Background Rheumatoid arthritis (RA) patients have an increased susceptibility to infection.
Objectives 1.To establish the rate of recurrent infection in RA patients recruited to the British Society of Rheumatology Biologics Registry Rheumatoid Arthritis (BSRBR-RA).2.To establish whether the organ class of index infection predicted future serious infection.
Methods The BSRBR-RA is a prospective observational cohort, previously described. Patients with at least one episode of serious infection requiring hospitalisation were included if they occurred whilst on anti-rheumatic drug therapy or within 5 drug half-lives of stopping. Infections were coded by MedDRA classification in to 7 categories. Infections occurring over 14 days after the first index infection were considered as new events. Event rates were calculated and compared using a Cox proportional hazards model with adjustments made for age, gender, disease duration, baseline DAS28 score, smoking status and seropositivity.
Results See Table 1.
In total, 21,943 subjects with 115,423 patient-years follow up were studied, 5365 subjects reported at least one serious infection. Comparing organ classes of prior infection at baseline, each group had comparable age, disease duration, baseline DAS28 and HAQ scores. The cohort characteristics are tabulated. The baseline annual rate of first serious infection was 4.6% (95% CI 4.5–4.7). Following an index infection, the annual rate of serious infection was 12.7% (95% CI 12.1–13.3). Respiratory infections were the most common (41.4% of all events). The system class of index infection was associated with the risk of a recurrent event; subjects who experienced sepsis had the highest risk of subsequent serious infection within 12 months:19.7%. Compared to an index respiratory tract infection, sepsis conferred a 33% increased hazard for recurrent serious infection within a year (HR 1.33, 95% CI 1.01–1.76). Increasing age was a significant predictor of infection recurrence.
Conclusions There is a high risk of recurrent infection in RA patients with past serious infection. Work is ongoing to determine whether organ class of recurrent infection event mirrors index events and the impact of biologic treatment decisions following the index infection.
Disclosure of Interest None declared