Managing multidrug resistant Gram-negative bacteraemia in immunocompromised hosts

Le résumé fourni par la source

PURPOSE OF REVIEW: Multidrug-resistant Gram-negative bloodstream infections (MDR-GNBSI) are increasingly frequent in immunocompromised hosts, particularly solid organ transplant (SOT) recipients and patients with haematological malignancy (HM) receiving intensive chemotherapy and/or cellular therapies. We summarize current evidence and practical approaches across empirical, quasi-targeted and targeted management. RECENT FINDINGS: Contemporary cohorts show rising rates of MDR-GNBSI in both SOT and HM patients. Appropriate therapeutic management is key to ensure good clinical outcome while minimizing ecological impact. Risk-adapted empirical therapy usually incorporates prior colonization and local epidemiology plus individual risk factors, thus screening could play central role. Rapid diagnostic testing can reduce time to appropriate and proportionate therapy, but currently the majority of them provide genotypic resistance that in Gram negatives can be difficult to interpret, in addition off-target resistance mechanisms should be considered mainly in nonfermenter bacteria. Thus, machine-learning models and structured BSI bundles are emerging tools to support early decision-making and stewardship. For targeted therapy, novel β-lactam/β-lactamase inhibitor combinations and cefiderocol expand options for carbapenem-resistant strains; pharmacokinetics/pharmacodynamics (PK/PD) optimization may improve target attainment and mitigate resistance selection. Evidence supporting oral step-down and shorter courses is growing, whereas routine follow-up blood cultures remain controversial due to observational bias. SUMMARY: MDR-GNBSI management in immunocompromised patients requires individualized, data-driven strategies integrating risk stratification, rapid microbiology, and PK/PD optimization. Pursuing antimicrobial stewardship principles in these populations remains a priority.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.