Cartilage Repair: What's Approved and What's Emerging
Patients ask regularly about regrowing cartilage, and about cartilage restoration in general, and whether it's an option for them.
Cartilage has no blood supply and doesn't repair itself, which I've written about separately. That's the reason this category of procedure exists at all: each one is an attempt to put something onto a joint surface the body can't rebuild on its own.
There are five procedures in use today and several more in development. Which one applies to a given knee has less to do with what's newest than with four measurements taken off an MRI, so I want to start there, because the rest of this only makes sense once you know what a surgeon is looking at.
What determines the recommendation
Size
Cartilage defects are measured in square centimeters. Under 2 centimeters squared opens one set of options, 2 to 4 centimeters squared opens another, and above 4 square centimeters narrows the field considerably. A defect that measures 1.5 square centimeters and one that measures 5 square centimeters are different clinical problems even though both are single defects in otherwise healthy knees.
Depth, and whether bone is involved
A defect confined to the cartilage layer is a different repair than one that has eroded into the subchondral bone underneath. When bone is involved, the procedure has to restore both tissues, which rules out the cell-based options and points toward the ones that transplant bone along with cartilage.
Location
The femoral condyles and the trochlea behave differently from the back of the kneecap. Patellofemoral defects generally do better with cell-based approaches than with plug transfers, partly because of the contours involved and partly because that surface loads differently.
Alignment
A knee that tracks off-axis loads a repaired surface unevenly, and this is the variable patients are most surprised by. If the mechanical axis is putting excess load through the compartment you're trying to repair, the alignment has to be corrected at the same time or the repair carries the same conditions that produced the defect.
Age and activity level factor in as well, though less decisively than the four above.
Current options
Marrow stimulation, including microfracture
Small holes are made in the bone beneath the defect to reach the marrow below, which opens a path for blood and marrow cells into a region that has no blood supply of its own. A clot forms in the defect and fills in over the following months.
What fills in is fibrocartilage rather than hyaline cartilage, and that distinction determines everything about how this procedure has come to be viewed. Fibrocartilage is the tissue your meniscus is made of, tough and built for compression, and it functions as a patch on a gliding surface rather than as a replacement for one. Published follow-up shows that the repair tissue tends to deteriorate within two to three years, and marrow stimulation alters the subchondral bone plate in ways that can compromise later attempts at repair, including plug integration in an osteochondral transfer and cell survival in a MACI implant.
Microfracture was the standard approach for focal defects for years and remains technically simple and widely available. Its role has narrowed considerably as options that restore hyaline cartilage have become accessible. If you've had it already, tell your surgeon at the first visit, because it changes what's reasonable to attempt next.
Osteochondral autograft transfer
A cylindrical plug containing intact cartilage and the column of bone beneath it is taken from a low-contact area of your own knee and press-fit into the defect. Because the plug is your own tissue and already fully formed, the surface that ends up in the defect is hyaline cartilage from the moment the procedure is finished rather than tissue that has to develop over months.
The limit is donor supply. One or two plugs cover a defect well, and outcomes decline while donor site problems increase once a surgeon needs more than two, which is what puts this procedure at its best under roughly 2 square centimeters. It handles bone involvement well, since it transplants bone and cartilage together in a single step.
Osteochondral allograft transplantation
The same principle using matched donor tissue instead of your own, which removes the donor supply ceiling entirely. A graft consists of viable hyaline cartilage supported by five to eight millimeters of subchondral bone, kept refrigerated while microbiologic and serologic testing is completed, which is why scheduling depends on tissue availability rather than only on the operating room calendar.
This handles large defects, defects in locations where your knee has no good donor site, and cases with significant bone loss. It works for small defects as well, though insurance carriers generally set a 2 square centimeters minimum, so in practice it's used above that threshold.
MACI
A two-stage cell-based procedure. In the first stage a small cartilage sample is taken arthroscopically and sent to a lab, where the chondrocytes are cultured over several weeks and seeded onto a collagen membrane. In the second stage that membrane is implanted into the defect and secured.
The controlled evidence comes from the SUMMIT trial, which compared MACI against microfracture in defects of 3 centimeters squared or larger and found better pain and function scores at both two and five years. In August 2024 the FDA expanded the approval to include arthroscopic delivery for full-thickness knee defects up to 4 centimeters squared, so defects in that range no longer require opening the knee.
MACI needs intact bone underneath since the membrane relies on a stable base, which is what makes it the preferred approach for patellofemoral defects and the wrong one when the subchondral bone is compromised.
Aragonite scaffold implants
A cell-free implant derived from purified coral exoskeleton, structured in two layers, with a denser lower portion that integrates with subchondral bone and a porous upper portion that recruits cells from the surrounding joint. New tissue grows into the scaffold while the implant gradually dissolves.
The FDA granted premarket approval in March 2022 for knee joint surface lesions. What separates it from everything else described here is the indication, which covers patients who also have mild to moderate arthritis in the joint rather than requiring healthy cartilage throughout.
Emerging
Minced cartilage implantation
Cartilage is harvested during surgery, minced into fragments, and implanted into the defect in a single operation rather than the two MACI requires, which removes both the second surgery and the weeks of waiting for cells to culture.
Comparisons against staged cell implantation have been favorable so far. Most of that work is retrospective, the patient numbers are small, and follow-up runs one to two years. That's a meaningful limitation in cartilage specifically, where a repair that holds at eighteen months and fails at six years is functionally a different procedure than one that lasts, and eighteen months is roughly where the current evidence stops.
Gene therapy for knee osteoarthritis
A gene therapy delivering an interleukin-1 receptor antagonist by single intra-articular injection has received Regenerative Medicine Advanced Therapy (RMAT) designation from the FDA, with Phase IIb/III study planned for 2026. Interleukin-1 drives inflammation and cartilage breakdown seen in arthritis, and the approach is designed to suppress that signal on a sustained basis after one injection.
RMAT designation accelerates development and review. It isn't approval, and this is a trial-stage therapy.
Drug-based regeneration
In November 2025 a Stanford team reported in Science that blocking a protein called 15-PGDH regenerated cartilage in aging mice, and that human tissue collected during knee replacement surgery produced new cartilage when exposed to the same treatment. An oral version of the drug has entered early trials for age-related muscle weakness, which is a different condition.
No trial has tested whether this regenerates cartilage in a living human knee, and no dose has been established for that purpose. Some of the coverage has suggested that patients delay treatment now to stay eligible later, and nothing in the study supports that.
Stem cells and exosomes
The FDA has approved no stem cell product for any orthopedic condition and no exosome product for anything. The only approved stem cell products in the United States are blood-forming stem cells derived from umbilical cord blood, approved for disorders affecting blood production, which excludes osteoarthritis, tendonitis, and knee, hip, back, and shoulder pain. The agency has issued repeated warnings about unapproved products derived from human cells or tissues marketed for these conditions.
Platelet-rich plasma is regulated differently, because it's prepared from your own blood during the same visit. We offer it and decide case by case whether it fits.
What to take from this
Every procedure described here, the established ones and the ones still in trials, treats focal defects, meaning damage confined to a defined area with healthy cartilage surrounding it. None of them rebuilds a joint surface that has worn through diffusely, and that distinction matters more than any difference between the individual techniques.
Which procedure fits your knee comes down to the measurements rather than the marketing. A 1.5 square centimeter defect on a femoral condyle with intact bone underneath and a 5 square centimeter defect with bone loss are different problems with different answers, and no article can tell you which one you have.
If you've been told you have cartilage damage and aren't sure what to do about it, bring the imaging to your appointment.
