Antimicrobial Drug Susceptibility Testing:

ANTMCROBAL AGENTs AND CHEMOTHERAPY, Jan. 1978, p. 61-69 pyright 1978 American Society for Microbiology Vol. 13, No. 1 Printed in U.S.A. Evaluation of the Micro-Media System for Quantitative Antimicrobial Drug Susceptibility Testing: a llaborative Study A. L. BARRY,1 * R. N. JONES,2 AND T. L. GAVAN3 Clinical Microbiology Laboratories, University of California (Davis)-Sacramento Medical Center, Sacramento, California 958171; Department ofpathology, Kaiser Foundation Hospital Laboratories, Portland, Oregon 972172; and Department ofmicrobiology, The Cleveland Clinic Found:ation, Cleveland, Ohio 44163 Received for publication 5 April 1977 Micro-Media Systems (MMS) has developed a procedure by which microdilution trays can be filled with dilutions of antimicrobial agents, frozen, and distributed to clinical laboratories. The trays are prepared in various distribution centers throughout the United States to supply clinical laboratories in the vicinity of each center. For use, trays are removed from the freezer, allowed to thaw, and then inoculated with inocula prepared as for any other susceptibility test, using a convenient disposable inoculator (ca. 5,ul per well). A collaborative study was planned to evaluate microdilution trays prepared in three Micro- Media Systems distribution centers. Microdilution minimal inhibitory concentrations (MCs) were compared to standard tube dilution tests (the international collaborative study group method). With gram-positive cocci, the two techniques gave essentially uivalent results. With gram-negative bacilli, the microdilution MCs were generally one doubling dilution lower than the standard tube dilution MCs. Similar results were seen with microdilution trays prepared with a oke Dynatech MC 2. nter- and intralaboratory reproducibility with the macro- and microdilution techniques were quite satisfactory, i.e., at least 96% of the end points were within a range of + 1 log2 dilution intervals. Quantitative broth dilution susceptibility necessary controls. Micro-Media Systems tests can be performed efficiently with a microdilution technique (1). By using the semiauto- centers across the United States, which prepare (MMS) has developed a number of distribution mated uipment now available, the wells in dilutions of antimicrobial agents, fill microdilution trays, and distribute the trays, frozen, microdilution trays can be filled and then stored frozen for at least several weeks (2, 6). to clinical laboratories in the near vicinity of Each well in a thawed tray may be inoculated each center. n a central laboratory, concentrated stock solutions of each drug are pre- with a multiple-inoculum replicator. After appropriate incubation, the minimal inhibitory pared, frozen, and distributed to the preparation centers where they are further diluted and concentration (MC) may be determined for each antimicrobial agent. Microdilution technology adapts the standard broth dilution tests the trays are frozen at -7 C and then held at dispensed into microdilution trays. Once filled, by decreasing the total volume of antimicrobial -2 C for as long as 6 days. They can be agent-containing broth from 1 or 2 ml in standard test tubes to.1 ml in wells in a disposable where they are held at -2 C, ready for use. delivered frozen to contracting laboratories plastic tray. Other investigators have demonstrated that it is possible to obtain essentially subcultured to a broth medium and the inocula n the clinical laboratory, test strains are uivalent results with microdilution and standard macrodilution techniques (6, 7, 8). test. The trays are then inoculated with a are standardized as for any other susceptibility Many clinical laboratories find it impractical convenient disposable inoculator capable of delivering about 5,ul to each well. After overnight or impossible to maintain stock solutions of antimicrobial agents, to fill microdilution trays incubation, the trays are examined with the on a regular basis, and to maintain all of the aid of a viewing box, and the MCs are deter- 61 Downloaded from http://aac.asm.org/ on May 4, 218 by guest

