Serum Anti-Tissue Transglutaminase IgA as Predictor Of  Duodenal Villous Atrophy in Adults with Suspected Coeliac Disease. The Bi.A.Ced Multi-Centre Prospective Study

Background: It remains controversial whether coeliac disease (CeD) in adults can be diagnosed using only serology. We conducted   a prospective study investigating this point focusing on serum anti-tissue transglutaminase IgA (anti-tTG IgA).   Methods: Patients with suspected CeD were enrolled by 14 centres and underwent local anti-tTG IgA measurements and endoscopic duodenal biopsy. Anti-tTG IgA was expressed as a multiple of each test’s upper limit of normal range (xULN) and defined high when > 1. CeD diagnosis was defined as the presence of villous duodenal atrophy.  Histology was evaluated by local pathologist and the central pathologist in discordant cases (high anti-tTG IgA without duodenal villous atrophy or non-high anti-tTG IgA with duodenal villous atrophy). Anti-tTG IgA was measured in a subgroup of patients by the central lab. Findings: Local data were complete for 436 patients. CeD prevalence was 79.6% by local centre histology and 81.4% after re-evaluation of 30 discordant cases; high anti-tTG IgA prevalence was 83.3%. Positive predictive value of duodenal villous atrophy increased with higher anti-tTG IgA levels using local data (from 9.6% for ≤1 xULN to 97.8% for >15 xULN, P <0.001) and after central re-evaluation (9.6% for ≤1 xULN to 100% for >15 xULN, P <0.001).  Using the > x10 ULN cut-off for anti-tTG for the serological diagnosis of CeD, this study correctly identified 161 (99.4%) of 162 patients without needing a biopsy. For the whole cohort, 36.9% would have avoided biopsy. Interpretation: Our data showed that biopsy could be reasonably avoided in adult CeD patients with anti-tTG IgA >10x ULN. Funding: None. Declaration of Interest: GLN is an employee of Werfen. All other authors declare they have no specific conflict of interest. Ethical Approval: The study was approved first by the Ethical Committee of the University of Salerno (Approval number 21, 15-02-2018) and later by the local Ethical Committees of all the participating centres.