Nosocomial sepsis: how to reconcile immediate broad-spectrum therapy with minimal ecological pressure and healthcare cost.

Purpose of review: To investigate the potential association between colonization of the rectal and oropharyngeal mucosa by multidrug-resistant (MDR) Gram-negative bacilli and the subsequent nosocomial sepsis due to the same pathogen in order to provide a rational basis for early de-escalation when standard clinical samples are negative.

Recent findings: Compelling metagenomic data shows that profound shifts in gut and respiratory microbiota occur over time in the context of antibiotic therapy, critical illness and intubation leading to predominance of P. aeruginosa and MDR-Enterobacterales. Detection of these microorganisms through culture or molecular methods in mucosal swab samples is associated with a clinically relevant risk of subsequent nosocomial sepsis caused by the same pathogens. Conversely, their absence confers a high negative predictive value (NPV, >95%) for infection due to these microorganisms.

Summary: In settings with a high prevalence of antimicrobial resistance, the empirical use of broad-spectrum antibiotics in sepsis is often necessary. However, in culture-negative sepsis, these agents are frequently continued to the full treatment duration, entailing potential collateral damage and a significant economic burden. In this context, clinical evidence suggests that failure to detect P. aeruginosa or MDR-Enterobacterales carries a high NPV for subsequent infection by these microorganisms. We propose an algorithm that ensures adequate empirical coverage while enabling antibiotic de-escalation in culture-negative cases based on colonization status.