What the 10-year RCT actually found

'Will this actually hold up long-term?' It is the question most active patients ask before committing to cartilage surgery, and for mosaicplasty versus microfracture it has a direct, randomised answer — which is unusual in this field.

Gudas et al. conducted a prospective randomised controlled trial comparing the two techniques in young athletes with osteochondral defects of the knee. The 10-year follow-up, published in the American Journal of Sports Medicine in 2012, is the only randomised comparison of mosaicplasty and microfracture to reach that horizon — a level of evidence that remains rare in cartilage surgery. At that point, the mosaicplasty group demonstrated significantly superior clinical outcome scores. Crucially, the two techniques had appeared broadly equivalent in the short term; that equivalence dissolved between five and ten years, as microfracture results declined and mosaicplasty results held.

Corroborating evidence comes from Pareek et al.'s 2016 systematic review of osteochondral autograft transfer outcomes at 10-year follow-up. IKDC and Lysholm scores — two widely used measures of knee function and symptoms — were significantly improved from baseline, supporting the finding that functional gains from OATS are durable rather than transient.

One nuance deserves equal weight: Tegner activity-level scores did not significantly improve in the Pareek analysis. Function recovers, but full restoration to the pre-injury level of sport is not guaranteed for every patient. That distinction matters when setting expectations before surgery.

Why the tissue type drives the long-run result

The answer lies in what tissue you actually end up with.

OATS is a transplant, not a repair stimulus. Cylindrical plugs of full-thickness cartilage and underlying bone are harvested from a low-load zone of the patient's own knee and press-fitted into the defect. The tissue transferred is mature hyaline cartilage — the same material that lines healthy joint surfaces — with its original collagen-II architecture and proteoglycan density intact from the moment it is placed.

Microfracture works on an entirely different principle. Small perforations are made through the subchondral bone plate to allow marrow stem cells to migrate into the defect and clot. The tissue that forms is fibrocartilage — a scar-like repair material that is biochemically distinct from native joint cartilage. It contains a higher proportion of collagen-I (the same collagen found in tendons and skin) rather than the collagen-II that gives hyaline cartilage its load-bearing resilience.

Under repetitive impact loading — the condition active patients place their knees in every training session — fibrocartilage fatigues. It lacks the structural density to resist the cyclic compression that hyaline cartilage tolerates routinely. This is not a theoretical concern: it is the most likely explanation for why microfracture outcomes, in multiple series, deteriorate between two and five years post-surgery, and why that deterioration continues. The transplanted hyaline cartilage in an OATS repair does not have the same vulnerability, which is precisely why short-term equivalence between the two techniques dissolves as follow-up lengthens.

Which patients are suitable for mosaicplasty

Two variables do most of the sorting work before any clinical assessment begins: age and defect size.

The evidence consistently identifies patients under 25 with lesions smaller than 2 cm² as the group most likely to return to their pre-injury level of sport after mosaicplasty. Both factors appear across multiple studies as positive prognostic markers, and they reflect a biological reality — younger joints integrate grafts more reliably, and smaller defects can be filled with fewer plugs, reducing the demands placed on the donor site.

Mosaicplasty remains a reasonable option for focal defects up to roughly 4 cm², managed as a mosaic of multiple plugs. The upper boundary matters, however. Beyond approximately 3–4 cm², available autograft is limited by how much tissue can safely be harvested from the knee's non-weight-bearing zones without creating its own morbidity. At that size threshold, the appropriate step-up is either MACI or osteochondral allograft (OCA), not a larger mosaic. The SUMMIT trial established that MACI produces significantly improved KOOS pain and function scores over microfracture for defects of 3 cm² or greater at both two and five years — useful context for patients sitting near that boundary when weighing their options.

Equally important is what mosaicplasty does not treat. It is a focal-defect procedure suited to joints that are otherwise well-preserved. Diffuse cartilage loss or established osteoarthritis falls outside its scope; the RCT evidence base is drawn from young athletic populations with discrete, symptomatic lesions.

Confirming which category a patient falls into requires MRI and clinical assessment — which is the natural starting point for the evaluation pathway.

Return to sport — rates, timelines, and realistic expectations

Roughly three in four patients who undergo cartilage restoration surgery return to sport. For mosaicplasty specifically, the rate is meaningfully higher than for alternatives: Campbell et al.'s systematic review of 1,117 patients found osteochondral autograft transfer associated with significantly better return-to-sport rates compared with other repair techniques. Pooled across all cartilage procedures, a meta-analysis of 2,549 patients puts the figure at around 76–78%, with approximately 72% reaching their pre-injury level; the average time to return, across all techniques, was 11.2 months.

The timeline is the trade-off that needs stating clearly. Mosaicplasty takes longer to recover from than microfracture — the evidence shows a significantly greater average duration to return to previous sporting activity in the mosaicplasty group. Patients choosing mosaicplasty on the strength of its durability evidence need to factor in that longer rehabilitation arc before the advantage materialises.

In high-demand sport, published series report more encouraging figures than population-wide averages suggest. In competitive football, return rates of around 83% have been described, with 80% returning to the same competitive level and between 87% and 100% maintaining the ability to play at five years post-operatively — numbers described as among the highest reported across cartilage restoration procedures.

Returning to sport is not, however, the same as returning to the same activity level. Functional outcome scores improve durably after mosaicplasty, but activity-level recovery is less consistent at ten years — a distinction worth making plain before a patient commits. The prognostic factors that influence which side of that line a patient falls on are the same variables central to patient selection: age and lesion size.

Donor-site morbidity, recovery demands, and cost context

Harvesting cartilage plugs from the patient's own knee creates a secondary wound at the donor area — typically the periphery of the femoral condyle or the trochlear groove. Donor-site symptoms, including pain, stiffness, and residual aching, are a documented complication and are more common when a greater number of plugs are taken. Published series have not converged on a single rate, because frequency depends on plug count and individual patient factors; it is a real enough risk to address explicitly before surgery, not to minimise.

The number of plugs that can safely be harvested per session is also finite. This limits the volume of defect that autograft can cover — beyond roughly 3–4 cm², both donor supply and morbidity constraints make a larger mosaic impractical.

Recovery is more demanding than a microfracture procedure, which involves no harvesting at all. A supervised physiotherapy programme lasting several months is the norm before sport becomes a realistic goal; compressing that timeline risks graft integration failure.

Against these trade-offs, cost gives useful context. In the UK private setting, OATS/mosaicplasty is priced at approximately £14,000 — above the £4,000 starting point for microfracture and well below the £28,000 range for osteochondral allograft. The difference broadly tracks procedural complexity and tissue source rather than clinical superiority at a given defect size.

Individual risk profiles, plug-count requirements, and likely recovery course are best established through clinical assessment.

Getting an accurate assessment in London

Ten years of follow-up data and a single direct randomised answer are rare in cartilage surgery — and the evidence that does exist points clearly in one direction for the right patient. Acting on it, though, requires knowing whether a given patient actually fits that profile. Defect size, depth, and subchondral bone condition all need to be characterised on MRI before any technique can be responsibly recommended; those findings cannot be inferred from symptoms alone.

The choice between mosaicplasty, MACI, and osteochondral allograft is then made on individual anatomy and activity goals, not a universal algorithm. Plug count, donor-site geometry, and defect boundaries each bear on the outcome, and assessing them demands clinical examination alongside imaging.

At the London Cartilage Clinic on Harley Street, Professor Paul Y. F. Lee assesses patients at this decision point and performs cartilage restoration procedures including OATS/mosaicplasty. Patients can arrange a consultation via londoncartilage.com.