How should lupus flares be measured? Deconstruction of the Safety of Estrogen in Lupus Erythematosus National Assessment-Systemic Lupus Erythematosus Disease Activity Index flare index

Objective. Accurate assessment of lupus flares is critical but problematic in clinical trials. This study examined the impact of modifications to the classic Safety of Estrogens in Lupus Erythematosus National Assessment (SELENA)-SLEDAI flare index (cSFI). Methods. Ninety-one SLE patient records were evaluated at two visits at which the SLEDAI and BILAG had been scored prospectively. The cSFI was compared with an experimental version (eSFI) that eliminated medication criteria and separated the mild/moderate flare category into its components by clinical judgement based on records. The revised SFI (SFI-R) and some physician's global assessments (PGAs) were also scored using chart notes. Results. eSFI-rated moderate flares had higher PGA and BILAG scores than those rated as mild. When medication criteria were excluded, 42 of 55 cSFI severe flares and 15 of 49 mild/moderate flares were downgraded in severity. Comparing flares that remained severe with those that were downgraded, disease activity was higher by PGA (P < 0.001), SLEDAI (P < 0.001), BILAG (P < 0.001), number of active BILAG organs (P < 0.04) and flaring SFI-R organs (P < 0.01). PGA (P < 0.001) and the number of SFI-R domains flaring (P < 0.001) were higher in mild/moderate eSFI flares than in those that were downgraded. Twenty-one of 83 (25%) medication changes occurred with no flare. Forty-six of 52 (88%) medication changes defining severe flare by cSFI involved patients rated by physicians with no, mild or moderate flares. Conclusion. A deconstructed flare index improves the discrimination of mild from moderate flares and selects more ill patients with true clinical worsening for each category of flare.