![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Previous Article | Next Article 
Journal of Clinical Microbiology, November 2005, p. 5808-5810, Vol. 43, No. 11
0095-1137/05/$08.00+0 doi:10.1128/JCM.43.11.5808-5810.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.
Department of Infection Control and Laboratory Diagnostics, Tohoku University Graduate School of Medicine, Sendai, Miyagi 980-8574, Japan
Received 18 May 2005/ Returned for modification 10 July 2005/ Accepted 20 August 2005
 |
ABSTRACT |
|---|
 Top Abstract TextReferences |
|---|
|
TEXT |
|---|
|
TopAbstractTextReferences |
|---|
It is important to select proper antibacterial agents for these pathogens (6, 10), but many clinical laboratories currently spend from 48 to more than 72 h to obtain results from susceptibility testing. If resistant bacteria (methicillin-resistant or vancomycin-resistant bacteria) could be identified at an earlier stage, the use of ineffective antibiotics could be avoided, and a choice could be made of the most suitable antibacterial agents for patients with infections caused by a given bacterium. Moreover, immediate reports are likely to lead to avoidance of the emergence of new resistant bacteria, and such information is also likely to be useful for clinicians.
RAISUS is a fully automated system of identification and susceptibility developed by Nissui Pharmaceutical Co., Ltd. The RAISUS system is a fully automated instrument from sample inoculation onto a microtiter plate, culture, identification, and MIC interpretation to collection of the test device once the prepared sample solution is set to a McFarland standard of 0.5. RAISUS identifies bacterial strains using a fluorescence reagent (enzyme substrate or fluorescence indicator), and an antimicrobial susceptibility test is performed using an oxidation-reduction (redox) indicator. In this study, we compared the performance of RAISUS and VITEK in bacterial identification and the performance of RAISUS and CLSI (formerly NCCLS) methods in assessing antimicrobial susceptibility, using clinical isolates of staphylococci, including MRSA and MR-CoNS. We examined the time required for identification and determination of susceptibility using RAISUS and studied the potential effectiveness of RAISUS in clinical bacteriological laboratories.
The clinical isolates used in this study were 103 strains obtained from urine, blood, sputum, spinal fluid, and other samples, which were collected from inpatients at Tohoku University Hospital, Miyagi, Japan, from 2000 to 2001. Antimicrobial susceptibility testing was conducted using nine agents: penicillin G (PEN), oxacillin (OXA), amoxicillin-clavulanic acid (AMC), meropenem (MEM), erythromycin, clindamycin (CLI), vancomycin (VAN), trimethoprim-sulfamethoxazole (SXT), and levofloxacin (LVX).
The MIC was determined by broth microdilution methods according to CLSI criteria (12). We prepared frozen microtiter plates using Mueller-Hinton broth for MIC measurement following CLSI broth microdilution susceptibility test standards. In obtaining such data, it is well-known that fundamental errors of antimicrobial susceptibility testing occur within a range from half to double the MIC (a single doubling dilution difference) (8, 9). Therefore, data within a single doubling dilution difference were considered to be in agreement, and this criterion was used to assess the essential agreement of MICs within 2 dilutions for the two methods. In addition, the measured MIC was classified into three categories: susceptibility (S), intermediate resistance (I), and resistance (R), according to the SIR concentration range in the CLSI methods (12). Agreement in clinical category (ACC) was evaluated using three criteria: very major error (VME), major error, and minor error (5).
Agreement between RAISUS and VITEK for identification of S. aureus was 97.8% (44/45), and the methods showed disagreement for only one strain. In identification of CoNS, RAISUS and VITEK showed agreement for 75.9% of the strains (44/58) and gave different results for 14 strains. Among the 14 strains, 10 strains identified as Staphylococcus capitis by RAISUS were identified as Staphylococcus epidermidis by VITEK. These strains were studied by DNA-DNA hybridization by the method of Ezaki et al. (2), which confirmed that all these strains were identified correctly by RAISUS.
The essential agreement of MICs within 2 dilutions for the two methods for S. aureus was lowest at 62.2% for AMC and highest at 100% for CLI, VAN, and LVX (Table 1). The lowest agreement in clinical category was 66.7% for AMC, and the highest was 100% for CLI, VAN, and SXT. For OXA, which is essential for MRSA identification, the MIC agreement was 84.4% (38/45) and ACC was 100% (45/45). Moreover, 95.8% (23/24) of strains identified as MRSA and 95.2% (20/21) identified as methicillin-sensitive S. aureus by CLSI methods were identified as MRSA and methicillin-sensitive S. aureus by RAISUS within 5 h after the measurements were started. The remaining two strains required 6.25 h for identification (Fig. 1).
|
|
|
Identification by RAISUS showed a high correlation (97.8%) with VITEK for S. aureus, but a correlation of only 75.9% for CoNS, suggesting that strain identification for CoNS may be more difficult than for S. aureus. The RAISUS results for all strains for which there was disagreement with VITEK results were consistent with DNA-DNA hybridization data.
The susceptibility of S. aureus determined by RAISUS correlated well with the results from CLSI methods for all agents except AMC. The growth curve for the well with 8 µg/ml AMC, which was the SIR breakpoint, indicated that the absorbance increase in the AMC well occurred after about 6 to 8 h, whereas the increase in absorbance of wells with other agents was noted in approximately 5 h. This phenomenon of delayed increase in the growth curve was specific for MRSA. Therefore, the cause for disagreement between RAISUS and CLSI methods for AMC may be associated with induced drug resistance. We assume that one of the reasons why the RAISUS system resulted in VMEs is the high affinity of AMC to PBP2' of MRSA (1). Furthermore, all 20 strains with VMEs for AMC, MEM, and erythromycin were MRSA, again suggesting induced drug resistance.
The susceptibility of CoNS determined by RAISUS correlated well with the results from CLSI methods. All eleven strains which gave VMEs for PEN, AMC, MEM, and LVX were MR-CoNS, and this was also considered to be due to induced drug resistance. However, there will be no problems, since MRSA and MR-CoNS isolates would be reported as resistant if CLSI guidelines are followed.
The total times for RAISUS identification and antimicrobial susceptibility testing using OXA were about 5 h for S. aureus and about 7 h for CoNS. The difference in identification times between S. aureus and CoNS might simply result from the large number of CoNS strains that developed slowly. We conclude that RAISUS shows excellent accuracy in identification of staphylococci and that rapid and effective reports can be achieved from RAISUS susceptibility data. Therefore, RAISUS can provide the necessary information for selection of appropriate antibacterial agents in a shorter time frame.
|
FOOTNOTES |
|---|
|
REFERENCES |
|---|
|
TopAbstractTextReferences |
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Antimicrob. Agents Chemother. | Clin. Microbiol. Rev. |
|---|---|
| Clin. Vaccine Immunol. | ALL ASM JOURNALS |
|---|
