GISA and Hetero-VRSA Coming Out of the Closet

abstract & commentary

Synopsis: Reduced susceptibility to vancomycin among clinical isolates of Staphylococcus aureus continues to add to the panoply of antibiotic-resistant organisms causing serious infections.

Source: Wong SS, et al. Diagn Microbiol Infect Dis 2000;36: 261-268.

Wong and colleagues in hong kong describe the clinical and microbiological aspects of four cases of bacteremia due to Staphylococcus aureus that demonstrated heterogenous resistance to vancomycin (hetero-VRSA). The patients were 53-78 years of age, each had severe underlying disease, and vancomycin administration failed to eradicate the organism. All four patients died.

Routine disk susceptibility testing failed to detect resistance to vancomycin (although the zone of inhibition around the teicoplanin disk was in the "intermediate" range). Heteroresistant strains were detected as satellite colonies around an aztreonam disk on the surface of Mueller-Hinton agar containing 4% NaCl and 4 m/mL vancomycin. The vancomycin MIC of the resistant subpopulations was, in each case, 8 m/mL, while those of teicoplanin were 8-24 m/mL. All isolates produced beta-lactamase and all possessed the mecA gene, which codes for methicillin resistance. Three of the isolates, obtained from patients at a single hospital, were indistinguishable by pulsed-field gel electrophoresis, while that obtained from a patient at a second hospital differed from the other three. Electron microscopy of the hetero-VRSA strains demonstrated a thickened cell wall when compared to MRSA strains. Checkerboard titration demonstrated significant synergy between ampicillin and vancomycin.

Comment by Stan deresinski, md, facp

The National Committee for Clinical Laboratory Standards (NCCLS) guidelines for susceptibility of S. aureus to vancomycin are given in the table.1 To date, there have been no reported isolations of S. aureus with MIC ³ 32 m/mL. These intermediately susceptible isolates have been termed vancomycin-intermediate S. aureus (VISA), but are now more generally known as glycopeptide-intermediate S. aureus (GISA). Wong et al distinguish between isolates with relatively homogenous reduced susceptibility for which they reserve the term GISA and those with heteroresistance, which they term hetero-VRSA.

Table-NCCLS Guidelines for Vancomycin Susceptibility1
Interpretation MIC
Susceptible £ 4 m/mL
Intermediate 8-16 m/mL
Resistant ³ 32 m/mL

Isolates ofS. aureuswith reduced susceptibility to vancomycin were first identified in Japan in 1996.2Since that time, they have also been described in many areas of the world, including the United States, Germany, France, Greece, Hong Kong, the United Kingdom, Italy, and Guatemala. Their prevalence varies greatly. In 1997, hetero-VRSA were found in hospitals throughout Japan, with 9.3% of MRSA from university hospitals and 1.3% from nonuniversity hospitals having this characteristic.2However, only 1.1% of MRSA yielded VISA subclones at one Italian hospital.3

As in the currently reviewed experience, disk susceptibility testing commonly fails to detect reduced susceptibility to vancomycin in these strains. The vancomycin agar screen plates used in this study, which are commercially available, are effective in detecting colonial growth from strains with reduced susceptibility, which should be confirmed by broth dilution testing. It has been recommended that this screening agar be used in all laboratories that use disk diffusion as their primary susceptibility method. The MicroScan method and the E test perform well when incubated for 24 hours.4

The mechanism of reduced resistance to vancomycin, which is distinct from that seen in enterococci, remains poorly defined. To date, hetero-VRSA and GISA (and hetero-VRCNS5) have only been detected on a background of methicillin resistance. VRSA and hetero-VRSA strains demonstrate an increased expression of PBP2, as well as an increased rate of cell wall turnover and they have abnormally thick cell walls, which become more apparent in GISA strains when the organism is exposed to sub-MIC concentrations of vancomycin.6,7 It has been proposed that the activity of vancomycin is impaired due to inability to reach its site of activity.8

Whether the virulence of these strains is altered is unclear. One in vitro study found, however, that reduced susceptibility to vancomycin is associated with increased biofilm formation and adherence to artificial surfaces when the organism is exposed to this glycopeptide.9

Infection with S. aureus with reduced vancomycin susceptibility on the background of methicillin resistance (and beta-lactamase production) reduces the available therapeutic choices. However, many of the strains isolated to date are susceptible to antibiotics whose mechanism of action does not target the cell wall, such as trimethoprim-sulfamethoxazole, rifampin, tetracyclines, quinupristin-dalfopristin, daptomycin and linezolid. The observation by Wong et al of synergy between ampicillin and vancomycin has been previously reported. Ampicillin has relatively high affinity for PBP2A and ampicillin-sulbactam has been demonstrated to be bactericidal in vitro against some VISA strains.10,11


1. National Committee for Clinical Laboratory Standards. 1999. Performance standards for antimicrobial susceptibility testing. NCCLS approved standard M100-S9. National Committee for Clinical Laboratory Standards, Wayne, PA.

2. Hiramatsu K, et al. Lancet 1997;350:1670-1673.

3. Marchese A, et al. J Clin Microbiol 2000;38:866-869.

4. Swenson JM, Hindler JA, Peterson LR. Special phenotypic methods for detecting antibacterial resistance. In: Murray PR, et al (eds). Manual of Clinical Microbiology. 7th ed. 1567.

5. Garrett DO, et al. Infect Control Hosp Epidemiol 1999; 20:167-170.

6. Pfeltz RF, et al. Antimicrob Agents Chemother 2000; 44:294-303.

7. Hanaki H, et al. J Antimicrob Chemother 1998;42:199-209.

8. Sieradzki K, Tomasz A. J Bacteriol 1997;179:2557-2566.

9. Wootton M, et al. San Francisco, CA; 39th ICAAC, Sept. 26-29, 1999.

10. Chambers HF. Clin Microbiol Rev 1997;10:781-791.

11. Hershberger E, et al. Antimicrob Agents Chemother 1999;43:717-721.