Histological assessment of gastric pseudopyloric metaplasia: Intra- and inter-observer consistency

The histological spectrum of oxyntic mucosal atrophy (a major determinant of gastric cancer risk) includes pseudopyloric metaplasia (PPM), which histological assessment has been regarded as unreliable. PPM consistently expresses Trefoil-Factor 2 (TFF2), which is histochemically detecteble (TFF2-IHC).

Intra- and inter-observer consistency in assessing PPM was examined using both hematoxylin & eosin (H&E) and TFF2-IHC.

Seventy-four oxyntic biopsy samples obtained from autoimmune gastritis were considered. Two serial histological sections obtained from the paraffin-embedded tissue-samples were stained with H&E and TFF2-IHC. Three pathologists (Alpha, Beta, Gamma) independently scored PPM by both staining and the Intra- and inter-observer consistency (H&E versus TFF2-IHC) was calculated using k-statistics and/or Spearman's coefficient.

Based on H&E-stain versus TFF2-IHC, intra-observer consistency in PPM assessement was ranked as consistently "good" (k-values: Alpha=0.79; Beta=0.78; Gamma=0.75). Based on H&E, the overall PPM inter-observer consistency among the 3 observers was ranked as "good" (k=0.77) (the inter-observer consistency for pairs of observers was as follows: Alpha versus Beta k=0.88; Alpha versus Gamma k=0.87; Beta versus Gamma k=0.80). Based on TFF2-IHC, the overall PPM inter-observer agreement was ranked as "excellent" (k=0.91) (the inter-observer consistency for pairs of observers was as follows: Alpha versus Beta k=1; Alpha versus Gamma k=0.91; Beta versus Gamma k=0.91).

Relying on either H&E staining or TFF2-IHC, pathologists assess PPM consistently.