly young, active patient population.

1.2 Intrinsic Limitations of the Generic Protocol

Spatial resolution trade-off: Standard protocol parameters balance resolution against acquisition time. Small cartilage lesions (< 5 mm), subtle meniscal tears (especially horizontal cleavage, peripheral tears), and partial ligament tears may approach the resolution limits of a standard non-dedicated protocol.

Post-operative assessment: The standard protocol is significantly limited in post-operative knees — metal artefact from surgical implants (screws, anchors, partial metallic meniscal sutures) requires metal artefact reduction sequences (MARS); post-operative tissue changes (scar, graft remodelling) require specific interpretation adjustments. Dedicated post-operative protocols are child pages.

Cartilage quantification: The generic protocol provides morphological cartilage assessment. Quantitative cartilage mapping (T2 mapping, T1rho, dGEMRIC) for osteoarthritis research or clinical trials requires dedicated sequences not included in the standard protocol.

3D sequences not universally included: 3D isotropic sequences (DESS, CUBE, SPACE, WATS) provide superior spatial resolution and multiplanar capability for cartilage and ligament assessment but are not universally acquired in standard protocols due to acquisition time constraints.

When a dedicated child protocol is required: Post-operative knee (ACL graft, meniscal repair, cartilage repair), MR arthrography (intraarticular contrast for loose body detection, cartilage fissure detection, labral-like structures), dedicated cartilage protocol, suspected bone tumour, suspected infection, and paediatric knee with physeal concerns.

2. Main Clinical Indications

2.1 Standard Indications

Acute knee trauma is the most frequent single indication for urgent knee MRI. ACR Appropriateness Criteria designate MRI without contrast as "usually appropriate" after negative or non-contributory radiographs in acute knee trauma to evaluate meniscal tears, ligament injuries, and bone marrow contusions [1]. Clinical decision rules (Ottawa Knee Rules) can reduce unnecessary radiography; when radiographs are negative and soft tissue injury is suspected, MRI provides definitive structural characterisation. The standard protocol is sufficient for the vast majority of acute trauma presentations.

Chronic knee pain is the dominant outpatient indication. ACR Appropriateness Criteria [2] stratify imaging based on radiographic findings: in patients with normal radiographs or joint effusion, non-contrast MRI is usually appropriate; in patients with OCD, loose bodies, or prior cartilage/meniscal repair, MRI or MR arthrography is appropriate; in patients with radiographic degenerative disease, MRI appropriateness decreases because findings often do not alter surgical decision-making in advanced OA.

ACL and cruciate ligament injury: MRI is the investigation of choice for suspected cruciate ligament rupture, providing direct visualisation of fibre continuity, bone marrow contusions (pivot-shift injury pattern), associated meniscal tears, and posterior capsular injury. The standard protocol is sufficient for complete characterisation of acute and subacute ACL tears and their associated findings.

Meniscal pathology: The standard protocol provides reliable meniscal assessment with high sensitivity (approximately 90–95%) and specificity (approximately 85–92%) for meniscal tears [3]. It is sufficient for most meniscal indications; post-operative meniscal repair assessment may require MR arthrography (child page).

Cartilage evaluation: The standard protocol detects moderate-to-severe cartilage loss reliably; sensitivity for early or superficial cartilage lesions is limited at standard parameters. Dedicated cartilage protocols with thin-slice 3D or high-resolution fat-suppressed sequences are the child page option for cartilage-centric referrals.

Suspected osteonecrosis, bone marrow oedema syndrome: MRI is the most sensitive modality for early osteonecrosis, subchondral insufficiency fractures, and transient bone marrow oedema syndrome. The standard protocol (T1 + PD/T2 fat-suppressed) is sufficient for initial detection and characterisation.

Patellofemoral disorders: Patellofemoral instability, cartilage, patellar tracking, and Hoffa fat pad pathology are all assessable on the standard protocol. Dedicated patellofemoral dynamic assessments are specialised child protocols.

Inflammatory arthropathy: Assessment of synovial proliferation, effusion, cartilage loss, and bone erosions in inflammatory joint disease. Post-contrast sequences are frequently added when inflammatory activity assessment will change management.

