Objective: To describe the effects of proprioceptive training on pain, stiffness, function, and functional test outcomes among patients with knee osteoarthritis (OA).

Data sources: All studies completed from 1946 to 2017 were obtained from 4 databases (PubMed, MEDLINE, CINAHL, and SPORTDiscus).

Study selection: Three reviewers independently identified appropriate studies and extracted data.

Data extraction: Methodologic quality and level of evidence were assessed using the Physiotherapy Evidence Database scale and Oxford Centre for Evidence-Based Medicine guidelines. The standardized mean differences (SMDs) and 95% confidence intervals (CIs) were calculated for pain, stiffness, function, and functional test outcomes.

Data synthesis: Seven randomized controlled trials involving 558 patients with knee OA met the inclusion criteria. The selected studies had Physiotherapy Evidence Database scores of 6 to 8. All randomized controlled trials had an Oxford Centre for Evidence-Based Medicine level of evidence of 2. Meta-analysis of the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) pain subscale (SMD = -0.56; 95% CI = -1.06, -0.07; P = .026), function subscale (SMD = -0.40; 95% CI = -0.59, -0.21; P < .001), and non-WOMAC walking speed test (SMD = -1.07; 95% CI = -2.12, -0.01; P = .048) revealed that proprioceptive training had significant treatment effects. Proprioceptive training was not associated with reductions in WOMAC stiffness subscale scores and did not improve non-WOMAC get-up-and-go scores.

Conclusions: Proprioceptive training effectively promoted pain relief and completion of functional daily activity among patients with knee OA and should be included in rehabilitation programs. Stiffness and other mobility measures were unchanged after proprioceptive training. Modified proprioceptive training programs are needed to _target stiffness and improve additional physical function domains.