Evaluation of the BD Max StaphSR Assay for Rapid Identification of Staphylococcus aureus and Methicillin-Resistant S. aureus in Positive Blood Culture Broths

Abstract

We evaluated the performance of the BD Max StaphSR assay for the direct detection of Staphylococcus aureus from blood culture medium. In a two-center trial, 155 blood cultures from the BD Bactec FX system and 212 from the bioMérieux BacT/Alert system were tested; 170 bottles yielded S. aureus, and all were identified correctly by the BD Max StaphSR assay. The assay required approximately 2.5 h, thus allowing rapid identification of blood cultures flagged positive.

TEXT

The BD Max StaphSR assay (BD, Quebec, Canada) is a fully automated, qualitative, in vitro diagnostic test for the direct detection of Staphylococcus aureus and methicillin-resistant S. aureus (MRSA) in patients at risk for nasal colonization. The test uses real-time PCR for the amplification of three DNA targets, i.e., SCCmec right-extremity junction (MREJ), thermostable nuclease (nuc), and methicillin resistance (mecA and mecC). Positivity for MREJ and mecA/mecC is required for the result “MRSA,” and detection of nuc or MREJ without mecA/mecC is interpreted as “positive for S. aureus.” The BD Max StaphSR assay was shown to have excellent sensitivity (96.4%) and specificity (93.6%), compared to combined direct and enriched cultures, for detection of S. aureus from anterior nares samples (1). In this study, we tested the performance of the StaphSR assay with a different sample type, namely, positive blood culture medium with Gram-positive cocci in clusters.

Bloodstream infections often result in the need for intensive care and are associated with high mortality rates and large economic burdens (2). S. aureus is one of the main causes of bloodstream infections (3, 4). Rapid identification of positive blood cultures is a prerequisite for timely targeted treatment of patients with sepsis (5–7). Quick differentiation of S. aureus from coagulase-negative Staphylococcus (CoNS) can reduce the costs of patient care and decrease hospitalization times (8). The use of molecular assays for direct detection of S. aureus in blood culture bottles has been shown to allow rapid diagnosis of S. aureus septicemia, which might improve patient treatment (6, 9, 10). The recently introduced BD Max system allows fully automated real-time PCR, but the currently available BD Max StaphSR assay is CE marked and FDA cleared only for use with nasal swabs (11).

In a first stage, we tested the use of the StaphSR assay with blood culture bottles (Bactec Plus Aerobic/F or Bactec Plus Anaerobic/F) that had been artificially inoculated with 1,000 CFU of MRSA, methicillin-susceptible S. aureus (MSSA), or methicillin-resistant coagulase-negative Staphylococcus (MR-CoNS). After bottles were flagged positive, aliquots from those bottles were tested by inoculating the sample buffer tube (SBT) of the assay. The blood culture bottles were tested again after an additional 18 to 24 h of incubation, to mimic delayed analysis and to ensure that high DNA levels did not result in competition with any of the PCR target amplifications. We tested three volumes, i.e., 15, 50, and 100 μl. All spiked blood cultures were correctly reported by the StaphSR assay as either MRSA positive or S. aureus positive (Fig. 1). All bottles with MR-CoNS were correctly reported as negative and gave positive signals in the channel for detection of mecA and/or mecC. Accurate results were obtained with all three volumes tested and with blood cultures tested directly after being flagged positive or after additional incubation for 18 to 24 h. Although blood culture medium is prone to interfere with PCR amplification (12), we did not observe any unresolved result due to inhibition of the internal sample processing control (SPC) with spiked blood cultures.

FIG 1.

C_T values from StaphSR assays performed with blood culture bottles spiked with the indicated bacteria. Assays were performed directly after Bactec Plus Aerobic/F blood culture bottles were flagged positive (direct [d]) or after an additional 18 to 24 h of incubation (24); 15, 50, or 100 μl of medium was used, and samples were run in triplicate. Indicated results are mean + SD. Missing bars, no amplification. Similar results were obtained for anaerobic bottles.

In a second stage, a prospective blind comparison of the StaphSR assay versus conventional culture for a consecutive series of positive blood culture bottles with microscopic results of Gram-positive cocci in clusters was performed. The study was performed at one site in Germany and one site in the United Kingdom. In total, 367 positive blood cultures were tested. One center used the BD Bactec FX system (testing 155 bottles [93 aerobic [AE], 58 anaerobic [AN], and 4 pediatric [PEDS]]) and the other the bioMérieux BacT/Alert system (testing 212 bottles [116 AE and 96 AN]). A 0.5-ml aliquot of positive blood culture broth was taken immediately after the bottle was flagged positive, at the same time as initiation of the routine diagnostic procedure. Aliquots were stored for up to 24 h at room temperature or 72 h at 4°C prior to testing, and 15 μl was used for the BD Max StaphSR assay. Storage did not alter the PCR results (data not shown). Blood cultures were selected to represent S. aureus and CoNS in approximately equal numbers. Duplicate patient samples were allowed if the samples were obtained in different draws. The molecular results were compared with the combined results of DNase plate testing (BD DNase test agar [Becton Dickinson, Heidelberg, Germany]), MRSA identification plate testing (BD BBL CHROMagar MRSA II medium [BD] or Oxoid Brilliance MRSA 2 agar [Oxoid, Basingstoke, United Kingdom]), latex agglutination testing (Pastorex Staph Plus kit [Bio-Rad, Munich, Germany] or Microgen Staph test [Microgen, Fleet, United Kingdom]), species identification by matrix-assisted laser desorption ionization–time of flight (MALDI-TOF) mass spectrometry (microflex system [Bruker Daltonics, Billerica, MA, USA]), susceptibility testing with a Vitek2 system (bioMérieux, Nuertingen, Germany), and eventually in-house PCR assays for mecA and femB.

