Ten-Year Analysis of Low-Dose Glucocorticoids in early RA Save

Abstract #1998 by C. Roubille et al from Montpellier, France is a 10-year prospective analysis that looked at the risk of severe outcomes related to very low-dose glucocorticoid (GC) in early rheumatoid arthritis (RA). The long-term risk of adverse events of low-dose GC treatment is still controversial; thus, this study aimed to look at the harm of very low- dose steroids in RA by analyzing data from the ESPOIR cohort. 

The ESPOIR cohort (Etude et Suivi des POlyarthrites Indifferenciees Recentes) is a French, prospective, multicenter, observational cohort of 813 patients between 2002 and 2005 that are naïve to DMARDs or GC and consisted of data locked from the 10-year time point. In this study, patients were stratified into 2 groups—with or without GC treatment—at least once during a median of 10-year follow-up. 

Among the 608 patients that were included in the study with a mean age of ~48 years and mostly women, 397 of the 608 pts (or about 65%) took at least one systemic GC treatment with a mean dose of 2.8 mg/day and median dose of 1.9 mg/day between t0 and end of follow-up. The cumulative GC dose of 8,468 mg was mainly during the first 6 months in 70% of the cohort, and the mean duration of GC treatment was ~45 months.  Notably, the patients in the GC cohort relative to without GC had more active disease (higher DAS28-CRP, higher HAQ), greater consumption of DMARDs, biologic agents, or NSAIDs, and higher CRP & ACPA levels. 

At 10 years, patients taking GCs experienced significantly more events that those without GC. Overall, 95 events were identified: 10 deaths, 18 cardiovascular disease (CV) events (which included MI, stroke, and HF), 32 fractures, and 35 severe infections. Also, there was a significant cumulative dose effect, particularly for the risk of severe infections. The risk of events per time was significantly associated with GC treatment, age, hx of HTN, and ESR. Hazard ratio in year 1 of GC use was 0.46, which increased to 6.83 by year 10. The authors used statistical methods to help reduce the impact of treatment selection bias and confounding variables. 

In summary, the 10-year analysis of the ESPOIR cohort showed a dose and time-dependent impact of very low-dose GC treatment in early RA, with a long-term high risk of severe outcomes including death, CV events, fractures, and severe infections.