Short-term exposure to outdoor air pollutants and rheumatoid arthritis activity in metropolitan areas in the North of Italy

Background: Air pollution is believed to cause oxidative stress and systemic inflammation, that could trigger autoimmunity in rheumatoid arthritis (RA). Several epidemiological studies investigated the possible role of air pollution in the outbreak of RA with controversial results. As far as we know, studies on the effects on disease activity of short-term exposure have not been published. Objectives: To evaluate the impact of short-term exposure to air pollutants (daily mean PM10, PM2.5, NO2 and O3) on disease activity in patients with RA. Methods: Consecutive patients with RA (ACR/EULAR Criteria 2010) resident in Lombardy (Italy) were enrolled. In each patient Disease Activity Score on 28 joints (DAS28), Simple Disease Activity Index (SDAI) were assessed. Daily PM10, PM2.5, NO2 and O3 concentrations, estimated by Regional Environmental Protection Agency at municipality resolution, were used to assign short-term exposure from day of visit back to 14 days. Multivariable linear regression models were performed to identify the day of the pollutants independently associated with disease activity indices, adjusting for the variables significant at the univariate analysis. β coefficients were reported for 1 μg/m3 increments of pollutants’ concentrations. Results: 422 RA patients were enrolled in the study between January and June 2018: 81.5% females, mean age 58.2±13.3 years, mean disease duration 16.1±11.5 years, 27.3% current smokers, 59.5% RF positivity, 54.5% ACPA positivity. Sparse punctual statistically significant negative associations emerged at the multivariate analysis between PM10, PM2.5, NO2 and the outcomes, although with very low estimates, whereas positive associations resulted for O3. Afterwards patients were stratified in 3 subgroups according to their ongoing treatment (no therapy, n=25, conventional synthetic Disease Modifying anti-Rheumatic Drugs -DMARDs-, n=108 and biological or targeted synthetic DMARDs, n=289). A statistical significance was found by analysing the influence of therapy on the interaction between PM2.5 and DAS28 (Figure below): a positive trend between PM2.5 and DAS28 appeared in the first two groups (no therapy, 0.013±0.007, p=0.06 and csDMARDs, 0.006±0.004, p=0.17), whereas a statistically significant inverse association was seen in the b/tsDMARDs group (-0.005±0.002, p=0.01). Therapy interaction was particularly evident in several days before the visit also for O3. Conclusion: The changes of the outcome measures related to the increase of the pollutants’ levels did not reach the minimal clinically important difference, therefore air pollution seems barely relevant on disease activity once the loss of tolerance is established in RA. O3 and PM/NO2 always exhibit an opposite performance having inversely proportional atmospheric concentrations, whereas the biological role of this substance is still matter of debate and will need further understanding. Therapy seems to be able to interact with the relation between air pollutants and the parameters considered.