Discordance Between Systemic Lupus Erythematosus Disease Activity Index Domain Weights and Their Association With...

Studies of damage accrual in patients with systemic lupus erythematosus (SLE) show associations with disease activity measured by the SLE Disease Activity Index 2000 (SLEDAI-2K), but these associations are imperfect. SLEDAI scores are powerfully influenced by weightings (1-8) assigned to each domain. We investigated whether the weightings of SLEDAI-2K domain scores aligned with the risk of organ damage accrual.

In patients with SLE attending a single center from 2007 to 2016, disease activity (SLEDAI-2K) and organ damage (Systemic Lupus International Collaborating Clinics/American College of Rheumatology damage index; SDI) were recorded prospectively at each visit and annually, respectively. Visits were dichotomized according to whether the visit occurred during a period of damage transition (increase in SDI in the same year). Associations were assessed using Fisher exact test (univariable analysis) and generalized linear mixed-effect models (multivariable analysis).

We analyzed 5,538 visits of 266 patients (86.5% female, 51.5% White). SLEDAI-2K domains positively associated with damage transition all had weights of ≤4 (odds ratio [95% confidence interval]: pericarditis, 5.39 [1.59-18.3] P < 0.01; pleurisy, 2.12 [1.14-3.94] P = 0.02; arthritis, 1.40 [1.02-1.94] P = 0.04; and leukopenia, 1.52 [1.04-2.24] P = 0.03). Higher-weighted SLEDAI-2K domains were negatively associated with damage transition, not associated, or were too infrequent to assess.

Associations of SLEDAI-2K domains with damage accrual were variable and were discordant with domain weightings. Alternative weighting strategies for disease activity instruments could improve their prognostic utility.