Vancomycin-resistant enterococci (VRE) are becoming a major concern in medical practice. Their increased prevalence and their ability to transfer vancomycin resistance to other bacteria (including methicillin-resistant Staphylococcus aureus) have made them a subject of close scrutiny and intense investigation. Colonization is usually acquired by susceptible hosts in an environment with a high rate of patient colonization with VRE (eg, intensive care units, oncology units). Vancomycin-resistant enterococci can survive in the environment for prolonged periods (>1 week), can contaminate almost any surface, and can be passed from one patient to another by health care workers. Whether VRE colonization leads to infection depends on the health status of the patient. Whereas immunocompetent patients colonized with VRE are at low risk for infection, weakened hosts (patients with hematologic disorders, transplant recipients, or severely ill patients) have an increased likelihood of developing infection following colonization. Quinupristin-dalfopristin and linezolid are among the anti-infective agents that have recently become available to treat infection caused by VRE. Other antimicrobials are currently under development. Molecular techniques such as polymerase chain reaction and standard culture studies are being used to detect VRE colonization, infection, and outbreaks.
The consequences of such co-colonization have already been realized. In 2003, physicians in Michigan reported the first case of vancomycin resistance transferred from VRE to MRSA in a patient with a wound colonized by both organisms.
Clearly, clinicians need to be aware of the importance of VRE. This review provides an overview of VRE, available treatments, and current detection techniques.
Infection with VRE (described in more detail subsequently) typically follows vancomycin-resistant enterococcal colonization, predominantly of the gastrointestinal tract. Colonization, which does not result in symptoms, may last for long periods and may serve as a reservoir for the transmission of VRE to other patients. Within hospitals, widespread colonization with VRE may occur with a comparatively small number of documented infections. Therefore, tracking colonization with VRE through active surveillance in high-risk units is an important component of preventing further transmission. At the Mayo Clinic, we perform active surveillance on patients in the ICU and on hematology, oncology, and abdominal transplantation wards.
Colonization is contingent on exposure to VRE and on being a “susceptible” host. With regard to exposure to VRE, the most important considerations are proximity to and duration of exposure to those already colonized with VRE. When the proportion of patients colonized with VRE on a particular ward (the so-called colonization pressure) is high (>50%), other risk factors for colonization (described subsequently) become less important.
These include patients who are severely ill and those receiving multiple and prolonged courses of antimicrobial agents. Colonization in these hosts often occurs in long-term care facilities and urban referral hospitals. Solid (especially abdominal) organ transplant recipients and hematology patients are at particularly increased risk for colonization with VRE. Health care workers and their household members are also at risk for VRE colonization.
Most patients colonized with VRE will remain colonized for prolonged periods. A Mayo Clinic study that defined clearance as negative rectal VRE cultures on at least 3 consecutive tests obtained more than 1 week apart showed spontaneous decolonization in only 18 (34%) of 53 liver and kidney transplant recipients.
Spontaneous decolonization occurs infrequently, and little progress has been made in finding a pharmacological method to eliminate VRE from a colonized patient. Antimicrobials such as oral bacitracin, ramoplanin, and novobiocin have shown limited success in permanently eradicating VRE from these patients.
Controlling transmission of VRE in the health care environment can be challenging. Vancomycin-resistant enterococci are capable of prolonged (>1 week) survival in the environment and can be transferred from environmental sites to staff hands. Vancomycin-resistant enterococci have been isolated from virtually every site and every object in health care facilities, including monitoring devices (call bells, electrocardiographic monitors, pulse oximeters, glucose meters, stethoscopes, electronic thermometers, blood pressure cuffs, keyboards, wall-mounted control panels), furniture (telephones, air cushions, headboards, tables, chairs, bed rails), toilet seats, doors, floors, linens, and other medical equipment (ventilator tubing, pumps, wash bowls, automated medication dispensers, intravenous poles).
Equally difficult is the maintenance of appropriate infection control measures. Nevertheless, during the course of caring for patients colonized with VRE, a change of gloves is necessary after contact with material that may contain high concentrations of the bacteria, such as stool, urine, bedpans, soiled linen, rectal thermometers, toilets, or other infected body sites. Strict isolation techniques should be used at all times: gloves and gown should be donned before entering and removed before exiting the patient's room. Hands should be sanitized immediately thereafter. Clothing and ungloved hands should not contact environmental surfaces potentially contaminated with VRE such as doorknobs or privacy curtains in the patient's room. Dedicated use of noncritical items (stethoscopes, thermometers) for individual patients decreases the risk of transmission of VRE from patient to patient. If such devices are used for other patients, they must first be thoroughly cleaned and disinfected.