2.2 Urgent Red Flags Requiring Expedited or Emergency Imaging

The knee is a peripheral joint without vital structures. There are no emergent knee MRI scenarios comparable to cord compression or cauda equina syndrome. However, the following presentations warrant expedited (same-day to 24 h) imaging:

3. Preparation Reference

3.1 Anatomy-Specific Preparation Items

Prior knee surgery and hardware: Prior surgical history is the most important preparation item for knee MRI. Anterior cruciate ligament reconstruction (interference screws, cortical buttons), meniscal repair (suture anchors), cartilage repair (microfracture, osteochondral autograft/allograft, ACI), partial knee arthroplasty, and total knee arthroplasty (TKA) all modify the expected imaging appearance and the technical approach. Patients with TKA require MARS protocols that are not part of the standard generic protocol. For patients with isolated metallic hardware (screws, anchors), the standard protocol is attempted, and MARS sequences are added if artefact is limiting.

Brace or external device removal: Knee braces, soft bandages, compression wraps, and patellar straps should be removed before the examination. These are not safety hazards but reduce positioning flexibility and may introduce susceptibility artefact at the coil-skin interface if they contain metallic components.

Clothing: Trousers and clothing over the knee should be removed or replaced with examination gown. Metallic fasteners or trouser rivets at the knee level produce focal artefact.

Pain management: Acute knee pain (fracture, haemarthrosis, severe meniscal tear) may limit patient ability to maintain the required position (approximately 15–20° of flexion within the coil). If clinically appropriate, analgesia before positioning improves compliance. Joint aspiration of a tense haemarthrosis before MRI is occasionally performed to relieve pain and improve imaging quality — this is a clinical decision, not a routine radiological recommendation.

Patient history modifying the protocol:

• Prior ACL reconstruction → post-operative protocol; metal artefact assessment
• Prior meniscal repair → MR arthrography consideration for tear vs. normal repair signal
• Prior cartilage repair → dedicated cartilage protocol
• Known malignancy or suspected bone tumour → whole-knee contrast protocol
• Suspected septic arthritis → contrast sequences
• Inflammatory arthropathy workup → post-contrast sequences
• Paediatric physeal concern → modified thin-slice coverage

3.2 Patient Positioning on the MRI System

Patient position: Supine, feet-first entry. The patient lies with the knee extended, in approximately 10–15° of external rotation. This position aligns the ACL with the sagittal imaging plane, improving ACL visualisation — a standard and widely used technique. Full extension minimises in-bore space requirements. Some institutions use slight knee flexion (5–10°) to reduce pressure on the posterior compartment and improve patient comfort during long examinations.

Coil selection: A dedicated multichannel knee coil is mandatory for diagnostic-quality knee MRI. Modern knee coils are 8- to 16-channel phased-array transmit/receive coils that provide uniform SNR coverage of the entire knee compartment. Coil selection is the single most important determinant of image quality in knee MRI. Do NOT attempt knee MRI with a body coil or a non-dedicated coil — image quality will be inadequate for meniscal and cartilage assessment.

Centering: The isocentre should be positioned at the joint line — approximately the level of the inferior patellar pole / proximal tibial plateau. This ensures maximum coil sensitivity at the menisci, cruciate ligaments, and articular cartilage surfaces.

External rotation: Approximately 10–15° of external rotation (natural lie of the lower limb in the supine position) aligns the sagittal imaging plane with the long axis of the ACL and with the medial-to-lateral axis of the menisci. A foam pad under the ankle maintains this rotation during the examination.

Immobilisation: Foam padding within the coil, around the knee, and under the ankle stabilises the limb and reduces motion. Velcro straps securing the coil are helpful. Instruct the patient to remain still and avoid bending or extending the knee during acquisitions.

Comfort strategies: Knee examinations are typically 20–30 minutes in duration. The supine position with the limb in the coil is generally well tolerated. For patients with acute pain, positioning with the knee in mild flexion or with additional padding reduces discomfort.

Pre-scan technologist checks:

1. Verify correct knee coil activation on console.
2. Confirm correct side (right/left) is documented and labelled.
3. Centre isocentre at joint line — verify on localiser.
4. Confirm approximately 10–15° external rotation on the localiser before proceeding.
5. Ensure no metallic objects remain in or near the knee (braces, supports).
6. Acquire three-plane localiser and verify joint line, patellofemoral joint, and entire proximal tibia are within FOV before starting diagnostic sequences.

4. Standard Protocol Design

The knee protocol is fundamentally different from spinal protocols in sequence philosophy: the dominant sequences are fluid-sensitive fat-suppressed (PD-FS or T2-FS), which detect joint fluid, bone marrow oedema, and tissue pathology simultaneously. A non-fat-suppressed sequence (PD or T1) provides the complementary anatomical detail and tissue contrast needed for meniscal signal characterisation and bone marrow baseline assessment.

