Gut Microbiota Dysbiosis in Suspected Food Protein Induced Proctocolitis—A Prospective Comparative Cohort Trial

Abstract

Objectives: In infants with suspected food protein induced proctocolitis (sFPIP) only a minority of patients are finally diagnosed with the disease following diagnostic dietary intervention (DDI). There is a need for a pathophysiological explanation for the cause of hematochezia in the majority of sFPIP infants. Methods: We prospectively recruited infants with sFPIP and healthy controls. Fecal samples were collected at inclusion, week 4 (end of DDI in sFPIP), and week 8. For 16S rRNA sequencing (515F/806R) we used Illumina MiSeq sequencing system. Amplicon sequence variants were generated using Qiime2 and DADA2. Qiime diversity alpha and beta group comparisons and linear discriminant analysis effect size analysis was performed. For shotgun metagenomic analysis on species level we used KneadData and MetaPhlAn2. Results: Fourteen sFPIP infants were compared to 55 healthy infants. At inclusion overall microbial composition of sFPIP infants differed significantly from controls (weighted UniFrac; Pairwise PERMANOVA, P = 0.002, pseudo-F = 5.008). On genus level healthy infant microbiota was significantly enriched with Bifidobacterium (B) compared to sFPIP patients (linear discriminant analysis [LDA] = 5.5, P < 0.001, 31.3% vs 12.1%). sFPIP stool was significantly enriched by Clostridium sensu stricto 1 over controls (LDA = 5.3, P = 0.003, 3.5% vs 18.3%). DDI caused a significant and sustained increase of Bifidobacterium (LDA = 5.4, P = 0.048, 27.9%) in sFPIP infants. Species level analysis revealed significant reduction of abundance of B longum in sFPIP patients, which after DDI was reversed by B. species other than B longum. Conclusions: We revealed a gut microbiota dysbiosis phenomenon in sFPIP infants. DDI induces a microbiota composition comparable to that of healthy infants. In most sFPIP infants hematochezia might be triggered by a gut microbiota dysbiosis phenomenon.