Are Stem Cell Injections Safe? What the Side-Effect Research Shows

Most people who ask us about stem cell therapy ask two questions. The first is whether it will work. The second, usually asked a little more quietly, is whether it could hurt them. This article is about the second question.

It is a fair thing to worry about. Stem cell treatments have a mixed public reputation, and some of the stories behind that reputation are real. But "stem cells" covers very different things, delivered in very different ways, by providers who range from regulated laboratories to people with no oversight at all. The safety record depends heavily on which of those you are talking about. Below we set out what the published safety data actually show, where the serious harms have come from, and which side effects you should realistically expect.

The quick answer

For cells given into a vein: two systematic reviews from the same Canadian research group, published eight years apart, found no link between MSC infusions and infection, organ complications, death, or cancer in controlled trials. The one consistent signal was a short-lived fever.[1][2]

For cells injected into a joint: a systematic review of culture-expanded cells in joints found that increased pain and swelling was the only side effect attributed to the cells themselves; most other problems related to the bone marrow harvest.[3] A 2024 meta-analysis of knee trials confirmed that injected MSCs may increase the risk of post-injection knee pain and swelling.[4]

Where the serious harms come from: published reports of blindness and abnormal tissue growth involved unproven treatments at clinics operating outside regulated clinical pathways.[5][6][7] Who makes the cells, how they are tested, and where they are injected matter more than the word "stem cell" on the brochure.

Why "is it safe?" has more than one answer

Asking whether stem cell therapy is safe is a bit like asking whether surgery is safe. It depends on the procedure. Three things change the risk profile considerably:

• The product. A same-day concentrate of your own bone marrow, a fat-derived preparation, and a laboratory-grown, counted dose of donor umbilical cord MSCs are different products with different risks. Our guide to what happens to cells in the lab explains how a characterized, release-tested product differs from an undefined one.
• The route. An intravenous drip, an injection into a knee, and an injection into the eye or spinal fluid carry very different consequences if something goes wrong.
• The oversight. Whether the laboratory and clinic answer to a regulator, test for contamination, and document what goes into each vial. We explain how this works in Colombia in our article on INVIMA and stem cell therapy regulation in Colombia.

What the trial data show for intravenous MSCs

The best-known safety analyses come from the Ottawa Hospital Research Institute and the Canadian Critical Care Trials Group, who reviewed every prospective trial they could find in which MSCs were given into a vein or artery.

The first review, published in PLoS One in 2012 by Lalu and colleagues, covered 36 studies with 1,012 participants across a range of conditions and healthy volunteers. Pooling the eight randomized trials, the authors found no association between MSCs and acute infusion reactions, organ complications, infection, death, or malignancy. There was a significant association with transient fever. They concluded MSC therapy appeared safe, while calling for larger trials with more rigorous adverse-event reporting.[1]

The update, published in EClinicalMedicine in 2020 by Thompson and colleagues, was considerably larger: 55 randomized controlled trials and 2,696 patients. MSCs were again associated with an increased risk of fever (about two and a half times the control rate), but not with other infusion reactions, infection, blood clots, death, or malignancy. No included trial had been stopped early for safety reasons.[2]

Two caveats are worth stating plainly. First, these reviews only included cells given into the bloodstream under trial conditions; they say nothing about products made without quality control. Second, "no detected association with malignancy" in trials of a few years' follow-up is reassuring, but it is not the same as decades of data. If you are considering an infusion, our article on IV stem cell therapy covers what the process involves.

What the trial data show for joint injections

Joint injections are the most common reason international patients contact us, so this is the evidence most of our readers need.

In 2013, Peeters and colleagues at Erasmus MC in Rotterdam published a systematic review in Osteoarthritis and Cartilage of intra-articular treatment with culture-expanded stem cells. Eight studies met their criteria, covering 844 procedures with an average follow-up of 21 months; all used the patient's own bone marrow-derived MSCs. Four serious adverse events were reported: an infection after the bone marrow aspiration that resolved with antibiotics, a pulmonary embolism two weeks after aspiration judged possibly related, and two tumours — neither at the injection site — judged unrelated. The main procedure-related problems were increased pain and swelling and dehydration after the marrow harvest. Increased pain and swelling was the only adverse event attributed to the cell product itself. The authors concluded that cultured stem cells in joints appear safe, with continued caution.[3]

The more recent picture is consistent. The 2024 meta-analysis by Sadeghirad and colleagues in Osteoarthritis and Cartilage — 16 randomized trials, 807 participants with knee osteoarthritis — found low-certainty evidence that MSC injections may increase the risk of any adverse event and of knee pain and swelling compared with placebo or usual care.[4] The same paper concluded that MSCs probably provide little to no improvement in knee pain, which we discuss in our comparison of stem cells vs PRP for knee arthritis. Safety and effectiveness are separate questions, and a treatment can be reasonably safe without being reliably effective.

