Performance of BD Max StaphSR for Screening of Methicillin-Resistant Staphylococcus aureus Isolates among a Contemporary and Diverse Collection from 146 Institutions Located in Nine U.S. Census Regions: Prevalence of mecA Dropout Mutants

Abstract

This study determined the performance of BD Max StaphSR and the rate of methicillin-resistant Staphylococcus aureus (MRSA) with an unrecognized staphylococcal cassette chromosome mec (SCCmec) right-extremity junction (MREJ) region among 907 methicillin-resistant S. aureus (MRSA) and 900 methicillin-susceptible S. aureus (MSSA) isolates. The rate of mecA/mecC dropout mutants was also evaluated. Only three MRSA isolates (99.7% sensitivity; 904/907) were classified as MSSA by the BD Max StaphSR assay, due to negative results for MREJ. Eight MSSA isolates (99.1% sensitivity; 892/900) were assigned as MRSA. However, six of these MSSA isolates had the mecA gene confirmed by PCR and sequencing (99.8% sensitivity; 898/900). Overall, 7.1% (64/900) of MSSA isolates showed results compatible with a mecA dropout genotype.

TEXT

Several studies have reported a decline in the incidence of hospital-acquired methicillin-resistant Staphylococcus aureus (HA-MRSA) and invasive infections in US and European hospitals (1–6). However, the incidence of community-onset (CO) MRSA infection has varied according to geographic region (7–10). Despite variability in the occurrence of CO-MRSA and HA-MRSA invasive diseases, S. aureus persists as the most common organism responsible for human infections, and methicillin resistance remains the most commonly identified resistance in medical institutions (11). Therefore, proper infection control practices and antimicrobial stewardship strategies play important roles in controlling MRSA infections (12, 13).

Screening for MRSA carriers has become an important tool for early detection and to help prevent MRSA spread (14). Early generations of molecular assays targeting the mecA gene may provide false-positive results due to the copresence of methicillin-resistant staphylococci other than S. aureus (i.e., coagulase-negative staphylococci [CoNS]) (15). Performance evaluations of second-generation assays targeting the staphylococcal cassette chromosome mec (SCCmec)-orfX right-extremity junction (MREJ) region reported the presence of S. aureus carrying a genetic element that lacked the mecA (so-called dropout) mutant, again resulting in false-positive reports (16). Newer approaches targeting both mec and MREJ region sequences have been developed to minimize the likelihood of false-positive results, thus minimizing unnecessary isolation precautions (17). However, a false-positive reaction can still occur in the presence of mixed populations of methicillin-resistant CoNS and a dropout S. aureus mutant.

This study aimed to (i) determine the relative percentage rate of mecA/mecC dropout mutants among methicillin-susceptible S. aureus (MSSA) isolates collected from U.S. hospitals and (ii) determine the relative percentage rate of MRSA with unrecognized MREJ region sequences. A total of 907 MRSA and 900 MSSA isolates were included (at least 100 MRSA and 100 MSSA from each U.S. Census region). Isolates were collected from 146 U.S. hospitals during the 2013 SENTRY Antimicrobial Surveillance Program (see Table S1 in the supplemental material). Diversity within this collection was provided by the selection of isolates from multiple medical centers within each US Census region and selection of isolates displaying distinct antimicrobial susceptibility profiles. Isolates were also recovered from multiple different clinical specimen types (>30 types).

Antimicrobial susceptibility testing for oxacillin and cefoxitin was performed by disk diffusion (18, 19) and broth microdilution (20), according to CLSI recommendations. These isolates were defined as MRSA or MSSA by the oxacillin and/or cefoxitin susceptibility results obtained by the reference broth microdilution and/or disk diffusion method (18–20). Isolates were subjected simultaneously to the BD Max StaphSR assay kit according to the manufacturer's instructions with a small modification. As nasal samples are the primary specimen type used for MRSA screening, swabs were artificially prepared by placing them in fresh bacterial suspensions containing ∼1 × 10⁴ CFU/ml. The extra inoculum was removed and the swab placed in the manufacturer's sample buffer tube. The remaining steps followed the manufacturer's recommendation for specimen preparation. The BD Max StaphSR assay targets the nuc and mecA/C genes and the MREJ region. Dropout mutants were defined as those reactive for the targeted nuc gene (S. aureus) and MREJ region and mecA/C negative by the BD Max StaphSR assay. Isolates showing discrepant results regarding bacterial identification or the methicillin (oxacillin) status between the BD Max StaphSR and phenotypic assays were repeated. Remaining discrepant results on repeat testing were evaluated further by using matrix-assisted laser desorption ionization–time of flight (MALDI-TOF) and in-house PCR assays for detection of nuc and/or mecA/C (21) and epidemiology typing (i.e., multilocus sequence typing [MLST], spa and SCCmec typing) (22).

Among the 1,807 S. aureus isolates included in the study, all but 2 (99.9%; 1,805/1,807) were correctly identified by the BD Max StaphSR assay as S. aureus. These two negative results (complete absence of amplification signals) were confirmed on a second attempt, while positive signals were recorded from the internal control Cy5.5 channel (sample processing control) and bacterial identification confirmed by MALDI-TOF and the presence of nuc. Among the MRSA subset (i.e., oxacillin- and/or cefoxitin-resistant results), 904 (99.7%) isolates were also genotypically characterized as MRSA by the BD Max StaphSR kit (Table 1). Three MRSA isolates were classified as MSSA by the assay. Although the system detected the presence of mecA/C and the nuc gene, the final MRSA-negative results provided by the BD Max StaphSR were due to negative results for the MREJ (complete absence of amplification signals), which was confirmed in a second attempt. These isolates were screened for mecA/C using a multiplex PCR assay and confirmed to harbor the mecA gene by sequencing analysis. These isolates were ST1 (t922), ST772 (t657), and ST8 (t008) and harbored SCCmec types V, V, and IV, respectively.

