Which Cartilage Repair Procedure Does My Knee Need?
The right cartilage procedure depends on lesion size, whether bone is involved, containment, alignment, age, and whether arthritis is already present — not on which implant is newest.
Overview
The right cartilage procedure isn't chosen from a menu — it's determined by the defect itself. Dr. Raffo weighs lesion size, bone involvement, containment, alignment, age, and whether early arthritis is present, and those factors point toward microfracture, MACI, osteochondral allograft, or Agili-C. This page lays out that decision framework directly.
Who this is for
Patients with a symptomatic cartilage defect of the knee — pain, swelling, or catching localized to one area of the joint, usually confirmed on MRI — are candidates for some form of cartilage restoration. Good candidates are typically active adults who want to preserve the native joint rather than move straight to replacement. Cartilage restoration is not for patients with widespread, multi-compartment arthritis; once the whole joint surface is worn rather than one focal defect, restoring a single patch of cartilage will not relieve symptoms, and a different conversation — including robotic total knee replacement — becomes appropriate.
How I approach it
I start every cartilage evaluation the same way: what is the lesion actually made of, and what does the rest of the knee look like around it. A 2 cm² contained chondral defect in a 25-year-old with neutral alignment is a completely different problem than an 18 cm² osteochondritis dissecans lesion with bone loss in the same knee, even though both get called "cartilage damage" by patients. I don't have a favorite implant — I have a framework, and the lesion tells me which technique it wants.
I also don't treat the defect in isolation. If a knee is malaligned into valgus, or the meniscus is deficient, or the patella is maltracking, fixing the cartilage without fixing the environment around it sets the graft up to fail regardless of which technique I use. That's why alignment correction, meniscal allograft, and tibial tubercle osteotomy come up as often as the cartilage procedure itself in my consults.
The operation
Cartilage restoration begins with arthroscopic evaluation to confirm lesion size, depth, and containment, often combined with the primary imaging findings from MRI. From there the technique depends on the decision points below: microfracture and MACI are performed through a mix of arthroscopic and open exposure; osteochondral allograft transplantation involves press-fitting a size-matched cadaveric bone-and-cartilage plug; and Agili-C is implanted as an off-the-shelf scaffold in a single stage. When the knee is malaligned or the meniscus is deficient, a concomitant procedure — high tibial osteotomy, tibial tubercle osteotomy, or meniscal allograft transplantation — is planned alongside the cartilage work rather than deferred.
Where Surgeons Disagree
How large does a defect need to be before microfracture is the wrong answer?
I move away from microfracture once a defect reaches the size ranges studied in the major cell- and graft-based trials, rather than defaulting to it as a first-line option for anyone with a hole in their cartilage.
Microfracture's long-term track record is not reassuring. A 12-year follow-up of 110 patients found 43 required additional surgery, including 7 knee replacements, and 50 had a poor long-term outcome by strict criteria (Solheim et al., Knee Surg Sports Traumatol Arthrosc 2016). A systematic review of ≥10-year microfracture outcomes found radiographic arthritis progression in 40–48% of patients and return to sport of only 17.2–20% (Gopinatth et al., J Exp Orthop 2024). Head-to-head against osteochondral autograft, microfracture failed in 66% of knees versus 51% for OAT, at a mean time to failure of 4.0 years versus 8.4 years (Solheim, Cartilage). The pivotal MACI trial enrolled only defects 3 cm² and larger, and that is the size range where MACI's advantage over microfracture was demonstrated and sustained to 5 years (Saris et al. 2014; Brittberg et al. 2018).
Where I’d be talked out of it
In a small, contained defect under roughly 2 cm² in an older or lower-demand patient, microfracture is still a reasonable, low-morbidity first step, and I'll use it there rather than push a two-stage cell therapy on someone who doesn't need it.
Does bone involvement change the procedure, not just the cartilage layer?
Once the subchondral bone itself is compromised — an OCD lesion, a large cystic defect, prior failed surgery with bone loss — I move to osteochondral allograft rather than a cartilage-only technique.
