Eosinophils in gastritis : A study on gastric tissue eosinophil density and blood leucocyte differential counts in patients with gastritis at the Kenyatta National Hospital
Kalebi, Ahmed Y
MetadataShow full item record
Background: Little is known of the significance of eosinophils infiltration in the gastritis, yet infiltration of the gastric mucosa by various inflammatory cells and gastric epithelial changes forms the basis for the evaluation of gastritis. Presently used guidelines for the histological assessment of gastritis do not include any objective measures for assessment of these cells. Recent studies however suggest that the eosinophil may be an important player in the tissue damage associated with gastritis. In addition, the assessment of peripheral WBC total and differential counts have found utility as diagnostic adjuncts in the screening and monitoring of various gastrointestinal diseases. However there is no sufficient published data on the pattern of WBC counts in gastritis. Objective: This study was set out to examine gastric tissue eosinophil densities in endoscopic gastric biopsies and peripheral WBC total and differential counts to elucidate their relation to established clinicopathological parameters of gastritis. Design: Prospective cross-sectional descriptive study Setting: Kenyatta National Hospital- Kenya's main referral and teaching hospital Subjects: Consecutive adult patients who underwent endoscopy during the study period. Methods: Structured questionnaires were used to collect clinical and demographic data. The peripheral WBC total and differential counts were determined on venous blood using an automated 5-part differential coulter counter, supported with a manual PBF check. The histological examination was done as per the Sydney System for graded variables of gastritis. The GTED was recorded as cells per HPF of the microscope, on formalin-fixed 4 um-thick paraffin H&E-stained sections submitted from corpus and antrum mucosa. Main outcome measures: Age and gender; history of dyspepsia, allergy, food intolerance; use of NSAIDs, antibiotics, antacids, protein pump inhibitors and H2 blockers and alcohol; endoscopic description of the gastric lining; graded histological parameters of gastritis: Inflammation, Activity, Atrophy, Intestinal metaplasia and H pylori ; gastric tissue eosinophil density as cells/HPF; Peripheral blood total and differential WBC counts. Statistical considerations: The X2 test, Student t-test, Mann-Whitney U test and One-way ANOYA were used to analyze differences and compare mean cell counts between various groups. The Wilcoxon signed-rank test was used to compare histological findings between the related variables in different biopsy sites. Determination of spearman's rho correlation coefficient was also made to examine the relationship between quantitative variables. P <0.05 was considered statistically significant. Results: -Gastric biopsies from 65 patients were evaluated as per the inclusion/exclusion criteria. The mean age was 43 years with an age range of 18 to 86 years, with a male to female ratio of 1: 1. Sixty patients (94%) had a positive history of dyspepsia and 21 (32%) were found to have duodenal ulcer on endoscopy. Most of the biopsies showed moderate to severe degree of active and chronic inflammation as per the criteria of the Sydney System. H pylori was found in 89% of the cases on histological examination. Significant histopathological discordance was observed between biopsies from the antrum and the corpus, as well as between multiple antral biopsies from respective patients. The antrum consistently showed higher grades for all the graded parameters compared to the corpus. Significant association was also found between histology and the use of PPI, NSAID, antibiotics and triple therapy, but not with H2B, antacids or alcohol. Duodenal ulcer was significantly associated with higher grades of inflammation in the corpus. The mean gastric tissue eosinophil density (GTED) was 5.9 ± 0.74 eosinophils/HPF in the corpus and 9.58 ± 0.93 eosinophils/HPF in the antrum. Ten biopsies (12%) from the corpus and one from the antrum did not have any eosinophils. Majority of the biopsies in the corpus had less than 5 eosinophils/HPF. The highest density observed was 23 eosinophilslHPF in the corpus and 31 eosinophils/HPF in the antrum. No statistically significant association was found between GTED and demographic or clinical data. Use of PPI was associated with lower GTED in the antrum, while duodenal ulcer was associated with higher GTED in the corpus. Significant positive correlation was found between the GTED and the degree of chronic inflammation, neutrophil activity and H pylori colonization. No significant correlation was found with degree of atrophy or intestinal metaplasia. The mean total and differential WBC counts in this population were within normal reference ranges: WBC 5.73xI09/L, Neutrophils 2.64xI09/L, Lymphocytes 2.15x109/L, Monocytes 0.33xI09/L. Eosinophils 0.18 xl09/L and Basophils 0.07 xl09/L. However, 12 patients (20%) had leucopenia, 13 (21%) had neutropenia, 15 (25%) had monocytopenia and 13 (21 %) had basophilia. No significant association (p>0.05) was seen between the WBC counts and clinical or endoscopic data, but significant association (p<0.05) with negative spearman's rho correlation coefficients were found between the total WBC and neutrophil counts with the use of PPI and H2B. There was significant positive correlation (spearman's rho correlation p<O.05) between the WBC counts and the graded variables. H pylori correlated positively (spearman's rho correlation p>O.05) to peripheral total WBC count, neutrophils, lymphocytes, basophils and eosinophils. Patients with leucopenia, neutropenia and eosinophilia were found to have significantly higher degrees of inflammation (t-test p<0.05) in the gastric mucosa than those with counts within the reference range. Conclusion: This study has shown that gastric tissue and peripheral blood eosinophil counts vary in gastritis in positive correlation to the degree of inflammation. The observations suggest that eosinophils are an important mediator in the inflammatory process of H pylori¬associated gastritis. The quantitative assessment of gastric tissue eosinophils may be of value as a histological marker of the level of inflammation, while peripheral blood eosinophil counts together with other leucocyte differential counts may have a pattern that is peculiar in gastritis. The study reaffirms that H pylori-associated gastritis is the most prevalent form of gastritis in KNH. Discordance observed between multiple biopsies taken from the antrum of respective patients implies that the number of antral biopsies submitted per respective patient influences the sensitivity for the diagnosis of atrophy and intestinal metaplasia. Recommendations: Quantitative histological assessment of gastric tissue eosinophils should be considered for inclusion in the present guidelines last updated more than a decade ago, as one of the objective measures of the degree of inflammation and tissue damage in gastritis. Owing to the paucity of published data in this area, further studies are needed and if confirmed through a wider intervention studies, peripheral WBC counts may prove to be a relatively inexpensive and simple way of monitoring disease progression in patients who are on treatment for H pylori-associated gastritis. The clinical value of peripheral WBC differential counts in patients with gastritis could be pursued through routine measurement in patients being followed-up to see whether changes correlate to disease progression. It is a recommendation from this study that endoscopists should take multiple biopsies from the antrum, including the incisura. which should be submitted separately for histopathological examination to increase the pick-Up rate for gastric atrophy and intestinal metaplasia.