62 BARRY, JONES, AND GAVAN mined as the lowest concentration of each drug that completely inhibits visible growth of the test organism. The procedure is essentially an adaptation of the standard broth dilution technique to a more convenient microdilution tray, by miniaturization. With the MMS system, the antimicrobial agent is subjected to several freeze-thaw cycles, and the filled trays are stored for as long as 6 days. Because deterioration of antimicrobial drugs may alter the final MC, especially at the lower concentrations, quality control organisms must be included with each batch of tests. A collaborative study was planned to evaluate samples of MMS antimicrobial microdilution trays prepared in three distribution centers (Campbell, Calif.; Portland, Oreg.; and Cleveland, Ohio). The microdilution MCs were compared to standard broth macrodilution tests as described by Ericsson and Sherris (5) in a report from an international collaborative study group (the CS method). Limited studies were also undertaken to compare microdilution MCs obtained with the MMS trays to those prepared with a Dynatech MC 2 (oke Engineering., Alexandria, Va.). MATERALS AND METHODS Test strains. Sixteen bacterial isolates were each tested by two methods on 3 separate days in each of the three participating laboratories. Tables 1 and 2 list the test strains and the MC range of each antimicrobial drug tested by both the CS technique and microdilution method. Strains designated with prefix K were obtained from Kaiser Foundation Hospital Laboratory (Oregon region), and all others were derived from ATCC stock cultures. These microorganisms were selected to maximize the number of strains with end points that fell within the seven dilution-step range of concentrations provided in the MMS microdilution trays. Only end points in wells 2 through 6 would be expected to give "onscale" end points if the technical variability in MCs was ±1 log2 dilution step. For each antimicrobial agent, at least three on-scale end points were obtained, and for all but three drugs (clindamycin, methicillin, and trimethoprim/sulfamethoxazole) at least one strain gave an end point at the lower concentrations (well 5 or 6). This is important since the lower concentrations of the antimicrobial drug might be more likely to lose activity during prolonged storage and reflect minor errors in the preparation of serial dilutions. n unsupplemented Mueller-Hinton broth, the currently recommended control strain of Pseudomonas aeruginosa (ATCC 27853) gave MCs with gentamicin, tobramycin, and kanamycin which were all "off scale." These results were not included in the analysis. A total of 63 drug-organism combinations were analyzed, with nine microdilution MCs and nine macrodilution ANTMCROB. AGENTS CHEMOTHER. MCs for each combination, i.e., each of three laboratories tested each strain on 3 separate days. CS macrodilution tests. The standard tube dilution reference method was the broth dilution technique outlined by Ericcson and Sherris (5). Each of the participating laboratories received Mueller-Hinton broth (Difco Laboratories, Detroit, Mich., control no. 628274) for the macrodilution tests. Also, all three participants received frozen portions of concentrated stock solution of each antimicrobial agent to be tested (most were 1,28 jag/ml; cephalothin and kanamycin were prepared at 2,56,ug/ml, and carbenicillin was 2,48 jag/ml). These stock solutions were prepared at Micro-Media Laboratories (Campbell, Calif.) from standard powders provided by the Food and Drug Administration as reference standards for assay work. Before distribution, each stock solution was tested by the bioassay technique of Bennett et al. (4), and the investigators were instructed to dilute the solutions by the assayed value. Reproducibility of end points in the three laboratories confirmed the absence of significant deterioration of the stock solutions in transit. All three participants prepared serial dilutions of the antimicrobial agents by the same dilution protocol, as outlined in Table 21 (p. 66) of Ericsson and Sherris (5). To prepare the inoculum, logarithmicphase broth cultures were adjusted to give a turbidity matching that of a MacFarland.5 standard and were then further diluted 1:5 in Mueller- Hinton broth (.5 ml + 25 ml of Mueller-Hinton broth). The final inoculum density was about 1 x 15 colony-forming units per ml, as confirmed by colony counts; the inoculum is comparable to that