415 - Antibiotic Resistance in Children with Tracheostomy and Home Ventilation: CDC Serious Threats

Publication Number: 2403.415

Anshu Punreddy, George Washington University School of Medicine and Health Sciences, Naples, FL, United States; Brennan Harmon, Children's National Health System, Washington, DC, United States; Aszia N. Burrell, Children's National research institute, Columbia, MD, United States; Marianna Said, Children's National Health System, Arlington, VA, United States; Francisco E. Herrera Carrillo, Boston Children's Hospital, Brookline, MA, United States; Robert J. Graham, Boston Children's Hospital, Boston, MA, United States; Marcos Pérez Losada, The George Washington University, Washington, DC, United States; Matthew Ross, Baylor College of Medicine, Houston, TX, United States; Kristi Hoffman, Baylor College of Medicine, Houston, TX, United States; Jonathan Cogen, University of Washington School of Medicine, Seattle, WA, United States; Rebecca Steuart, Medical College of Wisconsin, Milwaukee, WI, United States; Jonathan Mansbach, Boston Children's Hospital, Boston, MA, United States; Andrea Hahn, Children's National Hospital, Washington, DC, United States

Background: Acute respiratory infections (ARI) are the #1 cause of death and hospitalization in children with tracheostomy. Pseudomonas aeruginosa is among the most frequently isolated pathogens, is only susceptible to a limited number of antibiotics, and has multidrug antibiotic resistance, placing it on the Centers for Disease Control and Prevention (CDC) serious threat list.

Objective: Using paired tracheal aspirates collected before and during ARIs from children with tracheostomy and home mechanical ventilation (HMV), our study objectives were to 1) identify the most abundant antibiotic resistance genes (ARGs) in tracheal aspirates with a focus on P. aeruginosa, and 2) assess if ARGs were associated with ARI severity.

Design/Methods: We enrolled 312 children with HMV from 13 hospitals in the Multicenter Tracheostomy Collaboration (MyTRaCh) study. Using serial tracheal aspirates, we tested paired pre-ARI (0-31 days before ARI) and ARI (day 0-7 of symptoms) samples using full-length 16S rRNA gene and RNA sequencing. The
presence of ARGs was determined using ShortBRED, a protein profiling software. Bacteria associated with each ARG was determined using the Comprehensive Antibiotic Resistance Database. Differential abundance of ARGs across groups was determined using DESeq2.

Results: Among 63 children there were 91 ARIs with pre-ARI and ARI samples. The median age at study enrollment was 7.5 years (IQR 3.0-13.4 years), and 62.2% of children were male. The two bacteria with the highest relative taxonomic abundance across all samples were Serratia marcescens (19%) and P. aeruginosa (6%). P. aeruginosa was detected in 100% of samples. While 3% of ARG reads (12 genes) were associated with S. marcescens, 47% of reads (70 genes) were associated with P. aeruginosa. Although the abundance of ARGs did not significantly differ between pre-ARI and ARI, several P. aeruginosa ARGs were higher in those who visited the emergency department (Fig 1A) or required hospitalization (Fig 1B) at the time of ARI
(unadjusted p-value < 0.05). Most importantly, 30% of ARG reads (41 genes) encode for multidrug efflux pumps associated with multidrug-resistant P. aeruginosa (Table 1).

Conclusion(s): In a prospective, multicenter cohort of children with HMV, we detected P. aeruginosa in all tracheal aspirates. Almost 1/3 of ARGs detected cause multidrug resistance in P. aeruginosa, which is on the CDC "serious threat" list. These results emphasize the importance of identifying colonization versus infection to limit unnecessary antibiotic use against P. aeruginosa and prevent the furthering of antibiotic resistance.

Figure 1. Differential abundance of antimicrobial resistance genes based on severity of acute respiratory infection (ARI).
Panel A. Differential abundance of AMR genes at time of ARI based on need for ED visit; unadjusted p-value <0.05. Two of these four genes are associated with Pseudomonas aeruginosa. Panel B. Differential abundance of AMR genes at time of ARI based on need for hospitalization; unadjusted p-value <0.05. Six of these seven AMR genes are associated with P. aeruginosa. Positive log fold changes are associated with increased abundance in ED visit or hospitalization, whereas negative log fold changes are associated with decreased abundance in ED visit. Figure generated using Excel and Biorender.com.

Table 1. High-risk antibiotic resistance genes detected in acute respiratory infection.
In our dataset, the majority of antibiotic resistance genes (ARGs) identified as serious threats by the Centers for Disease Control and Prevention (CDC) were genes that encode for multidrug efflux pumps within Pseudomonas aeruginosa. This was despite P. aeruginosa only being the second most common bacteria identified at 6% relative abundance, while Serratia marcescens was the most abundant bacteria identified with a relative abundance of 19%. Streptococcus pneumoniae was present in our dataset with a relative abundance of 1%.