Massage for Knee Pain: What the Trials Found, Why the US Arthritis Guideline Changed Its Mind in 2026, and When a Sore Knee Needs a Doctor
Does massage help knee pain? For knee osteoarthritis, the best trial says yes for a while: eight weekly hour-long full-body massages beat both gentle hand-resting and usual care at eight weeks, and the edge was gone by a year. In September 2026 the American College of Rheumatology switched massage from 'conditionally recommended against' to 'conditionally recommended', on low-certainty evidence. Here is what was tested, what it does not cover, the knee signs that need a doctor, and what our Hội An spa menu offers.
Does massage help knee pain? If the pain is knee osteoarthritis, yes, for a while, and the main US guideline now agrees. In the largest trial, 222 adults with knee osteoarthritis were given eight weekly one-hour full-body massages, eight weekly sessions of a therapist simply resting their hands on the body, or nothing new. At eight weeks the massage group had improved more than both. By one year the groups no longer differed. In September 2026 the American College of Rheumatology changed its advice on massage for knee and hip osteoarthritis from "conditionally recommended against" to "conditionally recommended", with low certainty of evidence. For other knee pain, a twist, a fall, a swollen or hot joint, nobody has shown massage helps, and some of those knees need a doctor first.
Does massage help knee pain? The short answer
Claim
What was tested
Verdict
Massage eases knee osteoarthritis
222 adults aged 35 and over with X-ray confirmed knee osteoarthritis, eight weekly 60-minute whole-body Swedish massages, against hand-resting "light touch" and usual care (Perlman 2019)
Yes at eight weeks, against both comparisons
The benefit lasts
Same trial, followed to 52 weeks, some of the massage group kept going every other week
Not shown: no difference between groups at one year
More massage is better
125 adults, 30 or 60 minutes, weekly or twice weekly, for eight weeks (Perlman 2012)
Up to an hour a week; beyond that it levelled off
Doctors recommend it
American College of Rheumatology, 2019 then 2026
Against in 2019; a conditional yes in 2026, low certainty
One massage on holiday fixes a sore knee
Never tested
Unknown
Massage helps a twisted, swollen or injured knee
Not what these trials enrolled
Unknown; see a doctor first if the warning signs below apply
What did the knee massage trials actually test?
One research group did most of the work. Adam Perlman and colleagues first gave 68 adults with X-ray confirmed knee osteoarthritis standard Swedish massage, twice a week for four weeks and then weekly for four more, against a group who waited their turn; the massage group improved significantly on the WOMAC score (Archives of Internal Medicine, 2006). A second trial of 125 adults tried four doses over eight weeks, 30 or 60 minutes, once or twice a week, against usual care. On the WOMAC, a 0 to 100 score of knee pain, stiffness and function where lower is better, the 60-minute groups improved by 24.0 points against 6.3 with usual care. The authors picked one hour once a week as the best dose because more time added little (PLoS One, 2012). At 24 weeks, no measure differed from usual care.
Illustrative photo, not our spa. In the knee trials, between 20 and 27.5 minutes of each hour went to the legs, from the feet to the hips, and the rest to the back, neck, head and chest.
The massage was not a knee rub. The trial protocol gave the lower limbs 20 to 27.5 minutes of 55 minutes on the table, "From knee down including lower leg, ankle, and foot. From knee up including hips, pelvis, buttocks, and thigh", and 15 to 24 minutes to the back, head, neck and chest, using standard Swedish strokes (Ali and colleagues, Trials, 2012). It was a whole-body hour with extra time on the legs.
The third trial is the one that matters most, because it had a fair comparison. At four US sites, Duke and Rutgers universities among them, between June 2012 and May 2014, 222 adults aged at least 35 with X-ray confirmed knee osteoarthritis and knee pain of 40 to 90 on a 0 to 100 scale were split three ways: an hour of whole-body Swedish massage once a week for eight weeks; an hour of "light touch", in which the therapist gently placed their hands on the body in a set sequence with no massage strokes; or usual care (Perlman and colleagues, Journal of General Internal Medicine, 2019). At eight weeks, massage beat light touch by 8.16 points on the 0 to 100 WOMAC and usual care by 9.55. The authors applied a published threshold of 7 points as a change that matters to patients, so both differences cleared it, just, while noting that the threshold was validated for change in one person, not for differences between groups. Light touch itself did no better than usual care on the overall score.
Then the advantage faded. After eight weeks, the two treated groups were split again by lot: some carried on every other week and the rest stopped. At 52 weeks, "the omnibus test of any group difference in the change in WOMAC Global from baseline to 52 weeks was not significant": everyone, massage or not, had improved from where they started, and the groups ended in the same place. Side effects were few. Of 135 adverse events, 133 were mild or moderate, and one severe event was judged likely to be caused by the treatment: "post-massage pain in subject's hip and thigh". The most common minor complaint was knee pain after a session, in both the massage and light-touch groups.
