Antibiotic De-Escalation Can Decrease the Development of Antibiotic Resistance in Gram-Negative Bacteria

By Richard R. Watkins, MD, MS, FACP, FIDSA, FISAC

Professor of Medicine, Northeast Ohio Medical University, Rootstown, OH

SYNOPSIS: A retrospective cohort study found that de-escalation of broad-spectrum beta-lactam agents in patients with sepsis significantly reduced the subsequent development of antimicrobial resistance in gram-negative bacteria isolated from clinical cultures.

SOURCE: Teshome BF, Park T, Arackal J, et al. Preventing new gram-negative resistance through beta-lactam de-escalation in hospitalized patients with sepsis: A retrospective cohort study. Clin Infect Dis 2024; Jun 6:ciae253. [Online ahead of print].

The ongoing dissemination of antimicrobial resistance (AMR) is a grave threat to public health. Therefore, finding ways to mitigate the spread of AMR is an urgent global priority. One potential strategy is antibiotic de-escalation. This involves replacing a broad-spectrum antibiotic agent with one that has a narrower spectrum. Teshome and colleagues aimed to determine whether de-escalating broad-spectrum beta-lactam antibiotics to narrow-spectrum ones based on a previously developed antibiotic spectrum score would affect the development of AMR in gram-negative bacteria.

The study was a retrospective cohort investigation of patients with sepsis that was conducted at a single institution between Dec. 1, 2010, and Dec. 31, 2017. Patients included in the study were at least 18 years of age; had a discharge diagnosis for sepsis, severe sepsis, or septic shock; and received at least three consecutive days of beta-lactam antibiotics. The investigators employed a previously designed antibiotic spectrum score. This assigns a numerical score (1 to 11) based on organism-beta-lactam antibiotic activity pairings on each day a patient is exposed to a beta-lactam during a defined follow-up period, with a broader spectrum drug awarded a higher score. Patients were grouped into three categories: de-escalation of beta-lactam spectrum score (BLSS), no change in BLSS, or escalation of BLSS. The primary endpoint was the isolation of a new antibiotic-resistant gram-negative pathogen from a clinical culture within 60 days of assignment to the cohort.

There were 7,742 patients in the analysis. Most had no change in antibiotic spectrum (n = 4,802 [62.0%]), followed by de-escalation (n = 1578 [20.4%]) and escalation (n = 1,362 [17.6%]). The mean age in all groups was 60.3 years and 56.2% were male. Those in the escalation group had more days in an intensive care unit (ICU) and of mechanical ventilation, vasopressor administration, central venous catheter use, and urinary catheter use. In-hospital mortality was highest in the escalation group (41.3%) compared to the de-escalation (18.1%) and no change groups (26.6%).

New gram-negative resistance developed in 644 patients (8.3%), with a mean time to resistance of 23.7 days (standard deviation, 15.0 days). The lowest incidence rate occurred in the de-escalated group, 1.42 (95% confidence interval [CI], 1.16-1.68) per 1,000 patient-days. There was a statistically significant reduction in the development of new gram-negative resistance associated with de-escalation compared to no-change (hazard ratio [HR], 0.59; 95% CI, 0.48-0.73). The most common pathogen group isolated with new resistance was the Enterobacterales, specifically third-generation cephalosporins and multidrug resistance (i.e., multiple antibiotic classes). Carbapenem-resistant and multidrug-resistant Pseudomonas aeruginosa also were common, especially in escalation patients. Furthermore, in patients with at least one follow-up culture, there was a statistically significant reduction in the development of new gram-negative resistance associated with de-escalation compared to no change.

COMMENTARY

De-escalating broad-spectrum antibiotics to a more targeted and narrower-spectrum agent seems rational from an ecological standpoint. Thus, de-escalation has been clinical dogma for many years. However, the evidence for this practice is weak and the recommendations in guidelines for it have mainly been based on expert opinion. Therefore, the study by Teshome and colleagues is important because it answers a significant question faced by clinicians daily: Does antibiotic de-escalation improve outcomes? Indeed, de-escalation appears to reduce the development of new gram-negative resistance compared to no change or escalation. Although the investigators only focused on gram-negative pathogens in subsequent clinical cultures, their results likely will apply to gram-positive pathogens as well, but this requires further study.

The study had a few limitations. First, the retrospective design makes it susceptible to unmeasured confounding factors. Second, the time frame for the study (2010-2017) may not represent the change in antibiotic use patterns that were brought about by the COVID-19 pandemic. Third, the study was conducted at a single academic medical center in Missouri, which may limit its generalizability. Fourth, the investigators did not collect fecal samples that detect changes in the gut microbiome. This is the standard for identifying early emergence of AMR. Finally, the investigators did not control for multiple covariates that may have affected the results, such as pharmacodynamic and/or pharmacokinetic differences between antibiotics, outbreaks of infections that could have occurred in various units or wards, or how changes in the microbiome could affect new, clinically relevant AMR.

Antibiotic de-escalation is a useful tool in the clinician’s toolbox to combat the spread of AMR. Educating peers and trainees about this practice is an important role for the infectious disease physician and those involved with antibiotic stewardship. Further studies are needed to elucidate the effects of antibiotic de-escalation in different patient populations, such as organ transplant recipients and children.