How Effective Is Stem Cell Therapy for Knees?

Stem cell therapy for knee osteoarthritis consistently shows meaningful improvements in pain and function across clinical trials, but the size of the benefit varies depending on the type of cells used, the severity of the arthritis, and the patient’s age. Multiple meta-analyses of randomized controlled trials report statistically significant pain relief and better joint function scores lasting at least a year after injection. Whether those injected cells genuinely rebuild cartilage or primarily calm inflammation remains an open and more complicated question, and the therapy is still not standard-of-care in most countries.

What Stem Cells Do Inside the Knee

The term “stem cell therapy” in the knee context almost always refers to mesenchymal stem cells, or MSCs. These cells can be harvested from your own bone marrow, abdominal fat, or donated umbilical cord tissue. Once injected into the joint, they release growth factors and anti-inflammatory molecules that reduce swelling and promote tissue repair. They can also, in principle, turn into chondrocytes, the cells that make up cartilage.¹

In practice, researchers now believe the larger share of the benefit comes from the anti-inflammatory and signaling effects rather than from the cells physically becoming new cartilage. MSCs secrete molecules that coax nearby cells into repair mode and dampen the destructive enzymes that chew through cartilage in an osteoarthritic joint. This “paracrine” action is why even cell-free products derived from stem cells are being explored.²

Pain Relief and Functional Improvement

The most consistent finding across trials is that people feel meaningfully better after stem cell injections. In one study tracking patients for about two years, standard osteoarthritis scores improved from roughly 50 points down to about 30, and pain on a 0-to-10 scale dropped from nearly 5 to 2.³ A meta-analysis of randomized trials using fat-derived stem cells found that pain, function, and quality-of-life scores all favored the stem cell group over controls at both six and twelve months.⁴ A second meta-analysis using slightly different inclusion criteria confirmed those findings, reporting significant pain relief at six and twelve months with no increase in serious procedure-related complications compared to control groups.⁵

Patients treated with bone marrow concentrate injections show a similar pattern. A prospective trial comparing bone marrow aspirate concentrate and fat-derived stem cells found that both groups improved significantly on every measured outcome, including knee function scores and pain scales.⁶ Umbilical cord-derived cells follow the same trend: a systematic review found that average pain, function, and cartilage-assessment scores all improved after treatment, with no severe adverse reactions.⁷

Does It Actually Rebuild Cartilage?

This is where the story gets less tidy. Some trials do show structural improvements on MRI. A phase II randomized trial using fat-derived stem cells found a measurable increase in cartilage thickness in specific areas of the tibia, and clinical scores tracked with MRI improvement.⁸ A head-to-head comparison against hyaluronic acid injections was even more striking: in the stem cell group, cartilage lesion volume in the medial femoral condyle shrank by about 50 cubic millimeters over six months, while the hyaluronic acid group saw lesion volume grow by roughly 36 cubic millimeters.⁹

But the radiological picture across the broader literature is inconsistent. A systematic review of Wharton’s jelly-derived MSCs noted that while functional outcomes improved consistently, “radiological outcomes demonstrated mixed results.”¹⁰ In other words, patients feel better more reliably than MRI scans look better. Whether the pain relief depends on actual cartilage regrowth or mainly on reduced inflammation is still debated. One study found that clinical scores at follow-up did correlate with structural MRI improvement, suggesting the two are connected, but that single finding does not settle things.¹¹

How It Compares to PRP and Hyaluronic Acid

Platelet-rich plasma and hyaluronic acid are the most common injectable alternatives. A systematic review comparing MSCs directly against PRP found that stem cells produced better results on pain and function scores, and that adding PRP to the stem cell injection did not significantly improve outcomes beyond stem cells alone.¹² The randomized study mentioned earlier comparing fat-derived stem cells to hyaluronic acid found both groups improved in symptoms, but the stem cell group showed superior and sustained improvements in pain, stiffness, and function scores, plus actual cartilage regeneration on MRI.⁹

These head-to-head results look favorable for stem cells, but they come with caveats. Most comparison trials are small, follow-up periods are short, and the cost gap is enormous. PRP and hyaluronic acid injections typically run a few hundred dollars per session, while stem cell treatments can cost thousands.

