What the trial data show at ten years

At ten years, the gap between mosaicplasty and microfracture is not subtle. The Gudas prospective randomised controlled trial — the only RCT in young athletes to reach a genuine ten-year endpoint for these two procedures — recorded treatment failure in 14% of mosaicplasty patients compared with 38% of those who had microfracture, a difference that was both statistically significant and clinically meaningful. Return-to-sport figures told the same story: 84–100% of mosaicplasty patients were back competing at ten years, against 44–83% of the microfracture group, with the latter more likely to be playing at a reduced level or to have withdrawn from sport entirely. Gudas and colleagues subsequently extended this cohort to 15–17 years — among the longest RCT-level follow-up periods completed for either procedure — and the mosaicplasty advantage persisted across that timeframe.

Survival-curve data from the Solheim 2018 comparative cohort (microfracture n=119, osteochondral autograft transfer n=84) sharpen the picture further. Mean time to failure was 8.4 years (SD 4.8) in the OAT group and 4.0 years (SD 4.1) in the microfracture group (P=0.01). Kaplan-Meier analysis showed OAT survival remaining above 80% for the first seven years and above 60% at fifteen years. Microfracture followed a markedly steeper trajectory: survival dropped below 80% within twelve months of surgery and below 60% within three years — a pattern that diverges from OAT almost immediately post-operatively.

One important caveat: a 2026 BMJ Open systematic review found that when comparing OATS with microfracture on patient-reported outcome measures from RCTs alone, the evidence of a difference carried only low certainty. That qualification applies to PRO-level scoring in relatively small trials and does not overturn the survival-curve and failure-rate data above, which reflect durability rather than symptom scores at a single time point. The two bodies of evidence are measuring different things; the survival data are the more relevant lens when the question is how long each repair is likely to last.

Why fibrocartilage breaks down before hyaline cartilage

The reason the survival curves diverge so dramatically comes down to tissue type. When microfracture perforates the subchondral bone, marrow-derived cells migrate into the defect and produce repair tissue — but that tissue is fibrocartilage, dominated by type I collagen. Think of it as scar tissue for a joint: it fills the cavity and provides short-term relief, but it lacks the structural architecture of native articular cartilage. OATS transfers intact osteochondral plugs carrying genuine hyaline cartilage — type II collagen arranged in the layered, load-distributing configuration of the original surface.

The mechanical consequences follow a predictable timeline. In the first two years, both repairs can perform similarly: the fibrocartilage is intact, symptoms are often improved, and neither technique shows a clear clinical advantage at this stage. From around five years, the difference begins to emerge. Fibrocartilage is softer and less resistant to the cumulative stresses of walking, stair-climbing, or sport, and it fatigues progressively under repetitive load. By ten years the gap is consistent and measurable — precisely when the trial and cohort data described above crystallise it numerically.

Histological evidence reinforces this. A NICE meta-analysis of 33 studies with follow-up to ten years confirmed that microfracture produces poorer tissue quality at biopsy than other cartilage repair techniques.

There is a further structural consequence that goes beyond the initial repair: microfracture also disrupts the subchondral bone plate. That disruption can compromise the bone bed available to any subsequent procedure, which matters considerably when a first repair fails in a younger, high-demand patient.

Which patients are most likely to benefit from OATS

Defect size is the single most important factor a surgeon will weigh when choosing between OATS and microfracture. For smaller lesions — broadly under 2 cm² — the evidence picture is more mixed, and the technical simplicity of microfracture has historically made it the default choice. That calculus shifts as lesion area grows. A 2026 decision-tree analysis of knee chondral management identified lesion size, age, and activity level as the primary treatment-differentiating factors, with OATS favoured for larger or higher-demand scenarios. Mosaicplasty, using multiple plugs arranged to resurface the defect, can extend the OATS approach to lesions approaching 4 cm²; beyond that threshold, techniques such as MACI or fresh osteochondral allograft (OCA) typically enter the conversation.

Osteochondritis dissecans (OCD) of the medial femoral condyle is one of the clearest shared indications where OATS data are particularly robust. A 2024 study reporting minimum ten-year survivorship after single-plug osteochondral autograft for knee OCD provides some of the most specific long-term evidence available for this patient group.

Activity level is the next modifier. The Gudas trial was conducted exclusively in young athletes, which is why its return-to-sport figures are so striking — the population was precisely the one where repair tissue durability is most tested. Younger patients carry additional significance: more years of knee loading ahead means the difference between fibrocartilage and hyaline cartilage compounds over time, making durable tissue quality more consequential the earlier surgery is performed.

One consideration that should be discussed pre-operatively is donor-site morbidity. Harvesting osteochondral plugs from a lower-load zone of the same knee carries a reported complication rate of approximately 14–20%, including persistent discomfort at the harvest site. This is a genuine trade-off — microfracture carries no equivalent harvest risk — and patients should weigh it alongside the longer-term durability data.

The gap between evidence and surgical practice in the UK

Numbers rarely make a clinical argument as plainly as these: across 19 UK centres over five years, surgeons performed 1,579 microfracture procedures for isolated knee cartilage defects. Over the same period, the total OATS count was 29 — and the procedure was not available at more than half the participating centres. That figure comes from a collaborative UK study published in 2025 and represents a measurable gap between what the evidence supports and what most patients are actually offered.