Where the serious harms have come from

The frightening cases you may have read about are real, and we think every patient should know them. What they have in common is instructive.

Vision loss after eye injections. In 2017, Kuriyan and colleagues reported in the New England Journal of Medicine on three patients with macular degeneration who received injections of their own fat-derived "stem cells" into the eye at a US clinic. All three developed severe vision loss in both eyes, with complications including retinal detachment and bleeding inside the eye. A year later their vision ranged from 20/200 to no light perception.[5]

Abnormal growth in the spinal cord. A 2016 report in the same journal by Berkowitz and colleagues described a glioproliferative lesion — an abnormal growth of cells — in the spinal cord as a complication of "stem-cell tourism".[7]

A wider pattern. A 2018 review in Stem Cells Translational Medicine by Bauer and colleagues analysed scientific publications, case reports, and media reports of safety incidents linked to unproven stem cell interventions, and examined why clinics offering them have spread internationally — including promises of a cure for severe diseases and patients' willingness to pay large sums for that hope.[6]

Notice what these cases share: products with little or no characterization, delivered into delicate, high-consequence sites, outside a controlled clinical or regulatory framework, often for conditions where there was no evidence of benefit. That is a very different situation from a characterized cell product injected into a knee by a physician who has reviewed your imaging. It does not make joint injections risk-free. It does mean the worst outcomes are, to a large extent, avoidable by choosing carefully.

How to lower your risk

You cannot reduce the risk to zero, but you can stack the odds considerably. Before agreeing to any stem cell procedure, anywhere:

• Ask what the product is. Cell source, whether it is cultured, and the dose — with a certificate of analysis showing viability, identity, and sterility and endotoxin testing for your batch.
• Ask who regulates the lab and the clinic, and ask to see it in writing. In Colombia, that is INVIMA.
• Be very wary of high-risk injection sites. Eye, spinal fluid, and brain injections of unproven products are where the most serious published harms have occurred.
• Be wary of treatments for everything. A clinic that offers the same injection for dozens of unrelated diseases is a warning sign, not a credential.
• Ask what happens if something goes wrong, including after you fly home, and who you contact.
• Tell the clinic your full medical history, including medications such as blood thinners, previous cancers, and any current infection.

Our list of 12 questions to ask a stem cell clinic goes through these in more detail, and you are welcome to ask every one of them of us.

Who may be a reasonable candidate

Safety is part of candidacy. In general, the people for whom an injection carries the most favourable balance of risk and potential benefit are those who:

• Have a musculoskeletal condition, such as mild-to-moderate knee osteoarthritis or a tendon problem, confirmed on recent imaging
• Have already tried standard care — exercise therapy, weight management, simple pain relief — without enough benefit
• Are otherwise in reasonable general health, with no active infection or untreated serious illness
• Accept that the realistic outcome may be modest or no improvement, and that temporary pain and swelling are likely

The conditions we assess and how we decide who to treat are set out on our page about stem cell therapy for joint and orthopedic conditions.

When this may not be right for you

We advise against stem cell treatment — or at least against proceeding until something has changed — in these situations:

• Active infection, anywhere in the body, or a skin infection near the planned injection or harvest site.
• Active cancer, or cancer that is still under treatment. Stem cells are not a cancer treatment, and we do not offer them as one.
• Pregnancy or breastfeeding. There are no adequate safety data.
• Bleeding disorders or anticoagulants that cannot safely be paused, particularly if a marrow or fat harvest is planned.
• Uncontrolled diabetes or a serious uncontrolled illness that makes any procedure riskier.
• End-stage joint damage where surgery is the evidence-based option; see our comparison of stem cells vs knee replacement.
• An expectation of a cure. No stem cell treatment can promise one, and any clinic that does is giving you a reason to walk away.

The bottom line

In controlled trials, characterized MSCs have a reassuring safety record so far. The side effects you are most likely to experience are temporary: pain and swelling after a joint injection, soreness where marrow or fat was taken, or a brief fever after an infusion. Rare complications such as infection are mostly linked to the procedure rather than the cells.

The serious harms in the medical literature are real, but they cluster around unregulated products, injections into the eye and spinal cord, and clinics promising cures. Choosing a regulated provider that can document what is in the syringe — and that will tell you when you are not a good candidate — removes much of that risk. It does not remove all of it, and nobody honest will tell you otherwise. If you want a straight answer for your own situation, send us your history and imaging and we will tell you plainly whether we think the balance of risk and benefit is in your favour.