MACI restores the cartilage surface and is labeled "with or without bone involvement," but in a real-world series of over 5,000 knees, bony involvement was only addressed in 4.5% of cases — it is not what the implant is built around (Milliron/Flanigan, Cartilage 2025). OCA replaces cartilage and subchondral bone together as a single mature unit, which is why it is the standard for OCD: in a dedicated OCD series, the mean allograft area was 7.3 cm², far above the typical MACI lesion, and survivorship in that population was 95% at 5 years and 93% at 10 years (Sadr et al. 2016). Agili-C's FDA label permits bony defect depth up to 8 mm, which covers shallow bone involvement but not the deeper, larger lesions OCA is built for (FDA SSED, P210034).
Where I’d be talked out of it
If the bone involvement is genuinely shallow and the lesion otherwise fits the Agili-C label — 1 to 7 cm², depth under 8 mm — I'll consider the single-stage scaffold instead of committing a patient to an allograft procedure with its longer recovery and donor-tissue logistics.
Is early arthritis a reason to abandon cartilage restoration altogether?
Mild-to-moderate arthritis is not automatically a reason to skip cartilage restoration in favor of arthroplasty — but it does narrow the options to essentially one.
Neither the MACI label nor the OCA literature addresses use in a knee with radiographic osteoarthritis. Agili-C's FDA indication is explicitly for lesions "without severe osteoarthritis (Kellgren-Lawrence grade 0-3)" — KL grade 4 is excluded, but grades 0 through 3 are on-label (FDA PMA P210034). In the pivotal trial, patients with mild-to-moderate OA at baseline (45.5% of the Agili-C arm) still showed treatment failure of only 5.3% versus 27.8% for standard-of-care surgery in that same subgroup, and the FDA states superiority was confirmed across all subgroups (FDA SSED, P210034). Full detail is on the Agili-C page.
Where I’d be talked out of it
Once arthritis is severe — KL grade 4, or a joint that is bone-on-bone rather than focally worn — no cartilage restoration technique is appropriate, Agili-C included, and I'll have the robotic total knee replacement conversation instead.
Should I fix the alignment or the meniscus at the same time as the cartilage?
Yes, in the patients where the data supports it — I don't implant a cartilage graft into a knee I know is going to overload it.
In a study of 168 patients undergoing ACI or OCA with or without high tibial osteotomy, neutral mechanical alignment produced significantly longer graft survival than slight valgus for ACI grafts specifically (P = .003 and P = .05), and ACI failures were significantly more likely to have valgus alignment (P = .002) (Ackermann et al., Arthroscopy 2020). The same study found no significant alignment effect in the OCA group. On the patellofemoral side, a national database of over 60,000 patients found concomitant tibial tubercle osteotomy was performed in 22% of patellofemoral cartilage restorative procedures — and in 37% of ACI cases specifically — versus only 4.1% of palliative procedures (Markes et al., Arthrosc Sports Med Rehabil 2025). A meniscal allograft series found 34% of patients needed a concomitant OCA at the time of meniscal transplant, underscoring how often these procedures travel together (Wagner et al. 2023).
Where I’d be talked out of it
For a well-aligned knee with a stable meniscus and no patellar maltracking, adding an osteotomy or meniscal allograft the patient doesn't need only adds recovery time and risk, and I won't do it just because the literature shows it's common in this population overall.
Risks and honest tradeoffs
Cartilage restoration carries procedure-specific risks addressed in full on each technique's own page, but some tradeoffs apply across the whole framework. Failure rates rise with lesion size, bone involvement, and uncorrected alignment, and even the best-performing techniques do not return every patient to full sport. Osteochondral allograft, for example, carries a weighted mean reoperation rate of 30.2% across the literature, even though outright graft failure is lower at 18.2% (Familiari et al. 2018). No cartilage restoration technique — MACI, OCA, or Agili-C — has long-term data proving it prevents eventual conversion to arthroplasty in every patient; each carries a real, quantified rate of progression to knee replacement, detailed on the individual procedure pages.
Recovery and rehabilitation
Recovery timelines differ meaningfully by technique — MACI and OCA both require a period of protected weight-bearing measured in weeks, while Agili-C's published protocols allow a somewhat faster progression given its single-stage, cell-free design. General milestones across the framework:
- 0–6 weeks: Protected weight-bearing, and for two-stage procedures like MACI, the initial biopsy has typically already occurred weeks earlier.
- 6–12 weeks: Progressive weight-bearing and range-of-motion work begins in earnest.
- 3–6 months: Return to low-impact conditioning for most techniques.
