Medicines and rehab. 8 minute read.
GLP-1 medicines and knee arthritis: what changed in 2026
By the PocketPhysio editorial team. Published 2026-10-01, updated 2026-10-02.
Two questions in one search box
Two different people type this into a search box, and they want opposite things. One is already on a GLP-1 medicine and has noticed their knees, hips or hands complaining, and wants to know whether the drug did it. The other has an arthritic knee and has read a headline about weight loss injections helping, and wants to know whether that is real. The sections below take them in that order.
Where this page stops is the same in both cases. We write about exercise and rehabilitation. A dose, a comparison between medicines, a side effect to manage, or a decision to start, pause or stop goes to the prescriber or pharmacist. They hold your full record and we do not.
Aching joints on a GLP-1 medicine
First, tell your prescriber. Any new symptom that began after you started the medicine is theirs to judge against your history. Never stop or change a prescribed medicine on your own. Nothing below is a reason to delay that call.
Then look at what the product information says, because it is checked by regulators and free to read. The NHS medicine pages for semaglutide and tirzepatide, both last reviewed on 15 May 2026, list between them the common side effects as feeling or being sick, constipation, diarrhea, stomach pain, tiredness, itchy or red skin at the injection site and hair loss. Between them the serious lists cover allergic reaction, low blood sugar and inflamed pancreas, with gallbladder problems named for one medicine, and worsening diabetic eye disease and a sudden loss of vision, usually in one eye, for the other. Joint pain and muscle pain are on neither list, and they are not on the MedlinePlus entries for semaglutide or tirzepatide either.
An absent symptom on a leaflet is weak evidence on its own, and both NHS pages say plainly that these are not all the side effects and point you to the leaflet with your medicine. Not finding your symptom on a list is not proof the medicine had nothing to do with it, which is the whole reason the first paragraph of this section sends you to the prescriber.
The leaflets are not the only evidence. A 2026 systematic review pooled the safety data of 43 randomized trials of GLP-1 receptor agonists covering 100,488 participants, and found no difference between the treated groups and the control groups in reported gouty arthritis, rheumatoid arthritis, osteoarthritis, osteoporotic fracture, synovitis or disc protrusion (Cao and colleagues, 2026). Check what it measured before you lean on it. These were adverse events reported during the trials rather than conditions anyone examined for, the authors say so themselves, and joint ache as a symptom was not one of the six headings. What it argues against is a large difference between the groups in those six reported categories.
So your knee is sore and something is making it sore. The two things a physio looks at first in someone whose weight is dropping fast are both about load rather than chemistry, and neither is a research finding about these medicines.
The first is muscle. Rapid weight loss takes fat-free mass with it, and the thigh and hip muscles that protect a knee are part of that. A lighter body asks less of the joint while a weaker thigh asks more of it, so the two changes can cancel out or go the wrong way. Our topic on strength training on GLP-1 medicines sets out the evidence, with why muscle loss happens for the research and a plan for 2 to 3 sessions a week for what to do about it.
The second is that you have started moving again. People who lose a meaningful amount of weight often walk further, climb more stairs and pick up activities they had dropped, without noticing the size of the jump. A joint or a tendon that has been quiet for years registers that, and soreness from a change in activity usually tracks what you did a day or two ago instead of sitting there all the time. Writing down your walking for two weeks is a cheap way to see whether that is what is happening.
Either way, the ache gets looked into, not filed away. A hot, swollen knee with a fever is in the warning list below and is a same-day call, not something to wait out.
What the 2026 osteoarthritis guidance added
The American College of Rheumatology announced an updated guideline for osteoarthritis of the knee, hip and hand on 14 September 2026, replacing the 2019 version. Its summary document, approved by the ACR Board of Directors on 8 September 2026, marks exactly one recommendation as new, and that one is the reason this post exists. The sentence reads: "In people who have knee OA and obesity, we conditionally recommend use of GLP-1 RA to help achieve optimal weight in combination with diet and exercise." It is graded conditional, with moderate certainty of evidence.
