Abstract 2184: The Association of Cigarette Smoking with Disease Characteristics in African Americans with Early Rheumatoid Arthritis
Authors
T. R. Mikuls, L. B. Hughes, A. O. Westfall, G. Howard, G. S. Alarcón, V. M. Holers, D. L. Conn, E. A. Smith, G. S. Gilkeson, D. van der Heijde, B. Jonas, L. F. Callahan, S. L. Bridges, Jr
Purpose:
Smoking has been identified as a risk factor for the development of RA, as well as for seropositivity. However, the studies to date have been performed nearly exclusively in Caucasians. The current study examined the association of smoking in African American (AA) subjects with early RA with RF isotypes, anti-CCP antibody status, radiographic erosions and subcutaneous nodules.
Methods:
300 AA subjects with early RA (< 2 yrs) in the CLEAR registry were included. The association of smoking with outcomes (seropositivity, nodules, erosions [n=199]) at baseline was evaluated by multivariable regression, adjusting for shared epitope (SE) status. A smoking-SE interaction term was also included in the model.
Results:
Study subjects were predominantly women (83%). Mean age was 50.6 + 13.5 yrs. 14% had nodules and 27% had erosions. 22% were past, and 30% were current, smokers. 62% were positive for anti-CCP, 67% were positive for RF IgA, and 70% for RF IgM. Smoking status was not associated with autoantibody status nor with erosions. In the logit regression models, however, current smokers were more than twice as likely as never smokers to have subcutaneous nodules, and more than 70% as likely as never smokers to have high concentrations of RF IgA. There was no evidence for smoking-SE interactions in any of the statistical models. Only 42% of participants were positive for SE (1 or 2 alleles).
Conclusions:
Smoking was associated with some limited features of RA in AAs. However, the data did not support a role of smoking-SE interactions in mediating disease-specific features in AAs.
Editorial Comment:
This analysis examined patients with existing RA. Because there was no control cohort, the investigators were not able to examine the question of whether smoking is a risk factor for the development of RA. This is perhaps the most cogent question that begs an answer. However, with regard to smoking being a risk factor for clinical and laboratory features of RA, the results were mixed. Smoking was not associated with RF or anti-CCP status, only with high levels of RF IgA. Smoking was associated with nodules but not severity of disease, as measured by erosions. The absence of an interaction of smoking with SE for disease severity is perhaps the most important finding, as this clearly differs from findings in Caucasian populations. Furthermore, a lower prevalence of SE positivity was observed in this AA RA population, only 42% compared to about 2/3rds in most Caucasian RA populations studied to date. These results point out the danger in extrapolating conclusions from data in one racial population to another, and emphasize the need for more study of autoimmune diseases in minority populations.
