Interobserver Agreement Among Pathologists in the Histological Diagnosis of Coeliac Disease: A Systematic Review.

Aims. Histological assessment of duodenal biopsies remains central to the diagnosis of coeliac disease (CD), as detailed in the 2026 British Society for Gastroenterology Guidelines on the diagnosis and management of CD, which recommend them as part of an integrated diagnostic approach. However, duodenal biopsies demonstrate a spectrum of mucosal changes making interpretation inherently subjective. The reproducibility of histological diagnosis and classification has not been previously systematically evaluated. Methods. A systematic review was registered with PROSPERO (CRD42021257108) and conducted in accordance with PRISMA guidelines. MEDLINE, Embase and Cochrane Library Trials were searched from database inception to February 13th, 2026. Data on study design and measures of interobserver agreement were extracted from full-text articles and synthesised. Results. Twenty-three eligible studies evaluating interobserver agreement of histological diagnosis or classification of CD met the inclusion criteria. Considerable methodological heterogeneity limited direct comparison between studies and precluded meta-analysis. Interobserver agreement for the binary histological diagnosis of CD varied considerably between studies (κ=-0.07 to κ=1.00) and declined when observers graded specimens according to CD classification schemes. Factors affecting reproducibility included pathologist expertise, biopsy quality, endoscopic sampling strategy, and numbers of categories in each grading scheme used. Conclusions. Histological assessment remains central to the diagnosis of CD, however, its reproducibility is variable. Our results suggest that integrated diagnostic pathways that combine histological, serological, and clinical information could maximise diagnostic accuracy.