Osteochondritis Dissecans of the Knee

Osteochondritis dissecans (OCD) is an acquired, idiopathic lesion of subchondral bone that can affect the overlying articular cartilage. It ranges from softening of the cartilage to complete detachment of a bony fragment as a loose body. The knee is the most common location, accounting for 75% of all OCD.

Who. The classic patient is an active adolescent male athlete aged 10-20 with vague knee pain. Males outnumber females 2:1, and the incidence is 15-30 per 100,000. The lesion is bilateral in 20-30%, so always X-ray the other knee.

Juvenile or adult. OCD is divided by the state of the physis. Juvenile OCD occurs with the physis open and adult OCD with it closed, and the distinction decides both prognosis and treatment.

Cause. The cause is uncertain and multifactorial. Repetitive microtrauma in active adolescents is the most accepted theory, with overuse in sport. The other proposed contributors:

• Vascular insufficiency: ischaemia of the subchondral bone in a watershed area
• Genetic factors: familial clustering is reported, as are epiphyseal anomalies
• Aberrant ossification: an abnormal ossification centre
• Discoid meniscus, associated with lateral lesions

Where. The medial femoral condyle carries 75% of lesions, classically on its lateral aspect: the lateral wall of the intercondylar notch, at the femoral origin of the PCL. The other sites:

• Lateral femoral condyle: 20%
• Patella: 5%
• Trochlea: less than 1%
• Tibial plateau: rare

Why there. During knee motion the tibial spine or the PCL impinges against the lateral aspect of the medial femoral condyle, which is the site of the classic lesion.

The sequence. The disease starts in bone and reaches the joint last, and stability is lost at the step where the cartilage breaks:

1. A subchondral bone event: ischaemia or stress fracture of the subchondral bone
2. The overlying cartilage is initially intact, and the lesion is stable
3. The bone fragment separates from the parent bone
4. A cartilage breach allows fluid in, and the lesion is unstable
5. The fragment detaches completely as a loose body

Lesions are classified by their location on the condyle, by their arthroscopic appearance and by MRI. The arthroscopic and MRI systems both sort lesions into stable and unstable, which is the distinction that drives treatment.

The ROCK classification. The Research in OsteoChondritis of the Knee (ROCK) group's arthroscopic classification is a modern, reproducible system worth knowing alongside the older Guhl/ICRS and Cahill schemes. It divides lesions by probing into immobile (stable) and mobile (unstable) families, because mobility on probing, not appearance alone, drives the choice between drilling and fixation.

Why ROCK matters. It gives a shared vocabulary that correlates with intra-operative stability better than MRI alone, and probing is more reliable than a rim on a juvenile MRI. It also maps directly to treatment: immobile lesions that fail non-operative care are drilled, and mobile lesions need fixation of a salvageable fragment or restoration of an unsalvageable crater.

View larger

History. The pain is vague, poorly localised and activity-related, worse with impact and sport, and its onset is insidious and lasting. An intermittent effusion follows activity. Catching or locking suggests an unstable flap or a loose body.

Examination. Look for quadriceps atrophy if the problem is chronic and a mild to moderate effusion. The gait is antalgic, with the tibia externally rotated to avoid impingement. Tenderness lies over the classic site on the medial femoral condyle with the knee flexed 90°.

Wilson's sign is specific but has low sensitivity.

Radiographs are the initial screening tool. The views:

• AP
• Lateral
• Tunnel (notch): the most sensitive for classic MFC lesions, and the view to obtain when standard AP films are equivocal
• Skyline

What to look for. A crescent-shaped radiolucency, with a rim of sclerosis that suggests chronicity. Look for a calcified loose body in the joint, and record whether the physis is open or closed.

View larger

MRI confirms the diagnosis, assesses stability and plans surgery. The report should give size, location, physeal status, interface fluid or cysts, cartilage breach and loose bodies. Gadolinium can assess blood flow to the fragment but is rarely needed.

View larger

Signs of instability (Hefti criteria).

• Rim sign: high T2 fluid signal behind the fragment, greater than 1mm
• Multiple cysts underlying the bed
• A breach of the cartilage surface allowing fluid contrast
• The fragment elevated or displaced

Adult or child. The rim sign is the most specific criterion in adults, where the classic criteria together are 100% specific. In juveniles it is unreliable: the same criteria are only 11% specific (Kijowski), because a bright T2 rim in a child commonly reflects vascular granulation at a healing interface rather than fluid behind a mobile fragment. In a skeletally immature knee a single high-signal line is not enough; correlate interface fluid, cysts, cartilage breach and displacement.

View larger

View larger

The trap is the distal femoral ossification variant in a child under 10. It resolves with growth, so do not over-treat it.

View larger

The decision. Management turns on the physis and on stability. Diagnose on radiographs and MRI, decide from the physis whether the lesion is juvenile or adult, and assess stability on MRI fluid signal, remembering that the MRI criteria overcall instability in juveniles. The stable juvenile lesion is treated conservatively first. Surgery is indicated for unstable lesions, adult lesions and failed conservative treatment.