At the Mayo Clinic, we practice universal gloving for all hematology, blood and marrow transplant, and solid organ transplant patients. Furthermore, we perform twice-weekly mandatory polymerase chain reaction (PCR) testing on all patients admitted to those services.
Despite inconveniences, increased health care professional workloads, and bigger up-front costs, infection control measures decrease the spread of VRE and thus are cost-effective. A recent University of Maryland mathematical model showed that active surveillance in the ICU reduced VRE transmission by a projected 39%.
with the ratio of infected-to-colonized patients dependent on the specific patient population. It is highest in hematology patients and organ transplant recipients and approaches zero in healthier (immunocompetent) populations.
Portals of entry for VRE typically include the urinary tract, intra-abdominal (eg, gastrointestinal tract, biliary tree) or pelvic sources, wounds (surgical wounds, decubitus ulcers), and intravascular catheters.
At the Mayo Clinic, most cases of VRE infection occur in abdominal organ transplant recipients and hematology patients.
Urinary tract infections caused by VRE include cystitis, pyelonephritis, prostatitis, and perinephric abscess; most such infections are nosocomial and associated with urinary instrumentation. However, evidence of VRE in the urine in the absence of other findings (urgency, fever, flank pain, systemic symptoms) may have limited clinical importance, representing only asymptomatic bacteriuria and not requiring antimicrobial treatment.
which may result in colonization of intravascular catheters and subsequent intravascular catheter-related sepsis. In liver transplant recipients, the most common types of VRE infection include intra-abdominal infections associated with biliary leaks, stenoses, or obstruction; hepatic or perihepatic abscesses; stenosis or thrombosis of the hepatic artery; or perforated viscera.
Risk factors for VRE bacteremia include hemodialysis; organ transplantation; receipt of corticosteroids, chemotherapy, or parenteral nutrition; surgery; severe illness; long-term antibiotic administration; indwelling urinary catheters; neutropenia; and mucositis.
Mortality rates for patients with VRE bacteremia vary depending on the population at risk. Recipients of autologous peripheral blood stem cell transplants have been shown to have mortality rates as low as 10%.
Although no currently available antimicrobial agent can eradicate VRE colonization, several treatment options exist for VRE infection. Most isolates of VRE are resistant to penicillin and ampicillin; however, in unusual cases in which such agents are active, they can be useful therapeutic options. Several choices for current, and possibly future, treatment of infection with VRE are described subsequently. Antimicrobial susceptibility testing is recommended to verify the activity of any agent being used to treat VRE.
In late 1999, quinupristin-dalfopristin became the first antimicrobial agent available for the treatment of vancomycin-resistant Enterococcus faecium infection. Although E faecium (the most common vancomycin-resistant Enterococcus species) is susceptible to quinupristin-dalfopristin, most Enterococcus faecalis isolates and many other non-E faecium species are intrinsically resistant to this antimicrobial agent. Quinupristin-dalfopristin, a streptogramin, targets the bacterial 50S ribosome, thereby inhibiting protein synthesis. Although uncommon, resistance can develop through modification of the target binding site, enzymatic inactivation, and/or efflux. The most common adverse effects are arthralgias and/or myalgias, which can be debilitating and have limited widespread use of the agent.
Linezolid, the first of a new class of antimicrobial agents termed oxazolidinones, became available in 2000. It can be administered orally or intravenously and, unlike quinupristin-dalfopristin, has activity against both E faecium and non-E faecium species (E faecalis, Enterococcus casseliflavus, Enterococcus gallinarum). The oral formulation has excellent bioavailability. Currently, linezolid is the only oral agent approved by the Food and Drug Administration for treatment of infections caused by VRE. Linezolid inhibits ribosomal protein synthesis but at a different site from other agents that target the ribosome (chloramphenicol, macrolides, lincosamides, streptogramin, aminoglycosides, tetracycline). Consequently, existing mechanisms of resistance to these agents do not confer cross-resistance to linezolid.
