Which exercise works best for people with knee OA?

Aerobic exercise appears to be the best form of workout for people with knee osteoarthritis (OA), resulting in improved pain, function, gait performance, and quality of life (QoL), suggests a study.

“Exercise treatment offers obvious benefits for individuals with knee OA,” the researchers said. “With moderate certainty quality evidence, aerobic exercise might be the best exercise modality for improvements in pain, function, gait performance, and QoL.”

This systematic review and network meta-analysis obtained data from the databases of PubMed, Embase, Cochrane Library, Web of Science, CINAHL, PsycINFO, AMED, PEDro, Scopus, ClinicalTrials.gov, ICTRP, and ClinicalTrialsRegister.eu from inception to August 2024.

The research team identified randomized controlled trials (RCTs) comparing different exercise modalities, including aerobic exercise, flexibility exercise, mind-body exercise, neuromotor exercise, strengthening exercise, mixed exercise, and control group for patients with knee OA.

A total of 217 RCTs, including 15,684 participants, met the eligibility criteria. Based on moderate certainty evidence, aerobic exercise resulted in large improvement in pain at short-term (standardized mean difference [SMD], ‒1.10, 95 percent confidence interval [CI], ‒1.68 to ‒0.52) and mid-term follow-up (SMD, ‒1.19, 95 percent CI, ‒1.58 to ‒0.79) compared with control. [BMJ 2025;391:e085242]

Likewise, aerobic exercise improved function at mid-term (SMD, 1.78, 95 percent CI, 1.05‒2.51), gait performance at mid-term (SMD, 0.85, 95 percent CI, 0.55‒1.14), and QoL at short-term follow-up (SMD, 1.53, 95 percent CI, 0.47‒2.59).

Mind-body exercise potentially led to a huge increase in function at short-term follow-up (SMD, 0.88, 95 percent CI, 0.03‒1.73; moderate certainty). Neuromotor exercise also resulted in a large improvement in gait performance at short-term follow-up (SMD, 1.04, 95 percent CI, 0.51‒1.57; moderate certainty).

Moreover, strengthening (SMD, 0.86, 95 percent CI, 0.53‒1.18) and mixed exercise (SMD, 1.07, 95 percent CI, 0.68‒1.46) led to an increase in function at mid-term follow-up (moderate certainty).

Best treatment

Over long-term follow-up, flexibility exercise reduced pain (SMD, ‒0.99, 95 percent CI, ‒1.63 to ‒0.36; low certainty), aerobic exercise improved function (SMD, 0.87, 95 percent CI, 0.02‒1.72; low certainty), and mixed exercise resulted in increases in function (SMD, 0.56, 95 percent CI, 0.26‒0.86; low certainty) and gait performance (SMD, 0.57, 95 percent CI, 0.21‒0.92; moderate certainty).

Overall, aerobic exercise was consistent in having the highest probability of being the best treatment, as indicated by surface under the cumulative ranking curve values (mean 0.72) across outcomes.

Only a small proportion of studies (18 percent) reported safety outcomes, with no clear differences noted between exercise interventions and control.

“This study presents a comprehensive and up-to-date analysis of the effects of exercise intervention as a treatment strategy for knee OA, which could assist clinicians in prescribing therapy to improve treatment outcomes for patients,” the researchers said.

Adherence

A variety of factors can influence the outcomes of therapeutic exercise in patients with knee OA, including adherence to the program and inherent characteristics of the outcome measures.

An earlier study found that greater adherence resulted in large improvements in pain and function. [Arthritis Care Res (Hoboken) 2010;62:1087-1094]

“However, adherence was not consistently reported across the included trials in our study, which limits further analysis,” the researchers said.

“Additionally, commonly used outcome measures, such as Western Ontario and McMaster Universities Osteoarthritis Index or Knee injury and Osteoarthritis Outcome Score, may lack the required sensitivity to detect early neuromuscular or psychosocial improvements induced by exercise,” they added. [Arthritis Care Res (Hoboken) 2016;68:1224-1231; Arthritis Care Res (Hoboken) 2016;68:1224-1231]