Exercise-induced effects on inflammatory markers and brain-derived neurotrophic factor in patients with knee osteoarthritis. A systematic review with meta-analysis

Background: In the pathogenesis of knee osteoarthritis (KOA), inflammatory mediators play an important role. However, the precise underlying mechanism by which regular exercise therapy (ET) exert effects on the immune system in KOA patients is unknown.

Objectives: The aim of this systematic review was to investigate the basal and acute effects of ET on inflammatory biomarkers and brain derived neurotrophic factor (BDNF) in KOA patients.

Methods: PubMed, Web Of Science and PEDro were systematically searched for appropriate studies. If possible, a meta-analysis was performed or an approximation of the effect size (ES) was calculated. Risk of bias was scored using the Cochrane ROB 2.0 or ROBINS-tools.

Results: Twenty-one studies involving 1374 participants were included. Fifteen articles focused on basal exercise effects, four on acute effects, and two on both. Biomarker analysis (n=18) was performed in synovial fluid (n=4) or serum/plasma (n=17). A meta-analysis demonstrated that basal CRP was reduced in KOA patients 6-18 weeks weeks after ET (MD: -0.17;95%CI[-0.31;-0.03]), while IL-6 (MD: 0.21;95%CI[-0.44;0.85]), and TNF-α (MD: -0.57;95%CI[-1.47;0.32]), levels did not significantly change. Also, sTNFR1/2 did not change significantly after ET. For other biomarkers, insufficient data were available to perform a meta-analysis. Nevertheless, a low degree of evidence was found for a decrease in IL-6 (ES:-0.596 & -0.259 & -0.513), an increase in sTNFR1 (ES:2.325), a decrease in sTNFR2 (ES:-0.997) and an increase in BDNF (ES:1.412). Locally, intra-articular IL-10 (ES:9.163) increased, and IL1β (ES:-6.199) and TNF-α decreased (ES:-2.322) after ET. An acute exercise session elicited a myokine response (ES IL-6:0.314), and an increase in BDNF (no ES-data). No inflammatory effect (ES CRP:0.052; ES TNF-α:-0.019 & 0.081) following an acute bout of training was found. However, a single bout of exercise elicited a decrease in intra-articular IL-10 (no ES-data).

Conclusion: ET can induce circulatory and intra-articular anti-inflammatory effects in patients with KOA. The antiinflammatory properties have important implications for informing these patients and clinicians about the underlying effects of ET.