Regenerative Medicine · Health & Wellness
Stem Cell Injections for Knee Pain: What Clinics Offer and What Trials Show
Knee osteoarthritis is the most common reason people walk into a "regenerative" clinic. The pitch is easy to follow: cartilage is wearing away, stem cells can become cartilage, so an injection of stem cells should rebuild the joint and spare you a knee replacement. Each step sounds reasonable. The trials that have tested the whole chain tell a different story.
This guide covers what is actually in the syringe, what the strongest studies found, where the FDA stands, what has gone wrong for patients and how to tell a real study from a paid procedure. It is general information, not advice about your knee. Your own doctor, ideally one who knows your X-rays and history, is the right person to weigh options with you.
What clinics put in the syringe
"Stem cell injection" covers several different products. Asking which one you are being offered is the first useful question.
| What it is called | What it usually is | How it is obtained |
|---|---|---|
| Bone marrow concentrate (BMAC) | Your own marrow, spun in a centrifuge to concentrate the cell layer | A needle draw from the back of the hip bone, same day |
| "Fat stem cells" or SVF | Your own fat, broken down into a mixed cell pellet called stromal vascular fraction | Liposuction, usually from the belly, same day |
| "Amniotic" or "umbilical cord" product | Processed fluid or tissue from donated birth tissue, bought from a supplier | Thawed from a vial |
| Lab-grown cord tissue cells | Donor mesenchymal stromal cells multiplied in a manufacturing facility | Used in regulated trials |
| Platelet-rich plasma (PRP) | Your own blood, spun to concentrate platelets; no stem cells | A blood draw, same day |
Some clinics now add "exosome" injections to the menu. These are particles released by cultured cells and contain no cells at all.
Two points get lost in the marketing. First, true stem cells are a small minority of what is in a marrow or fat preparation; most of the cells are blood cells, immune cells and other supporting cells. Our explainer on bone marrow and mesenchymal stromal cells covers why many researchers now prefer "stromal cells" to "stem cells" for these. Second, birth-tissue products often contain no living cells. A 2019 study in The American Journal of Sports Medicine found no mesenchymal stem cells in the commercial amniotic fluid products it tested, and a 2021 analysis in the journal Cartilage found no live cells in any of nine amnion, umbilical cord and amniotic fluid products.
What the strongest studies found
Small, uncontrolled reports of knee injections are easy to find, and most describe patients feeling better. Knee pain responds strongly to placebo and to the ritual of a procedure, which is why trials with a control group matter most.
A head-to-head trial of three cell products
The largest US trial so far, called MILES and published in Nature Medicine in 2023, randomized 480 people with knee osteoarthritis at several centers. It compared bone marrow concentrate, fat-derived SVF and lab-grown umbilical cord tissue cells against a standard corticosteroid injection. At one year, none of the three cell products beat the steroid shot for pain. Every group improved by a similar amount, and MRI scans showed no meaningful structural differences. No serious procedure-related events were reported, though bruising at the harvest site was common in the fat group.
Pooled results from controlled trials
A 2024 systematic review in Osteoarthritis and Cartilage pooled 16 randomized trials with 807 participants that compared mesenchymal stem cell injections with placebo or usual care. Its authors concluded that the injections probably provide little to no improvement in pain or physical function. The average pain difference was about half the size that patients typically notice. The same review found that cell injections may increase the risk of adverse events, including knee pain and swelling, though the certainty of that finding was low.
A 2025 meta-analysis in Frontiers in Medicine looked only at trials with an inert placebo. It estimated that most of the improvement people report after these injections comes from contextual effects, meaning the placebo response and the care around the procedure, rather than from the cells.
And PRP?
PRP has many more trials than cell products, and their results conflict. The largest placebo-controlled trial, RESTORE, published in JAMA in 2021, gave 288 adults with mild to moderate knee osteoarthritis three weekly injections of either PRP or salt water. At 12 months, pain scores and cartilage volume were no different between the groups.
What guidelines say
The 2019 guideline from the American College of Rheumatology and the Arthritis Foundation, published in early 2020, recommends against stem cell injections and against PRP for knee and hip osteoarthritis. It does recommend exercise, weight loss for people who are overweight, anti-inflammatory medicines and corticosteroid injections, among other measures.
Approved, unapproved and the gray zone in between
The FDA has approved no stem cell product for knee osteoarthritis. As of September 2026 its list of approved cellular and gene therapy products includes one cell product used in knees, MACI, approved in 2016. It is a surgical implant of a patient's own cartilage cells grown in a lab, for specific cartilage defects. It is not a stem cell injection and not a treatment for arthritis.
