Background: So far the reliability of the anti-guinea pig and anti-human tissue transglutaminase antibodies for the coeliac disease diagnosis has been evaluated in selected groups of patients.
Aim: To compare the diagnostic accuracy of anti-human versus anti-guinea pig tissue transglutaminase in the coeliac disease screening of the general population.
Subjects: Two healthy Italian populations living in Marche region and in Western Sardinia.
Methods: Both anti-guinea pig and anti-human tissue transglutaminase were determined using an enzyme-linked immunosorbent assay-based commercially available kit (Eu-tTG, Eurospital, Trieste, Italy).
Results: During the period 1999-2001, 3541 subjects (1500 from "continental" Italy and 2041 from Sardinia) were screened for coeliac disease using both anti-guinea pig and anti-human tissue transglutaminase as first-level tests. Both these tests were negative in 3439/3541 sera, while 29 resulted positive for both of them and 73 showed discordant results. Overall, 50 intestinal biopsies were performed in 22, 21 and 7 subjects with positivity to both screening tests, to anti-guinea pig and to anti-human tissue transglutaminase alone, respectively. A coeliac disease diagnosis was made in 25 subjects giving an overall prevalence of 1:126 individuals. The anti-tissue transglutaminase specificity and sensitivity were 98 and 92% for guinea pig and 99.6 and 96% for human tissue transglutaminase, respectively.
Conclusions: The anti-human tissue transglutaminase test should definitely replace the anti-guinea pig-derived one as first-level screening tool for identifying all subjects who need the second-level investigations (small intestinal biopsy).