Impact of Exercise Therapy on Molecular Biomarkers Related to Cartilage and Inflammation in Individuals at Risk of, or With Established, Knee Osteoarthritis: A Systematic Review and Meta-Analysis of Randomized Controlled Trials

Bricca A, et al. · Arthritis Care & Research

Twelve RCTs, 57 comparisons at 4 to 24 weeks. Exercise therapy decreased molecular biomarkers in 17 comparisons (30%), left them unchanged in 36 (63%) and increased them in 4 (7%). Meta-analyses of nine biomarkers found non-significant reductions in C-reactive protein, CTX-II, TNF-alpha, soluble TNF receptors 1 and 2, C2C and cartilage oligomeric matrix protein against non-exercising controls, and no effect on interleukin-6 or its soluble receptor. Change was detected more often in synovial fluid (50% of comparisons) than in blood (36%) or urine (20%), so the compartment sampled may matter as much as the marker.

Population

12 RCTs (57 biomarker comparisons) in people at risk of or with knee OA

Intervention

Exercise therapy, 4-24 weeks

Comparator

Non-exercise controls

Limitations

Heterogeneous biomarkers and protocols, surrogate outcomes, and meta-analytic effects that are mostly non-significant with wide confidence intervals. The reviewers downgrade the overall quality of evidence to low and say plainly that this reflects how few randomised trials exist rather than flaws in the ones that do - though some failed to report dropouts, or whether the people analysing the samples were blinded. Funded by an EU Seventh Framework Programme grant, with no funder involvement in the review declared.