Weiru Yuan1, Hua Cao1, Weiping Li2, Xinyi Wu2, Jie Zheng3. 1. Department of Dermatology and Rheumatology, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, 197 Ruijin Er Road, Shanghai, 200025, China. 2. Laboratory of Dermatoimmunology, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China. 3. Department of Dermatology and Rheumatology, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, 197 Ruijin Er Road, Shanghai, 200025, China. jie-zheng2001@126.com.
Abstract
OBJECTIVES: Detection of antinuclear antibodies (ANA) by immunofluorescence assay (IFA) is increasingly substituted by multiplex bead-based immunoassay (MBA) and line-blot immunoassay (LIA). This study is to compare the diagnostic performance of MBA and LIA ANA assays on clinically characterized patient samples. METHODS: A total of 728 serum samples from 385 patients with systemic autoimmune rheumatic diseases (SARD), 204 patients with non-SARD diseases, and 139 apparently healthy subjects were tested with the BioPlex 2200 ANA Screen and EuroLine ANA Profile 3 as the representative MBA and LIA technologies and HEp-2 ANA IFA. Clinical data were collected independent of laboratory analysis and later related to the ANA test results. The clinical diagnostic performances were analyzed using Analyse-it software. RESULTS: The MBA demonstrated higher area under curve (AUC) compared to LIA (0.814 vs 0.761, p = 0.002) and HEp-2 IFA (0.814 vs 0.771, p = 0.008). The MBA and LIA ANA methods showed higher specificity (83.8% and 77.0% vs 67.6%, p < 0.001 and p = 0.005) but lower sensitivity (79.0% and 75.3% vs 86.5%, p < 0.001) compared to HEp-2 IFA. The MBA and LIA ANA revealed substantial to excellent agreements on specific antinuclear antibodies except anti-dsDNA, with the total agreement from 92.3 to 99.9% and Cohen's kappa from 0.71 to 0.98. The MBA demonstrated significantly higher sensitivity (58.1% vs 19.8%, p < 0.001) and comparable specificity (95.9% vs 97.5%, p = 0.221) on anti-dsDNA assay for the diagnosis of SLE compared to LIA. CONCLUSIONS: The MBA and LIA ANA assays have higher specificity but lower sensitivity compared to HEp-2 IFA. There are good agreements between MBA and LIA ANA for the specific antinuclear antibodies except for anti-dsDNA. The MBA ANA demonstrated better assay performance compared to LIA as the MBA possesses higher sensitivity and specificity in the diagnosis of SARD. Key Points • The multiplex bead-based immunoassay (MBA) ANA outperformed line-blot immunoassay (LIA) and traditional HEp-2 IFA. • There are good agreements between the MBA BioPlex 2200 ANA Screen and LIA EuroLine ANA Profile 3 for the most of specific antinuclear antibodies except anti-dsDNA. • Additional anti-dsDNA testing is suggested when EuroLine ANA Profile 3 is used for the aid of SLE diagnosis and management. • The positive predictive value of both multiplex ANA assays can be substantially increased without significantly affecting negative predictive value by using at least two specific antinuclear antibodies for reporting a positive ANA result.
OBJECTIVES: Detection of antinuclear antibodies (ANA) by immunofluorescence assay (IFA) is increasingly substituted by multiplex bead-based immunoassay (MBA) and line-blot immunoassay (LIA). This study is to compare the diagnostic performance of MBA and LIA ANA assays on clinically characterized patient samples. METHODS: A total of 728 serum samples from 385 patients with systemic autoimmune rheumatic diseases (SARD), 204 patients with non-SARD diseases, and 139 apparently healthy subjects were tested with the BioPlex 2200 ANA Screen and EuroLine ANA Profile 3 as the representative MBA and LIA technologies and HEp-2 ANA IFA. Clinical data were collected independent of laboratory analysis and later related to the ANA test results. The clinical diagnostic performances were analyzed using Analyse-it software. RESULTS: The MBA demonstrated higher area under curve (AUC) compared to LIA (0.814 vs 0.761, p = 0.002) and HEp-2 IFA (0.814 vs 0.771, p = 0.008). The MBA and LIA ANA methods showed higher specificity (83.8% and 77.0% vs 67.6%, p < 0.001 and p = 0.005) but lower sensitivity (79.0% and 75.3% vs 86.5%, p < 0.001) compared to HEp-2 IFA. The MBA and LIA ANA revealed substantial to excellent agreements on specific antinuclear antibodies except anti-dsDNA, with the total agreement from 92.3 to 99.9% and Cohen's kappa from 0.71 to 0.98. The MBA demonstrated significantly higher sensitivity (58.1% vs 19.8%, p < 0.001) and comparable specificity (95.9% vs 97.5%, p = 0.221) on anti-dsDNA assay for the diagnosis of SLE compared to LIA. CONCLUSIONS: The MBA and LIA ANA assays have higher specificity but lower sensitivity compared to HEp-2 IFA. There are good agreements between MBA and LIA ANA for the specific antinuclear antibodies except for anti-dsDNA. The MBA ANA demonstrated better assay performance compared to LIA as the MBA possesses higher sensitivity and specificity in the diagnosis of SARD. Key Points • The multiplex bead-based immunoassay (MBA) ANA outperformed line-blot immunoassay (LIA) and traditional HEp-2 IFA. • There are good agreements between the MBA BioPlex 2200 ANA Screen and LIA EuroLine ANA Profile 3 for the most of specific antinuclear antibodies except anti-dsDNA. • Additional anti-dsDNA testing is suggested when EuroLine ANA Profile 3 is used for the aid of SLE diagnosis and management. • The positive predictive value of both multiplex ANA assays can be substantially increased without significantly affecting negative predictive value by using at least two specific antinuclear antibodies for reporting a positive ANA result.
Authors: F J López-Longo; M Rodríguez-Mahou; M Escalona-Monge; C M González; I Monteagudo; L Carreño-Pérez Journal: Lupus Date: 2003 Impact factor: 2.911