4.1 Mandatory Core Sequences

Practical note on sequence nomenclature: There is significant variability in published protocols between "PD-weighted fat-saturated" (TE 30–50 ms) and "T2-weighted fat-saturated" (TE 50–80 ms) sequences. Both are acceptable fluid-sensitive sequences; PD-FS tends to give better meniscal signal and is more widely used. T2-FS provides higher fluid-to-tissue contrast. The ESSR and most published institutional protocols use PD-FS or intermediate-weighted FS as the primary fluid-sensitive sequence [4, 5]. This protocol uses "PD-FS" as the primary descriptor while acknowledging T2-FS as a valid alternative.

4.2 Conditional Sequences

4.3 Rationale Summary Per Sequence

Sagittal PD-FS (fat-suppressed proton density) — the primary diagnostic sequence. The sagittal plane provides the best cross-section through the body of the medial and lateral menisci, the cruciate ligaments, the articular cartilage surfaces of the femoral condyles and tibial plateaus, and Hoffa's fat pad. PD weighting (TE 30–50 ms) provides excellent tissue contrast between fibrocartilage (dark), joint fluid (bright), articular cartilage (intermediate-to-bright), hyaline cartilage surface irregularities, and bone marrow (dark with fat suppression).

What it detects well: Meniscal tears (signal extending to articular surface), ACL fibre continuity, posterior cruciate ligament, bone marrow oedema (bright on FS), subchondral bone marrow lesions (BML), articular cartilage surface irregularity, patellar tendon, quadriceps tendon, posterior capsule.

What it misses: Meniscal root tears and radial tears may be suboptimal on sagittal alone (coronal is complementary); subtle cartilage fissures at bone-cartilage margins may require 3D acquisitions.

Technologist note: The sagittal plane is planned parallel to the long axis of the lateral femoral condyle — this aligns the imaging plane with both the menisci and the ACL simultaneously. This is the critical planning step for the sagittal series.

Coronal PD-FS — the collateral ligament and meniscal root sequence. The coronal plane provides the best view of the collateral ligaments (MCL, LCL), the meniscal roots (anterior and posterior roots of both menisci), the posteromedial and posterolateral corners, the articular cartilage of the medial and lateral compartments, and the intercondylar notch.

What it detects well: Medial collateral ligament tears, lateral collateral ligament and fibular head avulsions, meniscal root tears, radial meniscal tears, joint line medial or lateral compartment cartilage loss, tibial bone marrow oedema patterns, posterolateral corner injuries.

Limitation: ACL/PCL are poorly assessed in the strict coronal plane; sagittal is primary for cruciate assessment.

Axial PD-FS — the patellofemoral and peripatellar sequence. The axial plane provides cross-sectional assessment of the patella, trochlear groove, patellar cartilage, trochlear cartilage, medial and lateral retinaculum, patellar tendon, suprapatellar bursa, and periarticular soft tissue.

What it detects well: Patellar articular cartilage (primary plane), trochlear groove morphology and dysplasia, patellar maltracking, retinacular injuries (medial patellar retinaculum in lateral patellar dislocation), suprapatellar effusion, periarticular soft tissue pathology.

Limitation: The menisci are not well assessed in the axial plane (sagittal and coronal are primary).

Coronal T1 (non-fat-suppressed) — the anatomical and bone marrow baseline sequence. T1-weighted non-fat-suppressed imaging provides bright marrow signal as the baseline against which T1-dark bone marrow lesions (oedema, infiltration, AVN, fractures) are identified. It also provides excellent tissue contrast for meniscal signal characterisation — menisci appear dark on T1 (as expected for fibrocartilage) and any T1 signal within the meniscal substance suggests intrameniscal degeneration or myxoid change.

What it detects well: Bone marrow T1 signal characterisation (marrow fat replacement, lesion characterisation), anatomical survey of joint, cortical bone integrity, periarticular osseous structures.

Limitation: Joint fluid is not bright on T1; inflammatory oedema is not bright on non-FS T1.

Sagittal PD (non-FS) — alternative to coronal T1 in some protocols. Some departments use sagittal non-FS PD in lieu of or in addition to coronal T1. The non-FS sagittal provides meniscal signal characterisation complementary to the fat-suppressed sagittal, as the non-suppressed background allows reliable grading of intrameniscal degeneration on the established Stoller grading system.

4.4 Sequence Matching and Cross-Sequence Consistency

The three fat-suppressed sequences (sagittal, coronal, axial PD-FS) do not need to share identical geometry — they are in orthogonal planes by definition. However, the FOV and coverage of all sequences must collectively ensure that every anatomical structure of clinical relevance is covered by at least two orthogonal sequences.