Linezolid is the anti-VRE drug used most commonly at the Mayo Clinic. However, its myelosuppressive adverse effects, especially thrombocytopenia, may limit its use in some patients. In addition, because linezolid is a weak monoamine oxidase inhibitor, a diet low in tyramine (as instructed in the package insert) is generally recommended while taking the medication. Linezolid has some potentially important drug-drug interactions, and careful review of the patient's medical regimen, in consultation with a pharmacist, is recommended before it is prescribed.
Linezolid has several characteristics that were initially perceived as important in preventing the emergence of resistance. It is a synthetic agent, and preexisting resistance similar to that seen with natural antibacterial agents (penicillin, vancomycin, etc) was therefore considered unlikely. In addition, because oxazolidinones inhibit protein synthesis by binding to domain V of ribosomal RNA, which is encoded by genes (ribosomal RNA gene) present in multiple copies (4 copies in E faecalis and 5-6 copies in E faecium), selection of mutational resistance was expected to require mutations in multiple copies of 23S ribosomal DNA, a hypothetically unlikely event.
Unfortunately, cases of linezolid-resistant enterococci have emerged and spread nosocomially. In 2001, for example, 7 clinical isolates of linezolid-resistant, vancomycin-resistant E faecium were reported from the Mayo Clinic.
The linezolid-resistant, vancomycin-resistant E faecium strain was identified in a liver transplant recipient whose course was complicated by vancomycin-resistant E faecium intra-abdominal infection treated with linezolid therapy. The strain was transmitted nosocomially to 6 other patients despite strict isolation of the index case, the use of private rooms, and universal gloving by health care workers before entering patients' rooms.
A cyclic lipopeptide fermentation product of Streptomyces roseosporus, daptomycin became available in 2003. It rapidly kills gram-positive bacteria by disrupting multiple aspects of bacterial membrane function.
Although daptomycin has in vitro activity against VRE, publishedclinical data regarding its use for the treatment of infections caused by VRE, a non-Food and Drug Administration-approved indication, are minimal. Daptomycin should not be used to treat pneumonia because clinical trials have shown a high failure rate in this setting.
Tigecycline, a broad-spectrum glycylcycline antimicrobial agent, became available in 2005. This novel tetracycline derivative has activity against gram-positive and gram-negative aerobic and anaerobic bacteria, including tetracycline-resistant isolates.
Similar to daptomycin, this compound has in vitro activity against VRE; however, clinical data on the treatment of infections caused by VRE are lacking.
Many VRE isolates are susceptible to nitrofurantoin, which has been used to treat VRE urinary tract infection but does not have useful activity in other VRE infections.
Future options for treatment of VRE infection may include mannopeptimycins and dalbavancin. Mannopeptimycins, a novel class of glycopeptides, are semisynthetic antimicrobials isolated from Streptomyces hygroscopicus. Like vancomycin, they prevent transglycosylation of cell wall peptidoglycans. They are active against a wide variety of gram-positive bacteria, including VRE, and have been shown to be bactericidal in vivo.
Dalbavancin, another semisynthetic glycopeptide, has gained attention because of its once-weekly dosing schedule. However, while active against VanB glycopeptide-resistant phenotype enterococci, dalbavancin has little activity against the more common VanA glycopeptide-resistant phenotype enterococci.
To date, 6 glycopeptide-resistant enterococcal phenotypes, VanA, VanB, VanC, VanD, VanE, and VanG, have been described. They can usually be distinguished on the basis of the level, inducibility, and transferability of resistance to vancomycin and teicoplanin (Table 1). The first 2 types are the most clinically relevant.
Table 1Resistance to Glycopeptides in Enterococci
From Patel R. Vancomycin-resistant enterococci in solid organ transplantation. Curr Opin Organ Transplant. 1999;4:271-280, with permission from Lippincott Williams & Wilkins.
Vancomycin complexes with the d-alanyl-d-alanine termini of normal peptidoglycan cell wall precursors, thereby inhibiting cell wall synthesis (Figure 1). The genes associated with high-level vancomycin resistance in enterococci encode a ligase responsible for the synthesis of the depsipeptide d-alanyl-d-lactate. This depsipeptide is incorporated into the terminal portion of the peptidoglycan cell wall precursor, limiting vancomycin-peptidoglycan precursor binding (Figure 1).
