Antiendomysial and antihuman recombinant tissue transglutaminase antibodies in the diagnosis of coeliac disease

Objective To investigate the value of serum antitissue transglutaminase IgA antibodies (IgA-TTG) and IgA antiendomysial antibodies (IgA-EMA) in the diagnosis of coeliac disease in cohorts from different geographical areas in Europe. The setting allowed a further comparison between the antibody results and the conventional small-intestinal histology. Methods A total of 144 cases with coeliac disease [median age 19.5 years (range 0.9-81.4)], and 127 disease controls [median age 29.2 years (range 0.5-79.0)], were recruited, on the basis of biopsy, from 13 centres in nine countries. All biopsy specimens were re-evaluated and classified blindly a second time by two investigators. IgA-TTG were determined by ELISA with human recombinant antigen and IgA-EMA by an immunofluorescence test with human umbilical cord as antigen. Results The quality of the biopsy specimens was not acceptable in 29 (10.7%) of 271 cases and a reliable judgement could not be made, mainly due to poor orientation of the samples. The primary clinical diagnosis and the second classification of the biopsy specimens were divergent in nine cases, and one patient was initially enrolled in the wrong group. Thus, 126 coeliac patients and 106 controls, verified by biopsy, remained for final analysis. The sensitivity of IgA-TTG was 94% and IgA-EMA 89%, the specificity was 99% and 98%, respectively. Conclusions Serum IgA-TTG measurement is effective and at least as good as IgA-EMA in the identification of coeliac disease. Due to a high percentage of poor histological specimens, the diagnosis of coeliac disease should not depend only on biopsy, but in addition the clinical picture and serology should be considered.