What a bone marrow lesion actually is

Seeing 'bone marrow oedema' or 'bone marrow lesion' on an MRI report can be alarming — and the fact that both terms are still in use makes it easy to assume something has been written down incorrectly. They describe the same finding. The older label, 'bone marrow oedema' (coined in 1988), suggested the problem was fluid pooling inside the bone, much like swelling in a bruised ankle. Later research showed that picture was misleading, and the term 'bone marrow lesion' — or BML — has largely replaced it in clinical practice, even though radiology reports may still use either phrase.

What is actually happening is more complex than fluid retention. Within the subchondral zone — the thin layer of trabecular bone that sits immediately beneath the articular cartilage — BMLs represent a cluster of overlapping processes: micro-scale fractures within the bone's internal scaffolding, localised inflammatory activity, and areas of fibrosis where normal tissue has been replaced with scar-like material. This is damage and disordered repair, not passive waterlogging.

The location matters because this zone is where the knee absorbs compressive forces during walking, stair-climbing, and other weight-bearing activities — it is the mechanical interface between the cartilage surface above and the deeper medullary bone below. Research has also identified disruption to mesenchymal stem cell populations within BML regions and abnormal patterns of bone remodelling, which distinguishes them from the straightforward degenerative bony changes seen in long-standing arthritis.

On MRI, a BML typically appears as a bright area on fluid-sensitive sequences (T2 or STIR), but the size or brightness of the signal does not reliably predict how much pain a patient experiences — a point that becomes important when considering what to do about them.

Why bone marrow lesions cause pain — and why it can be severe

Pain from a BML often catches patients off guard precisely because it does not follow the expected pattern. Knee pain attributed to worn cartilage typically worsens gradually alongside visible imaging changes — but many patients with BMLs describe a deeper, more persistent ache that seems out of proportion to what their scan shows in terms of cartilage loss. That disconnect has a structural explanation.

The working theory — widely cited but best understood as the most plausible mechanism rather than definitively proven — is that microfractures and localised inflammation within the subchondral bone elevate pressure inside the bone itself. Bone tissue contains pain-sensitive nerve fibres, and when intraosseous pressure rises, those fibres are stimulated. The result can be substantial pain in a knee where cartilage wear alone would not fully account for it. A patient with moderate cartilage loss and large BMLs may experience considerably more pain than someone with more severe cartilage loss but no lesions at all.

This intra-bone origin also helps explain why some patients with prominent BMLs gain only limited or short-lived relief from standard OA injections. Conventional intra-articular injections act on the synovial lining and joint fluid — they are not designed to reach the subchondral bone where the pressure and inflammatory activity are occurring, leaving the deeper pain source largely unaddressed.

The specific mediator pathways involved — including intraosseous nociceptor profiles and prostaglandin activity — remain an active area of research, and direct human data is limited.

Who develops knee BMLs and what drives them

Several different circumstances can lead to a BML, and understanding which applies can help make sense of what comes next.

Osteoarthritis is the most common setting. The majority of people with symptomatic knee OA show BMLs on sensitive MRI sequences — the same mechanical forces that gradually wear cartilage also stress the bone beneath it. Research by Radin and Rose, published in 1986, established a bidirectional relationship here: subchondral bone damage does not simply result from cartilage loss but actively accelerates it, which is why BMLs matter clinically even when cartilage changes appear modest on imaging.

Malalignment — where the knee tracks slightly inward (valgus) or outward (varus) — concentrates compressive load on one compartment. That focal overloading creates the mechanical environment in which BMLs tend to develop and persist, regardless of age or body weight.

Acute injury is a different scenario. A bone contusion from a twisting incident or ligament rupture can produce a BML in an otherwise structurally sound knee; these often resolve over weeks to months with appropriate load management.

Repetitive high-load activity on a knee that already has some structural compromise can prevent that natural resolution — the bone has insufficient recovery time between loading cycles.

A BML on MRI is not a diagnosis on its own. Its significance depends on where it sits, how it correlates with symptoms, and what a physical assessment finds. A scan finding without matching symptoms often warrants monitoring rather than immediate intervention.

How BMLs are assessed and diagnosed

A normal X-ray is one of the most common reasons patients find themselves uncertain about what is causing their knee pain. X-ray shows bone density and joint-space width — it cannot detect the microscopic changes within the subchondral marrow that define a BML. Significant bone-level pathology can coexist with a completely unremarkable plain film.

