Why knee cartilage cannot heal itself

The most common question from patients who discover a focal cartilage lesion on MRI is whether it will simply heal over time. The honest answer is: not meaningfully, and here is why.

Articular cartilage — the smooth, pearl-grey hyaline tissue lining the knee joint surfaces — contains no blood vessels and no nerve fibres. Because circulating repair cells arrive via the bloodstream, tissue that has no blood supply cannot trigger the standard wound-healing cascade that closes a skin cut or knits a broken bone. A focal defect therefore stays open, and under the repetitive mechanical load of walking, the exposed edges gradually widen rather than fill in.

When a defect reaches full thickness — exposing the bone beneath, classified as ICRS grade III or IV — spontaneous closure does not occur. Focal chondral lesions of this kind are identified in 11–35% of routine knee arthroscopies, and untreated they carry a meaningful risk of progressing to diffuse osteoarthritis across the joint.

This biological reality shapes the entire management approach. The aim of every option discussed in this article — from physiotherapy and injections through to scaffold-based restoration — is to control symptoms, protect the remaining cartilage, and slow structural progression. None replicates de novo hyaline cartilage by themselves. That distinction matters: it explains why clinicians think in terms of a rational treatment ladder rather than a single cure.

At one end sits conservative symptom management; next comes biologic and injection support; beyond that, cartilage scaffold and restoration procedures; and at the far end, joint replacement for diffuse end-stage disease. Where a patient sits on that ladder depends on defect size, depth, age, activity demands, and symptom severity — factors explored in the sections that follow. Focal, symptomatic defects sit firmly in the middle of this continuum; diffuse osteoarthritis is a different conversation.

What conservative management can realistically achieve

Starting conservative management early is not a holding pattern — it is active treatment with a clear mechanical rationale.

The single most effective non-surgical intervention is targeted physiotherapy, specifically progressive strengthening of the quadriceps and the posterior chain (gluteal muscles, hamstrings). Strong muscles act as dynamic shock-absorbers, reducing the peak force that travels through the focal lesion with every step. This offload effect is measurable: losing approximately 5 kg of body weight produces a meaningful reduction in knee joint-reaction force, making weight management a concrete clinical goal rather than a lifestyle suggestion.

Unloader bracing works by the same principle — shifting load away from the affected compartment. It is a useful short-term adjunct during higher-demand activity, though it does not alter the underlying defect.

Activity modification rounds out the mechanical strategy. Swapping high-impact loading (running on hard surfaces, heavy squat work) for lower-impact alternatives (cycling, swimming) reduces the repetitive peak stress on the damaged area without sacrificing cardiovascular or muscular fitness.

NSAIDs and simple analgesics address pain and inflammation during flare-ups. They improve tolerance for physiotherapy but have no effect on cartilage biology — an important distinction when setting patient expectations.

If three months of consistent, structured conservative care has not produced meaningful symptom improvement, a clinical review to discuss next steps is warranted. That threshold is explored further in the summary section.

What injection therapies do — and what they cannot

Three injectable options occupy this space: hyaluronic acid (HA), platelet-rich plasma (PRP), and bone marrow aspirate concentrate (BMAC). Their role is symptom management and biologic support — not structural cartilage restoration — and the evidence for each reflects that distinction.

Hyaluronic acid (viscosupplementation) restores viscosity to the synovial fluid, reducing friction within the joint and providing short-term pain relief. OARSI guidelines offer conditional endorsement, while AAOS and ACR are more reserved. What HA does not do is rebuild lost cartilage matrix — the lubricating effect is mechanical, not regenerative.

PRP is an autologous plasma concentrate rich in growth factors including TGF-β, IGF, and PDGF, which may stimulate chondrocyte activity and matrix production. A 2025 retrospective study of 60 athletes found that three ultrasound-guided PRP sessions, spaced three weeks apart, produced greater improvement on both VAS pain scores and KOOS functional scores than an equivalent HA course. However, 3D-MRI data from the same year (21 knees) adds important granularity: cartilage thickness increased by more than 0.1 mm in 43% of knees in the anteromedial femoral region, but fewer than 20% of knees showed any measurable gain in the posteromedial femoral and medial tibial zones — precisely the regions most affected in medial knee OA. True structural change from PRP injection alone is modest and regionally variable.

BMAC delivers mesenchymal stem cells and growth factors harvested from the patient's own bone marrow in a same-day procedure. A 52-knee study with two-year follow-up found significant improvement in VAS, IKDC, and KOOS scores — comparable to allogeneic umbilical-cord MSC implantation — supporting BMAC as a clinically meaningful option within this tier.

A 2026 international Delphi consensus — 75 multidisciplinary experts across 25 or more countries — placed PRP, BMAC, and HA as equally appropriate for selected patients with persistent symptoms. No superiority was demonstrated among formulations or injection strategies, and standardised preparation protocols remain an evidence gap common to all three options.

ChondroFiller injection: an injectable scaffold for focal repair

ChondroFiller occupies a distinct category in the treatment spectrum: it is an acellular injectable collagen scaffold — a CE-marked Class III medical device — rather than a biologic injection such as PRP or HA, and rather than a surgical procedure requiring a theatre or anaesthetic.

The distinction matters mechanistically. Where PRP and BMAC introduce growth factors or cells to support the existing joint environment, ChondroFiller provides a physical three-dimensional scaffold placed directly at the defect site under ultrasound guidance. Once positioned, the scaffold attracts the patient's own progenitor cells and creates a structural environment for matrix-induced chondrogenesis — a process in which the patient's own biology performs the repair work within the scaffold's architecture, rather than the scaffold itself substituting for cartilage.

