Why Grade 3 and 4 defects sit at a treatment crossroads

A cartilage grading report that reads Grade 3 or Grade 4 tends to prompt an immediate question: does this mean a knee replacement is inevitable? The short answer is no — but it does mean the damage is severe enough that the full range of treatment options, including both biological repair and partial resurfacing, now belongs on the table.

The ICRS grading scale measures how deeply a cartilage lesion penetrates the joint surface. Grade 3 means the defect extends more than half the thickness of the cartilage, down toward or through the calcified layer that sits just above the bone. Grade 4 means the damage has gone all the way through, exposing the underlying subchondral bone. That distinction matters clinically: once bone is involved, the repair strategy must account for the bone layer as well as the cartilage above it.

What these grades do not mean is that the whole knee joint is failing. Diffuse osteoarthritis — where wear is spread across multiple surfaces — is a fundamentally different condition. Grade 3–4 defects, as discussed here, are focal: concentrated damage in one area, often the result of a prior injury or a localised loading problem, while the rest of the knee may remain largely intact. That structural integrity in unaffected compartments is precisely what keeps joint-preserving options realistic.

Symptoms alone are an unreliable guide. Some patients with Grade 4 findings report surprisingly manageable pain; others with Grade 3 damage are significantly limited. Clinical assessment combined with imaging — including MRI that can characterise cartilage depth and bone oedema — is needed to establish the full picture before any pathway is chosen.

The treatment decision at this threshold is therefore not about whether to replace the entire knee. It is about whether the damaged compartment is better served by biological restoration — rebuilding the cartilage using the body's own cells or a scaffold — or by partial resurfacing, which removes the damaged surface and replaces it with a precisely fitted implant while leaving the rest of the joint untouched. Both are legitimate, evidence-supported routes; which fits better depends on defect size, bone involvement, alignment, and what the patient needs the knee to do.

Defect size: the first branch point in choosing a repair approach

Defect area is the first practical filter once Grade 3–4 damage has been confirmed. Moving through the size range clarifies which biological techniques remain viable — and where the conversation shifts toward partial resurfacing.

For the smallest lesions, under roughly 2 cm², marrow-stimulation by microfracture has historically been the entry-level procedure. At Grade 3–4, however, it is no longer a preferred option: fibrocartilage — the inferior repair tissue it produces — tends to deteriorate within two to three years, and the drilling required can damage the subchondral bone plate in ways that narrow future repair choices. That limitation is relevant if a patient eventually needs a more advanced procedure.

From approximately 2 cm² to 4 cm², several techniques carry meaningful evidence: osteochondral autograft transfer (OATS/mosaicplasty), AMIC, and cell-based approaches including ACI and MACI. The choice within this band depends on how much bone is involved, where in the knee the defect sits, and whether prior surgery has been performed.

At 3 cm² and above, MACI becomes the better-supported option. The SUMMIT trial demonstrated superior KOOS pain and function scores for MACI over microfracture at both two and five years in lesions of this size. Longer-term cohort data reinforce that picture: at 10-year follow-up, only 7.4% of MACI recipients in a 168-patient series had progressed to total knee replacement — a meaningful signal that biological repair at this threshold can defer arthroplasty by a considerable margin.

For large or bipolar defects — where opposing cartilage surfaces on both sides of the joint are affected — fresh osteochondral allograft (OCA) is the biological ceiling. Cohort data from 89 knees with a mean defect of 16.7 cm² show survivorship of 66.6% at 10 years and 58.9% at 15 years, substantially lower than outcomes for isolated focal defects. When defect size, bipolar involvement, or compartmental spread pushes OCA survivorship to these levels, UKA enters the discussion not as a fallback but as a structured alternative with its own evidence base — covered in the section that follows.

The biological repair toolkit for Grade 3–4 focal defects

Three groupings cover the biological toolkit at this grade — single-stage options, two-stage cell-based repair, and allograft — each suited to a different profile of defect size, bone involvement, and patient circumstance.