The non-fat-suppressed sequence (coronal T1 or sagittal PD) should match the corresponding fat-suppressed sequence in plane, angulation, FOV, and slice thickness so that direct signal comparison is possible: identifying a finding on fat-suppressed images and then assessing its T1 signal on the matching non-FS sequence is a standard diagnostic workflow.

Pre-/post-contrast matching: When contrast is administered, pre-contrast T1 fat-suppressed sequences must be acquired before injection and must precisely match post-contrast sequences in all geometric parameters. This is essential for enhancement characterisation in synovial disease, tumour, and infection.

Serial follow-up: For post-operative monitoring, cartilage repair follow-up, and OCD monitoring, identical geometric parameters, coil configuration, and field strength must be maintained across examinations. This is particularly important for cartilage lesion size estimation and meniscal repair signal evolution.

4.5 Fat Suppression in Knee MRI

Fat suppression is central to knee MRI protocol design — fundamentally more important in the knee than in any spinal region. The dominant clinical sequences are fat-suppressed, and fat suppression quality directly determines diagnostic sensitivity.

Why fat suppression is critical in the knee:

• Bone marrow oedema (the most common finding in acute knee trauma) is only reliably detected on fat-suppressed images because the bright marrow fat signal on non-FS sequences masks the subtle T2 prolongation of oedema
• Joint fluid (bright on T2-FS/PD-FS) provides the natural arthrographic effect that outlines cartilage surfaces and meniscal margins — essential for tear detection
• Hoffa's fat pad pathology (inflammation, fibrosis) is only visible against fat-suppressed background

Preferred fat suppression technique for knee MRI:

Spectral fat saturation (SPIR/SPAIR/ChemSat) is the most commonly used technique in knee MRI. At both 1.5T and 3T, the small FOV of the knee (14–18 cm) generally provides adequate B0 homogeneity for reliable spectral fat saturation, unlike the large FOV challenges encountered in spinal imaging. SPAIR is preferred over CHESS at 3T for better B0 robustness within the knee joint.

Dixon technique provides the most robust fat suppression and is increasingly the preferred approach at 3T for its B0-independence. Dixon T2 or PD sequences for the knee generate excellent water-only images with uniform fat suppression across the joint, even in patients with metallic hardware nearby.

STIR: Reliable fat suppression independent of B0; lower SNR than SPIR/SPAIR. Used in the knee when spectral fat saturation fails (adjacent metallic hardware, far from isocentre). A critical consideration: STIR cannot be used post-gadolinium — the same principle as spinal protocols.

Critical fat suppression pitfall: Incomplete fat suppression in the knee manifests as residual bright subcutaneous fat or bone marrow fat signal on PD-FS images. This reduces sensitivity for bone marrow oedema (which appears bright only when fat is suppressed). The technologist must verify uniform fat suppression on the first acquired image before proceeding with the full protocol. If fat suppression fails, shim adjustment or technique change (STIR, Dixon) is required before re-acquisition.

Fat suppression is not applied to: standard coronal T1 (bright marrow fat is the diagnostic signal); standard sagittal PD non-FS (meniscal signal characterisation uses non-suppressed background).

5. Optimisation Strategy

5.1 Artifact Reduction by Source

Popliteal artery pulsation artefact is the dominant motion artefact source specific to knee MRI. The popliteal artery runs immediately posterior to the knee joint, generating periodic phase-encoding ghosting that propagates through the posterior joint, potentially simulating posterior horn meniscal pathology, PCL signal change, or posterior capsular injury.

Physical cause: Cardiac-synchronous arterial pulsation of the popliteal artery produces phase shifts in the phase-encoding direction. Appearance: Ghost images of the popliteal artery displaced along the phase encoding direction at intervals proportional to the TR/cardiac cycle ratio. Reduction strategies: R-L phase encoding for axial sequences (displaces ghosts laterally); S-I phase encoding for sagittal and coronal sequences (displaces ghosts cranially/caudally); anterior or posterior saturation bands adjacent to the popliteal fossa; reducing TR reduces the number and spacing of ghost replications.

Fat suppression failure is the most common quality failure in knee MRI and the most clinically consequential. Incomplete fat suppression on PD-FS sequences reduces sensitivity for bone marrow oedema and joint fluid characterisation.

Physical cause: B0 inhomogeneity shifts the fat resonance frequency away from the saturation pulse frequency. More common at 3T than 1.5T; more common at the periphery of the FOV and near metallic hardware. Appearance: Residual bright subcutaneous fat or bone marrow fat signal on fat-suppressed sequences. Reduction strategies: Shimming restricted to the knee joint volume; use SPAIR instead of CHESS at 3T; use Dixon technique for most homogeneous suppression; use STIR in patients with hardware or at 1.5T if CHESS is unreliable; verify fat suppression on first image before proceeding.

