Emergence of Extended-Spectrum Beta-Lactamase–Producing Enterobacteriaceae Colonization: What Termites Can Teach Us About the Gut Microbiota
It is in this setting that the article published in this issue of Critical Care Medicine by Detsis et al (2) perform a meta-analysis with the purpose of determining the prevalence and risk factors for and the clinical impact of newly acquired ESBL-PE gut colonization in critically ill patients. The primary analysis (13 studies; 15,045 patients) revealed an overall 7% ESBL-PE acquisition rate ranging from 3% in the Americas to 21% in the Western Pacific region. When only large studies (from France and the United States) were analyzed—in order to eliminate a possible small study effect—the acquisition rate was only 3%. Secondary analyses could only be performed on subsets of the entire dataset; nonetheless, a pattern emerged with antibiotic use being a risk factor for ESBL-PE acquisition. Both treatment with a beta-lactam/beta-lactamase inhibitor or a carbapenem while in the ICU or a history of antibiotics usage during the prior year were associated with newly diagnosed ESBL-PE colonization. Even more startling were the findings that ESBL-PE gut colonized patients were significantly more likely to suffer an ESBL-PE infection and that once infected, these patients were 57% (95% CI, 1.25–1.98) more likely to die when compared with noncolonized patients. The reason why most clinicians probably have not witnessed these high rates of infection and associated mortality may be explained by practicing in a region where both the prevalence and acquisition rates are relatively low.
The overarching significance of this study is that it serves to highlight the important association between ESBL-PE gut colonization and subsequent infection with the same organism and begs the question of which interventions could possibly break this cycle of colonization leading to infection. The options are currently limited. Improvements in antibiotic stewardship preventing patients from receiving unnecessary exposure to antimicrobials is crucial. Besides the study at hand, ESBL-PE colonization associated with antibiotic usage has been previously described even among healthy individuals (3). The role for traditional infection control measures in preventing ESBL-PE colonization is less clear. The impact of cohorting and isolation, for example, was not evaluated in this meta-analysis. And while the authors noted that previous hospitalization was associated with increased risk of colonization, interestingly, the ICU prevalence of ESBL-PE colonization did not impact new acquisition rates. Taken together, these findings would suggest that the newly acquired ESBL-PE gut colonization is more a function of antibiotic exposure and less a consequence of nosocomial spread. Selective decontamination of the digestive tract is another potential approach; unfortunately, its efficacy has not been consistent across studies and concern has been raised about the possibility of this strategy promoting the emergence of drug resistance and subsequent infections with resistant organisms (4, 5).
Manipulation of the gut microbiota (the microbes that inhabit the gut ecosystem) may hold the key to treat colonization and subsequent infection with an assortment of multidrug resistant organisms (MDRO). Although many point to the success of fecal microbial transplantation (FMT) in the treatment of Clostridium difficile infection, the idea of altering the gut microbiota for the benefit of the host is as old as nature itself.