Document resource
Objectives Systemic lupus erythematosus (SLE) is a heterogeneous disease whose management often relies on glucocorticoids (GCs). Concerns about GC-related damage led the 2023 EULAR recommendations to advise maintenance doses 5 mg/day or less and encourage discontinuation when feasible. However, the impact of GC withdrawal in patients in remission remains uncertain.The aim of this study was to assess whether glucocorticoid discontinuation during remission influences flare risk in SLE, using a propensity score–matched cohort.Methods Prospective data from SLE patients followed between 1990 and 2023 were retrospectively analyzed. Remission was defined as clinical SLEDAI-2K=0 while receiving prednisone 5 mg/day or less and stable background therapy. Flares were defined as any increase in clinical SLEDAI-2K>0 and/or therapy escalation. Flare-free remission was the time between remission and flare. Flares on prednisone 5 mg were classified as occurring in on-GC patients; flares after withdrawal or tapering <5 mg as off-GC (intention-to-treat). A propensity score–matched (PSM) 3:1 analysis compared patients who discontinued GCs with those maintaining 5 mg/day or less, matching for clinically relevant variables. Flare-free remission was assessed through Kaplan–Meier and Cox regression; predictors of flares were evaluated using logistic regression.Results After PSM, 350 patients were included (255 off-GC; 95 on-GC). Patients who discontinued GCs had fewer flares (14.3% vs 31.3%, p=0.001). Kaplan–Meier curves showed longer flare-free remission in off-GC patients (p<0.001, figure 1). At Cox regression, GC continuation was associated with shorter flare-free remission (HR 4.805, p<0.001), and higher cumulative GC dose predicted shorter remission duration. Age at diagnosis, SDI at remission onset and belimumab use were not associated with flare-free survival. Off-GC patients had significantly lower cumulative GC exposure (23.03 g vs 35.19 g, p<0.001), less high-dose exposure (25.9% vs 52.6%, p<0.001), and fewer total (median 2 vs 3) and infectious hospitalizations (0 vs 1). In logistic regression, GC withdrawal did not increase flare risk (OR 1.517, p=0.28), while belimumab use (OR 0.223, p=0.009), longer remission duration and older age at diagnosis were protective.Abstract LBA:01:15 Figure 1Kaplan-Meier curve for flare-free survival in propensity score matched patients off-GC (blue) and on-GC (red)Abstract LBA:01:15 Table 1Propensity score match analysis: univariate analysis comparing matched patients who had flare and who did not. * data available only a portion of patientsConclusions GC discontinuation in SLE patients in remission is not associated with increased flare risk, particularly in those achieving sustained remission. Belimumab use during remission emerged as an additional protective factor.