Comparison of rapid identification assays for Staphylococcus aureus

Evaluation of six agglutination tests for Staphylococcus aureus identification depending upon local prevalence of meticillin-resistant S. aureus (MRSA)

Most routine laboratory detection of Staphylococcus aureus isolates is based on rapid agglutination test systems. Failure of agglutination assays to identify meticillin-resistant S. aureus strains (MRSA) has been demonstrated. The aim of this study was to evaluate six commercially available agglutination tests for the detection of meticillin-sensitive S. aureus (MSSA) and mecA-positive MRSA strains. The Dry Spot Staphytect Plus test (Oxoid), the Pastorex Staph Plus test (Bio-Rad), the Slidex Staph-Kit and Slidex Staph Plus test (bioMérieux), the Staphaurex Plus test (Remel) and the Staphylase Test (Oxoid) were used. As determined by pulsed field gel electrophoresis, 52 distinct MRSA strains from five countries, 83 MSSA strains and 150 coagulase-negative staphylococci were included. Species identification and determination of susceptibility patterns were performed using colony morphology, Gram stain, catalase testing, tube coagulase testing, DNase testing, mannitol fermentation, susceptibility testing towards oxacillin by Etest, coagulase gene PCR, fibrinogen receptor gene PCR and PCR of the mecA gene. Sensitivity of the agglutination tests ranged from 82.7 to 100.0 % for MRSA strains and 92.8 to 100.0 % for MSSA strains, respectively. Specificity of the test systems ranged from 91.3 to 99.1 %. None of the six agglutination assays produced correct reactions for all staphylococci tested. Only the Dry Spot Staphytect Plus test correctly identified all 52 MRSA strains. For the other tests kits, sensitivity of MRSA detection was lower than for MSSA isolates. Depending upon the local MRSA prevalence and the parameter of interest (sensitivity or specificity), these test systems may be useful for routine diagnostic purposes.

Evaluation of new agglutination test for identification of oxacillin-susceptible and oxacillin-resistant Staphylococcus aureus

A new agglutination test (Monostaph +; Bionor, Skien, Norway) has been developed. This new agglutination test has been compared with two other agglutination tests for the identification of 128 isolates of Staphylococcus aureus and 82 coagulase-negative staphylococci. The sensitivities of both Monostaph + and Pastorex Staph-Plus were excellent (98.7 and 97.4%, respectively) in detection of oxacillin-resistant Staphylococcus aureus. The specificity was 96.4% (two Staphylococcus epidermidis isolates and one Staphylococcus hominis isolate were false positive).

A prolonged outbreak of exfoliative toxin A-producing Staphylococcus aureus in a newborn nursery

An outbreak of erythromycin-resistant, exfoliative toxin-producing Staphylococcus aureus infection in a neonatal unit is described. The organism was coagulase positive but staphyloslide negative, and this unusual phenotype facilitated early recognition of the organism in the routine laboratory. In the initial outbreak there were 77 probable or confirmed cases, with a peak attack rate of 66%. Increased infection control measures were put in place and attempts were made to identify a staff carrier. No carriers were found and the major outbreak subsided. Sporadic cases occurred over the following 10 months, until May 1992, when a colonized staffperson was discovered. She was treated and no further cases occurred. The causative organism was subjected to typing by phage, enterobacterial repetitive intergenic consensus sequence polymerase chain reaction, and pulsed-field gel electrophoresis with two separate enzymes. The phage typing and genomic tests confirmed the presence of the same clone in the unit for 9 months. The organism possessed genes encoding exfoliative toxin A as determined by polymerase chain reaction.