Two cautions from the trial itself. Participants could not be blinded, since anyone can tell a massage from a hand resting on them, and the authors note this "may have led to differential dropout": by a year, 34% of the light-touch group had left the study against 11% of the massage group, many "because they did not feel the intervention was beneficial". And every participant was paid in gift certificates for three future massages.
Why did the US arthritis guideline change its mind?
Illustrative photo. Exercise remains the strongest recommendation for knee osteoarthritis in both the 2019 and 2026 US guidelines; massage is a conditional extra.
In 2019 the American College of Rheumatology and the Arthritis Foundation wrote: "Massage therapy is conditionally recommended against in patients with knee and/or hip OA." Their reason: "Studies addressing massage have suffered from high risk of bias, have included small numbers of patients, and have not demonstrated benefit for OA-specific outcomes." They also recorded a dissent: patient members of the panels "felt strongly that massage therapy was beneficial for symptom management" (Kolasinski and colleagues, Arthritis Care & Research, 2020).
The 2026 update, approved by the College's board on 8 September 2026 and posted on 14 September, reverses it: "In people with knee and hip OA, we conditionally recommend massage therapy", certainty of evidence low, marked as a "Change in recommendation direction since 2019 update". The summary does not give its reasons; it says the full paper will be submitted to the College's journals. Two reviews give a sense of the evidence it had to work with. A 2017 review of seven trials with 352 people with arthritis found "low- to moderate-quality evidence that massage therapy is superior to nonactive therapies in reducing pain", and that "It is unclear whether massage therapy is more effective than other forms of treatment" (Nelson and Churilla, American Journal of Physical Medicine & Rehabilitation). A 2026 review of 29 trials found that hands-on therapy added to exercise reduced knee osteoarthritis pain more than exercise alone in the short and medium term, with very low to moderate certainty, but "no significant long-term effects" (da Silva and colleagues, European Journal of Pain).
Treatment for knee osteoarthritis
ACR 2026
Certainty
Exercise
Strongly recommended
Moderate
Weight loss, if overweight
Strongly recommended
Moderate
Tai chi
Conditionally recommended
Very low
Locally applied heat or cold
Conditionally recommended
Low
Massage therapy
Conditionally recommended (was against in 2019)
Low
Acupuncture
Conditionally recommended
Moderate
TENS
Conditionally recommended against
Low
Platelet-rich plasma (PRP) or stem cell injections
Strongly recommended against
Low / moderate
So the honest summary: for knee osteoarthritis, massage has moved from a reason to hold back to a reasonable thing to try, behind exercise and weight loss, which both carry strong recommendations. The UK's NHS says much the same in plainer words: "Exercise is one of the most important treatments for people with osteoarthritis, whatever your age or level of fitness", and lists manual therapy, in which "a physiotherapist uses their hands to stretch, mobilise and massage the body tissues to keep your joints supple and flexible". Tâm Anh General Hospital, in its page on thoái hóa khớp gối (knee osteoarthritis), lists physiotherapy exercises, capsaicin cream, acupuncture and supplements, and does not mention massage at all.
Is it osteoarthritis, or something else?
The trials only tell you about osteoarthritis. The NHS says a doctor may suspect it if "you're aged 45 or older", "you have joint pain that gets worse the more you use your joints", and the stiffness "is not there in the mornings, or lasts less than 30 minutes". Its knee pain page sorts other causes by what happened: pain after "overstretching, overusing or twisting" is usually a sprain or strain; pain "between your kneecap and shin, often caused by repetitive running or jumping" suggests tendonitis; a knee that is "Warm and red, kneeling or bending makes pain and swelling worse" suggests bursitis. None of those was studied in the massage trials, so for them we do not know.
Why does a knee ache on a walking holiday?
A street in Hội An. A holiday of walking and cycling asks more of the knees than most working weeks do.
Mostly because a holiday changes how much you use your knees. The NHS describes osteoarthritis pain as getting "worse the more you use your joints", and long days on foot through the Old Town, or on a bicycle, are exactly that. The NHS first steps for a painful knee are to "put as little weight as possible on your knee – for example, avoid standing for a long time", to use "an ice pack or bag of frozen peas wrapped in a tea towel on your knee for up to 20 minutes every 2 to 3 hours", and painkillers such as paracetamol or ibuprofen gel or tablets; Tâm Anh likewise recommends "Chườm đá", an ice pack, to ease pain and swelling, and rest, especially after exercise or a long walk. A massage that evening may feel good. No trial has measured what a single massage does for a knee, and the trials that worked used eight weekly sessions.
When is a sore knee not a spa matter?
See a doctor, not a massage therapist, if any of these apply. The NHS lists them as reasons to get urgent advice:
Your knee is very painful, or you cannot move it or put any weight on it.
