Comparison of the serum-supplemented Todd-Hewitt and the new Haemophilus test media for broth microdilution susceptibility testing of Streptococcus pneumoniae.

Horse serum-supplemented Todd-Hewitt broth (STH) in use at Hôpital Ste-Justine for the last 12 years was compared to the recently proposed Haemophilus test medium (HTM), for broth microdilution susceptibility testing of Streptococcus pneumoniae. One hundred and twenty S. pneumoniae isolates from pediatric clinical specimens were used in this study. In general, the minimum inhibitory concentrations (MICs) in STH for 15 antimicrobial agents were quite comparable to those determined in HTM but tended to be higher. Drugs which generated MICs within +/- 1 log2 concentration differences in both media included penicillin, ampicillin, oxacillin, cefuroxime, cefotaxime, cefixime, clindamycin, chloramphenicol, trimethoprim-sulfamethoxazole, rifampin, ciprofloxacin and vancomycin. Cefaclor and tetracycline MICs tended to be > or = 2 log2 dilutions higher with STH for most of the isolates tested, while erythromycin MICs were often 2 log2 dilutions lower with STH than with HTM. Despite some differences in MICs noted above, few very major (0.4%), major (0.2%) and minor interpretive category errors (4.4%) were observed. The visual reading of the MICs for most of the 120 clinical isolates tested was generally easier in STH which was superior in supporting best the bacterial growth as detected by spectrophotometry. The risk of false susceptibility is thus decreased by using STH rather than HTM; furthermore, STH is free of the technical problems of the lysed horse blood Mueller-Hinton (LHB-MH) recommended by the NCCLS.