Background: Osteoarthritis is a leading cause of disability. Nonsurgical treatment is a key first step.
Purpose: Systematic literature review of physical therapy (PT) interventions for community-dwelling adults with knee osteoarthritis.
Data sources: MEDLINE, the Cochrane Library, the Physiotherapy Evidence Database, Scirus, Allied and Complementary Medicine, and the Health and Psychosocial Instruments bibliography database.
Study selection: 193 randomized, controlled trials (RCTs) published in English from 1970 to 29 February 2012.
Data extraction: Means of outcomes, PT interventions, and risk of bias were extracted to pool standardized mean differences. Disagreements between reviewers abstracting and checking data were resolved through discussion.
Data synthesis: Meta-analyses of 84 RCTs provided evidence for 13 PT interventions on pain (58 RCTs), physical function (36 RCTs), and disability (29 RCTs). Meta-analyses provided low-strength evidence that aerobic (11 RCTs) and aquatic (3 RCTs) exercise improved disability and that aerobic exercise (19 RCTs), strengthening exercise (17 RCTs), and ultrasonography (6 RCTs) reduced pain and improved function. Several individual RCTs demonstrated clinically important improvements in pain and disability with aerobic exercise. Other PT interventions demonstrated no sustained benefit. Individual RCTs showed similar benefits with aerobic, aquatic, and strengthening exercise. Adverse events were uncommon and did not deter participants from continuing treatment.
Limitation: Variability in PT interventions and outcomes measures hampered synthesis of evidence.
Conclusion: Low-strength evidence suggested that only a few PT interventions were effective. Future studies should compare combined PT interventions (which is how PT is generally administered for pain associated with knee osteoarthritis).
Primary funding source: Agency for Healthcare Research and Quality.