Antimicrobial De-escalation Is Rarely Applied to Patients in ICUs

Few patients in intensive care units undergo antimicrobial de-escalation for treatment of bacterial infections, according to a recent study, which also noted that the antimicrobial stewardship strategy did not negatively affect clinical cure results.

The study, published in Intensive Care Medicine, evaluated 1495 adult patients empirical antimicrobial therapy for bacterial infection in 152 intensive care units in 28 countries between October 2016 until May 2018 to determine how often antimicrobial de-escalation is performed and to estimate how it effects clinical cure.

“Antimicrobial de-escalation (ADE) is a very attractive antimicrobial stewardship strategy, especially in the critically ill patient, as it combines the goal of early adequate antimicrobial therapy with the aim to reduce the overall use of broad-spectrum antimicrobials,” corresponding author Liesbet De Bus, MD, intensivist at Ghent University Hospital in Belgium, told Contagion^®. “Intuitively, ADE is perceived as an ‘easy’ stewardship target as it can be performed during the course of the infection, on a point in time when more clinical and microbiological information becomes available to support the antimicrobial decision-making process.”

About 50% of patients in the study, known as the DIANA study, received combination therapy, and 26% were prescribed carbapenems. Within the first 3 days of therapy, 16% underwent ADE, 63% underwent no change and 22% underwent change other than ADE. Investigators also reported unadjusted mortality rate of 15.8% at day 28 in the ADE cohort compared with 19.4% among patients who underwent no change. The IP-weighted relative risk estimate for clinical cure comparing patients who underwent ADE with those who didn’t was 1.37 (95% CI 1.14—1.64).

“Our research clearly showed that performing ADE is not that straightforward in the critically ill-infected patient, as it was only done in 16% of ICU patients within 3 days following empirical prescription, despite the fact that half of the empirical prescriptions consisted of combination therapy and one-quarter contained a carbapenem,” De Bus told Contagion^®. “In addition, our study showed that ADE was not worse compared with no-ADE regarding clinical cure on day 7. As most studies on ADE, our study was observational in nature and as such we have to acknowledge that our results were also prone to bias. Compared to previous research however, maximal efforts were undertaken to reduce potential bias by using advanced appropriate statistical methods.”

De Bus said the low number of patients who underwent ADE was unexpected and the prevalence of broad-spectrum antimicrobial therapy among critically ill patients was striking.

“Over the past years, antimicrobial de-escalation has been promoted as an important stewardship strategy although a clear definition was (and still is) lacking and the ecological impact of different treatment strategies remains unknown up until now,” De Bus said. “In our study, ADE was defined based on the intention of the treating physician to narrow the antimicrobial spectrum. Although ADE as it was applied and defined in our study, seems to be a safe strategy for the individual patient, physicians should bear in mind that our research has shown that ADE is not always easy to perform in the critically ill within the first days following the start of the antimicrobial therapy.”

Previous studies have found higher instances of ADE, the study noted, adding that single-center studies with special interest in antimicrobial stewardship might account for the difference. Another factor that might play a role could be differences in how ADE is defined.

“Clear guidelines on the optimal de-escalation strategy, supported by randomized controlled trials are currently still lacking,” De Bus stated. “Antimicrobial de-escalation should not be considered as the only antimicrobial stewardship strategy of interest and at least equal attention should be paid to other strategies pursuing the same goal e.g. the development of treatment guidelines adjusted to local ecology, early source control, reducing overall treatment duration.”

“We need to elucidate barriers to potential ADE in the critically ill patient,” De Bus continued. “In addition, more basic research on the ecological impact of different antimicrobial treatment strategies in different patient populations is mandatory to enable the construction of clear ADE guidelines. Related to this, the optimal timing of ADE has to be studied.”