Abstract

Proprioception is signi ficantly impaired in knee osteoarthritis (KOA), contributing to reduced functionality. Strength training (ST) is essential in KOA by improving muscle strength, although it may also be effective in improving proprioception. The purpose was to determine the effect of ST on knee proprioception in KOA patients. Pubmed, CINAHL, Scopus, WOS, and PEDro were searched for randomized controlled trials (RCTs) (inception to March 2023). Comparisons for ST were physical exercise different from ST, non-exercise-based interventions, and no intervention. Methodological quality was assessed using the PEDro scale, and risk of bias (RoB) using the Cochrane tool. Meta -analyses were performed by comparison groups using the standardized mean difference (SMD) (Hedge's g ) with random effects models, also considering subgroups by proprioception tests. Finally, six RCTs were included. The mean PEDro score was 6.3, and the highest proportion of biases corresponds to performance, selection, and detection. The meta-analysis indicated that only when compared with non-intervention, ST signi ficantly improved knee proprioception for the joint position sense (JPS) (active + passive), JPS (passive), and threshold to detect passive motion (TTDPM) subgroups ( g = -1.33 [-2.33, -0.32], g = -2.29 [-2.82, -1.75] and g = -2.40 [-4.23, -0.58], respectively). However, in the knee JPS (active) subgroup, ST was not signi ficant ( g = -0.72 [-1.84, 0.40]). In conclusion, ST improves knee proprioception compared to non-intervention. However, due to the paucity of studies and diversity of interventions, more evidence is needed to support the effectiveness of ST. Future RCTs may address the limitations of this review to advance knowledge about proprioceptive responses to ST and contribute to clinical practice.