Incident Rheumatoid Arthritis in HIV Infection: Epidemiology and Treatment

Abstract

Objective: To assess the incidence, presentation, and management of rheumatoid arthritis (RA) in patients with HIV, including the use of disease-modifying antirheumatic drugs (DMARDs) in this immunosuppressed population.

Methods: Patients included in this study were from the Veterans Aging Cohort Study, a longitudinal cohort of veterans with HIV and age-, race-, and site-matched uninfected veterans. We identified all patients who had ≥1 rheumatologist-generated International Classification of Diseases, Ninth Revision (ICD-9) or Tenth Revision (ICD-10) code for RA and whose serum samples were tested for rheumatoid factor (RF) and anti-cyclic citrullinated peptide (anti-CCP) antibodies. To further confirm the diagnosis of RA, medical charts were reviewed to verify whether patients met the American College of Rheumatology/European Alliance of Associations for Rheumatology 2010 criteria for RA. We recorded DMARD use and adverse effects during the first contiguous course of treatment (i.e., >6 months of no interruption in DMARD treatment).

Results: This study included 56,250 patients with HIV and 116,944 uninfected individuals over 2,384,541 person-years. Of the 2,748 individuals in this cohort who were reviewed for a diagnosis of RA based on ICD-9 or ICD-10 codes, incident RA was identified in 215 individuals, including 21 patients with HIV. The incidence rate ratio of RA for patients with HIV compared to uninfected individuals was 0.29 (95% confidence interval 0.19-0.48). Most of the patients diagnosed as having RA (88%) were seropositive for RA-associated autoantibodies (RF and/or anti-CCP). However, high autoantibody titers were less frequent in RA patients with HIV compared to RA patients without HIV. In total, 5% of RA patients with HIV (1 of 21) had both high titers of anti-CCP and high titers of RF, compared to 41% of uninfected individuals (81 of 194). DMARDs were prescribed in 71% of RA patients with HIV (15 of 21) compared to 94% of RA patients without HIV (183 of 194). There was no indication that the DMARD safety profile was worse among RA patients with HIV who were prescribed DMARDs (n = 10 assessed) compared to RA patients without HIV who were prescribed DMARDs (n = 158 assessed).

Conclusion: In this cohort, incident RA was less common in patients with HIV compared to uninfected individuals. Moreover, compared to RA patients without HIV, the seropositivity rate and titers of RA-specific autoantibodies were lower among RA patients with HIV, and those with HIV were prescribed DMARDs less frequently than those without HIV.