OUP user menu

Detection of Vancomycin-Intermediate Staphylococcus aureus With the Updated Trek-Sensititre System and the MicroScan System
Comparison With Results From the Conventional Etest and CLSI Standardized MIC Methods

Rohan Nadarajah MT(ASCP), Linda R. Post MT(ASCP), Catherine Liu MD, Steven A. Miller MD, PhD, Daniel F. Sahm PhD, Geo F. Brooks MD
DOI: http://dx.doi.org/10.1309/AJCPMV1P0VKUAZRD 844-848 First published online: 1 June 2010


Vancomycin-intermediate Staphylococcus aureus (VISA) organisms have minimum inhibitory concentrations (MICs) of 4 to 8 μg/mL and are often associated with vancomycin treatment failure. Detection of VISA has remained problematic. A comparison of 4 methods to detect VISA was done. Of the 20 VISA isolates, the Clinical and Laboratory Standards Institute (CLSI) broth microdilution method yielded susceptible end points of 2 μg/mL for 7, MicroScan (Siemens Healthcare Diagnostics, West Sacramento, CA) for 2, Trek Sensititre (Trek Diagnostic Systems, Cleveland, OH) for 1, and Etest (AB Biodisk North America, Piscataway, NJ) for none. Comparison with the CLSI method showed essential agreement for 95% or more for the Etest, MicroScan, and Trek methods; categorical agreement was as follows: Etest, 60%; MicroScan, 65%; and Trek, 60%. Reliance on a single automated method for determining vancomycin MICs could lead to misclassification of some VISA isolates as vancomycin susceptible. At least 2 methods, including the Etest, should be used when confirming VISA because of slight differences in results from different methods around the end points of 2 and 4 μg/mL.

Key Words:
  • Staphylococcus aureus
  • Vancomycin
  • Susceptibility testing
  • Antibiotic resistance

During the past decade, it has become increasingly apparent that vancomycin minimum inhibitory concentrations (MICs) of 4 μg/mL or more for Staphylococcus aureus are often associated with vancomycin treatment failure.13 In 2006, the Clinical and Laboratory Standards Institute (CLSI) changed the interpretive criteria for vancomycin-intermediate S aureus (VISA) to MICs of 4 to 8 μg/mL (previously 8–16 μg/mL) and for sensitive to 2 μg/mL or less.4,5 The laboratory determination of VISA has, however, remained somewhat problematic,69 although the manufacturers of susceptibility test systems have improved their products or developed new ones to detect VISA.

To examine our ability to properly measure vancomycin inhibitory concentrations of VISA we performed an evaluation of the updated Trek Sensititre System (Trek Diagnostic Systems, Cleveland, OH). The results using previously identified VISA isolates were compared with results from tests using materials and methods likely to be available in many clinical microbiology laboratories: automated broth microdilution tests and conventional Etest strips (AB Biodisk North America, Piscataway, NJ). The results were also compared with MIC results from CLSI standardized broth microdilution tests.

Materials and Methods

S aureus Isolates

Our laboratory, like most other clinical microbiology laboratories, does not have enough VISA isolates to evaluate the ability of a susceptibility test system to detect them. Thus, we obtained 18 previously identified VISA isolates from the Network on Antimicrobial Resistance in Staphylococcus aureus (NARSA; National Institutes of Health Contractor Eurofins Medinet, Chantilly, VA). These isolates were categorized as “glycopeptide intermediate staphylococci” by NARSA based on the work of the isolate donors and their published data. Many of these isolates have been associated with vancomycin treatment failure, and all had elevated vancomycin MICs when tested by the donor laboratories. The specific NARSA isolate identification designations are listed in Table 1. The 18 isolates represented a broad geographic area as follows (No. of isolates): France (5), Japan (1), and United States (12). The numbers of isolates from the United States were as follows: Illinois, 1; Kentucky, 1; Michigan, 2; New Jersey, 2; New York, 1; North Carolina, 1; Ohio, 1; Rhode Island, 1; Texas, 1; and West Virginia, 1. The data pages for the specific isolates and more specific information, including references to publications for individual isolates, are listed on the NARSA Web site (http://www.narsa.net).