Single-stage options suit patients seeking one operation. OATS/mosaicplasty uses cylindrical plugs of the patient's own bone and cartilage, press-fitted into defects of roughly 1–4 cm² in a single procedure; donor-site discomfort at the harvest zone is the principal trade-off. AMIC (autologous matrix-induced chondrogenesis) adds a Type I/III collagen membrane scaffold to microfracture, supporting repair-tissue formation with less resource demand than cell-based procedures — registry data show equivalent outcomes at mean 6.8-year follow-up in patients aged 50–69 compared with those under 50, which makes age alone an insufficient reason to rule it out. ChondroFiller injection is also a single-stage option, delivered as an ultrasound-guided outpatient procedure rather than a theatre operation: an injectable collagen scaffold is placed into the defect, recruiting the patient's own progenitor cells through matrix-induced chondrogenesis. As with OATS and AMIC, candidate suitability — including defect size and depth — is confirmed at clinical assessment and imaging review.

Two-stage cell-based repair covers ACI and MACI: a first procedure harvests chondrocytes, which are cultured before re-implantation — in MACI, seeded onto a collagen membrane for more consistent fixation. The two-procedure timeline is more demanding, but across six RCTs the treatment failure rate for third-generation ACI ran at 0–1.8%, compared with 2.5–8.3% for microfracture in lesions ranging from 1.8 to 5.0 cm². MACI carries the strongest evidence once defect size reaches approximately 3 cm² and above.

Fresh osteochondral allograft (OCA) uses donor tissue to address bone and cartilage simultaneously in one stage, without the volume constraints of autograft. It is reserved for larger post-traumatic defects or cases where prior surgery has depleted suitable autograft. Availability of fresh-matched tissue and timing logistics are practical considerations.

Emerging one-step approaches — including HA-BMAC scaffold constructs, which showed durable results in 26 patients at mean 14-year follow-up, and next-generation single-stage ACI variants (STACI) — are showing early promise but remain investigational pending larger trials.

When partial knee replacement becomes the stronger case

Partial knee replacement — unicompartmental knee arthroplasty (UKA) — resurfaces only the damaged compartment, medial or lateral, while leaving the rest of the knee's native tissue untouched. That distinction matters: bone stock and ligament function are preserved, keeping future revision options open. The procedure suits a specific patient profile: single-compartment disease confirmed on imaging and examination, an intact anterior cruciate ligament, and an alignment problem that is correctible rather than fixed.

The long-term case for UKA is robust. Mobile-bearing medial UKA achieves 89% survivorship at 10 years and 84.1% at 15 years in series data from 219 knees. Perioperative safety compares favourably with total knee replacement: in a 274,411-patient analysis, TKA carried nearly twenty times the odds of blood transfusion and significantly higher rates of major complications, minor complications, and readmission at 30 days. Gait studies add a functional dimension — patients who have undergone both a UKA on one side and a TKA on the other accept 22% more load through the UKA limb at the weight-acceptance-to-midstance transition, reflecting a more natural loading pattern.

The figure that demands honest acknowledgement is the 39.9% one-year conversion rate to total knee replacement seen in US registry data. This does not reflect a problem with the implant itself; it reflects the sensitivity of patient selection. When single-compartment involvement is assumed rather than confirmed — or when alignment is inadequately assessed — early failure follows. Robotic-assisted systems have improved implant positioning accuracy and their use has grown substantially, but they have not yet resolved the selection challenge upstream.

For younger patients weighing UKA against high tibial osteotomy, the trade-off is explicit: UKA delivers better pain relief and a lower complication rate, whereas HTO preserves a greater range of motion and carries a lower long-term revision rate — considerations that carry more weight the further a patient is from typical arthroplasty age.

UKA, then, is the more appropriate recommendation when focal grade 3–4 damage is genuinely compartment-confined, alignment is correctible, ACL integrity is confirmed, and the patient's symptom burden and functional goals warrant a resurfacing approach rather than continued biological repair attempts.

How alignment, age, and activity shift the recommendation

Alignment comes first — and not as a surgical add-on. The load path through the knee determines whether repaired cartilage will survive long-term or be worn down by persistent abnormal force distribution. In one MACI series, half of all candidates required a concurrent osteotomy to bring the mechanical axis into the target range of approximately −2.5° valgus to 4.5° varus; patients who achieved that range were 2.5 times more likely to reach an acceptable symptom state than those who did not (70% versus 27%). Where malalignment is correctible, high tibial osteotomy combined with cartilage repair is a coherent joint-preserving strategy in younger patients. Where it is not — or where the deformity is too severe to correct adequately — UKA, which can accommodate moderate varus within defined limits, becomes the more durable option.

Age shapes the conversation without resolving it. Cartilage repair cohorts in the published literature are predominantly under 40; UKA cohorts skew older. The 40–55 range with large Grade 4 defects is the most clinically contested territory, where implant longevity, biological repair durability, and activity level all intersect without a single dominant answer from trial evidence. No reliable age cutoff separates the two pathways.