Evaluation of RapiDEC Staph for identification of Staphylococcus aureus, Staphylococcus epidermidis, and Staphylococcus saprophyticus

RapiDEC Staph is a test for presumptive identification of the principal human staphylococcal species, Staphylococcus aureus, S. epidermidis, and S. saprophyticus. The test includes control and test cupules for fluorogenic detection of coagulase and chromogenic substrates for alkaline phosphatase and beta-galactosidase. These tests identify S. aureus, S. epidermidis, and S. saprophyticus, respectively. Positive results with both chromogenic substrates provide a presumptive identification of S. xylosus or S. intermedius (S. xylosus-S. intermedius). Test cupules are inoculated with an organism suspension, and reactions are read after a 2-h incubation. RapiDEC-Staph was evaluated with 303 clinical and stock staphylococcal strains. Identifications were compared with those obtained by the tube coagulase test, a latex slide coagulase test (StaphAUREX), another commercial identification system (Staph-TRAC), and additional conventional tests. RapiDEC-Staph correctly identified 100% of 130 S. aureus strains, 70.3% of 74 S. epidermidis strains, and 81.3% of 32 S. saprophyticus strains. Four of five S. xylosus isolates were called S. xylosus-S. intermedius. Unidentified S. epidermidis and S. saprophyticus strains were called "Staphylococcus spp." Among the 62 other coagulase-negative staphylococci, 4 were misidentified as S. epidermidis and 7 were misidentified as S. saprophyticus. While the sensitivity and specificity of the fluorogenic coagulase test for S. aureus were 100%, failure to detect alkaline phosphatase activity in several S. epidermidis isolates resulted in fewer correct identifications by the RapiDEC-Staph test for this species.

Rapid detection of methicillin-resistant Staphylococcus aureus strains not identified by slide agglutination tests

Seventy-nine methicillin-resistant Staphylococcus aureus (MRSA) strains, isolated during 1980 to 1990, were classified as MRSA Aggl- (14 strains) and MRSA Aggl+ (65 strains) strains on the basis of test results in slide agglutination assays designed to detect fibrinogen-binding protein (clumping factor) and protein A on the staphylococcal surface. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis analysis revealed that lysostaphin digests of MRSA Aggl- strains contained a high-molecular-weight protein which was not detected in digests of MRSA Aggl+ strains. Immunization of rabbits with an MRSA Aggl- strain produced an antiserum which agglutinated all MRSA Aggl- strains and also 64 of 65 MRSA Aggl+ strains. Only 1 of 68 coagulase-negative staphylococci showed agglutination in this assay. The anti-MRSA Aggl- antiserum reacted mainly with a 230-kDa staphylococcal surface protein but also with a 175-kDa protein, probably formed by proteolysis of the former and a few slightly smaller proteins. These could not be immunologically detected in lysostaphin digests of MRSA Aggl+ strains. Purified antibodies reacting with the 230-kDa protein agglutinated all MRSA Aggl- strains, indicating that the protein is located on the surfaces of staphylococci. The results suggest a tentative role for the 230-kDa protein or its fragments as a novel target to develop more efficient rapid identification methods for S. aureus, including MRSA.

New latex reagent using monoclonal antibodies to capsular polysaccharide for reliable identification of both oxacillin-susceptible and oxacillin-resistant Staphylococcus aureus

A new latex agglutination test (Pastorex Staph-Plus, Sanofi Diagnostics Pasteur), consisting of a mixture of latex particles coated with fibrinogen and immunoglobulin G for the detection of clumping factor and protein A and latex particles sensitized with monoclonal antibodies directed to Staphylococcus aureus serotype 5 and 8 capsular polysaccharides, was compared with three commercially available rapid agglutination methods for the identification of 220 isolates of S. aureus (61 oxacillin resistant) and 128 isolates of coagulase-negative staphylococci. The sensitivity for identification of S. aureus was high with the Pastorex Staph-Plus test (98.6%) compared with those of the other tests, which ranged from 91.8 to 84.5%. Test sensitivities for the identification of oxacillin-resistant S. aureus were as follows: Pastorex Staph-Plus, 95.1%; Pastorex Staph, 73.8%; Staphyslide, 72.1%; and StaphAurex, 49.2%.