During the evaluation period, the StaphSR assay had rates of inhibition (failed detection of the sample processing control) of 1.7% (BD Bactec FX system) and 0.5% (BacT/Alert system). Culturing identified 153 blood culture bottles as positive for MSSA and 17 as positive for MRSA. All 170 bottles with S. aureus (MRSA or MSSA) were identified correctly by the StaphSR assay (Table 1); one test had to be repeated because the first result was indeterminate. We also observed mixed infections with growth of MSSA and MR-CoNS, which were correctly identified as MSSA. The remaining 197 bottles tested were negative for S. aureus and mainly yielded CoNS. Of those 197 bottles, 190 were negative in the first analysis. Among the seven bottles with discrepant results, one culture-negative and PCR-positive sample indicated S. aureus with weak amplification of nuc (threshold cycle [C_T] of 37.8); a subsequent blood culture from the same patient yielded S. aureus. Six samples, all within one BD Max run, gave very weak signals with the StaphSR assay, which were not reproducible upon retesting. Contamination was assumed, and the samples were judged negative. Overall, the rates of agreement between culture results and StaphSR molecular detection results were 98.1% initially and 100% after discrepancy analysis.

TABLE 1.

Results of the BD Max StaphSR assay with blood culture medium

Culture result	No. with BD Max result of:
	S. aureus		MRSA
	Positive	Negative	Positive	Negative
BD Bactec FX
S. aureus (MSSA)	72	0	0	72
MRSA	2	0	2	0
Negative for S. aureus	1a	80	0	81
bioMérieux BacT/Alert
S. aureus (MSSA)	81	0	0	81
MRSA	15	0	15	0
Negative for S. aureus	0 (6b)	116 (110b)	0	116

^aGrowth of S. aureus in another blood culture from the same patient was observed.

^bSix samples gave weak positive signals in the first analysis that could not be reproduced in repeat testing.

In this study, access to the C_T values in addition to the final results was obtained (see Table S1 in the supplemental material). The C_T values for the nuc target were <26 for all MRSA-positive blood cultures (Fig. 2A) and most MSSA-positive blood cultures (Fig. 2B), indicating sufficient availability of DNA well above the limit of detection. Among the MRSA-positive bottles, one sample showed high C_T values for both MREJ (C_T = 36) and mecA/mecC (C_T = 37.9) despite of a low C_T value for nuc (C_T = 11.9) (Fig. 2A, black symbols), which might be related to a mixed MSSA-MRSA culture. From 153 bottles growing MSSA, 110 samples were positive only for the nuc target. Ten samples showed amplification of nuc and mecA/mecC, indicating mixed MSSA and MR-CoNS infections. Thirty-three samples were positive for nuc and MREJ but negative for mecA/mecC. They showed split distributions of C_T values for MREJ (Fig. 2B); one group had C_T values of <20. Those samples, because they were negative for mecA/mecC, might represent “dropouts” (13). A second group showed C_T values of >27, suggesting nonspecific amplification. The latter samples might be misclassified as MRSA in assays relying only on SCCmec right-extremity junction amplification. Indeed, for high MREJ C_T values, discrepancies with culture results have been observed when singleplex MRSA assays are used, and the use of a C_T value of 34 as the threshold for positivity has been suggested (14). In our study, with low MRSA prevalence, such samples were not rare (21.6%). The data confirm the superiority of MRSA assays with combined detection of different targets (15, 16). The assay delivered results approximately 2.5 h after initiation, allowing rapid analysis of blood cultures flagged positive. In our study, the time from positive flagging to identification as MSSA/MRSA by conventional means was 37.3 h, on average (data from one site). Depending on its integration into the laboratory workflow, the BD Max assay might decrease this time considerably. Within an antibiotic stewardship program, the assay could help to deescalate antibiotic treatment, which was demonstrated to be beneficial for patients with invasive staphylococcal disease (17).

FIG 2.

Distribution of C_T values for the indicated targets (nuc, MREJ, and mecA/mecC). (A) C_T values for samples positive for MRSA by the StaphSR assay (n = 17). Black symbols, one sample with discrepant C_T values for nuc versus MREJ and mecA/mecC. (B) C_T values for samples reported as S. aureus. Among those, C_T distributions are shown for three groups, namely, 110 samples that were positive only for nuc, 33 samples that were also positive for MREJ, and 10 samples in which mecA/mecC was also detected.

We conclude that the automated BD Max StaphSR assay can be used for rapid detection of MSSA and MRSA in positive blood cultures. The assay was compatible with both the BacT/Alert and BD Bactec FX systems. Rapid detection of MRSA and MSSA from positive blood cultures helps clinicians to optimize antibiotic treatment, thus improving patient outcomes.