Influence of variations in test methods on susceptibility of Haemophilus influenzae to ampicillin, azithromycin, clarithromycin, and telithromycin

The National Committee for Clinical Laboratory Standards standard broth microdilution method for testing the susceptibility of Haemophilus influenzae to ampicillin, azithromycin, clarithromycin, and telithromycin was evaluated by altering one variable at a time. Variables that were tested included age of colony for inoculum preparation, inoculum density, test medium, incubation atmosphere, and incubation time. For the macrolide, azalide, and ketolide agents, incubation in 5 to 7% CO(2) most significantly affected the MICs, producing nearly twofold increases for clarithromycin and telithromycin and a greater than threefold increase for azithromycin. For ampicillin, a 10-fold increase in inoculum density increased the geometric mean MICs for beta-lactamase-negative strains from 1. 50 to 2.45 microg/ml. In addition, 206 H. influenzae strains were tested for their susceptibilities to the same drugs by the broth microdilution tests in two media, as well as by agar dilution tests, disk diffusion tests, and Etests, on six different agar media. The three standard methods with Haemophilus test medium (HTM) compared favorably with each other except for a high minor discrepancy rate (27%) by the disk diffusion test with ampicillin and clarithromycin. Agar dilution test MICs on the five comparative media were generally higher than those on HTM agar but were only rarely more than one twofold concentration higher. Etest MICs of azithromycin and telithromycin were more than twofold higher than agar dilution and broth microdilution MICs on HTM; ampicillin Etest MICs were nearly twofold lower. The use of media other than HTM agar appears to have a minimal effect on susceptibility test results for the ketolide, azalide, or macrolide drugs that we tested against H. influenzae.     

Improved medium for antimicrobial susceptibility testing of Haemophilus influenzae.

The need for complex growth media has complicated routine susceptibility testing of Haemophilus influenzae because of antagonism of certain antimicrobial agents by the medium or because of difficulties in interpretation of growth endpoints. Haemophilus test medium (HTM) is a simple, transparent medium for broth- or agar-based tests with H. influenzae. HTM incorporates Mueller-Hinton medium with additions of 15 micrograms of hematin per ml, 15 micrograms of NAD per ml, and 5 mg of yeast extract per ml as growth-promoting additives. Agar or broth microdilution MICs of 10 antimicrobial agents for a collection of 179 H. influenzae isolates determined by using HTM compared favorably with MICs determined by the conventional agar or broth dilution methods recommended by the National Committee for Clinical Laboratory Standards. Disk diffusion tests performed with HTM allowed accurate categorization of susceptible and resistant strains and were easier to interpret than tests performed with Mueller-Hinton chocolate agar. A particular advantage of HTM was the reliability of broth- or agar-based test results with trimethoprim-sulfamethoxazole. The results of the study suggest modification of current National Committee for Clinical Laboratory Standards MIC-interpretive criteria for H. influenzae with amoxicillin-clavulanate, chloramphenicol, and trimethoprim-sulfamethoxazole. Error rate-bounded analysis of MICs and disk diffusion zone sizes also suggest modified zone-interpretive criteria for ampicillin, amoxicillin-clavulanate, chloramphenicol, and tetracycline with HTM or conventional media. Interpretive zone sizes are newly proposed for cefaclor and rifampin disk diffusion tests.     

Effects of various test media on the activities of 21 antimicrobial agents against Haemophilus influenzae

As considerable variation in the antimicrobial susceptibility of Haemophilus influenzae has been reported, the effects of various test media on the susceptibility of H. influenzae were studied. MICs were determined by three laboratories for 21 antimicrobial agents against a panel of 100 selected isolates. Testing was performed using a reference NCCLS frozen broth microdilution method with Haemophilus test medium (HTM) broth and dried commercial MIC trays rehydrated with the following media: in-house and commercially prepared HTM broth, Mueller-Hinton broth with 2% lysed horse blood and NAD, IsoSensitest broth with 2% lysed horse blood and NAD, and IsoSensitest broth-based HTM. Overall, all results were very reproducible, with the MIC at which 50% of the isolates tested are inhibited (MIC(50)), MIC(90), and geometric mean MIC being within one doubling dilution by all six methods and at all three testing centers for 15 of the 21 agents tested. Interlaboratory differences were more marked than intralaboratory differences or differences among media. Cefprozil, cefaclor, and trimethoprim-sulfamethoxazole results differed the most, while results for ampicillin, amoxicillin-clavulanic acid, cefdinir, cefixime, ceftriaxone, and clarithromycin were the most reproducible. However, these variations in results caused considerable differences in susceptibility rates for agents for which NCCLS susceptible breakpoints were close to the geometric mean MIC, particularly for cefaclor and cefprozil. This was much less of a problem when pharmacokinetic-pharmacodynamic breakpoints were used. Reproducible susceptibility results were obtained for a wide range of agents against H. influenzae in three laboratories using a variety of media that support the growth of this fastidious species.     