used with the microdilution tests. The tests were incubated at 35 C for 16 to 18 h and then read for the presence or absence of turbidity. Microdilution susceptibility tests. Samples from one randomly selected production lot of microdilution trays were provided by the MMS distribution center closest to each participating laboratory, and all tests were completed within 3 weeks after manufacture. Additional samples from 1 different production lots of trays were tested 1 to 2 weeks after production and again after 6 days of storage at -2 C or less. One investigator (T.L.G.) included tests with microdilution trays prepared with a oke Dynatech MC 2 (Alexandria, Va.). These trays were inoculated with a oke semiautomated inoculator using a fixed dilution (.3 ml in 2 ml of water) of brain heart infusion broth culture (5 to 6 h,.5 ml), following the principle described for standardizing the inoculum for the agar overlay disk technique (3). The inoculum for the MMS microdilution tests was prepared from the same cell suspension used to inoculate the macrodilution tests. A logarithmic-phase culture was adjusted to match a MacFarland.5 turbidity standard and then diluted 1:5 in 25 ml of sterile water with.2% Tween 8. Each tray was then inoculated with a separate disposable inoculator that delivers ca. 5 ;L to each well. About 1 x 14 viable cells were delivered to each well, which contained ca..1 ml of broth; the final concentration of cells would Downloaded from http://aac.asm.org/ on May 4, 218 by guest

VOL. 13, 1978 thus be uivalent to that achieved with the CS method. After 16 to 18 h of incubation at 35 C, the trays were examined on an MMS viewing box, and the MC was recorded as the lowest concentration that completely inhibited growth of the test organism. Both micro- and macrodilution end points were independently read by two microbiologists, and when a discrepancy occurred, a third reader arbitrated. Discrepancies between readers rarely occurred. One investigator (R.N.J.) estimated the actual volume of broth in microdilution trays by aspirating the contents of each well, using a 5-Al Unimetric syringe, which was calibrated against 5-ul disposable glass capillary pipettes, to measure wthin ±3% at 1 pul. RESULTS Tube versus microdilution. The modes and ranges of MCs determined by the two dilution tests are recorded in Tables 1 and 2. Each macrodilution MC was compared directly to the matching microdilution MC, and the differences were expressed as an MC ratio (CS tube method/mms microdilution method). f the MC values of the two methods were identical, the ratio was 1; if the CS method gave larger MC values, the ratio was 2, 4, 8, etc.; and if the CS method gave lower MC values, the ratio was.5,.25, etc. Data submitted by all three participants are summarized in Table 3. n 31 of the tests, one of the MC values was beyond the range of concentrations tested; these data were excluded from the analysis since a valid comparison could not be made. f the off-scale end points were included in the calculations, the conclusions would not have been altered appreciably, i.e., if an MC s.25 Ag/ml was considered to be.25 Ag/ml or if an MC >16,ug/ml was considered to be 32 Ag/ml. With the gram-positive cocci, a total of 21 pairs of MC values could be compared; 96.2% of the strains gave MCs that were the same or that differed by ± 1 log2 dilution step. There was a definite trend for the CS method to give MCs about 1 log2 higher than the MMS microdilution trays; 96.5% of the strains gave MC ratios of 1, 2, or 4, i.e., a ratio of 2 ± 1 dilution interval. Somewhat greater disparity was seen with trays tested in California than with those studied in Ohio. A simple experiment was performed to help answer the question of whether the differences between macro- and microdilution MCs with gram-negative bacilli reflect differences in actual potency of the antimicrobial agents in the MMS trays or whether they represent a fundamental characteristic of the microdilution tech- MMS MCRODLUTON MCs 63 nique. Further tests were performed with two strains of Escherichia coli, which demonstrated marked differences between macro- and microdilution MCs with cephalothin and kanamycin. Serial dilutions of both antibiotics were prepared in Mueller-Hinton broth and then distributed into 1.