Metal susceptibility artefact from surgical hardware is the second most common quality problem. Titanium (interference screws, cortical buttons) produces less artefact than stainless steel but still causes signal void and geometric distortion adjacent to the implant.

Physical cause: Local B0 field distortion from ferromagnetic or paramagnetic implants. Appearance: Signal void at the implant site with surrounding artefact radiating in the frequency-encoding direction; geometric distortion. Reduction strategies: Increase bandwidth (reduces frequency-encoding displacement per susceptibility unit); use TSE (less susceptible than GRE); use MARS sequences (SEMAC/WARP/MAVRIC-SL) when standard protocol is non-diagnostic; 1.5T preferred over 3T for severe hardware artefact. For sequence-level protocol optimisation, vendor terminology and artefact management, see the dedicated MRIninja page Turbo Spin Echo (TSE/FSE) Sequence.

Chemical shift artefact at cartilage-fluid-fat interfaces (particularly at subchondral bone and periarticular fat) can simulate cartilage lesions or obscure true lesions.

Reduction: Adequate bandwidth; SPAIR or Dixon fat suppression provides good chemical shift management.

Motion artefact: Generally low in knee MRI as the patient is comfortable and the joint is immobilised in the coil. Involuntary knee movement (pain, discomfort) produces ghosting in the phase direction. Adequate padding and comfortable positioning before starting reduces this.

Truncation artefact (Gibbs ringing): High-contrast interfaces (cartilage-fluid, tendon-fluid) may generate oscillating signal bands simulating cartilage lesions or intrasubstance tendon signal. Adequate matrix size minimises this.

5.2 Protocol Efficiency and Throughput

Standard routine protocol: Sagittal PD-FS + Sagittal PD (or Coronal T1) + Coronal PD-FS + Axial PD-FS = approximately 15–25 minutes depending on resolution, parallel imaging, and number of sequences.

Premium protocol (cartilage focus): Standard protocol + 3D isotropic sequence (CUBE/SPACE/DESS/VISTA) + coronal oblique ACL = approximately 25–35 minutes.

Abbreviated protocol: Sagittal PD-FS + Coronal PD-FS alone (10–12 minutes). Evidence exists for abbreviated protocols with acceptable diagnostic accuracy for specific indications (meniscal tears, ligament injuries) [reviewed in 6]. The clinical utility depends on indication and patient population.

3D sequences: 3D isotropic knee MRI (CUBE/SPACE/VISTA at 0.5–1 mm isotropic) has demonstrated comparable or superior performance to 2D for cartilage assessment and ligament evaluation while providing multiplanar reformatting capability. Evidence from prospective studies supports their integration into clinical protocols [7]. However, 3D acquisitions are more susceptible to motion artefact and cost additional time (6–10 minutes per 3D volume).

AI-accelerated reconstruction: DNN-based reconstruction has demonstrated 2–4× scan time reduction with maintained diagnostic quality in prospective multi-vendor studies [8]. This emerging technology is enabling clinically practical 3-minute knee protocols without sacrificing diagnostic accuracy [9].

5.3 Field Strength Considerations

Clinical recommendation: 3T is preferred for cartilage-centric protocols, subtle meniscal pathology, and high-resolution ligament assessment. 1.5T is preferred for post-operative knees with metallic hardware. For routine internal derangement assessment, both field strengths are clinically diagnostic.

6. Contrast Use Principles Specific to Knee MRI

6.1 Non-Contrast Standard Protocol — Sufficient For

Non-contrast MRI is diagnostically sufficient for the vast majority of knee MRI indications: acute and chronic meniscal tears, ACL/PCL tears, collateral ligament injuries, bone marrow oedema, subchondral bone lesions, OCD, articular cartilage morphological assessment, effusion, patellofemoral disorders, occult fractures, and osteonecrosis.

6.2 Gadolinium Indicated — Knee-Specific Contexts

Intravenous GBCA:

• Suspected septic arthritis: Enhancement of synovial membrane and periarticular soft tissues characterises activity and extent; joint aspiration is still the definitive investigation.
• Suspected primary or metastatic bone tumour: Enhancement is essential for tumour characterisation, extent, and internal architecture.
• Inflammatory arthropathy (synovitis assessment): Post-contrast T1 fat-suppressed characterises synovial proliferation, pannus, and joint inflammation activity when this will change management.
• Pigmented villonodular synovitis (PVNS/TGCT): Enhancement pattern and extent assessment.
• Suspected avascular necrosis (advanced staging): Post-contrast enhancement delineation of viable and non-viable zones.

