Which technique produces better outcomes?

The evidence consistently places MACI ahead — particularly for larger cartilage defects. The most rigorous head-to-head comparison, the SUMMIT Phase 3 randomised controlled trial (144 patients, mean defect size 4.8 cm²), found that MACI produced significantly greater improvements in KOOS pain scores (KOOS — a 0–100 scale measuring how the knee affects daily life and activity) at two years: 82.5 versus 70.9 for microfracture (p=.001). Improvements in KOOS function followed the same pattern (60.9 versus 48.7, p=.001), with MACI outperforming microfracture across all five outcome subscales.

Those gains did not fade with time. Five-year follow-up data from the SUMMIT extension (Brittberg et al., 2018) confirmed that the advantages in both pain and function were fully maintained, suggesting durability rather than a short-term effect.

Lesion size is the key qualifier. For defects smaller than roughly 2–3 cm², the gap between the two techniques narrows, and microfracture remains a defensible first-line option. The SUMMIT population had a mean lesion of 4.8 cm², and it is at or above the 3 cm² threshold — where fibrocartilage formed by microfracture is most likely to underperform — that MACI's superiority becomes most clinically meaningful.

Why cartilage repair type matters: hyaline vs fibrocartilage

The difference in long-term outcomes comes down to the type of tissue each technique produces.

Microfracture works by making small holes in the bone immediately beneath the damaged cartilage surface. This draws bone marrow stem cells up into the defect, where they form a repair patch. The tissue that results is fibrocartilage — a harder, less specialised material that behaves more like scar tissue than true knee cartilage. Fibrocartilage has lower stiffness and wear resistance than native hyaline cartilage, and under the repetitive loads of walking, stair-climbing, and physical activity it tends to break down over time.

MACI takes a different biological route. A small sample of the patient's own cartilage cells (chondrocytes) is harvested, cultured in a laboratory, and seeded onto a collagen membrane that is fixed into the defect during a second procedure. The repair tissue that forms is significantly more hyaline-like. MRI and histological assessments from the SUMMIT trial confirmed meaningfully better defect fill and repair-tissue quality in the MACI group compared with microfracture — findings that correlate directly with the superior clinical outcomes observed at two and five years. Meta-analytic evidence supports the same pattern: ACI and MACI consistently show higher rates of adequate defect fill on imaging than microfracture.

There is also a practical structural concern with microfracture. Perforation of the subchondral bone plate — the thin layer of bone that underpins cartilage — can alter the local bone architecture in ways that make any subsequent repair procedure more technically demanding. For patients who may need revision surgery in future, this is a meaningful consideration that a specialist assessment should address.

The SUMMIT trial: what the head-to-head data show

Examining the SUMMIT trial in detail matters because it is the only Phase 3 randomised controlled trial to directly compare the two techniques in a well-powered, prospectively designed study — making it the strongest single source of evidence for this decision.

The trial enrolled 144 adults with symptomatic grade III or IV focal defects of at least 3 cm² and a baseline KOOS pain score below 55 (indicating meaningful daily-life limitation). Mean age was 33.8 years and the average defect measured 4.8 cm². KOOS — the Knee injury and Osteoarthritis Outcome Score — rates pain, function, and quality of life on a 0–100 scale, where higher scores reflect fewer symptoms.

Starting from comparably low baselines (around 37 for pain and 13 for function in both groups), MACI brought pain scores to 82.5 at two years — a level broadly consistent with near-normal daily activity — while microfracture reached 70.9. Function scores followed suit at 60.9 versus 48.7, both differences reaching p=.001. The secondary subscales — activities of daily living (87.2 versus 75.8, p<.001), knee-related quality of life (56.2 versus 47.3, p=.029), and symptoms (83.7 versus 72.2, p<.001) — all favoured MACI significantly. These subscale results are the section of the SUMMIT data most directly relevant to patients asking what the evidence means for day-to-day living.

Safety outcomes were comparable between groups; no serious adverse events attributable specifically to the MACI implant were reported. That said, the population was relatively young (mean 33.8 years) with larger defects, so readers with smaller lesions or different demographic profiles should discuss how closely they match this trial population with a specialist.

