Are you a candidate for OATS or mosaicplasty?

The practical question most patients ask at this stage is straightforward: does this procedure fit my situation? Both OATS and mosaicplasty take one or several small cylinders of bone and cartilage — the graft plugs — from a low-load zone of your own knee and seat them into the damaged area in a single operation. Whether that is the right approach depends on four intersecting factors.

The defect itself

Both techniques are designed for focal, contained, grade III–IV chondral or osteochondral lesions — in practice, defects roughly up to 2–4 cm² on the femoral condyle. The medial femoral condyle is the most frequently treated site. Defects larger than 20 mm in diameter, or those involving both sides of the joint ('kissing' or bipolar lesions), fall outside the typical operating envelope for arthroscopic OATS.

Patient biology

Candidates are usually younger active adults, typically under 50, with no significant background osteoarthritis (Kellgren-Lawrence grade 2 or below), no history of inflammatory arthritis, and no prior joint infection. A BMI above 40 is a relative — not absolute — contraindication, as is age above 50; individual surgeon judgement applies in borderline cases. Before surgery is considered appropriate, patients should have tried and not responded to a structured conservative programme — usually physiotherapy-led rehabilitation, activity modification, and where indicated, injection support over several months — since the operation is a restorative rather than a first-response measure.

Limb environment

Uncorrected varus or valgus malalignment is listed as a contraindication to OATS in isolation; if loading asymmetry is present, alignment correction may need to accompany the cartilage procedure for graft survival to be realistic.

How defect size shapes the decision

Defect area is the variable that most directly drives procedure selection once a patient clears the baseline candidacy hurdles. Within the working 1–4 cm² range, the picture changes meaningfully around the 2 cm² mark.

Below that threshold, osteochondral autograft holds a clear advantage over marrow-stimulation approaches. In a study of osteochondral damage no larger than 1.5 cm² occurring alongside ACL reconstruction, OATS produced significantly better Lysholm and IKDC scores than microfracture at every follow-up point out to 12 months (p<0.001). Mosaicplasty also performs durably in this size band across longer follow-up periods.

Above 2 cm², evidence from a 12-year comparative study shifts the calculus considerably. Patients treated with mosaicplasty for lesions exceeding 2 cm² recorded significantly lower IKDC objective scores and Tegner activity scores than those who received matrix-induced autologous chondrocyte implantation — MACT, a two-stage cell-based technique in which cartilage cells are cultured in a laboratory and reimplanted on a collagen membrane (p=0.031 and p=0.014, respectively). Mosaicplasty carried a 25% overall failure rate in that comparison; MACT carried 0%. Defect size did not affect outcomes in the MACT group, suggesting cell-based repair is more tolerant of larger lesion area. That asymmetry makes 2 cm² a practical upper threshold for mosaicplasty when MACI or ACI is available.

For defects at or above approximately 3 cm², clinical consensus generally favours MACI, ACI, or — where the lesion is very large or involves the underlying bone — fresh osteochondral allograft (OCA), which uses donor graft tissue rather than the patient's own. The constraint is partly biological: the non-weight-bearing harvest sites on the patient's condyle can supply only so much graft material in a single stage.

The 2 cm² figure is best read as a practical guide rather than an immovable biological law. Lesion geometry, bone involvement, and whether cell-based techniques are locally accessible all influence where the threshold is applied in individual cases.

Why multi-plug mosaicplasty has a built-in ceiling

The gap problem has a structural implication as defect size grows: the larger the mosaic, the greater the proportion of the repair that is fibrocartilage rather than transplanted hyaline tissue — the mechanical reason outcomes diverge above 2 cm², as covered in the previous section.

A finite harvest zone

Donor plugs are taken from the non-weight-bearing peripheral rim of the medial femoral condyle and the intercondylar notch — zones chosen because removing a cylinder of bone and cartilage there does not compromise the knee's load-bearing function. In published series, a typical procedure uses a mean of approximately 2.2 plugs to cover a mean defect of around 2.3 cm². That harvest zone is anatomically fixed. Once those sites are used, there is no equivalent fallback within the same knee in a single stage; graft availability is the ceiling, not surgical technique.

Donor-site considerations

Harvest-site discomfort and stiffness are recognised after surgery, and how troublesome they prove varies between individuals. Long-term data on donor-site outcomes are still being collected — this is an acknowledged gap in the evidence rather than a settled question. In practice, a pre-operative assessment includes a review of harvest-zone anatomy, which allows the surgeon to advise whether adequate graft supply exists and what the likely trade-off looks like for that specific knee before any decision is made.

How long OATS repairs last

Survival data from a long-term cohort study (n=203, follow-up to 18 years) give the clearest answer available: OAT/mosaicplasty repairs remained functional in more than 80% of patients through the first seven post-operative years, and in more than 60% at 15 years. Mean time to failure was 8.4 years for OAT — more than double the 4.0-year mean recorded for microfracture in the same dataset. Microfracture's survival curve fell below 80% within 12 months and below 60% within three years, underscoring why bone-marrow stimulation is increasingly set aside in favour of autograft for patients who will place sustained demands on the knee over years rather than months.

