Histopathology underlying environmental enteric dysfunction in a cohort study of undernourished children in Bangladesh, Pakistan, and Zambia compared with United States children

Document Type

Article

Department

Biological and Biomedical Sciences; Community Health Sciences; Paediatrics and Child Health

Abstract

Background: Environmental enteric dysfunction (EED) is an asymptomatic intestinal disorder associated with growth impairment, delayed neurocognitive development, and impaired oral vaccine responses.
Objectives: We set out to develop and validate a histopathologic scoring system on duodenal biopsies from a cohort study of children with growth failure in Bangladesh, Pakistan, and Zambia ("EED") with reference to biopsies from United States children with no clinically reported histologic pathology (referred to hereafter as "normal") or celiac disease.
Methods: Five gastrointestinal pathologists evaluated 745 hematoxylin and eosin slide images from 291 children with EED (mean age: 1.6 y) and 66 United States children (mean age: 6.8 y). Histomorphologic features (i.e., villus/crypt architecture, goblet cells, epithelial and lamina propria acute/chronic inflammation, Brunner's glands, Paneth cells, epithelial detachment, enterocyte injury, and foveolar metaplasia) were used to score each histopathologic slide. Generalized estimating equations were used to determine differences between EED, normal, and celiac disease, and receiver operating characteristic curves were used to assess predictive value.
Results: Biopsies from the duodenal bulb showed higher intramucosal Brunner's gland scores and lower intraepithelial lymphocyte scores than from the second or third parts of the duodenum (D2/3), so only D2/3 were included in the final analysis. Although 7 parameters differed significantly between EED and normal biopsies in regression models, only 5 (blunted villus architecture, increased intraepithelial lymphocytosis, goblet cell depletion, Paneth cell depletion, and reduced intramucosal Brunner's glands) were required to create a total score percentage (TSP-5) that correctly identified EED against normal biopsies (AUC: 0.992; 95% CI: 0.983, 0.998). Geographic comparisons showed more severe goblet cell depletion in Bangladesh and more marked intraepithelial lymphocytosis in Pakistan.
Conclusions: This scoring system involving 5 histologic parameters demonstrates very high discrimination between EED and normal biopsies, indicating that this scoring system can be applied with confidence to studies of intestinal biopsies in EED.

Publication (Name of Journal)

The American Journal of Clinical Nutrition

DOI

doi.org/10.1016/j.ajcnut.2024.02.028

Link to Full Text

Share

COinS