How outcomes hold up at 5 and 10 years

The durability question — how long will the improvement actually last? — is where the two techniques diverge most starkly over time.

Five-year follow-up of the SUMMIT trial (Brittberg et al., 2018) confirmed that the MACI advantages in KOOS pain and function were fully maintained, with no evidence of regression. Longer data from Wang et al. (2024), reporting minimum ten-year outcomes, found that 88–93% of patients remained satisfied with their MACI result, graft failure occurred in only 7–11%, and conversion to total knee replacement remained low at approximately 7–10%. MRI confirmed stable graft tissue between the two-year and ten-year marks in most patients — suggesting the repair consolidates rather than deteriorates.

Microfracture follows a different trajectory. The fibrocartilage patch tends to break down under load: around six in ten microfracture repairs are classified as failures by a median of four years, and 15–20% of patients go on to need further surgical intervention.

Some meta-analyses note partial convergence in questionnaire-based functional scores beyond five to seven years. The important context is that MRI assessments consistently distinguish the two techniques even at that stage — MACI repairs retain better defect fill and more hyaline-like tissue quality even when self-reported scores narrow. What that tissue-quality difference means for outcomes beyond a decade remains an open question in the literature.

For patients considering MACI, outcomes appear best in those with a healthy BMI, fewer previous knee surgeries, and defects no larger than 4 cm² (Weishorn et al., 2024).

Lesion size, patient fit, and when microfracture is still reasonable

Size determines the starting point, but it is not the only variable in case selection.

Below the 3 cm² line, mosaicplasty — in which plugs of healthy osteochondral tissue are transferred from a low-load area of the knee to fill the defect — offers a single-stage alternative to microfracture that avoids MACI's two-stage pathway. Published meta-analytic evidence, including Muthu et al. (2024), does not show a decisive MACI advantage within this smaller-defect range, and where the evidence is broadly comparable, the simpler, less resource-intensive procedure generally makes more practical sense. Defect geometry and location — whether the lesion is accessible for autograft transfer, for example — influence which of these single-stage options is more appropriate.

From 3 cm² to approximately 10 cm², MACI is both licensed and the better-evidenced choice. At this scale, fibrocartilage's susceptibility to load-related breakdown — discussed above — carries its greatest clinical consequence, and the additional procedural investment becomes proportionate to the expected long-term benefit.

Patient-level factors add a further layer beyond defect size. High BMI, complex defect geometry, and a history of multiple prior knee surgeries — in particular earlier marrow-stimulation procedures, which can alter the subchondral bone surface and compromise later repair — each affect how well any restoration approach is likely to perform (Weishorn et al., 2024). There is also an evidence ceiling by age: with a mean trial age of 33.8 years in SUMMIT, the comparative data are strongest in younger adults, and how closely those outcomes apply in patients over 50 with similar lesion sizes remains a genuine gap that warrants specialist assessment.

Practical trade-offs: staging, cost, and finding a specialist

The practical gap between the two procedures is worth naming plainly. Microfracture is performed in a single arthroscopic session; MACI requires two separate operations — a first-stage biopsy to harvest chondrocytes, a period of laboratory cell culture spanning several weeks, and then a second procedure to implant the seeded collagen membrane. That staging has direct consequences for planning and time away from work.

Cost follows the same pattern: the laboratory processing involved in MACI makes it substantially more expensive than single-stage microfracture. For smaller defects where published evidence shows comparable results, that additional burden is difficult to justify on clinical grounds alone. Where the defect is large enough for MACI's durability advantage to matter, the two-stage pathway and higher cost are the central trade-offs — not safety, which the SUMMIT trial found comparable between the two techniques, with no MACI-specific serious adverse events identified.

Recovery involves protected weight-bearing and structured rehabilitation in both cases. MACI programmes are generally longer, as the cellular scaffold needs time to mature and integrate before the joint is progressively loaded.

For symptomatic focal defects above 3 cm², the published evidence now makes MACI the stronger default in most clinical scenarios; the individual question is whether a two-stage pathway is practical and appropriate for a given patient. A cartilage specialist experienced across the full range of restoration options is best placed to work through that assessment. Search MSK lists knee cartilage repair specialists across the UK, organised by region and specialty, to help identify a clinician suited to your situation.