Abstract
Background
Broad-range polymerase chain reaction (PCR) sequencing is a promising tool for diagnosis of infectious conditions when traditional microbiologic strategies fail to identify a pathogen. Data on the optimal clinical scenarios in which to use this tool are limited.
Methods
We assessed the rate of organism identification from broad-range PCR testing sent from our quaternary care children’s hospital between March 2017 and June 2020. We completed a retrospective chart review to evaluate patients’ baseline demographic and clinical features as well as clinical significance of results (defined as influencing antimicrobial management) by specimen type.
Results
Among 184 total samples, 111 (60%) were obtained from immunocompromised patients. The median age of patients at the time of sample collection was 11.4 years (IQR 6.5-16.0). 128/181 (71%) samples were from patients known to be on ≥ 1 antimicrobial, including prophylaxis, in the 24 hours prior to sample collection. 52/184 (28%) patients ultimately had an infectious disease diagnosed by another testing modality. The most common PCR sample types were bronchoalveolar lavage (BAL) fluid (35), lung tissue (20), and bone (14). An organism was identified from 41 (22%) samples, but positive results for only 8 samples (4%) led to a change in antimicrobial management: addition of agents in 4 cases, cessation of agents in 2, and transition from one agent to another in 2. Negative results for 3 (2%) samples led to discontinuation of antimicrobials. Organisms were identified from 11 (31%) BAL samples, of which only 2 (6%) were judged to be clinically significant. No results from lung tissue, CSF (11), skin biopsies (6), or joint fluid (4) affected antimicrobial management.
Conclusion
We found that only 6% of broad-range PCR results influenced antimicrobial management in a diverse pediatric cohort. Our findings suggest that many positive results, especially in BAL fluid, do not lead to changes in antimicrobial management. Additional work is necessary to characterize the ideal clinical scenarios in which broad-range PCR should be used as over a quarter of patients had a causative infectious disease identified by another modality.
Disclosures
All Authors: No reported disclosures