Piglet cardiopulmonary bypass induces intestinal dysbiosis and barrier dysfunction associated with systemic inflammation

Abstract

The intestinal microbiome is essential to human health and homeostasis, and is implicated in the pathophysiology of disease, including congenital heart disease and cardiac surgery. Improving the microbiome and reducing inflammatory metabolites may reduce systemic inflammation following cardiac surgery with cardiopulmonary bypass (CPB) to expedite recovery post-operatively. Limited research exists in this area and identifying animal models that can replicate changes in the human intestinal microbiome after CPB is necessary. We used a piglet model of CPB with two groups, CPB (n=5) and a control group with mechanical ventilation (n=7), to evaluate changes to the microbiome, intestinal barrier dysfunction and intestinal metabolites with inflammation after CPB. We identified significant changes to the microbiome, barrier dysfunction, intestinal short-chain fatty acids and eicosanoids, and elevated cytokines in the CPB/deep hypothermic circulatory arrest group compared to the control group at just 4 h after intervention. This piglet model of CPB replicates known human changes to intestinal flora and metabolite profiles, and can be used to evaluate gut interventions aimed at reducing downstream inflammation after cardiac surgery with CPB.

Keywords: Barrier dysfunction; Congenital heart disease; Inflammation; Microbiome; Short-chain fatty acid.

Fig. 1.

Relative bacterial abundance between CPB/DHCA group and controls. (A) Bacteria at the phylum level in each sample pre- and post-surgery for the control group and the CPB/DHCA group. There is a slightly larger increase in the amounts of Proteobacteria in the CPB/DHCA group pre-operative to post-operative samples compared to the control group. (B) Bacteria at the genus level. The legend identifies SCFA-producing organisms, which are reduced in the CPB/DHCA group post-operative samples compared to the control group post-operative samples. CPB, cardiopulmonary bypass; SCFA, short-chain fatty acid.

Fig. 2.

α- and β-diversity plots in CPB/DHCA group and controls. (A) Observed operational taxonomic units (OTUs) in the CPB/DHCA group compared to the control group. There were no statistically significant differences in the total number of bacteria present between the two groups. (B) Phylogenetic diversity between the CPB/DHCA group and the control group. There was a significant decrease in phylogenetic diversity in the CPB post-operative samples compared to the control post-operative samples. (C) β-diversity via UniFrac distance matrix. There was a statistically significant difference in the β-diversity in the CPB group compared to the control group. The numbers indicate P-values using unpaired Wilcoxon rank sum test. PCoA, principal coordinates analysis.

Fig. 3.

LEfSE plot and cladogram of bacterial associations in CPB/DHCA group and controls. (A) LEfSE plot providing organisms associated with either the CPB/DHCA group (green) or the control group (CNT, red). The logarithmic score details the strength of the association of each organism to a specific group. (B) Cladogram of the LEfSE analysis with organisms in the shaded green area associating more strongly with the CPB/DHCA group and organisms in the shaded red area associating with the control group. The microbial compositions were compared at different taxonomic levels. LDA, linear discriminate analysis.

Fig. 4.

Changes in markers of EBD, cytokines and SCFAs between the CPB/DHCA group and controls. (A) A significant increase was seen in FABP2, claudin-2 and claudin-3 levels between the pre-operative and post-operative samples in the CPB group compared to controls. (B) A significant increase was seen in IL-1β, IL-6 and TNF-α levels between the pre-operative and post-operative samples in the CPB/DHCA group compared to controls. (C) A significant reduction was seen in acetic acid, butyric acid and propionic acid levels between the pre-operative and post-operative samples in the CPB group compared to controls. Two-way ANOVA with Holm–Sidak's multiple comparisons test was performed. FABP2, fatty acid-binding protein 2; IL, interleukin; TNF, tumor necrosis factor. ns, not significant; *P<0.05; **P<0.01; ****P<0.0001.

Fig. 5.

Changes to stool eicosanoids between CPB/DHCA group and controls. (A) A selection of eicosanoids with variation between the CPB/DHCA group and the control (CNT) group. Boxes indicate the 25-75th percentiles, whiskers show 1.5 times the interquartile range, the central line marks the median, and the mean is indicated by the yellow diamond. (B) A heatmap plot of the association of various eicosanoids with both pre-operative and post-operative samples of the CPB/DHCA group and controls. The Mann–Whitney U test was used to assess differences between the two groups. Conc, concentration; HETE, Hydroxyeicosatetraenoic acid; DiOH-PGF2, dihydro-prostaglandin 2; HOTrE, hydroxyoctadecatrienoic acid.

Fig. 6.

Canonical correlation analysis of the microbiome with EBD, cytokines and SCFA. (A) Network map (top) and heatmap (bottom) of the markers of EBD and associated organisms. (B) Network map (top) and heatmap (bottom) of inflammatory cytokines and associated organisms. (C) Network map (top) and heatmap (bottom) of SCFAs and associated organisms.

Fig. 7.

Mediation analysis of the microbiome on changes to EBD, cytokines, SCFAs and eicosanoids. (A) The three outcomes, PGD2, PGE2 and valeric acid, to be mediated by changes in the microbiome. Exposure is CPB, the mediator is the microbiome, and the outcomes are listed above. (B) The three markers of EBD, FABP2, claudin-2 and claudin-3, depicting no statistically significant mediation effect of the microbiome on the changes in EBD. CNT, control; FABP2, fatty acid-binding protein 2. Blue lines indicate the association between the microbiome and individual biomarkers using principal component axis 1, and shaded gray areas represent the 95% confidence intervals.