IMPACT of INTERSTITIAL LUNG DISEASE on SEVERE COVID-19 OUTCOMES for PATIENTS with RHEUMATOID ARTHRITIS: A MULTICENTER STUDY

ABSTRACT

Background: RA has been associated with poor COVID-19 outcomes, but few studies have investigated outcomes in RA features such as interstitial lung disease. Objectives: To assess COVID-19 outcomes in patients with RA overall, and those with and without ILD, compared to general population comparators. Methods: A multicenter, retrospective cohort study was conducted at Mayo Clinic (19 hospitals and affiliated outpatient centers in 4 states) and Mass General Brigham (14 hospitals and affiliated outpatient centers in New England). Consecutive patients with RA meeting ACR/EULAR criteria and a positive COVID-19 test from March 1, 2020 through June 6, 2021 were matched 15 on age, sex, race, and COVID-19 test date with general population comparators without RA. RA features assessed included RA-ILD per Bongartz criteria [1], duration, rheumatoid factor (RF), cyclic citrullinated peptide antibody (CCP), bone erosions, and treatments. The primary outcome was a composite of hospitalization or death following COVID-19 diagnosis. We used multivariable Cox regression to investigate the association of RA, and features such as ILD, with COVID-19 outcomes compared to matched comparators. Results: We analyzed 582 patients with RA and 2892 comparators without RA, all with COVID-19. Mean age was 62 years, 51% were female, and 79% were White. Mean RA duration was 11 years, 67% were seropositive (52% RF+ and 54% CCP+), 27% had bone erosions, 28% were on steroids, and 79% were on DMARDs. 50/582 (9%) patients with RA had ILD. The COVID-19 hospitalization or death rate for RA patients was higher than comparators (3.0 per 1,000 days [95% CI 2.5-3.6] vs. 1.9 per 1,000 days [95% CI 1.7-2.1], respectively). Overall, RA patients had a 53% higher risk of hospitalization or death than comparators after adjustment (95% CI 1.20-1.94). Among those with RA-ILD, the hospitalization or death rate was signifcantly higher than comparators (10.9 [95% CI 6.7-15.2] vs. 2.5 per 1,000 days [1.8-3.2], respectively). RA-ILD was associated with nearly 3-fold higher risk for hospitalization or death than comparators (multivariable HR 2.84 [95% CI 1.64-4.91], Table 1). There was a signifcant interaction between RA/comparator status and presence/absence of ILD for risk of severe COVID-19 (p<0.001, Figure 1). The elevated risk for severe COVID-19 was similar for RA subgroups defned by serostatus or bone erosions. Conclusion: We confrmed that RA was associated with severe COVID-19 outcomes compared to the general population. We found evidence that ILD may be an effect modifer for the relationship between RA and severe COVID-19 outcomes, but RA subgroups defned by serostatus and bone erosions had similarly elevated risk. These fndings suggest that ILD or its treatment may be a major contributor to severe COVID-19 outcomes in RA.