Genetic association of gut microbiota with osteoarthritis: a multivariable Mendelian randomization study considering medication use

<b>Background.</b> The interplay among human gut microbiota (GM) composition, osteoarthritis (OA) and OA-related medication use has been extensively discussed. However, to date, there has been no exploration of the genetic correlation among these three factors.<b>Hypothesis/Gap.</b> The potential causal link between GM and OA), and whether medications influence this relationship, remains unclear.<b>Methods.</b> We utilized bidirectional Mendelian randomization (MR) to explore the genetic associations between GM and OA. We leveraged genome-wide association study (GWAS) summary statistics from the MiBioGen and GO consortia, which provided data on GM taxa and OA cases, respectively. We identified outlier single-nucleotide polymorphisms using radial-MR and assessed causal associations using inverse variance weighting (IVW), weighted median and MR-Egger methods. Robust outcomes, consistent across these methods, were reported. We addressed potential biases through tests for horizontal pleiotropy and heterogeneity, supplemented by the Mendelian randomization pleiotropy residual sum and outlier method. Multivariable MR techniques were applied to adjust for OA medication use using UK Biobank data.<b>Results.</b> IVW estimates revealed a significant increase in hip OA risk for <i>Gordonibacter</i> and <i>Eubacterium</i> (brachy group) [odds ratio (OR): 1.09, 95% confidence interval (CI): 1.04-1.15, <i>P</i>=7.82E-04; OR: 1.09, 95% CI: 1.03-1.16, <i>P</i>=4.67E-03, respectively]. Conversely, <i>Senegalimassilia</i>, <i>Slackia</i> and <i>Streptococcus</i> exhibited protective effects (OR: 0.88, <i>P</i>=2.14E-02; OR: 0.88, <i>P</i>=3.33E-02; 0.91, <i>P</i>=4.29E-02). <i>Sutterella</i> increased the risk of knee OA (OR=1.15, 95% CI: 1.07-1.25, <i>P</i>=4.06E-04), while <i>Haemophilus</i> decreased it (OR=0.94, 95% CI: 0.88-1.00, <i>P</i>=4.26E-02). No significant heterogeneity or horizontal pleiotropy was observed in the results. Even after accounting for the potential confounding effect of medication, the results remained consistent. No reverse causation was detected.<b>Conclusions.</b> Our MR study reveals gut microbiome links to OA risk. Associations hold after adjusting for medication, indicating a potential causal connection between GM and OA.