Skip to main page content

• HOME
• Current Issue
• Archive
• Alerts
• About ASM
• Contact us
• Tech Support
• Journals.ASM.org

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Rahman, M. Articles by Möllby, R. Search for Related Content PubMed PubMed Citation Articles by Rahman, M. Articles by Möllby, R. Agricola Articles by Rahman, M. Articles by Möllby, R.

 Previous Article  |  Next Article 

Applied and Environmental Microbiology, April 2004, p. 2398-2403, Vol. 70, No. 4
0099-2240/04/$08.00+0     DOI: 10.1128/AEM.70.4.2398-2403.2004
Copyright © 2004, American Society for Microbiology. All Rights Reserved.

Evaluation of a Scanner-Assisted Colorimetric MIC Method for Susceptibility Testing of Gram-Negative Fermentative Bacteria

Mokhlasur Rahman,^1* Inger Kühn,¹ Motiur Rahman,² Barbro Olsson-Liljequist,³ and Roland Möllby¹

Microbiology and Tumor Biology Center, Karolinska Institutet, S-171 77 Stockholm,¹ Department of Bacteriology, Swedish Institute for Infectious Disease Control, S-171 82 Solna, Sweden,³ International Center for Diarrhoeal Disease Research, Mohakhali, Dhaka-1212, Bangladesh²

Received 28 October 2003/ Accepted 9 December 2003

Top Abstract Introduction Materials and Methods Results Discussion References

 
We describe the ScanMIC method, a colorimetric MIC method for susceptibility testing of gram-negative fermentative bacteria. The method is a slight modification of the National Committee for Clinical Laboratory Standards (NCCLS) recommended broth microdilution method that uses a redox indicator 2,3,5-triphenyltetrazolium chloride (TTC) to enhance the estimate of bacterial growth inhibition in a microplate and a flatbed scanner to capture the microplate image. In-house software was developed to transform the microplate image into numerical values based on the amount of bacterial growth and to generate the MICs automatically. The choice of indicator was based on its low toxicity and ease of reading by scanner. We compared the ScanMIC method to the NCCLS recommended broth microdilution method with 197 coliform strains against seven antibacterial agents. The interpretative categorical agreement was obtained in 92.4% of the assays, and the agreement for MIC differences (within ±1 log₂ dilution) was obtained in 96% for ScanMIC versus broth microdilution and 97% for a two-step incubation colorimetric broth microdilution versus the broth microdilution method. The method was found to be labor-saving, not to require any initial investment, and to show reliable results. Thus, the ScanMIC method could be useful for epidemiological surveys that include susceptibility testing of bacteria.

Top Abstract Introduction Materials and Methods Results Discussion References

 
One of the principal areas of medical concern has been the emergence (14) and transmission of multidrug resistant bacteria in humans (10, 27), since the wide use of antibiotics in hospitals and the community, agriculture, and animal feed has increased the number of resistant bacteria in the environment (12, 26). Obviously, resistant bacteria pose a public health hazard when they are present in the environment because of the likelihood of transmission to humans (4, 24). Thus, it is important to monitor the resistance of bacteria in the environment as well as in humans. To facilitate such studies, we developed a semiautomated colorimetric MIC method, the ScanMIC method, for determining susceptibility patterns for hundreds of strains of gram-negative fermentative bacteria.

Many different methods are available for bacterial susceptibility testing, including disk diffusion, agar dilution, broth microdilution, and antibiotic gradient disks (15). Broth microdilution is convenient and widely used for susceptibility testing of several antibiotics on a large number of bacterial isolates in a short time. Most of the commercially available automated systems, including MicroScan plates with the WalkAway-96 system (23) and Vitek automated microbiology system (bio Merieux Vitek, Inc.) (3), and semiautomated (the Wider system) (6) susceptibility testing systems are also based on the broth microdilution method. Some of those automated systems use image-processing technology to read the plates to avoid the time-consuming labor of visually reading and registering the MICs on a form, followed by manually entering them into the computer. In those semiautomated systems, spectrophotometers, digital cameras, or video cameras are used to capture the micro plate images (6, 17). However, a cheap, attractive alternative would be the use of a normal flatbed scanner, a method that was evaluated in this study.

