Validation of the Systemic Lupus Erythematosus Disease Activity Score (SLE-DAS) in Brazilian SLE patients

Abstract

Objectives: To evaluate the measurement properties of SLE-DAS, using the Brazilian-Portuguese version (SLE-DAS Pt-BR).

Methods: Cross-sectional, test-retest, multicenter study, that included patients ≥18 years old who met the classification criteria for SLE. Reliability was tested using type 2.1 intraclass correlation coefficient (ICC2.1), standard measurement error (SEM) and minimum detectable difference (MDD). For construct validity, Spearman correlation (rho) and/or percentage of agreement with chi-square test were used to correlate/associate SLE-DAS Pt-BR with SLEDAI-2K, PGA and the definitions of remission (DORIS) and low disease activity (LLDAS). Performance of SLE-DAS was assessed with PGA as reference standard, calculating sensitivity, specificity and accuracy.

Results: 365 patients were included: mean age 44.5 (SD 14.0) years, 92.9% female. 57.3% were in remission according to SLE-DAS Pt-BR. Excellent intra and inter-rater reliability was observed (ICC2.1= 0.999 and 0.945, respectively), with acceptable SEM/MDD. There was a strong correlation between SLE-DAS Pt-BR with SLEDAI-2K(rho=0.895) and PGA (rho=0.861). Sensitivity, specificity and accuracy of SLE-DAS Pt-BR were above 90% for remission and moderate/severe disease activity and above 80% for mild activity. Most individuals in remission and in moderate/severe activity according to SLEDAI-2K were also in those categories according to SLE-DAS Pt-BR (99.5% and 77.9%, respectively). Considering mild activity concordance was observed in 42.9% patients. Most patients considered in DORIS remission and LLDAS were also in remission and in low disease activity by SLE-DAS Pt-BR.

Conclusion: The present study demonstrated that the SLE-DAS Pt-BR is valid and capable of measuring SLE disease activity.

Keywords: Systemic lupus erythematosus; severity of illness index; translating; validation study.