Enterococcus, a globally important opportunistic pathogen, can be carried in the GI tract for a long period without any symptoms of infection and likewise persist in the hospital environment (16). Enterococcus is inherently resistant to a number of antimicrobial classes, and over recent decades there has been a significant increase in the rates of acquired antimicrobial resistance (AMR) in E. faecalis and E. faecium, including VRE (17, 18). Previous studies have demonstrated that enterococcus status at ICU admission was associated with risk for death or all–cause infection, and the gastrointestinal microbiome may have a role in risk stratification and early diagnosis of ICU infections (11). There is a need for active surveillance to better prevent the emergence and dissemination of VRE.
We performed a multicenter study investigating the prevalence of gastrointestinal colonization with VRE in hospital patients in Beijing and analyzing the molecular epidemiology of VRE. Although this study was performed on a selection of hospital patients, i.e., patients admitted to ICUs, the results are of critical importance since these patients are especially prone to colonization and (subsequent) infection. We reported a mean proportion of 31.1% (46 of 148) for culture positive intestinal carriage of VRE in ICU, and found that vanA-type VRE and vanM-type VRE were both detected in almost every hospital but vanA-type was more popular. We additionally identified age, longer ICU stay, use of an endotracheal tube, and prior glycopeptides exposure as significant risk factors. To our knowledge, this is the largest study investigating the prevalence of VRE colonization in ICU in China mainland, and the results are further strengthened by the multicenter design and comprehensive capture of potential risk factors through a standardized medical record review.
Several previous studies have reported a prevalence of VRE carriage on ICU admission ranging between 2.5% and 40% (19, 20, 21). In this study, we found the overall prevalence of VRE carriage at ICU was markedly high. Moreover, there were huge difference in the isolation rates of VRE among the six hospitals, and the lowest isolation rate was 14.7% (5/34) but the highest rate up to 85.7% (12/14). Hospital variations are not unusual for drug-resistant bacteria, as different compliance with isolation practices, infection control, and antibiotic stewardship programs that affect behavior among the health care personnel, account for this differences. In addition, the variations may be related to the patients’ characteristics prior to ICU admission (surgical wards, transplantation units, etc.), as well as, the limited sample size in part of the hospitals (22, 23). Altogether, the incidence proportion of VRE colonization was comparatively high in Beijing and might be neglected before. Previous studies have indicated that pathogens can be cultured from stool or swabs that predict specific infections, and VRE colonization has been established as a risk factor for subsequent infection. Our study results might strengthen the need for prompt identification of colonized patients and to carry out the effective multifaceted strategies to control VRE in healthcare institution.
E. faecium (42/46) predominated among colonizing VRE in this study, as reported elsewhere. A high level of glycopeptide resistance is more common in E. faecium than E. faecalis (24). In addition, E. casseliflavus accounted for two VRE strains, and another two VRE were singletons belonging to E. faecalis, and E. gallinarum, respectively. Among the tested VRE, 35 strains harbored vanA resistance gene and 8 strains carried vanM gene, and there was no vanB gene detected. Glycopeptide resistance in enterococci is mediated by van gene clusters, among which vanA and vanB are the most commonly reported worldwide (25, 26, 27, 28). In 2006, vanM was first reported as a new and prevalent resistance determinant in clinical enterococci in China (29). Subsequently, vanM-type VRE has spread rapidly around the country, especially in the cities of Shanghai and Hangzhou (24, 30). In 2011, a study reported a vanM-type E. faecium clinical strain was detected in Singapore, which indicated that vanM gene might spread to other countries (31). Epidemiology data for strains with vanM cluster remain rare in Beijing, and previous results were limited to single-center investigations. In this study, vanM-type VRE were detected in almost every participating hospital, thus suggesting that vanM cluster plays an important role in vancomycin resistance and van gene clusters dissemination in Beijing. However, our study showed that vanA cluster still was the dominant resistance determinant in enterococcus in Beijing.
The tested E. faecium, including vanA-type strains and vanM-type strains, all displayed the high vancomycin MIC values. Interestingly, these strains showed variable levels of resistance to teicoplanin, with the MIC ranged from 0.25 to 128 µg/ml. Although the strains carried the same type of van cluster, they displayed the different susceptibilities to teicoplanin. This phenotype variation may to some degree be related to the integrality of the van operon (32, 33). Most of the colonization VRE were resistant to several kind of antimicrobial agents, and they belong to multi-drug resistant (MDR) (resistant to three or more antimicrobial classes). Fortunately, linezolid, daptomycin, and tigecycline demonstrated complete in vitro activity against these strains.
MLST typing displayed two dominant STs, including ST78, and ST192, which were frequently identified in VREfm strains in China, and the most common ST78 occurred in each participating hospital (34, 35, 36). All but one E. faecium strain (ST922) in this study belonged to clonal complexes (CC) 17, which represents a lineage of a virulent VRE hospital clone that has been observed worldwide (2, 37–42). The results described here clearly suggest a clonal spread of the highly adapted and resistant lineage CC17 of E. faecium strains among hospitals. In addition, two E. faecium strains both carrying vanM resistance gene belonged to ST922, and they were detected in the same hospital. The fact that ST922 was not included in CC17, and no previous study has found ST922 strains carrying vanM cluster in China, goes some way to suggest a horizontal transfer of van cluster among E. faecium strains. Overall, these data indicate that clonal expansion and horizontal transfer of resistance genes have contributed to VRE increased prevalence in hospital.
This study identified four risk factors for VRE colonization at ICU: age, length of ICU stay, endotracheal tube use, and recent glycopeptides use, all of which agrees with the results of numerous other studies. Here, the attributed risk of VRE acquisition after glycopeptide treatment needs to be mentioned. Several previous studies have demonstrated that vancomycin might influence the likelihood of VRE colonization and transmission between patients. This phenomenon may be attributed to vancomycin being a narrow-spectrum antibiotic and prescribed generally with other antibiotics, such as cephalosporins, and carbapenems, which disrupt the normal gut flora, thus the likelihood of VRE colonization may have increased. Another possible reason is that part of gut enterococci belong to vancomycin-variable enterococci (VVE), which are a group of enterococci containing van resistant genes that exhibit a vancomycin-susceptible phenotype but are capable of shifting to a glycopeptide-resistant phenotype under vancomycin therapy, as previously stated (32, 33).
The study has some limitations. First, the study period was relatively short, and thus the population was not large enough, which may influence the determination of risk factors for VRE colonization. Second, the rectal surveillance swab was not applied on patients at the time of admission but every Tuesday morning during the study period, which can lead the VRE colonization status of some patients unclear. Another limitation is the problem of screening process. We evaluated the VRE colonization status by way of phenotype identification, however, some patients colonized by VVE were neglected and the prevalence of van resistant gene might be underestimated.