Arthritis & Rheumatism, Volume 62,
November 2010 Abstract Supplement

Abstracts of the American College of
Rheumatology/Association of Rheumatology Health Professionals
Annual Scientific Meeting
Atlanta, Georgia November 6-11, 2010.

Gene-Environment Interaction Determines Interstitial Lung Disease in Rheumatoid Arthritis.

Restrepo2,  Jose Felix, Rincon4,  Inmaculada del, Zuniga3,  Ricardo, Pogosian4,  Samvel, Escalante1,  Agustin

Univ of Texas HSC San Antonio, San Antonio, TX
Univ of Texas HSC, San Antonio, San Antonio, TX
University of Texas HSC, San Antonio, TX
UTHSCSA, San Antonio, TX


Interstitial lung disease (ILD) is a progressive fibrotic process of unknown cause that may affect rheumatoid arthritis (RA) patients. We examined the role of gene-environment interaction as a determinant of the presence of ILD in RA patients.


We performed a cross sectional study of prevalence and factors associated with ILD in a cohort of RA patients. ILD was diagnosed using chest X-ray, CT scan, MRI, or lung biopsy. Patients with ILD were compared to RA patients without ILD or other pulmonary diseases. We used the following independent variables: age, gender, rheumatoid factor (RF), disease duration, DAS 28, ESR, smoking status, SF36 general health, global assessment of disease severity and activity, Steinbrocker functional class, modified HAQ, current or past use of metothrexate and current use of prednisone. HLA-DRB1 shared epitope (SE) was assessed using sequence-specific DNA primers. We used t-test for continuous variables; chi square and odds ratio (OR) for categorical variables and logistic regression for multivariable analyses, adjusting for confounders.


The sample included 779 RA patients. Sixty nine of them (8.8%) had ILD. We compared them to the 563 patients without pulmonary disease. We excluded 147 patients who had other pulmonary conditions. Significant differences between the two groups included (mean in ILD vs non-ILD patients): age 60.2 vs 52.9 (p<0.0001), age of RA onset: 47.5 vs 42.7 (p<0.004), SF36 general health: 68 vs 62 (p<0.01), ESR: 58 vs 39 (p<0.0001),DAS28: 5.96 vs 5.37 (p<0.002), modified HAQ 2.04 vs 1.87 (p<0.05), diseases severity: 3.76 vs 2.77 (p< 0.0004)and Steinbroker functional capacity 2.40 vs 2.06 (p<0.001). For categorical variables, ILD was associated with nodules, OR (95% CI) = 1.85 (1.05, 3.2; p<0.001); rheumatoid factor,OR = 3.15 (1.32,9.10, p<0.006); male gender, OR = 3.33 (1.93, 5.7, p<0.000001), current prednisone, OR = 2.02 (95% CI, 1.17–3.5), smoking OR = 2.19 (1.23,4.04, and the SE, OR = 1.98 (0.99, 4.32). In multivariate analysis, male gender, OR = 3.08 (1.74, 5.44), RF, OR = 2.58 (1.04, 6.39), prednisone,OR = 1.77 (1.007–3.13); DAS28, OR 1.42 (1.17, 1.73) showed significant, independent association with ILD.

We tested the SE for interaction with smoking using product terms in logistic regression. Smoking was strongly associated with ILD in the presence, but not in the absence of the SE. (Table)

Shared EpitopeSmoking OR for ILD (95% CI)p value
Negative0.54 (0.15–1.9)0.152
Positive2.19 (1.1–4.5)0.033


ILD is frequent in RA, and is associated with aging, male gender, nodules, active disease, positive RF and current use of prednisone. The SE was necessary for smoking to be associated with ILD, providing an example of gene-environment interaction. Further research should focus on the use of variables associated with ILD for early identification of patients at risk for ILD. Mechanistic research into the SE x smoking interaction would be of great interest.

To cite this abstract, please use the following information:
Restrepo, Jose Felix, Rincon, Inmaculada del, Zuniga, Ricardo, Pogosian, Samvel, Escalante, Agustin; Gene-Environment Interaction Determines Interstitial Lung Disease in Rheumatoid Arthritis. [abstract]. Arthritis Rheum 2010;62 Suppl 10 :1035
DOI: 10.1002/art.28802

Abstract Supplement

Meeting Menu