Read that sentence slowly, because every clause is doing work. It is conditional, not strong, which in this grading system means the panel judged that reasonable people given the same evidence would choose differently. It is written for people who have knee osteoarthritis and obesity, not for knee osteoarthritis on its own. The target is a weight, not the joint. And the diet and exercise are inside the recommendation: this is not a medicine that replaces them.
The same document grades exercise as a strong recommendation for knee and hip osteoarthritis, at moderate certainty, and weight loss as a strong recommendation at moderate certainty for people who meet criteria for overweight or obesity. So the new line does not displace anything. It adds one more route to a goal that two strong recommendations were already pointing at, and a clinician has to decide with you whether it fits.
What the semaglutide knee osteoarthritis trial measured
The evidence behind that recommendation is a trial called STEP 9, funded by the medicine's manufacturer and published in the New England Journal of Medicine in 2024. It enrolled 407 adults who had obesity and a clinical and X-ray diagnosis of moderate knee osteoarthritis with at least moderate pain, at 61 sites in 11 countries, and assigned them 2 to 1 to once-weekly injectable semaglutide or a placebo injection. Both groups received counseling on physical activity and a reduced-calorie diet for 68 weeks. Average age was 56, average body mass index 40.3, and 81.6% were women.
The results, at week 68. Body weight fell 13.7% with the medicine and 3.2% with placebo. The WOMAC pain score, which runs from 0 to 100 with higher meaning worse and averaged 70.9 in these participants at the start, fell by 41.7 points against 27.5 points. The SF-36 physical function score, where higher means better, rose by 12.0 points against 6.5. All three gaps had a P value below 0.001.
Serious adverse events were similar in the two groups, and a side effect led 6.7% of the semaglutide group and 3.0% of the placebo group to stop the trial regimen permanently, mostly because of gut symptoms. The published abstract reports P values and no confidence intervals, so there are none to give you here.
Now the part the headlines left out. Look at the registered outcome list for this trial. It names body weight, the WOMAC pain score, the stiffness score, the function score, the SF-36 scores, waist circumference, the intensity of the pain and the amount of pain medicine taken. X-rays were taken at the start to confirm the grade of arthritis. There is no MRI, and no cartilage measure anywhere on the list.
So the trial says the medicine produced more weight loss and less reported pain over 68 weeks than placebo did. It does not say anything about the joint surface, because nobody looked.
The placebo group is worth a second look too. Its pain score improved by 27.5 points, roughly two thirds of the improvement in the treated group, on counseling plus a dummy injection. Part of that will be the counseling, part the natural swing of arthritis symptoms over 68 weeks, and part the effect of being in a trial and being asked about your knee every few weeks. Which part is which cannot be separated out, and that is not a criticism of the trial. It is the reason the trial needed a placebo arm at all.
Losing weight without losing the leg
Here is the clinical worry that sits underneath all of this, and it is not about the drug. A knee gets its protection from the muscles that cross it and the control that puts the foot down in the right place. Weight loss helps the first job by reducing the load; it does nothing for the second, and if it costs you quadriceps it can quietly undo some of what it gained.
So the plan does not change. The thigh and hip work carries on whatever the scale is doing: two or three sessions a week, in the amounts the exercise pages set, with your physio adjusting them to your knee. Our knee osteoarthritis program is the staged version of that, starting with lying and sitting exercises for a painful knee and working up to stairs and single-leg strength. The hip osteoarthritis program does the same for a hip. Neither of them mentions a medicine, and neither of them needs to.
What a physio watches for in someone losing weight fast is strength going the wrong way while the pain scores improve, because a thinner person with a weaker leg feels better for a while and then does not. How many chair stands you can do, how the stairs feel and whether you trust yourself stepping off a curb tell you more than a number on a scale.
What goes to your prescriber, and what comes here
Take to your prescriber: whether a medicine is right for you, what dose, what to do about a symptom, how long to stay on it, and whether to stop. Never stop or change a prescribed medicine on your own. That is the only thing this site will say on the subject. Food and protein targets belong with a dietitian.
