Comparison of the interobserver reproducibility with different histologic criteria used in celiac disease

Clin Gastroenterol Hepatol. 2007 Jul;5(7):838-43. doi: 10.1016/j.cgh.2007.03.019. Epub 2007 Jun 4.


Background & aims: The Marsh-Oberhuber classification of duodenojejunal mucosal lesions is currently used for celiac disease. A more simplified classification, which is based on 3 villous morphologies (A, non-atrophic; B1, atrophic, villous-crypt ratio <3:1; B2, atrophic, villi no longer detectable) and an intraepithelial lymphocyte count of >25/100 enterocytes, has recently been proposed. The aim of the study was to asses the interobserver agreement between different pathologists in classifying celiac disease lesions according to both Marsh-Oberhuber and the new classification system.

Methods: Sixty patients were selected for the study: 10 subjects without celiac disease, 13 celiac patients with normal villi but a pathologic increase in intraepithelial lymphocytes >25/100 and hyperplastic crypts, and 37 patients with celiac disease with villous atrophy. Sixty slides were sent to 6 pathologists, who were blinded to each other and were not given any clinical information. Each pathologist received the set of biopsy specimens on 2 separate occasions and had to evaluate them according to both grading systems in a random order. The kappa statistic was used to assess agreement between each pair of pathologists.

Results: Overall, mean kappa values were 0.35 (fair) for the Marsh-Oberhuber classification versus 0.55 (moderate) for the new classification system.

Conclusions: The new classification for duodenal pathology in celiac disease gives better interobserver agreement compared with the more cumbersome Marsh-Oberhuber classification and contributes to the validity of diagnosis in celiac disease.

Publication types

  • Comparative Study

MeSH terms

  • Adolescent
  • Adult
  • Aged
  • Biopsy / methods
  • Celiac Disease / classification
  • Celiac Disease / pathology*
  • Child
  • Child, Preschool
  • Duodenum / pathology*
  • Female
  • Humans
  • Infant
  • Male
  • Middle Aged
  • Observer Variation
  • Reproducibility of Results
  • Severity of Illness Index