- 9–14 months: Return to sport is assessed individually; pooled data across all cartilage procedures show a mean time to return to sport of 9.2 months, though this varies by technique — OAT averaged 6.6 months and ACI averaged 13.1 months in one pooled analysis (Kunze et al. 2025).
0–6 weeks
Protected weight-bearing, and for two-stage procedures like MACI, the initial biopsy has typically already occurred weeks earlier.
6–12 weeks
Progressive weight-bearing and range-of-motion work begins in earnest.
3–6 months
Return to low-impact conditioning for most techniques.
9–14 months
Return to sport is assessed individually.
pooled data across all cartilage procedures show a mean time to return to sport of 9.2 months, though this varies by technique — OAT averaged 6.6 months and ACI averaged 13.1 months in one pooled analysis (Kunze et al. 2025).
Alternatives I considered
For a genuinely small, contained defect in a lower-demand patient, I still use microfracture — it's a single-stage, low-morbidity procedure and remains appropriate when the lesion doesn't warrant a cell- or graft-based technique. For a knee with diffuse, multi-compartment arthritis rather than a focal defect, I don't reach for any cartilage restoration technique at all; that patient is better served by a conversation about robotic total knee replacement, where I can be equally direct about what the data does and doesn't support.
Ready to be seen?
Appointments are booked through Maryland Orthopedic Specialists, where Dr. Raffo practices.
Frequently Asked Questions
Clinical References
- Saris D, et al. Matrix-applied characterized autologous cultured chondrocytes versus microfracture: two-year follow-up. Am J Sports Med. 2014;42(6):1384-1394.
- Brittberg M, et al. Matrix-applied characterized autologous cultured chondrocytes versus microfracture: five-year follow-up. Am J Sports Med. 2018;46(6):1343-1351.
- Familiari F, et al. Clinical outcomes and failure rates of osteochondral allograft transplantation in the knee. Am J Sports Med. 2018;46(14):3541-3549.
- Sadr KN, et al. Osteochondral allograft transplantation in patients with osteochondritis dissecans of the knee. Am J Sports Med. 2016;44(11):2870-2875.
- US Food and Drug Administration. Premarket Approval P210034 — Agili-C.
- US Food and Drug Administration. Summary of Safety and Effectiveness Data, P210034 (Agili-C).
- Ackermann J, et al. Decreased graft survival in autologous chondrocyte implantation with concomitant realignment. Arthroscopy. 2020;36(8):2204-2214.
- Markes AR, et al. Concomitant tibial tubercle osteotomy decreases odds of revision after patellofemoral cartilage restoration. Arthrosc Sports Med Rehabil. 2025;7(6):101286.
- Solheim E, et al. Results at 10-14 years after microfracture treatment of articular cartilage defects in the knee. Knee Surg Sports Traumatol Arthrosc. 2016;24(5):1587-1593.
- Gopinatth V, et al. Long-term outcomes of microfracture for chondral defects of the knee: a systematic review. J Exp Orthop. 2024;11:e70060.
- Kunze KN, et al. High rate of return to sport for athletes undergoing articular cartilage restoration procedures of the knee. Am J Sports Med. 2025;53(10):2471-2482.
- Wagner KR, et al. Meniscal allograft transplantation at minimum 10-year follow-up. Am J Sports Med. 2023;51(11):2954-2963.
Related Procedures
MACI (Matrix-Induced Autologous Chondrocyte Implantation)
MACI is a two-stage cartilage repair that uses a patient's own cultured chondrocytes on a collagen membrane, best suited to larger, contained cartilage defects without significant bone loss.
Learn more kneeOsteochondral Allograft Transplantation
Osteochondral allograft transplantation replaces cartilage and the bone beneath it as a single donor unit, used for large lesions, bone loss, and osteochondritis dissecans that cartilage-only techniques can't address.
Learn more kneeCartiHeal Agili-C Implant
Agili-C is a single-stage, off-the-shelf cartilage scaffold and the only FDA-approved cartilage implant labeled for use in knees with mild-to-moderate arthritis, though long-term data beyond 2 years remains limited.
Learn more kneeRobotic-Assisted Total Knee Replacement
Robotic assistance improves alignment accuracy in total knee replacement, but the honest evidence shows no meaningful difference in patient-reported outcomes or revision rates. Dr. Raffo explains why he still uses it.
Learn more