Information on diagnosis and management of systemic lupus erythematosus derived from the routine measurement of 8 nuclear autoantibodies

• Published in: Journal of rheumatology Vol. 30; no. 8; p. 1761
• Main Authors: Ignat, Gheorghe Paul, Rat, Anne-Christine, Sychra, Jerry J, Vo, Jacqueline, Varga, John, Teodorescu, Marius
• Format: Journal Article
• Language: English
• Published: Canada 01.08.2003
• Subjects: Antibodies, Antinuclear - blood; Cohort Studies; DNA, Single-Stranded - immunology; Female; Humans; Linear Models; Longitudinal Studies; Lupus Erythematosus, Systemic - diagnosis; Lupus Erythematosus, Systemic - immunology; Lupus Erythematosus, Systemic - therapy; Male; Predictive Value of Tests; Reproducibility of Results; Retrospective Studies; Sensitivity and Specificity; Software
• ISSN: 0315-162X

To determine the value of routine measurement of a panel of 8 nuclear autoantibodies (ANA/8) for the diagnosis and management of patients with systemic lupus erythematosus (SLE). To estimate disease sensitivity of ANA/8, we studied 25 patients with new SLE and 114 with new and established SLE. To estimate disease specificity, 100 patients with other autoimmune rheumatic diseases were included. We used computerized statistical analysis of the level of 8 ANA in relation to clinical activity determined as Systemic Lupus Activity Measure disease activity scores (DAS). Data were collected retrospectively from the charts of 114 patients with 698 visits and evaluated by multiple and piece-wise linear regression analysis (PWLRA) and correlation and cluster analyses. The disease sensitivity of the 3 types of SLE profiles identified was 100% for new SLE patients (n = 25) and 87% for mixed SLE patients; the disease specificity was 98%. Autoantibody levels of anti-ssDNA, dsDNA, and Scl-70 were the best individual correlates of general and organ-specific DAS. Twenty-four percent (R2) of the variability in the general DAS was explained by the multiple regression (R = 0.49), with significant contribution made by anti-Scl-70 (beta = 0.39), dsDNA (beta = 0.17), Sm (beta = 0.10), and SSA (beta = 0.08). PWLRA indicated that for 68% of the 698 clinical presentations (average 6/patient), the observed DAS and the predicted DAS from autoantibody levels were both low and clustered; they were partially discrepant for the remaining 32%, which was explained by the relatively high correlation of DAS with prior changes in autoantibody levels (R = 0.6). The changes in DAS and in anti-dsDNA levels were significantly predicted by the multiple regression at one prior visit, with anti-ssDNA as the main contributor. The ANA/8 profile showed ~ 100% sensitivity and ~ 98% specificity for SLE and correlated with contemporary and subsequent changes in DAS and autoantibody levels. Among autoantibodies of this profile, anti-ssDNA (ssDNA) was the most sensitive indicator of SLE and the main contributor to prediction of subsequent changes in DAS.