Evaluation of the AdvanDx VRE EVIGENE Assay for Detection of vanA in Vancomycin-Resistant Staphylococcus aureus

Abstract

AdvanDx VRE EVIGENE, a commercial vanA/vanB DNA hybridization assay to identify vancomycin-resistant enterococci (VRE), was evaluated for the detection of vanA in Staphylococcus aureus. Performance was assessed using S. aureus, VRE, and vancomycin-intermediate and -susceptible isolates. The assay demonstrated 100% sensitivity and specificity when analyzed visually and by optical density.

Vancomycin continues to be an important antimicrobial agent for the treatment of methicillin-resistant Staphylococcus aureus infections and for empirical therapy of staphylococcal infections in areas where methicillin-resistant S. aureus prevalence is high (3, 8). The widespread use of vancomycin has led to the emergence of S. aureus with reduced susceptibility to vancomycin. Vancomycin-intermediate S. aureus (VISA), for which the vancomycin MICs are 4 to 8 μg/ml (5), have been reported worldwide (15). The VISA phenotype, associated with a marked thickening of the cell wall, is not mediated by van genes, and the specific genetic mechanisms are not well understood (6). A less frequent but more concerning phenomenon involves the six vancomycin-resistant S. aureus (VRSA) isolates, for which the vancomycin MICs are ≥16 μg/ml (5), observed in the United States (3, 13, 16-18). All six VRSA isolates carry the vanA resistance determinant, most likely acquired from coinfecting strains of vancomycin-resistant enterococci (VRE) (4, 17, 18).

Rapid identification of VRSA is important for directing appropriate therapy for infected patients and for proper implementation of infection control measures to prevent transmission. The VRE EVIGENE Detection assay (AdvanDx, Woburn, MA) has been evaluated previously for its ability to detect vanA and vanB genes in VRE (11). In this study, we evaluated the ability of the VRE EVIGENE assay to detect vanA in S. aureus. Our evaluation included the six independent VRSA isolates described to date as well as a variant of VRSA 3 (VRSA 3b) isolated from the same patient (17). The resistant phenotype of VRSA 3b is stable (vancomycin MIC of ≥256 μg/ml), while the vancomycin-resistant phenotype of VRSA 3a is unstable, resulting in a susceptible phenotype in the absence of vancomycin selection (17). The seven VRE isolates associated with the VRSA cases were also tested, as these isolates are the most likely donor organisms for the vanA gene (9, 17, 18). Only one isolate, VRSA 2, did not have an associated VRE. Other organisms evaluated included six VRE with different vancomycin resistance determinants (vanC to vanG), five geographically distinct VISA isolates, two vancomycin-susceptible S. aureus isolates, and one vancomycin-susceptible Enterococcus faecalis isolate (Table 1). Isolates were identified using standard biochemical methods (1). All isolates had been previously characterized for vancomycin resistance by reference broth microdilution (5) and for the presence of the van genes by PCR (7).

TABLE 1.

