Vancomycin-Resistant Staphylococcus aureus in the United States, 2002–2006
- Dawn M. Sievert1,2,
- James T. Rudrik1,
- Jean B. Patel2,
- L. Clifford McDonald2,
- Melinda J. Wilkins1, and
- Jeffrey C. Hageman2
- 1Michigan Department of Community Health, Lansing, Atlanta, Georgia
- 2Centers for Disease Control and Prevention, Atlanta, Georgia
- Reprints or correspondence: Dr. Dawn M. Sievert, Div. of Healthcare Quality Promotion, Centers for Disease Control and Prevention, 1600 Clifton Rd., MS A-24, Atlanta, GA 30333 (DSievert{at}cdc.gov).
Abstract
Background. This report compares the clinical characteristics, epidemiologic investigations, infection-control evaluations, and microbiologic findings of all 7 of the cases of vancomycin-resistant Staphylococcus aureus (VRSA) infection in the United States during the period 2002–2006.
Methods. Epidemiologic, clinical, and infection-control information was collected. VRSA isolates underwent confirmatory identification, antimicrobial susceptibility testing, pulsed-field gel electrophoresis, and typing of the resistance genes. To assess VRSA transmission, case patients and their contacts were screened for VRSA carriage.
Results. Seven cases were identified from 2002 through 2006; 5 were reported from Michigan, 1 was reported from Pennsylvania, and 1 was reported from New York. All VRSA isolates were vanA positive and had a median vancomycin minimum inhibitory concentration of 512 µg/mL. All case patients had a history of prior methicillin-resistant S. aureus and enterococcal infection or colonization; all had several underlying conditions, including chronic skin ulcers; and most had received vancomycin therapy prior to their VRSA infection. Person-to-person transmission of VRSA was not identified beyond any of the case patients. Infection-control precautions were evaluated and were consistent with established guidelines.
Conclusions. Seven patients with vanA-positive VRSA have been identified in the United States. Prompt detection by microbiology laboratories and adherence to recommended infection control measures for multidrug-resistant organisms appear to have prevented transmission to other patients.
Staphylococcus aureus and Enterococcus species are gram-positive, human commensal bacteria. S. aureus is commonly found on the skin and in the nares of healthy people. Enterococci are normally present in the human intestines. Both of these bacteria are opportunistic pathogens and have been among the most common causes of nosocomial infections [1, 2]. During the past 2 decades, these bacteria have developed resistance to commonly prescribed antimicrobial agents. Methicillin-resistant S. aureus (MRSA) infection first emerged in the United States in the 1970s, and by the 1990s, MRSA was considered to be endemic in most large urban medical centers [3, 4]. Vancomycin-resistant enterococci (VRE) were first reported in a US hospital in 1989 and rapidly became a common cause of health care-associated infections [5–7]. Although vancomycin could no longer be used to treat the growing number of VRE infections, it remained the only uniformly effective antimicrobial agent to treat the numerous MRSA infections [8–10]. In 1992, Noble et al. [11] demonstrated that conjugal transfer of the vanA gene, which mediates vancomycin resistance, from VRE to MRSA on the skin surface of hairless mice could be achieved, creating vancomycin-resistant S. aureus (VRSA). In an era of increasing rates of VRE and MRSA infection, the prospect of this transfer occurring spontaneously in vivo was of serious concern. In 1997, the first case of vancomycin-intermediate S. aureus infection was reported from Japan [12]. However, the mechanism of resistance was not mediated by vanA but rather by a change in cell physiology caused by genetic mutations and altered expression of certain genes, resulting in a characteristic thickened cell wall that prevented vancomycin from reaching its target [13, 14].
In June 2002, the Michigan Department of Community Health reported the first clinical case of vanA-mediated VRSA infection in the world [15, 16]. Since then, 6 additional cases have been confirmed by the Centers for Disease Control and Prevention (1 case each from Pennsylvania and New York and 5 cases from Michigan) [17–21]. Information regarding only the first 3 cases has been published. This report will compare the clinical characteristics, epidemiologic investigations, infection-control evaluations, and microbiologic findings of all 7 documented cases of VRSA infection in the world.