Who Responds Best

Disease severity matters more than almost any other factor. The prospective trial comparing bone marrow and fat-derived cells found that patients with moderate arthritis (grade 2 on the Kellgren-Lawrence scale) had significantly better outcomes than those with grade 3 or 4 disease.⁶ A separate meta-analysis concluded that stem cell implantation works better for grade 3 knees than grade 4, and that once more than three-quarters of patients in a study had advanced disease, the overall treatment effect became insignificant.¹³

Age and cartilage damage size are the other key predictors. One study found that patients under 60 with cartilage lesions smaller than about 6 square centimeters had significantly better clinical outcomes, and that these two factors were the strongest independent predictors of a good result.¹⁴ The practical message: if you are younger, your arthritis is still in the early-to-moderate range, and the damaged area is relatively small, stem cell therapy is most likely to help. If you have bone-on-bone arthritis across the entire joint, the evidence for meaningful benefit is weak.

How Long the Benefits Last

Most randomized trials only follow patients for six to twelve months, so long-term data is thin. One randomized trial of fat-derived MSCs showed clinically meaningful pain and functional improvement at twelve months but did not report beyond that.¹⁵ A small case report of three patients tracked for five years found that improvements peaked around six months, then gradually declined, but at five years the treated knees were still functioning better than they had before the injection.¹⁶

The longest data comes from a 15-year randomized study, though it compared two different delivery methods rather than stem cells versus no treatment. In that study, knees that received bone marrow stem cells injected directly under the cartilage surface (subchondral) had a yearly knee replacement rate of about 1.3%, while knees that received a standard intra-articular injection had a rate of about 4.6%. By the 15-year mark, 20% of the subchondral group had undergone knee replacement compared to 70% of the intra-articular group.¹⁷ That is a dramatic difference, but it speaks as much to the importance of delivery technique as to the cells themselves.

Safety

Across the board, stem cell knee injections appear safe in the short to medium term. A systematic review covering nearly 2,000 patients found an overall rate of transient side effects of about 12%, almost all of which were swelling or pain at the injection site that resolved within four weeks. No cases of infection, fat embolism, or other serious complications were reported. Umbilical cord-derived injections had a higher rate of these minor reactions (about half of patients) compared to bone marrow or stromal vascular fraction injections.¹⁸ Long-term safety data beyond a few years is still limited, but no signal of tumor formation, immune reactions, or other delayed serious events has emerged from published trials.

Choosing a Cell Source

Three main sources dominate the field, and each has trade-offs. Bone marrow aspirate concentrate is drawn from your hip bone under local anesthesia, processed, and injected the same day. It contains MSCs along with growth factors and platelets. Fat-derived cells come from a small liposuction procedure, yielding a much higher cell count per milliliter than bone marrow. Both cultured (lab-expanded) and uncultured fat-derived preparations show benefit, though one meta-analysis found that cultured cells reduced symptoms faster and with fewer side effects than the uncultured stromal vascular fraction.¹⁹

Umbilical cord-derived MSCs are the only commonly used allogeneic option, meaning they come from a donor. They avoid a harvesting procedure on the patient, have strong proliferative capacity, and raise no ethical issues since they come from tissue that would otherwise be discarded. A systematic review found that these cells can relieve pain, improve knee function, and potentially delay surgery.²⁰ Multiple injections of Wharton’s jelly MSCs appeared to produce better outcomes than a single injection.¹⁰

A meta-analysis attempting to determine whether lab-expanding cells before injection is necessary concluded that it is not: uncultured autologous MSCs performed comparably, and among uncultured options, fat-derived cells produced the best functional outcomes.²¹ Despite all these comparisons, a thorough review of the mechanistic and clinical literature concluded that no definitive winner has emerged among tissue sources, and the best cell dose and long-term durability remain unresolved.²²

Rehabilitation After the Injection

Getting a stem cell injection is not a “get it and forget it” procedure. The evidence suggests that what you do afterward matters. A meta-analysis found that rehabilitation was a significant modifier of outcomes: studies that included structured rehab programs reported larger improvements in physical function than those that did not.²³

Published rehab protocols typically start with a brief period of reduced loading (a few days to a week), then early return to full weight-bearing, sometimes using a pool or an assisted treadmill. Early, gentle loading helps promote collagen reorganization and tissue healing.²⁴ Strengthening work for the quadriceps and hip muscles is then phased in over weeks. If a clinic injects your knee and sends you home without a rehab plan, that is a red flag.

