Why OATS delivers real cartilage — not scar tissue

When a surgeon recommends OATS or mosaicplasty for a focal knee cartilage defect, the central reason comes down to tissue quality. Healthy joint surfaces are lined with hyaline cartilage — a dense, resilient material that absorbs load and allows smooth movement. OATS (osteochondral autograft transfer system) works by harvesting a cylindrical plug of that same hyaline cartilage, together with its underlying subchondral bone, from a low-demand area of the patient's own femoral condyle and press-fitting it precisely into the damaged site. The result is a repair built from the original biological material, not a substitute.

The contrast with marrow-stimulation techniques such as microfracture matters here. When small holes are made in the bone to recruit stem cells, the tissue that forms is fibrocartilage — a scar-like fill with a different fibre structure, lower stiffness, and reduced capacity to handle repeated loading. In many patients, this fibrocartilage begins to break down within two to three years, particularly under the mechanical demands of sport or an active lifestyle.

For defects too wide for a single plug — roughly 2–4 cm² — mosaicplasty tiles several smaller cylindrical grafts side by side to achieve wider coverage. The technique, which entered clinical practice in 1992, takes its name from the mosaic-like pattern the plugs form at the repair site.

Critically, both approaches are single-stage procedures: the graft is harvested and implanted in the same operation. That distinguishes them from cell-based methods such as MACI, where a biopsy is taken in a first procedure and cells are cultured before a separate re-implantation.

The typical candidate profile

Patients most likely to benefit from OATS or mosaicplasty share a recognisable profile: relatively young, still physically active, and living with a discrete full-thickness cartilage injury that has not responded to conservative care such as physiotherapy, activity modification, and appropriate joint injections.

Age matters, though not as a strict cut-off. Evidence from large outcome series — including Campbell and colleagues' review of more than 1,100 patients across studies — consistently favours younger patients, with the strongest results in those under 34. Patients in their forties and even into their early fifties are routinely considered; under 34 is best understood as a prognostic benchmark rather than a gate.

The nature of the lesion is equally important. OATS and mosaicplasty are designed for focal, full-thickness cartilage loss — classed as ICRS grade III (deep fissuring reaching bone) or grade IV (complete cartilage loss with exposed bone). Diffuse or multi-compartmental joint disease falls outside this scope; it requires a different clinical conversation.

Appropriate underlying causes include traumatic cartilage injury, osteochondritis dissecans (OCD), focal spontaneous osteonecrosis of the knee (SONK), or a contained focal-wear lesion from repetitive loading. What these share is localised, structurally defined damage within an otherwise preserved joint.

Activity demand is the primary variable shaping outcome differences between techniques. In randomised trials restricted to competitive athletes, OATS delivers substantially higher rates of good clinical outcomes and return to sport than microfracture; in mixed-activity populations, the gap narrows considerably. For patients whose lifestyle or sport places sustained mechanical demands on the knee, that context shifts the balance further towards genuine hyaline-cartilage restoration.

Defect size — the main deciding factor

Defect area is the single most reliable guide to which cartilage repair technique fits a given patient — and the thresholds map fairly cleanly onto what each approach can physically achieve.

For lesions under approximately 2 cm², a single OATS plug — typically 8–10 mm in diameter — provides complete coverage of the defect in one step. It also addresses any subchondral bone loss at the same time, because the graft carries its own bone column, making it well-suited to osteochondral rather than purely chondral injuries.

Between 2 cm² and 4 cm², a single plug cannot fill the area; this is the primary territory for mosaicplasty, where several smaller cylinders are tiled together to achieve wider coverage. The practical ceiling here is not arbitrary — it reflects the amount of cartilage that can safely be harvested from low-demand areas of the same knee without creating a clinically meaningful donor-site problem. Exceeding that harvest capacity is the real constraint, not a preference for a different method.

Above 4 cm², that ceiling is reached, and the balance shifts towards techniques that do not rely on autograft volume: MACI or ACI, which culture the patient's own chondrocytes and seed them onto a membrane, or osteochondral allograft (OCA), which uses donor tissue for very large or post-traumatic defects. For lesions of 3 cm² or larger, data from the SUMMIT trial showed improved KOOS pain and function scores at two and five years with MACI compared to microfracture — suggesting that at areas where mosaicplasty coverage becomes technically difficult, MACI earns its additional complexity by outperforming even the microfracture fallback that might otherwise be used.

Surgeons treat these thresholds as practical guidance rather than a rigid algorithm; lesion shape, depth, and location all influence the final planning decision.

When OATS is not the right fit

Several clinical factors can make OATS unsuitable — some as hard barriers, others as practical concerns that careful surgical planning can sometimes overcome.