In broth microdilution, as with many susceptibility tests, problems might arise when determining the endpoints of growth, which may lead to false interpretations of susceptibility and to decreased reproducibility (9). Since turbidity is measured visually, the endpoint of bacterial growth may be underestimated. This problem has been partly solved in some laboratories by using an indicator in the conventional broth microdilution method, a procedure referred to as a colorimetric MIC method (1, 2, 13). In a colorimetric MIC method, a redox indicator changes the color in response to bacterial growth, which enhances the detection of growth. Adding the indicator also helps to semiautomate the broth microdilution method through image processing. However, the indicators might affect bacterial growth, which in turn may change the interpretation of susceptibility (13, 25). In order to be able to select a suitable indicator, we compared five different indicators for toxicity and suitability for computerized image analysis.

In this study, we describe the ScanMIC method and evaluate the MIC results obtained by the ScanMIC method in comparison to conventional broth microdilution method as recommended by the National Committee for Clinical Laboratory Standards (NCCLS) (22) and to a two-step incubation colorimetric broth microdilution method for 197 gram-negative coliform bacterial strains against seven antibacterial agents.

Top Abstract Introduction Materials and Methods Results Discussion References

 
Bacterial isolates.
Environmental samples were taken at different sites of a model sewage treatment plant with duckweed for purification of hospital sewage water in Bangladesh. In addition, fecal samples were collected from children attending the hospital who were suffering from diarrhea. A total of 1,005 environmental and human coliform bacterial isolates were collected. The isolates were subjected to phenotyping by a biochemical fingerprinting system (18, 20), the PhenePlate system (PhPlate Microplate Techniques, Solna, Stockholm, Sweden); 197 strains, each representing a different PhP phenotype, were subjected to MIC determination. According to the PhP results, 95 strains were Escherichia coli, 18 strains were Klebsiella spp., 16 strains were Citrobacter spp., 15 strains were Proteus spp., 9 strains were Serratia spp., 7 strains were Enterobacter spp., and 37 strains were other genera of the Enterobacteriaceae family. The isolates were stored in BHI broth with 30% glycerol at –70°C and cultured on blood agar plates prior to susceptibility testing. NCCLS recommended reference strains Enterococcus faecalis ATCC 29212, Escherichia coli ATCC 25922 and 35218, Pseudomonas aeruginosa ATCC 27853, and Staphylococcus aureus ATCC 29213 were used as control strains in all experiments.

Antimicrobial agents.
The antimicrobial agents tested were purchased from Sigma (St. Louis, Mo.), each one with a specified assay potentiality. Each one was tested in a series of seven twofold dilutions. The concentrations were selected from two dilution steps below the susceptible breakpoint to two steps above the resistance breakpoint. The agents tested and ranges of concentrations tested were: ampicillin, 2 to 128 mg/liter; tetracycline, 1 to 64 mg/liter; chloramphenicol, 2 to 128 mg/liter; nalidixic acid, 2 to 128 mg/liter; cephalothin, 1 to 64 mg/liter; streptomycin, 2 to 128 mg/liter; and gentamicin, 1 to 64 mg/liter.

Preparation of microdilution plates of the antibiotics.
Weighing of antibiotic powders, solution preparation, and preparation of microdilutions of antimicrobial agents (twice the desired final concentration) in round-bottomed microplates were performed according to the NCCLS recommendations (22). Furthermore, in the ScanMIC method, a growth indicator, 2,3,5-triphenyltetrazolium chloride (TTC; Sigma), was also added to the microplates, and the microplates were dried overnight with dry air at 25°C. All antimicrobial microdilution trays were stored in the cold room and used within 2 weeks after preparation.

Procedure for susceptibility testing.
Bacteria were cultured overnight on blood agar plates. Three colonies were transferred to 5 ml of Mueller Hilton broth (Becton Dickinson) without blood and incubated at 37°C for 4 h to reach the exponential phase of growth. From these cultures, bacterial turbidity was adjusted to 0.5 on the McFarland turbidity standard as measured by absorbance (0.08 to 0.1 at 625 nm) in a spectrophotometer (Hitachi U-1100), corresponding to approximately 10⁸ CFU/ml. The adjusted bacterial suspensions were first diluted 1:200 in Mueller-Hilton broth, and 100 µl was added to each well (5 x 10⁴ CFU/well) in the preprepared microplate (broth microdilution or ScanMIC microplate), followed by incubation for 16 h at 37°C.