Activity of Gatifloxacin against Haemophilus influenzae and Moraxella catarrhalis, Including Susceptibility Test Development, E-Test Comparisons, and Quality Control Guidelines for H. influenzae

In vitro antimicrobial activity and susceptibility testing interpretation criteria and quality control were studied for gatifloxacin, a new 8-methoxy fluoroquinolone, tested against Haemophilus influenzae. Moraxella catarrhalis (600 strains) and H. influenzae (1,400 strains) from the SENTRY Antimicrobial Surveillance Program in North America (Canada and the United States) were also tested against gatifloxacin and 12 other antimicrobial agents. Gatifloxacin (MIC at which 90% of the isolates are inhibited [MIC90], /=18 mm) was also suggested for H. influenzae testing. No interpretive errors were observed. Quality control guidelines for H. influenzae ATCC 49247 were determined by using the NCCLS M23-T3 (1998) study design. The results from the nine-laboratory protocol suggested the following control ranges: for broth microdilution tests, 0.004 to 0.03 microg/ml; for disk diffusion testing, 33 to 41 mm. Gatifloxacin appears to be a potent anti-Haemophilus fluoroquinolone compound with in vitro testing interpretive criteria that will produce accurate results (disk diffusion, broth microdilution, and E-test).     

Antimicrobial activity of cefmetazole (CS-1170) and recommendations for susceptibility testing by disk diffusion, dilution, and anaerobic methods

Cefmetazole, formerly CS-1170, was found to have antimicrobial activity slightly superior to that of cefoxitin but a clinically usable antimicrobial spectrum that should be considered identical to that of cefoxitin. Disk diffusion and dilution test methods with cefmetazole correlated highly (r, greater than or equal to 0.95) with cefoxitin results. The recommended 30-micrograms cefmetazole disk interpretive breakpoints for susceptibility and resistance were greater than or equal to 18 mm (MIC, less than or equal to 8.0 micrograms/ml) and less than or equal to 14 mm (MIC, greater than or equal to 32 micrograms/ml), respectively. Cefmetazole and cefoxitin should be considered to be in the same antimicrobial spectrum class, requiring separate testing for other cephalosporins such as cephalothin, cefamandole, cefuroxime, and cefotetan. Recommended interpretive criteria performed well for fastidious organisms (Haemophilus influenzae, Neisseria meningitidis, and Branhamella catarrhalis) and for broth microdilution tests with anaerobes. Cefmetazole and cefoxitin broth disk elution tests for anaerobic bacteria produced higher rates of false susceptibility results.     

Evaluation of in vitro methods for testing ceftriaxone against anaerobic bacteria, including quality control guidelines.

Tests with 100 anaerobic bacterial isolates demonstrated comparability between ceftriaxone MICs obtained with the reference agar dilution procedure and those obtained with a broth microdilution susceptibility testing procedure. The aerobically incubated thioglycolate disk elution test was also evaluated. Six 30-micrograms ceftriaxone disks in 5 ml of thioglycolate separated strains for which MICs were less than or equal to 32 micrograms/ml from those for which MICs were greater than or equal to 64 micrograms/ml (6% overall discrepancies). Quality control limits for ceftriaxone agar dilution tests were determined to be as follows: Bacteroides fragilis ATCC 25285, 32 to 128 micrograms/ml; and Bacteroides thetaiotaomicron ATCC 29741, 64 to 256 micrograms/ml.     

Susceptibility of Haemophilus influenzae to piperacillin-tazobactam combinations: interpretive criteria and quality control limits for standardized tests.

In vitro studies evaluated methods for testing the susceptibility of Haemophilus influenzae to piperacillin-tazobactam combinations. Ampicillin-resistant beta-lactamase-nonproducing strains of H. influenzae may be presumed to be relatively resistant to combinations of piperacillin-tazobactam, even though they frequently appear to be susceptible by disk diffusion methods. Other ampicillin-resistant or -susceptible strains were predictably susceptible; i.e., 130 such strains gave zones of inhibition > or = 26 mm in diameter, and MICs for these strains were < or = 0.125/4.0 micrograms/ml (< or = 1.0/0.12 micrograms/ml when an 8:1 ratio was tested). A resistant category has yet to be defined. For quality control purposes, H. influenzae ATCC 49247 should give zones of inhibition 32 to 38 mm in diameter, and broth microdilution MICs should be 0.12/4.0 to 0.5/4.0 micrograms/ml.     

