(P-525) Mycophenolate and Azathioprine Are Associated with Risk of Hospitalized Viral Respiratory Infection in Rheumatoid Arthritis and Systemic Lupus

Background: Rheumatoid arthritis (RA) and systemic lupus erythematosus (SLE) patients may be at a higher risk of severe infections because of their underlying condition and/or use of immunosuppressive medications. More recently, in the context of COVID-19 pandemic, there was an increased interest in the antiviral properties of some drugs used to treat these conditions.

Objectives: To evaluate the risk factors for hospitalization for viral respiratory infections among RA and SLE patients.

Methods: We studied adult RA and SLE individuals identified in the MarketScan database (2011-2018) with an outpatient diagnosis (including emergency room, ER) of viral pneumonia or other viral respiratory infections. We required patients to be covered in the medical/pharmacy plan one year before time zero (date of outpatient infection). The outcome was hospitalization for viral infection within the 30 days after time zero. We used multivariate Poisson regression models to assess risk factors associated with the outcome, including recent use (in the 90 days prior to time zero) of relevant medications: hydroxychloroquine (HQN) or chloroquine (CQ), methotrexate, mycophenolate (MMF), azathioprine (AZA), other immunosuppressants (cyclophosphamide, sulfasalazine, and leflunomide), corticosteroids, biologics, and NSAIDs, age, sex, setting in which viral infection was initially identified (ER or other), underlying condition (RA or SLE), and comorbidities.

Results: We identified 63,971 episodes of outpatient viral respiratory infections among 54,561 RA/SLE patients (80% female, average age 57.4 years, standard deviation 13.9). During the 30-day period following outpatient infection, we found 480 occurrences of hospitalization for viral respiratory infections. In adjusted multivariate analyses, use of MMF (adjusted RR, aRR 2.54, 95% CI: 1.51–4.28), AZA (aRR 1.90, 95% CI: 1.12–3.22) and corticosteroids (aRR 1.57, 95% CI: 1.29–1.91) were significantly associated with the risk of hospitalized viral respiratory infections. HQN/CQ was not associated with the outcome studied (aRR 0.89, 95% CI: 0.702–1.12). In the same model, comorbidities, ER presentation, and older age were also significantly associated with hospitalized viral infection.

Conclusions: Among RA/SLE patients with an outpatient viral infection, MMF and AZA, as well as corticosteroids, comorbidities, ER presentation and older age, were all significantly associated with subsequent need for hospitalization. Our results for HCQ/CQ are consistent with recent clinical trials showing no protective effect of these drugs to reduce the risk of severe COVID-19.