Antibiotic Strategies in the Era of Multidrug Resistance

George Karam; Jean Chastre; Mark H. Wilcox; Jean-Louis Vincent


Crit Care. 2016;20(136) 

In This Article


A fundamentally important challenge in clinical medicine is how to control or prevent antibiotic resistance. Several strategies may be beneficial.

Rapid Diagnostics

Rapid treatment of sepsis is associated with decreased mortality in patients with serious infections, but diagnosis of sepsis can be difficult and delayed in critically ill patients. Hence broad-spectrum empirical antibiotics are often started before microbiology results are available, resulting in some patients receiving unnecessary antibiotic treatment. More rapid, culture-independent identification methods are being developed,[44] which, in addition to providing quicker identification of the pathogen causing an infection, can also rapidly identify the susceptibility patterns of the organism. The benefits of such information are twofold: if the organism possesses resistance mechanisms, appropriate therapy based on susceptibilities can be more quickly accomplished; and, based on susceptibility information, the duration of broad-spectrum therapy may be limited. By enabling earlier, accurate diagnosis of infection and of the causative microorganism(s), these tests may help optimize antibiotic prescriptions, and in so doing potentially reduce selection pressure and thus resistance.

Colonization Prevention

Embracing the principle that colonization is the antecedent event leading to clinical infection, the clinician is poised to incorporate an important infection prevention opportunity into clinical practice. In an observational cohort study in two ICUs with endemic carbapenemase-producing Enterobacteriaceae, patients were screened with perineal swabs at admission and twice-weekly thereafter.[45] Patients colonized with carbapenemase-producing Enterobacteriaceae had a 1.8 times greater hazard of dying in the ICU than noncolonized patients, primarily because of an increased length of stay. In a study designed to prevent colonization and infection by KPC-producing Enterobacteriaceae in four long-term acute-care hospitals with high endemic KPC prevalence, a bundled intervention was tested using a stepped-wedge design.[46] Patients were screened for rectal KPC colonization on admission and every other week. Contact isolation and geographic separation of KPC-positive patients was implemented in ward cohorts or single rooms. All patients were bathed daily with chlorhexidine gluconate. Healthcare workers were educated, and their adherence was monitored. During the intervention period of the study, the incidence rate of KPC colonization fell from 4 to 2 acquisitions per 100 patient-weeks (p = 0.004 for linear decline). Compared with the preintervention period, there were decreases in the isolation of KPC in any clinical culture, KPC bacteremia, and all-cause bacteremia. The demonstration that prevention of colonization can have significant clinical benefits in the era of carbapenemase infections has significant implications for ICU and non-ICU settings.

Heterogeneity of Antibiotic Usage

The basis for this approach was noted in the mid-1990s through a program of informatics at LDS Hospital in Salt Lake City, UT, USA.[47] Subsequently, other groups found that patterns of antibiotic use in which the same antibiotic was given repeatedly (i.e., homogeneity) were associated with higher rates of resistance than when there was variability in the antibiotics given by the same prescriber among patients (i.e., heterogeneity).[48] Several studies have evaluated the role of empiric antibiotic rotation protocols in reducing development of antibiotic resistance. In a surgical ICU, Bennett et al.[49] reported that monthly cycling of four antibiotics (PTZ, imipenem/cilastin, ceftazidime, and ciprofloxacin) as the primary antibiotic to treat suspected Gram-negative infections was associated with an overall improvement in the antibiotic susceptibility profile of Gram-negative organisms compared with the medical ICU in the same hospital where cycling was not performed. The 2007 antimicrobial stewardship guideline from the Infectious Diseases Society of America/Society of Healthcare Epidemiology of America[50] stated that there were "insufficient data to recommend the routine use of antimicrobial cycling as a means of preventing or reducing antimicrobial resistance over a prolonged period of time".

Ironically, formal cycling may impose antibiotic selection pressures during those periods when a particular agent is preferred. However, increased diversity of prescribing has been shown to correlate with reduced levels of resistance.[51] The challenge therefore remains how increased antibiotic heterogeneity can be achieved in a way that delivers benefits rather than risks.

Short Duration Antibiotic Courses

Since the classic paper in the 1980s elucidating the concept that clinical resistance in Gram-negative pathogens occurs on the basis of selection by antibiotics of spontaneously mutant strains of bacteria which possess resistance mechanisms,[8] there has been increasing awareness about the importance of appropriately limiting the duration of antibiotic therapy. This strategy has taken two important forms in recent years. One has been de-escalation of therapy, in which the spectrum of empiric antibiotics is narrowed when microbiological data become available to minimize the selective pressure of antibiotics. A second strategy has been antimicrobial stewardship, which targets the duration of therapy as an important means of achieving optimal clinical outcomes. Current guidelines recommend a course of 7–10 days for most severe infections,[52] although some recent data support the use of shorter courses in certain infections, such as intraabdominal infections.[53] Many clinicians, however, remain hesitant about prescribing fewer fixed days of antibiotics for patients with severe bacterial infection, and prefer to customize antibiotic duration based on the clinical course of the disease and/or using serial determinations of a biological marker of infection, such as procalcitonin (PCT). Adapting the antimicrobial treatment duration to PCT kinetics has been demonstrated as useful in several randomized trials targeting patients with acute respiratory infection.[54] Nevertheless, PCT kinetics should only be used as a tool to support clinical judgment.