The procedure is delivered as an outpatient ultrasound-guided injection. It is indicated for focal chondral defects typically up to approximately 3 cm², with the scaffold formulation extendable to cover defects up to 6 cm²; it is not appropriate for diffuse osteoarthritis, where the pattern of joint involvement is fundamentally different from a discrete focal lesion.

Published outcome data include IKDC score improvements of approximately 30 points — on the 0–100 scale, a gain of this magnitude typically moves a patient from moderate to mild functional limitation — alongside MOCART cartilage repair scores in the range of 70–87 on MRI, indicating good-to-excellent structural fill at the repair site. The reported complaint rate is approximately 0.06% across published series, consistent with the device's established safety record.

Those figures apply to suitable candidates treated at appropriate defect stages. Individual outcomes depend on defect size, depth, subchondral bone integrity, and post-injection rehabilitation adherence — variables that require a consultant assessment to evaluate properly.

Microfracture and AMIC: what the evidence now shows

For much of the past two decades, microfracture was the default first procedure when focal knee cartilage defects required active intervention. The technique involves making small penetrating holes in the subchondral bone, allowing marrow-derived cells to flood the defect and form a repair tissue. It is single-stage and arthroscopic — but the tissue it produces is fibrocartilage rather than hyaline cartilage. That distinction matters: fibrocartilage lacks the organised collagen architecture and compressive stiffness of native hyaline cartilage, and under the cyclic load of daily knee function it tends to soften and fragment within two to three years.

The evidence has also identified a second problem. Creating the penetrating holes disrupts the subchondral bone plate, which provides the structural foundation for any subsequent repair procedure. Patients who have already undergone microfracture may find that more advanced options are technically more complex and outcomes less predictable as a result — a practically important point for anyone who had the procedure years ago and is now considering further treatment.

AMIC (autologous matrix-induced chondrogenesis) was developed to address microfracture's durability limitation while retaining its single-stage arthroscopic format. A collagen matrix scaffold placed over the marrow-stimulation site guides migrating progenitor cells into more organised repair tissue, with medium-term outcomes consistently superior to microfracture alone. A prospective multisite registry study with a mean follow-up of 6.8 years found no significant difference in Lysholm scores or pain VAS between patients aged 50–69 and those under 50 following AMIC — relevant given the widespread assumption that older patients are unsuitable for cartilage repair.

Both procedures require a theatre setting and a structured, protected weight-bearing recovery phase. Neither represents a non-surgical pathway.

How the right treatment gets chosen

Treatment selection for a focal knee cartilage defect comes down to a short list of variables, applied in order.

Defect grade and size come first. ICRS grade I–II lesions — partial-thickness, with the subchondral bone intact — respond better to conservative physiotherapy and injection pathways. Full-thickness grade III–IV defects, or lesions larger than roughly 2 cm², are unlikely to be adequately managed by injection alone and typically require a restorative scaffold or surgical procedure.

Three months of consistent conservative care without meaningful improvement is a defined action signal. German DGOU guidelines (2023) identify this as the point at which surgical referral should be considered — a window that reflects how most patients who will respond to conservative management do so within it.

Age adjusts the picture less than many patients expect. As noted in the AMIC discussion above, repair outcomes in patients aged 50–69 are comparable to those in younger cohorts; activity level, BMI, joint alignment, and comorbidities are individually more informative than chronological age.

Prior microfracture changes the technical options. Disruption of the subchondral bone plate narrows subsequent repair pathways and should always be disclosed at assessment.

Diffuse osteoarthritis sits outside this framework entirely. Generalised cartilage loss is a fundamentally different clinical problem from a discrete focal lesion; when damage is widespread, joint preservation or replacement surgery is the more appropriate conversation.

A structured assessment — using AI-assisted MRI cartilage analysis (onMRI™) to characterise defect size and depth, and objective biomechanical load mapping (MAI Motion®) to identify how force is distributed across the joint — provides the information needed to match the right pathway to the individual defect before any decision is committed to.

1. [1] Clinical Phase I/II Trial of SVF Therapy for Cartilage Regeneration: A Cellular Therapy with Novel 3D MRI Imaging. (2023). https://doi.org/10.3389/fcell.2023.1106279 https://doi.org/10.3389/fcell.2023.1106279
2. [2] Conservative Management of Osteochondral Lesions — DGOU Working Group Recommendations (2023). (2023). https://doi.org/10.1177/19476035231161806 https://doi.org/10.1177/19476035231161806
3. [3] 3D-MRI Analysis of Cartilage Thickness Changes After PRP Injection in Medial Knee Osteoarthritis: A Preliminary Report. (2025). https://doi.org/10.1371/journal.pone.0321067 https://doi.org/10.1371/journal.pone.0321067
4. [4] Bridging Conservative Management With Injectable Orthobiologic Therapies: International Consensus Meetings on Cartilage Repair (2026). (2026). https://doi.org/10.1177/19386400261456925 https://doi.org/10.1177/19386400261456925
5. [5] Treatments for Knee Cartilage Injuries in Athletes Using PRP and Hyaluronic Acid: A Retrospective Analysis. (2025). https://doi.org/10.37990/medr.1604362 https://doi.org/10.37990/medr.1604362
6. [6] Comparison of Bone Marrow Aspirate Concentrate and Allogenic hUCB-MSC Implantation on Chondral Defect of Knee: 2-Year Follow-Up. (2020). https://doi.org/10.1177/0963689720943581 https://doi.org/10.1177/0963689720943581
7. [7] Immunohistochemical Analysis of Knee Chondral Defect Repair After Autologous Particulated Cartilage and PRP Treatment in Sheep. (2023). https://doi.org/10.3390/ijms242015157 https://doi.org/10.3390/ijms242015157