NICE guidance, supported by the meta-analysis of 33 studies with follow-up to ten years noted earlier in this article, identifies microfracture as producing poorer histological outcomes than other repair techniques. The implementation gap is difficult to reconcile with those findings for medium-to-large defects.

The reasons are structural rather than negligent. OATS and mosaicplasty require specialist training, specific instrumentation, and confidence in plug harvest and press-fit technique. Microfracture is simpler, faster, and deeply embedded in surgical training pathways as a default response to a focal defect. That kind of institutional inertia changes slowly, and resource constraints in busy services reinforce it.

For younger, active patients with lesions approaching or exceeding 2 cm², this matters in practical terms. If microfracture has been recommended without discussion of alternatives, it is entirely reasonable to seek a specialist opinion at a centre with OATS or mosaicplasty expertise before accepting it as the only option — not as a challenge to the treating surgeon, but as a question the ten-year evidence supports asking.

The downstream cost when microfracture fails in younger patients

The consequences of choosing microfracture in a younger, active patient extend well beyond the initial repair. When the procedure fails and salvage surgery becomes necessary, the most common next step — autologous chondrocyte implantation (ACI) — produces inferior results compared to primary ACI. In one series of 92 patients with a mean age of 30.1 years, the average time from microfracture to ACI was 21.2 months, and outcomes were consistently poorer than in patients who had received ACI as their first procedure. That inferiority is partly explained by the subchondral plate damage discussed earlier in this article: as already noted, microfracture disturbs the calcified layer that subsequent repair procedures depend upon, compressing the biological window available for any future intervention.

The durability problem is compounded by questions about whether microfracture confers meaningful early benefit at all. A 2025 multicentre double-blinded RCT found microfracture was not superior to arthroscopic debridement alone for femoral cartilage lesions smaller than 2 cm² over two years — precisely the size range where microfracture has historically been applied most freely. For medium-to-large defects, a 2024 systematic review found limited efficacy at mean ten-year follow-up, reinforcing the survival-curve data covered earlier.

For patients who have already had microfracture and are experiencing renewed symptoms or functional decline, early specialist review matters. Options including MACI, osteochondral allograft, or joint-preservation surgery are more straightforward to plan before further cartilage loss or subchondral deterioration accumulates. Delay rarely improves the range of available choices.

Getting an assessment for focal knee cartilage repair

The evidence reviewed in this article converges on a practical point: for focal knee cartilage defects approaching or exceeding 2 cm² in patients with significant activity demands, technique selection matters enormously over a ten-year horizon. Choosing a procedure on the basis of availability or surgical familiarity, rather than defect characteristics and patient profile, carries measurable long-term costs — both in repair durability and in the quality of any salvage options that remain.

A thorough pre-operative assessment should include MRI with cartilage-specific sequencing. T2 mapping and cartilage segmentation can characterise defect depth, geometry, and subchondral bone integrity more precisely than standard sequences, informing whether OATS, MACI, or another pathway is most appropriate. Functional evaluation and a full procedural history — including any prior marrow-stimulation surgery — complete the picture a specialist needs before technique selection.

Lincolnshire Knee, part of the MSK Doctors group, accepts patients without GP referral at sites in Sleaford NG34 and Grantham NG31. Where indicated, AI-driven knee MRI analysis through onMRI™ can support cartilage assessment. Book an appointment at lincolnshireknee.co.uk.

1. [1] Randomized Study of Long-term (15-17 Years) Outcome After Microfracture Versus Mosaicplasty in Knee Articular Cartilage Defects. (2018). https://doi.org/10.1177/0363546517745281 https://doi.org/10.1177/0363546517745281
2. [2] Comparative effectiveness of osteochondral restoration, cell-based regeneration and cell-free implants for knee chondral defects: a systematic review. (2026). https://doi.org/10.1136/bmjsem-2025-002889 https://doi.org/10.1136/bmjsem-2025-002889
3. [3] Outcomes of Autologous Chondrocyte Implantation in the Knee following Failed Microfracture. (2016). https://doi.org/10.1177/2325967116S00125 https://doi.org/10.1177/2325967116S00125
4. [4] Treatment options and outcomes for paediatric knee cartilage lesions: a systematic review. (2025). https://doi.org/10.1016/j.knee.2025.08.020 https://doi.org/10.1016/j.knee.2025.08.020
5. [5] Long-Term Survival after Microfracture and Mosaicplasty for Knee Articular Cartilage Repair: A Comparative Study Between Two Treatments Cohorts. (2018). https://doi.org/10.1177/1947603518783482 https://doi.org/10.1177/1947603518783482
6. [6] Articular fibrocartilage — Why does hyaline cartilage fail to repair?. (2019). https://doi.org/10.1016/j.addr.2018.12.015 https://doi.org/10.1016/j.addr.2018.12.015
7. [7] A decision tree model for chondral lesion management in the knee based on clinical and mathematical integration. (2026). https://doi.org/10.1016/j.jisako.2026.101131 https://doi.org/10.1016/j.jisako.2026.101131
8. [8] The management of cartilage defects of the knee and injection therapy — A collaborative and retrospective study. (2025). https://doi.org/10.1016/j.knee.2025.11.013 https://doi.org/10.1016/j.knee.2025.11.013