Main

Demmer RT, Molitor JA et al. J Clin Periodontol 2011; 38: 998–1006

Multivariable logistic regression analysis and proportional hazards models (the time interval before the event), 'generally' did not show associations between periodontal disease and new cases of rheumatoid arthritis (RA). In addition, there was no dose-responsiveness. This study used data obtained from NHANES I. In this, 9,702 US men and women were followed longitudinally for 20 years from 1971. At baseline only, the Periodontal Index and tooth loss (surrogate marker for historical periodontal disease) were used to quantify periodontal disease. The investigators used self-reporting for rheumatoid arthritis that had been diagnosed by a physician. Although therefore, this link is 'biologically plausible' and has been shown in animal models, 'it is unlikely that periodontal infections represent a strong RA risk factor in the general US population.'