TY - JOUR
T1 - A novel histological index for evaluation of environmental enteric dysfunction identifies geographic-specific features of enteropathy among children with suboptimal growth
AU - Liu, Ta Chiang
AU - Vanbuskirk, Kelley
AU - Ali, Syed A.
AU - Kelly, M. Paul
AU - Holtz, Lori R.
AU - Yilmaz, Omer H.
AU - Sadiq, Kamran
AU - Iqbal, Najeeha
AU - Amadi, Beatrice
AU - Syed, Sana
AU - Ahmed, Tahmeed
AU - Moore, Sean
AU - Ndao, I. Malick
AU - Isaacs, Michael H.
AU - Pfeifer, John D.
AU - Atlas, Hannah
AU - Tarr, Phillip I.
AU - Denno, Donna M.
AU - Moskaluk, Christopher A.
N1 - Publisher Copyright:
© 2020 Liu et al.
PY - 2020/1
Y1 - 2020/1
N2 - Background A major limitation to understanding the etiopathogenesis of environmental enteric dysfunction (EED) is the lack of a comprehensive, reproducible histologic framework for characteriz-ing the small bowel lesions. We hypothesized that the development of such a system will identify unique histology features for EED, and that some features might correlate with clinical severity. Methods Duodenal endoscopic biopsies from two cohorts where EED is prevalent (Pakistan, Zambia) and North American children with and without gluten sensitive enteropathy (GSE) were pro-cessed for routine hematoxylin & eosin (H&E) staining, and scanned to produce whole slide images (WSIs) which we shared among study pathologists via a secure web browser-based platform. A semi-quantitative scoring index composed of 11 parameters encompassing tis-sue injury and response patterns commonly observed in routine clinical practice was con-structed by three gastrointestinal pathologists, with input from EED experts. The pathologists then read the WSIs using the EED histology index, and inter-observer reliability was assessed. The histology index was further used to identify within-and between-child variations as well as features common across and unique to each cohort, and those that cor-related with host phenotype. Results Eight of the 11 histologic scoring parameters showed useful degrees of variation. The over-all concordance across all parameters was 96% weighted agreement, kappa 0.70, and Gwet’s AC 0.93. Zambian and Pakistani tissues shared some histologic features with GSE, but most features were distinct, particularly abundance of intraepithelial lymphocytes in the Pakistani cohort, and marked villous destruction and loss of secretory cell lineages in the Zambian cohort. Conclusions We propose the first EED histology index for interpreting duodenal biopsies. This index should be useful in future clinical and translational studies of this widespread, poorly under-stood, and highly consequential disorder, which might be caused by multiple contributing processes, in different regions of the world.
AB - Background A major limitation to understanding the etiopathogenesis of environmental enteric dysfunction (EED) is the lack of a comprehensive, reproducible histologic framework for characteriz-ing the small bowel lesions. We hypothesized that the development of such a system will identify unique histology features for EED, and that some features might correlate with clinical severity. Methods Duodenal endoscopic biopsies from two cohorts where EED is prevalent (Pakistan, Zambia) and North American children with and without gluten sensitive enteropathy (GSE) were pro-cessed for routine hematoxylin & eosin (H&E) staining, and scanned to produce whole slide images (WSIs) which we shared among study pathologists via a secure web browser-based platform. A semi-quantitative scoring index composed of 11 parameters encompassing tis-sue injury and response patterns commonly observed in routine clinical practice was con-structed by three gastrointestinal pathologists, with input from EED experts. The pathologists then read the WSIs using the EED histology index, and inter-observer reliability was assessed. The histology index was further used to identify within-and between-child variations as well as features common across and unique to each cohort, and those that cor-related with host phenotype. Results Eight of the 11 histologic scoring parameters showed useful degrees of variation. The over-all concordance across all parameters was 96% weighted agreement, kappa 0.70, and Gwet’s AC 0.93. Zambian and Pakistani tissues shared some histologic features with GSE, but most features were distinct, particularly abundance of intraepithelial lymphocytes in the Pakistani cohort, and marked villous destruction and loss of secretory cell lineages in the Zambian cohort. Conclusions We propose the first EED histology index for interpreting duodenal biopsies. This index should be useful in future clinical and translational studies of this widespread, poorly under-stood, and highly consequential disorder, which might be caused by multiple contributing processes, in different regions of the world.
UR - http://www.scopus.com/inward/record.url?scp=85078561422&partnerID=8YFLogxK
U2 - 10.1371/journal.pntd.0007975
DO - 10.1371/journal.pntd.0007975
M3 - Article
C2 - 31929525
AN - SCOPUS:85078561422
SN - 1935-2727
VL - 14
SP - 1
EP - 21
JO - PLoS Neglected Tropical Diseases
JF - PLoS Neglected Tropical Diseases
IS - 1
M1 - e0007975
ER -