Background: Antinuclear autoantibody determination relies on an initial screening step using immunofluorescence on HEp2 cells. This may be followed by anti-deoxyribonucleic acid (DNA) determination, if the fluorescence of nuclei is homogeneous. We assessed the validity of a restricted algorithm and compared this to a more comprehensive algorithm that accepted any nuclear pattern as a positive indicator.
Methods: Our retrospective study analyzed routine results for antinuclear antibodies (ANA) and their anti-DNA identification [double stranded nuclear DNA (ds-DNA), membrane associated DNA (mDNA), nucleosomes] for 58 systemic lupus erythematosus (SLE) patients (690 sera). We included 158 patients with systemic or organ-specific autoimmune diseases (888 sera), 44 with viral disease (88 sera), 34 cancer patients (89 sera) and 111 patients with inflammation that served as controls (122 sera) for a total of 1187 samples.
Results: 1) Anti DNA antibodies are not associated only with a homogeneous pattern, but can also be seen with a speckled or nucleolar pattern. 2) The observed pattern is typical for a particular patient rather than for a specific pathology. 3) A homogeneous pattern does not necessarily indicate SLE, nor does the presence of circulating anti DNA antibodies. 4) Determination of various specificities of anti DNA antibodies, whatever the immunofluorescent pattern, increases the sensitivity and specificity for SLE.
Conclusions: If diagnosis is based exclusively on a homogenous pattern, preselection would have missed identification of SLE as high levels of anti DNA antibodies were also associated with speckled or nucleolar pattern.