MRI resolves this. Specialist sequences that suppress the normal fat signal within bone marrow — known as STIR or proton-density fat-saturated sequences — make any abnormal inflammatory change stand out as a bright area on the scan. When a report refers to 'increased signal intensity' or 'bone oedema' beneath the joint surface, this is what those phrases describe: the marrow signal deviating from its normal baseline on these suppression sequences. MRI is considered the gold standard for detecting and characterising BMLs.

Size is noted in most MRI reports and carries clinical weight: larger lesions tend to be associated with higher pain scores and a greater risk of joint deterioration over time. That said, size alone does not determine the treatment path.

A specialist assessment places the scan finding alongside several other factors — the condition of the overlying cartilage, the presence of synovitis, limb alignment, and the patient's functional history and goals. Before imaging, a thorough clinical assessment covers pain location, how symptoms behave under load, prior injury, and what has or has not helped previously. The MRI is one important input into that picture, not a standalone verdict.

Conservative management: reducing load on the bone

Reducing the forces that created the problem is the central logic of conservative BML management — and understanding this helps explain why these measures are more than a holding pattern.

The subchondral bone heals slowly, but it struggles to do so when compressive loading continues at the level that caused the damage in the first place. Every strategy at this stage is, in essence, an attempt to alter that mechanical environment and give the bone's own repair processes the conditions they need.

Unloader bracing can help where varus or valgus malalignment is concentrating force on one compartment. By gently redistributing load across the joint, a well-fitted brace reduces peak stress at the affected subchondral zone — addressing the mechanical driver rather than just the symptoms.

Activity modification during an acute phase is typically recommended, though this does not mean complete rest. High-impact or repetitive-loading activities are usually reduced; low-load movement is generally encouraged, as it maintains circulation and joint nutrition without significantly adding to the compressive burden.

Weight management is directly relevant: the force transmitted through the knee per stride is a multiple of body weight, so even a modest reduction can meaningfully change the loading environment in the subchondral bone.

Targeted strengthening — particularly of the quadriceps and hip abductors — improves how load is shared across the joint, supporting each of the above strategies simultaneously.

How long conservative management takes to produce a result varies, and resolution timelines are not precisely defined by the available evidence; some lesions settle over weeks, others persist considerably longer. When pain remains functionally limiting after a structured conservative period, specialist assessment is appropriate — both to review whether further mechanical modification is possible and to evaluate whether a more targeted intervention is warranted.

Injection and procedural options when conservative care is not enough

The injection landscape shifts when the target is subchondral bone rather than the joint space. Corticosteroid and hyaluronic acid injections act on the synovial lining and intra-articular environment — both can ease OA-related pain, but neither is designed to reach the bone layer where BML pathology sits. That does not make them irrelevant, but it helps explain why some patients with prominent BMLs gain only partial or short-lived relief from these widely available options.

A 2022 study published in the Journal of Arthritis (Maulana, Cole, and Lee) found a reduction in patellofemoral bone marrow lesions following a single intra-articular injection of polyacrylamide hydrogel (Arthrosamid®). The proposed mechanism is indirect: by augmenting the synovial membrane and improving load distribution within the joint, the hydrogel appears to create a mechanical environment more conducive to subchondral healing. This is a single study, and its findings should be read as encouraging rather than definitive.

Sub-chondroplasty takes a more direct route. Calcium phosphate scaffold material is injected into the BML itself under imaging guidance, aiming to restore structural integrity to the damaged trabecular zone — a meaningful distinction from injections that act on the joint space rather than the bone. Published evidence comes primarily from case series and shorter follow-up studies; large randomised controlled trials are limited, and outcomes vary by lesion size, location, and the condition of the surrounding joint. The procedure is typically considered after a meaningful trial of conservative and injection-based management has not produced sufficient relief.

Bisphosphonate medications — drugs developed principally for bone-remodelling conditions such as osteoporosis — have also been investigated in clinical trials for BML volume reduction. Their application to knee BML management remains investigational rather than established MSK practice, and individual suitability involves factors beyond the lesion itself.

Which of these options, if any, is appropriate depends on lesion location, symptom severity, and what previous management has achieved — all of which require specialist assessment to evaluate.

1. [1] Trabecular Oedema (Bone Marrow Lesion). https://en.wikipedia.org/?curid=73122483 https://en.wikipedia.org/?curid=73122483
2. [2] Osteoarthritis. https://en.wikipedia.org/?curid=504841 https://en.wikipedia.org/?curid=504841