One of the most common messages we receive a few weeks after a joint injection is some version of: "It's been three weeks and my knee doesn't feel any different. Did it work?" It is a fair question, and the honest answer is that three weeks is usually too early to tell.

Stem cell injections do not behave like a cortisone shot, which often eases pain within days, or like a painkiller, which works within hours. In the clinical trials that exist, improvement was measured over months, not days. This article walks through what the published studies show about timing: what the first days usually feel like, when trials first recorded meaningful change, how long improvements lasted, and when it is reasonable to decide whether a treatment helped you. Most of the evidence comes from knee osteoarthritis, which is the most studied joint, so that is where we focus.

The honest headline: in the controlled knee trials, improvement was usually measured at 3 to 12 months, not in the first few weeks. Some patients improve, some improve only modestly, and some notice little or no change at all. The first week or two can bring extra soreness rather than relief.

The quick answer

Expect soreness first and any improvement later. Mild pain and swelling at the injection site, and at the collection site if your own cells were harvested, are the most commonly reported side effects and usually settle on their own.[1] In the randomized knee trials, the earliest comparisons against a placebo or another injection were made at around three months, and most trials judged results at six to twelve months.[3][4][5]

That means a fair checkpoint is roughly three months for an early read and six months for a more confident one. How long any benefit lasts beyond a year is genuinely uncertain, and the largest recent review of the randomized trials found that, on average, the improvement in pain over placebo was small.[11]

Why it doesn't work like a painkiller

Mesenchymal stem cells (also called mesenchymal stromal cells, or MSCs) are not thought to work by turning into new cartilage in large amounts. The leading explanation is that they release signalling molecules that calm inflammation in the joint lining and influence how the surrounding tissue behaves. One Toronto trial, for example, found lower markers of cartilage breakdown and less synovial inflammation on MRI at higher cell doses, while cartilage scores on MRI did not change over the year.[6]

Processes like that take time to show up as less pain or easier stairs, which is one reason trials measure outcomes at months rather than days. It also explains why "it hurts less" and "the cartilage grew back" are separate questions. We cover the second one in detail in our article on whether stem cells can regrow cartilage.

A realistic timeline, phase by phase

The table below summarises what patients commonly experience and what the published studies actually measured at each stage. Individual experiences vary widely.

Phase What you may notice What the evidence says
First few days Soreness, stiffness or swelling at the injection site; tenderness where cells were collected Pain and swelling are the most frequently reported side effects in joint cell-therapy studies[1]
Weeks 1 to 6 Often little change. Some people feel better early, which is not necessarily the cells Placebo injections alone relieve osteoarthritis pain noticeably, especially when given by needle[2]
Months 3 to 6 The window in which responders tend to notice less pain or better function Placebo-controlled trials assessed patients at 3 and 6 months and found greater improvement with cells[3][4]
Months 6 to 12 Results are usually judged here Several trials reported improvement that held or grew to 12 months, often more clearly at higher doses[5][6][7]
Beyond 1 year Some keep their improvement; others drift back towards where they started Limited, mostly small and uncontrolled follow-up studies, with mixed durability[8][9]

The first few days: soreness is common

A 2013 systematic review of intra-articular therapy with cultured stem cells, covering 844 procedures, found that increased pain and swelling was the only adverse event reported as related to the cells themselves. Procedure-related events included pain and swelling and, after bone marrow collection, dehydration. Serious events were rare.[1] In practice, this means many patients feel slightly worse for a few days before they feel any better. That is not a sign the treatment has failed. A joint that becomes hot, red and increasingly painful, or pain with fever, is different and needs urgent medical review, because infection, while uncommon, is a known risk of any joint injection. Our article on whether stem cell injections are safe covers the side-effect data in more depth.

Weeks one to six: don't over-read early changes

Some people feel nothing and assume the treatment failed. Others feel much better within days and assume it worked. Neither is a reliable signal yet.

A large meta-analysis of 198 randomized osteoarthritis trials found that placebo treatments relieved pain substantially, and the effect was bigger when the placebo was given by injection or needle.[2] Early relief after any joint injection is real relief, but it does not tell you whether the cells are responsible.

Months three to six: the first real checkpoint

The placebo-controlled trials give the clearest picture of when a difference first appears:

  • Emadedin et al., 2018. 43 patients with knee osteoarthritis received either bone marrow-derived cells or a saline injection in a triple-blind trial. Those given cells had significantly greater improvement in pain, function and painless walking distance than the placebo group, assessed at 3 and 6 months.[3]
  • Lee et al., 2019. 24 patients received either fat-derived cells or saline. The cell group showed a significant improvement in function and pain scores at 6 months; the saline group did not change significantly. On MRI, the cartilage defect did not change in the cell group but enlarged in the control group.[4]

Both were small trials, and both stopped at six months, so they tell us when a difference can appear but not how long it lasts.

