Improving the clinical utility of antinuclear antibody reporting by incorporating likelihood ratios in interpretative comments

Abstract

Background The interpretation of antinuclear antibody (ANA) test results is complex with current reporting adding little useful clinical information. Aims To improve the quality of ANA interpretative comments through the use of likelihood ratios (LRs) at different ANA titres for diagnosing systemic rheumatic diseases. Currently, no other laboratory in Australia implements this tool. Methods A total of 369 consecutive blood samples referred for ANA testing over a 1-week period were simultaneously assessed for extractable nuclear antigens (ENA) and dsDNA. The ANA antibodies were screened using Hep 2000 (Immunoconcepts). ENA antibodies were screened using an in-house counterimmunoelectrophoresis (CIE) method and characterised by FIDIS (Biomedical Diagnostic). Anti-dsDNA antibodies were measured by the FARR assay (Amerlex commercial kits). All commercial assays were performed according to the manufacturers’ instructions. Results Sixty-three patients (17%) were positive for ANA testing, 51(81%) female and 12 (19%) male. ANA testing was positive in 18.3% (37/202) of the age group ≥50 years and in 15.5% (26/167) of the age group Conclusions 1. A more effective method of interpreting ANA results is through LR which increases significantly with ANA positivity. The LR contrasts the proportion of individuals with and without systemic rheumatic diseases at a given level of ANA. It provides an independent measure of the post-test probability with the pre-test probability. 2. No significant difference in ANA and ENA positivity between both age groups ( 3. The correlation between ANA and ENA was more significant in nucleolar, centromere, PCNA and SSA patterns than in speckled and homogenous patterns.