D14-06 Lung Microbiome Changes in Varying Stages of Pediatric Acute Respiratory Distress Syndrome

Abstract Pediatric Acute Respiratory Distress Syndrome (pARDS) is caused by a disruption in the alveolar-epithelial barrier either by a direct or indirect insults. Recent studies including the PARDIE study demonstrated a stepwise increase in mortality with increasing disease severity. ARDS severity is classified as mild, moderate or severe. The lung-gut axis has benn studied for it’s role in severity of ARDS. Respiratory disease progression is linked to the changes in status of the lung microbiome from eubiosis to dysbiosis. Objective The objective of this study was to evaluate changes in the lung microbiome with worsening ARDS severity. The hypothesis was that patients with severe ARDS will have a less diverse microbiome as compared to those patients with mild or moderate disease. Design/Methods In this single-center, prospective, longitudinal study, patients aged 1 months to 23 years who met standard criteria for ARDS were enrolled over a 18-month period. This study was approved by the local IRB. ARDS severity based on levels of hypoxemia were assessed by a P/F or S/F Ratio. Those with diagnoses of chronic respiratory failure or airway disease were excluded. A nasopharyngeal swab was collected in TrizolTM solution and stored at -80o for subsequent analysis. Microbiome analysis was conducted by 16s rRNA sequencing. Krona based microbial composition data and Principle Component Analysis (PCA)- based beta-diversity analyses were used to compare microbiome structure across ARDS severity (mild, moderate, severe). Results 25 children who satisfied inclusion/exclusion criteria were enrolled. 1 patient was unenrolled due to inadequate samples. Median (IQR) age was 14 (3, 17) years, 56% were male. Median BMI was 20.2 (15.7, 26.4). 11 subjects (46%) were classified as mild, 9 (37%) as moderate, and 4 (17%) as severe ARDS. Comparing mild to moderate disease, there is partial overlap seen on the PCA analysis. Krona analysis demonstrates dominance of Bacteroidetes which typically represents healthy airway microbiome. Comparing moderate to severe disease, PCA separation is more pronounced. Krona data reveal firmicutes dominance and emerging proteobacteria. Comparing mild to severe disease, substantial separation is seen in PCA analysis reflecting significant divergence in microbiome diversity. There is a significant drop in Bacteroidetes and rise in Firmicutes and Actinobacteria. When comparing Mild/Moderate to Severe disease there is again significant PCA separation. Firmicutes and actinobacteria expand while proteobacteria emerge reflecting dysbiosis in the severe ARDS group. Conclusions This pilot study of the lung microbiome in pediatric ARDS demonstrates dysbiosis with increasing severity of ARDS. This abstract is funded by: None