Novel immunoenzymatic assay for identification of coagulase- and protein A-negative Staphylococcus aureus strains

A purified monoclonal antibody (MAb) which specifically reacts with Staphylococcus aureus glucosaminidase was obtained. This MAb was utilized to develop an immunoenzymatic assay for the identification of S. aureus strains. The sensitivity of this assay, based on the simultaneous detection of S. aureus glucosaminidase and protein A, was evaluated by analyzing a total of 196 strains, 26 of which did not exhibit one or more of the following properties: protein A, clumping factor, and staphylocoagulase. All strains yielded positive results by the MAb-based immunoenzymatic test. The assay's ability to differentiate between S. aureus and other staphylococci was then analyzed by testing a total of 277 non-S. aureus strains that yielded negative results. Our data demonstrate that this immunoenzymatic assay can be used as a single S. aureus identification criterion, particularly useful for those strains negative for clumping factor, staphylocoagulase, or protein A.

Comparative evaluation of five agglutination techniques and a new miniaturized system for rapid identification of methicillin-resistant strains of Staphylococcus aureus

The speciation of methicillin-resistant Staphylococcus aureus (MRSA) poses a significant diagnostic problem when rapid identification methods such as slide agglutination tests, are used, because of the high proportion of false-negative reactions. 150 perfectly identified MRSA strains were tested on 5 commonly used agglutination reagents ("Bacto staph latex test", "Monostaph", "Pastorex staph", "Staphaurex", and "Staphyslide test") in comparison with a new micromethod ("RAPIDEC staph") which detects a type of staphylocoagulase within 2 hours by a fluorescence test. The "RAPIDEC staph" reagent enabled identification of all the MRSA while the agglutination tests gave poorer results: "Monostaph" correctly identified 64.6% of strains, "Staphyslide", 59.3%, "Bacto staph latex test", 44.6%, "Pastorex staph", 38.6% and "Staphaurex", 28.6%. These results show that agglutination slide tests are not reliable enough for the identification of MRSA which are more and more encountered in hospital wards. The authors recommend not to use slide agglutination methods. They suggest the tube test for coagulase which is the reference technique, although it is time-consuming and not well standardized. The results of this evaluation encourage the use of the "RAPIDEC staph" reagent since it is an easy-to-use, reliable technique for the rapid identification of Staphylococcus aureus.

Coagulase testing compared with commercial kits for routinely identifying Staphylococcus aureus

Five commercial Staphylococcus aureus identification kits--Staphaurex (Wellcome), Staphylase (Oxoid), Staphyslide (bioMèrieux), Biostaph (Medlabs) and Bacto Latex (Difco)--were evaluated for the routine identification of S aureus from primary plates in the routine microbiology laboratory. Comparison was made with two methods of tube coagulase testing and five slide methods for detecting clumping factor (slide coagulase testing). Performances were assessed for two groups of organisms, staphylococcal species alone and a combined staphylococcal and non-staphylococcal species group. The effects of growth on selective media and storage of isolates at room temperature and 4 degrees C were investigated. Selective media cannot be recommended, nor can storage of isolates before testing. Ranked according to efficiency value with the combined staphylococcal and non-staphylococcal species group, the kits and coagulase methods performed as follows (the figures in parentheses are the efficiency values for the staphylococcal group alone): tube coagulase reference method 100% (100%), tube coagulase SJH method 99% (99%), Staphaurex 94% (97%), Staphylase 93% (96%), slide coagulase method No 4 93% (94%), slide coagulase method No 5 93% (93%), Bacto Latex 92% (95%), Staphyslide 92% (95%), and Biostaph 87% (91%). It is concluded that a commercial S aureus identification kit should not replace tube coagulase testing for the routine identification of the organism from primary plates and that, even the kits with the best performances, have little advantage over a good slide coagulase test method.