-ml tubes (13 by 1 mm) and.1 ml went into each of five separate wells in a microdilution tray. n this way both micro- and macrodilution tests could be performed as before, but this time the antibiotic solutions were exactly the same. The resulting MCs are recorded in Table 4. The macrodilution CS method consistently gave MC values of about 1 log2 dilution step greater than that obtained with the microdilution technique, as was observed with the MMS microdilution trays. MMS versus oke trays. One of the investigators (T.L.G.) routinely uses microdilution trays prepared with the oke Dynatech MC 2. MCs were determined with the oke trays at the same time the MMS trays were tested. The method of standardizing the inoculum for the two types of trays differed, but the final inoculum density was about the same. Table 5 presents a direct comparison of MCs with the two microdilution techniques and of the CS macrodilution tests with both microdilution techniques. The two microdilution techniques gave essentially the same result with gram-negative bacilli, and both tended to give MCs that were about 1 log2 dilution step lower than those with the macrodilution CS technique. With gram-positive bacteria, both microdilution methods gave MCs that were essentially the same as those with the macrodilution method, and the two microdilution tests gave comparable results. Reproducibility. Because each microorganism was tested on 3 separate days in each of the laboratories, we were able to estimate the inter- and intralaboratory variability of the two dilution techniques. The first, second, and third determination made in each laboratory were compared directly, and the range of differences between the three laboratories is summarized in Table 6. nterlaboratory reproducibility ofresults with both techniques was essentially comparable; if anything, the microdilution test was a little more reproducible than the macrodilution test. The same conclusions would be drawn whether or not off-scale end points were included in the tally. Table 7 summarizes the variability between triplicate values obtained within each of the three independent laboratories on separate days. n all three laboratories, at least 95% of Downloaded from http://aac.asm.org/ on May 4, 218 by guest

6!. m :3 4) X, w w 11 co 14 CA ao 4,-.., - m v m v m v -1.- z,- toe 9- c M c*w 7-1- 9- e V- X io 1q cc4 e -4 ~ 1 2 r. co 9 as 8 u _4..4 s u la s :9 1._.,.~ a.",a._ C) CS 4) c1 c~44 C) o ka *- UX _; c A 13oz C4 e4 _- - Lo ko cq C, R V- Li - - ^11 " V slw C" " ' C 4 V 1e c 4 coc -,to Cs e c 1 c 4 t v " t co 41 - V- 4-1 t 1 '4 *. 1N o -_C;U oq o 1- C4 _1 41 V 1 V cc'q 1 T e - - -V- 1 _- _- _q -- _' ; o _ eo uz1 _ 4 z1z t _ C4 UZ4 Ci e 311 A A A A -4 cq u1 R 'O 1i1 RR. j A C4 C C-4 C4 3C4 11 co - Go - cq CO CO.o 4-4 4) i Downloaded from http://aac.asm.org/ on May 4, 218 by guest - -2,w qc- co 1co l ~~~~~~~c r4 A. 4 4 4 64

C~o. 2i WE _._ C).w r- 7q w t r-4 c4 ", " ecq " " to ;5 o CQ C14 F- E- z z _1, e- zz tt s Z on May 4, 218 by guest. ' CO C.). -6 C.). At C.) 4 1, Ce q C CM Cq CC A A Zs Cq _ 1 ~ ~ ~ ~ ~ ~ _* 4 ~ LO1Lo o &o e1 q _ 1 1 U13 U13 * Lo Lo (6 C; *; 6 zz Q oo 3o > CO Downloaded from http://aac.asm.org/ VOL. 13, 1978 MMS MCRODLUTON MCs 65 L. r. co 1E i cc -, CO w to - C- A - - F- - F- F W -- i " - -,1- -1- as 2 *3 Ce C. CO ~~~~~~~~~e C14 t- m cq t

66 BARRY, JONES, AND GAVAN TABLz 3. mparison of macrodilution (CS) and microdilution (MMS) MCs, matched pairs of tests performed in three separate laboratories MC ratio (% of strains)p Testing laboratory.25.5 1 2 4 8 Gram-negative strains Cleveland 1.8 26.8 55.6 15.7 Portland 4.4 21.2 55.7 17.7.9 Sacramento 2.8 45.8 3.2 3.1 Total 2.2 23. 52.7 2.8 2.5 Gram-positive strains Cleveland 22.2 54.2 23.6 Portland 5.6 72.2 18. 4.2 Sacramento 6.1 35.8 51.5 6. 