Oxacillin resistance screening agar base for detection of methicillin-resistant Staphylococcus aureus.

In a recent letter, Simor et al. reported results obtained by using oxacillin resistance screening agar base (ORSAB; Oxoid Limited, Basingstoke, England) for the detection of methicillin-resistant Staphylococcus aureus (MRSA) from clinical specimens (A. E. Simor, J. Goodfellow, L. Louie, and M. Louie, Letter, J. Clin. Microbiol. 39:3422, 2001). ORSAB is a modification of a mannitol-salt agar supplemented with oxacillin, in which mannitol-positive isolates turn blue due to an acid-dependent chromogenic component (aniline blue). When specimens from persons thought to be at high risk for MRSA colonization were screened with this agar base, 102 of 104 MRSA-positive clinical specimens (98%) were correctly identified. In total, 138 clinical specimens yielded blue, mannitol-fermenting colony types; therefore, the positive predictive value of ORSAB-positive specimens for MRSA has been 74%. During a period of 8 months, we tested ca. 4,200 staphylococcal isolates, including ca. 2,200 isolates of S. aureus from unselected clinical specimens, on ORSAB. A total of 131 strains of MRSA were included; all but one (99%) of the MRSA isolates were detected with ORSAB, the one exception being negative for mannitol fermentation, as assessed with the ATB ID 32 Staph kit (bioMérieux, Marcy-l’Etoile, France). In total, 266 isolates were blue on ORSAB. The results of identification of these isolates (using the ATB ID 32 Staph kit) are shown in Table 1; data obtained from the three different predominant specimen types are presented. The overall predictive value of ORSAB-positive isolates was 48.9% (130 of 266 isolates were positive). Predictive values with respect to specimen types were as follows: for specimens from nares and throat, 65.2% (60 of 92 isolates were positive); for skin and soft tissue specimens, 47.8% (32 of 67 isolates were positive); and for urine specimens, 33.3% (23 of 69 isolates were positive). While the sensitivity of ORSAB observed by us was as high as that reported by Simor et al., predictive values in our study were much lower. This may be explained by two facts: (i) our study was not restricted to persons at high risk for MRSA colonization but included all submitted clinical specimens and (ii) there were only a few urinary tract specimens ( 1% of all specimens) included in the study of Simor et al.; in our study, where these specimens were about 20% of all specimens, the predictive value for these specimens was much lower than that for specimens from nares and throat. Simor et al. state that, in contrast to specimens planted on ORSAB, specimens planted on mannitol salt agar supplemented with 2.0 g of oxacillin per ml “required additional work-up of mannitol-fermenting colonies that were subsequently determined not to be MRSA.” Taking into account that in our study the second most predominant ORSABpositive species was Staphylococcus haemolyticus (133 of 266