Staphylococcus aureus strains which are not identified by rapid agglutination methods are of capsular serotype 5

A total of 183 recent Staphylococcus aureus clinical isolates were tested with three commercially available rapid agglutination methods. The capsular polysaccharide type and resistance to oxacillin of these isolates were also determined. Seven isolates were not identified correctly by agglutination methods. All isolates not identified by the rapid methods were of capsular serotype 5, and of these isolates, six were resistant to oxacillin. The results suggest that these agglutination kits can be improved by the use of antibodies reactive with S. aureus capsular polysaccharide.

Efficacies of rapid agglutination tests for identification of methicillin-resistant staphylococcal strains as Staphylococcus aureus

Four commercially available rapid agglutination tests for the identification of Staphylococcus aureus were compared with the tube coagulase test for the identification of 300 methicillin-resistant isolates of staphylococci. Isolates tested included 207 methicillin-resistant S. aureus and 93 coagulase-negative staphylococci, collected from five medical centers. Strain variability was documented by phage typing and antimicrobial susceptibility patterns. Results of rapid identification tests ranged between 82 and 86% sensitivity, significantly poorer than the 98% sensitivity which the tube coagulase test provided.

Evaluation of several commercial biochemical and immunologic methods for rapid identification of gram-positive cocci directly from blood cultures

To develop laboratory methods for the rapid identification of gram-positive cocci from blood cultures, several commercial immunological and biochemical tests for identifying staphylococci and streptococci from two different blood culture systems (Thiol/TSB, Difco Laboratories; BACTEC, Johnston Laboratories, Inc.) were evaluated and compared with conventional identification methods. A total of 44 cultures contained Staphylococcus aureus as determined by conventional methods. Commercial immunological methods (six tested) ranged in overall sensitivity from 38.6 (Staphyloslide; BBL Microbiology Systems) to 77.3% (Staphaurex; Wellcome Diagnostics). All methods tested had 100% specificity. A total of 30 isolates of group D streptococci were tested with immunological and biochemical identification systems. The overall sensitivity ranged from 14.2% (Phadebact Group D; Pharmacia Diagnostics) to 100% with Streptex (Wellcome; immunological) and Identicult A-E (Scott Laboratories, Inc.; biochemical). The results of this study suggest that some reagents can be used to provide rapid identification of gram-positive cocci from blood cultures 24 h earlier than standard methods.

Problems with rapid agglutination methods for identification of Staphylococcus aureus when Staphylococcus saprophyticus is being tested

Six rapid agglutination tests for identification of Staphylococcus aureus were evaluated by using 62 strains of S. aureus, 63 strains of S. saprophyticus, and 67 strains of other coagulase-negative staphylococci. S. saprophyticus was responsible for 19 of 26 false-positive results and 20 uninterpretable reactions. Thus, urinary staphylococcal isolates that are positive by rapid agglutination tests may require other confirmatory tests for the identification of possible S. saprophyticus.

Comparison of seven coagulase tests for identification of Staphylococcus aureus

Four coagglutination tests for the identification of Staphylococcus aureus were compared with the ordinary slide and tube coagulation tests using two groups of staphylococcal strains (one isolated in the authors' laboratory and the other identified by a reference laboratory). The correct identification rate in the two groups was respectively for slide test 96.9% and 92.7%, tube coagulation (citrate plasma) 99.0% and 94.8%, tube coagulation (EDTA plasma) 99.0% and 91.7%, API Staphase III 100% and 91.7%, Staph Rapid 97.9% and 91.3%, Staphyslide 99.0% and 94.6%, Staphaurex 96.9% and 92.7%. The sensitivity of Staph Rapid, Staphyslide and Staphaurex was slightly higher than that of the other tests. These three tests and the slide test were considerably more rapid as regards identification than the other tests.

Comparison of two agglutination tests for differentiation between coagulase positive and coagulase negative staphylococci

Two agglutination tests, Monostaph (Bionor A/S, N-3700 Skien, Norway) and Staphaurex (Wellcome), for the identification of Staphylococcus aureus have been evaluated. Using the coagulase test as reference method both tests were equally reliable and in complete aggrement with the coagulase tube test when tested on 216 strains of Micrococcaceae.

