[A comparative study of antinuclear antibody measurement using different cells as nuclear substrates]

In an attempt to evaluate possible variations in the antinuclear antibody (ANA) titers resulting from nuclear substrate difference, we measured the ANA titer of 104 normal subjects using substrates from mouse liver cell (MLC) and HEp-2 cell (CSI, USA). We also performed similar experiments on 50 sera from 50 patients with rheumatic disease using mouse liver cell, HEp-2 cell (CSI), HEp-2 cell (AFT, Japan), and HEp-2 cell (AI, USA) as substrates. The ANA titer assay using HEp-2 cell (CSI), which is routinely used in our laboratory, was regarded as the reference method. The results showed that when HEp-2 (CSI) was used, 98.1% normal subjects had ANA titer lower than 1:320, therefore a cut-off value of 1:320 was used. While using the MLC method, 96.2% normal subjects had ANA titer lower than 1:80, and the cut-off value was set at 1:80 accordingly. The geometric mean ANA titers of the 50 sera from 50 rheumatic disease patients were 676 for HEp-2 cell (CSI), 503 for HEp-2 cell (AFT), 368 for HEp-2 cell (AI), and 265 for MLC, respectively. Although statistically significant difference (p less than 0.05) in geometric mean titer existed between reference method and the other 3 methods, the latter 3 methods still had good sensitivity (97.2% for HEp-2 cell (both AFT and AI); 91.7% for MLC), and specificity (100% for HEp-2 cell (both AFT and AI); 71.4% for MLC) for selecting out sera with abnormal ANA titer. We also evaluated the relationships between ANA titer and anti-double stranded DNA antibody (anti-dsDNA) concentration in another 210 sera from 151 patients with rheumatic disease. Statistically, there is no distinct correlation between anti-dsDNA and ANA of all fluorescence staining patterns except for peripheral pattern, in which case a significant correlation could be demonstrated.