Evaluation of a New Chromogenic Medium, MRSA Select, for Detection of Methicillin-Resistant Staphylococcus aureus

Abstract

We compared MRSA Select to mannitol-salt agar with 8 μg/ml cefoxitin for the detection of methicillin-resistant Staphylococcus aureus (MRSA) from 6,199 clinical samples submitted for MRSA screening. The sensitivities and specificities of MRSA Select and mannitol-salt agar with cefoxitin were 98% and 92% versus 90% and 78%, respectively (P < 0.0001). Most (96%) MRSA were detected after overnight incubation using MRSA Select.

The clinical burden and costs associated with infection due to methicillin-resistant Staphylococcus aureus (MRSA) in hospitalized patients can be substantial (2, 10). It is now recognized that screening high-risk hospitalized patients to detect MRSA colonization is an effective infection control measure to reduce the risk of transmission of the organism within health care facilities (1, 14, 17). Traditional methods for the detection of MRSA from surveillance and screening specimens have included the use of a variety of selective and differential solid media (3, 12, 18, 19, 23). The use of broth culture enhancement may improve the yield of MRSA, but it is more laborious and may delay identification of the organism (7, 18, 22). The addition of the cephamycin antibiotic cefoxitin into mannitol-salt agar appears to improve the sensitivity for isolation of MRSA (20, 21; L. Louie, D. Soares, M. Vearncombe, H. Meaney, and A. Simor, Abstr. 105th Gen. Meet. Am. Soc. Microbiol., abstr. C105, p. 123, 2005).

More recently, chromogenic media incorporating chromogenic enzymatic substrates and a variety of antimicrobial agents have become available for detection of S. aureus, including methicillin-resistant strains (4-6, 8, 11, 13, 16, 21). Some of the evaluations of these media involved only stored collections of isolates, while others included a relatively small number of clinical specimens. We evaluated a new, commercially available chromogenic medium, MRSA Select (Bio-Rad Laboratories, Marnes-la-Coquette, France) and compared it to mannitol-salt agar with 8 μg/ml cefoxitin (MRSA MSA modified agar; Quelab Laboratories, Inc., Montreal, Quebec, Canada) for its ability to detect MRSA from patient specimens submitted for MRSA screening. The objectives of this evaluation were as follows: (i) to determine the sensitivities of the two culture media for detecting MRSA, (ii) to determine the abilities of the media to detect MRSA after overnight and 48-h incubation, and (iii) to determine which screening culture medium requires less workup of isolates subsequently determined not to be MRSA (specificities of the media).

A total of 6,199 specimens submitted for MRSA screening (surveillance) from 1,883 hospitalized patients were included in the study, consisting of nares swabs (n = 2,483), perianal/rectal swabs (n = 2,312), samples from vascular catheter exit sites (n = 647), and skin/soft tissue (n = 632), sputum (n = 58), and urine (n = 67) samples. All specimens were plated onto MRSA Select and mannitol-salt agar with 8 μg/ml cefoxitin (MSFOX) using the Isoplater 180 (Vista Technologies Inc., Edmonton, Alberta, Canada). The order of which plates were plated first alternated between the two media every 100 specimens. All plates were incubated at 35°C in ambient air; MRSA Select plates were stored and incubated in the dark. The plates were read independently at 18 to 24 h and at 48 h by experienced technologists, blinded to results obtained by each other. Suspicious colonies (small pink colonies on MRSA Select; yellow or pale yellow colonies on MSFOX) were identified as MRSA using conventional methods, including Gram staining, Pastorex Staph latex (Bio-Rad), tube coagulase (BD BBL coagulase plasma; Becton Dickinson Company, Sparks, MD), and latex agglutination for PBP2a (MRSA-Screen; Denka Seiken, Tokyo, Japan).

MRSA was isolated from 181 (3%) of the 6,199 specimens and from 77 (4%) of the 1,883 patients that were screened. The numbers of MRSA isolated from each medium after overnight and 48-h incubations are summarized in Table 1. MRSA Select detected a total of 177 (98%) MRSA compared to 163 (90%) isolates that grew on MSFOX (P = 0.003). Although MRSA Select failed to recover MRSA from four specimens, this did not result in misdiagnosis of any of these patients, as other specimens from these patients collected concurrently yielded MRSA using MRSA Select. Of the 177 MRSA recovered on MRSA Select, 169 (96%) grew after overnight incubation, whereas only 125 (77%) MRSA were isolated with MSFOX after overnight incubation (P < 0.0001). The yields of MRSA from various anatomic sites were similar for the two media, except for perianal/rectal swabs, where MRSA Select detected 64 MRSA and MSFOX detected only 55 (P = 0.003). The sensitivities and specificities for all specimens and by specimen type after overnight and 48-h incubations are summarized in Table 2.

TABLE 1.