The Clinic Landscape and What to Watch Out For

Hundreds of clinics in the United States market stem cell knee injections directly to consumers, often at out-of-pocket costs ranging from $3,000 to $10,000 or more per knee. A survey of 317 U.S. centers offering these services found widespread direct-to-consumer advertising despite only low-level evidence supporting many of the specific products used.²⁵ The FDA regulates stem cell products, and many preparations used in clinics operate in a regulatory gray zone. A journal editorial in orthopedic surgery flagged the disconnect between the volume of clinical marketing and the evidence base behind it.²⁶

Some practical things to ask before committing: What exactly is being injected? (A same-day bone marrow concentrate is a very different product from lab-cultured cells or an off-the-shelf “amniotic” vial that may contain few viable cells.) Is there published trial data for this specific preparation? What is the provider’s imaging protocol before and after? And does the quoted price include rehab? The best-run programs offer MRI follow-up and structured physical therapy. The worst simply inject and invoice.

Cost-Effectiveness

Whether stem cell therapy is a good financial deal depends on the alternative. A health-economic analysis compared umbilical cord blood-derived MSCs against microfracture surgery for knee cartilage defects. It found that stem cell therapy cost about $14,400 more than microfracture but produced a quality-adjusted life-year gain of about 0.86, resulting in a cost per QALY of roughly $16,800. At standard willingness-to-pay thresholds, the analysis concluded there was a 99% probability that stem cells were cost-effective from a healthcare payer perspective.²⁷ That analysis was done in a South Korean context and may not translate directly to U.S. pricing, but it is one of the few rigorous economic evaluations of the therapy.

Exosomes and Cell-Free Approaches

If the stem cells’ main trick is the cocktail of signaling molecules they release, then an obvious question follows: can you skip the cells entirely and just deliver the molecules? That is the premise behind exosome therapy. Exosomes are tiny vesicles secreted by stem cells that carry proteins, lipids, and RNA fragments capable of influencing nearby cells. Preclinical studies have shown that MSC-derived exosomes can reduce cartilage damage, suppress inflammatory markers, and promote cartilage matrix formation in animal models of osteoarthritis.²⁸

A network meta-analysis of rat studies found that exosome therapy significantly reduced joint cartilage damage scores and boosted type II collagen production, with exosomes from umbilical cord MSCs showing the strongest anti-inflammatory effect.²⁹ Human trial data is still extremely early, but the theoretical advantages are appealing: exosomes can be standardized, stored, and dosed more precisely than living cells, and they carry less risk of unwanted cell behavior. Researchers are also working on hydrogel scaffolds that can hold stem cells in place and improve their survival and retention within the joint, addressing the reality that many injected cells simply die or wash out of the joint space before they can do much work.³⁰

What Biomarker Research Is Revealing

One of the field’s challenges is figuring out how to measure whether therapy is working at a biological level, beyond asking patients how they feel. Researchers have started analyzing synovial fluid, the lubricant inside the knee joint, before and after stem cell treatment. A case series studying umbilical cord MSCs found significant changes in markers related to cartilage breakdown, blood vessel remodeling, and immune modulation about six weeks after injection.³¹ In lab experiments, the secretome (the full set of molecules released by the cells) produced broader anti-inflammatory and anti-cartilage-degradation effects than the cells themselves, reducing levels of multiple destructive enzymes and inflammatory signals.³² This line of research could eventually let clinicians test whether a patient’s joint is responding at the molecular level, rather than waiting months to assess symptoms or run another MRI.