Small Intestine Bacterial Overgrowth in Bangladeshi Infants Is Associated With Growth Stunting in a Longitudinal Cohort

Abstract

Introduction: Small intestine bacterial overgrowth (SIBO) is common in children from low-income countries and has been cross-sectionally associated with growth stunting. We sought to determine whether SIBO was associated with poor growth and neurodevelopmental in a longitudinal analysis.

Methods: We measured SIBO by glucose hydrogen breath test (GHBT) at 18, 52, 78, and 104 weeks of life in a prospective longitudinal birth cohort of Bangladeshi children. Sociodemographic information and measures of enteric inflammation were analyzed as covariates. Diarrheal samples were tested for enteropathogens using polymerase chain reaction. Regression models were created using standardized mean GHBT area under the H2 curve (AUC) to determine associations with linear growth and cognitive, language, and motor scores on the Bayley-III Scales of Infant and Toddler Development at 2 years. We also investigated associations between GHBT AUC and enteropathogen exposure.

Results: A 1-ppm increase in standardized mean GHBT AUC was associated with a 0.01-SD decrease in length-for-age Z score (P = 0.03) and a 0.11-point decrease in Bayley language score (P = 0.05) at 2 years of age in adjusted analysis. Enteroaggregative Escherichia coli, Enteropathogenic Escherichia coli, Giardia, and Enterocytozoon bieneusi were associated with increased GHBT AUC, whereas Clostridium difficile, norovirus GI, sapovirus, rotavirus, and Cryptosporidium were associated with decreased GHBT AUC. None were consistent across all 4 time points.

Discussion: SIBO in the first 2 years of life is associated with growth stunting and decreased language ability in Bangladeshi infants and may represent a modifiable risk factor in poor growth and neurodevelopment in low-income countries.

Trial registration: ClinicalTrials.gov NCT02745327.

Figure 1.

Glucose hydrogen breath test output over the first 2 years of life. We analyzed 3 separate outputs of the glucose hydrogen breath test. Prevalence (as determined by > 12 ppm over the subject's baseline) (a), mean (with SE shown by whiskers) maximum change in breath hydrogen from baseline (b), and mean (with SE shown by whiskers) area under the hydrogen curve (AUC) (c). All measures demonstrated an increasing trend from 18 weeks of age to 78 weeks with a slight drop at 104 weeks. Differences across time were significant for all measures (χ² P value for prevalence and Welch's ANOVA P value for mean maximum change in breath hydrogen and mean GHBT AUC; all were P < 0.001***).

Figure 2.

Effects of glucose hydrogen breath testing area under the H₂ curve (AUC) on length-for-age Z score over time. Children were stratified by their standardized mean glucose hydrogen breath test (GHBT) AUC and the top and bottom quartiles compared with their time to stunting (length-for-age Z score < −2 SD). Children with higher standardized mean GHBT AUC had significantly quicker time to stunting over the first 2 years of life (P = 0.02).

Figure 3.

Association of pathogen carriage with glucose hydrogen breath test area under the H₂ curve (AUC). Glucose hydrogen breath test (GHBT) AUC was compared between children with pathogens of interest in their diarrheal stool within the 3 months before each GHBT and those without. Mann-Whitney U tests were performed to determine whether there was a significant difference in standardized GHBT AUC between pathogen exposed and unexposed children. Each pathogen at each time point was assessed independently. * P value of 0.01 to 0.05, ** P value of 0.001 to 0.01, and *** P value < 0.001.