1.5 Total 11.4 54.3 3.5 3.3.5 a MC ratio, CS/MMS. Data based on 117 MC determinations with gram-negative bacilli and 72 tests with gram-positive cocci. A total of 31 off-scale end points were excluded from these data; inclusion of such data would not appreciably affect the results. TABLE 4. Macro- versus microdilution MCs with the same dilutions ofantibiotic distributed into tubes and trays and inoculated with dilutions of the same standardized cell suspension w S~~~~~~~~~~~~~~MC (ZAg) MC method E. coli K 38 ATCC 25922 Cep" Kanf Cep Kan Macrodilution Microdilutionb 32 16 (8-16) 8 6c (2-8) 16 8 (8-8) 4 2 (1-2) a Cep, Cephalothin; Kan, kanamycin. b Values represent the mode of five separate determinations on one tray; range is noted in c parentheses. Mode of 6 = same number of values at 4 and at 8. TABLE 5. ANTMCROB. AGENTS CHEMOTHZR. mparison of macrodilution (CS) and microdilution (MMS and oke) MCs, matched pairs of tests performed at the Cleveland Clinical Foundationa Methods compared (no. of MC ratio (% of testsp tests).12.25.5 1 2 4 8 Gram-negative strains CS/MMS (18) 1.9 26.9 56.5 14.8 CS/oke (14) 4.8 8.7 24. 43.3 18.3 1. MMS/oke (1) 5. 6. 16. 51. 22. Gram-positive strains CS/MMS (72) 23.6 52.8 23.6 CS/oke (58) 5.2 32.8 36.2 22.4 3.4 MMS/oke (59) 1.7 28.8 49.2 15.3 5.1 a Microdilution trays were either provided by MMS (Cleveland) or prepared at the Cleveland Clinic Foundation with a oke MC 2 dispenser. b Excluding off-scale values. Downloaded from http://aac.asm.org/ on May 4, 218 by guest the triplicate MC determinations were reproducible within 1 log2 dilution interval, and all but one MC value fell within a range of 2 dilution steps (mode, + 1 dilution). To further test lot-to-lot reproducibility of MMS microdilution trays, additional samples of trays from 1 different production lots were tested in triplicate 1 to 2 weeks after preparation and again after 6 days of storage at - 7 C (3 lots) or at -2 C (2 lots). Five lots of trays

VOL. 13, 1978 for gram-positive organisms and five lots of trays for gram-negative organisms were used to test the appropriate reference strains. The trays were all collected from two different MMS distribution centers. We were unable to document any significant differences in the MC end points with different lots of trays, i.e., all MCs were within + 1 dilution interval from MMS MCRODLUTON MCs 67 the expected mode. Furthermore, after 6 days of storage, the trays performed the same as 1- to 2-week-old trays (±1 dilution). Two lots of trays were also tested 2 weeks after the expiration date; a significant loss of activity was not demonstrated by MC end points. Filling accuracy. One investigator (R.N.J.) estimated the volume of broth in the microdi- TABLz 6. nterlaboratory variation in broth dilution MC determinationsa Determination MC range (as 2 x dilutions)' Dilutions 1 Dilution 2 Dilutions 3 Dilutions Microdilution (MMS) lst test 19 (19)" 37 (35) 5 (3) 2 () 2nd test 24 (24) 27 (26) 11 (7) 1 (1) 3rdtest 17 (17) 31 (27) 12 (8) 3 (1) Total no. 6 (6) 95 (88) 28 (18) 6 (2) Total% 32 (36) 5 (52) 15 (11) 3 (1) Macrodilution (CS) lst test 2 (2) 31 (3) 1 (8) 2 (2) 2ndtest 18 (17) 31 (29) 12 (12) 2 (1) 3rd test 9 (9) 38 (37) 13 (11) 3 (3) Total no. 47 (46) 1 (96) 35 (31) 7 (6) Total % 25 (26) 53 (54) 18 (17) 4 (3) a Range of MC values obtained from three separate laboratories with each organism tested on 3 different days. b Expressed as the number of strains in each category. Strains with end points off scale were considered to have MCs at the lowest concentration tested if there was no growth in that well or at 1 dilution above the highest concentration tested if growth appeared in all wells. c Numbers in parentheses represent the total number of strains in each category when all data with one or more values off scale are excluded. TABLz 7. ntralaboratory variation in broth dilution MC determinationsa Testing laboratory MC range (as 2x dilutions)' Dilutions 1 Dilution 2 Dilutions 3 Dilutions Microdilution (MMS) Cleveland 42 (41)" 18 (15) 3 (1) () Portland 27 (27) 32 (31) 3 (3) 1 () Sacramento 27 (27) 33 (25) 3 (3) () Total no. 96 (95) 83 (71) 9 (7) 1 () Total % 51 (55) 44 (41) 5 (4) 1 () Macrodilution (CS) Cleveland 33 (33) 3 (3) () () Portland 3 (3) 3 (3) 3 (2) () Sacramento 31 (29) 29 (29) 2 (2) 1 (1) Total no. 94 (92) 89 (89) 5 (4) 1 (1) Total % 5 (49) 47 (48) 3 (2) 1 (1) a Range of MC values between triplicate tests in each of three separate laboratories. " Expressed as the number of strains in each category. Strains with end points off scale were considered to have MCs at the lowest concentration tested if there was no growth in that well or at 1 dilution above the highest concentration tested if growth appeared in all wells. c Numbers in parentheses represent the total number of strains in each category when all data with one or more values off scale are excluded. Downloaded from http://aac.asm.org/ on May 4, 218 by guest