It is badly swollen or has changed shape, for example after a fall or a sudden change of direction.
It locks, gives way or painfully clicks. The NHS adds that painless clicking is normal.
You have a very high temperature, or feel hot, cold or shivery, with redness or heat around the knee. The NHS says this "can be a sign of infection".
The pain is throbbing in one leg, usually the calf or thigh, with swelling, especially after a journey of more than 4 hours by plane, car or train. These are the NHS signs of a DVT, a blood clot in the leg; the NHS says to get urgent medical advice, and to go to an emergency department if you are also short of breath or have chest pain.
The pain has not improved within a few weeks. The NHS says to see a GP.
Planning a trip around this? See dates at our quiet riverside hotel on the Thu Bồn. Check availability →
What our spa menu offers for a sore knee, and what it does not
On our menu
60 min
90 min
The menu's own words
With knee osteoarthritis in mind
Relaxing full-body massage with oil
450,000₫
600,000₫
"slow, flowing oil massage — medium pressure, head to toe"
The closest to the trials' hour: oil, the whole body, the legs included. The menu does not call it Swedish massage
Vietnamese herbal full-body massage
520,000₫
700,000₫
"warm compresses of lemongrass, ginger and local herbs"
Warm compresses; the 2026 guideline conditionally recommends locally applied heat, but nobody has tested herbal compresses for knees
Himalaya hot stone full-body massage
520,000₫
700,000₫
"Warm Himalayan stones glide over tight muscles"
Warmth plus massage; not tested for knees
Thai traditional massage (with or without oil)
570,000₫
770,000₫
"assisted, yoga-like movements"
Bends and stretches the joints; the menu does not say whether knee stretches can be left out, so ask
Vietnam Impression bamboo massage
570,000₫
770,000₫
"firmer than hands, especially good for legs and back"
Firmer than hands, in the menu's words; the trials used hands
Foot massage
450,000₫
600,000₫
"from toes to calves"
Stops at the calves, below the knee
Prices are per person, VAT is included and there is no service charge; treatments starting from 19:00 carry a 20% evening surcharge, so a 450,000₫ hour becomes 540,000₫. At the 26,000₫ rate we use across the site, the oil massage is about US$17 for 60 minutes and US$23 for 90. Prices are from our live spa menu, read on 19 November 2026, and the quoted words are the menu's descriptions, not findings. We are a hotel spa: there is no doctor or physiotherapist on the premises and nothing on the menu is a medical treatment. The tested dose was one hour a week for eight weeks; a holiday gives you one or two, and what that does for a knee has not been measured. On the consultation form, which the online booking leads straight to, you choose the pressure, Gentle, Medium, Strong or Very strong, and write what to avoid: write "knee" and which side. More on the oil hour is on relaxing oil massage in Hội An, and the six whole-body choices side by side on full body massage in Hội An. We have male and female therapists; ask at booking.
What nobody has measured
A single massage for a sore knee. Every trial above used a course of weekly sessions and measured after weeks, not hours.
Massage for knee pain that is not osteoarthritis, such as a sprain, a sore kneecap after running or a swollen bursa. The trials enrolled X-ray confirmed osteoarthritis only.
Herbal compresses, hot stones, bamboo or Thai stretching for knee osteoarthritis. The trials used hands-on Swedish strokes.
The reasons behind the 2026 change. The summary gives the verdict and its certainty; the full paper has not been published yet.
Questions people ask before booking
Should you massage a swollen knee?
Not if it is badly swollen, hot or red, or swelled up after an injury; the NHS lists those as reasons to get urgent advice, and "redness or heat around your knee" with a fever can be a sign of infection. The trials excluded anyone with a significant knee injury in the past six months, recent steroid injections or a rash or open wound over the knee. For the mild swelling of osteoarthritis, the trials did massage the legs, but tell the therapist and ask them to keep off the joint if it hurts.
Can I get a massage after a knee replacement?
Ask your surgeon first. The largest trial excluded people with a replacement in the knee being studied, so there is no evidence either way from it. If your surgeon is happy, write the side and the date of the operation on our consultation form.
Do you offer physiotherapy or sports massage for knees?
No. Nothing on our menu is called physiotherapy, sports massage or deep tissue massage, and there is no doctor or physiotherapist on the premises. The closest to the tested treatment is the relaxing full-body massage with oil, 450,000₫ (about US$17) for 60 minutes.
How to book
Nghê Prana Riverside Hotel & Spa, Hẻm 384 Nguyễn Tri Phương, Cẩm Nam, Hội An, on the Thu Bồn, 3.2 km from the Old Town. The spa is open from 9:00 to 21:00 every day and takes visitors as well as hotel guests. Book on the online form or by phone or WhatsApp on +84 896 687 961; an hour's notice is enough. Coming as two? The couple spa page shows the private room for two. Free pickup to the spa from anywhere within 5 km of us in a straight line, which takes in the Old Town, hotel guest or not: send your location to +84 896 687 961 on WhatsApp or Zalo and we reply within 15 minutes (9:00–21:00).