Intraarticular contrast (MR arthrography): MR arthrography with diluted intraarticular gadolinium (or saline-only "indirect arthrography") is a conditional technique for:

• Post-operative meniscal repair: The non-contrast standard protocol cannot reliably distinguish a healed repair from a recurrent tear; intraarticular contrast extends into tear gaps.
• Loose body detection and localisation.
• Suspected cartilage flap or fissure (direct MRA is more sensitive than standard MRI for cartilage surface lesions).

6.3 Post-Contrast Acquisition Timing

For intravenous GBCA: standard timing 3–5 minutes post-injection. In synovial disease evaluation, early imaging (< 5 minutes) captures active enhancement; delayed imaging (10–20 minutes) allows discrimination of fibrous from inflammatory tissue.

For MR arthrography: diagnostic sequences are acquired immediately after joint distension (within 15–30 minutes before contrast is absorbed). Patient should perform 5–10 minutes of mild weight-bearing activity to distribute the contrast solution within the joint.

7. Reporting Essentials

7.1 Interpretation Framework

Knee MRI is inherently a compartmental analysis: the knee is divided into the medial tibiofemoral compartment, lateral tibiofemoral compartment, and patellofemoral compartment. Each compartment is assessed systematically. The axis of clinical interpretation depends on the indication:

Traumatic presentation (acute injury, twisting mechanism):

• Bone marrow: contusion pattern predicts mechanism (pivot shift = lateral BML + ACL tear; hyperextension = anterior tibial BML + PCL tear)
• Cruciate ligaments: continuity, signal, fibre orientation
• Menisci: peripheral, mid-body, and body tears; root tears
• Collateral ligaments and corners

Degenerative/chronic pain presentation:

• Menisci: degenerative horizontal cleavage, radial tears, root tears, complex tears
• Cartilage: compartmental cartilage loss grading
• Subchondral bone: BMLs, subchondral cysts
• Periarticular pathology: tendinopathy, bursitis, cysts

7.2 Mandatory Reporting Checklist

Medial tibiofemoral compartment:

• [ ] Medial meniscus: body, anterior horn, posterior horn, meniscal roots; tear type and location
• [ ] Medial collateral ligament: deep and superficial layers, femoral and tibial attachments
• [ ] Medial articular cartilage: femoral and tibial surfaces; grade, location
• [ ] Medial bone marrow: contusions, BMLs, subchondral cysts

Lateral tibiofemoral compartment:

• [ ] Lateral meniscus: body, anterior horn, posterior horn (including popliteal hiatus); root tears
• [ ] Lateral collateral ligament, iliotibial band, biceps femoris attachment
• [ ] Fibular head: styloid process (arcuate sign), fibular head avulsion
• [ ] Lateral articular cartilage; lateral bone marrow

Cruciate ligaments:

• [ ] ACL: femoral attachment, midsubstance, tibial attachment; fibre orientation and signal
• [ ] PCL: morphology, signal, femoral and tibial attachments

Patellofemoral compartment:

• [ ] Patellar cartilage (all facets): grade, size, location
• [ ] Trochlear cartilage: grade
• [ ] Patellar position and tilt (axial)
• [ ] Retinacular structures: medial patellofemoral ligament (MPFL)

Tendon and soft tissue:

• [ ] Patellar tendon
• [ ] Quadriceps tendon
• [ ] Hoffa fat pad signal
• [ ] Posterolateral corner (LCL, popliteofibular ligament, popliteus complex)
• [ ] Posteromedial capsule

Osseous and periarticular:

• [ ] Tibial plateau: cortical integrity, fractures
• [ ] Femoral condyles: subchondral integrity
• [ ] Patella: osseous anatomy, bipartite patella
• [ ] Proximal fibula

Joint cavity:

• [ ] Effusion: size and character
• [ ] Synovial thickening or nodularity
• [ ] Loose bodies: location and number

Technical items:

• [ ] Fat suppression quality
• [ ] Metal artefact if present and impact
• [ ] Side (right/left) correctly labelled

7.3 Structured Reporting

Indication → Technique (field strength, sequences, coil, side, contrast if used) → Comparison with prior studies → Findings (systematic by compartment as above) → Impression (concise, answering the clinical question) → Limitations → Critical communication if needed.