For broth microdilution, the MIC was defined as the lowest concentration of an antimicrobial agent at which no growth was detected as turbidity seen visually. For ScanMIC, the microplates were scanned with a reflective flatbed scanner (UMAX-Astra 6450) connected to a Windows-based personal computer. The images were analyzed with software developed in-house (available through PhPlate Microplate Techniques AB, Sweden). The software measures the intensity of the red color and the diameter of each pellet formed by bacterial growth, resulting in a numerical value representing the amount of pellet or formazan formation (pellet value). The background (value for the blank wells) is first deducted from the values for all other wells. After this, the software determines the lowest concentration showing a selected degree of growth inhibition, and from this concentration, the MIC for the antibiotic is calculated. In this investigation, 100% growth inhibition was used as limit in the MIC determination (22).

In another assay, referred to as the two-step incubation colorimetric broth microdilution method, the indicator (TTC) was added to each well of the microplates after 16 h of incubation of the broth microdilution MIC plates. The bacteria were sucked up and down with the multipipette, after which the plates were incubated for 3 h longer. The microplates were scanned, and the MIC was defined as described above for the ScanMIC method. Broth microdilution, ScanMIC, and two-step incubation colorimetric broth microdilution were performed on the same day for the same set of isolates, including reference strains. Duplicate tests were also performed on the following day for the subset of bacteria.

Growth indicators, toxicity, and kinetic curve assay.
The indicators investigated were TTC, 3-[4,5 dimethylthiazol-2-yl] 2,5-diphenyltetrazolium bromide (MTT; Sigma), 2,3-bis[2-methoxy-4-nitro-5-sulfophenyl]-2H-tetrazolium carboxanilide inner salt (XTT; Sigma), 2-[4-iodophenyl-]3-[4-dinitrophenyl]-5-phenyltetrazolium chloride (INT; Sigma), and resazurin (Eastman).

For toxicity testing of the indicators, inocula of the reference strains were adjusted (5 x 10⁴ CFU/ml) in Mueller Hilton broth as described above, and 100 µl was incubated for 14 h at 37°C in each well of the 96-well microplates, supplemented with different concentrations of the indicators. After incubation, the absorbance values were measured with a spectrophotometer (I EMS; Labsystems, Helsinki, Finland) at 540 nm. All experiments were repeated three times, and the mean values were calculated.

For the checkerboard experiment (kinetic curves), different concentrations of the adjusted inoculum were incubated with different concentrations of TTC in a 96 well microplate. The microplate was incubated in the spectrophotometer at 37°C, and absorbance values were measured every 30 min for 14.5 h with in-house-developed software.

Agreement with reference method.
The ScanMIC method was compared to the reference broth microdilution method in two different ways. First, the distribution of differences in MIC results was determined. Agreement between MICs was calculated as follows: if the MICs obtained by the reference method and the tested method were identical, the difference was 0; if the tested method gave a MIC one dilution step larger than the reference method, the difference was +1, and so on; for lower MICs, the differences were negative. The overall essential agreement was calculated within ±1 log₂ dilution, i.e., the percentage of the isolates that gave identical results between plus and minus one dilution step (2). MICs lower than the lowest tested concentration were considered equal to the lowest concentration, and MICs higher than the highest tested concentration were considered equal to the highest tested concentration. In the second method, interpretative categorical results, obtained according to the NCCLS recommended interpretive standards for susceptible, intermediate, and resistant (SIR) for Enterobacteriaceae (22), were compared for the ScanMIC method versus the reference method. This analysis compared the results with regard to clinical interpretation SIR classification, and "absolute interpretative agreement," obtained when both methods resulted in the same SIR categories. Absolute interpretative agreement, essential agreement, as well as minor, major, very major, and essential errors were also calculated and rated as described by Murray et al. (21). For the details of these calculations, see the footnotes to the tables.

Top Abstract Introduction Materials and Methods Results Discussion References

 
Indicator selection.
In this study, toxic effects to reference strains caused by five different indicators were examined. Figure 1 shows the absorbance values of an E. coli strain (ATCC 25922) grown with different concentrations of the indicators. It was found that MTT and INT were more toxic to all the reference strains than TTC, XTT, and resazurin, since growth was completely inhibited by MTT and INT at concentrations of >0.007% and >0.03%, respectively. TTC, resazurin, and XTT at concentrations of <0.125% did not exhibit any inhibitory effects on bacterial growth. It was also observed that at concentrations below 0.003%, such small amounts of indicator did not affect the absorbance or change the color of the bacterial pellet, as detected by image analysis. Again, resazurin, and XTT did not form a pellet upon reduction. Resazurin shifted into three different colors during reduction, and XTT is by far more expensive than TTC. Furthermore, resazurin and XTT also exhibited a background color, which makes them less suitable for image analysis. Thus, resazurin and XTT were less suitable indicators for the present assay, and therefore we chose the indicator TTC.