Methods
Patients and medical history. A patient with VRSA infection was defined as an individual from whom an S. aureus isolate was recovered for which the vancomycin MIC was ⩾32 µg/mL before 2006 and ⩾16 µg/mL after the vancomycin breakpoint was changed in 2006 [22–24]. Information regarding medical history, clinical course, treatment modalities (including antimicrobial and surgical therapy), and outcome was obtained through medical record review and interviews with patients, close family and friends, and medical personnel.
Epidemiologic investigations. A comprehensive epidemiologic contact investigation was conducted for each case patient with VRSA infection and their contacts. Specimens were collected from the nares, axilla, groin, wounds, and rectum of each case patient to determine both VRSA and VRE colonization status. To assess the extent of transmission beyond the case patients, nare, skin wound, and catheter exit site swab specimens were collected from all available patient contacts during the period of transmissibility. This period was defined as the time from the last date of a culture negative for VRSA through the date that appropriate infection-control precautions were implemented following the isolation of VRSA. Places where the case patients resided or visited within the period of transmissibility were identified as facilities where transmission could have occurred. These facilities included patient homes, hospitals, rehabilitation and long-term care facilities, physician offices, dialysis centers, an infusion center, a wound care clinic, and a nail salon. Patient contacts considered to be at risk for transmission included people who had direct physical contact with patients, shared the same living space, or had the same health care providers during the period of transmissibility. Persons who worked in laboratories where VRSA organisms were initially identified were also considered to be at risk. Overall, at-risk individuals included physicians, nurses, therapists, other patients, family members, friends, laboratory technologists, and a manicurist. On-going surveillance cultures were performed for the case patients and all individuals who remained at risk because of direct patient contact throughout the investigations.
Infection-control policies and procedures. Infection-control policies and procedures were assessed in all health care facilities where the patients with VRSA infection had received care during their periods of transmissibility. These assessments were conducted through interviews with infection control supervisors, review of written procedures, and direct observation of health care worker practices.
Laboratory procedures. Isolate identification and antimicrobial susceptibility testing were conducted at local clinical laboratories and then confirmed at state health department laboratories. VRSA isolates were then sent to the Centers for Disease Control and Prevention (Atlanta, GA) for further characterization. Species identification was determined using standard biochemical and molecular methods [25]. Antimicrobial susceptibility testing was performed using the reference broth microdilution method according to the Clinical and Laboratory Standards Institute [22–24]. Genomic DNA from VRSA isolates was isolated by the silica-gel membrane method (Qiagen DNeasy) and was used as a template for PCR to detect the presence of mecA, vanA, vanB, vanC, and vanD [26–28]. PFGE was performed using SmaI-digested DNA from VRSA isolates, and banding patterns were analyzed and classified as described elsewhere [29]. All isolates underwent typing of the staphylococcal cassette chromosome mec (SCCmec) genetic element using PCR [30].
