A comparative study on the reliability of an automated system for the evaluation of cell-based indirect immunofluorescence

Abstract

Background: Automated interpretations systems for anti-nuclear antibody (ANA), anti-double stranded DNA antibody (dsDNAab), and anti-neutrophil cytoplasmic antibody (ANCA) assessment by indirect immunofluorescence (IIF) have been recently introduced. The aim of this study was to compare the diagnostic performance of the automated IIF reading system AKLIDES with both traditional visual interpretation of IIF by laboratory experts and confirmatory tests.

Methods: Visual and automated autoantibody interpretations of IIF findings using AKLIDES pattern recognition algorithms were performed for ANA on HEp-2 cells (n=182), dsDNAab on Crithidia luciliae (n=44) and ANCA on human neutrophils (n=46). All serum samples tested by IIF for ANCA and dsDNAab were also assessed with the corresponding enzyme-linked immunosorbent assays (ELISAs). Out of the 182 sera tested for ANA by IIF, 116 were also assessed for antibodies to extractable nuclear antigens (ENA) by ELISA and dot immunoassay (DIA).

Results: ANA testing showed an excellent agreement between visual and AKLIDES reading (98.9%). The overall agreement of dsDNAab testing on C. luciliae substrate slides was 91.0%, whereas ANCA showed a concordance of 89.1%. There was a remarkable agreement of AKLIDES findings for dsDNAab with confirmatory tests.

Conclusion: Visual and automated interpretations of IIF findings for ANA, ANCA, and dsDNAab demonstrated a good agreement when assessing patients with suspected autoimmune diseases. Automated interpretation systems such AKLIDES may improve laboratory efficiency and support standardization of IIF in clinical laboratories.