VanA-type glycopeptide resistance is characterized by acquired inducible resistance to both vancomycin and teicoplanin. It is mediated by transposon Tn1546 or closely related genetic elements. Tn1546 contains the vanA gene clusterthat encodes 8 polypeptides (Figure 2). This transposon may be located on plasmids or bacterial chromosomes.
The transfer of high-level (VanA) vancomycin resistance from E faecalis to S aureus via Tn1546 was described recently.
VanB-type glycopeptide resistance is characterized by acquired inducible resistance to various concentrations of vancomycin but typically not to teicoplanin. The vanB gene cluster, as described in E faecalis V583, has homology to the vanA gene cluster; it consists of genes encodingpolypeptides assigned to the regulation of glycopeptide resistance genes (vanRB and vanSB), synthesis of the depsipeptide d-alanyl-d-lactate (vanHB and vanB), and hydrolysis of precursors of normal peptidoglycan (vanXB and vanYB). The vanB sequence varies among different enterococcal isolates.
MOLECULAR TESTS FOR DETECTION OF VRE COLONIZATION
At the Mayo Clinic, VRE colonization is identified by PCR on samples obtained from perianal, perirectal, or rectal swabs or from stool specimens. Bacterial DNA is extracted using the automated MagNA Pure instrument (Roche Diagnostics Corporation, Indianapolis, Ind). Then, the LightCycler instrument (Roche Diagnostics Corporation) is used to detect vanA and vanB using a rapid real-time PCR assay (Figure 3). This method is more sensitive and faster (~3.5 vs >72 hours) than culture for detecting VRE colonization.
The assay detects the presence of genes associated with vancomycin resistance in enterococci, vanA and vanB. A positive result indicates colonization but not necessarily infection with VRE.
Molecular testing for vanA and vanB may not always detect VRE (ie, if the bacteria are present in very low quantities). Furthermore, organisms other than enterococci (such as enteric anaerobes) can carry these van genes, leading to false-positive results.
Therefore, some positive results may not actually represent the presence of VRE. The frequency with which this phenomenon occurs is unknown, but it is considered rare.
In theory, PCR testing should decrease the spread of VRE by more rapid identification and earlier isolation of colonized patients. However, the cost-effectiveness of this practice is unknown because some PCR-positive patients have minimal bacterial burden and thus present a low risk for the spread of VRE.
INDICATIONS FOR VRE CULTURES
Conventional aerobic bacterial cultures are used to isolate VRE from clinical specimens such as blood for the diagnosis of VRE infection. This approach allows antimicrobial susceptibility testing for selection of appropriate antimicrobial treatment.
Culture for VRE is also required to “fingerprint” isolates in outbreak investigations. Vancomycin-resistant enterococci are genetically diverse. Nosocomial outbreaks may be monoclonal, oligoclonal, or polyclonal; certain clones may establish themselves as endemic strains. Pulsed-field gel electrophoresis is used commonly to evaluate clonality. With use of this molecular technique, the DNA of the bacterium is “cut” into large fragments (>30 kilobase) using restriction enzymes. The fragments are then separated in a gel using electrophoresis with periodic reorientation of the electric field. Smaller DNA pieces move faster, larger pieces lag behind, and ultimately, the DNA separates into multiple bands on the gel. Band patterns from different isolates can then be compared to assess whether they are similar. Indistinguishable band patterns suggest the possibility that 2 isolates are clones of one another, while different patterns imply a more distant relationship.
Colonization with VRE occurs in susceptible hosts when colonization pressure is high. Real-time PCR analysis of perianal, perirectal, or rectal swab specimens allows for rapid detection of colonization. Infection, with associated high mortality rates, may follow colonization in high-risk individuals. Culture techniques are used to isolate VRE in order to diagnose infection, assess antimicrobial susceptibility, and identify clonality in the case of nosocomial outbreaks. Linezolid is the antimicrobial used most commonly to treat infection with VRE. Other antimicrobials such as quinupristin-dalfopristin, daptomycin, tigecycline, and nitrofurantoin are also prescribed. Currently, no accepted treatment for colonization has been determined.
Knowledge about VRE is important for all health care professionals. With their increasing prevalence, capacity for prolonged survival in the environment, ability to over-come infection-control procedures, and capability of transferring vancomycin resistance to S aureus, VRE represent an important infectious disease threat.