When joint preservation is still on the table

The question a knee surgeon is really asking at a first consultation is not 'which operation?' but 'is there still a native joint worth preserving?' For a substantial number of patients, the answer is yes — and that single clinical judgement separates candidates for cartilage restoration from those heading towards joint replacement.

Knee care follows four broad stages: symptom management, biologic and injection support, cartilage restoration, and joint replacement. ACI (Autologous Chondrocyte Implantation) sits at stage three. It is designed for symptomatic, focal, full-thickness cartilage defects — areas where the damage is discrete and the surrounding joint remains largely intact. NICE Technology Appraisal TA477 (2017) defines the formal clinical envelope: a defect larger than 2 cm², minimal osteoarthritic change elsewhere, no prior cartilage repair surgery, and typically a patient under 50 who remains active.

Total knee replacement (TKR) belongs at stage four. It is appropriate when cartilage loss is diffuse across multiple compartments, joint deformity is fixed, or pain is unrelenting at rest — a picture that places the patient well outside ACI territory.

The central distinction is structural, not just philosophical: ACI uses the patient's own cells to regenerate native cartilage tissue. TKR removes those surfaces permanently and replaces them with metal and plastic implants. Once that step is taken, it cannot be undone. For patients who still have a preservable joint, delaying it — or avoiding it altogether — is a legitimate clinical goal with robust supporting evidence.

What ACI does that a replacement cannot

At a biological level, these two procedures are solving different problems. ACI works by repairing what is already there. In the first stage, a small sample of healthy cartilage is harvested from the knee and sent to a laboratory where the chondrocytes — the cells that produce and maintain cartilage — are expanded in culture over several weeks. In the second stage, those cells are reimplanted into the damaged area, held in place by a Type I/III collagen membrane (the MACI variant). Over time, the cells integrate with the surrounding tissue and generate hyaline-like cartilage — structurally closer to the original surface than the fibrocartilage produced by marrow-stimulation techniques such as microfracture.

TKR takes a fundamentally different approach. The worn surfaces are removed entirely and replaced with metal and polyethylene components. Recovery is faster — meaningful independence typically returns within three to six months — but the activity ceiling is permanently lower. High-impact sport, sustained running, and heavy pivoting are routinely discouraged to protect the implant, and revision surgery if components wear is considerably more complex than a first replacement.

ACI, by contrast, preserves native bone stock and joint anatomy. That matters practically: if a patient does eventually need a knee replacement, the surgeon is starting with an intact joint rather than managing the consequences of an earlier implant.

The honest counterweight is time. ACI cartilage reaches full maturation at around 18 months. Patients return to low-impact activity progressively over that window, with high-impact sport typically restricted for nine to twelve months. For someone weighing their options, that timeline is a real practical cost alongside the clinical benefits.

Who qualifies for ACI and who does not

Running through a rough eligibility checklist before a consultation can help patients arrive with the right questions.

When ACI is appropriate

NICE Technology Appraisal TA477 (2017) sets the formal clinical threshold, and it is relatively specific:

• A focal, full-thickness cartilage defect larger than 2 cm² — the kind of discrete lesion that leaves a clean boundary between damaged and healthy tissue
• Minimal osteoarthritic change in the surrounding joint — the wider knee must still be worth preserving
• No prior cartilage repair surgery at the same defect site — a point explored further below
• An active patient, typically under 50 — age is not an absolute cut-off but reflects the activity-demand profile that makes joint preservation most worthwhile

Defect size also has an upper limit in practical terms. The strongest ACI and MACI evidence covers defects up to roughly 10 cm². Beyond that — particularly where there is significant bone loss beneath the cartilage — fresh osteochondral allografts tend to be the more appropriate option, as they replace both the cartilage and the underlying bone stock simultaneously.

When ACI is not appropriate

Several factors shift the clinical calculus against ACI:

• Diffuse or multi-compartment arthritis — once cartilage loss is widespread rather than focal, there is no clean defect to repair, and the cost-effectiveness modelling underpinning TA477 no longer applies
• BMI above 30 — excess load independently raises the risk of graft failure and is a first-order consideration in case selection, not a minor footnote
• Prior failed cartilage treatment at the same site — data from the German Cartilage Registry (5,961 patients) show that patients with a history of failed cartilage surgery at the defect site have significantly lower KOOS scores at every follow-up time point compared to those with no prior cartilage surgery. The clinical implication is that ACI works best when it is the first cartilage repair intervention, not a rescue after microfracture has already disrupted the subchondral bone plate

Alignment and complexity

Varus or valgus malalignment does not automatically exclude a patient, but it does add surgical complexity. Where abnormal limb alignment is concentrating load across the damaged compartment, a concurrent osteotomy — high tibial or distal femoral — is often needed to protect the repair and give the graft a realistic chance of surviving. This combination is established practice rather than an experimental add-on.

Many patients sit on the borderline between these categories. A specialist assessment, including MRI analysis of defect morphology and surrounding cartilage integrity, is needed to determine where any individual patient actually falls.

How long ACI results actually last

The durability question is the one that matters most to patients weighing ACI against eventual replacement. The evidence is specific enough to give honest answers, but those answers come with meaningful caveats about study design.

The foundational dataset is Minas et al. (2009), a prospective cohort of 153 patients (mean age 38.3 years, mean defect 4.9 cm²) followed for up to 11 years after ACI for early-stage knee osteoarthritis. At five years, 92% were functioning well and had avoided joint replacement; only 8% progressed to arthroplasty over the entire follow-up period, with WOMAC subscales improving by 50–75% from baseline. For a population that was already showing early OA at the time of treatment, those figures set the benchmark for what well-selected ACI can achieve.