Evaluation of AdvanDx VRE EVIGENE assay

Organism	Source	Reference	van allelea	Expected result	Optical densityb		Interpretation
					Mean	SD	Optical density	Visual
VRSA 1 from MI	Foot wound	3	vanA	+	2.035	0.143	+	+
E. faecalis (VRSA 1)	Foot wound	3	vanA	+	0.972	0.056	+	+
VRSA 2 from PA	Foot wound	16	vanA	+	2.352	0.038	+	+
VRSA 3a from NY	Urine	17	vanA	+	0.772	0.082	+	+
VRSA 3b from NY	Nephrostomy tube	17	vanA	+	2.318	0.106	+	+
Enterococcus faecium (VRSA 3)	Rectal	17	vanA	+	0.866	0.045	+	+
VRSA 4 from MI	Gangrenous toe	13	vanA	+	1.078	0.054	+	+
E. faecalis (VRSA 4)	Rectal	18	vanA	+	1.237	0.05	+	+
VRSA 5 from MI	Abdominal wound	18	vanA	+	2.229	0.053	+	+
E. faecalis (VRSA 5)	Abdominal wound	18	vanA	+	1.08	0.175	+	+
E. faecium (VRSA 5)	Rectal	18	vanA	+	0.703	0.15	+	+
VRSA 6 from MI	Foot wound	18	vanA	+	2.349	0.052	+	+
E. faecalis (VRSA 6)	Rectal	18	vanA	+	1.229	0.105	+	+
Enterococcus avium (VRSA 6)	Rectal	18	vanA	+	0.76	0.09	+	+
E. faecalis (A256)		14	vanA	+	0.794	0.243	+	+
E. faecalis (V583)		12	vanB-1	+	0.734	0.064	+	+
E. faecalis (SF 300)		10	vanB-2	+	2.158	0.007	+	+
Enterococcus gallinarum (VR42)		12	vanC-1	−	0.148	0.011	−	−
Enterococcus casseliflavus (ATCC 25788)		12	vanC-2	−	0.186	0.038	−	−
E. faecium (N97-0330)		2	vanD-3	−	0.167	0.033	−	−
E. faecium (N03-0072)		2	vanD-5	−	0.169	0.007	−	−
E. faecalis (N00-0410)		2	vanE	−	0.159	0.029	−	−
E. faecalis (24E01247)		2	vanG	−	0.135	0.022	−	−
E. faecalis (ATCC 29212)		5	NA	−	0.184	0.028	−	−
S. aureus (ATCC 25923)		5	NA	−	0.127	0.014	−	−
S. aureus (ATCC 43300)		5	NA	−	0.163	0.035	−	−
VISA 27606			NA	−	0.135	0.009	−	−
VISA 30605			NA	−	0.147	0.011	−	−
VISA 08116			NA	−	0.173	0.033	−	−
VISA 30607			NA	−	0.127	0.009	−	−
VISA 963sm			NA	−	0.11	0.019	−	−

^aNA, not applicable.

^bPositive result was an optical density at 490 nm of ≥0.35 (Victor² Multilabel Plate Reader; Perkin-Elmer, Boston, MA).

All isolates evaluated in this study were grown overnight on nonselective trypticase soy agar containing 5% defibrinated sheep blood (Becton Dickinson, Sparks, MD) at 35°C. Isolates were tested according to the manufacturer's protocol by using the included assay reagents provided in drop-dispensing bottles in an enzyme-linked immunosorbent assay microtiter format with both visual and spectrophotometric interpretation criteria. Wells with no color change or exhibiting a faint pink color were considered negative, while dark pink to red wells were positive. For comparison, optical density (A₄₉₀ of ≥0.35 is positive) was also measured with a microplate reader (Wallac1420 Victor²; Perkin-Elmer, Boston, MA).

Positive controls included E. faecalis strains carrying vanA, vanB-1, and vanB-2. VRE carrying vanC to vanG alleles and vancomycin-susceptible S. aureus and E. faecalis strains were included as negative controls (Table 1). All testing was performed in triplicate.

The VRE EVIGENE assay demonstrated 100% sensitivity and specificity (n = 31) for the detection of vanA in S. aureus and vanA or vanB in enterococci by both visual interpretation and optical density measurements (A₄₉₀). VRSA and VRE containing vanA and VRE containing vanB determinants were positive by both interpretation methods. VRE carrying van alleles vanC-1, vanC-2, vanD-3, vanD-5, vanE, and vanG were negative, as were the vancomycin-susceptible isolates. VISA isolates (n = 5) were also negative with the VRE EVIGENE assay, as expected (6) (Table 1).

One limitation of this study is the small number of VRSA isolates, all of which were vanA positive. No vanB-positive S. aureus isolates have been described to date, and therefore, the ability of the VRE EVIGENE assay to detect vanB in S. aureus could not be assessed. Additionally, interpretation of results based solely on a visual color change observation is subjective and could lead to indeterminate results that would require additional testing, although this phenomenon was not observed in our evaluation.

Nevertheless, the VRE EVIGENE assay reliably and accurately detects vanA in S. aureus isolates. The assay is easy to perform, requiring approximately 15 min of hands-on time, with a 4-h turnaround time. The assay is not yet approved by the Food and Drug Administration for detecting vanA or vanB in enterococci or for detecting vanA in S. aureus, but the VRE EVIGENE assay may be a useful nondiagnostic method for reference laboratories to further evaluate isolates with elevated vancomycin MICs.