Seropositivity Needs to Be Critically Considered

Laass and colleagues recently reported in Deutsches Ärzteblatt International a prevalence of celiac disease of 0.9% in children and adolescents in Germany (1). This prevalence was deduced from the results of antibody measurements in more than 12 000 serum specimens by using a test whose origin was not reported. The conclusion of a celiac disease diagnosis was drawn from seropositivity. The authors assumed a specificity of 95% for their antibody test. However, even if the specificity is much higher (99.5%) the test will yield a number of false positives (n=60 to 65) that will be within the range of cases of seropositive children as reported by the authors (n=98). In the discussion section of the article, the authors attempted to assess the proportions of false positives and false negatives, but this seems speculative—among other reasons, because the comparison was made with results that—as the authors reported—were obtained by using a different testing system (2). The prevalence of celiac disease as reported in the article is consistent with study results from other countries, but deducing a prevalence of celiac disease from seropositivity should be critically considered. The calculated prevalence may represent an upper limit. The lower limit would then result from the number of 47 children whose antibody concentration was above the 10-fold of the cut-off value suggested by the test manufacturer, plus the 8 children with a clinical history. This amounts to a total of 55 children or a prevalence of celiac disease of 0.43%.