Results
Case descriptions. Five (71%) of the 7 cases of VRSA infection were reported from Michigan (table 1). Four (57%) of the 7 case patients with VRSA infection were female, 5 (71%) were white, and the median age was 58 years (range, 40–78 years). All 7 case patients had a history of previous MRSA and enterococcal (4 with VRE) infections or colonization. Most case patients had several underlying conditions, including 5 patients (71%) with chronic skin ulcers, 4 (57%) with diabetes, and 3 (43%) with chronic renal failure; 2 (29%) were considered to be obese. One of the case patients (patient 2) had not received vancomycin therapy during the 5 years prior to VRSA infection, 1 had not received vancomycin therapy during the previous 4 months (patient 3), 4 had received vancomycin therapy for 5–9 weeks in the previous 3 months, and 1 had intermittently received vancomycin therapy for ∼10 years. VRSA was isolated from specimens of either ulcers or wounds for 6 (86%) of the case patients and from urine specimens from a nephrostomy tube for the other case patient. In all but 1 case patient (patient 3), specimen collection was prompted by a wound or ulcer that appeared infected or was healing (in patient 5). Enterococcal species were recovered from the sites from which specimens were culture positive for VRSA for 5 of the 7 case patients; 3 of the isolates were VRE. All culture specimens were polymicrobial and included a variety of gram-negative organisms. At the time of specimen collection, 4 case patients (57%) were inpatients, 2 (29%) were outpatients, and 1 (14%) was a long-term care resident. Case patients were observed for a median of 10 weeks (range, 5–56 weeks) after initial VRSA isolation. Follow-up was concluded after culture results remained negative for VRSA for at least 3 consecutive weeks without antimicrobial therapy, when the culture site healed, or at the time of death. One case patient persistently tested positive for VRSA, and 6 case patients became culture negative for VRSA following multimodal therapy, including wound care (n=6), surgical intervention (n=5), and antimicrobial therapy (n=5).
Table 1
Clinical aspects of vancomycin-resistant Staphylococcus aureus (VRSA) infection in patients in the United States, 2002–2006.
Epidemiologic investigation results. Two case patients were colonized with VRSA at body sites other than the initial culture sites; 1 case patient (patient 5) had VRSA isolated from a specimen from the groin, adjacent to the VRSA-infected wound site, and 1 case patient (patient 6) had VRSA isolated from a specimen from the nares. After showering with chlorhexidine for 5 days (patient 5) and using nasal mupirocin twice daily for 5 days (patient 6), VRSA was not recovered again from either site. VRE isolates (5 Enteroccocus faecalis and 1 Enteroccocus faecium; 1 patient was colonized with both) were recovered from specimens from either a wound, nephrostomy tube, or gastrointestinal tract for 5 (71%) of the case patients.
The median number of case patient contacts for whom surveillance cultures for VRSA were performed was 42 (range, 23–371 case patient contacts) (table 2). Fourteen percent to 30% of case patient contacts were colonized with methicillin-susceptible S. aureus, 3%–8% were colonized with MRSA, and none were colonized with VRSA.
Table 2
Contact investigation results for patients with vancomycin-resistant Staphylococcus aureus in the United States, 2002–2006.
Infection-control findings. Standard infection-control precautions were in place at all of the involved health care facilities prior to identification of the cases of VRSA infection, and the policies and practices were appropriate for the respective settings. Prior to recovery of VRSA, most case patients had known MRSA infection or colonization; therefore, contact precautions were being used. Once VRSA was identified, contact precautions were reinforced, and some enhanced measures were instituted at all facilities and for all staff having continued contact with the infected patients. For inpatients, the enhanced precautions included placing patients in private rooms. For outpatients, treatment was administered in dedicated rooms or areas separate from other patients and during the last appointment of the day. For all patients, enhanced measures included dedicated staff and equipment, new gloves and gowns for each patient interaction, masks with eye protection when the potential for splashing and/or spraying of infectious material existed (e.g., during wound care in the podiatry clinic for patient 1), and thorough cleaning and disinfecting of all patient rooms and equipment after each use and after discharge from the hospital. Enhanced contact precautions remained in place throughout the follow-up period for both inpatients and outpatients. Follow-up was discontinued if culture results remained negative for VRSA for at least 3 consecutive weeks without antimicrobial therapy or if the primary culture site healed. However, contact precautions remained in place if outpatient care continued.
Laboratory results. The 7 VRSA isolates had a median vancomycin MIC of 512 µg/mL (range, 32–1024 µg/mL) (table 3). All isolates were susceptible to ⩾5 antimicrobial agents approved by the Food and Drug Administration for treating S. aureus infection, including linezolid, quinupristin-dalfopristin, and trimethoprim-sulfamethoxazole; 6 isolates were susceptible to daptomycin. All of the isolates acquired a vanA-containing Tn1546-like element by independent genetic events [15, 21, 31]. In each isolate, the vanA gene was localized to a plasmid, which ranged in size from 40 to 120 kb. Five isolates belonged to MRSA lineage USA100 and contained SCCmec type II, and 1 belonged to MRSA lineage USA800 and contained SCCmec type IV. The VRSA isolate from patient 6 did not belong to any current PFGE type in the database, and the SCCmec type could not be determined using standard primers.