TABLE 1.

BD Max StaphSR assay performance compared with phenotypic methicillin (oxacillin and cefoxitin) susceptibility results

Isolates (no. tested)a (n = 1,807)	Distribution of isolates by BD Max StaphSRb
	MRSA	MSSA
MRSA (907)	904	3
MSSA (900)	8c	892

^aMethicillin-resistant (MRSA) and -susceptible (MSSA) S. aureus clinical isolates defined by the oxacillin and/or cefoxitin susceptibility results obtained by the reference broth microdilution and/or disk diffusion methods according to CLSI guidelines (M02-A12, M07-A10, and M100-S25).

^bSensitivity and specificity of 99.7% (904/907) and 99.1% (892/900), respectively.

^cSix MSSA isolates were mecA/C positive using an in-house PCR screening assay, and genes were confirmed to be mecA on sequencing analysis. This would provide a corrected specificity of 99.8% (898/900).

A total of 892 (99.1%) MSSA isolates had BD Max StaphSR results in agreement with the methicillin phenotype (Table 1). Eight MSSA isolates characterized as MRSA by the system had results as follows: 6 isolates with confirmed susceptible oxacillin MIC results by broth microdilution (MIC, ≤0.25 to 1 μg/ml) and oxacillin (16 to 24 mm)/cefoxitin (22 to 29 mm) disk tests had mecA/C-positive results by PCR, which were confirmed to be mecA by sequence (see Table S2 in the supplemental material). The remaining 2 MSSA strains characterized by the system as MRSA were negative for mecA/C by PCR.

A total of 64 (7.1%) MSSA isolates showed results compatible with a dropout genotype (i.e., mecA/C negative and MREJ region positive). These putative dropout mutants were distributed among 51 institutions in 32 states in all nine U.S. Census regions (Table 2). Higher rates of dropout mutants among the MSSA population were observed in the East South Central (13.0%) and East North Central (11.0%) regions. The Mountain region had the lowest rate (2.0%), while the remaining regions had rates between 4.0% and 9.0%.

TABLE 2.

Distribution of dropout mutants among MSSA clinical isolates included in the study

U.S. Census region	No. of isolates	Mutantsa
		No.	%
1. New England	100	4	4.0
2. Mid-Atlantic	100	7	7.0
3. East North Central	100	11	11.0
4. West North Central	100	5	5.0
5. South Atlantic	100	9	9.0
6. East South Central	100	13	13.0
7. West South Central	100	6	6.0
8. Mountain	100	2	2.0
9. Pacific	100	7	7.0
Total	900	64	7.1

^aThe dropout mutants were defined as isolates with a negative signal from the carboxy-X-rhodamine (ROX) channel (mecA/C negative) and a reactive signal from the 6-carboxyfluorescein (FAM) channel (MREJ region positive).

BD Max StaphSR showed high sensitivity (99.7%) for the detection of MRSA compared with the phenotypic methicillin (oxacillin and/or cefoxitin) results. Three MRSA isolates showed MSSA results by the BD Max StaphSR, which were due to nonreactive signals for the MREJ region. One isolate belonged to clonal complex (CC) 8 (ST8-MRSA-IV), while the other two isolates were associated with CC1 (ST1-MRSA-V and ST772-MRSA-V). ST772 is a single-locus variant of ST1 (23), and a previous study reported that the assay did not recognize the MREJ region of 6.8% of tested isolates, which included those belonging to CC93, CC6, or CC1 (ST772) (24).

Nevertheless, these results suggest a low prevalence of MREJ regions among isolates in the United States that are not recognized by the primers and probes utilized by the systems. The isolates included in this study were collected from 146 medical centers across different geographic locations (nine U.S. Census regions). In addition, isolates were recovered from multiple specimen types and exhibited distinct antimicrobial susceptibility profiles. These broad selection criteria were intentionally applied to provide maximum strain variability, and the results indicate the ability of the system to correctly identify S. aureus and MRSA among diverse collections of organisms. However, several studies have documented the overwhelming presence of USA300 (CC8) and USA100 (CC5) carrying SCCmec types IV and II, respectively, in the United States (22, 25–27), while the MRSA population in Europe, Latin America, and Asia-Pacific countries seem to be more heterogeneous (22, 25, 28). Therefore, validation prior to clinical use in regions other than the United States seems prudent.

A total of eight MSSA strains were assigned as MRSA by BD Max StaphSR (99.1% specificity). However, six out of eight isolates in fact carried the mecA gene, which would provide a corrected specificity rate of 99.8%. Other studies performed in the United States and Europe have reported sensitivity and specificity rates of ≥94.3% and ≥97.7%, respectively (29, 30). However, it is important to mention that these studies evaluated the performance of the BD Max StaphSR from nasal swab samples or directly from blood specimens. Moreover, an overall rate of dropout mutants at 7.1% was documented, with higher rates in the East South Central and East North Central regions. When applying different methodologies, previous studies documented a prevalence of 4.6% for dropout mutants in a worldwide collection of isolates (15), with 3.5% to 3.8% in Canada (31, 32), 5.1% in Germany (33), and 8.3% among isolates collected from arrestees in a correctional institution in the United States (34). The results described herein and elsewhere emphasize the importance of correctly identifying dropout mutants to minimize false-positive results and thus limit unnecessary expenses of infection control practices.

Funding Statement

This study was funded by BD Diagnostics grant number BDS-MXMSRSA. Rodrigo E. Mendes, Amy A. Watters, Paul R. Rhomberg, David J. Farrell, and Ronald N. Jones are employees of JMI Laboratories, which received compensation fees for manuscript preparation.