View larger version (16K): [in this window] [in a new window]

FIG. 1. Representative curves generated by the growth responses, measured as absorbance at 540 nm after 14 h of growth, of E. coli ATCC 25922 (5 x 104 CFU/well) with different concentrations of the indicators. Resa, resazurin.

To optimize the concentrations of indicator and bacteria, checkerboard experiments were carried out. Figure 2 shows the 14.5-h kinetic curves in a microtiter plate with different concentrations of TTC and bacterial inocula, and 0.005% TTC was chosen as the optimal concentration. It was also noted that the initial concentration of the inocula was of less importance.

View larger version (36K): [in this window] [in a new window]

FIG. 2. Checkerboard titration of bacterial growth in relation to inoculum size and concentration of TTC in a microplate, where each row shows decreasing concentrations of indicator and each column shows decreasing inoculum size. The kinetic curves were generated from absorbance values (A540) measured automatically every 30 min from 0 to 14.5 h. E. coli ATCC 25922 was used for the experiment. The experiments were repeated three times, and mean values were used to construct the kinetic curves.

ScanMIC versus broth microdilution method.
The MICs of the seven antibacterial agents against 197 coliform strains tested in duplicate were used to evaluate the accuracy of the ScanMIC method compared to the NCCLS recommended broth microdilution method. A total of 2,712 organism-antimicrobial agent combinations were analyzed. Table 1 shows the distribution differences in MICs by the ScanMIC method versus the broth microdilution method. The overall essential agreement (±1 log₂ dilution) was 95.8%, and the essential agreements ranged from 92.4% for tetracycline to 99.0% for nalidixic acid. The best essential agreements were observed for nalidixic acid. The MICs of cephalothin and gentamicin tended to be slightly lower in the ScanMIC method than the broth microdilution method, and the MICs of chloramphenicol, nalidixic acid, streptomycin, and ampicillin tended to be slightly higher in the ScanMIC method than the broth microdilution method.

View this table: [in this window] [in a new window]

TABLE 1. Distribution of differences in the MICs determined by the ScanMIC method versus the NCCLS recommended broth microdilution method

Two-step incubation colorimetric broth microdilution method versus broth microdilution method.
In colorimetric MIC methods, indicators may be used in different ways. The indicator may be added to the microplate before the bacteria; i.e., the microplate is incubated once. This procedure is referred to as a one-step incubation colorimetric broth microdilution method, including our ScanMIC method. Alternatively, the indicator may be added to the broth microdilution plate after overnight incubation with bacteria, followed by incubation for another 3 h, i.e., a two-step incubation colorimetric broth microdilution method.

The MICs of the seven antibacterial agents against 197 coliform strains tested in duplicate were used to evaluate the accuracy of the two-step incubation colorimetric broth microdilution method compared to the broth microdilution method. A total of 2,712 organism-antimicrobial agent combinations were analyzed. Table 2 shows that the MIC differences and the overall essential agreement (±1 log₂ dilution) were 96.6%. In addition, the MICs were also read visually in parallel with the scanner. These two reading methods resulted in >99% concordance for the MICs obtained (data not shown)

View this table: [in this window] [in a new window]

TABLE 2. Distribution of differences in the MICs determined by the two-step incubation colorimetric broth microdilution method versus the NCCLS-recommended broth microdilution methoda

Agreement analysis by interpretative categories.
MICs were converted to interpretative categories of susceptible, intermediate, and resistant for both methods to evaluate the categorical agreements between the ScanMIC method and the reference broth microdilution method. Table 3 shows the number of susceptible, intermediate, and resistant strains; minor, major, very major, and essential errors; and absolute and essential category agreement between the ScanMIC and broth microdilution methods. Overall, absolute agreement in interpretative categories was obtained for 92.4%, essential agreement for 98.3%, and minor errors for 6%. Major errors were only seen in 1.7%, and very major errors were only seen in 0.9%. The absolute agreement was lowest (82%) for cephalothin; intermediate (90%) for ampicillin, tetracycline, and streptomycin; and excellent (>98%) for chloramphenicol, nalidixic acid, and gentamicin.