Absolute exclusions

Diffuse arthritis affecting more than one compartment, so-called 'kissing lesions' (cartilage damage on opposing joint surfaces that would face each other after repair), inflammatory joint disease such as rheumatoid arthritis, active joint infection, and joint-line malalignment that cannot be corrected all sit in this category. In each case the mechanical or biological environment cannot reliably support a graft, regardless of how well-defined the lesion appears.

Relative concerns — and when they can be addressed

A BMI above roughly 30–35 kg/m² increases loading through the repair site and may affect the healing environment; it does not automatically disqualify a patient, but surgeons weigh it carefully in planning. Uncorrected lower-limb malalignment carries a similar logic: excess force directed through the graft raises failure risk, yet in selected patients a corrective osteotomy (HTO to address varus, DFO for valgus) can be carried out at the same operation, restoring acceptable loading before the cartilage transfer proceeds. Where the meniscus is substantially deficient — more than approximately 50% — some surgeons consider a concurrent meniscal transplant to re-establish normal joint mechanics rather than treating it as an outright barrier.

Patients who have previously undergone microfracture present a further consideration: marrow-stimulation can alter the architecture of the underlying subchondral bone, which may complicate graft integration. Whether this is clinically significant depends on the extent of change, and specialist imaging is needed to assess it — it is not grounds for automatic exclusion, but it does warrant careful evaluation before proceeding.

What the long-term evidence shows

The strongest long-term comparison comes from Solheim's 2018 study of 203 patients followed for up to 18 years — Level III evidence, so not a randomised trial, but the largest and longest direct survival analysis available for these techniques. OAT repairs maintained survival above 80% for the first seven years and above 60% at 15 years. Microfracture, by contrast, fell below 80% within 12 months and below 60% within three years (log rank P<0.001). Overall failure rates were 51% for OAT versus 66% for microfracture (P=0.01), and mean time to failure was 8.4 years compared with 4.0 years — a gap that persisted when the analysis was restricted to patients matched for age and lesion size, reducing the risk that selection differences alone explained the result.

The mechanism behind this divergence becomes clearest under high physical demand. Fibrocartilage — the marrow-derived tissue microfracture produces — has lower stiffness and reduced load-distribution capacity compared with genuine hyaline cartilage. Under the pivoting, high-impact forces of sport, that structural inferiority becomes clinically apparent. Ten-year randomised data in athletes shows roughly double the rate of good or excellent outcomes with OATS compared to microfracture, with comparable return-to-sport advantages. In general, mixed-activity populations, however, RCTs have not shown a statistically significant difference, identifying activity demand as the primary variable driving the gap rather than a universal treatment effect. Large RCTs in non-athletes with follow-up beyond five years remain sparse, so how generalisable the survival advantage is outside sporting populations is not yet firmly established.

Systematic review evidence from Pareek et al. (2016) supports the functional picture: IKDC and Lysholm scores showed significant improvement from baseline at 10 years in OAT patients, though Tegner activity scores did not change significantly. A Campbell et al. analysis covering 1,117 patients reported significantly higher return-to-sport rates for OAT than for comparator techniques.

A 2025 retrospective study by Mukai and colleagues tested mosaicplasty in a different patient group: 39 knees with early-stage spontaneous osteonecrosis of the knee (SONK) and varus alignment, mean age 69.1 years. Over a mean follow-up of 90 months, the procedure produced significant long-term clinical improvement without graft failure — a small but notable signal that the technique's range may extend beyond the classic young-athletic profile. Given the retrospective design and fewer than 40 knees, the finding warrants confirmation in larger cohorts before it reshapes routine practice.

Choosing a specialist and next steps

Finding the right surgeon matters as much as selecting the right technique. OATS and mosaicplasty are specialist procedures — volume and regular practice in cartilage restoration are worth confirming rather than assumed from a general orthopaedic background. UK surgeons who work in this field typically hold subspecialty interests in sports orthopaedics or knee preservation; filtering by specialty and region when searching for a consultant is a practical starting point.

A pre-consultation MRI is the standard first step: it maps the lesion's size, depth, and the condition of the subchondral bone beneath, which together determine whether OATS or mosaicplasty is technically feasible, whether alignment correction is needed alongside the cartilage transfer, and whether a cell-based alternative such as MACI should be considered instead.

Timing rewards early referral. A contained focal defect may enlarge or develop subchondral bone changes if load continues through it unchecked — and a more complex lesion progressively narrows the restoration options. Arriving at a specialist assessment while the defect remains discrete keeps a single-stage procedure realistically available. The evidence reviewed across this article — on tissue quality, size thresholds, and long-term survival — gives patients a concrete basis for that consultation: the right questions to ask are about surgical volume, defect characterisation on MRI, and whether alignment or meniscal status needs to be addressed at the same time.