(833) Using Next-Generation Sequencing to Examine Pathogen Kinetics and Detection in Complicated Pneumonia

Introduction: Pediatric pneumonia can be severe and result in complications, including empyema and lung abscess formation. Conventional blood and pleural cultures often have limited sensitivity. Next-generation sequencing (NGS) may broadly detect pathogens from multiple body fluids despite antibiotic pre-exposure; however, the optimal timing and impact of sample type on diagnostic yield is unknown. The goal of this study is to understand how the detection of bacterial DNA changes over time in samples with varying proximity to the lungs.

Methods: This is a prospective, single-center descriptive pilot study conducted at a tertiary children’s hospital from July 2022 to June 2023. Children aged 3 months through 17 years admitted to the pediatric ICU with a primary diagnosis of complicated pneumonia, who were endotracheally intubated and/or had a tube thoracostomy were eligible. Plasma, endotracheal, nasopharyngeal, and pleural fluid samples were collected at three time points during hospitalization. After separate DNA and RNA extraction, combined libraries were prepared, enriched with RPIP (Illumina), sequenced and quantitative organism detections were analyzed after discharge.

Results: Conventional microbiological culture identified a putative organism in 9 of 10 patients. Pathogens included: S. pneumoniae (2), H. influenzae (1), S. aureus (2), Streptococcus spp. (3), P. aeruginosa (1). The yield of pleural cultures was highest (8), followed by respiratory culture (5) and blood culture (2). NGS identified the same bacterial pathogen as traditional testing in all samples evaluated to date. The treated organism was detected by NGS in upper respiratory samples (NP swabs and/or sputum), plasma, and pleural fluid at the earliest timepoint. NGS detected more bacterial diversity than culture in nearly all samples, but bacterial diversity was greatest in the NP swabs. NGS made additional viral detections not detected by standard PCR panels.

Conclusions: NGS detected a putative pathogen in all samples regardless of pre-treatment or time collected. Conventional culture methods only identified the pathogen reliably in invasively obtained pleural fluid or endotracheal aspirates. Future application of NGS may allow for non-invasive pathogen detection at a broader range of time points and more targeted antibiotic coverage.