Table 3
Laboratory aspects of vancomycin-resistant Staphylococcus aureus infections in patients in the United States, 2002–2006.
Discussion
This report describes the clinical and laboratory characteristics of 7 patients from whom vanA-containing VRSA isolates were recovered. Most of these patients had several characteristics in common, including chronic underlying conditions, history of MRSA and VRE infection or colonization, and previous exposure to vancomycin. All but 2 patients had VRE recovered at the time of VRSA isolation, either from a culture site or rectal specimen. One of the 2 patients who did not present with VRE infection or colonization at the time when VRSA was identified had a history of cultures positive for VRE; the other patient had a history of cultures positive for Enterococcus species, but unfortunately, susceptibility testing was never conducted. It is thought that, in all instances, VRE strains likely donated the vanA gene to S. aureus strains within a polymicrobial biofilm (e.g., wound, nephrostomy tube, or gastrointestinal tract) [15, 21]. The transferred vanA gene was able to be maintained in the S. aureus strains, even without the continued presence of VRE and vancomycin.
Although most of the case patients exhibited superficial signs of infection at the culture sites, none developed systemic signs or symptoms. This lack of systemic involvement may explain the low attributable morbidity and mortality. However, a recent study by Fox et al. [32] demonstrated that these strains were highly virulent in an experimental endocarditis rabbit model, have the ability to cause significant disease, and should be treated promptly when isolated. Currently, not enough data are available to recommend a standard treatment regimen for VRSA infection. All of the VRSA isolates were susceptible to a number of Food and Drug Administration-approved antimicrobial agents, and a recent study of experimental endocarditis using the first Michigan VRSA strain and a recent in vitro study using the VRSA strain from New York suggest that therapy with a synergistic combination of vancomycin plus a β-lactam may be an option [32, 33]. In addition to the variety of antimicrobials used as therapies for the VRSA infections, it is notable that surgical debridement and wound care were important treatment modalities in this case series.
Although researchers have been unable to demonstrate in vitro transfer of vanA from the VRE strains obtained from these VRSA-infected patients to S. aureus strains, conjugal transfer of vanA plasmid from VRSA strains to vanA-negative S. aureus strains has been demonstrated [15]. This suggests that, once the initial conjugal transfer of vanA from VRE strains to S. aureus strains occurs, the spread of vanA between S. aureus strains might be more likely. Therefore, to ensure implementation of infection-control precautions to prevent transmission, clinical microbiology laboratories need to issue prompt notification of a potential VRSA isolate to infection-control personnel [34, 35]. In addition, the extent of transmission that could occur before infection-control precautions are implemented should be assessed. Guidance has been developed to assist in developing a plan for this contact investigation [35]. To date, no transmission of VRSA beyond any of the case patients reported here has been identified.
