Airway Microbiome Modulates Corticosteroid Delivery Induced Recovery After Inhalational Burn Injury

Introduction:: Laryngeal burn injury associated with the inhalation of smoke results in dysphonia, dysphagia, and difficulty breathing, which significantly impacts the quality of life. Clinical approaches to treat these complications involve intubation, however, endotracheal tubes (ETTs) increase the risk of bacterial adhesion and biofilm formation on the surface of the ETT. Additionally, the post-burn environment and damage to the airway epithelium reduces mucociliary clearance and the high inflammation often leads to the development of airway stenosis and contributes to fibrotic scarring in the tissue. Corticosteroids are the clinical standard for suppressing chronic swelling and long-term fibrosis with little to no understanding of how it affects the laryngeal microbiome.

It has been recognized that microbiota composition correlates to respiratory conditions demonstrating there are changes in community diversity amongst specific microbial strains and diseases. The objective of the study was to assess changes in the upper respiratory microbiome with inhalational burn injury and ETT placement in a physiologically relevant swine model. A secondary aim was to evaluate the effect of localized corticosteroid delivery via a drug-eluting ETT to combat inflammation and alter the diversity of microbiota in the evolving treatment of the injury.


Materials and Methods:: Inhalational injury was simulated in 18 Yorkshire crossbreed swine by administering heated air (150-160°) in alternating 30-second intervals for 5 minutes. This injury and healing model was approved by the U.S. Air Force 59th Medical Wing Institutional Animal Care and Use Committee (Protocol FWH20210102AR). After the injury, ETTs with or without dexamethasone-loaded polycaprolactone (PCL) electrospun fibers were secured within the larynx for 3 or 7 days (n=4 per time/ETT type). Laryngeal tissue was sectioned along the mid-region of the vocal fold, stained with hematoxylin and eosin (H&E), and blindly reviewed by pathologists for the assessment of inflammation (1: mild, 2: moderate, 3: severe). Swabs were collected from the airway before and after injury in addition to the surface of the ETT following the end of study. The V3-V4 region of the 16S rRNA gene was sequenced (Illumina MiSeq platform) and an amplicon sequence variant (ASV) table was produced using the DADA2 pipeline in R Studio. Relative abundance was evaluated at the phylum level and the alpha diversity was assessed with Shannon indices. The beta diversity was evaluated using Bray-Curtis dissimilarity index and visualized with principal coordinate analysis (PCoA). Statistical analysis for alpha diversity was conducted using two-way ANOVA by ETT Type and duration of placement for each swab cohort (pre-injury, post-injury, and ETT surface) followed by Tukey’s post hoc test used to determine significant differences at p < 0.05. Permutational analysis of variance (PERMANOVA) based on Bray-Curtis distance was conducted to assess differences in community composition.


Results, Conclusions, and Discussions:: Histological outcomes are presented in Figure 1. Pathology assessment indicated an increased degree of inflammation in the groups with dexamethasone-coated ETT placement in comparison to regular ETT placement at 7 days. While corticosteroids are used for their anti-inflammatory effects, we suspect the fiber coating may cause increased airway irritation from the rough surface. Additionally, we did not assess if the inflammation/macrophages observed were pro-regenerative. The relative abundance of phylum varies amongst pre-injury, post-injury, and ETT groups (Figure 2A). Firmicutes were the predominant community of phyla within the upper airway. Proteobacteria has previously been found to flourish in inflammatory environments, however, we distinguished a decreased abundance in Proteobacteria before and after burn injury. There was similar composition observed between the post-injury and ETT groups with increased Fusobacteria from 3 to 7 days.

There were no significant differences in alpha diversity (Figure 2B) when considering the duration and ETT type in pre-injury and post-injury groups. When observing the ETT groups, dexamethasone-coated ETTs had a significantly higher Shannon diversity than the regular ETTs. These differences may be contributed to the structure of the electrospun fiber coating which offers increased surface area contact and allows for greater bacterial adhesion. The PCoA plot showed 36.6% of the total variance between the samples (Figure 2C). Differences in community composition based on Bray-Curtis distance were statistically significant between groups (p< 0.05) indicating microbial changes with injury, ETT type, and time. The intra-variability between samples collected post-injury was greater than pre-injury samples regardless of ETT type. These outcomes demonstrate that inflammation impacts diversity, especially over longer periods of time.

The microbial diversity of the surface of ETTs differed from the airway and there were changes between groups over time and with dexamethasone delivery. Our findings lead us to believe inflammation caused by laryngeal burn injury and intubation can directly modulate the bacterial composition and diversity of the upper respiratory microbiome. We further believe that this modulation directly impacts the regenerative environment of the upper airway when paired with the localized delivery of corticosteroids in changing the temporal progression of inflammation.


Acknowledgements (Optional): : This work is supported in part by the US Department of Defense through the 59MDW/ Office of the Chief Scientist Office (AC20COV10 and AC19EC0), the National Science Foundation CBET#1847103, and the Jacobson endowment at the University of Texas at San Antonio.


References (Optional): :