Abstract
Gut dysbiosis and systemic inflammation are increasingly recognized in the pathogenesis of knee osteoarthritis (OA). While altered gut microbiota profiles have been reported, the link between specific microbial taxa and physical functioning in OA patients remains understudied. This study investigated the association between gut microbiota composition and clinical/functional outcomes in mild to moderate knee OA. We analysed faecal samples and clinical data from 18 OA patients and 20 age-matched controls, and we used a machine learning approach to assay associations with functional tests. OA patients showed altered gut microbiota diversity. Machine learning identified Blautia luti, Collinsella aerofaciens, and Butyricicoccus faecihominis as discriminating taxa. Correlation analysis revealed distinct associations between genera of Bacteroidetes and Firmicutes with clinical and functional parameters. Species-level analysis identified species like as F. prausnitzii and Blautia luti as relevant for physical performance and pain. For the first time, this study provides novel evidence linking specific gut bacterial taxa to physical functioning and OA symptoms in mild to moderate knee OA. These findings highlight the potential role of the gut-joint axis and warrant further research into underlying mechanisms and microbiota-targeted interventions for OA management.