Tissue Transglutaminase Levels Are Not Sufficient to Diagnose Celiac Disease in North American Practices Without Intestinal Biopsies

Background Celiac serology is crucial for the diagnosis of celiac disease in children. The American guideline for celiac disease in children suggested that positive serology should be followed by confirmatory intestinal histology. The relationship between high tissue transglutaminase titers and celiac disease in children has not been well investigated in children from North America. Aims In the present study, we investigated whether different tissue transglutaminase titers in symptomatic children could predict celiac disease without the confirmation of intestinal histology. Methods Data from biopsy confirmed celiac children were collected from four different clinics in North America. Clinical, serological, histological, and follow-up data were collected. The accuracy rates of various tissue transglutaminase titers to predict celiac disease in children were calculated. Results The data from 240 children were calculated. The sensitivity, specificity, positive predictive value, negative predictive value, and accuracy rate of tissue transglutaminase titers at ≥10× upper limit of normal were 75.4, 48.8, 87.7, 29.0, and 70.8 %, respectively. Similar data were noted in the other tissue transglutaminase titers (≥3× upper limit of normal, >100 U/ml, or >100 U/ml and >10× upper limit of normal). Conclusions The positive predictive value of tissue transglutaminase titers at ≥3× upper limit of normal or higher was too low to predict celiac disease in children. Our data suggested that in routine clinical practice, high titers of tissue transglutaminase are not sufficient to diagnose celiac disease in North American children without intestinal biopsies.