What ICRS grading actually decides

Both Grade 3 and Grade 4 knee cartilage damage can fall within the injectable ChondroFiller pathway — but the depth of that damage changes what the scaffold is being asked to do, and which protocol is appropriate.

The ICRS (International Cartilage Repair Society) system grades cartilage from 0 (normal, undamaged surface) through to 4 (full-thickness loss, bone exposed). Grades 3 and 4 are the clinical target: Grade 3 lesions extend greater than 50% into the cartilage layer but leave the subchondral bone plate — the hard bone beneath — structurally intact. Grade 4 lesions breach that plate entirely.

That boundary matters because ChondroFiller® liquid is an acellular collagen scaffold: placed under ultrasound guidance in an outpatient appointment, it gels in the joint and provides a biological framework for the body's own cells to migrate into. When the subchondral plate is intact, the scaffold has an organised substrate beneath it. When the bone is exposed, the biological environment changes, and the protocol typically needs to adapt.

Defect grade is the starting point for candidacy assessment, not the whole picture. Defect size, the health of the surrounding cartilage borders, and MRI findings all shape the final decision — and those factors are explored in the sections that follow.

Grade 3 damage — why an intact subchondral plate helps the scaffold work

The intact subchondral plate in a Grade 3 lesion is not incidental — it is the biological foundation the scaffold builds from.

Grade 3 damage spans four subtypes (3A through 3D), from lesions stopping just above the calcified cartilage layer to surface blistering with deep structural loss beneath. What unites all four is a preserved subchondral bone plate. That preservation matters because ChondroFiller works through matrix-induced chondrogenesis — a process in which the collagen scaffold creates a structured environment that recruits the patient's own progenitor cells rather than delivering them directly. Those cells migrate in from the vascularised subchondral bone beneath and from the viable cartilage borders around the defect. With the plate intact, those pathways remain open, giving the scaffold the cellular supply it needs to support gradual tissue remodelling.

The injectable, in-situ gelling format suits this biology well. Delivered under ultrasound guidance into a fluid joint, the scaffold is placed without the aggressive debridement that arthroscopic surgical placement demands. Where surgery exposes a clean bone bed by removing existing tissue, this approach adds to what remains intact — preserving structure that still holds value.

One subtype warrants a closer look before confirming the injection route. Grade 3D — surface blistering overlying deep structural loss — sits at the boundary with Grade 4. MRI review of subchondral bone signal is a sensible precaution at this subtype: signal changes beneath a 3D lesion can indicate that the plate is under greater stress than the grade alone reflects, and that warrants careful clinical assessment before proceeding.

Grade 4 damage — conditional candidacy and when augmented protocols apply

Reaching Grade 4 does not automatically close the injectable pathway — but it does change the treatment logic, and patients deserve to understand why.

When a lesion breaches the subchondral bone plate, two things shift. First, the organised tissue substrate beneath the scaffold is gone: there is no longer an intact, vascularised bone plate to anchor the collagen gel or to supply progenitor cells upward in a structured way. Second, bone marrow constituents enter the defect directly. That is not inherently harmful — marrow contains growth factors and mesenchymal cells — but it is a different biological environment from Grade 3, one in which an acellular scaffold working alone carries greater uncertainty.

Standalone ChondroFiller monotherapy at Grade 4 is therefore less well supported than at Grade 3. The primary published evidence for injectable collagen scaffold use in Grade IV knee defects comes from Weninger et al. 2025, which demonstrated clinical benefit from pairing a collagen scaffold with regenerative cells — one study rather than a definitive body of evidence, but a meaningful signal that cellular augmentation matters at this depth.

This is the rationale for augmented combination protocols such as NanoACi, developed by Professor Paul Lee at London Cartilage Clinic. In this approach, ChondroFiller is combined with autologous cartilage micrografts (Mytocel MSK) and platelet-rich fibrin (Arthrozheal) in a single needle-delivered session. Each component serves a distinct role: the collagen scaffold provides structural matrix, the micrografts contribute autologous chondrocytes, and the platelet-rich fibrin supplies growth-factor support. The full three-part protocol has not yet been evaluated in randomised trials as a combined entity; component-level evidence underpins its rationale, and outcomes are being tracked prospectively through the NanoACi 100 programme.