Quantitative antimicrobial susceptibility testing of Haemophilus influenzae and Streptococcus pneumoniae by using the E-test.

The E-test (PDM Epsilometer; AB Biodisk, Solna, Sweden) is an antimicrobial agent gradient-coated plastic test strip which allows MIC determinations on agar media. The test is performed in a manner similar to the agar disk diffusion procedure. A collection of Haemophilus influenzae and Streptococcus pneumoniae strains possessing various resistance mechanisms was used to evaluate the E-test method. H. influenzae strains were tested with both Haemophilus test medium (HTM) and PDM ASM II chocolate agar, while the S. pneumoniae strains were tested on Mueller-Hinton sheep blood agar. E-test MICs for a total of 10 antimicrobial agents were compared with broth microdilution MICs determined according to National Committee for Clinical Laboratory Standards methods. In general, E-test MICs for both species were quickly and easily interpreted and agreed within one log2 MIC increment in 89.8% of tests with H. influenzae and in 80.4% of pneumococcal tests. The majority of disagreements between the E-test and conventional MICs occurred with trimethoprim-sulfamethoxazole because of trailing and diffuse E-test MIC endpoints with both species. Ampicillin MICs for beta-lactamase-producing H. influenzae determined by the E-test differed at times from those determined by conventional testing because of the vagaries of interpreting colonies growing within the E-test inhibition ellipses. E-test penicillin MICs for pneumococci tended to be 1 to 2 log2 dilutions lower than those determined by using Mueller-Hinton broth supplemented with lysed horse blood. Nevertheless, strains of both species with documented resistance to the study drugs were detected by E-tests, i.e., 0.7% of the tests had very major errors with H. influenzae and 0.8% had very major errors with S. pneumoniae. Thus, the E-test represents a potential alternative method for antimicrobial susceptibility testing of these two fastidious bacterial species.     

Problems with the disk diffusion test for detection of vancomycin resistance in enterococci.

A total of 53 strains of enterococci, including recently isolated strains with high-level resistance to vancomycin, were tested for vancomycin susceptibility by broth microdilution and disk diffusion using Mueller-Hinton media with and without supplementation with 5% blood. By using currently published parameters of the National Committee for Clinical Laboratory Standards for the disk diffusion test, we found that strains for which MICs were 8 to 32 micrograms/ml were incorrectly placed in the susceptible or intermediate category, which caused both very major (1.9%) and minor (11.5%) errors. When we used newer, recently proposed breakpoints for vancomycin, we found 13.5% minor errors but no very major errors. Changing disk diffusion breakpoints to less than or equal to 14 mm for resistant [corrected] and greater than or equal to 15 mm for susceptible [corrected] would eliminate the problem for the strains with MICs of 32 micrograms/ml but not for those with MICs of 8 micrograms/ml. For those strains, it is necessary to perform an MIC test to differentiate them from strains with MICs of less than or equal to 4 micrograms/ml.     

Comparison of agar dilution, tube dilution, and broth microdilution susceptibility tests for determination of ramoplanin MICs.

Standard broth microdilution (with and without bovine serum albumin [BSA] supplementation), tube dilution, and agar dilution susceptibility tests were compared for determining ramoplanin MICs. With a data base of 246 clinical isolates of gram-positive bacteria from 33 U.S. sites, it was shown that (i) agar and tube dilution susceptibility tests gave essentially the same results (93.9% of the test results were within 1 doubling dilution of equivalence), (ii) broth microdilution susceptibility tests gave results up to 5 doubling dilutions higher than agar or tube assays, and (iii) this data skewing could be reversed by BSA supplementation (final concentration, 0.02%) of the broth microdilution test medium.     

Interpretive criteria and quality control parameters for testing of susceptibilities of Haemophilus influenzae and Streptococcus pneumoniae to trimethoprim and trimethoprim-sulfamethoxazole. The Antimicrobial Susceptibility Testing OC Group.

