Mast Cell and Eosinophil Counts in Gastric and Duodenal Biopsy Specimens From Patients With and Without Eosinophilic Gastroenteritis

Background & Aims: Mast cells are believed to contribute to the development of eosinophilic gastrointestinal disorders (EGIDs). We quantified mast cells and eosinophils in biopsy specimens from patients with EGIDs and without known esophageal or gastrointestinal disease to investigate associations between these cell types and EGID and its features. Methods: We conducted a retrospective study of patients with EGID (n = 52) and of children and adults who underwent upper endoscopy and were found to have no evidence of gastrointestinal or systemic conditions (n = 123). We re-reviewed archived gastric and duodenal biopsy specimens to quantify mast cells (by tryptase immunohistochemistry) and eosinophils. We calculated the specificity of cell count thresholds for identification of patients with EGIDs and evaluated the correlation between mast cell and eosinophil counts and clinical and endoscopic features. Results: In the gastric biopsy specimens from patients without esophageal or gastrointestinal diseases, the mean mast cell count was 18.1 ± 7.2 cells per high-power field (hpf), and the peak mast cell count was 21.9 ± 8.2 cells/hpf. In the duodenal biopsy specimens from patients without esophageal or gastrointestinal diseases, the mean mast cell count was 23.6 ± 8.1 cells/hpf and the peak mast cell count was 28.1 ± 9.3 cells/hpf. The mean and peak eosinophil counts in gastric biopsy specimens from patients without disease were 3.8 ± 3.6 eosinophils/hpf and 5.8 ± 5.0 eosinophils/hpf; the mean and peak eosinophil counts in duodenal biopsy specimens were 14.6 ± 8.9 eosinophils/hpf and 19.5 ± 11.0 eosinophils/hpf. A mean count of 20 eosinophils/hpf in gastric biopsy specimens or 30 eosinophils/hpf in duodenal biopsy specimens identified patients with EGIDs with high specificity. Gastric and duodenal biopsy specimens from patients with EGIDs had significant increases in mean mast cell counts compared with biopsy specimens from patients without EGIDs. There was a correlation between mean mast cell and eosinophil counts in duodenal biopsy specimens (R = 0.47; P =.01). The mean mast cell and eosinophil counts did not correlate with symptoms or endoscopic features of EGIDs. Conclusions: We identified thresholds for each cell type that identified patients with EGIDs with 100% specificity. The increased numbers of mast cells and eosinophils in gastric and duodenal tissues from patients with EGIDs supports the concept that these cell types are involved in pathogenesis. However, cell counts are not associated with symptoms or endoscopic features of EGIDs.