Summary of screening culture results of MRSA Select and mannitol-salt agar with cefoxitin for 6,199 specimens

Medium	No. (%) of MRSA (n = 181) after incubation for:		No. (%) of specimens with the following culture result after incubation for the indicated time:
	18-24 h	48 h	No growth or no suspicious colonies		Growth of suspicious colonies (not MRSA), requiring further workup
			18-24 h	48 h	18-24 h	48 h
MRSA Select	169 (93)	177 (98)	6,002 (99.5)	5,552 (92)	28 (0.5)	470 (8)
Mannitol-salt agar with cefoxitin	125 (69)	163 (90)	5,627 (93)	4,676 (77)	447 (7)	1,360 (23)

TABLE 2.

Sensitivities and specificities of MRSA Select and mannitol-salt agar with cefoxitin for all specimens by specimen type after 18 to 24 h and 48 h of incubation

Specimen type (no. of specimens) [no. of MRSA]	Medium	Sensitivity (%) after incubation for:		Specificity (%) after incubation for:
		18-24 h	48 h	18-24 h	48 h
Overall (6,199) [181]	MRSA Select	93a	98b	99.5	92a
	Mannitol-salt agar with cefoxitin	69a	90b	92	78a
Catheter exit site samples (647) [14]	MRSA Select	100	100	99.8	97
	Mannitol-salt agar with cefoxitin	50	86	98	90
Nares swabs (2,483) [57]	MRSA Select	90	97	99.6	93
	Mannitol-salt agar with cefoxitin	70	95	97	89
Perianal/rectal swabs (2,312) [64]	MRSA Select	97	100b	99.3	90
	Mannitol-salt agar with cefoxitin	66	86b	85	58
Sputum samples (58) [4]	MRSA Select	100	100	100	94
	Mannitol-salt agar with cefoxitin	50	75	100	93
Skin/soft tissue samples (632) [42]	MRSA Select	91	95	99.5	95
	Mannitol-salt agar with cefoxitin	81	93	96	85

^aThe values are significantly different for the two media (P < 0.0001).

^bThe values are significantly different for the two media (P = 0.003).

The number of specimens with suspicious colonies requiring further investigation that turned out not to be MRSA differed substantially between the two media, with 1,360 of 6,199 (23%) for MSFOX compared to only 470 of 6,199 (8%) for MRSA Select (P < 0.0001) (Table 1). The vast majority of pink colonies that were not MRSA on MRSA Select appeared at 48 h of incubation and were infrequently found after overnight incubation.

Traditional methods for detection of MRSA from screening and surveillance specimens have used a variety of broth-based or selective and differential solid media. Mannitol-salt agar with oxacillin has been available for many years and is commonly used (7, 12). However, some strains of S. aureus may be inhibited by the salt component of the medium, and this may reduce its sensitivity (9). Mannitol-salt agar with cefoxitin appears to be superior for MRSA screening, probably because cephamycins are better inducers of PBP2a (15, 20, 21). Several other medium formulations have also been evaluated, with variable results and often delayed identification of MRSA (3, 7, 19, 22, 23).

The present study adds to available information describing the performance and utility of screening media incorporating chromogenic enzyme substrates for the isolation and detection of MRSA (4-6, 8, 11, 13, 21). We found that MRSA Select has improved sensitivity and specificity for the recovery of MRSA from screening specimens regardless of the body site sampled, with reduced time to detection. The majority (96%) of the MRSA isolates could be detected on MRSA Select after 18 to 24 h of incubation, suggesting that more prolonged incubation of these plates would provide only a small increase in yield. In a recent evaluation of MRSA Select for MRSA screening, Stoakes et al. found that only 3% of pink colonies at 24 h gave false-positive results (were not MRSA) (21). In the current study, 28 (14%) of the 197 specimens that yielded suspicious pink colonies after 18 to 24 h were not MRSA, although this represented only 0.5% of all the specimens cultured. Most of these organisms were subsequently found to be enterococci or diphtheroids, although there were a few members of the family Enterobacteriaceae and coagulase-negative staphylococci. On the basis of these results, we suggest that pink colonies growing on MRSA Select should be examined by Gram staining and latex agglutination, if appropriate.

Even though the cost of MRSA Select plates is greater than that of MSFOX or other nonchromogenic media, the total costs of materials, reagent, and processing of screening specimens for MRSA using MRSA Select in our laboratory during this evaluation were nearly equivalent ($9,758 Canadian) to the cost of using MSFOX ($9,106 Canadian). This is because there was a much greater number of suspicious colonies growing on MSFOX, which required subculture and additional testing (tube coagulase and latex agglutination tests) to exclude the presence of MRSA. In addition, work flow efficiency in the laboratory was improved, as less technologist time was required to examine and process the MRSA Select plates.

In conclusion, MRSA Select is a chromogenic culture medium with improved sensitivity and specificity for isolation of MRSA from surveillance and screening specimens compared to other currently available culture media. Almost all isolates can be detected after overnight incubation, with decreased laboratory technologist workload and associated cost savings. The reduced time to detection of MRSA could facilitate more rapid implementation of recommended infection control barrier precautions (1, 14).