[A simple disk diffusion test to identify beta-lactamase-negative, ampicillin-resistant Haemophilus influenzae--application of cephalexin, cefsulodin and cefaclor disks].

Currently, beta-lactamase-negative (BLN) ampicillin-resistant (AR) strains of Haemophilus influenzae are prevalent in Japan. BLNAR strains are defined by the presence of specific mutation(s) in the ftsI gene but are not phenotypically distinguishable by ampicillin (ABPC) susceptibility. In the present study, we showed that cephalexin (CEX), cefsulodin (CFS), and cefaclor (CCL) disk diffusion tests can be effectively used to identify BLNAR strains. A total of 169 clinical isolates of BLN H. influenzae, consisting of 113 of BLNAR and 56 of BLN, ampicillin-susceptible (AS), were included. All the isolates were genetically defined by detection of the TEM gene and partial sequencing of the ftsI gene. The Clinical and Laboratory Standards Institute (CLSI) standard broth microdilution and disk diffusion tests for ABPC provided 20% and 19% false susceptible rates, respectively. Alternatively, 34 cephem agents were tested using disk diffusion. Of the agents tested, CEX, CFS, and CCL disks could effectively discriminate between BLNAR and BLNAS isolates. All the BLNAS isolates showed visible growth inhibitory zones around CEX and CFS disks, but 108 (95.6%) and 106 (93.8%) BLNAR isolates did not. The results indicated 100% predictive values (PVs) for BLNAR and PVs for BLNAS were 91.8% for CEX and 88.9% for CFS. The CLSI-based interpretations for CCL (> or =20 mm) also highly correlated with BLNAR and BLNAS, PVs for BLNAR and for BLNAS being 100% and 93.3%, respectively. With simplicity and discriminability of the test method, we recommend a CEX disk diffusion test in combination with a rapid beta-lactamase test to identify BLNAR isolates in clinical laboratories.

[Multicenter-evaluation of optimal cell density to determine antimicrobial susceptibility for Haemophilus influenzae by the automated RAISUS system when the early-harvested bacterial cells were used].

The fully automated microbial system, RAISUS (Nissui Pharmaceutical, Tokyo, Japan) can provide antimicrobial susceptibility test results for the isolates of Haemophilus influenzae. It is known that viable cell concentrations (colony forming unit/ml) of H. influenzae significantly vary depending on the incubation period. For the rapid reporting of antimicrobial susceptibility test results, we evaluated optimal cell density when we prepared the cell suspension using the early-harvested (6 hour incubation) cells for RAISUS. A total of 180 clinical isolates, comprising of 33 ampicillin-susceptible isolates, 114 beta-lactamase negative but ampicillin-resistant isolates and 33 beta-lactamase positive and amoxicillin/clavulanic acid susceptible or -resistant isolates, were included. All the isolates were genetically defined according to the detection of TEM gene and specific mutation (s) in fts I gene. The isolates were incubated on chocolate agar plates for 6 hours, and then the cell suspensions were prepared and adjusted to 0.5, 0.25 and 0.125 McFarland standards through serially dilutions. The respective cell suspensions were tested by the RAISUS AST panels. The % agreements between RAISUS and Clinical and Laboratory Standards Institute standard microdilutions in ampicillin category interpretations were 66.7%(McFarland 0.5), 77.8% (McFarland 0.25) and 83.9%(McFarland 0.125). When the McFarland 0.125 cell suspensions were inoculated, the majority of discrep ant interpretations were minor errors (15.0%) and the occurrence of major error was 3.4%. There was no very major error throughout the study. Essential agreement in MIC determinations (with or within +/- 1 doubling dilution) for 11 beta-lactam antimicrobial agents tested improved to 95.2% by McFarland 0.125 when compared to 77.4% by McFarland 0.5. It was also demonstrated that the viable cell concentrations prepared from 6 hour incubation cultures were 2.5 to 6.5 times higher than those from 22 hour-incubations. With these results, it can be concluded that the early harvested cell suspension of H. influenzae is applicable to RAISUS antimicrobial susceptibility test with lower cell density (McFarland 0.125). With this adjustment, the antimicrobial susceptibility test for H. influenzae will be completed by RAISUS within 26 hours after primary isolation.

[Detection of vancomycin-resistant enterococci by a fully automated microbiology system, RAISUS].

A fully automated microbiology system, RAISUS recently developed (Nissui Pharmaceuticals Co., Ltd., Tokyo) was evaluated for identification of enterococci and for detection of vancomycin-resistant enterococci (VRE). When a total of 124 enterococcal isolates were tested, RAISUS correctly identified 122 (98.4%) isolates. Two isolates resulted in species-identifications disagreed with the reference but agreed as belonging to the genus of Enterococcus. When a total of fifty-seven VRE isolates confirmed to be positive for vanA and/or vanB genes were tested against vancomycin, the current RAISUS susceptibility program version 1.76 could detect 41 (71.9%) isolates of VRE as having > or = 32 microg/ml MIC for vancomycin, but one was intermediate (MIC, 8.0 microg/ml) and the remaining 15 vanB-type isolates were incorrectly interpreted as vancomycin-susceptible (MIC, < or = 4.0 microg/ml). The test program based on the algorism to determine bacterial growth in the presence of vancomycin was developed and evaluated. With this test program, all the VRE isolates positive for vanA and/or vanB genes were identified as being vancomycin-resistant or intermediate interpretation. However, eight of 19 clinical isolates of E. casseliflavus and E. gallinarum intrinsically possessing vanC gene were determined as being < or = 4.0 microg/ml MIC for vancomycin. With the influence of program revision, RAISUS became to incubate the test plate longer than with the current program, but 50% of enterococcal isolates including vancomycin-resistant and vancomycin-susceptible isolates were determined within 5 hour-incubations and 90% were within 9 to 10 hour-incubations. With these results, we can conclude that the revised test program for enterococcal isolates could rapidly and correctly identify vancomycin-resistance, and will be applicable to the routine susceptibility test in clinical laboratories.