Two recent VISA isolates from patients at the University of California San Francisco (UCSF) were included. One patient had a spinal tissue infection and the other a scalp infection, bringing the total tested to 20 isolates.

Three additional NARSA isolates were tested (NRS2, NRS35, and NRS64). These had been submitted as heterogeneous VISA (hVISA) with vancomycin MICs of 4 to 8 μg/mL.1012 These isolates had MICs of 2 μg/mL or less when initially tested at NARSA.

Broth Microdilution MICs

The CLSI standardized broth microdilution tray method used cation-adjusted Mueller-Hinton broth with serial log2 dilutions of vancomycin.4 Incubation was at 35°C for 24 hours. At NARSA, the CLSI standardized frozen microdilution MIC trays were obtained from Trek Diagnostic Systems. The frozen CLSI standardized trays for the UCSF isolates were obtained from the Microbiology Laboratory, University of California, Los Angeles.

At NARSA, inocula for the frozen MIC panels were prepared according to CLSI guidelines (M7-A8). Briefly, bacterial suspensions were prepared in Mueller-Hinton broth using growth from 16- to 20-hour-old blood agar plates. Suspensions were made in sterile deionized water to match the 0.5 McFarland standard by using a nephelometer, and this suspension was used to deliver a final inoculum of approximately 5 × 105 colony-forming units (CFU)/mL into each microdilution tray well. At UCSF, inocula were made by picking colonies from 5% sheep blood agar plates 24 hours old or less. The colonies were emulsified in 0.5 mL of brain heart infusion (BHI) broth to yield a turbid suspension. The BHI broth was added to 25 mL of sterile deionized water with 0.02% polysorbate-80, which was placed into the seed tray of a Dynatech Products MIC-2000 Inoculator (no longer manufactured). A 0.001-mL loop was used to sample the growth control well and streaked onto a blood agar plate to check purity; colony counts were used to confirm the inoculum concentrations of approximately 5 × 105 CFU/mL. Trays were incubated at 35°C for 24 hours. End points were read as growth or no growth by using a reflection plate reader with diffuse overhead light.

Conventional Epsilometer (Etest)

The conventional vancomycin Etest strips used at NARSA and UCSF used cation-adjusted Mueller-Hinton agar, a 0.5-McFarland inoculum measured with a nephelometer, and 24-hour incubation. The inoculum was sampled and streaked onto a 5% sheep blood agar plate to check purity. The Etest consists of a predefined gradient of vancomycin from 0.016 to 256 μg/mL applied to a thin inert plastic strip, which includes printed gradient concentrations. MICs were read by direct observation. End points were not changed to a log2 format.

MicroScan WalkAway

The MicroScan WalkAway SI 96 is an automated microbiology instrument that uses reconstituted dried panels and interpretive software to determine MICs of bacteria (Siemens Healthcare Diagnostics, West Sacramento, CA). The MicroScan PM 26 MIC trays (Siemens Healthcare Diagnostics) have 96 wells in a proprietary nonstandard size microdilution tray. The wells contain dilutions of the antimicrobial agents in cation-supplemented Mueller-Hinton broth that have been dried and stored at room temperature. Seven wells contained vancomycin in serial 2-fold dilutions from 0.25 to 16 μg/mL. Inocula were made by picking 3 to 5 colonies from 5% sheep blood agar plates 24 hours old or less. The colonies were emulsified in 3 mL of sterile deionized water to the turbidity of a 0.5-McFarland standard by using a nephelometer; 100 μL of suspension was added into a 25-mL tube of inoculum water with Pluronic (BASF, Florham Park, NJ). The RENOK system (Siemens Healthcare Diagnostics) was used to reconstitute the wells. A 0.001-mL loop was used to sample the growth control well and to streak onto a 5% sheep blood agar plate to check purity, yielding approximately 5 × 105 CFU/mL. Incubation was for 24 hours. MIC interpretations were made by the MicroScan software.

View this table:
Table 1

Trek Sensititre

The Trek Sensititre system is a microdilution method that uses fluorescence technology. Growth is determined by measuring the generation of a fluorescent product from a nonfluorescent substrate. A standardized suspension of the test organism is added to wells in the test plate that contain the substrate. Bacterial growth produces an enzyme that cleaves the substrate releasing the fluorophore, which is detected.

