Background Rheumatoid arthritis (RA) is a systemic autoimmune disease, the most severe manifestation being interstitial lung disease (ILD). Therapeutic options are undefined and ILD is one of the leading causes of mortality for RA patients. A better characterization of RA-ILD–associated factors may improve the risk stratification of RA patients.
Objectives To identify variables associated with ILD occurrence in RA.
Methods RA-ILD patients (RA-ILD+) in a French multicentric cohort were compared to monocentric RA cases without ILD (RA-ILD-). All patients fulfilled 2010 ACR/EULAR criteria for RA and had a HRCT chest scan for suspected ILD or during a pre-biologic assessment. ILD was defined by HRCT results. Because ILD occurrence increases with RA duration, RA-ILD+ and RA-ILD- patients were matched by disease duration (1:1). Demographic data, clinical features of RA and treatment modalities before the ILD diagnosis were collected. Results significant on univariate analysis (P<0.15) were selected for multivariate analysis to identify variables independently associated with RA-ILD. Duration of methotrexate (MTX) and anti-TNF agent use was adjusted to mean MTX or anti-TNF duration before ILD diagnosis.
Results Among 253 consecutive RA patients, 81 were male (32.02%), 112 (48.07%) were smokers, mean age at RA onset was 48.11 ± 16.33 years, and mean RA duration 11.69 ± 10.10 years; 199 patients (85.78%) were ACPA-positive and 171 (71.85%) had erosive status; 134 (60.63%) received MTX and 63 (28.25%) a TNF inhibitor. Among the 253 patients, 138 (54.55%) had ILD. Among the RA-ILD+ patients, 72 (74.23%) had UIP and 25 (25.77%) NSIP. On univariate analysis, ILD was associated with older age at RA onset (52.69 ± 15.00 vs 42.86 ± 16.26 years, P=0.0001), male sex (40.68% vs 21.74%, P=0.001), ACPA rate ≥60 N (20.63% vs 6.41%, P=0.012), less use of anti-TNF agents and MTX (14.55% vs 41.59%, P=0.001, and 50.94% vs 69.57, P=0.005, respectively) and lower MTX dosage (15.26 ± 4.44 vs 16.83 ± 4.31 mg/week, P=0.046). There was no difference for erosive status. On multivariate analysis, ILD was independently associated with older age at RA onset (P<0.0001), male sex (P=0.0006), less use of MTX (P=0.015) with lower dosage (P=0.005) and less use of anti-TNF agents (P=0.004).
Conclusions As previously reported, ILD was independently associated with male sex as well as older age at RA onset and, of interest, less use of MTX and anti-TNF biologics before ILD diagnosis, even though patients did not differ in erosive status. Larger prospective studies are required to further elucidate the role of DMARDs and anti-TNF biologics in ILD occurrence and progression in RA patients.
Disclosure of Interest None declared