TEXT

Top Abstract Text References

Early detection of emerging resistance mechanisms requires quantitative susceptibility testing, either zone diameter measurements or full panel MICs, as recommended by the ASM Task Force on Antibiotic Resistance (2). However, most commonly used automated methods provide only breakpoint values of antibiotic susceptibility. Single disk diffusion provides more endpoint values, about 35, than most full-panel dilution methods, about 12, but measuring zone sizes is tedious, time-consuming, and fraught with transcription errors. An automated method of measuring zone sizes would obviate these limitations.

In our clinical microbiology laboratory, we evaluated the Sirscan (i2a, Montpelier, France), an automated image analyzer that measures zone diameters and provides a user-programmed expert system that screens the results of each isolate. The program also extrapolates zone sizes to MICs and reports both.

Susceptibility tests. Testing was performed by the Kirby-Bauer single disk diffusion method according to National Committee on Clinical Laboratory Standards (NCCLS) guidelines (4), using 150-mm round plates of Mueller-Hinton agar purchased from BBL, Becton-Dickinson, Cockeysville, Md.

Strains tested. Strains were fresh clinical isolates from the Clinical Microbiology Laboratory of the Miriam Hospital. The number of isolates tested per species are as follows: 101 isolates of Escherichia coli, 74 isolates of coagulase-negative staphylococci, 53 isolates of Staphylococcus aureus, 28 isolates of Pseudomonas aeruginosa, 25 isolates of Klebsiella pneumoniae, 15 isolates of Serratia marcescens, 13 isolates of Klebsiella oxytoca, 13 isolates of Proteus mirabilis, 12 isolates of Enterobacter cloacae, 8 isolates of Citrobacter freundii, 8 isolates of Acinetobacter baumanii, 8 isolates of S. maltophilia, 7 isolates of Enterobacter aerogenes, and 1 isolate each of Serratia liquefaciens, Proteus penneri, and Salmonella sp. Thirty-five percent were urine isolates, 21% were from sputum, 19% were from blood, 11% were from wounds, and the remainder came from miscellaneous sources.

Zone diameter readings with a ruler. A technologist experienced in reading zones with a ruler placed the Kirby-Bauer (K-B) plate (150-mm diameter and round) on a view box with reflected light against a black background and measured the zone diameters manually. She dictated the readings to a colleague who recorded the results. When performing repeat readings of multiple plates, the technologist cycled the different plates.

Caliper readings. Six technologists in the clinical microbiology laboratory participated in these studies. They placed the K-B plates on the viewer and measured zones by using a digital caliper (Sylvac; Fowler Tools and Instruments, Boston, Mass.) connected to a foot pedal. When the pedal was depressed, the zone diameter automatically entered a database (WHONET) on a desktop personal computer.

Sirscan readings. After putting the K-B plate on a sliding tray, a keystroke initiates the zone readings. An image of the plate appears on the screen, with the zone diameters encircled by a green (susceptible), yellow (intermediate), or red (resistant) circle. At this point the reader can modify the zone readings. Another keystroke automatically enters the values into the Sirscan database. The timing tests performed on the quality control strains and the three clinical isolates were done without modifying the Sirscan readings. In the prospective study of 368 clinical isolates, the reader did not adjust the Sirscan readings (automatic readings) with the first 114 isolates; the remainder were adjusted as judgment dictated (reviewed readings).

Timing studies. A stopwatch was used to time all readings. Timing began when the technologist commenced reading the zone sizes of the K-B plate on the viewer. With the Sirscan, timing began when the keystroke initiated the reading of the K-B plate in the loading tray.

Time required to measure zone diameters. Table 1 shows the results of readings of clinical isolates of S. aureus, E. coli, and Morganella morganii by the three different methods. The median time to read the 12 disk diameters on the plates with gram-negative bacilli was 39 s by ruler, 28 s by caliper, and 11 seconds by Sirscan.

Zone size variation upon repeated testing. Zone sizes of control strains varied by more than 2 mm in 1 of 340 (0.3%) determinations by Sirscan and 6 of 297 (2.0%) readings by caliper. With three clinical isolates, zone diameters varied more than 2 mm in 4 of 297 (1.3%) readings by Sirscan and 7 of 330 (2.1%) readings by caliper. Variation exceeded 3 mm in 4 of 1,254 (0.3%) readings.

