Ioannis Parodis, Rama Andraos, Matteo Bottai, Christopher Sjöwall
Longer time in LLDAS-5 was robustly associated with less damage accrual. Crude full-cohort analyses suggested a numerically stronger inverse association and slightly better model fit for LLDAS-5, but this apparent advantage was not sustained after multivariable adjustment. These findings support LLDAS-5 as a clinically meaningful, guideline-aligned target, while its incremental prognostic value over conventional LLDAS-7.5 requires further study.
OBJECTIVE: To examine whether a modified Lupus Low Disease Activity State with a prednisone-equivalent threshold ≤5 mg/day (LLDAS-5) is more strongly associated with lower damage accrual than conventional LLDAS using a ≤7.5 mg/day threshold (LLDAS-7.5) in systemic lupus erythematosus (SLE).
METHODS: We analysed 332 patients, including 105 incident cases. For each patient, yearly Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index (SDI) worsening occasions and average proportion of follow-up spent in LLDAS-5 and LLDAS-7.5 were derived. Patient-level Poisson regression with follow-up years as the exposure and Huber's robust variance estimator estimated rate ratios (RRs). Parallel models were fitted for the two definitions. Full-cohort multivariable models adjusted for age, sex, ethnicity, SLE duration, baseline Systemic Lupus Erythematosus Disease Activity Index 2000 and baseline SDI.
RESULTS: In the full cohort, 142/332 patients experienced at least one SDI increase (190 occasions). Crude RRs were 0.61 (95% CI 0.37 to 1.00; p=0.052) for LLDAS-5 and 0.73 (95% CI 0.39 to 1.37; p=0.332) for LLDAS-7.5. Multivariable RRs were 0.56 (95% CI 0.37 to 0.86; p=0.007) and 0.47 (95% CI 0.27 to 0.85; p=0.012), respectively, with similar model fit. Among incident cases, 29/105 patients accrued damage (29 occasions); both LLDAS-5 (RR 0.31, 95% CI 0.11 to 0.89; p=0.029) and LLDAS-7.5 (RR 0.29, 95% CI 0.09 to 0.93; p=0.036) were associated with lower damage accrual.
CONCLUSION: Longer time in LLDAS-5 was robustly associated with less damage accrual. Crude full-cohort analyses suggested a numerically stronger inverse association and slightly better model fit for LLDAS-5, but this apparent advantage was not sustained after multivariable adjustment. These findings support LLDAS-5 as a clinically meaningful, guideline-aligned target, while its incremental prognostic value over conventional LLDAS-7.5 requires further study.