Conservative treatment. For the stable juvenile lesion: activity modification, with no impact, for 3-6 months, and serial radiographs or MRI. Success is 50-70%.

View larger

Which stable lesions fail. With a standardised non-operative protocol, roughly two-thirds of stable juvenile lesions heal by six months and about one-third fail, so a stable lesion justifies a non-operative trial but does not guarantee healing. Smaller lesions, less than 2cm, heal better. Size is best expressed relative to the femoral-condyle width, as a normalised ratio, rather than as an absolute millimetre value, because a given lesion is more significant in a smaller condyle; larger normalised lesions are less likely to heal.

Predictors of failure. Beyond a larger normalised size, an effusion and mechanical symptoms (giving-way, catching, locking) at presentation predict failure to heal. Chronological age alone is a weaker predictor than physeal status and these lesion features.

In practice. A small, quiet, stable lesion in an open-physis knee can be watched with serial imaging and activity restriction. A large lesion with an effusion or mechanical symptoms should be counselled as high-risk for failing conservative care and moved earlier toward drilling.

View larger

View larger

Surgery: can the fragment be saved? Salvageability is assessed arthroscopically. With good bone stock and cartilage the aim is preservation: retrograde drilling for a stable lesion, screw or pin fixation for an unstable one. A necrotic or comminuted fragment is removed as a loose body and the defect treated by marrow stimulation (microfracture), or by OATS or ACI for large defects.

Drilling. Indicated for a stable lesion that has failed conservative treatment, usually in a juvenile. Drilling may be retrograde or transarticular.

Retrograde drilling. The goal is to create vascular channels for healing without breaching the cartilage. Under fluoroscopy, a K-wire is drilled from the extra-articular femur into the lesion, avoiding penetration of the articular surface. The channels bring vascularity to the lesion and stimulate a healing response.

View larger

View larger

Internal fixation. Indicated for an unstable but salvageable fragment, a flap. The goal is to compress the fragment onto its bed to allow osseous union. At arthroscopy the fibrous tissue is debrided from the base with a curette and the fragment reduced and fixed, with the heads countersunk below the cartilage surface. The implant is a trade between compression and removal:

• Bioabsorbable nails or pins: no removal needed, lower compression
• Compression screws (Herbert): high compression, may need removal

View larger

Excision and restoration. Indicated for a loose body or a fragmented, necrotic fragment that cannot be saved. The loose body is removed and the defect restored according to its size:

• Microfracture: small defects, less than 2cm²
• OATS (mosaicplasty): medium defects, with cylindrical plugs
• ACI/MACI: large defects, greater than 2-4cm²

Excision creates fibrocartilage, which is less durable.

View larger

View larger

View larger

View larger

Weight bearing depends on what was done:

• Drilling: immediate weight bearing as tolerated, as the lesion is stable
• Fixation: touch-down for 6 weeks to protect the repair
• Grafting: non-weight-bearing for 6 weeks

Motion. Early passive range of motion prevents stiffness, and CPM is often used after cartilage repair.

Return to sport. After conservative treatment, at 3-6 months once MRI shows healing; after surgery, at 6-9 months once CT or MRI shows bone union. The criteria are no pain, full range of motion and radiographic healing.

View larger

View larger

Prognostic factors. Physeal status matters most, and lesion size is discussed under Management. Classic MFC lesions heal better than patellar or trochlear lesions, and stable lesions heal better than unstable ones.

Success rates after surgery.

• Drilling: 80-90% success in the failed conservative juvenile lesion
• Fixation: 80-90% success if the bone quality is good
• Excision: good short-term pain relief but poor long-term results, with a risk of OA

Global Epidemiology

• Incidence roughly 15-30 per 100,000 children and adolescents; rising with year-round single-sport participation worldwide.
• Peak age 10-15 years (juvenile, open physis); a smaller adult cohort presents with closed physes.
• Male predominance approximately 2:1; bilateral in 20-30% (image the contralateral knee).
• Classic site is the lateral aspect of the medial femoral condyle; lateral femoral condyle lesions associate with discoid meniscus.

Side-by-Side Guidance

Registry and Restoration Data

• OCD is a leading indication for cartilage restoration registries (e.g. German KnorpelRegister DGOU, and ICRS-aligned datasets) tracking microfracture, OAT/mosaicplasty, ACI/MACI and osteochondral allograft outcomes.
• Registry signals mirror the literature: fragment preservation and osteochondral restoration outperform excision for long-term joint survival.

High- vs Limited-Resource Practice

Related pages: Osteochondral Defects of the Knee for the cartilage restoration techniques used when the fragment cannot be preserved, and Articular Cartilage Injuries for the underlying repair biology; Osteochondritis Dissecans of the Capitellum and Osteochondral Lesion of the Talus for the same disease at the two other classic sites — note that the talus uses entirely separate size thresholds and that neither site's evidence transfers to the knee; Physeal Injuries: Salter-Harris and Distal Femoral Physeal Injuries for the skeletal maturity that is the single strongest prognostic factor here; Discoid Meniscus and Patellofemoral Instability for the paediatric knee differentials; and Knee Osteoarthritis for the endpoint that fragment excision accelerates.