Effect of a 6-Month Functional Food Intervention on the Microbiota of Stunted Children in East Nusa Tenggara, Indonesia-A Randomized Placebo-Controlled Parallel Trial

Abstract

We have previously shown a difference between the gut microbiota composition of stunted and non-stunted children in East Nusa Tenggara, Indonesia. The current study aimed to perform an intervention with a probiotic, Lactiplantibacillus plantarum IS-10506, and its UHT-treated postbiotic compared to placebo in order to accomplish catch-up growth in the stunted children, possibly through modulation of the gut microbiota. Apart from the maltodextrin (placebo), probiotic, and postbiotic in chocolate milk, all groups also received a functional and nutritional biscuit and had access to newly constructed water wells as well as soap to improve hygiene. The results show that independent of treatment, the stunted children had a significantly higher increase in height and zlen (corrected for age) compared with their age- and gender-matched controls but a significantly lower increase in weight. Several potential pathogenic taxa declined in all groups, among which was Escherichia/Shigella (adjusted.p = 6.44 × 10^-15), but so did some beneficial taxa, such as Bifidobacterium and Akkermansia. Faecalibacterium, which was already higher in the stunted children at baseline, increased independent of treatment. Changes in the relative abundance of several taxa of the microbiota correlated with the changes in anthropometric measures. In conclusion, although there was no difference between the interventions, understanding the dynamics and the role of the gut microbiota in this process might allow healthcare providers to develop targeted nutritional strategies aimed at optimizing health outcomes for children at risk of stunting, thereby addressing a critical global health issue.

Keywords: Faecalibacterium; Indonesian children under 5 year; SCFA; butyrate; gut microbiota; nutritional status; stunted.

Figure 1

CONSORT flowchart of the study.

Figure 2

Change (delta) in anthropometric parameters (A) weight, (B) length, (C) BMI, (D) zlen, (E) zwei, (F) zbmi, and (G) zwfl after the 6-month intervention for the non-stunted and stunted group.

Figure 3

Response (in relative abundance) of several relevant taxa at baseline and at the end of the intervention (endline) with the three treatments, split up by children that were stunted or not: (A) Escherichia/Shigella *, (B) Haemophilus *, (C) Enterobacter *, (D) Desulfovibrio, (E) Bifidobacterium *, (F) Akkermansia, (G) Lactobacillus, and (H) Faecalibacterium *. The asterisk (*) indicates that these taxa were also significant when linear mixed modelling was applied (Table 3).

Figure 4

(A) Variable importance plot and response (in relative abundance) of taxa with a mean decrease accuracy of >4 in the random forest analysis at baseline and at the end of the intervention (endline) with the three treatments, split up by children that were stunted or not: (B) Catenibacterium *, (C) Ruminococcus torques group *, (D) Muribaculaceae *, (E) uncharacterized taxon in the family Oscillospiraceae, (F) Fusicatenibacter *, and (G) Collinsella *. The asterisk (*) indicates that these taxa were also significant when linear mixed modelling was applied (Table 3).

Figure 5

Heatmap of the significant Spearman correlations between taxa and anthropometric measures. * indicates significance (q < 0.1). A blue color indicates a positive correlation, red a negative correlation.