The short answer on MACI durability

For most patients, the functional gains from MACI hold across a full decade. That is the headline finding from two independent prospective cohort studies published in 2024 — Ebert et al. (University of Western Australia, 204 patients) and Wang et al. (American Journal of Sports Medicine, minimum 10-year follow-up) — both of which confirm that improvements in pain, function, and quality of life established within the first two years do not meaningfully erode over the following eight.

The pattern is a plateau, not a slide. Scores on validated outcome measures rise steeply in the first 24 months as the implanted cells mature and integrate into the surrounding tissue, then remain stable through year 10. This two-year maturation window is clinically important: the first couple of years are a distinct recovery phase, and the decade of documented stability comes after that phase, not instead of it.

This should not be read as a claim that MACI lasts forever. The current evidence base reliably covers 10 years, with some data extending to 17 years or more, but no long-term randomised controlled trial data yet exists beyond the 2-year mark. What the evidence does support — and what the rest of this article explores in detail — is that when MACI works, it tends to keep working for a clinically meaningful period.

Pain, function, and satisfaction at 10 years

The 92%/76% satisfaction split from the Ebert 2024 cohort does more than confirm durability — it clarifies that MACI delivers two related but distinct outcomes, and not equally.

For patients whose primary goal is relief from daily pain — walking without aching, managing stairs, sleeping through the night — the evidence is strongly favourable: nine in ten report lasting satisfaction at ten years. For those whose benchmark is a return to pre-injury sport, the picture is still positive but more qualified. One in four patients in this prospective cohort of 204 did not report sporting satisfaction at the decade mark, whether because of activity modification, changes elsewhere in the joint, or expectations that outpaced what cartilage repair can realistically deliver. A recreational runner hoping to resume training at their previous volume faces a meaningfully different calculation from someone whose goal is simply to walk a golf course comfortably.

The Ebert study assessed participants at two, five, and ten years — a design that tracks trajectory rather than capturing a single endpoint. Outcomes did not merely look good at ten years; they were consistent across all three assessment points following the initial maturation window, which is what lends the finding its clinical weight.

Wang et al. (AJSM, 2024) independently confirmed durable patient-reported improvements at minimum ten-year follow-up in a separate cohort, providing corroboration across different patients and institutions. Crucially, both studies are prospective cohort designs rather than randomised trials, and the populations studied were those already considered suitable candidates for MACI — a selection factor that clinicians take into account when contextualising these figures for any individual.

What MRI shows about graft integrity

Imaging tells a broadly consistent story. The Ebert 2024 study assessed 151 grafts using MOCART — a standardised MRI scoring system that rates cartilage fill, tissue integrity, and surface quality on a 0–100 scale specifically designed for graft surveillance. Across every MOCART variable measured, there was no statistically significant change between the two-year and ten-year scans. The structural plateau mirrors the functional one described in the patient-reported data.

Pooled findings from multiple prospective studies put the proportion of patients maintaining good-to-excellent tissue infill on MRI at the ten-year mark somewhere between 81% and 93%. In the Ebert cohort specifically, 9.3% of grafts — 14 of the 151 assessed — showed MRI-confirmed failure at ten years. That is a minority outcome, and it is worth noting that imaging failure and clinical failure are not the same thing: some patients with suboptimal MRI appearances continue to report acceptable function and pain relief, while a small number with structurally intact grafts remain symptomatic.

This distinction matters because MRI provides an important but incomplete window into how a graft is performing. Clinical assessment — how the knee feels and functions in daily life — remains the primary guide to whether the repair is working, with imaging used alongside rather than instead of that judgment.

Reoperation and joint replacement rates

Numbers give the longevity picture its sharpest edges. Across the major prospective cohort studies, the all-cause reoperation rate following MACI at ten years sits at roughly 9%, and progression to total knee arthroplasty ranges from 7.4% to 12%. For a procedure performed predominantly in active adults carrying substantial chondral defects — patients already at meaningful joint risk at baseline — those figures represent a low conversion burden rather than a warning signal.

The TKA rate warrants a brief note on interpretation. Not every conversion reflects a failed graft; a proportion will reflect disease progression elsewhere in the joint — advancing osteoarthritis in adjacent compartments, for example — that would have required arthroplasty regardless of the repair outcome. Graft failure and joint failure are distinct events, and the published data do not always separate them cleanly.

All of these figures come from prospective cohort registries — Level 4 evidence. No ten-year randomised controlled trial data for MACI exists yet, which means direct comparison against a control arm at the decade mark is not currently possible. That gap is real, but it does not invalidate what the observational record shows: that the large majority of patients with successful MACI grafts have not required further major surgery ten years on, in a population where doing nothing would itself have carried a significant risk of joint deterioration.

Why graft location changes the picture

Not all MACI procedures carry the same long-term track record, and the dividing line runs through anatomy. Ebert et al.'s 2024 ten-year cohort found that patients with tibiofemoral repairs — grafts placed on the femoral condyle or tibial plateau — achieved significantly better outcomes on KOOS measures than those with patellofemoral repairs. The difference is large enough to affect how the headline satisfaction figures should be read: the cohort's tibiofemoral cases predominate, so the 92% pain-satisfaction figure skews toward them.

Patellofemoral MACI is not without benefit, but the ten-year evidence base for it is thinner and average satisfaction is lower. Patients in this group face a more genuinely uncertain long-term picture — not a contraindication, but a prompt for careful pre-operative discussion about realistic expectations specific to their repair site.

Defect size adds another layer. MACI is most extensively studied in the 2–10 cm² range; lesions outside those boundaries introduce additional uncertainty about how well the procedure's published durability data translate. Age at the time of surgery, rehabilitation compliance, and whether the treatment is funded through an insurance or self-pay pathway can all influence how durably the graft integrates over time.

The practical upshot: a patient with a patellofemoral lesion should ask their surgeon directly what the ten-year data look like for that specific location, because the aggregate numbers — strong as they are — are largely a tibiofemoral story.

How MACI compares with older techniques

Placing MACI within its treatment lineage clarifies why the evidence base carries the weight it does for larger focal defects.

Microfracture — once the default first-line intervention — stimulates fibrocartilage rather than the hyaline-like tissue that healthy articular cartilage requires. That fibrocartilage is mechanically inferior; in published series it begins to ossify and break down within two to three years, and the drilling process can damage the subchondral bone plate, complicating any subsequent repair attempt. It retains a role in narrow indications but is no longer considered appropriate for defects much above 2 cm².

The SUMMIT Phase 3 RCT — 144 adults aged 18 to 54 — set MACI directly against microfracture and found MACI superior at two years on KOOS pain and function scores. That advantage held at five-year follow-up, particularly for defects measuring 3 cm² or more, which is precisely the size range where microfracture's structural limitations are most clinically consequential.

MACI's immediate predecessor, first-generation ACI, provides useful lineage context: a Minas cohort of 210 patients demonstrated 71% survivorship and improved function in 75% at ten years. MACI's collagen membrane scaffold addresses several technical drawbacks of the periosteal patch used in that earlier generation, and its published outcomes suggest at least comparable durability.

One variable crosses both ACI-family approaches: patients who have previously undergone marrow-stimulation surgery tend to show higher failure rates if they subsequently proceed to cell-based repair. Prior surgical history is therefore a material factor in determining which restoration pathway is appropriate — not a formality, but a clinical consideration that shapes individual prognosis meaningfully.