September 21, 2020

An article review from the ESICM Journal Review Club

Antimicrobial de-escalation in the ICU and its effects on clinical outcome in a global cohort

Antimicrobial de-escalation (ADE) is a key component of antimicrobial stewardship and aims to provide early adequate antimicrobial therapy while reducing the spreading of multidrug resistant pathogens, but the actual application of such strategy on everyday practice worldwide is not known. Moreover, the real impact of ADE strategy on patient’s outcomes has not being completely elucidated.

In the DIANA study (DetermInants of Antimicrobial use aNd de-escAlation in critical care) De Bus and the co-authors conducted a large international, multicentre observational cohort study with the aims to determine the frequency of ADE on empirically prescribed antibiotic therapy within the first 3 days and to estimate the effect of ADE on clinical cure by day 7.

Between October 2016 and May 2018, 152 ICUs (90% teaching hospitals) in 28 countries (Europe 48%, Asia 38%, America 9%, Oceania 5%) and 1495 patients were included in the analysis. In the overall population, a combination of antimicrobial agents was prescribed in 50% of patients and carbapenems in 26%.

Antimicrobial treatments between day 0 and day 3 were classified as “no change”, “ADE” (discontinuation of antimicrobials from the initial empirical combination therapy or antimicrobial change to narrow the spectrum of activity) or “other change”.  ADE took place in the first three days in only 16% of patients, mainly by discontinuation of components (one or more) of a combination therapy (52%); and no changes occurred in 63% of patients.

The rate of clinical cure by day 7 (survival and resolutions of all infection-related symptoms) was higher in the ADE compared to the “no change” group [57.9% vs 42.7%, RR 1.34 (1.18-1.52); p<0.001]. The inverse probability weighted relative risk estimate for clinical cure by day 7 used to control for time varying confounders comparing ADE with no-ADE patients (no change or change other than ADE) was 1.37 (95% CI 1.14–1.64).




  • The study succeeded in including a large population of critically ill patients from a high number of countries, providing a solid representation of real-life utilisation and impact of ADE strategy.


  • The heterogeneity of the population and practices, the limited number of patients included per centre and differences in antibiotic stewardship across the participating centres.


  • Within 3 days of empirically prescribed antimicrobial therapy, ADE was performed only in a minority of critically ill patients (16%).
  • No deleterious effect of ADE was suggested on clinical cure by day 7, despite possible confounders.

This article review was prepared and submitted by Dr Filippo Annoni, Cliniques universitaires de Bruxelles/Hôpital Erasme, Brussels, on behalf of the ESICM Journal Review Club.

  1. De Bus, L., Depuydt, P., Steen, J. et al. Antimicrobial de-escalation in the critically ill patient and assessment of clinical cure: the DIANA study.   

Comment on this news