Weight loss is an important treatment for knee osteoarthritis in people living with overweight or obesity. It can reduce joint load, improve metabolic health and help some patients move with less pain. But can intensive weight loss provide the same relief as knee replacement once arthritis is severe and a patient has already been judged eligible for surgery?
The 2026 INKA randomized trial tested that question directly (Henriksen et al., 2026). It compared a supervised low-calorie dietary program with knee arthroplasty in patients who had obesity, severe knee osteoarthritis and a clinical indication for either total or partial knee replacement. Both groups improved, but they did not improve to the same degree. At six months after treatment began, arthroplasty produced substantially greater gains in patient-reported pain and function.
This does not make weight management unimportant. It clarifies what weight loss can and cannot be expected to do at different stages of disease. For a patient with early or moderate arthritis, weight reduction may be a central part of non-surgical care. For a patient with advanced arthritis, major functional loss and a sound indication for replacement, weight loss may improve health and symptoms without reversing the structural problem or reliably matching the benefit of surgery.
Intensive weight loss can improve knee osteoarthritis symptoms, but the INKA randomized trial found that knee arthroplasty provided greater improvements in pain and function among patients eligible for surgery.
Key Takeaways
- In the INKA randomized trial, knee arthroplasty improved patient-reported pain and function more than an intensive dietary weight-loss program among patients already eligible for surgery (Henriksen et al., 2026).
- Intensive diet still produced meaningful weight loss and symptom improvement, but it did not match the surgical group’s results at six months (Henriksen et al., 2026).
- Weight management remains an important part of osteoarthritis care and can complement other treatments (Messier et al., 2013; National Institute for Health and Care Excellence [NICE], 2022).
- Decisions about surgery should consider symptoms, function, individual risks and patient preferences rather than body mass index alone (American College of Rheumatology & American Association of Hip and Knee Surgeons [ACR/AAHKS], 2023).
What the INKA trial compared
Researchers in Denmark enrolled adults with a body mass index of at least 30 kg per square metre and knee osteoarthritis severe enough for an orthopaedic surgeon to recommend knee arthroplasty. Participants also had to be motivated to attempt weight loss. The investigators planned to recruit 200 people but stopped at 92 because recruitment was slower than expected. Forty-five participants were assigned to intensive diet and 47 to knee arthroplasty (Henriksen et al., 2026).
The diet intervention was more intensive than ordinary lifestyle advice. It began with eight weeks of complete meal replacement at about 810 calories per day, followed by four weeks at about 1,200 calories per day with two meal-replacement products daily. A dietitian supervised weekly group sessions that covered behaviour, education and long-term dietary planning.
The surgical arm reflected routine practice. The operating surgeon chose the prosthesis, procedure and rehabilitation pathway. Importantly, 64% of participants had been planned for unicompartmental, or partial, knee arthroplasty rather than total knee replacement. The trial therefore compared intensive diet with knee arthroplasty as a category, not with one standardized total-knee procedure.
The primary outcome was the Oxford Knee Score, a 0 to 48 patient-reported measure of knee pain and function in which a higher score is better. The researchers assessed the main outcome six months after treatment began and defined a four-point margin for non-inferiority (Henriksen et al., 2026). In simple terms, the diet program had to come within four points of the surgical result to be considered meaningfully comparable.
Surgery improved knee symptoms more
At the primary assessment, the adjusted Oxford Knee Score was 28.2 in the intensive-diet group and 36.5 in the arthroplasty group. The difference was 8.2 points in favour of surgery, with a 95% confidence interval from 5.0 to 11.5 points. That result crossed the prespecified four-point margin and showed that intensive diet was clinically inferior to arthroplasty for pain and function at six months (Henriksen et al., 2026).
Other patient-reported measures pointed in the same direction. At six months, the surgery group reported better pain, function, knee-related quality of life and treatment satisfaction. The gap narrowed somewhat by 12 months, partly because the diet group continued to improve. Even then, the Oxford Knee Score remained 6.3 points higher in the arthroplasty group.
The trial also showed that intensive dietary treatment was not futile. The diet group lost an average of 11.0 kilograms, or 10.4% of body weight, by six months. The arthroplasty group lost an average of 1.3 kilograms. At 12 months, the diet group remained an average of 7.1 kilograms below baseline, although some weight had been regained (Henriksen et al., 2026).
The correct interpretation is therefore specific. Intensive weight loss produced substantial weight reduction and meaningful improvement in many participants, but it did not provide symptom and function results comparable with arthroplasty in this group of surgical candidates.
Why disease stage changes the treatment question
Weight loss can reduce the mechanical demand placed on the knee. It may also improve metabolic factors associated with osteoarthritis and make exercise easier. These benefits matter throughout the course of the disease (Messier et al., 2013). They are especially valuable when joint structure and movement remain reasonably preserved and pain is not yet severely limiting daily life.
Advanced osteoarthritis is different. A patient may have marked joint-space loss, fixed varus or valgus deformity, restricted movement, night pain and loss of walking capacity. Weight reduction can lower load and improve health, but it cannot reliably restore lost cartilage, reopen a severely narrowed joint space or correct a fixed bony deformity. Symptom improvement should not be confused with reversal of the disease.
