(864) Intestinal Microbiome Composition and Clinical Outcomes in Infants After Cardiopulmonary Bypass

Introduction: Infants who require cardiopulmonary bypass (CPB) often have immune dysfunction causing post-op infection and death. The intestinal microbiome modulates immune response, but its role has not been studied in infants after CPB. We designed a pilot study to describe the composition of the intestinal microbiome in infants after CPB and its relationship with clinical outcome. We hypothesized that infants with postop infection would have decreased abundance of anaerobes in the Bifidobacterium genus and increased abundance of bacteria from the Enterococcus genus compared to infants without infection.

Methods: We prospectively enrolled 33 infants < 1 year of age who underwent cardiac surgery with CPB. We collected serial fecal samples after CPB. We described bacterial density and performed 16S rRNA amplicon sequencing.

Results: We enrolled 33 infants, 9 of whom were neonates. Ten had a complicated course (6 with infection), and median ICU length of stay was 9 days. Samples were collected a median of 6.5 days after bypass. The mean abundance of DNA of fecal samples (by ddPCR) was lower in neonates. Neonates with infection had lower DNA concentration (p < 0.001) than those without infection. Community composition (by PERMANOVA) was significantly different in infants with postop infection (p = 0.022). Random forest modeling identified Klebsiella and Dysgonomonas as discriminating taxa in infants with infection. We determined that Klebsiella was enriched in infants without infection (p < 0.001), while Dysgonomonas trended towards increase in infants with infection (p = 0.10). Other measures of community composition, including abundance of Bifidobacterium and Enterococcus genera, did not differ by infection status.

Conclusions: Infants who develop infection following surgical repair of congenital heart disease requiring cardiopulmonary bypass have different composition of their intestinal microbiome than those without infection. We confirmed the feasibility of longitudinally studying gut microbiota in this understudied population and the biologic plausibility of composition of the gut microbiome explaining variable risk of infection. Understanding features of the microbiome that cause immune dysregulation and postoperative infection is the next step in reducing morbidity in this vulnerable population.