Defining the physician global assessment threshold equivalent to remission in patients with systemic lupus erythematosus.

Abstract

The optimal threshold of the physician global assessment (PGA) for remission in SLE has never been evaluated systematically. The aim of this study was to assess the ideal PGA threshold associated with physician remission and to investigate its impact on remission rates in our Lupus cohort. In this monocentric cross-sectional study, patients with SLE were evaluated for physician remission by asking the treating physicians if they considered their patient in remission regardless of objective remission criteria. Further, two objective remission definitions were applied: 1) DORIS remission using a PGA of < 2 (0-10) (corresponding to < 0.5 on a VAS 0-3 used in DORIS); 2) DORIS remission with omission of PGA (modDORIS). A receiver operating characteristic analysis and regression analyses were performed to assess the ideal PGA threshold and factors influencing PGA. Of the 233 patients included, 126 patients (54.0%) were in physician remission, 42.5% in DORIS remission and 67.0% in modDORIS remission. A PGA of < 2 NRS 0-10) had the highest sensitivity (79%) and specificity (81%) for physician remission and modDORIS (AUC 0.85 and 0.69). PGA of patients fulfilling any of the remission definitions was associated with pain and hypocomplementemia. Damage was numerically higher in patients in modDORIS only; no association between PGA and damage was found in regression analysis. Using a PGA threshold of < 2 (0-10), corresponding to < 0.6 (0-3) resulted in best prediction of physician remission. PGA levels seem to be influenced by pain and complement levels but not disease damage.