Two hundred twenty-eight strains of Haemophilus influenzae and 234 strains of Streptococcus pneumoniae were tested by broth microdilution and disk diffusion methods for susceptibility to trimethoprim (TMP) and TMP-sulfamethoxazole (SMX) to evaluate proposed criteria. Data are presented to support the proposed TMP MIC breakpoints of < or = 2.0 micrograms/ml for susceptibility and > or = 4.0 micrograms/ml for resistance for both species and TMP-SMX MIC breakpoints of < or = 2.0-38 micrograms/ml for susceptibility and > or = 4.0-76 micrograms/ml for resistance. Corresponding zone diameter breakpoints for H. influenzae for both drugs are proposed: < or = 10 mm = resistant; > or = 16 mm = susceptible. A 10-laboratory study documented reproducibility of such tests with standard control strains. The following control limits are proposed for tests of H. influenzae ATCC 49247 against TMP; MIC, 0.12 to 0.5 microgram/ml; zone diameter, 27 to 33 mm. The current limits for TMP-SMX were confirmed. For tests of S. pneumoniae ATCC 49619, MICs of TMP were 1.0 to 4.0 micrograms/ml and the current TMP-SMX MIC range was confirmed. Disk susceptibility tests of either drug against pneumococci were not reproducible, and consequently neither quality control limits nor interpretive criteria could be established. Endpoint interpretation and lot-to-lot variability in Mueller-Hinton agars were significant factors leading to interlaboratory variability.     

Haemophilus test medium versus Mueller-Hinton broth with lysed horse blood for antimicrobial susceptibility testing of four bacterial species

Studies were undertaken to determine whether broth microdilution susceptibility tests could be standardized by using a single medium for testing fastidious respiratory pathogens. Mueller-Hinton broth with lysed horse blood and the broth version of Haemophilus Test Medium (HTM) were directly compared. Ten orally administered agents were found to give essentially identical results in both media but minor differences were noted. Because the tests are easier to read when HTM broth is used, that medium is to be preferred for routine testing ofHaemophilus influenzae, Streptococcus pneumoniae, Streptococcus pyogenes andMoraxella catarrhalis isolates by the microdilution procedure.     

Multilaboratory evaluation of screening methods for detection of high-level aminoglycoside resistance in enterococci. National Committee for Clinical Laboratory Standards Study Group on Enterococci.

Since the early 1970s, the synergistic activity of an aminoglycoside with a cell wall-active agent has been predicted by determining the ability of an enterococcus to grow in the presence of high levels of the aminoglycoside (usually > or = 2,000 micrograms/ml). However, a variety of media and concentrations of aminoglycosides has been used for this screening procedure. In the present study, we sought to optimize the agar dilution, broth microdilution, and disk diffusion tests used to detect high-level gentamicin and streptomycin resistance in enterococci. For dilution tests, brain heart infusion agar or broth gave the best growth and performance. For agar dilution, 500 micrograms of gentamicin per ml, 2,000 micrograms of streptomycin per ml, and an inoculum of 1 x 10(6) CFU/ml were optimal, while for broth microdilution, 500 micrograms of gentamicin per ml, 1,000 micrograms of streptomycin per ml, and an inoculum of 5 x 10(5) CFU/ml were best. Growth of more than one colony in the agar dilution test was determined to be the best indicator of high-level resistance. For disk diffusion, Mueller-Hinton agar, 120-micrograms gentamicin disks, and 300-micrograms streptomycin disks with breakpoints of no zone for resistance and > or = 10 mm for susceptibility gave the best sensitivity and specificity if results for strains with zones of 7 to 9 mm are considered inconclusive, indicating that a broth or agar test should be performed to determine susceptibility or resistance.     

Comparison of five methods, including the PDM Epsilometer test (E test), for antimicrobial susceptibility testing of Pseudomonas aeruginosa.

The antimicrobial susceptibilities of 100 clinical isolates of Pseudomonas aeruginosa to six antipseudomonal antibiotics were tested by five methods: the National Committee for Clinical Laboratory Standards (NCCLS) methods for broth microdilution, agar dilution, and agar disk diffusion; the Vitek Automicrobic System method (Vitek Systems, Hazelwood, Mo.); and the PDM Epsilometer test (E test) (AB Biodisk, Solna, Sweden). The E test results showed excellent correlation with agar dilution results, with over 90% agreement within 1 doubling dilution between the E test and reference agar dilution MICs for all antimicrobial agents tested. The E test results also showed good correlation with the results from the reference agar disk diffusion method, with 90 to 99% complete agreement and 100% essential agreement on categories for all antibiotics tested (essential agreement is the agreement obtained when minor discrepancies are ignored). Comparison of categories with the E test and broth microdilution methods, using the broth microdilution method as the reference method, gave only 59% complete agreement for gentamicin, with 28 minor discrepancies and 13 very major discrepancies. Some discrepancies were observed between results from the E test and broth methods for gentamicin, with the broth microdilution and Vitek methods giving higher MICs than the E test and other methods using agar. The most recent NCCLS guidelines for broth dilution testing have reduced the recommended levels of cation supplementation, which may enhance future agreement between results for the aminoglycosides and P. aeruginosa on broth and on agar. We found that the E test offers a simple, labor-efficient, and accurate method for MIC determination on an agar medium.     

