Getting an MRI report that mentions a bone marrow lesion

Receiving an MRI report that mentions a 'bone marrow lesion' or 'bone marrow oedema' can be unsettling, particularly when the wording is unfamiliar. Both terms describe the same finding: a patch of abnormal signal detected in the subchondral bone — the layer of bone immediately beneath the cartilage surface — on fat-suppressed MRI sequences such as STIR or T2 fat-saturated imaging.

The phrase 'bone marrow oedema' was the earlier label, coined when radiologists first noticed increased fluid-like signal in that region. It has largely been replaced by 'bone marrow lesion' following studies showing that the underlying changes are structural — microfractures in the trabecular bone, localised fibrosis, and altered remodelling — rather than a straightforward fluid shift. In practice, both terms continue to appear on radiology reports and carry the same meaning.

Crucially, this MRI finding is not a diagnosis in itself. Radiologists describe what the scan shows; a clinician then interprets that description alongside your symptoms, how long they have been present, and findings on physical examination. Small BMLs are seen in active people and those without significant knee symptoms, so the presence of the finding alone does not confirm a particular condition or predict a particular outcome. What the report says and what it means for you are two different questions — and the second one needs clinical assessment to answer properly.

Why BMLs are linked to knee pain

Studies tracking hundreds of knee OA patients over several years have established a consistent pattern: BMLs are significantly more prevalent in knees that hurt than in those that do not. The Boston Osteoarthritis of the Knee Study — one of the largest epidemiological cohorts of its kind — found this association to be among the most replicated MRI-pain relationships in OA research. Importantly, the relationship appears dose-dependent: within painful knees, greater BML size tends to correspond with greater pain intensity, suggesting lesion burden matters, not simply lesion presence.

The biological explanation centres on what is happening inside the subchondral bone itself. When trabecular microdamage accumulates, pressure within the bone rises. This elevated intraosseous pressure is thought to stimulate the periosteal and endosteal nociceptors — the nerve endings that supply bone tissue — producing deep, poorly localised pain. Alongside this, local inflammatory mediators including cytokines and prostaglandins appear to amplify pain signalling within the subchondral microenvironment.

Perhaps the most clinically meaningful insight is that BMLs are not fixed findings. They can enlarge, shrink, or resolve over time — and when they do, pain tends to follow. Lesion enlargement has been associated with worsening pain, while natural resolution has corresponded with pain improvement in longitudinal data. This dynamic relationship is a large part of why clinicians take the finding seriously: it suggests BML activity and symptomatic experience are genuinely connected.

That said, not every BML is painful. Lesion size, location, and the surrounding joint environment all influence whether a finding translates into symptoms. The causal direction — whether BMLs drive pain or both arise from the same underlying mechanical insult — remains a subject of ongoing debate.

What BMLs mean for long-term joint health

Beyond their relationship with pain, BMLs carry prognostic weight that extends over months and years. Research has shown that lesion burden in a given compartment independently predicts accelerated loss of the cartilage directly above it — the structural tissue that cushions the joint during movement. Larger, persistent lesions are associated with faster joint space narrowing on X-ray and a higher likelihood of eventually requiring joint replacement, even after accounting for other markers of OA severity.

One way to picture this: if cartilage is the shock-absorbing surface, the subchondral bone is its foundation. A BML signals that the foundation is under significant mechanical stress, and that the overlying surface is at risk of deteriorating more quickly as a result. The lesion is not simply recording today's damage — it is indicating the direction of travel.

For this reason, BML burden has been adopted as a formal endpoint in OA clinical trials, measured alongside pain questionnaires such as WOMAC and KOOS. Researchers track whether treatments reduce lesion size, not just because smaller lesions feel better, but because persistent or enlarging lesions suggest the joint continues to degrade.

For patients, the practical implication is straightforward: a BML that is enlarging or failing to resolve — particularly in the context of ongoing pain and OA — is a meaningful signal that the joint warrants active management rather than a 'wait and see' approach. Early assessment helps establish whether conservative measures, load reduction, or more targeted interventions are appropriate at that stage.

Why BMLs develop: loading, alignment, and subchondral bone

Subchondral bone handles an enormous mechanical workload during everyday movement — absorbing, distributing, and transmitting load between the cartilage surface and the broader skeletal structure. When that load exceeds what the trabecular architecture can tolerate, microfracture and accelerated remodelling follow: this structural breakdown is the substrate from which BMLs arise.