Performance of four slide agglutination methods for identification of Staphylococcus aureus when testing methicillin-resistant staphylococci

Four commercially available slide agglutination systems for identifying Staphylococcus aureus were compared with the conventional slide (clumping factor) and tube coagulase tests. The systems evaluated included Bacto Staph Latex (Difco Laboratories, Detroit, Mich.), Staphyloslide (BBL Microbiology Systems, Cockeysville, Md.), Mini ID Accu-Staph (Carr-Scarborough Microbiologicals, Inc., Decatur, Ga.), and Staphaurex (Wellcome Diagnostics, Research Triangle Park, N.C.). A total of 338 clinical isolates, including methicillin-resistant S. aureus (n = 149), methicillin-susceptible S. aureus (n = 78), methicillin-resistant, coagulase-negative staphylococci (n = 45), and methicillin-susceptible, coagulase-negative staphylococci (n = 66), were tested by each method. The slide test for clumping factor, the 4-h tube coagulase test, Bacto Staph Latex, Staphyloslide, Mini ID Accu-Staph, and Staphaurex detected 212 (93.4%), 218 (96%), 223 (98.2%), 223 (98.2%), 221 (97.4%), and 224 (98.7%) of the S. aureus (44% methicillin-resistant) isolates, respectively. There were no false-positive results with any of the methods when the 111 strains of coagulase-negative staphylococci were tested. The results of this evaluation suggest that the four slide identification methods tested can provide rapid and accurate identification of methicillin-resistant S. aureus strains.

Reliability of latex agglutination tests for identification of Staphylococcus aureus resistant to oxacillin

Commercial latex agglutination tests (LATs) for the simultaneous detection of clumping factor and protein A are gaining increased acceptance as a means of identifying Staphylococcus aureus. We evaluated two LATs (Accu-Staph; Carr-Scarborough, Decatur, Ga.; Staphaurex; Wellcome, Dartford, England) with particular emphasis on their ability to correctly identify oxacillin-resistant S. aureus. We tested 59 oxacillin-resistant S. aureus, 136 oxacillin-susceptible S. aureus, and 92 coagulase-negative staphylococcal strains with the two LATs and with thermonuclease, slide clumping factor, tube coagulase, and protein A hemagglutination tests. Clumping factor and protein A were present in 96.9 and 82.1% of our S. aureus strains, respectively. Accu-Staph correctly identified 92.8% and Staphaurex correctly identified 91.3% of S. aureus strains. No significant difference in LAT positivity rates, presence of clumping factor, or presence of protein A was found between oxacillin-resistant and -susceptible S. aureus. Overall, there were 31 false-negative LATs for 20 S. aureus strains, 14 with Accu-Staph and 17 with Staphaurex. Ninety-five percent of these strains possessed either clumping factor or protein A or both when these factors were determined independently. There were five false-positive LATs for four strains of coagulase-negative staphylococci (three Staphylococcus epidermidis and one Staphylococcus warneri), four with Accu-Staph and one with Staphaurex. Clumping factor was present in one S. warneri strain. Thus, the specificities of Accu-Staph, Staphaurex, and the clumping factor test were 95.6, 98.9, and 98.9%, respectively. Our results indicated that LATs identify oxacillin-resistant and -susceptible S. aureus equally well; however, they offer no greater sensitivity or specificity than the clumping factor test for identification of S. aureus.

Failure of rapid agglutination methods to detect oxacillin-resistant Staphylococcus aureus

Although a latex agglutination test (StaphAurex) and a hemagglutination test (Staphyloslide) correctly identified all strains of Staphylococcus aureus that were susceptible or had intermediate susceptibility to oxacillin, 17 of 73 (23%) and 18 of 73 (25%) strains of oxacillin-resistant S. aureus were not identified by StaphAurex and Staphyloslide, respectively. All strains not detected were resistant to trimethoprim-sulfamethoxazole and rifampin.