7.4 Incidental Findings — Clinical Decision Framework

Usually benign, no action required: Typical meniscal degeneration without surface extension (grade 1–2 intrameniscal signal); small subchondral cysts without associated cartilage loss; medial plica (if non-inflamed); perimeniscal cyst (if small and asymptomatic context); bipartite patella (if non-tender); popliteal cyst (Baker's cyst — note and describe, no further action unless complicated).

Requires documentation and follow-up: OCD lesion (stability characterisation; refer for clinical correlation and potential treatment planning); subchondral insufficiency fracture (clinical fracture risk assessment; orthopaedic referral); intraosseous ganglion cyst (document; exclude aggressive features).

Urgent/clinically important: Unexpected bone tumour or aggressive bony lesion — direct communication; displaced meniscal fragment causing locked knee (mechanical issue requiring surgical consultation); suspected pathological fracture — communication; unexpected infection findings.

8. MRI Technologist Pearls

8.1 Sequence Order Logic

Recommended standard order:

1. Three-plane localiser
2. Sagittal PD-FS — acquired first as the primary sequence; used as reference for other plane planning
3. Sagittal PD (non-FS) or Coronal T1 — anatomical reference sequence
4. Coronal PD-FS — planned from axial localiser after verifying sagittal quality
5. Axial PD-FS — final mandatory sequence

Rationale: Sagittal PD-FS is the highest diagnostic priority sequence and is acquired when the patient is freshest and most cooperative. If the patient cannot complete the examination, the sagittal series provides the maximum diagnostic yield.

8.2 Positioning Tricks

• External rotation: ensure approximately 15° external rotation (natural supine limb position) before closing the coil — this aligns the sagittal plane with the ACL.
• Foam padding under ankle: stabilises external rotation and supports the limb during long acquisitions without discomfort.
• Verify fat suppression on first sequence: immediately review the first sequence on the console to confirm fat is uniformly dark before proceeding. If fat is bright, adjust shim and repeat.
• Correct side labelling: mark right or left before starting — this is the most consequential labelling error in MSK imaging. Label on the localiser acquisition before starting diagnostic sequences.
• Posterior saturation band for popliteal artery: apply a saturation band over the popliteal fossa for all three planes to reduce pulsation ghosting.

8.3 Fast Salvage Protocol

Core minimum (two-sequence): Sagittal PD-FS + Coronal PD-FS = 6–10 minutes; adequate for the majority of internal derangement questions.

8.4 Common Avoidable Errors

9. Quality Control Checklist

Coverage:

• [ ] Sagittal: full width from medial to lateral capsule; both collateral ligaments visible on extreme slices
• [ ] Coronal: full extent from patella anteriorly to 2 cm posterior to femoral condyles
• [ ] Axial: full patella superiorly to tibial tuberosity inferiorly; fibular head visible

Image quality:

• [ ] Fat suppression uniform on all fat-saturated sequences (subcutaneous fat and bone marrow uniformly dark)
• [ ] No major motion artefact
• [ ] Popliteal artery pulsation ghosting minimised or absent
• [ ] No significant metal artefact obscuring critical structures (if hardware present: document extent)
• [ ] Menisci clearly defined on sagittal PD-FS
• [ ] Cruciate ligaments visible on sagittal
• [ ] Articular cartilage surfaces clearly delineated

Sequence completeness:

• [ ] Sagittal PD-FS: acquired, reviewed
• [ ] Coronal PD-FS (or T2-FS): acquired
• [ ] Axial PD-FS (or T2-FS): acquired
• [ ] Coronal T1 (or sagittal PD non-FS): acquired
• [ ] Any additional sequences per indication completed

Contrast (if used):

• [ ] Pre-contrast T1-FS acquired before injection
• [ ] Injection time documented
• [ ] Post-contrast timing documented

Labelling:

• [ ] Right/left side correctly labelled on all series
• [ ] Patient identifiers correct
• [ ] Series labels correct on PACS (not default numbers)

11. Evidence Gaps and Ongoing Debate

2D vs. 3D equivalence for routine clinical knee MRI: Evidence generally supports comparable diagnostic performance, but most studies are single-centre and use expert readers. Multi-site real-world equivalence, particularly for subtle meniscal pathology and cartilage fissures, is less well established.

Optimal TE for knee MRI (PD vs. T2-FS): No high-quality randomised study has demonstrated clinical superiority of PD-FS (TE 30–50 ms) over T2-FS (TE 50–80 ms) for the full range of knee pathology. Both approaches are widely used; the PD-FS preference is based on soft tissue SNR efficiency and established meniscal grading calibration, but not definitive comparative evidence.

