Background We determined the prevalence of anti-nuclear autoantibodies (ANAs) in the German adult population and examined the association between ANAs and cardiovascular and metabolic disorders. Methods We used data and blood samples from the pretest phases of the German National Cohort, obtained from six of the 18 study centers (n = 1199). All centers applied standardized instruments including face-to-face interviews, anthropometric measurements and collection of blood samples. Self-reported histories of diabetes mellitus, heart attack and elevated blood cholesterol and/or lipids were recorded. Height, weight and blood pressure were measured. ANAs were detected using a semi-automated system (AKLIDES®; Medipan GmbH, Dahlewitz, Germany). A positive ANA was defined as a titer ≥ 1:80. ANA were classified as weakly (1:80 or 1:160), moderately (1:320 or 1:640) or strongly (≥1:1280) positive. Specific autoantibodies against nuclear antigens were detected with second-step assays according to the ANA staining pattern. Associations between the assessed disorders and ANA positivity and pattern were examined using sex and age-adjusted mixed-effects logistic regression models. Results Thirty-three percent (95% confidence interval; 31–36%) of the 1196 participants (measurements could not be obtained from three samples) were ANA positive (titer ≥ 1:80). The proportions of weakly, moderately and strongly positive ANA were 29%, 3.3% and 1.3%, respectively. ANA positivity was more common among women than men across all titers (χ2, p = 0.03). ANA positivity, even when stratified according to height of titer or immunofluorescent pattern, was not associated with diabetes, elevated blood cholesterol and/or lipids, obesity or hypertension. Second-step autoantibody assays were positive in 41 of the 83 samples (49%) tested, with anti-DFS70 (n = 13) and anti-dsDNA (n = 7) being most frequent. These subgroups were too small to test for associations with the disorders assessed. Conclusions The prevalence of ANA positivity in the German general population was similar to values reported from other countries. Contrary to other studies, there was no association with selected self-reported and objectively measured cardiovascular and metabolic variables.