Functional recovery follows a broadly encouraging trajectory. Across 2-to-10-year follow-up analysis, 75–78% of patients achieved the minimum clinically important difference (MCID) on the IKDC score — the threshold at which a patient notices a meaningful improvement in daily life — and those gains were largely maintained at the 10-year mark.

For relative comparison with cell-based methods, the picture is more nuanced. No randomised controlled trial has established that OATS is superior to MACI or ACI on patient-reported outcomes; a network meta-analysis of 19 RCTs found no statistically significant difference between the techniques. The evidence supports equivalence rather than hierarchy, with defect size and clinical context — not durability alone — determining which approach is used.

For a patient under 45 with a contained femoral condyle defect under 2 cm², the durability outlook from published series is genuinely encouraging. No surgical cartilage repair is permanent, but an expected functional lifespan measured in years to a decade-plus is a reasonable frame for shared decision-making.

When alignment or ligament problems change the plan

Cartilage repair does not happen in isolation from the rest of the knee. Where the joint carries uneven load — typically because of varus (bow-legged) or valgus (knock-kneed) malalignment — the graft site bears disproportionate stress, and evidence suggests that mosaicplasty alone in this mechanical environment produces inferior results.

Adding a high tibial osteotomy (HTO) to mosaicplasty, which shifts load away from the damaged compartment, produces measurably better MRI-based graft integration compared with mosaicplasty performed without alignment correction — including higher MOCART scores, less bone marrow oedema, improved plug union, and reduced plug necrosis. The osteotomy does not repair cartilage; it changes the mechanical environment so that the graft has a realistic chance of surviving. A varus knee assessed for mosaicplasty that does not have concurrent alignment correction may therefore be a suboptimal candidate for the cartilage procedure in isolation.

Concurrent ACL insufficiency raises a similar issue. For patients with a small focal defect alongside an ACL tear, combined OATS and ACL reconstruction is supported by clinical evidence — in one study of defects measuring 1.5 cm² or less, the combined approach was safe and effective at one-year follow-up.

The practical implication is that a thorough biomechanical assessment — including review of limb alignment and ligament stability — is part of candidacy evaluation, not a secondary consideration. Where instability or malalignment is identified, the treatment plan may expand to include adjunct procedures before or alongside cartilage repair.

Getting a proper assessment at Lincolnshire Knee

Confirming candidacy for OATS or mosaicplasty brings together the four axes covered across this article: defect characteristics (size, grade, location, bone involvement), patient biology (age, BMI, OA burden), prior treatment history, and limb environment (alignment, ligament stability). No single scan or measure settles the question on its own.

At Lincolnshire Knee — part of the MSK Doctors group — that evaluation is consultant-led and begins with detailed MRI characterisation. The clinic's onMRI™ platform applies AI-assisted cartilage segmentation and T2 mapping to quantify defect depth and assess surrounding cartilage quality, helping to define lesion grade non-invasively before any procedural decision is reached. Where alignment or loading patterns are a consideration, MAI Motion® objective gait analysis can provide quantified biomechanical data to clarify whether an adjunct procedure such as HTO should be part of the plan.

The outcome of that assessment may point toward OATS, mosaicplasty, an alternative cartilage repair pathway, or a combined approach — decisions made on the basis of the full clinical picture rather than any single factor.

Patients can be seen without a GP referral at Sleaford NG34 or Grantham NG31. Book a consultant assessment at lincolnshireknee.co.uk.

1. [1] Randomized Study of Long-term (15–17 Years) Outcome After Microfracture Versus Mosaicplasty in Knee Cartilage Defects (2018). (2018). https://doi.org/10.1177/0363546517745281 https://doi.org/10.1177/0363546517745281
2. [2] Mosaicplasty with HTO for Knee Subchondral Insufficiency Fracture (2022). (2022). https://doi.org/10.1016/j.arthro.2022.07.020 https://doi.org/10.1016/j.arthro.2022.07.020
3. [3] Mosaicplasty versus MACT for Knee Cartilage Defects: Long-Term Clinical and Imaging Evaluation (2020). (2020). https://doi.org/10.3390/app10134615 https://doi.org/10.3390/app10134615
4. [4] Long-Term Survival after Microfracture and Mosaicplasty for Knee Articular Cartilage Repair (2018). (2018). https://doi.org/10.1177/1947603518783482 https://doi.org/10.1177/1947603518783482
5. [5] Effect of Autogenous Osteochondral Mosaicplasty on Balance Control in Knee Cartilage Defect Patients (2023). (2023). https://doi.org/10.1186/s13018-023-03821-6 https://doi.org/10.1186/s13018-023-03821-6