The effect of selective decontamination on the intestinal microbiota as measured with IS-pro: a taxonomic classification tool applicable for direct evaluation of intestinal microbiota in clinical routine

Abstract

Selective decontamination of the digestive tract (SDD) is aimed at elimination of potential pathogenic microorganisms. In this study, the effect of SDD on gut microbiota was evaluated in a large homogenous group of elective colorectal cancer surgery patients. Rectal swabs were taken from 118 patients undergoing colorectal surgery. These patients were randomly assigned to receive perioperative SDD or to the control group (no SDD). Rectal swabs were taken prior to surgery, 3 days after commencing administration of SDD. Gut microbial profiles were obtained with the IS-pro technique, a standardized microbiota profiling assay applicable in clinical routine. Differences in abundance for different taxonomical groups and diversity between the groups were assessed. Unsupervised and supervised classification techniques were used to assess microbial signatures, differentiating between the SDD group and the control group. Patients in the SDD group had different gut microbial signatures than in the control group, also in phyla that are not a target for SDD. Escherichia coli, Sutterella spp., Faecalibacterium prausnitzii, and Streptococcus spp. were the species that differed the most between the two groups. The SDD group showed clustering into two subgroups. In one subgroup, a decrease in Proteobacteria was observed, whereas the other subgroup showed a shift in Proteobacteria species. This study shows that SDD not only decreases colonization of the gastrointestinal tract with potential pathogenic Gram-negative microorganisms, but also reduces the abundance of normal colonizers of our gastrointestinal system and leads to a shift in total microbiota composition.

Keywords: Colorectal cancer surgery; IS-pro; Microbiota; SDD.

Fig. 1

Phylum abundance analysis. Boxplot of the phylum abundance analysis. Total Proteobacteria and FAFV group abundance was significantly reduced in the SDD group versus the control group (P = 0.0002, respectively P<0.0001)

Fig. 2

UPGMA clustering analysis. Unsupervised clustering based on Proteobacteria, depicted as a heat map of all microbiota profiles per patient, colored by phylum. A clear subclustering into two subgroups for SDD is shown. One SDD subgroup shows reduction of Proteobacteria, compared to controls. And one SDD subgroup showed no clear reduction in Proteobacteria, but a shift in Proteobacteria composition

Fig. 3

Partial least squares discriminant analysis for all phyla, Proteobacteria, Bacteroidetes, and the FAFV group. PLS-DA showed clear separation of SDD samples versus control samples, based on total microbiota analysis

Fig. 4

Abundance bar chart of the 4 species, discriminating most between SDD and control samples. Samples of the SDD group were associated with a decreased abundance of E. coli, F. prausnitzii, Sutterella spp., and Streptococcus spp.