Interobserver agreement of estimating the extent of intestinal metaplasia in patients with chronic atrophic gastritis

Abstract

The extent of gastric intestinal metaplasia (GIM) can be used to determine the risk of gastric cancer. Eleven international gastrointestinal expert pathologists estimated the extent of GIM on haematoxylin and eosin (H&E)- and Alcian blue-Periodic acid Schiff (AB-PAS)-stained slides of 46 antrum biopsies in 5% increments. Interobserver agreement was tested with the intraclass correlation coefficient (ICC). Correlation between standard deviation and extent of GIM was evaluated with the Spearman correlation. The interobserver agreement was very good (ICC = 0.983, 95% confidence interval (CI) 0.975–0.990). The use of AB-PAS did not increase the agreement (ICC = 0.975, 95% CI 0.961–0.985). Cases with a higher amount of metaplastic epithelium demonstrated a higher standard deviation (rs = 0.644; p < 0.01), suggesting lower diagnostic accuracy in cases with extensive GIM. In conclusion, estimating the extent of GIM on H&E-stained slides in patients with chronic atrophic gastritis can be achieved satisfactorily with high interobserver agreement, at least among international expert gastrointestinal pathologists.