AI-accelerated knee protocols: DNN reconstruction and compressed sensing enable substantial scan time reduction with maintained image quality in early prospective studies, but multisite validation across real-world patient populations (including motion, metal, body habitus variation) is incomplete.

Field strength for specific indications: Whether 3T systematically outperforms 1.5T for clinical outcomes (not just image quality scores) in routine internal derangement assessment is not established by randomised comparative data.

Abbreviated protocols: Evidence supports abbreviated protocols for specific indications, but the clinical safety margin of omitting sequences in a general population protocol is uncertain.

MR arthrography vs. non-contrast 3T: The diagnostic advantage of direct MR arthrography over high-field 3T non-contrast MRI for post-operative meniscal repair assessment remains actively debated, with some evidence suggesting 3T non-contrast may be equivalent for experienced readers.

Reporting thresholds for meniscal degeneration vs. tear: The clinical significance of grade 2 meniscal signal (not surface-extending) in older patients and its appropriate reporting threshold is an area of ongoing expert debate.

12. Evidence-Based References

A. Guidelines / Consensus / Society Recommendations

[1] ACR Expert Panel on Musculoskeletal Imaging. ACR Appropriateness Criteria® Acute Trauma to the Knee. J Am Coll Radiol. 2012;9(2):107–109. PMID: 22305695. (Evidence Level: High — Guideline) Relevance: Primary ACR guidance for acute knee trauma imaging.

[2] Fox MG, Chang EY, Amini B, et al. ACR Appropriateness Criteria® Chronic Knee Pain. J Am Coll Radiol. 2018;15(11S):S302–S312. PMID: 30392599. (Evidence Level: High — Guideline) Relevance: Stratified MRI appropriateness for chronic knee pain by radiographic findings.

[4] ESSR Musculoskeletal Working Group. ESSR MRI Protocols — Knee. European Society of Musculoskeletal Radiology. (Evidence Level: High — Consensus protocol) Relevance: European consensus reference protocol for knee MRI positioning and parameters.

B. Systematic Reviews / Meta-analyses

[3] Oei EH, Nikken JJ, Verstijnen AC, Ginai AZ, Myriam Hunink MG. MR imaging of the menisci and cruciate ligaments: a systematic review. Radiology. 2003;226(3):837–848. PMID: 12601218. DOI: 10.1148/radiol.2263011892. (Evidence Level: High — Systematic review) Relevance: Establishes sensitivity/specificity of knee MRI for meniscal and cruciate assessment.

C. Important Original Studies

[6] Fox MG (Global Reading Room). Knee MRI Protocols. AJR Am J Roentgenol. 2021;217(5). DOI: 10.2214/AJR.21.27238. (Evidence Level: Moderate — Multi-institutional survey) Relevance: Documents real-world protocol variation at major academic centres.

[7] Ristow O, Steinbach L, Sabo G, et al. Isotropic 3D fast spin-echo imaging versus standard 2D imaging at 3.0 T of the knee. Eur Radiol. 2009;19(5):1263–1272. PMID: 19125252. DOI: 10.1007/s00330-008-1265-3. (Evidence Level: Moderate — Prospective comparative study) Relevance: 3D vs 2D comparative evidence for clinical knee MRI.

[8] Kim H, et al. Highly accelerated knee MRI using DNN-based reconstruction. Sci Rep. 2023;13(1):17476. PMC: 10570285. DOI: 10.1038/s41598-023-44271-4. (Evidence Level: Moderate — Prospective multi-vendor study) Relevance: DNN acceleration enables 44–50% time reduction with maintained diagnostic quality.

[9] Becker M, et al. 3-Minute Knee MRI Protocol with AI SuperResolution Reconstruction. Diagnostics. 2025;15(10):1206. DOI: 10.3390/diagnostics15101206. (Evidence Level: Low/Technical — Prospective feasibility) Relevance: Demonstrates clinical feasibility of ultra-abbreviated knee protocols.

D. Technical MRI Papers

[He-2024] He M, et al. Design of a sample rapid MRI acquisition protocol for knee osteoarthritis. Osteoarthritis Cartilage Open. 2024. DOI: 10.1016/j.ocarto.2024.100472. (Evidence Level: Technical) Relevance: Contemporary parameter optimisation for knee MRI in clinical trials.

E. Landmark Historical References

[Stoller] Stoller DW. Magnetic Resonance Imaging in Orthopaedics and Sports Medicine. 3rd ed. Lippincott Williams & Wilkins. 2007. Relevance: Foundational reference establishing meniscal signal grading system (grades 1–3) universally used in clinical knee MRI reporting; basis of current reporting conventions.

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

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