What does a bone marrow lesion on my MRI actually mean

Seeing the words “bone marrow lesion” (or the older phrase “bone marrow oedema”) on a knee MRI report can sound alarming, but in osteoarthritis it usually refers to a stressed patch of bone just beneath the joint surface (the subchondral bone), not a tumour. In plain English, the radiologist is typically describing an area that looks “bruised and overworked” under worn cartilage, rather than a neat, fluid-filled cyst.

On the scan itself, osteoarthritis-related bone marrow lesions tend to appear as blurred, bright areas on fluid-sensitive sequences (fat-suppressed T2 or proton-density imaging), with a lower or intermediate signal on T1 images. That pattern is one reason the term “lesion” is preferred: it points to an abnormal marrow compartment, not simply extra water in the bone.

Tissue studies of these MRI-defined regions show the changes are mixed and active: tiny microfractures and thickened trabecular bone, fibrosis, small subchondral cysts, increased blood vessels, low-grade inflammation, and more nerve fibres around vessels than in unaffected bone. This fits with the idea that osteoarthritis can involve a “bone–cartilage unit”, where the bone and cartilage influence each other under repeated mechanical load.

Although these lesions are common in knee osteoarthritis, similar MRI appearances can occur after injury and in other conditions, so the finding needs to be weighed alongside symptoms, examination findings, and the rest of the report.

How closely are bone marrow lesions linked to knee pain

In knee osteoarthritis, bone marrow lesion (BML) burden is one of the MRI findings most consistently linked with how much the knee hurts. Across large cohort studies summarised in major reviews, people with larger or more numerous BMLs generally report more frequent and more severe knee pain than those with small or absent lesions, and this relationship can be stronger than the link between pain and X‑ray severity alone. That evidence is more direct than the “is this why it hurts?” framing suggests: BMLs are often a meaningful clue rather than a vague incidental comment on a report.

The exceptions are clinically important and show why an MRI cannot be read as a simple pain map. Reviews describe patterns such as severe radiographic osteoarthritis with relatively little pain when BMLs are minimal, versus marked symptoms in someone whose X‑ray changes are only moderate but whose MRI shows prominent BMLs. In other words, the same Kellgren–Lawrence stage on X‑ray can sit alongside very different symptom levels depending (in part) on subchondral changes seen on MRI.

Follow‑up MRI data also support a dynamic link: when BMLs enlarge over time, pain often worsens; when they shrink or fully resolve, pain frequently improves. This “move together” pattern is one reason BMLs are treated as potentially modifiable, not just a static scar.

Biology helps explain the association. Histology from MRI‑defined BML regions (including work using the OABS system) has found hypervascularity and increased perivascular nerve fibres (for example, PGP9.5‑positive nerves) alongside microdamage and remodelling changes—features that can plausibly generate pain in load‑bearing bone, unlike articular cartilage which has no nerves. Even so, the overall pain picture can still be shaped by other osteoarthritis features discussed in reviews (such as synovitis/effusion, meniscal pathology, muscle function, and pain sensitisation), so a sizeable BML is best seen as one of the most plausible MRI contributors to pain rather than the sole explanation.

Do bone marrow lesions tell me how my osteoarthritis will progress

What BMLs can—and can’t—predict

Longitudinal MRI evidence summarised in major reviews links baseline bone marrow lesions (BMLs) with a higher chance of later cartilage loss and joint‑space narrowing than in otherwise similar knees without BMLs. The risk signal is stronger when lesions are larger or present in multiple areas, which is why BML “burden” is often treated as a marker of more active structural disease rather than a harmless MRI footnote.

At a population level, greater BML burden has also been associated with faster overall osteoarthritis progression and a higher likelihood of ultimately needing knee joint replacement in follow‑up. That said, even a “high‑risk” MRI pattern does not translate into a precise personal timetable (for example, whether surgery is likely in 1 year versus 10 years) because symptoms, alignment, muscle function, activity demands, and other joint changes also influence outcomes.

These lesions are not fixed scars. Follow-up MRI studies show that BML volumes can change over time, indicating they may expand or regress alongside shifts in mechanical loading and/or treatment.

A central uncertainty remains in the recent syntheses: BMLs may be both a driver of progression and a marker of an overloaded, already‑damaged joint, and the balance probably varies by person and context. In practical terms, a large, persistent BML in a weight‑bearing region is often interpreted as a red flag for stressed subchondral bone—reason enough for more focused risk‑factor work, closer review, and timely discussion of options, without implying that surgery is inevitable.

How do bone marrow lesions fit into non surgical knee OA care

A bone marrow lesion (BML) on MRI doesn’t usually rewrite the non‑surgical osteoarthritis (OA) plan; it tends to change the emphasis. The core pathway set out in international guidance (including OARSI and the ACR/Arthritis Foundation) and echoed in a JAMA review still centres on education and self‑management, structured land‑based exercise, weight management where relevant, and sensible use of pain relief—without a separate “BML-specific” algorithm.

Same pathway, different emphasis on load

Where a BML is present, clinicians often focus harder on reducing focal subchondral stress while keeping the knee moving. That can mean stepping down high‑impact loading (for example repeated jumping or hard downhill walking), pacing flare‑provoking tasks, and using support measures when indicated—such as a walking aid for longer distances, or a brace aimed at unloading a more affected compartment. Physiotherapy commonly targets gait and alignment strategies, alongside progressive strengthening of the quadriceps and hip muscles and graded aerobic work.

