Identification of a serum transglutaminase threshold value for the noninvasive diagnosis of symptomatic adult celiac disease patients: a retrospective study

Abstract

Background: A celiac disease (CD) diagnosis is based on duodenal histology, with the exception of children showing anti-tissue transglutaminase (anti-tTG) serum levels exceeding ten times the cut-off. Our aim was to reproduce this simplified approach in adults, identifying an anti-tTG threshold value useful to diagnose CD without endoscopic procedures.

Methods: A total of 671 adult CD patients were subjected to blood sampling to determine anti-tTG serum levels, as well as to endoscopy with biopsy to perform duodenal histology. The anti-tTG serum levels/cut-off ratio was compared with the degree of duodenal lesions.

Results: Anti-tTG serum levels/cut-off ratio determined in patients with type IIIc was significantly higher than that measured in patients with type IIIb (p < 0.001), IIIa (p < 0.001), II (p < 0.05) and 0 (p < 0.001) of Marsh-Oberhuber histological classification. A significant correlation (r = 0.297, p < 0.0001) was found between the anti-tTG serum levels/cut-off ratio and the degree of duodenal lesions. The anti-tTG serum levels/cut-off ratio was classified as an accurate parameter (AUC = 0.715, p < 0.0001), with the best diagnostic performance obtained considering the threshold value >3.6 (sensitivity = 76.8 %, PPV = 97.2 %).

Conclusions: The anti-tTG serum levels/cut-off ratio correlates with the degree of duodenal lesions and, if used with the threshold value >3.6, could avoid endoscopy with biopsy in about 75 % of seropositive adults waiting for CD diagnosis. However, since this procedure could also imply CD diagnosis in almost 3 % of seropositive patients with normal villous architecture, a consensus opinion is needed to suggest its use in the diagnosis of adult CD.

Keywords: Adult celiac disease; Anti-transglutaminase; Diagnosis; Duodenal lesions; Threshold value.