View this table: [in this window] [in a new window]

TABLE 3. Overall results by interpretative categories for reference method and errors and agreement between the ScanMIC method and reference method

Top Abstract Introduction Materials and Methods Results Discussion References

 
The broth microdilution method with commercially prepared antibiotic panels has become popular for susceptibility testing in clinical and veterinary microbiology laboratories. The ScanMIC method is a colorimetric broth microdilution method, and one major advantage of our system is the use of a simple flatbed scanner for reading the susceptibility test results. The use of a scanner offers a number of advantages; it is cheap, readily available, and easy to connect to a computer, and the whole microplate image can be saved for further analysis and verification. The method was used in a practical study and found to be easy to use for susceptibility testing of a large number of isolates.

In a recent study, our group evaluated different redox indicators and the use of a flatbed scanner for quantification of microbial growth in microplates (11). It was shown that bacterial growth inhibition could be easily and reproducibly measured in round-bottomed microplates with a flatbed scanner, and it was also shown that TTC was an adequate indicator for measuring microbial growth with the scanner (11). Different indicators have been used to determine bacterial growth since the 1940s (25) as well as for bacterial, yeast, and fungal susceptibility testing (8, 19). Tetrazolium salts have been used in broth microdilution to enhance the detection of bacterial growth in different studies (13). However, it has also been reported that tetrazolium indicators may be toxic to several types of bacteria (13, 25). Our study again confirmed that high concentrations of indicators may have negative effects on bacterial growth (Fig. 1). It was also noticed that small amounts of indicator did not always reflect bacterial growth (pellet) properly. A wide range of indicator concentrations were found to be useful, and a concentration of 0.005% (wt/vol) TTC was chosen for this assay.

Highly automatic and semiautomatic commercial systems for bacterial susceptibility testing have been used for the last 30 years (6). In the highly automated systems, bacterial inoculum preparation, inoculum dilution and distribution to the antibiotic panel or card, incubation, final reading, and calculation of the MICs are done by fully automated machines. Those machines and devices are expensive, and some may need an experienced person to handle them properly. The antibiotic panels may be expensive, and the flexibility of the antibiotics is limited (5). Usually, the machines have been designed for susceptibility testing of hospital isolates, and in practice, they are only suitable for laboratories with a constant high load of samples to be tested. On the other hand, semiautomated systems are run more or less manually, but the results are normally obtained through the use of various plate-reading devices, such as a spectrophotometer, fluorometer, or digital camera, and the data are automatically transferred to a computer for image analysis, calculations, and storage of MICs (5, 6). Such systems are less time-consuming and require less-subjective judgment of results. Furthermore, semiautomated systems do not carry high initial costs, and the antibiotic panel may be prepared in house or by local companies, so the antibiotics to be tested can easily be selected according to the local demand.

In this study, we evaluated the use of a simple flatbed scanner to capture the microplate image (Fig. 3) and developed software to transfer the microplate images into MICs (Table 4) that were printed on a final result sheet (Table 5). Our ScanMIC method could be regarded as a semiautomated susceptibility testing method, which avoids visual reading and manual registering of the MICs. A further advantage of the software analysis is the fact that other limits than 100% inhibition can be used for the MIC calculation. For laboratories that only occasionally need to investigate the antibiotic susceptibility of a large number of isolates, the ScanMIC method may be an optimal choice. Furthermore, because of its low instrument cost, it may also be suitable in developing countries and small laboratories.

View larger version (82K): [in this window] [in a new window]

FIG. 3. Scanned image of an incubated ScanMIC microplate (susceptibility testing). Each column contains a concentration gradient of one antibiotic. Row A did not contain any antibiotic and served as a growth control. The MIC of each antibiotic can be seen as the first row of each column showing no growth (no red pellet). Columns 1 to 12 contained ampicillin, tetracycline, chloramphenicol, nalidixic acid, cephalothin, streptomycin, gentamicin, ampicillin, tetracycline, chloramphenicol, nalidixic acid, and cephalothin, respectively.

