Rapid Detection of Methicillin-Resistant Staphylococcus aureus (MRSA) and Methicillin-Susceptible Staphylococcus aureus (MSSA) Using the KeyPath MRSA/MSSA Blood Culture Test and the BacT/ALERT System in a Pediatric Population

Positive BacT/ALERT Pediatric FAN (fastidious antibiotic neutralization) blood culture bottles (bioMérieux, Inc, Durham, North Carolina) were prospectively tested in the microbiology laboratory at the Children's Hospital of Philadelphia (Philadelphia, Pennsylvania) during 3 periods: November to December 2010, April to May 2011, and February to March 2012. Staphylococcus aureus isolates were identified using colonial morphology, catalase reaction, and Staphaurex latex testing (Remel, Lenexa, Kansas). The rapid Alere PBP2a test (Alere Scarborough, Inc, Scarborough, Maine) and the Vitek 2 system (bioMérieux) were used to determine methicillin-susceptibility status.

The MMBT was performed according to the manufacturer's instructions. The MMBT uses an S aureus–specific bacteriophage cocktail that infects S aureus cells. Within 24 hours of the BacT/ALERT instrument signaling blood culture positivity, 10 μL of positive blood culture samples were pipetted into tubes containing predispensed “ID” reaction medium (with bacteriophage and culture medium) and “RS” reaction medium (with bacteriophage, culture medium, and cefoxitin). The tubes were incubated at 35°C in ambient air for 5 hours (±20 minutes), during which, bacteriophage rapidly replicated causing cellular lysis and increasing bacteriophage concentrations. The contents of the ID and RS reaction tubes were then dispensed into the ID and RS wells of the detector device. Detection of phage amplification by phage-specific antibodies in both the ID and RS wells indicated the presence of MRSA. Detection in only the ID well indicated the presence of MSSA.

The sensitivity, specificity, positive predictive value, and negative predictive value of MMBT in detecting S aureus, MSSA, and MRSA were calculated with 95% confidence intervals. Routine pathogen identification and susceptibility testing results were used as the gold standard.

Results are summarized in the Table. Of 188 positive blood cultures tested, 119 (63%) had Gram-positive cocci in clusters. All (119 of 119; 100%) grew Staphylococcus spp. Of these 119 positive results, 46 (39%) grew S aureus (26 MSSA [57%] and 20 MRSA [43%]), and the remainder grew coagulase-negative Staphylococcus spp. In the remaining positive blood cultures (n = 69; 37%), the organisms isolated included other Gram-positive pathogens (Streptococcus spp, Corynebacterium spp), Gram-negative pathogens (Enterobacteriaceae, nonfermenting organisms), Candida spp, and Fusarium spp.

In positive blood cultures with Gram-positive cocci in clusters, MMBT had a sensitivity of 87% (95% confidence interval [CI], 74%–94%) for the detection of S aureus, 81% (95% CI, 62%–92%) for MSSA, and 95% (95% CI, 75%–100%) for MRSA. Specificity was consistently 100%. Negative predictive value was 92% (95% CI, 84%–96%) for S aureus, 94% (95% CI, 88%–98%) for MSSA, and 99% (95% CI, 94%–100%) for MRSA. Positive predictive value was consistently 100%. Positive blood cultures that did not grow S aureus uniformly tested negative in both the ID and RS wells.

The MMBT failed to detect S aureus in 6 of the 46 blood cultures (13%) with S aureus present (5 MSSA [83%] and 1 MRSA [17%]). In 3 (50%) of these cases (2 MSSA [67%] and 1 MRSA [33%]), the ID well tested negative for S aureus initially but tested positive when MMBT was repeated within the recommended 24-hour testing window. In the 3 remaining cases (all MSSA), MMBT was negative when repeated. We did not observe any cases of MRSA designated as MSSA or vice versa.

Assays that rapidly and accurately detect MSSA and MRSA from positive blood cultures have the potential to improve clinical outcomes in patients with MSSA bacteremia and may help to limit inpatient transmission of MRSA. To date, the FDA-approved BD GeneOhm StaphSR (BD Diagnostics, San Diego, California) and the currently unavailable Cepheid Xpert MRSA/SA BC (Cepheid, Sunnyvale, California) polymerase chain reaction products have dominated the rapid blood culture MSSA/MRSA detection market. The latter has undergone 2 product recalls, most recently, in January 2012, because of high rates of invalid results.⁹  In addition, clinical data generated after FDA approval revealed cases of MRSA bacteremia testing falsely as MSSA by both assays.¹⁰  This phenomenon was attributed to failure of the assay primers to detect variants of SCCmec.¹¹  Misidentification of MRSA as MSSA would be considered a serious hazard to patients because it could prompt inappropriate de-escalation of vancomycin therapy to cefazolin or cloxacillin, leaving the patient on ineffective therapy for a potentially life-threatening condition. Instances of MSSA testing positive for MRSA have also been reported. This observation has been attributed to amplification of the SCCmec target in the absence of the mecA gene (the “empty cassette” phenomenon).¹²  Data in the literature suggest that matrix-assisted laser desorption ionization time-of-flight mass spectrometry (MALDI-TOF MS) may accurately detect S aureus directly from positive blood cultures.¹³  However, MALDI-TOF MS is not FDA-approved nor capable of reliably performing antibiotic susceptibility testing at this time. An alternative approach to rapid diagnosis of MSSA and MRSA bacteremia is clearly desirable.

By using bacteriophage technology, the KeyPath MMBT is able to circumvent the issue of DNA target variation. In this study, however, we observed 6 instances in which blood cultures that were positive for S aureus tested negative. The control lines on the detection device were appropriately reactive in all 6 cases. In 3 cases, repeat testing yielded positive results. We believe that additional incubation of the blood samples allowed bacterial replication to continue, enhancing phage replication to a level that was detectable by the assay. In the 3 remaining cases, we believe that MSSA was present at levels that were too low to generate sufficient phage replication to allow detection. The failure of the ID well to detect S aureus indicated that reporting of negative S aureus results was problematic. An option might be to report S aureus-negative results as “preliminary,” awaiting results from traditional culture procedures. These findings were consistent with data presented by Kingery et al.¹⁴ 

This study did have limitations. First, specimen accrual was challenging. In an attempt to achieve a larger sample size, the original study protocol was modified to add a third study period. We feel that this was related, in part, to the relatively low levels of central line–associated bloodstream infections at our institution, which were achieved through implementation of aggressive infection-control initiatives during the past decade. Second, the turnaround time of this test is 5 hours. For MMBT to translate successfully to accelerate optimization of antimicrobial therapy, immediate testing of positive blood cultures and the availability of clinicians to modify antimicrobial therapy would be required, ideally, on a 24-hour basis. Many centers may not have such resources available.¹⁵  Third, the use of a commercially available polymerase chain reaction product would have allowed a gold standard definition that included a molecular standard. We opted to rely on conventional identification and susceptibility testing methods because of concerns pertaining to the performance of existing FDA-cleared products.

In summary, at this time, there is considerable need for accurate, reliable, and rapid MSSA/MRSA detection tests that can be performed directly on positive blood cultures. The KeyPath MMBT provides an option. The positive predictive value of 100% for MSSA and MRSA detection suggested that positive results could be reported immediately to clinicians, but the sensitivity of this assay currently limits the ability to immediately report negative results. Further research aimed at reducing these false-negative S aureus results would be beneficial.

We thank the technologists in the microbiology laboratory at The Children's Hospital of Philadelphia for their assistance with specimen testing. The KeyPath MMBT kits used in this study were donated in kind by MicroPhage, Inc.