Skip to main content
  • Poster presentation
  • Open access
  • Published:

De-escalation of antimicrobial therapy in Gram-negative sepsis: easier said than done?


Appropriate and timely de-escalation of antimicrobial therapy has long been recognised as an important element in the optimal management of sepsis. When a causative pathogen has been isolated and its susceptibility profile known, the most suitable single therapy should be instituted to prevent the development of super-infection with pathogenic or resistant organisms, as well as reduce toxicity and costs.


All blood cultures analysed on the BD BACTEC system were evaluated over a 6-week period (17 April 2010 to 30 May 2010). Organisms in positive blood cultures were then further identified by MALDI-TOF mass spectrometry (Brucker) and susceptibility testing was performed using the MicroScan WalkAway system (Siemens). The demographics, treatment regimens and clinical outcomes of all episodes of clinically significant Gram-negative bacteraemia were prospectively audited.


Two hundred and seventy sets of blood cultures were positive during our study period, representing 246 individual bacteraemic episodes. A total 143/270 were considered contaminants, 42/270 were significant Gram-positive bacteraemias, 1/270 Candida albicans, and 84/270 Gram-negative bacteraemia. Of the latter, 70 were individual episodes, of which two died prior to susceptibility results being available. Of the survivors, following knowledge of the susceptibility profile, only 20.5% (14) were de-escalated to a narrower-spectrum agent and only 31% (21) were converted to suitable oral agents when practical. Twelve per cent (10) were treated with combination therapies even though single agents remained highly active. Of the group that failed to de-escalate antimicrobial therapy (54 patients), the 30-day mortality rate was 9% (5/54) versus 7% (1/14) in the group (14 patients) that adhered to the surviving sepsis guidelines. Likewise, the former group were more likely to develop diarrhoea, 38% (21/54) versus 21% (3/14), with three patients positive for Clostridium difficile toxin in the former group (none in the latter). Multidrug-resistant organisms and fungal colonisation occurred more frequently in the first group - 38% (21/54) and 22% (12/54) versus 21% (3/14) and 0% (0/14), respectively.


Surviving sepsis guidelines have reiterated the need for timely use of appropriate empirical antimicrobials as well as the importance of de-escalation of therapy when the causative agent has been identified. However, as is the case in our study, this has not always been the prevailing practice. Many factors underlie this deviation from recommended guidelines, including worsening clinical condition, reported penicillin allergy as well as multiple co-morbidities.

Author information

Authors and Affiliations


Rights and permissions

This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Cite this article

Phee, L., Gordon, N., Hornsey, M. et al. De-escalation of antimicrobial therapy in Gram-negative sepsis: easier said than done?. Crit Care 15 (Suppl 1), P219 (2011).

Download citation

  • Published:

  • DOI: