4. Eosinophils in autoimmune atrophic gastritis – typically numerous and a potentially helpful diagnostic feature

Abstract

<h3>Background</h3> Autoimmune atrophic gastritis (AAG) is an autoimmune condition characterised by loss of gastric parietal cells and subsequent replacement by metaplastic intestinal, gastric or pancreatic acinar epithelium. A lymphocyte rich inflammatory infiltrate localised to the parietal cell bearing gastric body and fundic mucosa and sparing the gastric antrum is typical. We have noted eosinophils to be a prominent component of the inflammatory process, and recently experimental evidence has suggested activation of the Th17 effector cell mechanism to be responsible. <h3>Aims</h3> To investigate eosinophils in AAG and their usefulness as a diagnostic adjunct. <h3>Methods</h3> Gastric body/fundic biopsies were collected in a consecutive prospective manner of AAG (<i>n</i>=52; diagnosed on basis of body only inflammation and suggestive history and/or antiparietal cell/ anti intrinsic factor antibodies and <i>H. pylori</i> immunohistochemical stain negative); <i>H. pylori</i> gastritis (HPG; <i>n</i>=49; immunohistochem-ical stain ± urease positive); multifocal atrophic gastritis (MAG; <i>n</i>=16; <i>H. pylori</i> immunohistochemical stain negative and atrophy of gastric antrum) and normal controls (NC; <i>n</i>=50; endoscopy and biopsy normal). Eosinophils were countered as the maximum in 1 HPF (field diameter 0.4mm) and maximum total in any 3 HPF. <h3>Results</h3> Maximum mean eosinophil numbers counts were as follows: 1 HPF, AAG 44; HPG 9; MAG 19; normal controls 3; 3 HPF, AAG 104; HPG 23; MAG 43, NC 6. There was a significant difference between the counts in AAG and those in NC, HPG and MAG (all <i>p</i> < 0.05). In 7 AAG cases (13.5%) eosinophil counts were ≤11 (the maximum count in NC). Eosinophils were common in both evolving and established disease as evidenced by the extent of metaplasia in biopsies. <h3>Conclusions</h3> A prominent increase in eosinophils is frequently identified in AAG and is greater than other conditions that comprise the histological differential diagnosis. This may serve as an additional diagnostic marker in histologically equivocal gastric body biopsies or where supportive clinical history is not available.