The Prevalence is Only Estimated

The prevalence reported in the study is 0.9% (1). The methods and interpretation of the results prompt a certain amount of criticism, however. The title of the article includes the term “prevalence of celiac disease”. What was studied in actual fact, however, was the frequency of positive transglutaminase antibodies (tTG-AB), and the prevalence of celiac disease was estimated from this. Celiac disease is an immunologically mediated enteropathy. For a diagnosis without histological confirmation, the ESPGHAN guideline (2012) has recommended a tTG-AB titer above the 10-fold cut-off, and additionally the confirmation of positive endomysium antibodies (EMA-AB). The study determined only the tTG-AB, although JH, co-initiator of the study, the Robert Koch-Institute, which conducted the analysis, pointed out the necessity of also measuring EMA-AB several times. When both antibodies were measured, the sensitivity and specificity in mass screenings were substantially improved (2, 3). When the frequency of the tTG-AB values above the 10-fold cut-off is considered in the present study, the prevalence of celiac disease according to Table 1 is 0.4% (95% confidence interval [CI]: 0.3%; 0.5%), which seems to reflect reality. Such an assessment, while including positive EMA-AB, is also consistent to two cohort studies in German adults, of 0.2% and 0.3% from 2010 (4). A recent Canadian study found biopsy-proofed celiac disease in 13.3% of children with up to threefold-cut-off positive tTG-AB but negative EMA-AB (2). As the maximum specificity of tTG-AB is 95%, 5% of false results are to be expected. The conclusion by the authors of the study (1) that the prevalence of celiac disease of 0.9% is comparable with that in other European countries, is thus not correct, as enormous differences exist in this respect between European countries and further abroad.