The tests were done in the Automated Reading and Incubation System (ARIS 2X), an automated plate handling, incubating, and reading module (Trek Diagnostic Systems). Inocula were made by picking 3 to 5 colonies from 5% sheep blood agar plates 24 hours old or less. The colonies were emulsified in sterile deionized water, and turbidity was measured in the AutoInoculator nephelometer (Trek Diagnostic Systems) and adjusted to a 0.5-McFarland standard. The standardized primary inoculum volume of 30 μL was added to the cation-adjusted Mueller-Hinton broth; this is a newly US Food and Drug Administration–approved increase in inoculum volume designed to enhance the detection of VISA and methicillin-resistant organisms (previously approved volume, 10 μL). The GPALL1F plates (Trek Diagnostic Systems) contained vancomycin in 8 wells in 2-fold dilutions from 0.25 to 32 μg/mL. These dried plates were stored at room temperature. The plates were reconstituted with the S aureus–inoculated Mueller-Hinton broth by using the AutoInoculator. A 0.001-mL loop was used to sample the growth control well and to streak onto a 5% sheep blood agar plate to check purity and colony count, representing approximately 5 × 105 CFU/mL. The plates were loaded into the ARIS instrument. After 24 hours’ incubation the end points were read by the instrument.

Calculation of Essential and Categorical Agreements

By using the CSLI method and Etest as reference standards, we calculated essential agreement as the percentage of isolates having MICs within ± 1 log2 dilution and categorical agreement as the percentage of isolates having identical interpretations of sensitive, intermediate, or resistant. These calculations were performed using Trek dried panels tested at UCSF and not those tested at NARSA because 2 of the isolates were not tested by this method at NARSA.


The data are shown in Table 1. Of the 20 VISA isolates, the CLSI method yielded susceptible end points of 2 μg/mL for 7. The MicroScan and Trek yielded MICs of 2 μg/mL for 2 isolates and 1 isolate, respectively. All 7 isolates with CLSI broth dilution MICs of 2 μg/mL had MICs of 3 μg/mL or more when tested by the Etest.

When the CLSI method was used as the reference standard, there were essential agreements (no more than 1 log2 dilution difference in result) of 95% or more for the Etest, MicroScan, and Trek Sensititre (dried panel) methods. The categorical agreements (same sensitive, intermediate, or resistant interpretations) were as follows: Etest, 60%; MicroScan, 65%; and Trek, 60%.

When the Etest was used as the reference standard, there was essential agreement of 100% for the MicroScan and Trek Sensititre (dried panel) methods. The categorical agreements were 85% for the MicroScan results and 90% for the Trek results.

The UCSF and NARSA results from testing with the dried Trek panels were both included in Table 1 to compare the results using the same system in different laboratories. The results were identical for 16 of the 18 NARSA isolates. One isolate yielded a 2-μg/mL MIC at NARSA and a 4-μg/mL MIC at UCSF, whereas the other isolate yielded an 8-μg/mL MIC at NARSA and a 4-μg/mL MIC at UCSF. The 3 hVISA isolates all gave vancomycin MICs of 2 μg/mL or less by all 4 methods used in this study, and no isolated colonies were seen within the zone of inhibition on Etest plates (data not shown).


The Etest, MicroScan, and Trek yielded very similar results, whereas the CLSI broth microdilution method often yielded slightly lower MICs. This was demonstrated by the categorical discrepancies in which the CLSI method yielded 2 μg/mL susceptible end points for 7 isolates, whereas the MicroScan, Trek, and Etest yielded MICs of intermediate susceptibility for nearly all of the same isolates. It has been previously reported that the Etest and MicroScan (and the Phoenix automated system, Becton Dickinson, Franklin Lakes, NJ) gave results that tended to be 1 dilution higher than the CLSI broth microdilution reference method.9 Because of the clinical importance of likely vancomycin treatment failure with MICs of 4 μg/mL or more, the use of the CLSI method seems to be less than optimal for detecting VISA. It is important for the clinical laboratory to accurately identify VISA because alternatives to vancomycin are recommended for the treatment of such infections. For these reasons, a second susceptibility test using a different method should be considered when reduced vancomycin susceptibility is detected in S aureus. This is particularly important for patients with isolate MICs of 2 μg/mL or more and when there is limited or no clinical response to vancomycin therapy.