Comparison of zone sizes measured by a Sirscan automated reader with those measured by caliper. Over a 4-month period, 368 fresh clinical isolates were tested prospectively by both methods in the Clinical Microbiology Laboratory. Correlation of zone diameter sizes was very good with penicillin, vancomycin, and ampicillin disks (Fig. 1). There was also good correlation of diameters for oxacillin and amoxicillin-clavulanate, although Sirscan failed to detect light growth around the oxacillin disk in 3 of 131 (2.3%) isolates and around the amoxicillin-clavulanate disk in 2 of 219 isolates (0.9%) (Fig. 2). Correlation of zone sizes was good, but not as tight, for imipenem and cephalothin, perhaps due to species-specific variation in growth around these disks (Fig. 3). In this regard, zone diameters around imipenem disks were 2 to 3 mm larger by Sirscan than by caliper with the control clinical isolates of M. morganii, but not with the E. coli or P. aeruginosa isolates.

View larger version (40K): [in this window] [in a new window]   FIG. 1.   Correlation of zone diameters as measured with a Sirscan image analyzer to caliper readings. Dark lines represent the NCCLS breakpoints. , automatic reading; X, reviewed reading. (A) Penicillin disk with 127 clinical isolates of staphylococci. (B) Vancomycin disk with 127 staphylococci. (C) Ampicillin disk with 209 gram-negative isolates.

View larger version (26K): [in this window] [in a new window]   FIG. 2.   Correlation of zone diameters as measured with a Sirscan image analyzer to caliper readings. , automatic reading; X, reviewed reading. (A) Oxacillin disk with 53 clinical isolates of S. aureus and 74 coagulase-negative staphylococci. (B) Amoxicillin-clavulanate disk with 127 staphylococci and 87 gram-negative bacilli. The NCCLS breakpoints are not shown because they differ for the different groups of bacteria.

View larger version (27K): [in this window] [in a new window]   FIG. 3.   Correlation of zone diameters as measured with a Sirscan image analyzer to caliper readings. Dark lines represent the NCCLS breakpoints. , automatic reading; X, reviewed reading. (A) Imipenem disk with 241 gram-negative isolates. (B) Cephalothin disk with 210 gram-negative isolates.

View larger version (54K): [in this window] [in a new window]   FIG. 4.   Correlation of zone diameters as measured with a Sirscan image analyzer to caliper readings. Dark lines represent the NCCLS breakpoints. , automatic reading; X, reviewed reading. (A) Ciprofloxacin disk with 278 staphylococci and gram-negative bacilli. (B) Aztreonam disk with 241 gram-negative isolates. (C) Trimethoprim-sulfamethoxazole disk with 366 staphylococci and gram-negative bacilli. (D) Nitrofurantoin disk with 115 staphylococci and gram-negative bacilli. Norfloxacin readings are not shown.

View larger version (24K): [in this window] [in a new window]   FIG. 5.   Correlation of zone diameters as measured with a Sirscan image analyzer to caliper readings. Dark lines represent the NCCLS breakpoints. , automatic reading; X, reviewed reading. (A) Erythromycin disk with 127 clinical isolates of staphylococci. (B) Clindamycin disk with 127 clinical isolates of staphylococci.

View larger version (53K): [in this window] [in a new window]   FIG. 6.   Correlation of zone diameters as measured with a Sirscan image analyzer to caliper readings. Dark lines represent the NCCLS breakpoints. , automatic reading; X, reviewed reading. (A) Piperacillin disk with 28 Pseudomonas and 124 other gram-negative isolates. (B) Cefuroxime disk with 209 gram-negative isolates. (C) Cefotaxime disk with 209 gram-negative isolates. (D) Gentamicin disk with 364 isolates of staphylococci and gram-negative bacilli. Mezlocillin readings are not shown.

Discussion. The Sirscan image analyzer reads zone diameters more than twice as rapidly as skilled technologists using a computer interfaced caliper system. Reproducibility of measurements was excellent, within about 1 mm for Sirscan and 3 mm for caliper readings. The latter, however, tended to vary with the technologist, unlike readings made with the Sirscan.