The FDA's information page for patients names osteoarthritis and knee pain among the conditions for which no stem cell, SVF, cord blood, amniotic fluid, Wharton's jelly or exosome product has been approved.
So why can clinics sell these at all? US rules split human cell and tissue products into two groups, named after sections of the Public Health Service Act.
- 361 products are regulated only for donor screening and infection control. To qualify, a product must be minimally processed and must do the same basic job in the patient that it did in the donor, among other conditions.
- 351 products are regulated like drugs. They need clinical trials under an investigational new drug application (IND) and FDA approval before they can be sold.
In plain words: a tissue that is cleaned and used for its natural purpose can skip approval. Cells that are broken out of tissue, grown in culture or injected to do a new job generally cannot. Grinding up amniotic membrane and injecting it into a knee to "regrow cartilage" is not the membrane's natural job. Federal appeals courts ruled in 2021 and again in 2024 that fat-derived SVF sold by two clinic businesses was a drug that needed approval. Bone marrow concentrate drawn and returned the same day sits in a grayer area, which is one reason clinics lean on it, but no marrow product has been approved as an arthritis treatment.
Harms on record
Knee injections carry less risk than injections into the eye or spinal fluid, but they are not risk free.
- Infection. In 2018 the CDC reported 12 patients in Texas, Florida and Arizona hospitalized with infections, including joint infections, after injections of contaminated umbilical cord blood products. The CDC report notes that bacteria grew from unopened vials.
- Flares and swelling. Painful swelling after injection is the most common complaint in trials and appeared more often with cells than with placebo in the 2024 review.
- Harvest-site problems. Liposuction and marrow draws add bruising, pain and a small infection risk at a second site.
- Delay and expense. Time spent on repeat injections can push back care with stronger evidence, and the money is rarely refundable.
A 2021 review by the Pew Charitable Trusts collected reports of 360 people who had adverse events after unapproved stem cell interventions between 2004 and 2020, drawn mostly from medical journals, government reports and news coverage. Our guide to the warning signs of unproven stem cell clinics covers more of that record.
Real trial or paid procedure
| A regulated clinical trial | A cash-pay clinic protocol | |
|---|---|---|
| Who pays for the product | Usually the sponsor; charging for it needs FDA authorization | You, often up front |
| FDA paperwork | An IND number for an investigational cell product | Often none, or "FDA registered," which is not approval |
| Comparison group | Placebo, steroid or usual care | None |
| Ethics review | An institutional review board you can name | Rarely mentioned |
| Results | Planned endpoints, reported publicly | Testimonials |
| Who can join | Strict criteria; many people are turned away | Almost anyone who can pay |
A ClinicalTrials.gov listing on its own proves little, because some businesses register patient-funded studies. Our guide to reading a ClinicalTrials.gov listing shows which fields to check.
Questions to bring to the appointment
- Exactly which product is this, where does it come from and how many living cells does each dose contain?
- Is it FDA approved for knee osteoarthritis? If not, what is the IND number?
- Which published, controlled trials support it for a knee like mine?
- What does the full course cost, including imaging, the harvest, repeat shots and follow-up?
- What are the risks, and who treats me and pays if I get an infection?
- Will you send my records to my own doctor?
Write the answers down and take them to your primary care doctor, orthopedic surgeon or rheumatologist before you agree to anything.
A note on cost
Insurers generally do not pay for these injections, so patients usually pay cash. The best-known price survey, published in The Journal of Knee Surgery in 2018, used a simulated patient to contact US clinics. Among the 65 that quoted a price, a single same-day knee injection was reported at about 5,160 dollars on average, with quotes ranging from about 1,150 to 12,000 dollars. Clinics in that survey also claimed success rates averaging about 80 percent, far above what controlled trials have shown. Prices have likely changed since, so ask for a written total.
Options with evidence
None of this means knee pain has to be accepted. The measures with the most support are less glamorous: strengthening and aerobic exercise, often guided by a physical therapist, weight loss where it applies, anti-inflammatory medicines, steroid injections for flares, braces or a cane and, for advanced arthritis that limits daily life, joint replacement, which has decades of outcome data behind it.
If you still want to try a cell product, the safest place is a registered trial with a control group, an IND and no charge for the product. Searching ClinicalTrials.gov for "knee osteoarthritis," with the filters set to recruiting and interventional studies, brings up the studies now enrolling, and the contact listed for each can tell you whether you might qualify.
This explainer is general information, not medical advice. Decisions about treatment, donation or trials belong with a clinician who knows your history.
