Knee osteoarthritis is the most common form of arthritis, globally affecting 365 million people and representing one of the main causes of chronic pain and functional limitation in adults over 50 years. The non-pharmacologic therapeutic arsenal is vast — physical therapy, exercise, laser, ultrasound, electrical stimulation, knee braces, hydrotherapy — but direct comparisons between these modalities are scarce.
A network meta-analysis published in PLOS One by Chen X et al. (2025) — widely echoed by ScienceDaily in March 2026 — performed the most comprehensive comparison ever published: 139 randomized clinical trials, 9,644 patients, and 12 simultaneous interventions in a single network analysis. The result ranks the available options for pain, physical function, joint stiffness, and total WOMAC score, providing input for individualized therapeutic decisions.
The scale of the meta-analysis: 12 interventions, 139 studies
The comprehensiveness of this NMA is its main differential. While most meta-analyses compare 2 to 4 interventions, this frequentist analysis with random-effects models allowed simultaneous ranking of:
The 12 physical interventions compared
- Low-Level Laser Therapy (LLLT)
- High-Intensity Laser Therapy (HILT)
- TENS (Transcutaneous Electrical Nerve Stimulation)
- Interferential Current (IFC)
- Short Wave Diathermy
- Therapeutic ultrasound
- Lateral Wedged Insole
- Knee Brace
- Physical exercise (aerobic and/or strengthening)
- Hydrotherapy / aquatic exercise
- Functional taping (Kinesio Taping)
- Extracorporeal shock wave therapy (ESWT)
The primary outcome was the WOMAC score (Western Ontario and McMaster Universities Osteoarthritis Index) for pain, function, and stiffness — the global reference questionnaire for osteoarthritis. Secondary outcomes included the VAS (Visual Analogue Scale) for pain at rest and pain during activity. Network consistency was verified by inconsistency tests and funnel analysis.
NMA OF 12 THERAPIES FOR KNEE OA — STUDY DATA
Results: knee brace and hydrotherapy at the top of the hierarchy
The SUCRA analysis (Surface Under the Cumulative Ranking Curve) — which quantifies the probability of each intervention being the best on each outcome — revealed a consistent hierarchy with two highlighted winners:
HIERARCHY OF INTERVENTIONS BY OUTCOME (SUCRA)
| OUTCOME | BEST RANKED INTERVENTION |
|---|---|
| WOMAC pain | Knee Brace — highest probability of being the best intervention · superiority over 11 of 12 comparators in direct analyses |
| WOMAC physical function | Knee Brace — significant difference vs. 11 of 12 comparators · clinically relevant benefit in daily activities |
| WOMAC joint stiffness | Knee Brace — best ranked · followed by exercise and hydrotherapy |
| WOMAC total score | Hydrotherapy — best for composite outcome of pain + function + stiffness · superiority vs. interferential current, ultrasound, and placebo |
The authors’ conclusion is direct: "The knee brace may be the most recommended therapeutic option for knee osteoarthritis, followed by hydrotherapy and exercise." Therapeutic ultrasound presented the worst results on multiple outcomes, questioning its routine use in knee OA.
Why does the knee brace surpass exercise, TENS, and laser?
The surprising result for the knee brace has biomechanical support. Predominantly medial knee osteoarthritis (internal compartment) — which represents 80% of cases — is aggravated by the knee adductor moment, a force that compresses the medial compartment during gait. Offloading braces reduce this moment by 10 to 30%, directly relieving cartilage overload.
Hydrotherapy, in turn, benefits from aquatic buoyancy that eliminates 90% of body weight on the joints during exercise, allowing range of motion without painful overload — especially advantageous in patients with moderate to severe OA and high body mass index.
Frequently Asked Questions
No. There are several types of knee braces for knee OA: (1) offloading braces — they shift the load from the most affected compartment (usually medial) to the lateral; these are the most effective for OA with varus deviation (knee inward), but need to be prescribed and adjusted individually; (2) stabilization sleeves (sleeve braces) — they provide compression and proprioception, without load redistribution; more modest effect; (3) lateral wedged insoles — a more accessible alternative, with smaller evidence. The NMA evaluated knee braces in general, but individualized prescription determines the clinical result.
In a limited way, depending on the country and region. Some rehabilitation centers, specialized services, and rehabilitation hospitals offer hydrotherapy through public networks. Coverage is typically unequal between regions. In addition, aquatic gymnastics in community pools or fitness centers covered by health plans is an alternative with proven effect — it does not need to be hospital-supervised hydrotherapy to have clinical benefit.
Acupuncture has independent and robust evidence for knee osteoarthritis. A network meta-analysis published in the BMJ in 2024 with 80 randomized clinical trials and more than 11,000 patients confirmed that acupuncture is superior to sham and to no treatment for pain and function in knee OA. Its mechanism is distinct from the physical therapies of this NMA: while knee brace and exercise act peripherally (mechanical unloading, muscle strengthening), acupuncture centrally modulates pain perception via the descending inhibitory system. The combination is clinically synergistic.
This NMA positioned ESWT favorably for knee OA — especially for function. The proposed mechanism includes neoangiogenic stimulation in subchondral bone, modulation of inflammatory mediators (IL-1β, PGE2), and direct analgesic effect by hyperstimulation of nociceptors. The literature has studies with variable results, partly due to differences in protocols (focused vs. radial shock waves, number and intensity of sessions). For patients with moderate OA who have not responded to other conservative measures, ESWT is a valid alternative before considering surgery.
The protocols of the studies in this NMA varied widely, but the guidelines suggest that an initial program of 8 to 12 weeks with 2 to 3 sessions per week produces measurable benefit. After this phase, maintenance with a home program (independent exercise initially supervised by a physical therapist or physical educator) is essential to preserve the gains. The knee brace can be used continuously during activities that provoke pain (walking, climbing stairs) without time limit, provided it is tolerated.
Sources consulted
- Chen X, Fan Y, Tu H, Luo Y. Network meta-analysis of non-pharmacological therapies for knee osteoarthritis. PLOS One. 2025;20(6):e0324864. DOI: 10.1371/journal.pone.0324864. PMID: 40531843.
- Bannuru RR, Osani MC, Vaysbrot EE, et al. OARSI guidelines for the non-surgical management of knee, hip, and polyarticular osteoarthritis. Osteoarthritis Cartilage. 2019.
- Kolasinski SL, Neogi T, Hochberg MC, et al. 2019 ACR/Arthritis Foundation guideline for management of osteoarthritis. Arthritis Care Res. 2020.
- Mazzei DR, Ademola A, Abbott JH, Sajobi T, Hildebrand K, Marshall DA. Are education, exercise and diet interventions a cost-effective treatment to manage hip and knee osteoarthritis? A systematic review. Osteoarthritis Cartilage. 2021.
- Zhao L, Liu J, Zhang F, et al. Effects of long-term acupuncture treatment on patients with knee osteoarthritis: a network meta-analysis. BMJ Open. 2024.
- ScienceDaily. Non-pharmacological therapies for knee osteoarthritis ranked by network meta-analysis. March 2026.
Founded in 1989 by physicians trained at the University of São Paulo (USP) and specialized in China, CEIMEC is a Brazilian national reference in the teaching and practice of medical acupuncture. With more than 3,000 physicians trained over 35 years, it collaborates with HC-FMUSP and is recognized by the Brazilian Medical College of Acupuncture (CMBA/AMB).