68 BARRY, JONES, AND GAVAN lution wells after they thawed for 2 to 3 min at room temperature (Table 8). Samples of trays obtained from three MMS preparation centers contained 64 to 14 JA of broth in each well. The control wells (Hi and 2) contained the least amount of broth. oke trays were provided by E. H. Gerlach (St. Francis Hospital, Wichita, Kans.). They were prepared with a oke MC 2 dispenser and shipped, frozen, to Portland, Oreg. The wells in these trays contained 66 to 123 ul of broth. One tray from each lot was then incubated for 15 to 18 h at 35C, covered with a blank tray. As much as 25% of the volume was lost from the wells along the outer edges of the tray, whereas only 1 to 3% of the broth medium was lost from the wells in the center of the tray. By storing the trays at -2 C without sealing tape, 7.2 to 9.8% of the volume was lost by evaporation, but, when the trays were covered with sealing tape, the moisture loss was negligible. The latter estimate was made by sampling two trays from each of five different lots; one of the trays was covered with sealing tape at the time of manufacture. Both trays were wrapped in plastic bags, frozen at -7 C, and then stored at -2 C for 9 to 49 days. The amount of moisture lost was not related to the length of storage. To estimate the effect of altering the volume of broth over a wide range, special trays were prepared with 5, 1, 15, and 2,ul of broth containing the various antimicrobial agents. The actual concentration of drug was not changed, but the volume was. These trays were inoculated with the appropriate test strains, and the MCs were determined. n all cases, the MCs varied no more than 1 dilution interval from the median, regardless of the volume of broth in the wells. Finally, sample trays TABLz 8. Estimated volume ofbroth in microdilution trays prepared in four different laboratories Prepn No. of wells Mean vol per well (pl) center sampled M MMS Calif. 1 93. (8-99P Oreg. 1 9.7 (64-14) Ohio 6 86. (74-96) oke Kans. 8 13.7 (66-123) a From each preparation center, two trays from each of three to five lots were sampled and 1 wells were sampled from each tray, following a prescribed diagonal pattern. b Numbers in parentheses represent range. ANTimicRoB. AGENTS CHEMOTHZR. (uncovered) were placed in an incubator at 35 C for 4.5 h, at which time about 4% of the volume was lost by evaporation and the drug concentration increased almost twofold. These trays gave MCs that were ual to or 1 dilution step lower than that obtained with the control trays that had not been subjected to such excessive evaporation. DSCUSSON Microdilution susceptibility tests are performed routinely in an increasing number of clinical laboratories. f each laboratory undertakes the responsibility for preparing stock solutions of antimicrobial agents, filling microdilution trays, and controlling the storage conditions with aduate quality control procedures, there should be a reasonable degree of inter- and intralaboratory reproducibility. Even greater reproducibility should be expected if the antimicrobial dilutions are prepared in a central laboratory and dispensed into microdilution trays in large batches, with more stringent controls, than those that can be applied in the average clinical laboratory. n this sense, the MMS system offers an important step in the direction of standardization; each contracting laboratory would theoretically receive trays with drugs of nearly identical potency. n the present study, different lots of microdilution trays prepared in three geographically separate distribution centers performed essentially the same. n fact, there was less interlaboratory variation with the MMS microdilution trays than with the standard CS macrodilution technique; even though all tube dilution tests were prepared from the same concentrated stock solutions of antimicrobial agents. ntralaboratory variation in MCs was quite satisfactory with both dilution techniques. Both the MMS microdilution trays and those prepared with a oke Dynatech MC 2 gave essentially the same end points, confirming the acceptability of the MMS microdilution trays. With the gram-positive cocci, the microdilution MCs were essentially uivalent to those obtained with the standard CS macrodilution technique. However, with the gram-negative bacilli, the microdilution MCs were about 1 dilution step lower than those obtained with the standard tube method. There was no clear tendency for such discrepancies to occur with one particular group of drugs or microorganisms: it appears to be a general phenomenon for most gram-negative bacilli. t would be difficult to believe that the discrepancies with trays designed for testing gram-negative organ- Downloaded from http://aac.asm.org/ on May 4, 218 by guest

