Causal association between periodontitis and risk of rheumatoid arthritis and systemic lupus erythematosus: a Mendelian randomization

• Published in: Zeitschrift für Rheumatologie 2020-11, Vol.79 (9), p.929-936
• Main Authors: Bae, S.-C., Lee, Y. H.
• Format: Article
• Language: English
• Subjects: Arthritis, Rheumatoid - epidemiology; Arthritis, Rheumatoid - genetics; Genome-Wide Association Study; Humans; Immunology; Internal Medicine; Laboratory Medicine; Lupus Erythematosus, Systemic - epidemiology; Lupus Erythematosus, Systemic - genetics; Medicine; Medicine & Public Health; Mendelian Randomization Analysis; Original Articles; Orthopedics; Periodontitis - epidemiology; Periodontitis - genetics; Polymorphism, Single Nucleotide - genetics; Rheumatology
• ISSN: 0340-1855; 1435-1250
• DOI: 10.1007/s00393-019-00742-w

Objective To investigate whether periodontitis is causally associated with risk of rheumatoid arthritis (RA) or systemic lupus erythematosus (SLE). Methods We performed two-sample Mendelian randomization (MR) analysis using the inverse variance-weighted (IVW), weighted median, and MR-Egger regression methods on publicly available summary statistics datasets using a periodontitis genome-wide association study (GWAS) as an exposure and RA and SLE GWASs on individuals of European descent as outcomes. Results We selected 7 or 20 single-nucleotide polymorphisms from a periodontitis GWAS as instrumental variables for RA or SLE. The IVW method results support a causal association between periodontitis and RA (beta = 0.168, SE = 0.080, p  = 0.035) and SLE (beta = 0.0001, SE = 0.0001, p  = 0.046) risk; however, the weighted median approach did not indicate a significant causal association. MR-Egger regression revealed that directional pleiotropy was unlikely to be biasing the RA (intercept = −0.115, p  = 0.078) or SLE results (intercept = 4.68E-05, p  = 0.394); no significant causal association was found between periodontitis and RA and SLE. The MR estimates from the IVW, weighted median, and MR-Egger regression analyses were not consistent. Conclusion Only the results of MR analysis by the IVW method indicated that periodontitis is likely causally associated with an increased risk of RA and SLE incidence. Our MR showed weak causal association between periodontitis and RA or SLE. These findings may assist in elucidating the underlying mechanisms of the effects of periodontitis on RA and SLE incidence.