Bring here: which movements hurt, how far you can walk, what the knee does on stairs, and what you want to get back to. Book a physiotherapy assessment if the knee is not improving after six to eight weeks of regular exercise, if it gives way, if pain is stopping you sleeping, or if getting out of a low chair has become harder than it was before the weight started coming off. Anything in the warning list below is dealt with first, at the speed each line sets.
See a doctor promptly if
- Emergency: if chest pain, pressure or tightness does not ease quickly when you rest, spreads to your arm, neck, jaw, stomach or back, or comes with sweating, feeling sick, feeling light-headed or being short of breath, call emergency services straight away, as this can be a heart attack. If it eases quickly and none of these happen, get medical advice the same day, and do not exercise again until you have been checked.
- Emergency: any sign of a stroke, even if it goes away. That means a face that droops on one side, an arm you cannot lift or keep up, slurred or muddled speech, weakness or numbness on one side of your body or face, sudden blurred vision, double vision or loss of sight, trouble swallowing, sudden dizziness with unsteadiness or falling over, a sudden severe headache, or a sudden fall where your legs give way but you do not black out (a drop attack). Call emergency services straight away.
- Emergency: sudden severe stomach pain together with pain spreading to your back, bloating that will not settle, a fast heartbeat, trouble breathing, or skin around your belly button, waist or upper outer thigh that looks blue or bruised (this can be harder to see on brown or black skin). Call emergency services or go to an emergency department, and do not drive yourself. If that sudden severe pain does not go away or keeps coming back, or comes with a high temperature or chills, without any of the signs above, get an urgent medical appointment or call your health service's urgent advice line the same day.
- Same day: shakiness, sweating, tingling lips, hunger, a pounding heartbeat, blurred vision, feeling weak or confusion during or after a session, if you take insulin or other diabetes medicines. These can be signs of low blood sugar, and the risk of it is higher when insulin or a sulfonylurea is taken alongside one of these medicines. Treat it the way your diabetes team has told you, then contact them or your prescriber the same day. If someone cannot be woken, cannot swallow safely or has a seizure, call emergency services.
- Same day: you cannot keep fluids down, you are being sick or have diarrhea that will not settle, you are peeing less than usual or your pee is dark yellow and strong smelling, you feel unusually tired, you feel dizzy when you stand up, or you are breathing quickly or your heart is racing. Contact your prescriber or get medical advice the same day, as these can be signs of dehydration. Call emergency services or go to an emergency department if your skin, lips or tongue look blue, gray, pale or blotchy (on brown or black skin this can be easier to see on the palms of the hands or soles of the feet), your skin feels cold, you are breathing fast or finding it hard to breathe, you are confused, or you are unusually sleepy or hard to wake.
- Same day: dizziness or feeling faint during or after exercise, without any of the emergency signs above. Stop the exercises and get medical advice the same day. If you faint, it does not settle within a few minutes of sitting still, or you cannot stand or walk steadily, call emergency services.
- Same day: you blacked out or fainted, even for a moment. Get medical advice the same day. Call emergency services instead if the person cannot be woken within 1 minute, has not fully recovered or has trouble speaking or moving, has chest pain or a pounding, fluttering or irregular heartbeat, fainted while exercising or lying down, is shaking or jerking, or was badly hurt.
- Same day: the knee is hot, red and swollen, or you have knee pain with a high temperature or feel hot, cold or shivery. Get medical advice the same day, and go to an emergency department if you feel very unwell. This can be a sign of infection in the joint, which needs treatment quickly.
- Same day: a calf or thigh that is newly swollen, warm, tender, red or darker than usual, has swollen veins that are sore to touch, or has a throbbing or cramping pain that does not fit your injury or condition, or that feels different from normal muscle soreness after exercise. This can be a sign of a blood clot (deep vein thrombosis), especially after surgery, an injury, time in a cast or boot, a long trip or a spell of being much less mobile than usual. Get medical advice the same day. If you are also short of breath or have chest pain, call emergency services.
