In vitro susceptibility testing of Haemophilus influenzae: review of new National Committee for Clinical Laboratory Standards recommendations.

In June 1987, the Antimicrobial Susceptibility Testing Subcommittee of the National Committee for Clinical Laboratory Standards (NCCLS) established a fastidious-organism working group expressly for the purpose of addressing issues related to in vitro susceptibility tests with aerobic or microaerophilic bacteria that do not grow well on unsupplemented Mueller-Hinton (MH) media. The fastidious-organism working group considered Haemophilus influenzae, for which no standardized method of susceptibility testing was available, for either disk diffusion or dilution susceptibility testing. Previously, numerous media with various inoculum sizes and conditions of incubation had been employed (24). Clearly, medium composition is crucial to susceptibility test results with Haemophilus spp. (1, 4-6, 17). Perhaps the most widely used medium for disk testing ofH. influenzae, at least in the United States, has been chocolatized MH agar (MHchoc) (25). Indeed, the NCCLS had promulgated zone diameter interpretive criteria for four antimicrobial agents, ampicillin, amoxicillin-clavulanate, ampicillin-sulbactam, and chloramphenicol, when H. influenzae is tested on MHchoc (21). Similarly, the broth medium, MH supplemented with lysed horse blood (3 to 5%) (MH-LHB), appeared to be used more commonly than other media for dilution tests with this organism (25). MIC interpretive criteria existed for the same four antimicrobial agents (20). The first issue considered by the NCCLS was the optimum medium composition for susceptibility tests with Haemophilus spp. Concerns pertaining to MH-choc included its opacity, its undefined chemical composition, and its lack of utility for trimethoprim-sulfamethoxazole (TMP-SMX) disk diffusion tests due to the presence of large amounts of thymidine and thymidine analogs. The major shortcomings of MH-LHB broth for dilution tests with Haemophilus spp. were its lack of commercial availability and the logistical complexities of preparing this medium in house. For these reasons, the NCCLS considered alternative media for standardized disk diffusion and dilution susceptibility tests with Haemophilus spp. In January 1988, the use of Haemophilus Test Medium (HTM), a new medium first developed by Jorgensen and colleagues (16), was explored. The agar form of HTM consisted of MH agar, 15 ,ug of hematin per ml, 15 jig of NAD per ml, and 5 ,ug of yeast extract per ml. The broth version of HTM consisted of cation-adjusted MH broth, 0.2 IU of thymidine phosphorylase per ml, and the same supplements used in HTM agar. In preliminary studies with 15 antimicrobial agents, laboratory-prepared agar and broth versions of HTM were stable and appeared to support the growth of fresh clinical isolates of H. influenzae (16, 17). Other presumed advantages of HTM were its transparent nature, its ability to support disk diffusion tests with TMPSMX, its lack of expense, and its chemical composition, which was more defined than that of MH-choc; in addition, HTM permitted the use of the same basal medium in both disk diffusion and broth dilution tests, and the broth version of the medium could be manufactured commercially. On the basis of a series of preliminary studies (23a), the NCCLS adopted HTM as the recommended medium for disk diffusion and dilution susceptibility testing of Haemophilus spp. in January 1988 (22, 23). In addition, methods for both procedures were described. With disk diffusion tests, HTM plates should be incubated at 35°C in CO2 (5 to 7%) for 16 to 18 h prior to zones of inhibition being read. Inocula are to be prepared from a saline suspension of the test organism equivalent in turbidity to a 0.5 McFarland standard. Because of the importance of accurate inoculum concentrations, it was recommended that this initial suspension be prepared with a bench-top nephelometric device directly from colony growth on a 20to 24-h chocolate agar culture. The method chosen for dilution tests with HTM was a broth microdilution procedure (final volume per well, 100 pl) with trays incubated for 24 h at 35°C in ambient air and a final inoculum concentration of 5 x 105 CFU/ml. Using HTM and these methods, a three-center collaborative study was conducted to develop interpretive criteria for disk diffusion and broth microdilution results (9). On the basis of the results of this study, interpretive criteria for 18 antimicrobial agents tested against Haemophilus spp. were adopted in June 1988 by the NCCLS (22, 23). These antimicrobial agents included ampicillin, amoxicillin-clavulanate, ampicillin-sulbactam, cefuroxime, cefamandole, cefonicid, cefaclor, ceftriaxone, cefotaxime, ceftizoxime, ceftazidime, imipenem, aztreonam, chloramphenicol, tetracycline, TMPSMX, rifampin, and ciprofloxacin. In June 1989, interpretive criteria for cefixime were adopted, as were quality control (QC) criteria for the antimicrobial agents, which now numbered 19 (22, 23). These QC criteria had been developed in a six-center collaborative study using a non-typeable P-lactamase-negative, ampicillin-resistant (BLNAR) strain of H. influenzae, ATCC 49247, as a challenge organism (7). This strain was selected because it yielded adequate interand intralaboratory precision in the controlled, six-center collaborative study. It grew well on HTM and yielded zone sizes, and the MICs of most antimicrobial agents for this strain were midrange because of its BLNAR characteristics. During the past 2 1/2 years, the NCCLS has adopted