Background:Rheumatoid arthritis (RA) is a systemic inflammatory disease of the joints and other organs, including the lungs.1 Interstitial lung disease (ILD) is a lung injury pattern associated with significant symptom burden and poor outcomes in RA.2 Better understanding of its risk factors could help with disease prevention and treatment.Objectives:Using a population-based cohort, we sought to ascertain the incidence and risk factors of RA-associated ILD (RA-ILD) in recent years.Methods:The study included adult residents of Olmsted County, Minnesota with incident RA between 1999 and 2014 based on the 1987 ACR classification criteria.3 Study subjects were followed until death, migration, or 4/30/2019. ILD was defined by the presence of bilateral interstitial fibrotic changes (excluding biapical scarring) on chest computed tomography (CT). In the absence of chest CT imaging, a physician’s diagnosis of ILD in conjunction with chest X-ray findings suggestive of ILD and a restrictive pattern on pulmonary function testing (defined as a total lung capacity less than the lower limit of normal) was considered diagnostic of ILD. Evaluated risk factors included age, sex, calendar year, smoking status, body mass index (BMI) and presence/absence of rheumatoid factor (RF) and anti-citrullinated protein antibodies (ACPA). Cumulative incidence of ILD was adjusted for the competing risk of death. Cox models were used to assess the association between potential risk factors and the development of RA-ILD.Results:In Olmsted County, 645 residents were diagnosed with RA between 1999 and 2014. Seventy percent of patients were females, and 30% were males; median age at RA diagnosis was 55.3 [IQR 44.1-66.6] years, and most patients (89%) were white. Fifty-three percent of patients were never-smokers, and 64% had seropositive RA. Forty percent were obese (i.e., BMI ≥30 kg/m2); median BMI was 28.3 [IQR 24.3-33.0] kg/m2.In the cohort, ILD was identified in 73 patients. The ILD diagnosis predated RA diagnosis in 22 patients (3.4%) who were excluded from subsequent analyses. Final analyses included the remaining 623 patients with no ILD preceding, or at the time of RA diagnosis. Over a median follow-up interval of 10.2 [IQR 6.5-14.3] years, 51 patients developed ILD. Cumulative incidence of ILD, adjusted for the competing risk of death, was 4.3% at 5 years; 7.8% at 10 years; 9.4% at 15 years; and 12.3% at 20 years after RA diagnosis (Figure 1).Age, and history of smoking at RA diagnosis correlated with the incidence of ILD; adjusted hazard ratios (HRs) were 1.89 per 10-year increase in age (95% confidence interval 1.52-2.34) and 1.94 (95% confidence interval 1.10-3.42), respectively. On the other hand, sex (HR: 1.21; 95% CI: 0.68-2.17), BMI (HR: 0.99; 95% CI: 0.95-1.04), obesity (HR: 0.89; 95% CI: 0.50-1.58), and seropositivity (HR: 1.15; 95% CI: 0.65-2.03) did not demonstrate significant associations with ILD.Conclusion:This study provides a contemporary estimate of the occurrence of ILD in a well-characterized population-based cohort of patients with RA. Our findings of a lack of association between sex, obesity and seropositivity with ILD may indicate a change in established risk factors for ILD and warrant further investigation.References:[1]Shaw M, Collins BF, Ho LA, Raghu G. Rheumatoid arthritis-associated lung disease. Eur Respir Rev. 2015;24(135):1-16. doi:10.1183/09059180.00008014[2]Bongartz T, Nannini C, Medina-Velasquez YF, et al. Incidence and mortality of interstitial lung disease in rheumatoid arthritis - A population-based study. Arthritis Rheum. 2010;62(6):1583-1591. doi:10.1002/art.27405[3]Aletaha D, Neogi T, Silman AJ, et al. 2010 Rheumatoid arthritis classification criteria: An American College of Rheumatology/European League Against Rheumatism collaborative initiative. Arthritis Rheum. 2010;62(9):2569-2581. doi:10.1002/art.27584Figure 1.Cumulative incidence of ILD in patients diagnosed with RA between 1999 and 2014, adjusted for the competing risk of death. Abbreviations. ILD: interstitial lung disease; RA: rheumatoid arthritis.Disclosure of Interests:Bilal Samhouri: None declared, Robert Vassallo Grant/research support from: Research grants from Pfizer, Sun Pharmaceuticals and Bristol Myers Squibb, Sara Achenbach: None declared, Vanessa Kronzer: None declared, John M Davis III Grant/research support from: Research grant from Pfizer., Elena Myasoedova: None declared, Cynthia S. Crowson: None declared
Estimating the correlation between semi‐competing risk survival endpoints
