We have developed a modified cyclodextrin solid (MCS) medium using the selective antibiotic cefdinir. MCS medium exhibited higher sensitivity (95.6%; any culture-positive sample as reference) and greater inhibition of nasopharyngeal flora than did Bordet-Gengou agar (65.2%, P ‫؍‬ 0.009) or cyclodextrin solid medium (39.1%, P < 0.001).Pertussis is an acute respiratory infection caused by Bordetella pertussis. Since the introduction of the acellular pertussis vaccine, the number of reported pertussis cases has drastically decreased. However, occasional outbreaks have still been reported (2-4, 7), and adult pertussis has emerged (2, 3).Culturing B. pertussis from clinical specimens is the "gold standard" for diagnosis of pertussis, although this remains an insensitive method (12, 13). In addition, isolation of clinical strains is required for epidemiological analysis (including phenotypic and genotypic characterization). It is also required for the determination of the appropriate vaccine strain and the antimicrobial susceptibility of isolates in order to control the spread of pertussis (12, 13). Bordet-Gengou agar (BG agar) was the standard culture medium for isolation of B. pertussis but has the problem of low selectivity. More-selective media, such as BG agar with 40 g/ml of cephalexin (cefalexin) (BG agar) and Regan-Lowe agar with 40 g/ml of cephalexin (RL agar; charcoal agar based), have been described; however, these media have a short shelf life because they contain blood (6, 15). Cyclodextrin solid medium with 5 g/ml of cephalexin (CS medium) does not contain blood products and is reported to have improved selectivity and a long shelf life (1). However, the rate of detection of B. pertussis by this medium was lower than that achieved with other conventional media (10). Moreover, ␤-lactam (especially narrow-spectrum cephalosporins like cephalexin)-resistant bacteria, such as Haemophilus influenzae, Streptococcus pneumoniae, Staphylococcus aureus, and Moraxella catarrhalis, have recently emerged (8, 11). Therefore, cephalexin may no longer be a suitable selective reagent for isolation of B. pertussis from nasopharyngeal specimens. To address this problem, we have attempted to improve the selective isolation of B. pertussis when performing direct plating of clinical specimens. Modified CS medium (MCS medium) was prepared by replacing cephalexin with cefdinir (Asteras Pharma Inc., Tokyo, Japan), which inhibits the growth of M. catarrhalis (9, 16); vancomycin; and amphotericin B at final concentrations of 4 g/ml, 8 g/ml, and 4 g/ml, respectively. The concentration of cefdinir was set by the MIC data from 50 clinical strains of M. catarrhalis and 40 clinical strains of B. pertussis. MCS medium was composed of basic medium and supplement. The basic medium contained the following: 10.7 g of sodium glutamate, 0.24 g of L-proline, 2.5 g of NaCl, 0.5 g of KH 2 PO 4 , 0.2 g of KCl, 0

We determined antimicrobial susceptibilities and analyzed molecular epidemiology of 26 strains of Bordetella pertussis clinically isolated and then performed pulsed-field gel electrophoresis (PFGE) in Japan (Japanese Pertussis Surveillance Group Participants), from 2001 to 2002. The MICs of erythromycin, clindamycin, tetracyclines, fluoroquinorones, trimethoprim-sulfamethoxazole and rifampicin of all isolates against these showed 1 microgram/ml or less. Sparfloxacin is the most potent agent, of which the MICs showed 0.008-0.016 microgram/ml. Results of DNA fingerprinting by pulsed-field gel electrophoresis (PFGE) differentiated three types (Type I; 11 strains (42%), type II; 14 strains (54%) and type III; 1 strains (4%)). However, no relation between regions and identical PFGE patterns was found in this study. Further, surveillance of the antimicrobial susceptibilities and molecular epidemiology of B. pertussis will be required.