Context.—

Eosinophilic diseases of the gastrointestinal tract (EGIDs), eosinophilic gastritis (EoG), and eosinophilic duodenitis (EoD) are rarely suspected clinically and infrequently detected by pathologists.

Objective.—

To determine whether histories of allergic or eosinophilic disorders and requests to rule out EoG and EoD affect pathologists' awareness of eosinophils in gastrointestinal biopsies.

Design.—

Thirty-one community-based pathologists were given 16 sets of biopsies from gastric and duodenal mucosa with elevated eosinophils, Helicobacter pylori gastritis, atrophic gastritis, normal stomach and duodenum, lymphocytosis, and celiac disease. Participants were assigned to 3 groups: group A did not receive histories of allergic or eosinophilic conditions; group B received similar histories plus a clue of possible allergic or eosinophilic conditions; and group C received the same histories as B and was asked to rule out EoG/EoD. A list of gastric and duodenal diagnoses and a space for comments were provided. Results were analyzed descriptively.

Results.—

Pathologists correctly diagnosed most noneosinophilic gastrointestinal disorders, indicating competence in gastrointestinal pathology. With respect to EoG and EoD, pathologists in group C performed significantly better that those in groups A and B. The combined odds ratio with 95% CI was 12.34 (2.87–53.04), P < .001, for A versus C and 4.02 (1.60–10.09), P < .02, for B versus C.

Conclusions.—

Most pathologists neither reported gastric/duodenal eosinophilia nor diagnosed EoG/EoD, even when provided histories of eosinophilic disorders. Requests to rule out EoG/EoD resulted in only 4 of 11 participants evaluating and counting eosinophils in some cases. Simple evidence-based histopathologic criteria are needed before pathologists can be expected to consider and diagnose EGIDs.

This content is only available as a PDF.

Author notes

The authors have no relevant financial interest in the products or companies described in this article.

Allakos Inc provided an unrestricted educational grant, which was used in part to provide participant pathologists with a small honorarium for reading the study slides.