It was previously reported that the Trek Sensititre system (and the Vitek 2 system, bioMérieux, Durham, NC) “tended to categorize VISA strains as susceptible”9 with MICs that tended to be lower than in the present study. Subsequently, the Trek Sensititre modified the Food and Drug Administration–approved inoculum volume from the 10 μL previously recommended and used at NARSA to the newly recommended 30 μL in current use. The larger volume inoculum was used at UCSF and may have been associated with the 1 isolate that had a 2-μg/mL MIC at NARSA and a 4-μg/mL MIC at UCSF.

Results from several other methods have been reported for the detection of VISA. One method is the modified Etest using an inoculum made to a 2-McFarland standard.6 The medium recommended for this high inoculum Etest is BHI agar; because this medium may not be readily available in clinical laboratories, the modified Etest was not included in our study. Similarly, prototype Etest strips have been specifically designed for glycopeptide resistance detection.13,14 These strips are for research use only and were not included in this study.

The CLSI disk diffusion method and BHI agar with 6 μg/mL of vancomycin should not be used for the detection of VISA. All of the NARSA isolates gave zone sizes of susceptible when tested by the CLSI disk diffusion method (data not shown). Similarly, screening with BHI agar with 6 μg/mL of vancomycin yielded mixed results, as expected (data not shown). Comparable results have been shown by others.9

The 3 NARSA probable hVISA isolates had vancomycin MICs of 2 μg/mL or less by all 4 methods used in this study. The hVISA strains generally have vancomycin MICs in the susceptible range, often 2 μg/mL,15 but contain low numbers of S aureus with vancomycin MICs in the intermediate range; detection may require a population analysis profile.5,16 Inability to detect hVISA is a limitation of current susceptibility testing methods. The relative clinical significance and treatment outcome of hVISA methicillin-resistant S aureus when compared with vancomycin-susceptible methicillin-resistant S aureus is unknown. Several retrospective studies suggest that infection with hVISA is associated with treatment failure,17,18 whereas others, including a recent prospective study,19 found no differences in outcome.20 In the absence of a practical, readily available detection method and consistent data regarding treatment outcomes, efforts at routine identification of hVISA are not warranted in clinical laboratories at this time.

Of the methods tested, the CLSI method is most likely to misclassify VISA isolates as susceptible, whereas the Etest is the most sensitive for detection of VISA. Laboratories should maintain caution when reporting vancomycin MICs using the CLSI method or automated test systems because these may fail to identify some VISA isolates from patients at risk for vancomycin treatment failure. Confirmation of VISA isolates around the end points of 2 and 4 μg/mL would increase the sensitivity for VISA detection because of slight differences in results from one method to another near the breakpoint for sensitivity. If an automated system such as Trek or MicroScan (or Phoenix) were used initially, then another automated test or, preferably, an Etest seems advisable for confirmation of S aureus with intermediate vancomycin susceptibility.


We acknowledge the National Institutes of Health Network on Antimicrobial Resistance in Staphylococcus aureus, contract NO1-AI-95359, for the provision of VISA strains. We thank Contract Administrator Mai Nguyen for helpful assistance.


Upon completion of this activity you will be able to:

  • describe the clinical importance of infection with vancomycin-intermediate Staphylococcus aureus (VISA).

  • outline the clinical testing methods used to identify VISA strains.

  • list the agreements for VISA testing among various test methods and recommend an algorithm with improved detection sensitivity.

The ASCP is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. The ASCP designates this educational activity for a maximum of 1 AMA PRA Category 1 Credit ™ per article. This activity qualifies as an American Board of Pathology Maintenance of Certification Part II Self-Assessment Module.

The authors of this article and the planning committee members and staff have no relevant financial relationships with commercial interests to disclose.

Questions appear on p 977. Exam is located at www.ascp.org/ajcpcme.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
View Abstract