Longer-term data for the MACI variant comes from a 2024 systematic review of 168 patients (mean age 37) with minimum 10-year follow-up — reaching 17 years in some participants. Patient-reported outcomes remained significantly improved throughout, and the TKA conversion rate was 7.4%. In practical terms, roughly nine in ten appropriately selected patients had not needed a knee replacement by the time most studies closed.

For larger defects — the 4–12 cm² range that stretches the upper boundary of ACI eligibility — a 2025 Phase III trial of hydrogel-based ACI reported KOOS improvement from 39.8 pre-operatively to 84.7 at five years (a mean gain of 44.1 points, p<0.0001), with 92.8% of patients classified as responders and an overall treatment failure rate of just 1%.

The longest available horizon comes from London Cartilage Clinic data: an 82% success rate with 74% of patients retaining their native knee at 20 years.

Two important qualifications apply to all of these figures. First, most long-term ACI datasets are case series drawn from specialist centres with careful patient selection — inherent selection bias means outcomes in broader practice may differ. No head-to-head randomised trial directly compares ACI with TKR in the overlapping patient group, typically aged 45–55 with early multi-focal OA. Second, and equally important for anyone considering the procedure: the 74% native-knee retention figure at 20 years means that roughly one in four patients did eventually require a replacement. ACI, at its best, substantially delays joint replacement for the majority — it does not guarantee avoiding it.

The harder decision for patients aged 45–55

Patients in their late 40s and early 50s often fall into a gap that neither the ACI eligibility criteria nor the standard indications for replacement quite capture. This is the grey zone — where the evidence is thinnest and the clinical consultation is most complex.

The activity-level contrast illustrates why age alone is an unreliable guide. A 52-year-old recreational runner with one isolated focal defect and surrounding cartilage that is otherwise intact presents a very different clinical picture from a 48-year-old with diffuse medial compartment changes and mechanical symptoms at rest. The first patient may well be an ACI candidate if lesion morphology supports it; the second is unlikely to benefit meaningfully from cartilage restoration alone.

Unicompartmental knee replacement (UKR) occupies a legitimate middle ground for patients in this bracket whose arthritis is confined to one compartment but is already too widespread for cartilage repair. UKR preserves considerably more bone stock than total replacement and imposes fewer functional restrictions — though it remains a replacement procedure, not a joint-preservation one, and should be framed as such.

Timing is a first-order consideration. As the eligibility section above sets out, prior failed cartilage treatment at the defect site materially worsens ACI prognosis. A patient who delays intervention until symptoms become severe risks narrowing — or closing — the joint-preservation window. Earlier treatment, while the defect remains focal and the surrounding cartilage is intact, consistently produces better outcomes than later salvage.

Specialist MRI analysis — assessing defect morphology, surrounding cartilage quality, and compartmental load distribution — alongside objective biomechanical assessment helps clarify which pathway is clinically realistic for an individual patient, rather than relying on age as a proxy for either suitability or urgency.

Getting an accurate assessment before committing to either path

Choosing between ACI and replacement rests on the quality of the pre-surgical assessment — and specifically on how precisely the lesion has been characterised before any decision is reached.

Plain X-ray routinely underestimates cartilage damage. A defect that appears modest on a weight-bearing radiograph may, on MRI with dedicated cartilage sequences such as T2 mapping and cartilage segmentation, prove to involve the subchondral bone layer or extend into surrounding OA-affected tissue — either finding shifts the clinical picture substantially. AI-assisted MRI analysis, such as onMRI™, can quantify defect boundaries and tissue composition beyond a standard read, helping to confirm whether a lesion falls within the ACI eligibility envelope or outside it.

Biomechanical factors carry equal weight. Varus or valgus malalignment concentrates load through the damaged compartment; without correction, cartilage repair in that compartment is likely to fail regardless of technique. Objective gait analysis — measuring actual load distribution through the knee during walking, as MAI Motion® provides — can identify mechanical contributors that clinical examination alone may miss and clarify whether alignment surgery should accompany any repair.

Consultant-led assessment integrates these imaging and biomechanical findings with the patient's activity demands and long-term goals before any surgical pathway is recommended. Lincolnshire Knee, part of the MSK Doctors group, accepts patients without a GP referral. To book an assessment, visit lincolnshireknee.co.uk.

1. [1] Treatment of Large Cartilage Defects in the Knee by Hydrogel-Based ACI: 5-Year Follow-Up Phase III Trial. (2025). https://doi.org/10.1177/19476035251334737 https://doi.org/10.1177/19476035251334737
2. [2] Minced Cartilage Implantation vs ACI for Knee Cartilage Lesions: Matched-Pair Analysis. (2025). https://doi.org/10.1002/ksa.70210 https://doi.org/10.1002/ksa.70210
3. [3] 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
4. [4] Effect of Previous Knee Surgery on ACI Outcomes: German Cartilage Registry Study. (2022). https://doi.org/10.1177/03635465211070536 https://doi.org/10.1177/03635465211070536
5. [5] Cost-Effectiveness of New ACI Technique vs Standard ACI and Microfracture in Knee Cartilage Defects. (2023). https://doi.org/10.1080/13696998.2023.2194805 https://doi.org/10.1080/13696998.2023.2194805