Use of a molecular syndromic panel for the etiological diagnosis of bloodstream infections in critically ill patients: impact on clinical outcomes and antimicrobial use from a multicenter, prospective study

Background: Bloodstream infections (BSI) frequently occur in critically ill patients and remain associated with high mortality rates. Methods: A prospective, observational, multicenter investigation was carried out through the MULTI-SITA platform to evaluate how the addition of a molecular syndromic panel, the BIOFIRE® Blood Culture Identification 2 (BCID2) Panel (BCID2 panel), to conventional blood culture diagnostics influenced key clinical endpoints and antimicrobial use. The primary endpoint was all-cause mortality within 30 days following the onset of BSI. Results: Among the 126 included critically ill patients with BSI, multivariable analysis identified SOFA score (hazard ratio [HR] 1.21, 95% confidence interval [CI] 1.09-1.34, p < 0.001), other concomitant infections requiring antibiotic therapy (HR 2.61, 95% CI 1.18-5.79, p = 0.018), and BSI caused by Acinetobacter spp. (HR 4.35, 95% CI 1.10-17.18, p = 0.036), Staphylococcus aureus (HR 3.38, 95% CI 1.26-9.08, p = 0.016), or Stenotrophomonas maltophilia (HR 15.97, 95% CI 2.43-104.54, p = 0.004) as associated with increased mortality, while no association was observed between use of the BCID2 panel and mortality (HR 0.97, 95% CI 0.42-2.25, p = 0.941). In secondary analyses, the use of the BCID2 panel resulted in more events of either early antimicrobial de-escalation or initiation of appropriate therapy at day one from BSI onset compared with a standard blood culture workflow (55.6% vs. 31.7%, p = 0.015). Conclusion: In critically ill patients with BSI, implementation of a molecular syndromic panel influenced antimicrobial management decisions without affecting mortality. Additional investigations are warranted to clarify the long-term antimicrobial stewardship implications of treatment strategies guided by molecular syndromic panels-based diagnostic workflows.