Skip to the line map
Before the Needle

Can Stem Cell Therapy Regrow Cartilage in Arthritic Knees?

  • Updated November 14, 2026
  • Stop 1 of 6 on the Joints, tendons and spine line

Not on current evidence. Randomized trials of cell injections for knee osteoarthritis have reported mixed pain and function results, but they have not shown consistent regrowth of durable joint cartilage. A 2026 meta-analysis of 28 randomized trials found no significant difference on one MRI structural score, and the American College of Rheumatology’s 2026 update strongly recommends against stem cell injections for knee and hip osteoarthritis. Feeling better and growing new cartilage are different outcomes, and no cell injection is FDA-approved for knee osteoarthritis.

What does cartilage regrowth mean in knee osteoarthritis?

The smooth surface covering the ends of the bones in the knee is hyaline cartilage. In osteoarthritis it thins, frays and develops defects. “Regrowth” would mean restoring that tissue: more cartilage thickness or volume, of a quality close to the original, that lasts and supports function over years.

Repair is harder than it sounds. When cartilage does fill in after an injury or procedure, it can be fibrocartilage, a tissue with a different composition and organization that may not provide the same function or durability as the hyaline cartilage of a normal joint surface. So a claim of “regrowth” needs three things checked: whether there is more tissue, what kind of tissue it is, and whether it lasts.

What kinds of cells have been tested?

Trials have used several different preparations, and they are not interchangeable:

  • Culture-expanded mesenchymal stromal cells from bone marrow, fat or umbilical cord tissue, grown in a laboratory to larger numbers before injection
  • Bone marrow aspirate concentrate (BMAC), a same-day concentrate of your own marrow containing a mixture of cells, explained in what BMAC is and what it contains
  • Stromal vascular fraction (SVF), a mixed cell preparation processed from fat tissue

A result for one of these does not transfer to the others. The naming problem behind “stem” versus “stromal” is explained in whether mesenchymal stem cells and mesenchymal stromal cells are the same thing.

What have human randomized trials measured?

Two very different families of outcome appear in these trials:

What is measured Examples What a positive result shows What it does not show
Symptoms and function Pain scales, WOMAC, KOOS People reported less pain or better function at a time point That cartilage changed
Joint structure on MRI Cartilage thickness or volume, T2 mapping, whole-organ scores such as WORMS A measurable imaging change on that specific score That the tissue is hyaline cartilage, that it lasts, or that symptoms improved
Tissue examined directly Biopsy or second-look arthroscopy What the tissue looks like under examination That the finding applies to all patients or products

When structure is measured, trials use different MRI methods and scores, which makes pooling difficult and is one reason a single pooled structural verdict is hard to reach.

Why does feeling better not prove cartilage has regrown?

Knee pain in osteoarthritis comes from several sources, including inflammation of the joint lining and changes in the bone beneath the cartilage. Cartilage itself has no nerves. An injection can reduce pain through effects on inflammation, through the act of injection, or through expectation, without any change to cartilage.

In placebo-controlled trials, people in the placebo group can improve too. That is why a trial without a control group cannot tell you whether a cell injection did anything beyond what the injection process or time would have done.

What do recent reviews say about structural outcomes?

A 2026 meta-analysis of 28 randomized controlled trials of mesenchymal cell-based therapy for knee osteoarthritis reports, in its abstract, a non-significant result for MRI-based WORMS scores. WORMS is a whole-organ MRI score that rates cartilage and other joint structures. That finding is specific to that score; it does not mean every MRI measure in every trial was null. What it does mean is that pooled randomized evidence on this structural measure did not show a difference.

A separate 2026 narrative review of MRI and cartilage findings summarizes structural results across studies. Because it is a narrative review rather than a systematic meta-analysis, it maps what has been reported rather than providing a pooled estimate, and it reports limited structural findings.

Guideline bodies have weighed the full picture:

  • The American College of Rheumatology’s 2026 recommendation summary strongly recommends against stem cell injections in knee and hip osteoarthritis, with moderate certainty of evidence. That recommendation stands even though some individual studies report symptom improvement.
  • The <a href="https://www.healthquality.va.gov/HEALTHQUALITY/guidelines/CD/OA/Osteoarthritis-CPG2026-Guidelinefinal_20260618.pdf”>VA/DoD clinical practice guideline for hip and knee osteoarthritis, Version 3.0 (May 2026) takes a different formal position from the ACR. Its Recommendation 22 finds insufficient evidence to recommend for or against intra-articular stem cell injections (mesenchymal, adipose-derived and bone marrow-derived) and micro-fragmented adipose tissue; the strength and direction is “Neither for nor against.” That is a statement that the evidence is inadequate, not an endorsement.
  • The VA/DoD 2026 patient companion, Orthobiologics at a Glance, says studies have not consistently shown lasting improvements in pain, function or joint health for stem cell injections, and that bone marrow aspirate concentrate has not consistently shown meaningful improvements compared with standard treatments.

Across these sources, the common thread is inconsistency: some trials report improvement, others do not, some follow-up periods are short, products differ, and structural results do not add up to demonstrated regeneration.

Which evidence-based knee-OA care options provide context?

The same 2026 ACR summary strongly recommends exercise for knee osteoarthritis, with moderate certainty. Weight management, activity modification and other clinician-guided care are part of standard management. If you are comparing injection types for symptom relief, the comparison of PRP and hyaluronic acid injections for knee osteoarthritis covers that separate question. Knee pain that began with a twist, or comes with catching or locking, may point to a meniscus problem rather than osteoarthritis alone; recovery for that is covered in how long a meniscus tear takes to heal without surgery.

Pain, MRI and cartilage: keeping the three apart

Does less knee pain mean new cartilage? No. Pain can improve for reasons unrelated to cartilage, and cartilage itself does not contain pain nerves.

Can MRI show whether cartilage increased? MRI can measure cartilage thickness, volume and some quality markers, but the methods vary, small changes can fall within measurement error, and an MRI change does not by itself confirm that the new tissue is durable hyaline cartilage.

Questions worth taking to an orthopedic specialist: Which exact product is being offered, and in which randomized trials was it tested? Did those trials measure cartilage structure, and how? What did the control group show? Is the product FDA-approved for knee osteoarthritis, or offered under an IND in a registered study?

Every stop on the Joints, tendons and spine line

  1. Can Stem Cell Therapy Regrow Cartilage in Arthritic Knees?
  2. How Long Does a Meniscus Tear Take to Heal Without Surgery?
  3. Frozen Shoulder vs. Rotator Cuff Tear: Symptoms and Key Differences
  4. Tendinitis vs. Tendinosis: Why the Difference Matters
  5. Spinal Stenosis vs. Herniated Disc: How to Tell the Difference
  6. Can Stem Cell Injections Help Degenerative Disc Disease? What Studies Show

Reader comments

Leave a note on this stop

Your email address will not be published. Required fields are marked *