View this table: [in this window] [in a new window]

TABLE 4. Raw dataa

View this table: [in this window] [in a new window]

TABLE 5. Final result sheet provided by the softwarea

Data on accuracy and reproducibility are important in the evaluation of any new susceptibility testing method. According to the NCCLS, more than 95% of the MICs obtained from a new MIC test method should fall within the 3 log₂ dilutions of the values obtained with the reference method (15). However, NCCLS has not yet set standards for comparisons of automated susceptibility systems, but a guideline has suggested that the overall essential agreement must be >90%, very major errors must be <3%, and the rate of minor and major errors must not exceed 7% (7). Again, in the evaluation of a new susceptibility testing method, an adequate number of resistant strains should be tested because the susceptibility level of the tested population may affect the evaluation of susceptibility test errors (15).

In our study, with all 197 coliform strains tested against seven antibacterial agents, a mean of 44% strains were susceptible, 6% were intermediate, and 50% were resistant. The ScanMIC method was found to be comparable to the reference method, with an overall essential agreement of 96% (±1 log₂ dilution) and 92% interpretative category agreement. Neither minor, major, nor very major errors exceeded the recommended limits. Thus, the ScanMIC method results meet the performance criteria for susceptibility testing. Similar values for overall absolute categorical agreement (91%), distribution difference agreement (91%), major errors (1.9%), and very major errors (2.4%) were reported earlier by Baker et al. for comparisons between a colorimetric broth microdilution and the agar dilution method (1). The ScanMIC results showed better accuracy than the results found by Baker et al., but that is to be expected, since the ScanMIC method and the reference method were based on the broth dilution method.

It may be argued that the influence of the indicator on the action of the antibiotic and the influence of the software may give rise to false interpretations of the data and thus influence the accuracy of the MIC results. However, since these effects are clearly linked in the experimental situation, we have not been able to measure these underlying effects separately. To verify their influence on the accuracy of the results, the two-step incubation colorimetric broth microdilution method was performed (see Results section). It might be expected that the two-step incubation colorimetric broth microdilution method would show better accuracy than a one-step incubation colorimetric broth microdilution method, since possible effects of the indicator on bacterial growth or of interactions between the antibiotic and indicator could be escaped by adding the indicator after bacterial growth had occurred. The overall essential agreement between the two-step incubation colorimetric broth microdilution method and the broth microdilution method was found to be as high as 96.6% (Table 2). An earlier study by Johnson et al. reported an overall essential agreement of 93% when comparing the routine broth microdilution MIC method with a two-step incubation colorimetric broth microdilution method (13). In our study, as expected, the agreement between the two-step incubation colorimetric broth microdilution method and the broth microdilution was higher than that obtained for the ScanMIC method and the broth microdilution method. However, the overall effects on accuracy were within accepted limits.

In recent years, increasing bacterial resistance to antimicrobial agents has become a topic of great interest in the scientific community (28). In many countries, national surveillance and research programs have been initiated to monitor resistance in isolates from humans, foods, animals, and environment (16). To investigate the representative flora from each sample, it is important that all investigators be able to study large numbers of isolates. Therefore, a rapid and efficient method is needed to screen large numbers of isolates for antibiotic resistance patterns. When cheap, easy, and labor-saving susceptibility testing methods for large numbers of bacterial isolates are needed, the ScanMIC method is an alternative.

In conclusion, the achievement of this study is the development of a simple, cheap, labor-saving semiautomated susceptibility testing method for gram-negative fermentative bacteria. The method used concentration gradients of antibiotics in 96-well microplates, tetrazolium salt as a growth indicator, a flatbed scanner to read the results, and in-house-developed software to calculate and store the MICs. The high agreement between our the ScanMIC method and the NCCLS reference method for MIC determination leads us to conclude that the ScanMIC method could become an acceptable and useful method for epidemiological surveys of the resistance patterns of large numbers of bacteria in the hospital and in the environment. It is to be noted, however, that the method has so far been evaluated only for coliform bacteria with seven antibiotics.

 
This work was supported by SIDA/SAREC grant 1999-255 for research fellowships and the Karolinska Institutet fund.

We thank John Albert and G. B. Nair at the International Center for Diarrhoeal Disease Research in Bangladesh for laboratory support; Prism Bangladesh, Ltd., for sampling support; Lena Gezelius at the Swedish Institute for Infectious Disease Control for technical support; and Jenny Gabrielson at the Karolinska Institutet for valuable discussion.