Why “non–weight‑bearing for weeks” advice online may not apply

People searching the older term “bone marrow oedema” often find protocols for non‑OA bone marrow oedema syndromes. A 2024 narrative review describes conservative first‑line care in those settings as no weight‑bearing for 3 to 6 weeks plus analgesics/anti‑inflammatories. In chronic knee OA, however, strict unloading is not routinely recommended because it can drive deconditioning; relative load modification is usually preferred.

Alongside exercise and load management, standard symptom tools remain in play—topical or oral NSAIDs where appropriate, and occasional intra‑articular corticosteroid or hyaluronic acid injections—used as support rather than a replacement for the guideline basics.

Are there treatments that directly target bone marrow lesions

Several approaches have been tested with the specific aim of reducing a knee osteoarthritis (OA) bone marrow lesion (BML) on MRI. The practical framing here is a hierarchy rather than a menu: systemic “bone drugs” and procedures are still experimental, while the most reliable way to influence a BML signal remains changing the loads the joint sees over time (the core OA plan already covered, just applied more deliberately to the painful compartment).

Bone-active medicines (e.g. bisphosphonates)

Trials have looked at drugs that alter bone turnover, including bisphosphonates such as zoledronic acid, in people with knee OA who also have BMLs. Reviews report that some studies found short-term reductions in BML size and pain, but results are mixed, and potential downsides of systemic therapy (including side-effects and suitability) limit routine use. These agents are not part of standard OA care pathways or major guidelines at present.

Targeted load reduction (biomechanics and bracing)

Biomechanical strategies—such as interventions intended to shift forces away from the most affected compartment—are used because BMLs are thought to reflect focal subchondral stress. Evidence for BML-specific benefit exists but is generally small and heterogeneous, and it is often difficult to separate “BML shrinkage” from overall symptom improvement and activity change across months of care.

Extracorporeal shock wave therapy (ESWT)

ESWT has been studied for painful BMLs, with some small studies reporting improvements in pain and MRI appearance. A key limitation noted in conservative-treatment reviews is that protocols vary (energy, number of sessions, targeting), and overall trial quality remains limited, so it has not become a mainstream OA recommendation.

Intraosseous “regenerative” injections

Image-guided injections delivered into the BML region (for example platelet-rich plasma (PRP) or bone-marrow–derived cell concentrates) are being explored in early-phase and often uncontrolled studies. Early reports suggest possible improvements in pain and MRI findings, but the evidence base is still low certainty, and these techniques are not endorsed in current OA guidelines.

Across all these BML-focused options, the main unanswered question is long-term impact: it has not been shown convincingly that directly reducing a BML reliably delivers durable improvements in function or lowers the eventual risk of joint replacement beyond what well-delivered OA care can achieve.

How should I use my MRI report when planning next steps

An MRI report is most useful when it helps link a symptom pattern (for example, pain that is clearly worse with loading on stairs) to a treatable focus in the knee. In osteoarthritis, a prominent or persistent bone marrow lesion (BML) is typically interpreted as a sign of high subchondral stress and, at a population level, is associated with faster structural change and a higher likelihood of later knee replacement—but it is still only one part of the overall picture.

To address the “too many questions” feel flagged in review, this section shifts into a declarative checklist: what the report can practically contribute to planning, without getting stuck on a single phrase such as “bone marrow oedema”.

• Match the BML to the pain story: reports often specify the compartment (for example medial tibia or medial femoral condyle). When the lesion location and the main pain triggers line up, clinicians may treat it as a stronger pain candidate than cartilage signal alone.
• Use it to calibrate intensity of non-surgical care: the core pathway in major guidance (for example OARSI and a JAMA review) still centres on education, structured land-based exercise, weight management where relevant, and appropriate analgesia—without a separate “BML algorithm”. A sizeable BML may simply justify a more closely supervised strengthening and load-management plan.
• Plan monitoring around function, not just imaging: MRI studies show that BML volume can change over time, so follow-up often focuses on pain, walking tolerance, stairs, and flares; repeat imaging is usually reserved for cases where it would change management.

When pain remains high despite well-delivered conservative care over months, a marked BML can contribute to a lower threshold for discussing escalation—optimising injection strategies and, where appropriate, obtaining a surgical opinion—while balancing overall health and goals. A BML often means “stressed and potentially changeable bone”, not a standalone verdict that surgery is inevitable.

1. [1] Bone marrow lesions in osteoarthritis: From basic science to clinical implications. (2023). https://doi.org/10.1016/j.bonr.2023.101667 https://doi.org/10.1016/j.bonr.2023.101667
2. [2] Osteoarthritis bone marrow lesions. (2022). https://doi.org/10.1016/j.joca.2022.09.007 https://doi.org/10.1016/j.joca.2022.09.007
3. [3] The osteoarthritis bone score (OABS): A new histological scoring system for the characterisation of bone marrow lesions in osteoarthritis. (2022). https://doi.org/10.1016/j.joca.2022.01.008 https://doi.org/10.1016/j.joca.2022.01.008