%0 Journal Article
%J Am J Clin Nutr
%D 2017
%T A methodologic framework for modeling and assessing biomarkers of environmental enteropathy as predictors of growth in infants: an example from a Peruvian birth cohort
%A Colston, J. M.
%A Penataro Yori, P.
%A Colantuoni, E.
%A Moulton, L. H.
%A Ambikapathi, R.
%A Lee, G.
%A Rengifo Trigoso, D.
%A Siguas Salas, M.
%A Kosek, M. N.
%K *Nutritional Status
%K biomarkers
%K Biomarkers/metabolism
%K body height
%K Carrier Proteins/*metabolism
%K Child, Preschool
%K Environment
%K environmental enteropathy
%K environmental exposure
%K Feces
%K Female
%K Growth Disorders/*etiology/metabolism
%K Humans
%K Infant
%K Infection/*complications/metabolism/pathology
%K inflammation
%K Intestinal Diseases/*complications/metabolism/pathology
%K Intestines/metabolism/pathology
%K Mal-ed
%K Male
%K Malnutrition/complications
%K mixed-effects model
%K Models, Biological
%K Neopterin/*metabolism
%K nutritional status
%K Permeability
%K Peroxidase/*metabolism
%K Peru
%X Background: Environmental enteropathy (EE) impairs the gut's absorptive capacity and immune function and causes decelerations in statural growth that manifest gradually over time.Objective: To illustrate an approach for assessing emerging biomarkers of EE, we separately assessed the associations between 3 such markers and subsequent nutritional status.Design: Stool samples were routinely collected between January 2010 and November 2014 from a cohort of 303 Peruvian infants and analyzed for concentrations of the biomarkers alpha-1-antitrypsin (AAT), myeloperoxidase, and neopterin. For each marker, a mixed-effects linear regression model was fitted for length-for-age z scores (LAZs) obtained from anthropometric assessments that incorporated covariate predictors, polynomial terms for age, and product interaction terms to test associations over varying lag lengths. The biomarkers' contribution to the models was assessed with the use of the likelihood ratio test and partial R2 statistics.Results: Test statistics for the combined inclusion of the 4-model terms that involved the biomarker were highly statistically significant for AAT (28.71; P < 0.0001) and myeloperoxidase (62.79; P < 0.0001) over a 3-mo lag and moderately so for neopterin (13.97; P = 0.0074). AAT and myeloperoxidase seemed to interact strongly with age, with the magnitude and direction of the effect varying considerably over the first 3 y of life. The largest proportion of the variance explained by any biomarker (2.8%) and the largest difference in LAZ predicted between the 5th and 95th percentile (0.25) was by myeloperoxidase over a 2-mo lag.Conclusions: Of the 3 fecal biomarkers studied, 2 that related to intestinal function-AAT and myeloperoxidase-were associated with small but highly statistically significant differences in future statural growth trajectories in infants in this cohort, lending further evidence to the EE hypothesis that increased gut permeability and inflammation adversely affects subsequent nutritional status. This association exhibited a complex interaction with age. This trial was registered at clinicaltrials.gov as NCT02441426.
%B Am J Clin Nutr
%V 106
%P 245-255
%8 Jul
%@ 0002-9165
%M 28592604