For Grade 4 knees, the clinical message is that candidacy is conditional rather than excluded — and honest counselling means being clear about what augmentation adds and what remains to be confirmed.

The candidacy variables that cut across both grades

Grade alone does not settle the candidacy question — the size of the area affected, the health of the cartilage surrounding the defect, and what the MRI shows about the bone beneath all feed into the same clinical decision.

Defect size is an independent variable. Small, focal lesions and diffuse advanced wear across a compartment are positioned differently within the injection pathway. A focal defect with a contained footprint — typically up to 6 cm² in published ChondroFiller guidance — is the classical target for injectable scaffold placement. Diffuse wear without a clear focal boundary shifts the treatment rationale: the scaffold functions less as a targeted repair and more as a protective biological layer across a broadly compromised surface.

Border integrity may matter as much as grade itself. Healthy surrounding cartilage is not simply cosmetic — it contains the gel on delivery and supplies progenitor cells for the repair process. A Grade 3 defect with substantially degraded borders may therefore be a weaker injection candidate than a well-bordered, focal Grade 4 lesion with otherwise stable surrounding tissue. Grade does not automatically outrank border quality in clinical decision-making.

In Grade 4 knees, subchondral bone oedema on MRI adds a further layer to the assessment, beyond what the grade classification alone captures. Oedema signal beneath an exposed defect can indicate ongoing bone stress or early insufficiency changes that affect how the scaffold will behave — a factor that straightforward grading cannot reflect.

Patient-level variables — alignment, BMI, and activity demands — remain part of the picture throughout. Even when grade, defect size, and borders are all favourable, a knee that is significantly malaligned or mechanically overloaded presents a different treatment environment. That is why a detailed assessment, including MRI review and clinical examination, is needed before any candidacy determination: the imaging grade is the starting point, not the whole answer.

What the outcomes evidence shows — and where the gaps are

The improvement figures in published ChondroFiller knee literature are consistent and come from a meaningful patient volume — but they do not allow grade-by-grade comparison, and that distinction matters for how they should be read.

Across four knee clinical studies, IKDC scores improved by approximately 30 points over 12 months, well above the minimum clinically important difference of 16.7 points. The Jerosch et al. prospective PMCF study reported a mean improvement of 32.4 IKDC points, sustained and slightly increased at three-year follow-up, with patients reaching a functional score of 80. MOCART imaging scores ranged from 81.6 to 84.3 at one year — indicating greater than 80% defect fill and good integration with surrounding native cartilage — with scaffold maturation evidenced by MOCART rising from 65.3 at four weeks to 81.6 at one year. These figures derive from pooled Grade III/IV populations. Grade-stratified comparative outcome data for ChondroFiller injection does not currently exist in the published literature: the evidence supports the combined Grade III/IV target as a class, not as grade-separated efficacy claims.

The safety record across more than 20,000 implantations over ten-plus years provides a credible secondary anchor: approximately 0% complication rate and a 3–8% reoperation rate in knee studies. These figures, again, reflect pooled multi-grade populations rather than grade-specific cohorts.

Two gaps bear directly on Grade 4 candidacy counselling. Formal RCT-level evidence for the combined augmented protocol is not yet available. And the NanoACi 100 prospective programme — the mechanism for generating grade-specific outcomes data — has not yet reported results.

Getting a candidacy assessment in Lincolnshire

A scan report showing Grade 3 or Grade 4 damage is useful context — it is not, on its own, a candidacy decision.

A proper assessment brings together clinical history, a physical examination of the knee, weight-bearing X-rays to evaluate joint-space narrowing and limb alignment, and MRI reviewed for cartilage depth, border integrity, and subchondral bone signal. Where available, advanced cartilage mapping — including T2 relaxometry to assess tissue quality — can clarify defect characteristics that standard sequencing may not fully resolve. All of the variables covered in this article: grade, defect size, border quality, bone oedema, alignment, and activity level, feed into that structured review. No single number substitutes for it.

Lincolnshire Knee is part of the MSK Doctors group and accepts patients without a GP referral. Assessment clinics run at Sleaford NG34 and Grantham NG31, with consultant-led appointments and on-site imaging available. Book an assessment at lincolnshireknee.co.uk.