 
* Corresponding author. Mailing address: Microbiology and Tumor Biology Center, Karolinska Institutet, Box 280, S-171 77 Stockholm, Sweden. Phone: 46 8 5248 71 53. Fax: 46 8 5248 7908. E-mail: mokhlasur.rahman{at}mtc.ki.se.

Top Abstract Introduction Materials and Methods Results Discussion References

 

1
1. Baker, C. N., S. N. Banerjee, and F. C. Tenover. 1994. Evaluation of Alamar colorimetric MIC method for antimicrobial susceptibility testing of gram-negative bacteria. J. Clin. Microbiol. 32:1261-1267.[Abstract/Free Full Text]
2
2. Baker, C. N., and F. C. Tenover. 1996. Evaluation of Alamar colorimetric broth microdilution susceptibility testing method for staphylococci and enterococci. J. Clin. Microbiol. 34:2654-2659.[Abstract]
3
3. Barry, J., A. Brown, V. Ensor, U. Lakhani, D. Petts, C. Warren, and T. Winstanley. 2003. Comparative evaluation of the VITEK 2 Advanced Expert System (AES) in five UK hospitals. J. Antimicrob. Chemother. 51:1191-1202.[Abstract/Free Full Text]
4
4. Bates, J., J. Z. Jordens, and D. T. Griffiths. 1994. Farm animals as a putative reservoir for vancomycin-resistant enterococcal infection in man. J. Antimicrob. Chemother. 34:507-514.[Abstract/Free Full Text]
5
5. Berke, I., and P. M. Tierno, Jr. 1996. Comparison of efficacy and cost-effectiveness of BIOMIC VIDEO and Vitek antimicrobial susceptibility test systems for use in the clinical microbiology laboratory. J. Clin. Microbiol. 34:1980-1984.[Abstract]
6
6. Canton, R., M. Perez-Vazquez, A. Oliver, B. Sanchez Del Saz, M. O. Gutierrez, M. Martinez-Ferrer, and F. Baquero. 2000. Evaluation of the Wider system, a new computer-assisted image-processing device for bacterial identification and susceptibility testing. J. Clin. Microbiol. 38:1339-1346.[Abstract/Free Full Text]
7
7. Chapin, K. C., and M. C. Musgnug. 2003. Validation of the automated reading and incubation system with sensititre plates for antimicrobial susceptibility testing. J. Clin. Microbiol. 41:1951-1956.[Abstract/Free Full Text]
8
8. Clancy, C. J., and M. H. Nguyen. 1997. Comparison of a photometric method with standardized methods of antifungal susceptibility testing of yeasts. J. Clin. Microbiol. 35:2878-2882.[Abstract]
9
9. Cunha, B. A. 1997. Problems arising in antimicrobial therapy due to false susceptibility testing. J. Chemother. 9(Suppl. 1):25-35.
10
10. Donabedian, S. M., L. A. Thal, E. Hershberger, M. B. Perri, J. W. Chow, P. Bartlett, R. Jones, K. Joyce, S. Rossiter, K. Gay, J. Johnson, C. Mackinson, E. Debess, J. Madden, F. Angulo, and M. J. Zervos. 2003. Molecular characterization of gentamicin-resistant enterococci in the United States: evidence of spread from animals to humans through food. J. Clin. Microbiol. 41:1109-1113.[Abstract/Free Full Text]
11
11. Gabrielson, J., M. Hart, A. Jarelov, I. Kühn, D. McKenzie, and R. Möllby. 2002. Evaluation of redox indicators and the use of digital scanners and spectrophotometer for quantification of microbial growth in microplates. J. Microbiol. Methods 50:63-73.[CrossRef][Medline]
12
12. Goodyear, K. L. 2002. Veterinary surveillance for antimicrobial resistance. J. Antimicrob. Chemother. 50:612-614.[Free Full Text]
13
13. Johnson, T. L., B. A. Forbes, M. O'Connor-Scarlet, A. Machinski, and K. D. McClatchey. 1985. Rapid method of MIC determinations utilizing tetrazolium reduction. Am. J. Clin. Pathol. 83:374-378.[Medline]
14
14. Jones, R. N. 2001. Resistance patterns among nosocomial pathogens: trends over the past few years. Chest 119:397S-404S.[Abstract/Free Full Text]
15
