Why MACI is always two stages — and what that means in practice

Patients often arrive at clinic having read about 'single-stage versus two-stage MACI' and wondering which version they might be offered. The short answer is that no single-stage variant of MACI exists — the procedure is, by biological necessity, always two separate operations. The comparison that actually matters is between MACI and other techniques that can be done in a single session, such as AMIC or the emerging STACI platform. Understanding why MACI requires two stages makes that broader choice much clearer.

The first stage is a brief arthroscopic procedure, typically around 30 minutes, in which the surgeon harvests a small sample of cartilage — around 200–300 mg — from a low-load region of the knee where the joint surface is healthy. The joint itself is not repaired at this point; the biopsy site heals without intervention.

That tissue is then sent to a specialist laboratory, where over the following 4–8 weeks the chondrocytes (cartilage cells) are carefully expanded in culture and seeded onto a porcine Type I/III collagen membrane. The result is a living, cell-rich scaffold tailored to the patient.

The second stage implants this scaffold directly into the prepared cartilage defect, where it is secured using fibrin sealant — no sutures are passed through cartilage, which reduces disruption to the repair site.

Historically, MACI was described as 'simpler' than first-generation autologous chondrocyte implantation (ACI), but that label referred only to what the surgeon does during the second operation — handling a membrane rather than harvesting and fixing a periosteal flap. It never meant fewer operations. The two-stage structure is the architecture of the biology, not a design choice that can be shortened.

Which patients are suitable for MACI

Three questions govern whether MACI is appropriate for a given patient: how large is the defect, how old and active is the patient, and what has previously been done to that part of the knee?

Defect size — the primary gate

Lesions below roughly 1.5–2 cm² can usually be addressed with single-stage options such as OATS or nano-drilling, where the two-stage commitment of MACI is rarely justified. Between 2 and 4 cm², MACI competes directly with single-stage alternatives on both clinical and practical grounds. Above approximately 4–4.5 cm², the conversation shifts toward osteochondral replacement rather than cell-based repair.

For lesions of 3 cm² or more, the SUMMIT trial established that MACI produces better KOOS pain and function scores than microfracture at both two and five years — making defect size not just a threshold question but a guide to which comparator is most relevant.

Age, biology, and joint condition

MACI is generally appropriate for patients up to around 55 years old with focal, contained defects and healthy cartilage at the margins. Both the surrounding tissue and the subchondral bone beneath the defect need to be in reasonable condition. Diffuse or advanced osteoarthritis places a patient outside the scope of cell-based repair, regardless of age.

Prior treatment to the same site

A history of microfracture to the same area is a material factor that surgeons weigh carefully. Evidence — examined in detail in a later section — suggests that MACI applied to a previously microfractured knee may not achieve the same long-term durability as it does in treatment-naïve cases. This caution is distinct from the SUMMIT finding above: the trial compared MACI against microfracture as competing first-line treatments, whereas the durability concern arises when microfracture has already been performed and MACI is being considered as a salvage or revision option.

Single-stage alternatives and when they compete with MACI

Single-stage is not automatically inferior to MACI — it is differently positioned, and for certain patients it represents the more pragmatic choice.

AMIC — the most established comparator

AMIC (autologous matrix-induced chondrogenesis) combines microfracture with immediate placement of a collagen membrane in a single surgical episode. A matched-pair retrospective analysis (Schneider, n=48) found no significant difference between MACI, AMIC, and minced cartilage implantation across VAS pain and all KOOS domains at two years — all three techniques produced meaningful improvements from baseline. That equivalence at the two-year mark is clinically important and should not be minimised when counselling patients whose primary concern is near-term recovery.

MACI's advantage becomes more apparent for larger defects and over longer follow-up. The SUMMIT trial demonstrated superiority over microfracture for lesions of 3 cm² or more at both two and five years — though that comparison is against marrow stimulation alone, not membrane-augmented single-stage repair.

STACI — emerging, not established

STACI (single-treatment autologous chondrocyte implantation) processes chondrocytes intraoperatively, removing the 4–8 week culture interval entirely. It is positioned for highly active patients under roughly 40–45 with focal defects. Critically, no head-to-head randomised trial comparing STACI with MACI exists at long-term follow-up — a material gap when advising younger patients who might otherwise be MACI candidates. STACI carries 'next-generation' rather than established status, and its long-term durability remains to be confirmed in larger series.

The practical weight of one fewer anaesthetic

Removing a second general anaesthetic and a second rehabilitation block has genuine real-world weight. For working patients, the difference in life disruption between one operation and two belongs in the clinical conversation alongside tissue-quality considerations — it is a legitimate factor, not a secondary afterthought.

What 10-year evidence actually shows

For most patients, the central question is simple: will the procedure still be making a difference a decade from now? The clearest answer comes from two large published datasets.

Ebert et al. 2024 prospectively enrolled 204 patients and followed 168 of them to 10 years, with MRI review of 151 grafts — the most detailed long-term cohort for MACI to date. Patient-reported outcomes improved significantly by 2 years (p<0.0001) and then held steady: there was no statistically significant further change between 2 and 10 years. That plateau is worth presenting plainly — it reflects durable stabilisation, not gradual decline. At the 10-year mark, 92% of patients were satisfied with their pain relief and 76% were satisfied with their ability to participate in sport. MRI MOCART scores were similarly stable from 2 to 10 years, indicating that surviving grafts maintained their structural appearance rather than deteriorating on imaging.

