The defect size that changes the decision

Defect area is the number that shifts the decision. For focal cartilage lesions smaller than roughly 2 cm², microfracture and OATS produce broadly comparable results in the short term, and the long-term gap between them, though present, is relatively modest. Push the lesion area past 2–4 cm², and the data begin to diverge materially: marrow stimulation becomes progressively less reliable, while OATS continues to hold its advantage at extended follow-up.

This size threshold is not arbitrary. The 2021 Cartilage Book notes that for defects under 2–4 cm² both microfracture and osteochondral autografting are reasonable treatment options, but that mosaicplasty has been shown to produce higher clinical scores than microfracture in long-term follow-up. Active patients with larger lesions face a durability problem that defect size alone explains before technique even enters the conversation — the bigger the void, the harder it is for marrow-derived cells to sustain a structurally adequate repair across years of repeated loading.

OATS addresses this differently. The procedure harvests cylindrical plugs of full-thickness hyaline cartilage on a bone base from a low-load zone of the same knee, then press-fits them into the prepared defect — restoring the joint surface with the patient's own tissue in a single stage, with no donor cost and no second operation. It is that structural difference — genuine hyaline cartilage on a bony scaffold rather than a marrow-derived fibrous repair — that drives the 10-year outcome argument explored in the sections that follow.

What microfracture actually repairs — and why that matters at five years

Perforating the subchondral bone plate is microfracture's entire mechanism. The surgeon uses an awl to punch through the calcified layer beneath the damaged cartilage surface, releasing bone marrow blood and mesenchymal cells into the defect. Those cells form a clot that slowly matures — over weeks to months — into repair tissue. That tissue is fibrocartilage, not the native hyaline cartilage it replaces.

The distinction matters biomechanically. Fibrocartilage is stiffer, less elastic, and organised differently at the collagen-fibre level than articular hyaline cartilage. Under low or moderate load it can function adequately, which explains why two-year results from microfracture are often reported as acceptable. The problem is durability. Evidence suggests that fibrocartilage repair tissue can deteriorate after three to five years of repetitive joint loading — precisely the timeframe in which short-term trial data tend to end and longer follow-up begins to reveal the gap.

For active patients, that timeline is compressed further. High-impact and high-frequency loading accelerates fibrocartilage breakdown, making the physically active group the population most exposed to this failure mode.

There is a second consequence that bears noting. Perforating the subchondral bone plate is not a neutral act: it alters the bone architecture beneath the joint surface, and this disruption can complicate any revision cartilage procedure the patient may require later. Microfracture is therefore not simply a failed first attempt — in some cases it reduces the technical options available if further surgery becomes necessary.

How OATS restores the original tissue architecture

The structural logic behind OATS starts with what native articular cartilage actually is: not a single layer but a two-part osteochondral unit — bone below providing a vascular, load-distributing base, and hyaline cartilage above providing the smooth, durable articulating surface. Microfracture leaves that two-part architecture partially intact but fills the cartilage void with fibrous repair tissue. OATS replicates both layers simultaneously.

A cylindrical plug is harvested from a region of the same knee that bears lower contact stress — typically the far medial or lateral condyle — and press-fitted into a matching channel prepared at the defect site. Each plug carries a full-thickness column of hyaline cartilage sitting on its own bone base, and because it is press-fitted under compression, it integrates with the surrounding subchondral bone while the cartilage surface aligns flush with the native joint. The result is structurally continuous with adjacent tissue in a way that a marrow-derived fibrous fill is not.

Hyaline cartilage withstands compressive and shear forces differently from fibrocartilage: its collagen architecture is organised to distribute load across the joint surface rather than concentrate it. Under repetitive active use — the condition that exposes microfracture's limitation over time — this matters considerably.

For smaller contained lesions, a single cylindrical plug suffices. Larger defects up to roughly 4 cm² require multiple plugs arranged in a mosaic pattern. That is where the technique carries a genuine caveat: the spaces between individual grafts cannot be filled with additional plug tissue, and those gaps tend to fill instead with fibrocartilage. The greater the number of plugs required, the larger the proportion of the repaired surface that carries the same repair-tissue limitation OATS was chosen to avoid. Precise placement and sizing reduce this problem but cannot eliminate it entirely.