How we checked: on 19 November 2026 we read the American College of Rheumatology's 2026 osteoarthritis guideline summary (PDF, approved 8 September 2026, posted 14 September 2026); the full texts of the 2019 ACR/Arthritis Foundation guideline (Kolasinski 2020, Arthritis Care & Research), Perlman 2019 (Journal of General Internal Medicine), Perlman 2012 (PLoS One) and Ali 2012 (Trials); the abstracts of Perlman 2006 (Archives of Internal Medicine), Nelson and Churilla 2017 (American Journal of Physical Medicine & Rehabilitation) and da Silva 2026 (European Journal of Pain); the NHS pages on knee pain (reviewed 21 December 2023), osteoarthritis and its treatment (reviewed 20 March 2023) and DVT (reviewed 30 April 2026); and Tâm Anh General Hospital's page on knee osteoarthritis (updated 31 May 2025). NICE's osteoarthritis guideline refused automated reading, so we do not cite it. We also read our live spa menu. We are a hotel spa, not a clinic; nothing here is medical advice.
American College of Rheumatology (2026). 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 8 September 2026, posted 14 September 2026; massage therapy p. 2; PDF read 19 November 2026). American College of Rheumatology. View source
Kolasinski SL, Neogi T, Hochberg MC, Oatis C, Guyatt G, Block J, et al. (2020). 2019 American College of Rheumatology/Arthritis Foundation Guideline for the Management of Osteoarthritis of the Hand, Hip, and Knee (author manuscript, full text read 19 November 2026). 72(2):149-162. Arthritis Care & Research. View source
Perlman A, Fogerite SG, Glass O, Bechard E, Ali A, Njike VY, Pieper C, Dmitrieva NO, et al. (2019). Efficacy and Safety of Massage for Osteoarthritis of the Knee: a Randomized Clinical Trial (full text read 19 November 2026). 34(3):379-386. Journal of General Internal Medicine. View source
Perlman AI, Ali A, Njike VY, Hom D, Davidi A, Gould-Fogerite S, Milak C, Katz DL (2012). Massage therapy for osteoarthritis of the knee: a randomized dose-finding trial (full text read 19 November 2026). 7(2):e30248. PLoS One. View source
Perlman AI, Sabina A, Williams AL, Njike VY, Katz DL (2006). Massage therapy for osteoarthritis of the knee: a randomized controlled trial (abstract). 166(22):2533-2538. Archives of Internal Medicine. View source
Ali A, Kahn J, Rosenberger L, Perlman AI (2012). Development of a manualized protocol of massage therapy for clinical trials in osteoarthritis (Table 1, 60-minute protocol; full text read 19 November 2026). 13:185. Trials. View source
Nelson NL, Churilla JR (2017). Massage Therapy for Pain and Function in Patients With Arthritis: A Systematic Review of Randomized Controlled Trials (abstract). 96(9):665-672. American Journal of Physical Medicine & Rehabilitation. View source
da Silva VR, Ramalho RB, de Noronha M, Bracht MA, Tavares LF, Selistre LFA (2026). Additional Effects of Manual Therapy to Exercise on Pain and Function in Individuals With Knee Osteoarthritis: A Systematic Review and Meta-Analysis (abstract). 30(9):e70397. European Journal of Pain. View source
NHS (2023). Knee pain (page last reviewed 21 December 2023; read 19 November 2026). NHS website. View source
NHS (2023). Osteoarthritis: overview (page last reviewed 20 March 2023; read 19 November 2026). NHS website. View source
NHS (2023). Osteoarthritis: treatment (page last reviewed 20 March 2023; read 19 November 2026). NHS website. View source
NHS (2026). DVT (deep vein thrombosis) (page last reviewed 30 April 2026; read 19 November 2026). NHS website. View source
Bệnh viện Đa khoa Tâm Anh (tư vấn chuyên môn: PGS.TS.BSCC Đặng Hồng Hoa) (2025). Thoái hóa khớp gối: Nguyên nhân, triệu chứng, chẩn đoán và phòng ngừa (cập nhật 31 May 2025; read 19 November 2026). Bệnh viện Đa khoa Tâm Anh. View source
Every room at Nghê Prana is designed around the science of sleep. Blackout curtains, nightly aromatherapy turndown, and riverside quiet — experience what real rest feels like.
Riverside hotel rooms on the Thu Bồn, ten minutes by bicycle from the Ancient Town. Whether it's one night between Hue and Da Nang or a full week of doing nothing — we kept your room quiet.
Free cancellation · Direct from the family who built the hotel