Comparative evaluation of identification systems for testing methicillin-resistant strains of Staphylococcus aureus

Several commercial systems are available to distinguish between Staphylococcus aureus and the coagulase-negative species of the Micrococcaceae family. Four latex agglutination systems (Accu-Staph, SeroSTAT, Staphaurex, and Staphylatex) and two hemagglutination systems (Hemastaph and Staphyloslide) were compared for their performance in the rapid identification of 232 isolates of staphylococci, including 114 of methicillin-resistant S. aureus. Accu-Staph, Staphaurex, and Staphyloslide correctly identified 100% of the methicillin-resistant S. aureus isolates; Hemastaph and Staphylatex, 99.1%; and SeroSTAT, 94.7%. Most reactions were easy to interpret, although 15% of the SeroSTAT reactions were weak. Autoagglutination occurred only with isolates of coagulase-negative staphylococci. False-positive reactions were rare and occurred only with systems which did not detect autoagglutination. Five of these six systems appear to be adequate for the rapid identification of S. aureus, including methicillin-resistant isolates.

Evaluation of rapid coagulase methods for the identification of Staphylococcus aureus

Four rapid latex agglutination assays, StaphAurex (Wellcome Diagnostics, Research Triangle Park, N.C.), Bacto Staph (Difco Laboratories, Detroit, Mich.), SeroSTAT (Scott Laboratories, Inc., Fiskeville, R.I.), Veri-Staph (Zeus Technologies, Raritan, N.J.), and two hemagglutination tests, Staphyloslide (BBL Microbiology Systems, Cockeysville, Md.) and Hemastaph (Remel, Lenexa, Kans.), were compared with the conventional slide coagulase, tube coagulase (TC), and thermonuclease (TNase) tests for the identification of Staphylococcus aureus. A total of 118 clinical isolates of S. aureus (52 methicillin resistant), 50 S. epidermidis, 5 S. capitis, 2 S. hominis, 3 S. simulans, 6 S. saprophyticus, and 2 S. warneri were tested. The slide coagulase, TC and TNase tests detected 115 (97.5%), 117 (99.2%), and 118 (100%) of the S. aureus isolates, respectively. All showed 100% specificity. The StaphAurex, Veri-Staph, Staphyloslide, Hemastaph, SeroSTAT, and Bacto Staph assays correctly identified 117 (99.2%), 117 (99.2%), 116 (98.3%), 110 (93.2%), 108 (91.5%), and 107 (90.7%) of the S. aureus isolates, respectively. For methicillin-resistant S. aureus isolates, StaphAurex, Veri-Staph, Staphyloslide, Hemastaph, SeroSTAT, and Bacto Staph showed 1 (2%), 1 (2%), 2 (4%), 7 (13.5%), 7 (13.5%), and 8 (15.4%) false-negative results, respectively. All the commercial agglutination assays demonstrated false-positive results with strains of S. capitis, S. saprophyticus and S. warneri. The overall accuracy of the commercial agglutination assays compared with TC and TNase ranged from 90.7 to 99.2%. We recommend that negative reactions with the rapid commercial test kits for methicillin-resistant Staphylococcus isolates be confirmed with the TC or TNase test.

Comparison of a yellow latex reagent with other agglutination methods for the identification of Staphylococcus aureus

A new commercial yellow latex agglutination reagent (Bacto-Staph) was compared with the slide and tube coagulase tests and three other commercial reagents for the identification of 283 Staphylococcus aureus and 54 non-S. aureus staphylococcal strains. Test sensitivities for the identification of S. aureus were as follows: tube coagulase, 99.6%; slide coagulase, 98.6%; Bacto-Staph, 99.6%; Staphylatex, 98.6%; Sero STAT Staph, 98.2%; and Staphyloslide, 97.5%. No false-positive reactions were observed with any of the commercial reagents.