This is why the phrase weight loss versus knee replacement can be misleading. The treatments do not do the same job. Weight management addresses load, conditioning and metabolic health. Arthroplasty resurfaces the damaged joint when pain and disability have become severe enough to justify an operation. Previous randomized evidence has also shown that total knee replacement combined with non-surgical treatment provides greater improvements in pain and function than non-surgical treatment alone in appropriately selected patients with knee osteoarthritis (Skou et al., 2015). The relevant question is whether non-surgical care still offers enough likely benefit for this patient at this stage, not whether one treatment is universally better.
Crossover is part of the clinical message
During the trial, 13 of the 45 participants assigned to intensive diet, or 29%, crossed over to knee arthroplasty. By 12 months, treatment failure was recorded in 31.1% of the diet group and 2.1% of the arthroplasty group. These findings help explain why the groups appeared more similar later: some people originally assigned to diet had already received surgery (Henriksen et al., 2026).
Crossover also reflects a practical problem. Once a patient has severe symptoms and has been told that replacement is appropriate, maintaining non-surgical treatment may be difficult if pain and disability continue despite meaningful weight loss. Delaying surgery can be reasonable when symptoms remain acceptable, when the patient prefers to wait, or when medical optimization is needed. Delay becomes less useful when it simply extends severe disability without changing the likely treatment.
What the trial does not prove
The INKA trial was relatively small and did not reach its planned recruitment target (Henriksen et al., 2026). Only about 42% of eligible patients enrolled, and those who enrolled had somewhat less severe symptoms than those who declined. The results may therefore not apply equally to people with the most severe symptoms or to patients who are not motivated to lose weight.
The study was open-label, which is unavoidable when comparing surgery with diet. The operation was not standardized, and most participants had partial rather than total replacement planned. The intensive program was also unusually structured, using meal replacement and close dietitian supervision. Its results should not be assumed to describe casual dietary advice or unsupervised dieting.
Subgroup analyses suggested that diet may have been more competitive among participants with a baseline body mass index below 35, but the trial was not large enough to make that a firm selection rule. A subgroup signal should guide further research, not become an automatic threshold for denying or recommending surgery.
Adverse-event incidence was similar between groups, but this should also be interpreted cautiously because the safety analysis was small. The surgical group reported wound-related events and three surgical-site infections (Henriksen et al., 2026). Surgery carries real risks, while severe calorie restriction and obesity-related medical conditions also require appropriate supervision.
How this should guide a real decision
A sound decision begins with diagnosis and function. The clinician should confirm that knee osteoarthritis is the main source of pain, assess the distribution of disease and deformity, examine movement and stability, and understand how much the knee limits walking, stairs, sleep, work and independence. Weight-bearing radiographs should support the clinical picture, but an X-ray alone should not decide the operation.
For early or moderate arthritis, structured exercise, education, weight management where relevant, medicines and selected injections may provide useful symptom control. Current injections should not be presented as proven cartilage-regrowth treatments or as guaranteed ways to avoid future surgery. Their role depends on arthritis stage, the source of pain, realistic goals and the expected duration of benefit (NICE, 2022).
For severe arthritis, the patient should not be told that weight loss is pointless. Improving weight, strength, diabetes control, nutrition and general health may reduce risk and support rehabilitation. But optimization should not become an indefinite barrier when pain and disability remain substantial and the indication for arthroplasty is sound. Body mass index is a risk factor to discuss and manage, not a complete treatment decision by itself (ACR/AAHKS, 2023).
The INKA trial gives patients and clinicians a more honest framework. Intensive weight loss can improve symptoms and health, even in severe arthritis. In patients already eligible for knee arthroplasty, however, it did not match the improvement in knee pain and function achieved with surgery (Henriksen et al., 2026). The treatment plan should respect both findings.
Final Thoughts
The INKA trial helps clarify a decision that many patients with severe knee osteoarthritis face: intensive weight loss can be beneficial, but it is not necessarily a substitute for knee arthroplasty when surgery is already indicated (Henriksen et al., 2026). The choice should be guided by the severity of pain and disability, response to non-surgical care, overall health and the patient’s priorities. Weight management can remain part of the treatment plan whether a patient proceeds with surgery or continues conservative care (ACR/AAHKS, 2023; NICE, 2022).
References
Henriksen M, Christensen R, Poulsen AS, et al. Intensive diet or knee arthroplasty for severe knee osteoarthritis: the INKA randomised trial. Osteoarthritis and Cartilage. Published online July 10, 2026. https://doi.org/10.1016/j.joca.2026.07.004
Skou ST, Roos EM, Laursen MB, et al. A randomized, controlled trial of total knee replacement. New England Journal of Medicine. 2015;373:1597-1606. https://doi.org/10.1056/NEJMoa1505467
Messier SP, Mihalko SL, Legault C, et al. Effects of intensive diet and exercise on knee joint loads, inflammation, and clinical outcomes among overweight and obese adults with knee osteoarthritis. JAMA. 2013;310:1263-1273. https://doi.org/10.1001/jama.2013.277669
American College of Rheumatology and American Association of Hip and Knee Surgeons. Clinical practice guideline for the optimal timing of elective hip or knee arthroplasty. 2023. https://rheumatology.org/indications-for-total-hip-and-knee-arthroplasty-guideline
National Institute for Health and Care Excellence. Osteoarthritis in over 16s: diagnosis and management. NICE guideline NG226. 2022. https://www.nice.org.uk/guidance/ng226