VOL. 13, 1978 isms resulted from the use of excessively high concentrations of drugs at all three MMS preparation centers because the gram-positive panels did not show the same phenomenon. n addition, the data reported in Tables 5 and 8 support the fact that we observed a general phenomenon common to the microdilution technique, regardless of how the trays are prepared (MMS, oke MC 2, etc.). We can suggest two factors that might contribute to such discrepancies between the two methods. At concentrations just below the MC, bacterial growth was often reduced, but not completely inhibited. n the test tube, a faint, barely visible turbidity was often seen at one concentration of drug, and complete inhibition occurred at the next higher log2 dilution. Because of the optics involved in reading end points in microdilution trays, the barely visible turbidity might not be seen, and thus the microdilution MCs would be 1 log2 dilution step lower than the tube dilution MC. n addition, minor differences in the actual inocula densities might contribute some variability in end points. The absolute number of viable cells delivered to the microdilution trays was about one-tenth of that delivered to the tubes, but the final concentration of cells should be comparable. From a practical point of view, most techniques that involve serial twofold dilutions are considered satisfactorily controlled if the results vary no more than ± 1 dilution. The macro- and microdilution MCs were uivalent within the acceptable range of + 1 dilution with 85% of the tests (74% of gram-negative strains and 96% of the gram-positive strains). Both inter- and intralaboratory reproducibility with both methods also showed an acceptable MMS MCRODLUTON MCs 69 range of variation (+ 1 dilution). We conclude that the MMS microdilution tests were entirely satisfactory and the MCs were essentially uivalent to those obtained with a standard broth dilution technique. ACKNOWLEDGMENT We express our gratitude to L. J. Effinger, R. E. Badal, R. Packer, M. J. Telenson, S. Diedrich, K. McHale, M. Teplitz, and P. Lieberman for their invaluable assistance in collecting the data summarized in this report. LTERATURE CTED 1. Barry, A. L. 1976. The antimicrobic susceptibility test: principles and practices, p. 95-99. Lea & Febiger, Philadelphia. 2. Barry, A. L., L. J. Effinger, and R. E. Badal. 1976. Short-term storage of six penicillins and cephalothin in microdilution trays for antimicrobial susceptibility tests. Antimicrob. Agents Chemother. 1:83-88. 3. Barry, A. L., F. Garcia, and L. D. Thrupp. 197. An improved single-disk method for testing antibiotic susceptibility of rapidly growing pathogens. Am. J. Clin. Pathol. 53:149-158. 4. Bennett, J. V., J. L. Brodie, E. J. Benner, and W. M. M. Kirby. 1966. Simplified, accurate method for antibiotic assay of clinical specimens. Appl. Microbiol. 14:17-177. 5. Ericsson, H. M., and J. C. Sherris. 1971. Antibiotic sensitivity testing report of an international collaborative study. Acta Pathol. Microbiol. Scand. Sect. B. (Suppl.) 217, 9 p. 6. Gavan, T. L., and D. A. Butler. 1974. An automated microdilution method for antimicrobial susceptibility testing, p. 88-93. n A. Balows (ed.), Current techniques for antibiotic susceptibility testing. Charles C Thomas, Springfield, ill. 7. Gavan, T. L., and M. A. Town. 197. A microdilution method for antibiotic susceptibility testing: an evaluation. Am. J. Clin. Pathol. 53:88-885. 8. Gerlach, E. H. 1977. Dilution test procedures for susceptibility testing, p. 45-51. n A. Bondi, J. T. Bartola, and J. E. Prier (ed.), The clinical laboratory as an aid in chemotherapy of infectious disease. University Park Press, Baltimore. Downloaded from http://aac.asm.org/ on May 4, 218 by guest