Evaluation of Screening and Commercial Methods for Detection of Methicillin Resistance in Coagulase-Negative Staphylococci

Abstract

The National Committee for Clinical Laboratory Standards recommends 48 h of incubation by the oxacillin salt agar screen (OSAS) method for the detection of methicillin-resistant coagulase-negative staphylococci (CoNS). An earlier identification of methicillin resistance is desirable. The time to detection of the mecA gene by PCR was compared with the times to detection by OSAS, by the oxacillin disk diffusion (ODD) method, and with MicroScan Gram Positive Combo type 6 panels (MicroScan Inc. Sacramento, Calif.) and Vitek GPS-SA cards (bioMérieux Vitek Inc., Hazelwood, Mo.). The combination of the Vitek card and the ODD method detected 92 of 99 methicillin-resistant strains of CoNS at 24 h; however, 6 mecA-positive strains were phenotypically methicillin susceptible. We conclude that most methicillin-resistant CoNS can be detected and the results can be reported after overnight incubation by a combination of methods.

The recent data from National Nosocomial Infection Surveillance indicate the increasing importance of gram-positive organisms (13). The increase in methicillin resistance has accompanied the increase in the prevalence of staphylococci as nosocomial pathogens. In North America, the proportion of strains of Staphylococcus aureus resistant to methicillin rose from 2.4% in 1975 to 29% in 1991 (2), and more than 50% of isolates of coagulase-negative staphylococci (CoNS) are resistant to beta-lactams (13). The use of vancomycin has drastically increased (17) because it is the drug of choice for the treatment of methicillin-resistant staphylococci. The emergence of vancomycin-resistant Enterococcus sp. (VRE) has further compounded the problem (3, 9). To check the spread of VRE, the Hospital Infection Control Advisory Committee has recommended that the use of vancomycin be curtailed (5). Early identification of methicillin resistance in staphylococci could help to control the unnecessary use of vancomycin. Thus, a rapid and reliable method for the detection of methicillin resistance is needed. Methicillin-resistance S. aureus (MRSA) can be reliably detected after 24 h of incubation; however, for CoNS 48 h is often required before resistance can be ruled out (11).

We compared the time to mecA detection by PCR with the times to detection by the National Committee for Clinical Laboratory Standards (NCCLS) oxacillin salt agar screen (OSAS) method, by the oxacillin disk diffusion (ODD) method, and with MicroScan Gram Positive Combo type 6 panels (DADE MicroScan Inc. Sacramento, Calif.) and Vitek GPS-SA cards (bioMérieux Vitek Inc., Hazelwood, Mo.). The objective of the study was to determine which method or combination of methods could accurately and most rapidly predict methicillin resistance in CoNS.

One hundred eighty-one strains of CoNS isolated from consecutive blood cultures were studied. The strains were identified with MicroScan panels (6). Additional biochemical tests recommended in the Manual of Clinical Microbiology (7), susceptibilities to desferrioxamine (8), and cellular fatty acid profile (15) were used whenever necessary. The isolates were saved at −70°C and were subcultured twice before testing.

For the OSAS method, plates were spot inoculated with a direct colony suspension with a turbidity equivalent to that of a 0.5 McFarland standard according to the recommendations of NCCLS (12). The plates were incubated at 35°C and were read at 24 and 48 h. The ODD test was performed as described by NCCLS (12), and the zone diameters were determined at 24 and 48 h of incubation.

Oxacillin susceptibility testing with the MicroScan Gram Positive Combo type 6 panels and Vitek GPS-SA cards were performed according to the manufacturers’ instructions. MicroScan panels were read at 24 and 48 h with an AutoScan 4 instrument. A Vitek 120 Reader/Incubator was used to process the GPS-SA cards; the reader incubator read the cards within 24 h whenever the growth was sufficient.