Manual approaches to rapid microbiology results

Rapid generation of microbiology results does not require sophisticated automated instruments. Many common laboratory procedures, such as microscopic examination of clinical specimens, use of manual screening tests, macroscopic examination of isolated colonies, and performance of rapid biochemical tests, can provide the physician with clinically useful information. These tests are neither expensive nor time-consuming and should be used routinely to generate rapid test results.

Evaluation of various rapid agglutination methods for the identification of Staphylococcus aureus

A latex agglutination test (SeroSTAT Staph; Scott Laboratories, Fiskeville, R.I.) and two hemagglutination tests (Staphyloslide; BBL Microbiology Systems, Cockeysville, Md.; and Hemastaph; Remel, Lenexa, Kans.) were compared with the slide coagulase (SC) and tube coagulase (TC) tests at room temperature (22 to 25 degrees C) and at 37 degrees C for the rapid identification of Staphylococcus aureus. A total of 380 clinical strains of staphylococci were tested. The TC test performed at room temperature yielded the largest number of TC-positive results (n = 239), and based on this observation 239 organisms were classified as S. aureus and 141 were classified as non-S. aureus. The SC, TC (37 degrees C), SeroSTAT Staph, Staphyloslide, and Hemastaph tests correctly identified 210 (87.9%), 221 (92.5%), 238 (99.6%), 239 (100%), and 236 (98.7%) of the S. aureus isolates, respectively. Of the S. aureus isolates that were TC positive at room temperature 68% required 24 h of incubation before coagulase production was detected. There was one false-negative SeroSTAT Staph result and one false-negative Hemastaph result. The Staphyloslide test yielded two noninterpretable results (both organisms were later confirmed as non-S. aureus), whereas there were six noninterpretable results recorded with the Hemastaph test (four organisms were classified as non-S. aureus, and two were classified as S. aureus). The SeroSTAT Staph, Staphyloslide, and Hemastaph tests were all more sensitive than the conventional SC and TC (37 degrees C) tests and were considerably more rapid than the TC test at either temperature.

Variation in the abilities of automated, commercial, and reference methods to detect methicillin-resistant (heteroresistant) Staphylococcus aureus

The abilities of commercial MIC, automated, and reference methods for in vitro detection of methicillin-resistant Staphylococcus aureus were determined on 49 strains from eight hospitals. Micro-Media, MicroScan, Sensititre, Sceptor, API Uniscept KB, Abbott MS-2, Vitek AMS, Autobac MTS, NCCLS disk diffusion, and broth microdilution antimicrobial susceptibility testing procedures were evaluated. All testing was performed by using manufacturers' or reference procedures, and results were determined at no later than 24 h of incubation at 35 degrees C. With NCCLS disk diffusion, all strains were resistant to oxacillin (1 microgram), 47 (96%) were resistant to methicillin (4 micrograms), and 48 (98%) were resistant to nafcillin (1 microgram). The percentages of strains resistant to methicillin (greater than 8 micrograms/ml) were 98% with API Uniscept KB, 86% with Sceptor, MicroScan, and Autobac MTS, 84% with Sensititre, 71% with Micro-Media, and 70% with NCCLS MIC. Abbott MS-2 detected 86% of strains resistant to methicillin (greater than 5 micrograms/ml). With oxacillin (greater than 2 micrograms/ml), 90% were detected with Vitek AMS and 70% were detected with NCCLS MIC. With nafcillin (greater than 2 micrograms), 82% were resistant with Micro-Media, 57% were resistant by NCCLS MIC, and 50% (three of six) were resistant by MicroScan. Two strains from one hospital and one strain from another gave susceptible results with all automated and commercial methods. All strains from three centers were detected by all methods. Variability also occurred among the systems with cephalothin, clindamycin, gentamicin, chloramphenicol, and trimethoprim sulfamethoxazole.