Donor-site morbidity is a separate and honest consideration: harvesting healthy cartilage from any site in the joint introduces a secondary defect, and although the donor region is chosen to minimise functional impact, it is not without risk.

What ten-year data actually show

Longitudinal trial evidence anchors the comparative argument firmly at ten years. The landmark RCT by Gudas and colleagues — which enrolled active patients with femoral condyle defects and randomised them to either OATS or microfracture — remains the primary long-term data source underpinning the preference for osteochondral autograft transfer in this population. Its long-term follow-up, reported around 2012, confirmed that mosaicplasty produced significantly higher clinical scores than microfracture at ten years, and that the gap between procedures was clinically meaningful rather than marginal.

The important detail is when that gap appears. Early results in the Gudas cohort did not show the same stark separation: at two to three years, microfracture patients can report outcomes that look broadly comparable. The divergence accumulates as the follow-up clock runs — consistent with what the biology predicts. Fibrocartilage repair tissue, adequate under early post-operative conditions, comes under increasing mechanical stress with continued activity, and its structural inferiority becomes apparent in the data progressively rather than suddenly.

For defects of 3 cm² or greater, the SUMMIT trial adds a complementary perspective. That trial compared MACI against microfracture and found significantly improved KOOS pain and function scores at both two and five years in the larger-lesion group — reinforcing that marrow stimulation fails most visibly where the defect area is greatest. SUMMIT data extend to five years rather than ten, and it tested MACI rather than OATS, but the directional finding aligns: size amplifies the long-term disadvantage of fibrocartilage repair.

The active-patient dimension is where the ten-year gap becomes most consequential. Repetitive high-load activity applies the mechanical stress that fibrocartilage is least equipped to withstand over time, which is precisely why physically active patients — the group generating the greatest cumulative joint loading — are those in whom the OATS durability advantage is most pronounced at extended follow-up.

The return-to-sport trade-off active patients need to weigh

Durability and speed of recovery pull in opposite directions, and for an active patient that tension is real. Return-to-sport time is significantly longer after mosaicplasty than after microfracture — a finding from the same evidence base that shows OATS winning at ten years. Microfracture's shorter rehabilitation reflects the nature of its repair: marrow cells consolidate faster than transplanted osteochondral plugs integrate, so patients may return to training sooner. The difficulty is that the tissue supporting that earlier return is less durable under load.

Who genuinely needs to weigh this most carefully? Prognostic data point clearly to two variables: age under 25 and lesion size under 2 cm². Both independently favour a successful return to the same level of sport, whichever procedure is chosen. For a younger patient with a small, contained defect in that range, microfracture's shorter recovery may carry less long-term risk than in someone older with a larger lesion.

Beyond 2 cm², the calculus shifts. The longer rehabilitation after OATS is the cost of investing in repair tissue that holds up across a decade of active use. Accepting that timeline is not the same as prolonging injury — it is building the structural foundation that avoids a revision procedure or an earlier decline in joint function.

Patients should ask their surgeon for realistic return-to-training milestones before choosing based on speed alone. The right answer depends on lesion size, age, sport, and how much time the patient can afford to invest now to protect function later.

Who OATS suits — and where other options fit

Three variables determine whether OATS is the right step: defect size, patient age and BMI, and whether conservative measures have already been tried and failed. OATS is not a first intervention — it is indicated for patients who remain symptomatic after non-operative management, typically for contained lesions in the 1–4 cm² range, in individuals under 50 with a BMI below 40 and no more than Kellgren-Lawrence grade 2 osteoarthritis at the affected joint.

Outside those parameters, the choice shifts. Lesions larger than 4 cm² exceed what mosaicplasty can practically address; ACI, MACI, or fresh osteochondral allograft (OCA) are the options typically considered for larger or post-traumatic defects. At the other end of the size range, a smaller contained focal lesion in the right patient may be appropriate for a ChondroFiller injection — an ultrasound-guided outpatient collagen scaffold treatment that avoids an operating theatre for eligible patients.

None of these decisions reduces to a straightforward checklist. Defect size, activity level, age, underlying bone quality, and what has already been tried all shape the recommendation, and the evidence supporting each option differs in depth and follow-up duration. A specialist assessment is the appropriate starting point; the London Cartilage Clinic on Harley Street offers that evaluation for patients considering cartilage restoration, and further information is available at londoncartilage.com.