Annals of Intensive Care (Mar 2025)

Relative faecal abundance to predict extended-spectrum β-lactamase-producing Enterobacterales related ventilator‑associated pneumonia

  • Pierre Bay,
  • Paul-Louis Woerther,
  • Vincent Fihman,
  • Ségolène Gendreau,
  • Pascale Labedade,
  • Antoine Gaillet,
  • Florian Jolly,
  • Guillaume Carteaux,
  • Nicolas de Prost,
  • Jean-Winoc Decousser,
  • Armand Mekontso-Dessap,
  • Keyvan Razazi

DOI
https://doi.org/10.1186/s13613-025-01456-w
Journal volume & issue
Vol. 15, no. 1
pp. 1 – 12

Abstract

Read online

Abstract Background Antimicrobial stewardship (AMS) for ventilator-associated pneumonia (VAP) in carriers of extended-spectrum β-lactamase-producing Enterobacterales (ESBL-E) presents significant challenges. The abundance of ESBL-E rectal carriage has emerged as a potentially valuable tool for predicting ESBL-E-related VAP. Methods This single-center, retrospective study was conducted between October 2019 and April 2023 in the medical ICU of a university hospital. The relative abundance of ESBL-E rectal carriage (RAC) was calculated as the ratio of ESBL-E counts to the total number of aerotolerant bacteria. The aim was to evaluate the predictive value of RAC for diagnosing ESBL-E-related VAP in patients with confirmed VAP who were ESBL-E carriers. Results During the study period, 478 patients with ESBL-E carriage were admitted to the ICU, of whom 231 (48%) required mechanical ventilation. Eighty-three patients (17%) developed a total of 131 confirmed VAP episodes, of which 62 episodes (47%) were ESBL-E-related VAP. The median interval between the last rectal screening and VAP occurrence was 4 [3–7] days. RAC was not associated with ESBL-E-related VAP in the entire cohort (p = 0.39). Similar findings were observed in several sensitivity analyses, including the following subsets: recent and high-quality screening (interval between screening and VAP ≤ 7 days and bacterial load on rectal swab > 104 CFU/mL, p = 0.21); first VAP episodes only (p = 0.41); cases involving Escherichia coli exclusively (p = 0.08) or other ESBL-E strains (p = 0.29); and VAP associated with Gram-negative bacteria (p = 0.26) or Enterobacterales (p = 0.34). However, in a multivariable model, rectal colonization with non-Escherichia coli ESBL strains was independently associated with ESBL-E-related VAP (adjusted odds ratio [aOR] 1.213 [95% CI 1.005–1.463], p = 0.045). Conclusion RAC was not associated with confirmed VAP in ESBL-E carriers. Further studies are needed to explore effective strategies for improving AMS in ESBL-E carriers with suspected VAP.