Tentative criteria for confirming the in vitro susceptibilities of Haemophilus influenzae and Neisseria gonorrhoeae to two fluoroquinolones (sparfloxacin and levofloxacin), including quality control parameters.

Sparfloxacin and levofloxacin were evaluated against 150 Haemophilus influenzae isolates and 149 Neisseria gonorrhoeae isolates in order to define susceptibility testing parameters. Sparfloxacin-susceptible H. influenzae strains were defined as those for which the MICs were < or = 0.25 microgram/ml and the zones were > or = 30 mm, and N. gonorrhoeae susceptible strains were those for which the MICs were < or = 0.03 microgram/ml and the zones were > or = 39 mm (5-micrograms disks). Levofloxacin-susceptible strains of H. influenzae included those for which the MICs were < or = 0.12 microgram/ml and the zones were > or = 32 mm and N. gonorrhoeae susceptible strains were those for which the MICs were < or = 0.12 microgram/ml and the zones were > or = 37 mm (5-micrograms disks). Criteria for a resistant category cannot yet be defined for either quinolone. In multilaboratory studies with different lots of Haemophilus Test Medium, replicate tests with the standard control strain of H. influenzae (ATCC 49247) were evaluated. For sparfloxacin disk tests, the proposed zone size limits were 33 to 42 mm and broth microdilution MIC limits were 0.004 to 0.016 microgram/ml, whereas for levofloxacin tests, zone size limits were 32 to 41 mm and broth microdilution MIC limits were 0.008 to 0.03 microgram/ml. Other multilaboratory studies evaluated tests with supplemented GC agar and N. gonorrhoeae ATCC 49226; for both drugs, zone size limits were 44 to 52 mm and agar dilution MIC limits were 0.004 to 0.016 microgram/ml.     

Antimicrobial susceptibility testing of less commonly isolated Haemophilus species using Haemophilus test medium.

Haemophilus test medium (HTM) was developed recently for dilution and disk diffusion antimicrobial agent susceptibility testing of Haemophilus influenzae. The application of HTM to the testing of other, less frequently encountered Haemophilus species recovered from humans was evaluated in this study by using commercially prepared HTM (BBL Microbiology Systems, Cockeysville, Md.) in broth microdilution and agar disk diffusion susceptibility tests with 18 antimicrobial agents. A total of 93.3% of 90 isolates belonging to six Haemophilus species provided acceptable growth in HTM agar disk diffusion tests, while only 63.3% (57 of 90) provided acceptable growth in the broth microdilution tests. However, HTM agar dilution testing provided an alternative method for those strains (primarily H. haemolyticus) which failed to grow adequately in broth. Based on the latest National Committee for Clinical Laboratory Standards guidelines (standard M2-T4) for interpretation of HTM disk tests of H. influenzae, the overall very major, major, and minor errors for all 18 drugs and six species tested were 0.2, 0.7, and 3.4%, respectively. Thus, the use of HTM in agar or broth susceptibility tests can be recommended for testing the less commonly encountered Haemophilus species by using the same test conditions and interpretive guidelines developed for H. influenzae.     

Relationship between in vitro susceptibility test results for chloramphenicol and production of chloramphenicol acetyltransferase by Haemophilus influenzae, Streptococcus pneumoniae, and Aerococcus species.