- Same day: back pain after a minor fall or strain if you are older, have low bone density (osteoporosis) or have taken steroid tablets for a long time. Get medical advice the same day.
- Within a few days: the pain is there at night or at rest and keeps getting worse. See your doctor within a few days. If you have had cancer, now or in the past, or are losing weight without trying, see your doctor sooner, within a day or two, and mention it. This applies even if the pain is not getting worse. If you are being treated for cancer now, contact your cancer team the same day.
Common questions
Can a GLP-1 medicine cause joint pain?
Tell your prescriber about any new symptom that started after you began the medicine, including an aching joint, because they are the only person who can judge it against your history. Never stop or change a prescribed medicine on your own. What the published information says is that joint pain is not on the patient side effect lists: the NHS pages for semaglutide and tirzepatide, both last reviewed 15 May 2026, list between them nausea, vomiting, constipation, diarrhea, stomach pain, tiredness, injection site reactions and hair loss among the common ones, and allergic reaction, low blood sugar and inflamed pancreas among the serious ones, and the MedlinePlus entries for the two medicines do not list joint pain, muscle pain or aching either. Those lists are not the last word, and the NHS says so on both pages: these are not all the side effects, and the full list is in the leaflet that comes with the medicine, so a symptom that is missing from a patient leaflet is not a symptom that cannot happen to you. There is more than absence to go on here: a 2026 systematic review pooled the safety data of 43 randomized trials covering 100,488 participants and found no difference between people taking a GLP-1 receptor agonist and controls in reported gouty arthritis, rheumatoid arthritis, osteoarthritis, osteoporotic fracture, synovitis or disc protrusion (Cao and colleagues, 2026), and because those were adverse events reported during the trials rather than conditions anyone went looking for, with joint ache as a symptom not among the six, the fair reading is that an aching knee needs an explanation of its own rather than being filed under side effects.
Does losing weight help knee pain?
Yes, and it is one of the few things the osteoarthritis guidelines agree on. NICE recommendation 1.3.5 tells clinicians to advise people with osteoarthritis who are living with overweight or obesity that weight loss will improve their quality of life and physical function and reduce pain, to support them to choose a goal, and to explain that any amount of weight loss is likely to help but that losing 10% of body weight is likely to be better than 5%. The 2026 American College of Rheumatology guideline summary strongly recommends weight loss for people with knee or hip osteoarthritis who meet criteria for overweight or obesity, and strongly recommends exercise, both graded strong with moderate certainty of evidence. How the weight comes off is a separate question from whether it helps, and that first question belongs with your doctor.
What did the semaglutide knee osteoarthritis trial find?
STEP 9 was a 68-week double-blind randomized trial at 61 sites in 11 countries, funded by the medicine's manufacturer, and it enrolled 407 adults with obesity and a clinical and X-ray diagnosis of moderate knee osteoarthritis with at least moderate pain, assigned 2 to 1 to once-weekly semaglutide or a placebo injection, with counseling on physical activity and a reduced-calorie diet in both groups. At week 68 body weight was down 13.7% with semaglutide against 3.2% with placebo, and the WOMAC pain score, measured on a 0 to 100 scale where higher is worse and averaging 70.9 at the start, had fallen by 41.7 points against 27.5. The SF-36 physical function score had risen by 12.0 points against 6.5, all three comparisons with P less than 0.001. Serious adverse events were similar in the two groups, and a side effect led 6.7% of the semaglutide group and 3.0% of the placebo group to stop the trial regimen permanently, most often for gut symptoms. Both groups improved on pain, which is what a trial with counseling in both arms should show, and the published abstract gives P values and not confidence intervals, so none are quoted here.
Do these medicines repair knee cartilage?
Nothing in the trial tested that. STEP 9's registered outcome measures are body weight, WOMAC pain, stiffness and function, SF-36 scores, waist circumference, pain intensity and pain medication use. X-rays were used at the start to confirm the grade of arthritis, and there is no MRI or cartilage outcome on the list at all. So the fair statement is that in this trial the medicine reduced weight and reduced reported pain, and nobody looked at the joint surface. Anyone telling you these drugs regrow cartilage has gone past the evidence.