Joint alignment shapes precisely where that stress lands. In knees with varus — a bow-legged posture — the mechanical axis shifts medially, concentrating load on the inner compartment. Medial BMLs are consistently more prevalent in varus knees, and the pattern follows the load rather than chance. The same principle operates in reverse: when alignment is corrected surgically, or when weight loss reduces the mechanical demand placed on the joint, BML burden tends to diminish. The evidence that load-modifying interventions reduce lesion size is one of the stronger arguments that mechanics, not secondary inflammation alone, drives BML formation.

BMLs do not develop in isolation. They typically emerge alongside cartilage damage and meniscal changes as part of the evolving structural picture of OA — which is why those same validated MRI scoring frameworks referenced earlier (WORMS, BLOKS, and the more recent MOAKS) incorporate BML grading as a distinct tissue assessment. Their inclusion in these systems reflects a broad clinical consensus that subchondral pathology is a central feature of OA staging, not an incidental one.

Managing a knee with bone marrow lesions

Conservative management comes first and carries direct structural relevance: reducing excess body weight and correcting lower-limb malalignment both lessen the mechanical demands on affected compartments, and — as established earlier — load reduction demonstrably shrinks BML burden in a meaningful proportion of patients. Activity modification and physiotherapy to strengthen the surrounding musculature reinforce these gains.

When conservative measures prove insufficient, specialist assessment opens a range of more targeted options.

Bone-modifying agents

Bisphosphonates, principally zoledronate, have been studied most extensively in this setting. A single intravenous infusion reduced both pain intensity and BML size at six months in patients whose lesions exceeded 1 cm in trials. The important caveat is that these gains did not persist to 12 months, suggesting a time-limited therapeutic window rather than a durable structural effect. Strontium ranelate has shown a reduction in BML score specifically in patients who had lesions at baseline, pointing to the potential value of MRI-based patient selection before starting any bone-targeted therapy.

Subchondral injection

Sub-chondroplasty — the injection of a calcium phosphate scaffold directly into the area of subchondral bone defect identified on MRI — targets the BML as the primary structural lesion rather than addressing it indirectly. Unlike pharmacological routes, it aims to restore mechanical integrity at the site of failure. Early case series have reported pain relief and functional improvement in patients with persistent subchondral defects, including some who had not responded adequately to conservative care. Larger controlled trials are still in progress, so sub-chondroplasty sits within the emerging rather than established evidence tier — a distinction with practical meaning, since outcomes across patient groups have not yet been fully characterised.

Intra-articular approaches

Intra-articular injections — including hydrogel scaffolds designed to adhere to the synovial membrane and cushion the joint — may reduce BML burden indirectly by altering the mechanical and inflammatory environment within the joint cavity. A 2022 study reported patellofemoral BML reduction following a single polyacrylamide hydrogel injection in advanced knee OA, suggesting that load redistribution at the synovial level has downstream effects on subchondral tissue. This mechanistic route is an active area of investigation rather than an established indication.

No universally agreed BML size threshold currently exists for triggering treatment decisions. In practice, management is guided by symptom burden, whether lesions are enlarging or stable, and the patient's broader OA stage — all factors that require specialist assessment to weigh properly.

When to seek a specialist opinion

A BML reported on an MRI scan is not a verdict — it is a finding that needs to be read alongside the patient's symptoms, activity level, and the broader state of the joint before its significance can be established.

The clearest indication for specialist review is ongoing knee pain that is worsening or beginning to limit everyday activity, particularly where a lesion sits in a loaded compartment. The evidence is consistent: enlarging BMLs track with worsening pain and accelerating cartilage loss, so leaving a symptomatic lesion unassessed carries real structural risk over time.

Assessment can also answer questions that a single scan cannot. A BML visible on one MRI provides no information about whether it is stable or progressing — that determination requires correlating the imaging with symptom trajectory over weeks to months, and sometimes a follow-up scan. A specialist can weigh the lesion against any accompanying cartilage thinning, meniscal damage, or alignment issues to judge whether the structural picture as a whole is advancing.

Not every BML warrants immediate intervention. Incidental findings in younger or more active patients with minimal symptoms may need monitoring and load management rather than treatment; context — symptom burden, lesion size, pain behaviour over time — determines the appropriate response.

The three questions a structured assessment typically aims to resolve are: does this lesion plausibly explain the pain, or is another source more likely? Is it stable, or is there evidence it is enlarging? And given the patient's OA stage and functional goals, which management pathway fits? These are not questions that an imaging report alone can settle — but they are answerable with the right clinical assessment.

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