One of the most prominent, outstanding questions is why 5 of the 7 cases of VRSA infection occurred in Michigan. This regional emergence likely occurred because of the convergence of several factors, including population characteristics, antimicrobial pressure, and the presence of VRE strains that are more likely to donate vanA operon. Michigan has a large population of individuals with chronic underlying conditions, such as diabetes and end-stage renal disease, both of which are conditions that appear to be associated with these infections. It is estimated that 590,000 adults (7.8% of the adult population) in Michigan have received diagnoses of diabetes [36]. Often, patients with diabetes develop other chronic health conditions, including impaired sensation in the feet, leading to unrecognized injuries (such as foot ulcers) from which VRSA has been isolated. In addition, diabetes is the leading cause of end-stage renal disease. Compared with the other states, Michigan has the eighth highest prevalence of diabetes, the ninth highest incidence of end-stage renal disease, and the ninth greatest number of dialysis and transplant facilities [37]. Patients receiving dialysis are at high risk of infection with invasive MRSA, leading to increased vancomycin exposure and, because of their reduced renal clearance of the drug, prolonged exposure to subtherapeutic levels of vancomycin. A recent report noted that the rate of invasive MRSA infection among patients receiving dialysis is higher than that among any other known patient population and is 100 times higher than that among the general population [38]. In addition to these patient population factors, Michigan was 1 of the first locations in the United States to document MRSA infection occurring outside of health care facilities in the 1980s [39]. Therefore, the early use of vancomycin in Michigan for treatment of these MRSA infections may have provided increased selective pressure for the development of vancomycin-resistant organisms.
The demographic characteristics and risk factors of Michigan's population may favor the emergence of VRSA, yet other regions have similar populations and have not witnessed this emergence. Therefore, specific characteristics of either the S. aureus or VRE strains circulating in Michigan may lead to a greater propensity for them to donate and/or acquire vanA. Although each VRSA strain has been distinct, the associated VRE strains from the Michigan patients contain a broad host-range Inc-18 conjugative plasmid that may be more likely to donate vanA to other bacterial species, compared with VRE strains from other regions [31, 40]. Studies are currently underway to examine the prevalence of these plasmids in enterococcal isolates from Michigan and nationally.
In summary, VRSA infection continues to be a rare occurrence. A few specific existing factors seem to have predisposed these case patients to VRSA infection, including prior MRSA and enterococcal infections or colonization, underlying conditions (such as chronic skin ulcers and diabetes), and previous treatment with vancomycin. Further studies are necessary to investigate the specific characteristics of the S. aureus and VRE plasmids isolated from these case patients. Appropriate antimicrobial prescribing by health care providers, adherence to recommended infection-control guidelines, and, ultimately, the control of both MRSA and VRE are necessary to prevent further emergence of VRSA strains.
Acknowledgments
We thank Sandip Shah, Glenn Fink, Stephen Haskell, Robert Jacobson, and Barbara Robinson-Dunn, for identification of the vancomycin-resistant Staphylococcus aureus isolates; Greg Fosheim, Sigrid McAllister, Linda McDougal, Nancye Clark, and Karen Anderson, for microbiologic characterization of the vancomycin-resistant Staphylococcus aureus isolates; Teri Lee Dyke, Shane Bies, Linda Burbank, Debbie Reid, Judy Moshos, John Torresan, Paula Keller, and Yma Wood, for assistance with contact investigations; and Dr. Tom Madhavan, Dr. Farris Gulli, and Carol Benton, for assistance with patient follow-up.
Potential conflicts of interest. All authors: no conflicts.
Footnotes
-
(See the editorial commentary by Tenover on pages 675–7)
-
The findings and conclusions in this report are those of the authors and do not necessarily represent the views of the Centers for Disease Control and Prevention.
- Received August 15, 2007.
- Accepted October 31, 2007.
- © 2008 by the Infectious Diseases Society of America
References
This Article
- AbstractFree
- » Full Text (HTML)Free
- Full Text (PDF)Free
Classifications
-
- Articles and Commentaries
Services
Citing Articles
Google Scholar
PubMed
Related Content
Share
- Email this article
-
CiteULike
Connotea
Delicious
Facebook
Google+
Mendeley
Twitter
Published on behalf of
IDSA Members: For your free access to this journal, log in via the IDSA members area.
The Journal
Impact Factor: 8.186
5-Yr impact factor: 7.898
For The Media
For Authors
- Services for authors
- Self-archiving Policy
- IDSA Conflict of Interest policy
- Instructions for Authors
- ICMJE Form
- Submit Now!
Open access options for authors visit Oxford Open
This journal enables compliance with the NIH Public Access Policy
Corporate Services
Alerting Services
Related IDSA Publications
