On this page
- Two very different safety questions
- What trials report
- What trial safety data does not cover
- Procedure risks, which are separate from cell risks
- Where documented harm has concentrated
- The evidence-substitution harm
- Questions that reduce your risk
- What to do if something goes wrong abroad
- Frequently asked questions
- Requesting a medical evaluation
Marketing pages tend to compress this into one reassuring sentence. Stem cell adverse events deserve more than that — not because the treatments are dangerous in trial conditions, but because the harms that have actually happened to patients mostly happened outside trial conditions, and for reasons that are predictable.
Two very different safety questions
The first is what happens in a controlled trial with characterised cells, a defined dose and monitored patients. The answer there has generally been reassuring in the short term.
The second is what happens in a commercial clinic with an unsupervised preparation process, a cell product nobody has characterised, and a patient who was never a trial candidate. The answer there is different.
Conflating the two is how “stem cells are safe” gets said. The trial data does not cover the clinic sector — see what the evidence levels mean.
What trials report
Across the randomised literature, the commonest events are transient and local: pain and swelling at an injection site lasting days, brief fever after an infusion.
The 2025 knee osteoarthritis systematic review assessed safety alongside efficacy — Efficacy and safety of mesenchymal stem cells in knee osteoarthritis — as did the phase III adipose trial — Clinical Efficacy and Safety of the Intra-articular Injection of Autologous Adipose-Derived Mesenchymal Stem Cells.
Safety has also been reported in trials outside orthopaedics — in diabetic kidney disease — Safety and Preliminary Efficacy of Mesenchymal Stromal Cell (ORBCEL-M) Therapy in Diabetic Kidney Disease — and in traumatic brain injury — Mesenchymal Stromal Cell Implants for Chronic Motor Deficits After Traumatic Brain Injury.
What trial safety data does not cover
Trial populations are selected. People with active malignancy, active infection, significant organ failure or poorly controlled comorbidity are generally excluded — so the safety record does not describe what happens when those people are treated anyway.
Follow-up is mostly months. Rare events and late events are precisely what short follow-up misses, and a rare event is not rare to the person it happens to.
Trial products are also characterised, which is the least intuitive point here and the most important. Safety data attaches to a specific product made a specific way in a specific facility. It does not attach to a category of biology, and it does not transfer to a preparation nobody has tested.
And trial products are characterised. A preparation nobody has tested is not covered by safety data for a preparation somebody did. This is the least intuitive point and the most important one: safety data attaches to a specific product made a specific way, not to a category of biology.
Procedure risks, which are separate from cell risks
An intra-articular injection carries a small risk of septic arthritis, which is uncommon and serious.
An autologous harvest is a second procedure with its own risks — bleeding, infection and discomfort at the donor site, whether that is abdominal fat or the pelvic marrow.
An intrathecal injection is more invasive again. Route determines a large part of the risk, and route is often the least discussed part of an offer — see MSC therapy for what each route is actually used for.
Where documented harm has concentrated
Contamination of preparations made outside supervised facilities. Infection following injection. Products given by routes and into sites never studied for them.
Treatments sold for conditions with no supporting evidence at all, where the harm is partly clinical and partly the money and time spent instead of on something that works.
The broader pattern has been examined in the medical tourism literature — Patient care without borders: a systematic review of medical and surgical tourism.
The evidence-substitution harm
Some of the most serious harm in this field is not a complication at all. It is a patient with a progressive condition choosing an unproven infusion over treatment that works.
A Cochrane review of stem cell treatment after heart attack is instructive here — Stem cell treatment for acute myocardial infarction — because Cochrane reviews are designed to be unenthusiastic, and reading one is a useful corrective after reading a clinic page.
In cerebral palsy, a meta-analysis found it necessary to explicitly separate cord blood from mesenchymal stromal cells — distinguishing cord blood infusions from mesenchymal stromal cells — because the two had been conflated in ways that made the evidence look stronger than it was.
Questions that reduce your risk
Most of the documented harms would have been caught by asking a few things beforehand.
- Where were these cells prepared, and under what supervision?
- What release testing was done — sterility, endotoxin, viability?
- Has this route been studied for this condition?
- What happens if I have a reaction, and who manages it?
- What are the exclusion criteria, and have I been assessed against them?
A clinic that has not asked about your malignancy history or active infections has not assessed you — see safety studies.
What to do if something goes wrong abroad
Keep every document you are given: consent forms, product details, batch identifiers, dose, route and dates. These are what any doctor treating you afterwards will need.
Seek local medical care first rather than waiting to get home, particularly for anything febrile after an injection or infusion. A hot, swollen joint after an injection is an emergency wherever you are, and delay costs more than the flight change does.
Tell your own doctor what you had, precisely. “Stem cells” is not enough information for anyone to act on — see the international patients page and the research library.
Expect some awkwardness. Doctors at home are often sceptical of treatment bought abroad, and that scepticism is easier to work with if you arrive with documentation rather than a brand name.
Frequently asked questions
Is stem cell therapy dangerous?
In trials, short-term tolerability has generally been good. The documented harms cluster in the unregulated sector, around contamination and untested routes.
What is the commonest side effect?
Transient pain and swelling at an injection site, or a brief fever after an infusion. Both usually settle within days.
What about long-term risks?
Genuinely not well characterised, because most follow-up is measured in months. That is an honest gap rather than a reassurance.
Can an injection cause infection?
Yes. Septic arthritis after a joint injection is uncommon and serious, and it is a risk of the procedure rather than of the cells.
Is a harvest risky?
It is a second procedure with its own risks at the donor site. That is part of the trade-off against using donated cord cells.
What should I keep after treatment?
Everything — consent forms, product and batch details, dose, route and dates. Your own doctor will need specifics, not the phrase “stem cells”.
Requesting a medical evaluation
Nothing on this page establishes whether any treatment is appropriate for you. That needs your history, your imaging and your current medications read by a clinician. If you would like that review, send your records and we will tell you honestly whether there is anything worth discussing — including when the answer is no.
This page is general information, not medical advice, and does not create a doctor–patient relationship. Regenerative treatments discussed here are in most cases investigational. Discuss any treatment with a clinician who knows your history.