Activity demands add a third dimension — and here a concrete distinction matters. A patient returning to physically demanding manual work, with sustained load-bearing through a working day, weighs UKA's more predictable and faster return to function differently from someone in a largely sedentary role who can accommodate a longer MACI rehabilitation timeline and places greater priority on preserving a native joint for the decades ahead. Recovery trajectory is not a secondary consideration; for some patients it is the deciding one.

Getting the assessment right before choosing a pathway

Choosing between biological repair and partial resurfacing at Grade 3–4 cannot rest on a standard MRI report alone. Defect geometry, bone involvement, limb alignment, and functional status each carry independent weight in the decision — and they need to be assessed together rather than in isolation.

Advanced cartilage MRI analysis, including T2 mapping and cartilage segmentation, provides objective detail on defect depth and subchondral bone condition that routine reporting may understate. Objective biomechanical assessment adds loading and alignment data that imaging alone cannot capture — two inputs that, taken together, make the difference between a confident recommendation and an uncertain one.

A thorough specialist consultation maps findings across the full decision range: biologic repair and scaffold options, osteotomy where alignment requires correction, partial resurfacing, and total replacement. The goal is a structured choice rather than a default one reached by exclusion.

Lincolnshire Knee is part of the MSK Doctors group and accepts patients without referral. Assessment and imaging are available at Sleaford NG34 and Grantham NG31. Book an assessment at lincolnshireknee.co.uk.

1. [1] Midterm Survivorship and Clinical Outcomes in Fresh OCA Transplantation for Large Bipolar Lesions of the Knee. (2025). https://doi.org/10.1177/03635465241313139 https://doi.org/10.1177/03635465241313139
2. [2] Differences in Clinical and Functional Outcomes Between OCA Transplantation and ACI for Focal Articular Cartilage Defects. (2022). https://doi.org/10.1177/23259671211058425 https://doi.org/10.1177/23259671211058425
3. [3] The unicompartmental knee is the preferred side in individuals with both a UKA and TKA. (2019). https://doi.org/10.1007/s00167-019-05814-7 https://doi.org/10.1007/s00167-019-05814-7
4. [4] Autologous minced cartilage repair for chondral and osteochondral lesions — minimum 5-year follow-up. (2023). https://doi.org/10.1007/s00167-023-07546-1 https://doi.org/10.1007/s00167-023-07546-1
5. [5] ACI, MACI, OAT and OCA improve knee function and pain — systematic review and meta-analysis. (2024). https://doi.org/10.1002/ksa.12525 https://doi.org/10.1002/ksa.12525
6. [6] Epidemiological Trends and Outcomes of Unicompartmental Knee Arthroplasty Among 104 Million US Patients. (2024). https://doi.org/10.1016/j.arth.2024.06.048 https://doi.org/10.1016/j.arth.2024.06.048
7. [7] 15-Year Follow Up of Mobile Bearing Medial Unicompartmental Knee Arthroplasty. (2023). https://doi.org/10.1016/j.arth.2023.01.024 https://doi.org/10.1016/j.arth.2023.01.024
8. [8] Minimum 10-Year Outcomes of Matrix-Induced Autologous Chondrocyte Implantation in the Knee. (2024). https://doi.org/10.1177/03635465231205309 https://doi.org/10.1177/03635465231205309
9. [9] Third-Generation ACI versus Microfracture for Focal Chondral Defects of the Knee — Systematic Review of RCTs. (2022). https://doi.org/10.1016/j.arthro.2022.02.011 https://doi.org/10.1016/j.arthro.2022.02.011
10. [10] Patellofemoral Joint Preservation: Cartilage Restoration, Unloading, and Avoiding Complications. (2026). https://doi.org/10.1097/JSA.0000000000000459 https://doi.org/10.1097/JSA.0000000000000459
11. [11] Neutral to slightly undercorrected mechanical leg alignment provides superior long-term results in M-ACI. (2024). https://doi.org/10.1002/ksa.12226 https://doi.org/10.1002/ksa.12226
12. [12] Unicompartmental Knee Arthroplasty vs Total Knee Arthroplasty: 30-day Outcomes. (2022). https://doi.org/10.1016/j.artd.2022.06.017 https://doi.org/10.1016/j.artd.2022.06.017