Mechanical-based therapies may reduce pain and disability in some patients with knee osteoarthritis: A systematic review with meta-analysis

Background: Mechanical-based therapies are not yet recommended to manage osteoarthritis (OA). This systematic review and meta-analysis aim to assess the effects of passive mechanical-based therapies (isolated or combined with other therapies) on patients with knee OA compared to placebo, other isolated or combined interventions. Methods: Pubmed, Cochrane, Web of Science and EMBASE were searched up to December 2020. We included randomized and non-randomized trials using therapeutic ultrasound, phonophoresis, extracorporeal shockwave therapy (ESWT) and vibration (single or combined with other therapies) compared to placebo, and/or other physical therapies groups. Biochemical, patient-reported, physical and imaging outcome measures were retrieved. We judged risk of bias using the RoB2 tool for randomized studies, the ROBINS-I tool for non-randomized studies, and the GRADE to interpret certainty of results. Results: We included 77 clinical studies. Ultrasound and ESWT statistically improved pain and disability comparing to placebo (combined or not with other therapies), and when added to other therapies versus other therapies alone. Ultrasound was statistically inferior to phonophoresis (combined or not with other therapies) in reducing pain and disability for specific therapeutic gels and/or combined therapies. Vibration plus exercise statistically improved pain relief and function versus exercise alone. All meta-analyses showed verylow certainty of evidence, with 15 of 42 (38%) pooled comparisons being statistically significant (weak to large effect).