Outcomes of COVID-19 in patients with rheumatoid arthritis: A multicenter research network study in the United States

Abstract

ObjectivesTo investigate outcomes of Coronavirus Disease-2019 (COVID-19) in patients with rheumatoid arthritis (RA) as compared to the general population. Additionally, outcomes were explored among RA patients stratified by sex, race, and medications use through sub-cohort analyses. MethodsThis comparative cohort study used a US multicenter research network (TriNetX) to extract data on all adult RA patients who were diagnosed with COVID-19, and adults without RA who were diagnosed with COVID-19 (comparative cohort) anytime from January 20, 2020 to April 11, 2021. COVID-19 outcomes were assessed within 30 days after its diagnosis. Baseline characteristics that included demographics and comorbidities were controlled in propensity score matching. ResultsA total of 9730 RA patients with COVID-19 and 656,979 non-RA with COVID-19 were identified. Before matching, the risk of all outcomes including mortality (RR: 2.11, 95%CI: 1.90 to 2.34), hospitalization (RR: 1.60, 1.55 to 1.66), intensive care unit-ICU admission (RR: 1.86, 1.71 to 2.05), mechanical ventilation (RR: 1.62, 1.44 to 1.82), severe COVID-19 (RR: 1.89, 1.74 to 2.06), acute kidney injury (RR: 2.13, 1.99 to 2.29), kidney replacement therapy/hemodialysis (RR: 1.40, 1.03 to 1.89), acute respiratory distress syndrome-ARDS (RR: 1.76, 1.53 to 2.02), ischemic stroke (RR: 2.62, 2.24 to 3.07), venous thromboembolism-VTE (RR: 2.30, 2.07 to 2.56), and sepsis (RR: 1.97, 1.81 to 2.13) was higher in RA compared to non-RA. After matching, the risks did not differ in both cohorts except for VTE (RR: 1.18, 1.01 to 1.38) and sepsis (RR: 1.27, 1.12 to 1.43), which were higher in the RA cohort. Male sex, black race, and glucocorticoid use increased the risk of adverse outcomes. The risk of hospitalization was higher in rituximab or interleukin 6 inhibitors (IL-6i) users compared to tumor necrosis factor inhibitors (TNFi) users, with no significant difference between Janus kinase inhibitors (JAKi) or abatacept users and TNFi users. ConclusionThis large cohort study of RA-COVID-19 found that the risk of all outcomes was higher in the RA compared to the non-RA cohort before matching, with no difference in the majority of outcomes after matching, implying the risk being attributed to adjusted factors. However, the risk of VTE and sepsis was higher in RA cohort even after matching, indicating RA as an independent risk factor. Male sex, black race, and glucocorticoid use were associated with adverse outcomes in RA with COVID-19. Rituximab or IL-6i users were associated with an increased risk of hospitalization compared to TNFi users.