Months six to twelve: when results are usually judged

Trials that followed patients for a year suggest that, when benefit occurs, it tends to hold through the first year, and that dose matters:

  • Lamo-Espinosa et al., 2016. 30 patients were randomized to hyaluronic acid alone or hyaluronic acid with a low or high dose of bone marrow-derived cells. Pain scores improved in both cell groups across the year. On the broader function score, improvement in the control and low-dose groups was not sustained beyond 6 months, while the high-dose group was still improved at 12 months.[5]
  • Chahal et al., 2019. Patients with late-stage knee osteoarthritis received one of three doses of bone marrow-derived cells. Pain, symptoms and quality of life improved over 12 months, with the highest dose reaching clinically meaningful improvements on most measures. Cartilage scores on MRI did not change.[6]
  • Matas et al., 2019. 26 patients received hyaluronic acid, a single dose of umbilical cord-derived cells, or two doses six months apart. At 12 months, only the two-dose group had significantly less pain than the hyaluronic acid group, and MRI scores did not differ. Several authors worked for the company that produced the cells.[7]

After the first year: durability is the open question

Longer follow-up exists but is thinner. In a two-year follow-up of 18 patients given fat-derived cells, improvements in pain and function were reported up to two years. However, outcomes tended to slip after the first year in the low- and mid-dose groups and only held steady in the high-dose group, and the authors themselves raised concerns about durability.[8] A 2021 reanalysis of two Spanish pilot trials reported that much of the effect seen at one year remained at two and four years; one author sits on the board of a company that makes cells for clinical use, and two of its employees were co-authors.[9]

Put plainly: some patients maintain their improvement for years, others gradually return towards their starting point, and nobody can yet predict reliably which group you will fall into.

How big is the improvement, on average?

Timing is only half the picture; it is also worth knowing how large the change tends to be. Here the reviews disagree, and the more recent and rigorous one is more cautious.

A 2020 meta-analysis of six randomized trials (203 patients) of culture-expanded cells found a significant improvement in pain at 6 to 12 months, but limited evidence for better function or cartilage repair.[10] A larger 2024 review of 16 randomized trials (807 participants) concluded that, compared with placebo or usual care, the injections probably make little or no difference to pain at 3 to 6 months or at 12 months, may improve physical function at 12 months (low certainty), and may increase the risk of side effects such as knee pain and swelling.[11]

We think patients deserve to see both. Individual trials show that some people improve meaningfully. Pooled across all the trials, the average advantage over placebo is modest and uncertain. That is why we are careful never to promise a result, or a timeline for one.

What can affect how quickly, and whether, you respond

  • Severity of the joint damage. Trials included a range of severities, but results in advanced, bone-on-bone joints are less predictable, and joint replacement has stronger evidence there.
  • Cell dose and type. Several of the trials above saw clearer or longer-lasting effects at higher doses, though the best dose is not settled.[5][8]
  • One injection or more. One trial found a second dose at six months outperformed a single dose at one year.[7] That is one small study, not a rule.
  • Rehabilitation and load. Strength work, weight management and sensible activity changes still matter, with or without an injection.
  • The source of the pain. If much of your pain comes from the back, a torn meniscus that is catching, or nerve irritation, a joint injection may not touch it.

Who may be a reasonable candidate

Based on the published studies, the people most likely to be reasonable candidates for a joint cell injection tend to:

  • Have mild-to-moderate osteoarthritis confirmed on X-ray or MRI, rather than advanced bone-on-bone disease
  • Have tried physiotherapy, weight management and standard medication without enough relief
  • Want to delay or avoid surgery and accept that surgery may still be needed later
  • Can wait three to six months to judge the result, and won't be disappointed by a modest improvement
  • Are in reasonable general health, with no active infection or cancer

When this may not be right for you

A stem cell injection may not be the right choice, or the timing may be wrong, if:

  • You need relief within days or weeks, for example before a trip, an event or a return to physical work. The evidence does not support quick results.
  • Your joint is severely worn, deformed or unstable. A joint replacement often has stronger evidence for advanced disease.
  • You have mechanical symptoms, such as a knee that locks or gives way, which may need a surgical assessment first.
  • You have an active infection, active cancer, are pregnant or breastfeeding, or have a bleeding disorder that makes cell collection or injection unsafe.
  • You would only consider it worthwhile with a guaranteed outcome. No honest clinic can offer one.

How to judge whether it worked

Because early impressions are unreliable, it helps to measure. Before treatment, write down your pain on a 0 to 10 scale, how far you can walk, whether you can manage stairs, and which activities you avoid. Repeat the same notes at six weeks, three months and six months.

If there is little or no change by six months, that is useful information too. It may mean the treatment has not helped you, and the next step is usually a conversation about other options rather than more of the same by default.

How our medical team assesses knees, hips, shoulders and other joints, and what treatment involves in Medellín, is set out on our page on stem cell therapy for knee, shoulder and joint injuries.

The bottom line

Stem cell injections for joints are slow, not instant. The first days often bring soreness, the first weeks are hard to read, and the controlled trials measured meaningful change at roughly three to twelve months. Some patients improve and keep that improvement; others improve modestly or not at all, and how long any benefit lasts beyond a year is still being studied. Going in with that timeline in mind makes it much easier to judge your result fairly.

If you are weighing whether an injection is worth it for your joint, send us your imaging. We will tell you honestly what the evidence suggests for your situation, how long you should expect to wait before judging the result, and when we think another option would serve you better.