Optimization of the diagnostic algorithm of antinuclear antibodies

Introduction: The antinuclear antibody test is a powerful tool for diagnosing rheumatic diseases. Antinuclear antibodies are determined in the laboratory by an algorithm or sequence that starts with a screening test and continues with the identification of the most common antinuclear specificities. But how to interpret the discordant results between the two levels of study of antinuclear antibodies? Objective: To determine the less frequent antinuclear specificities in positive patients of ANA screening and negative of the most common specificities. Methods: A prospective study was done on 88 consecutive patients referred for the routine ANA screening with a positive result of screening by enzyme-linked immunosorbent assay (ELISA) but negative for anti-double-stranded DNA (dc, IgG) and common extractable anti-nuclear antigens (ENAc). The corresponding serum samples were evaluated by indirect immunofluorescence on human laryngeal epidermoid carcinoma cells (IFI-HEp-2) and by ELISA for the individual detection of specific ANA. Results: The ANA test by IFI / HEp-2 was positive in 56/88 (63.6 %) and the antinuclear specificities were detected in 57/88 (64.8 %) samples, in decreasing Anti-Nucs order: 16/88 (18.2 %); anti-centromere (CENP-B): 15/88 (17.0 %); -histona: 15/88 (17 %); -PM / Scl: 13/88 (14.8 %); -ADNsc: 11/88 (12.5 %) and -ENAc individual: 8/88 (9.1 %). The sensitivity of IFI-HEp-2 for antinuclear specificities was 0.83 (95 % CI: 0.72-0.93). No history of rheumatic disease associated with ANA was read in (26/31) 83.9 % patients with negative subserology. Conclusions: The majority of patients with discordant results between the first and second level of ANA were positive of less common antinuclear specificities, but of recognized diagnostic value.