Graft failure — defined as delamination — occurred in 9.3% of grafts (14 out of 151). That figure deserves to be stated directly: MACI is not uniformly successful, and around 1 in 11 grafts will not survive to 10 years.

A systematic review (Wang et al. 2024) covering 188 defects across 10–17 years of follow-up produced broadly consistent numbers: 9.0% all-cause reoperation and 7.4% progression to total knee arthroplasty. The convergence across two independent datasets gives those headline figures reasonable weight.

A location-specific caveat

Tibiofemoral MACI produced better long-term clinical scores than patellofemoral MACI at 10 years, even though MRI structural appearances were similar between the two sites. This is a clinically important split — patients with patellofemoral defects should understand that the functional track record at 10 years is less favourable than the tibiofemoral picture, despite comparable healing on imaging.

For historical perspective, Snow's 2026 randomised trial follow-up of first-generation ACI — the predecessor technique — reported survivorship of 71% and improved function in 75% of patients at 10 years. MACI series sit modestly above that benchmark, consistent with the membrane-based refinement reducing some of the procedural variability of earlier periosteal-patch implantation.

Signals of concern and honest evidence gaps

Not all signals from the 10-year data are reassuring.

The most striking comes from a small New Zealand prospective series — 15 patients enrolled, with only 4 reaching 10-year follow-up. MRI fill averaged 90% at 2 years, dropped to 72% at 5 years, and fell to 49% at 10 years. All four decade-point patients had bone oedema and bone cysts; 73% of biopsies showed fibrocartilage rather than hyaline-like tissue, and 80% of grafts were softer than the surrounding cartilage at re-arthroscopy. The critical confounder is that every patient in this cohort had undergone microfracture before MACI implantation. The series is too small to generalise, but its findings are consistent with a broader pattern: prior marrow stimulation appears to compromise subsequent cell-based repair, likely by altering the subchondral bone environment. For a patient with a previous microfracture, this is the specific question worth raising explicitly at specialist assessment.

A separate structural risk is intralesional bony overgrowth (ILBO) — bony regrowth beneath the graft surface. At 5 years, ILBO was identified on MRI in 44.4% of patients following cartilage repair procedures including MACI, and was associated with higher T2 signal values in the opposing cartilage, suggesting a degeneration risk at that surface over time. At the 5-year mark it had not translated into measurable differences in clinical scores, but it warrants longer follow-up.

Biomechanical recovery appears to lag behind clinical scores: some recent gait analysis data from spheroid-MACI cohorts suggest kinematic recovery may remain incomplete even when pain and function scores have normalised — a consideration specifically for patients whose goal is return to impact sport rather than pain relief alone.

Patellofemoral MACI at 10 years remains understudied. The long-term cohorts are dominated by tibiofemoral cases, and patients with patellar or trochlear defects face a genuine evidence gap that structural MRI equivalence between sites does not resolve.

Getting an assessment in London

Ten years of evidence shifts the consultation conversation from 'does MACI work?' to more specific questions: where exactly is the defect, has the subchondral bone been disturbed before, and what does that mean for this particular patient's outcome trajectory? The long-term data support cautious optimism for tibiofemoral defects in appropriate candidates — and genuine uncertainty for patellofemoral cases and anyone with a prior microfracture on record.

Arriving at assessment with full MRI imaging and a detailed surgical history, particularly any record of previous marrow stimulation, gives a specialist the material needed to work through those distinctions accurately. A MACI assessment involves defect sizing, subchondral evaluation, and a structured comparison with single-stage alternatives; it is not a single-appointment conclusion.

Professor Paul Y. F. Lee manages complex focal cartilage cases at the London Cartilage Clinic on Harley Street; consultations can be arranged via londoncartilage.com.

1. [1] Comparison of Three Different Techniques for the Treatment of Cartilage Lesions — MACI vs AMIC and Arthroscopic Minced Cartilage — 2-Year Follow-Up. (2025). https://doi.org/10.3390/jcm14072194 https://doi.org/10.3390/jcm14072194
2. [2] Minimum 10-Year Outcomes of Matrix-Induced Autologous Chondrocyte Implantation in the Knee. (2024). https://doi.org/10.1177/03635465231205309 https://doi.org/10.1177/03635465231205309
3. [3] 10-Year Prospective Clinical and Radiological Evaluation After Matrix-Induced Autologous Chondrocyte Implantation and Comparison of Tibiofemoral and Patellofemoral Graft Outcomes. (2024). https://doi.org/10.1177/03635465241227969 https://doi.org/10.1177/03635465241227969
4. [4] Consensus on Rehabilitation Guidelines among Orthopedic Surgeons in the United States following MACI for Knee Cartilage Lesions. (2020). https://doi.org/10.1177/1947603520968876 https://doi.org/10.1177/1947603520968876
5. [5] A Prospective Outcome, MRI and Biopsy Study of MACI Cartilage Transplantation. (2017). https://doi.org/10.1177/2325967117S00186 https://doi.org/10.1177/2325967117S00186
6. [6] Long-Term Impact of Intralesional Bony Overgrowth on Opposing Cartilage Integrity: Five-Year Results Following Cartilage Repair. (2025). https://doi.org/10.1177/19476035251335008 https://doi.org/10.1177/19476035251335008