For PCR, an overnight growth of organisms in 1.5 ml of brain heart infusion broth was pelleted and washed once with sterile saline. The pellet was suspended in 1 ml of distilled water, and an equal volume of Chelex 100 was added. The suspension was boiled for 7 min and centrifuged at 18,000 × g for 15 min, and 1 μl of the supernatant was used in 20 μl of amplification mixture. The primers used to detect the mecA and femA genes and the overlapping sequence of staphylococcal insertion element IS431 have been published by Vannuffel et al. (16). The last set of primers served as an internal control for the amplification process. The PCR mixture and amplification conditions were identical to those published by Vannuffel et al. (16). The bands were detected with UV light after electrophoresis of the amplified products through a 6% polyacrylamide gel containing ethidium bromide. The presence of a 309-bp band was considered a positive result. A positive PCR product was confirmed by digestion with the restriction enzyme AatII, which produced 219- and 85-bp fragments.

Of the 181 CoNS tested, 132 (72.9%) were S. epidermidis. Other species represented were S. hominis (23 strains), S. warneri (9 strains), S. capitis (7 strains), S. haemolyticus (6 strains), S. lugdenensis (3 strains), and S. simulans (1 strain). Eighty-two of the 181 strains were methicillin susceptible and 99 were methicillin resistant, as determined by the presence of the mecA gene. The OSAS method, the ODD test, and the MicroScan and Vitek systems accurately identified these mecA-negative strains as methicillin susceptible at both 24 and 48 h.

At 24 h, the Vitek system assigned the highest number of mecA-positive strains to the category of methicillin resistance and the OSAS method assigned the lowest number of strains to this category. After 48 h of incubation, the rates of detection of methicillin-resistant strains improved for all three methods for which results were evaluated at 24 and 48 h. The combination of Vitek cards plus the OSAS method was the most sensitive of all combinations at 24 h of incubation. The sensitivities of the four possible combinations were identical at 48 h. The numbers of isolates identified as methicillin resistant by each method individually and in combination after 24 and 48 h of incubation are indicated in Table 1. All non-PCR methods tested failed to classify six mecA-positive S. epidermidis strains as methicillin resistant.

TABLE 1.

Detection of methicillin-resistant CoNS

Methicillin resistance detection method	No. (%) of strains resistant at the following times:
	24 h	48 h
OSAS	41 (41.4)	77 (77.7)
ODD	66 (66.7)	89 (89.9)
MicroScan	78 (78.8)	86 (86.9)
Vitek	89 (89.9)	Not applicable
Vitek-OSAS	92 (92.9)	93 (93.9)
Vitek-ODD	90 (90.9)	93 (93.9)
MicroScan-OSAS	85 (85.9)	93 (93.9)
MicroScan-ODD	85 (85.9)	93 (93.9)

Several studies have demonstrated that PCR is a sensitive method for the detection of methicillin resistance in S. aureus and strains of CoNS (4, 10, 16). However, most laboratories are not in a position to perform the test. Phenotypically, the detection of methicillin resistance in staphylococci is complicated by the heterogeneous expression of the resistance. Although all cells possess the mecA gene, only 1 in 10⁵ to 10⁷ organisms expresses the resistance phenotypically (1). The OSAS method is a sensitive and cheap technique for the detection of methicillin-resistant S. aureus and CoNS. Commercially available systems are less reliable. In one study that used a mecA gene probe to identify MRSA, the MicroScan and Vitek systems failed to detect 17 and 36 of 254 MRSA strains, respectively (14). The reliability of the OSAS method for the detection of PCR-proven mecA-positive strains of CoNS has been demonstrated. However, for some strains 48 h of incubation is required (18).

None of the individual test methods or various combinations were able to detect all methicillin-resistant strains. However, six strains of S. epidermidis were consistently susceptible to oxacillin by all methods at 24 h and remained so after 48 h of incubation. We suspect that these strains were unable to express the mecA gene and are therefore phenotypically methicillin susceptible. Strains that are unable to express the mecA gene have been described previously (14). If the six strains in our study are excluded from our analysis, then the sensitivity of all combination methods was 100% at 48 h. The combination of Vitek and an OSAS plate was able to identify 92 of 93 phenotypically methicillin-resistant strains at 24 h.

We conclude that commercially available susceptibility test methods should be used in combination with an ODD method or an OSAS plate. Isolates of CoNS resistant by any method at 24 h can be reported as such. The accuracy of the combination of Vitek cards and the OSAS method was 98.9% (if the six strains mentioned above are excluded from analysis); therefore, oxacillin susceptibility results could be released at 24 h if this combination were used.