Background/Purpose: Glucocorticoids (GC) are rapidly effective in suppressing disease activity in rheumatoid arthritis (RA), and are often used as ‘bridging’ therapy while slower acting conventional synthetic (cs)DMARD are initiated. However, discontinuation of bridging GC carries a risk for disease flares if csDMARDs are not (yet) (as) effective. If a flare occurs, the options are to restart GC and thereby risk GC related toxicity, or to switch to other therapies, for instance a biological (b)DMARD. To evaluate if initially starting GC in real life practice leads to a higher risk of GC and bDMARD use in the long term, this study looked at data from the electronical health records (EHR) of patients newly diagnosed with RA.

Methods: Data from 1404 newly diagnosed RA patients were available from the EHR of the Leiden University Medical Center in the Netherlands. Treatment was non-protocolized and reflected daily practice. Patients were included in this analysis if their first visit took place in or after the year 2000 (because of availability of biological (b)DMARDs) and if they had at least 3 follow-up visits in 2 years. The patients who started and who did not start with GC as initial treatment were compared in terms of GC use within 2 years, bDMARD use within 2 years and duration until first bDMARD course using regression models and cox proportional hazard models. Analyses were adjusted for age, gender, year of first visit, ACPA status and disease activity score (DAS28(3) based on 3 components) at baseline.

Results: 1240/1404 patients fulfilled the inclusion criteria of this study. Fifty-one percent of these patients started GC as initial treatment (iGC group) which was combined with methotrexate in 56% of these patients. Eventually 59% of all patients used GC at least once during their first 2 years of treatment. Four percent of the iGC group eventually used a bDMARD in 2 years of follow-up compared to 1% of the niGC group. Patients who did not start GC as initial treatment (niGC group) were younger, more often female and had a lower DAS28(3) compared to patients in the iGC group (table 1), but were comparable in terms of anti-citrullinated protein antibodies (ACPA) and rheumatoid factor (RF) positivity. Whether or not patients were treated with initial GC was not associated with an increased risk to use GC later on during the disease course (aOR 0.93 95%CI 0.60; 1.45). However, patients in the iGC group had a higher risk of starting a bDMARD later on than the niGC group (aHR 3.76 (95%CI 1.67; 8.47). The median duration to the first course of a bDMARD was 409 days in the iGC group and 460 days in the niGC group (p=0.79).

Conclusion: In this daily practice cohort of newly diagnosed RA patients, starting with GC bridging as part of the initial treatment was not associated with GC use later in the disease course. However, in adjusted models patients starting GC as initial treatment were more likely to start a bDMARD later on, although this was not earlier in the treatment course compared to patients who did not start GC initially. Although models were adjusted for amongst others DAS28(3), confounding by indication (affecting both initial and later therapy choices) could have played a role in this observational data.

« Back to ACR Convergence 2022

ACR Meeting Abstracts - https://acrabstracts.org/abstract/is-glucocorticoid-bridging-associated-with-later-glucocorticoid-and-biological-dmard-use-in-patients-with-rheumatoid-arthritis/