Duodenal Eosinophilia in Eosinophilic Esophagitis

Abstract

Purpose: Eosinophils are present throughout the GI tract with the exception of the esophagus. The significance of increased eosinophils in the duodenum is not well understood. Duodenal eosinophilia has been seen in dyspepsia but has not been well studied in eosinophilic esophagitis (EoE). The aim of this study is to characterize gastric and duodenal eosinophilia in patients with EoE and non-ulcer dyspepsia (NUD). Our hypothesis is that there is increased duodenal eosinophilia in EoE and NUD compared to control patients. We also hypothesize that there is no difference in gastric eosinophilia among these groups. Methods: A retrospective chart review was done to gather information on baseline characteristics. Gastric and duodenal biopsy histopathology was prospectively reviewed. Eosinophils were counted in five high power fields. The mean eosinophil count was calculated for each patient. Each group had about thirty patients. The NUD patients met Rome III criteria. A small group of eosinophilic gastroenteritis (EG) patients were also compared to controls. Patients with malignancies, hypereosinophilic syndrome, inflammatory bowel disease, infections, H. pylori, myeloproliferative disease, collagen vascular disease, celiac disease and alcohol abuse were excluded from the study. Results: Duodenal eosinophil counts for EoE patients ranged from 1 to 54, with a mean of 8. Ranges in the other three groups were smaller. The median duodenal eosinophil counts for EoE (5.5) and NUD (4.3) were higher than the control group (4), p<0.05. There was no statistically significant difference between the median duodenal eosinophil counts of EG (4.8) and controls (4). The median gastric eosinophil count for EG (28.7) was higher than the control group (4.2), p<0.05. There was no statistically significant difference between the median gastric eosinophil counts of EoE patients (4.4), NUD patients (3) and control patients (4.2). Conclusion: Patients with EoE and NUD have increased duodenal eosinophilia compared to control patients. The median eosinophil count of the EG group was higher than the control group and NUD group. However, there was probably no statistically significant difference between EG and control patients due to the relatively small number of EG patients. The similarity of gastric eosinophil counts among the EoE, NUD and control patients is consistent with prior studies. The high gastric eosinophil count of the EG group is consistent with their diagnosis. The median eosinophil counts in this study may be lower than prior studies due to higher percentage of female patients, younger age, and exclusion of patients with H. pylori. Duodenal eosinophilia in EoE is a novel finding and additional studies will need to done.