Host transcriptional signatures predict etiology in community-acquired pneumonia: Potential antibiotic stewardship tools

Research output: Contribution to journalJournal articlepeer-review

Documents

  • William W Siljan
  • Dhanasekaran Sivakumaran
  • Christian Ritz
  • Synne Jenum
  • Tom Hm Ottenhoff
  • Elling Ulvestad
  • Jan C Holter
  • Lars Heggelund
  • Harleen MS Grewal

Background: Current approaches for pathogen identification in community-acquired pneumonia (CAP) remain suboptimal, leaving most patients without a microbiological diagnosis. If better diagnostic tools were available for differentiating between viral and bacterial CAP, unnecessary antibacterial therapy could be avoided in viral CAP patients.

Methods: In 156 adults hospitalized with CAP classified to have bacterial, viral, or mixed viral-bacterial infection based on microbiological testing or both microbiological testing and procalcitonin (PCT) levels, we aimed to identify discriminatory host transcriptional signatures in peripheral blood samples acquired at hospital admission, by applying Dual-color-Reverse-Transcriptase-Multiplex-Ligation-dependent-Probe-Amplification (dc-RT MLPA).

Results: In patients classified by microbiological testing, a 9-transcript signature showed high accuracy for discriminating bacterial from viral CAP (AUC 0.91, 95% CI 0.85-0.96), while a 10-transcript signature similarly discriminated mixed viral-bacterial from viral CAP (AUC 0.91, 95% CI 0.86-0.96). In patients classified by both microbiological testing and PCT levels, a 13-transcript signature showed excellent accuracy for discriminating bacterial from viral CAP (AUC 1.00, 95% CI 1.00-1.00), while a 7-transcript signature similarly discriminated mixed viral-bacterial from viral CAP (AUC 0.93, 95% CI 0.87-0.98).

Conclusion: Our findings support host transcriptional signatures in peripheral blood samples as a potential tool for guiding clinical decision-making and antibiotic stewardship in CAP.

Original languageEnglish
JournalBiomarker Insights
Volume17
Issue numberJanuary-December
Pages (from-to)1-8
Number of pages8
ISSN1177-2719
DOIs
Publication statusPublished - 2022

Bibliographical note

© The Author(s) 2022.

    Research areas

  • Faculty of Science - Pneumonia, Gene expression signatures, Bacteria, Viruses, Antimicrobial stewardship, Clinical decision-making

Number of downloads are based on statistics from Google Scholar and www.ku.dk


No data available

ID: 310222190