Two-step approach.

The authors reference as the basis for their summary the different evidence based guidelines that have been developed over recent years, and a literature search, whose search strategy included, among others, all publications from the past 10 years identified by using the search terms “celiac disease” and “diagnosis” (1). Unfortunately they did not include our article that was published in 2013, even though this article was based on long years of experience and aimed to reach a definitive diagnosis using a minimum number of biopsies (2). The best diagnostic test is that which results in the fewest false-positive and false-negative diagnoses; for this reason we’d suggest the following approach: The first step: simultaneous measuring of IgA and IgG antibodies specific for deamidated gliadin peptides, IgA antibodies specific for human tissue transglutaminase (in addition, total IgA). Most patients will either have a positive reaction to all three tested antigens or will test negative to all three of the specific antibody tests. In both these groups, biopsy is therefore unnecessary, since the positive predictive value (ppv) is 99% and the positive likelihood ratio (lr+) 87, whereas the negative predictive value (npv) is 98% and the negative likelihood ratio (lr-) 0.01. The results become even more meaningful (ppv 99%, lr+ 86; npv 100%, lr- 0.00) (2) if a fourth test is done for IgA endomysial–specific antibodies (2). The second step is small bowel biopsy. It is necessary only in patients with contradictory antibody results—that is, in patients who were positive in one or two tests only. This “two-step approach” reduces the proportion of patients requiring a biopsy to one-fifth (3, 4).