513 A Survey of Bacterial Microbiome Dysbiosis in Burn ICU Patients

Abstract

With the advancement of nucleic acid sequencing technologies and bioinformatic power, a boom of the “-omic” era has revolutionized how we understand host microbiome interactions. Dysbiosis of the skin and gut microbiota have correlated with many immune disorders and chronic diseases. These associations are often not universal, making it critical for unique populations to be studied individually. Numerous aspects of the response to burn injury and resulting care can influence normal commensal bacteria on the skin and gut. Healthy volunteers were swabbed in two locations on each forearm for validation of non-invasive skin bacterial sampling techniques, with a subset utilized for bacterial population comparison with burn patient samples. Patients with >10% total body surface area (TBSA) burns were sampled using sterile swabs on the burn wound and a corresponding spared site of the skin, as well as one peri-anal and rectal swab. These four locations were sampled on days 0, 3, 7, 14, and 21 of the patient’s ICU stay or up to discharge. Metadata collected includes demographics, medications, nutrition plan, and progress notes. A next generation 16S amplicon sequencing protocol was utilized in conjunction with bioinformatics performed using Nephele under the QIIME pipeline. Twenty healthy volunteers were recruited for method validation. Four male burn patients were recruited with a mean age of 42 (SD=13.4) years and mean total body surface area (TBSA) burn of 24.1% (SD=8.1). Comparison of healthy volunteer skin swabs (n=18) to burn wound swabs (n=12) found a significant reduction in percent relative abundance of the phylum Actinobacteria (p<0.001) and increase in the order Bacillales (p<0.05), composed of primarily Staphylococcus. Varying abundance of non-culturable bacteria were sequenced. The Shannon Diversity Index was most variable among wound swabs (n=12) (mean [SD]: 4.09 [2.15]). No significant difference between pooled taxa percent abundance for rectal and peri-anal sampling (p=0.297) was observed. Proportions of bacterial phyla that are “normal” colonizers of the skin, such as Actinobacteria and Firmicutes, are perturbed on the burn wound and spared skin sites sampled from the four subjects. The clinical significance of non-culturable bacteria remains unknown. These findings suggest that bacterial dysbiosis occurs on the burn wound following injury and remains perturbed during routine care. Rectal and peri-anal swabs did not significantly differ in taxa relative abundance, indicating that less invasive sampling could be used. Profiling microbial populations immediately post burn and throughout care is the first step in elucidating mechanisms by which bacterial dysbiosis occurs and if this perturbation impacts clinical outcomes.