Evaluating the performance of the Alere PBP2a SA Culture Colony Test with the Vitek 2 Antimicrobial Susceptibility Test Card System as reference standard in coagulase-negative Staphylococcus species

Highlights • The PBP2a immunochromatographic (IC) assay was 66.7% sensitive and 100% specific for Staphylococcus epidermidis compared to microdilution minimum inhibitory concentration method.• The PBP2a assay is 100% sensitive and specific in predicting the susceptibility of non-epidermidis CoNS.• A larger number of isolates is needed to confirm our findings of high accuracy in predicting the susceptibility of non-epidermidis CoNS by PBP2a IC assay.


Background
The main hallmark of methicillin-resistant Staphylococcus aureus (MRSA) is the possession of the mecA gene, which produces penicillin-binding protein 2a (PBP2a).This confers resistance to beta-lactam agents, except for the fifth-generation cephalosporins ceftaroline and ceftobiprole.
The Alere PBP2a SA Culture Colony Test (Alere Scarborough, Inc., Scarborough, ME, USA) is an FDA-cleared, simple, qualitative, in vitro immunochromatographic (IC) assay that uses monoclonal antibodies for the rapid detection of PBP2a in S. aureus isolates.The test uses very sensitive recombinant monoclonal antibody fragments (rFabs) to detect the PBP2a protein from S. aureus .The clinical performance of the PBP2a IC test was compared with cefoxitin disk diffusion in S aureus isolates and showed a sensitivity of 99.1% and specificity of 99.2% on tryptic soy agar plates [ 1 ].
Because it takes about 5 minutes to identify whether an S. aureus isolate is MRSA without using an expensive high-tech machine, Fargo VA Healthcare System, an

Methods
Most coagulase-negative Staphylococcus (CoNS) species also possess the mecA gene; however, only a few studies have investigated the diagnostic performance of IC assays that target the PBP2a of CoNS [ 2 , 3 ].The aim of this study was to validate the performance of the PBP2a IC test for clinically significant CoNS isolates using the Vitek 2 AST Card System (BioMérieuex, Inc., Durham, NC, USA) based on the broth microdilution minimum inhibitory concentration method (BMMICM) as a reference standard.
Eighty clinical CoNS isolates at Fargo VA Healthcare System that met the inclusion criteria were tested with the Alere PBP2a SA Colony Test, and the results were compared with those of Vitek 2 AST, from October 2014 to September 2016.The criteria for including CoNS clinical isolates in this study were positive blood culture of ≥ 2 sets; positive urine culture of ≥ 10 5 CFU/mL of pure CoNS; positive wound culture of pure CoNS; and positive abdominal, pleural, and synovial fluid culture.These isolates were included because they were considered clinically significant based on American Society for Microbiology (ASM) recommendations [ 4 ].Exclusion criteria were a positive blood culture of only one set, positive urine culture of < 10 5 CFU/mL, and positive urine culture of ≥ 10 5 CFU/mL, but not pure CoNS or wound cultures that grew a CoNS isolate as one of the polymicrobial flora.
To verify the accuracy and examine the performance of the PBP2a SA Culture Colony Test, we randomly selected 60 S. aureus isolates, with Vitek 2 AST as the reference standard, before commencing the coagulasenegative Staphylococcus study.Both the PBP2a SA Culture Colony Test and Vitek 2 AST were performed according to the manufacturers' instructions.
This study was approved by the University of South Dakota's Institutional Review Board and the Fargo VA Health Care System's Research and Development committee.

Results
The 60 S. aureus isolates selected randomly as controls demonstrated 100% concordance between the PBP2a IC assay and Vitek 2 AST.
Eighty CoNS isolates from 80 unique patients were studied, and 2 samples were excluded owing to the failure  1 .Because the primary objective of this study was to solely investigate the performance of the PBP2a SA Culture Colony Test for CoNS isolates, we did not look into the clinical implications of our microbiological findings.
There was concordance for 68 CoNS isolates between the PBP2a IC assay and the Vitek 2 AST.However, there was discordance for 10 S. epidermidis isolates, which showed as negative on the PBP2a assay, despite oxacillin resistance (MIC > 4 μg/mL) showing on the Vitek 2 AST.A breakdown of the sources of the 10 S. epidermidis isolates is as follows: 7 from urine, 2 from wounds, and 1 from abdominal fluid.There was 100% concordance between the PBP2a IC assay and Vitek 2 AST for all non-epidermidis CoNS isolates.
The overall performance of the PBP2a IC assay was 66.7% sensitivity and 100% specificity for S. epidermidis when using the broth microdilution minimum inhibitory concentration method (BMMICM) as a reference standard ( Table 2 ).The positive predictive value was 100%, the negative predictive value was 76.2%, and the prevalence of oxacillin-resistant isolates was 48%.

Discussion and conclusion
Our study showed that the concordance rate between the PBP2a IC assay and BMMICM was 100% for Staphylococcus aureus and non-S.epidermidis coagulase-negative Staphylococcus .However, the IC assay failed to detect oxacillin resistance in 10 (33%) of the S. epidermidis isolates that showed oxacillin resistance by BMMICM.Two studies, one conducted in Japan and the other in France, found that an immunochromatographic assay for detecting PBP2a among non-Staphylococcus aureus isolates was 100% accurate.The smaller sample numbers of coagulase-negative Staphylococcus isolates in Matsui et al.'s and Mantion et al.'s studies ( n = 53 and 25, respectively) might explain why both studies were unable to detect any difference between the two methodologies [ 5 , 6 ].Another study conducted by Arnold and colleagues found that the Alere PBP2a culture colony test was sensitive and specific for non-aureus Staphylococcus species, but the studied isolates did not include S. epidermidis [ 7 ].
A recent study demonstrated that the methicillinresistant isolates of CoNS that initially showed as negative in the IC assay could be induced to show as positive by harvesting colonies from around the cefoxitin disk (cefoxitin induction) [ 3 ].We did not perform cefoxitin induction in our study and that might explain the high negative PBP2a rate among the oxacillin-resistant isolates.Another plausible explanation for the discordance between the PBP2a IC assay and BMMICM in our study is the heterogenous expression of the mecA gene by various species of CoNS [ 8 , 9 ].Suzuki and coworkers showed that CoNS can manifest oxacillin resistance via mechanisms other than the production of PBP2a [ 10 ].
In conclusion, our study demonstrated that the PBP2a IC assay has low sensitivity in determining the susceptibility of S. epidermidis to oxacillin.In contrast, the assay is highly accurate in predicting the susceptibility of non-epidermidis CoNS to oxacillin.However, because of the small sample size, the diagnostic accuracy for non-epidermidis CoNS needs further assessment with a larger number of isolates to confirm our findings.
This study was the result of work performed at the Fargo VA Health Care System and supported with resources and the use of the Health Care System's facilities.The contents do not represent the views of the U.S. Department of Veterans Affairs or the United States Government.

Funding
The authors did not receive any funding support for this study.
-bed Veterans Affairs Hospital in Fargo, North Dakota, has been using the PBP 2a IC test to rapidly identify MRSA from S. aureus clinical isolates.This has routinely been confirmed with the Vitek 2 Antimicrobial Susceptibility Test (AST) reference standard since August 2011.

Table 1
Sources and species of the 78 coagulase-negative Staphylococcus isolates.

Table 2 2
×2 table of PBP2a and oxacillin results for the 62 Staphylococcus epidermidis isolates.