Haemophilus influenzae, Streptococcus pneumoniae, and Aerococcus species were tested for susceptibility to chloramphenicol by standard broth microdilution and disk-diffusion methods. MICs and zone diameter breakpoints were correlated with production of chloramphenicol acetyltransferase (CAT). A comparison of MICs and zone diameters indicated that the interpretative criteria for H. influenzae and S. pneumoniae should be an MIC of less than or equal to 4 micrograms/ml or a zone diameter greater than or equal to 25 mm for susceptible strains and an MIC of greater than or equal to 8 micrograms/ml or a zone diameter of less than or equal to 20 mm for resistant strains; for Aerococcus species, interpretative criteria should be an MIC of less than or equal to 8 micrograms/ml or a zone diameter of greater than or equal to 20 mm for susceptible strains and an MIC of greater than or equal to 32 micrograms/ml or a zone diameter of less than or equal to 12 mm for resistant strains. All but four strains of H. influenzae and one strain of S. pneumoniae that were resistant to chloramphenicol by these criteria produced CAT. For Aerococcus species, however, chloramphenicol-resistant strains were negative for CAT as determined by a commercially available disk test. When comparing susceptibility results with CAT production, thiamphenicol was a better indicator of the presence of the enzyme than chloramphenicol and may be useful in assaying resistance to chloramphenicol.     

Evaluation of a proprietary broth medium for microdilution susceptibility testing of nutritionally fastidious bacteria.

For microdilution susceptibility tests with nutritionally fastidious microorganisms, a new clear broth medium developed at Micro-Media Systems, Inc., Potomac, Md., was evaluated in a three-laboratory collaborative study. Replicate tests were performed with 80 isolates (51 Streptococcus spp., 27 Haemophilus influenzae isolates, and 2 Neisseria meningitidis isolates) against 15 antimicrobial agents. In standard 100-microliters volumes, results of tests in the new broth medium were comparable to those in the reference medium (Mueller-Hinton broth with 2 to 3% lysed horse blood), but MICs were somewhat easier to read in the new broth medium. Results of similar tests in smaller panels, containing 40 microliters in each well, were less satisfactory; i.e., growth failures and poorly defined endpoints were more commonly encountered. With drugs other than erythromycin or clindamycin, the 40-microliters panels provided MICs which compared favorably with those obtained by standard reference methods.     

Evaluation of in vitro methods for testing susceptibility of anaerobes to ampicillin-sulbactam and amoxicillin-clavulanic acid.

A total of 97 anaerobic bacteria were tested for susceptibility to ampicillin, ampicillin-sulbactam, and amoxicillin-clavulanic acid by broth microdilution and disk elution methods, the results of which were compared with those of the reference agar dilution method. With the broth microdilution method, approximately 95% of MICs were within 1 dilution of those of the reference agar method, with a definite (0.6 to 0.7 dilution) trend toward lower MICs. The disk elution test performed satisfactorily, but additional anaerobic isolates resistant to ampicillin-sulbactam and/or amoxicillin-clavulanic acid (currently rare) are needed to assure the predictability of resistance by the disk elution test.     

Multisite reproducibility of results obtained by two broth dilution methods for susceptibility testing of Mycobacterium avium complex

A multicenter study was conducted to assess the interlaboratory reproducibility of susceptibility testing of Mycobacterium avium complex (MAC) by broth microdilution using two different media (cation-adjusted Mueller-Hinton broth with 5% oleic acid-albumin-dextrose-catalase and 7H9 broth with casein) and by macrodilution using the BACTEC 460TB and 12B media at pH 6.8 and 7.3 to 7.4. Ten well-characterized strains of MAC (four macrolide susceptible, six macrolide resistant) were tested against clarithromycin and azithromycin (the latter only by BACTEC 460TB, pH 6.8). At each site, strains were tested against clarithromycin three times on each of three separate days (nine testing events per isolate) by using a common lot of microdilution trays and BACTEC 12B medium, pH 6.8; strains were tested once on three separate days against clarithromycin in 12B medium at pH 7.3 to 7.4 and against azithromycin. Agreement among MICs (i.e., mode +/- 1 twofold dilution) was 100% for all strains and both drugs when BACTEC 460TB was used, regardless of the pH of the medium, but varied when microdilution with either medium was used, particularly with susceptible strains. Agreement based on interpretive category, with NCCLS-recommended breakpoints, was 100% for all strains with the BACTEC 460TB method (both drugs and both pH values) and with microdilution using 7H9 broth. With microdilution and Mueller-Hinton broth, agreement by interpretive category was 100% for eight isolates and >90% for two; errors occurred only in laboratories where personnel had minimal experience with this technique. MAC susceptibility testing may be performed by broth macrodilution or microdilution at either pH, with NCCLS-recommended interpretive breakpoints. However, because visual interpretation of broth microdilution end points is subjective, it is more prone to reader error; therefore, this method requires greater expertise than the BACTEC 460TB. Both techniques require appropriate validation and continued documentation of proficiency.