15. Jorgensen, J. H. 1993. Selection criteria for an antimicrobial susceptibility testing system. J. Clin. Microbiol. 31:2841-2844.[Free Full Text]
16
16. Katla, A. K., H. Kruse, G. Johnsen, and H. Herikstad. 2001. Antimicrobial susceptibility of starter culture bacteria used in Norwegian dairy products. Int. J. Food Microbiol. 67:147-152.[CrossRef][Medline]
17
17. Korgenski, E. K., and J. A. Daly. 1998. Evaluation of the BIOMIC video reader system for determining interpretive categories of isolates on the basis of disk diffusion susceptibility results. J. Clin. Microbiol. 36:302-304.[Abstract/Free Full Text]
18
18. Kühn, I., G. Allestam, T. A. Stenstrom, and R. Möllby. 1991. Biochemical fingerprinting of water coliform bacteria, a new method for measuring phenotypic diversity and for comparing different bacterial populations. Appl. Environ. Microbiol. 57:3171-3177.[Abstract/Free Full Text]
19
19. Meletiadis, J., J. F. Meis, J. W. Mouton, J. P. Donnelly, and P. E. Verweij. 2000. Comparison of NCCLS and 3-(4,5-dimethyl-2-thiazyl)-2, 5-diphenyl-2H-tetrazolium bromide (MTT) methods of in vitro susceptibility testing of filamentous fungi and development of a new simplified method. J. Clin. Microbiol. 38:2949-2954.[Abstract/Free Full Text]
20
20. Möllby, R., I. Kühn, and M. Katouli. 1993. Computerized biochemical fingerprinting — a new tool for typing of bacteria. Rev. Med. Microbiol. 4:231-241.
21
21. Murray, P. R., A. C. Niles, and R. L. Heeren. 1987. Comparison of a highly automated 5-h susceptibility testing system, the Cobas-Bact, with two reference methods: Kirby-Bauer disk diffusion and broth microdilution. J. Clin. Microbiol. 25:2372-2377.[Abstract/Free Full Text]
22
22. National Committee for Clinical Laboratory Standards. 2000. Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically. Approved standard M7-A5, vol. 20, no 2. National Committee for Clinical Laboratory Standards, Wayne, Pa.
23
23. Saiman, L., J. L. Burns, D. Larone, Y. Chen, E. Garber, and S. Whittier. 2003. Evaluation of MicroScan Autoscan for identification of Pseudomonas aeruginosa isolates from cystic fibrosis patients. J. Clin. Microbiol. 41:492-494.[Abstract/Free Full Text]
24
24. Shea, K. M. 2003. Antibiotic resistance: what is the impact of agricultural uses of antibiotics on children's health? Pediatrics 112:253-258.[Abstract/Free Full Text]
25
25. Tengerdy, R. P., J. G. Nagy, and B. Martin. 1967. Quantitative measurement of bacterial growth by the reduction of tetrazolium salts. Appl. Microbiol. 15:954-955.[Medline]
26
26. van den Bogaard, A. E., and E. E. Stobberingh. 2000. Epidemiology of resistance to antibiotics. Links between animals and humans. Int. J. Antimicrob. Agents 14:327-335.[CrossRef][Medline]
27
27. Walker, R. A., A. J. Lawson, E. A. Lindsay, L. R. Ward, P. A. Wright, F. J. Bolton, D. R. Wareing, J. D. Corkish, R. H. Davies, and E. J. Threlfall. 2000. Decreased susceptibility to ciprofloxacin in outbreak-associated multiresistant Salmonella typhimurium DT104. Vet. Rec. 147:395-396.[Free Full Text]
28
28. Walker, R. D., and C. Thornsberry. 1998. Decrease in antibiotic susceptibility or increase in resistance? J. Antimicrob. Chemother. 41:1-4.[Free Full Text]

---

Applied and Environmental Microbiology, April 2004, p. 2398-2403, Vol. 70, No. 4
0099-2240/04/$08.00+0     DOI: 10.1128/AEM.70.4.2398-2403.2004
Copyright © 2004, American Society for Microbiology. All Rights Reserved.

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Rahman, M. Articles by Möllby, R. Search for Related Content PubMed PubMed Citation Articles by Rahman, M. Articles by Möllby, R. Agricola Articles by Rahman, M. Articles by Möllby, R.