Should I keep doing knee exercises while the weight is coming off?
Yes, and this is the part worth protecting. Some of what leaves the scale is muscle, which is the subject of our pages on strength training on GLP-1 medicines, including why muscle loss happens and a plan for 2 to 3 sessions a week. A lighter body asks less of the knee, and a thigh that has lost strength asks more of it, so the two changes can cancel out. The staged exercises are on our knee osteoarthritis program, with the numbers on each exercise page.
Should I ask my doctor about a GLP-1 medicine for my knee arthritis?
That conversation belongs with your doctor, and this page will not take a position on it, but the 2026 guideline summary gives you the shape of the question. In people who have knee osteoarthritis and obesity, it conditionally recommends use of a GLP-1 receptor agonist to help achieve the best weight for the person, in combination with diet and exercise, graded conditional with moderate certainty of evidence. Conditional means the panel judged that reasonable people given the same evidence would choose differently, so it is a discussion rather than a default, and the recommendation is written for people who have both conditions and alongside diet and exercise, not instead of them. Bring what your knee stops you doing, what you have already tried, your other conditions and your other medicines, and note that nothing on this site recommends, compares, doses, starts or stops a medicine. Never stop or change a prescribed medicine on your own.
References
- American College of Rheumatology. 2026 update of the American College of Rheumatology recommendations for the management of osteoarthritis of the knee, hip, and hand. Guideline summary, approved by the ACR Board of Directors on 8 September 2026. Read 1 October 2026. https://assets.contentstack.io/v3/assets/bltee37abb6b278ab2c/bltb3d12c34020da842/oa-guideline-summary-2026.pdf
- American College of Rheumatology. American College of Rheumatology updates osteoarthritis guidance, reinforcing personalized, comprehensive care. Press release, 14 September 2026. https://rheumatology.org/press-releases/american-college-of-rheumatology-updates-osteoarthritis-guidance-reinforcing-personalized-comprehensive-care
- Bliddal H, Bays H, Czernichow S, et al. Once-weekly semaglutide in persons with obesity and knee osteoarthritis. New England Journal of Medicine. 2024;391(17):1573-1583. doi:10.1056/NEJMoa2403664
- ClinicalTrials.gov. Effect of subcutaneous semaglutide 2.4 mg once-weekly compared to placebo in subjects with obesity and knee osteoarthritis (STEP 9). NCT05064735. Accessed 1 October 2026. https://clinicaltrials.gov/study/NCT05064735
- Cao M, Lin C, Cai X, Lv F, Yang W, Ji L. The association between glucagon-like peptide-1 receptor agonists and reported musculoskeletal adverse events: a systematic review and meta-analysis of randomized controlled trials. Therapeutic Advances in Musculoskeletal Disease. 2026;18:1759720X261428147. doi:10.1177/1759720X261428147
- National Institute for Health and Care Excellence. Osteoarthritis in over 16s: diagnosis and management (NG226). 2022. Recommendation 1.3.5 read 1 October 2026. https://www.nice.org.uk/guidance/ng226
- NHS. Semaglutide: a medicine to manage type 2 diabetes or treat obesity. Page last reviewed 15 May 2026. https://www.nhs.uk/medicines/semaglutide/
- NHS. Tirzepatide: a medicine to manage type 2 diabetes and treat obesity. Page last reviewed 15 May 2026. https://www.nhs.uk/medicines/tirzepatide/
- MedlinePlus. Semaglutide injection. AHFS Patient Medication Information. Last revised 15 May 2026. https://medlineplus.gov/druginfo/meds/a618008.html
- MedlinePlus. Tirzepatide injection. AHFS Patient Medication Information. Last revised 15 February 2026. https://medlineplus.gov/druginfo/meds/a622044.html
Written and checked by the PocketPhysio editorial team. Last updated 2026-10-02.
This page is general education, not a diagnosis or a personal treatment plan. See a physiotherapist or doctor for advice about your own situation.