Tailored Testing for Celiac Disease

Celiac disease is a gluten-sensitive enteropathy occurring in genetically susceptible individuals and affecting almost 1% of the population. The small-intestine biopsy, currently the gold standard test, is a rather invasive and expensive test to serve as the initial step in the diagnosis of celiac disease (1, 2). Therefore, current guidelines recommend serologic testing for autoantibodies against tissue transglutaminase (TGA) or endomysium (EMA) as the first step (1, 2). Most studies have reported excellent sensitivity and specificity for these tests (3, 4), but some studies have reported unsatisfactory sensitivity, particularly among patients with mild intestinal damage (5, 6). The HLA types DQ2 or DQ8 are strongly associated with celiac disease, but few researchers have tested this association as a basis for diagnosis (7).

In this issue, Hadithi and colleagues (8) compared the test performance of HLA-DQ typing and serology for diagnosing celiac disease. Unlike participants in many previous studies, all participants in Hadithi and colleagues' study had duodenal biopsy as the primary diagnostic test. The investigators thereby avoided the selection bias inherent when physicians use a positive serologic test result to decide whom to refer for biopsy, which is one of the several limitations of many studies (4). Subsequent to the biopsy, all participants had serologic testing and HLA-DQ typing. The major strength of the study is the large number of unaffected participants, which provides a precise estimate of specificity for serologic tests. The specificity was 99% for TGA or EMA and 57% for HLA-DQ typing, both with narrow 95% CIs. By contrast, HLA typing had excellent sensitivity: 100% versus 81% for serology. The sensitivity of antigliadin antibodies (AGA) was inferior to EMA and TGA. These data show clearly that serologic tests and HLA typing perform quite differently as tests.

On the basis of these …