ABSTRACT
BACKGROUND: Automated systems for antinuclear antibody analysis are being introduced. The aim was to evaluate whether automated quantitative reading of fluorescence intensity is clinically relevant and allows for value-added reporting of test results. METHODS: Consecutive samples (n=260) were used to correlate fluorescence intensity with end-point titer. Moreover, 434 samples from controls (150 healthy blood donors, 150 chronic fatigue syndrome, and 134 diseased controls) and 252 samples (obtained at diagnosis) from patients with systemic rheumatic diseases were screened for antinuclear antibodies (1:80) on HEp-2 cells using NOVA View, and likelihood ratios were calculated for fluorescence intensity result intervals. RESULTS: There was a significant correlation between end-point titer and fluorescence intensity. Likelihood ratios for a systemic rheumatic disease increased with increasing fluorescence intensity. The likelihood ratio for a systemic rheumatic disease was 0.06, 0.18, 0.51, 5.3, and 37.5 for a fluorescence intensity of ≤66, 67-150, 151-300, 301-1000, >1000, respectively. A range of 31%-37% of the patients with Sjögren's syndrome, systemic sclerosis or systemic lupus erythematosus had fluorescence intensities >1000. CONCLUSIONS: Estimation of fluorescence intensity by automated antinuclear antibody analysis offers clinically useful information. Likelihood ratios based on fluorescence intensity test result intervals aid with the interpretation of automated antinuclear antibody analysis and allow value-added reporting.
