fetching data ... 
Background: We previously conducted a randomized controlled trial (RCT) comparing glucocorticoid (GC) withdrawal with immunosuppressant (IS) withdrawal in Systemic Lupus Erythematosus (SLE) patients in sustained remission. The study demonstrated that IS withdrawal is noninferior to GC withdrawal in terms of flare rates at 52 and 104 weeks. Despite various studies on drug withdrawal in SLE, data on outcomes of flares following withdrawal remain limited.
Objectives: This study evaluated the outcomes of flares and identified predictors of multiple flares following GC or IS withdrawal in SLE.
Methods: We conducted a follow-up study of our RCT comparing GC and IS withdrawal in SLE patients in sustained remission >1 year. In the trial, 117 patients were randomized 1:1 to GC (n=58) or IS (n=59) withdrawal. Eligibility included age >18 years, stable GC dose (≤7.5 mg/day prednisolone), IS maintenance ≥3 years, and remission ≥1 year. This study focused on patients experiencing flares, classified using the SELENA-SLEDAI Flare Index (SFI), until the last available follow-up. Flare outcomes included hospitalization, renal biopsy, infections, cumulative steroid doses, excess steroids for flare management, and damage accrual (SDI). Predictors of multiple flares were analyzed using binary logistic regression.
Results: After a median follow-up of 40 months, 49.1% of the GC group and 61.4% of the IS group remained flare-free. Hospitalizations were higher in the GC group (9 vs. 4). Other outcomes, including renal biopsy (p=0.32), remission rates (p=0.57), cumulative prednisolone dose (p=0.690), excess prednisolone dose (p=0.476), SDI increase (p=0.433), infections (p=0.906), and deaths, were similar (Table 1). Infections were more frequent in the flare group (OR 6.15, 95% CI: 1.63–23.12, p=0.003). Patients with flares required higher excess prednisolone (p=0.003), especially with multiple flares (p<0.05). Those with multiple flares had higher renal biopsy rates (p=0.04) and prednisolone use (p<0.05). Remission rates (p=0.98), SDI increase (p=0.304), and infection rates were comparable. Male gender (OR 8.01, 95% CI: 1.07–59.74, p=0.04) and low baseline C4 (OR 5.76, 95% CI: 1.33–24.83, p=0.02) independently predicted multiple flares (Table 2).
Conclusion: Although drug withdrawal is vital in SLE remission, determining which drug to withdraw and in whom remains complex. These findings underscore the need to challenge conventional practices and prioritize personalized, patient-centered strategies.
Outcomes of flare following glucocorticoid and immunosuppressant withdrawal
| Outcomes of flares | GC withdrawal (n=29) | IS withdrawal (n=22) | P value |
|---|---|---|---|
| Duration of follow up, months, median (range) | 40 (29-43) | 40 (28-43) | - |
| No. of hospitalisations for flare, n | 9 | 4 | - |
| Single flares, n(%)
| 23 (65.7)
| 12 (34.3)
| 0.093 |
| Need for renal biopsy, (n%) | 10 (35.7) | 10 (50) | 0.322 |
| Remission, (n%) | 28 (96.6) | 20 (90.9) | 0.57 |
| Cumulative prednisolone dose (g), median (range) | 17.75 (6-50.7) | 17.8 (10.9-55.9) | 0.690 |
| Excess prednisolone dose after flare (g), median (range) | 1.25 (0-39) | 2.48 (0-6) | 0.476 |
| Increase in SDI, n(%) | 8 (27.6) | 4 (18.2) | 0.433 |
| Infections, n(%) | 7 (24.1) | 5 (22.7) | 0.906 |
| Deaths, n(%) | 0 | 1(0.04)* | - |
| Lost to follow up/default/protocol violation, n(%) | 10 | 5 | - |
| Variable | Flares (n=51 ) | No flares (n=63 ) | P value |
| Increase in SDI, n(%) | 12 (23.5) | 17 (27) | 0.674 |
| Cumulative prednisolone dose (g), median (range) | 17.8 (6-55.9) | 16.1 (2.7-37.9)
| 0.247 |
| Excess prednisolone dose for flare (g), median (range) | 2.38 (0-39) | 0.27 (0-23) | 0.006 |
| Infections, n(%) | 12 (23.5) | 3 (4.8) | 0.003 |
SFI, SELENA-SLEDAI flare index, SDI, SLICC/ACR Damage Index, *death due to sepsis
Comparison of single versus multiple flares, and predictors of multiple flares following drug withdrawal
| Variable | Single flares (n=35) | Multiple flares (n=17) | P value | ||||
|---|---|---|---|---|---|---|---|
| No. of hospitalisations for flare, n | 1 (20) | 7 (63.6) | 0.110 | ||||
| Need for renal biopsy, n(%) | 9 (29) | 10 (58.8) | 0.040 | ||||
| Remission, n(%) | 31 (93.9) | 16 (94.1) | 0.980 | ||||
| Cumulative prednisolone dose (g), median (range) | 17.45 (6-50) | 21 (10.9-55.9) | 0.150 | ||||
| Excess prednisolone dose after flare (g), median (range) | 0.64 (0-24) | 4 (2-39) | <0.050 | ||||
| Increase in SDI, n (%) | 9 (25.7) | 2 (11.8) | - | ||||
| Infections, n(%) | 8 (22.9) | 4 (23.5) | 0.304 | ||||
| Predictors of multiple flares | Univariate analysis | Multivariate analysis | |||||
| Variable | OR | 95%CI | P value | OR | 95%CI | P value | |
| IS withdrawal group | 1.62 | 0.61-4.29 | 0.33 | 1.56 | 0.47-5.16 | 0.46 | |
| Age at enrolment (years) | 0.99 | 0.85-1.05 | 0.96 | 1.01 | 0.95-1.07 | 0.70 | |
| Gender (male) | 7.5 | 1.17-48.15 | 0.03 | 8.01 | 1.07-59.74 | 0.04 | |
| Disease duration (months) | 0.99 | 0.98-1.01 | 0.84 | 0.99 | 0.98-1.01 | 0.85 | |
| Low C3 at baseline | 3.72 | 1.22-11.34 | 0.02 | 2.35 | 0.47-5.16 | 0.46 | |
| Low C4 at baseline | 7.79 | 2.23-27.01 | <0.01 | 5.76 | 1.33-24.83 | 0.02 | |
IS, Immunosuppressant
REFERENCES: NIL.
Acknowledgements: The authors thank Mrs Kalaichelvi Vigneswaran and Mrs Rubini Vijayan for their assistance with patient scheduling and follow-up.
Disclosure of Interests: None declared.
© The Authors 2025. This abstract is an open access article published in Annals of Rheumatic Diseases under the CC BY-NC-ND license (