Disk Susceptibility Studies with Cefazolin and Cephalothin

ANTIMICROBiAL AGENTS AND CHEMOTHEMRAPY, Jan. 1974, p. 63-67 Copyright i 1974 American Society for Microbiology Vol. 5, No. 1 Printed in U.SA. Disk Susceptibility Studies with Cefazolin and Cephalothin PAUL ACTOR, JOSEPH GUARINI, JOSEPH URI, JUDITH DICKSON, JOHN F. PAULS, AND JERRY A. WEISBACH Smith Kline & French Laboratories, Philadelphia, Pennsylvania 19101 Received for publication 14 September 1973 Cefazolin and cephalothin disk susceptibility and minimal inhibitory concentration determinations were conducted on 591 clinical isolates. Cefazolin demonstrated superior activity, as shown by lower minimal inhibitory concentrations, and a greater percentage of isolates inhibited in the disk susceptibility test. The cephalothin antibiotic class disk by the standard Bauer-Kirby method failed to detect susceptibility to cefazolin in a significant percentage of Escherchia coli, Enterobacter species, and Enterococcus isolates. A separate cefazolin disk with a susceptibility cut-off point of 18 mm is recommended. An alternative to a separate cefazolin disk would be a reinterpretation of the cephalothin susceptibility disk zone diameters so that it would more adequately predict cefazolin activity. Standard regulatory methods for tfie testing of antibiotic disk susceptibility have been published in the Federal Register (3). Representing the cephalosporin antibiotics is a disk containing 30,g of cephalothin. This disk is to be used in the determination of cephalothin, cephaloridine, cephaloglycine, and cephalexin susceptibilities. The use of such a representative, or class, disk assumes that it will cover adequately the susceptibilities within the antibiotic class. Because our laboratory is in the process of developing a new cephalosporin, cefazolin, we decided to undertake studies to compare the 30-,gg cephalothin disk with a 30-/ig cefazolin disk. The primary purpose of these studies was to determine the adequacy of the cephalosporin class disk in predicting cefazolin response as well as to detect potential differences in the two antibiotic disks. A battery of susceptibility tests comparing the two antibiotic disks was performed against a variety of gram-positive and gram-negative clinical bacterial isolates. In addition, the minimal inhibitory concentrations of cefazolin and cephalothin for each isolate were determined in order to correlate these values with the data obtained in the disk susceptibility studies. MATERIALS AND METHODS Bacterial isolates. Isolates used in these studies were obtained from hospitalized patients and sent to our laboratory from a variety of geographical locations in the United States. These samples included some isolates from patients later treated with cefazolin. Upon receipt, the isolates were classified and tested for response to 12 commercial antibiotics. A single passage of the organism was made, and samples were frozen at -70 C for additional trials. Antibiotic disks. Cephalothin 30-,ug disks were purchased from BBL. The 30-iAg cefazolin disks were prepared in the BBL laboratories under controlled conditions conforming to their commercial production of susceptibility disks. The disks were assayed for stability over a 1-year period in both BBL and our laboratories and found to have no significant loss in potency when stored at 4 C. The assays were performed by the FDA performance plate test method described in the Federal Register for cephalosporin disks (2). Disk susceptibility tests. The procedure employed was essentially that of Bauer and Kirby (1) and later was detailed in the Federal Register (3). Clinical isolates first were grown in Trypticase soy broth, adjusted to a standard optical density, and swabbed on the surface of a petri dish containing Mueller-Hinton agar (BBL). Duplicate disks were placed on the agar surface with sterile forceps. All plates were incubated at 37 C for 18 h. Zones of inhibition that developed were measured with a Fisher-Lilly zone rqader. A duplicate test was carried out on another day to account for day-to-day variation. In each experiment, two stock reference strains, one of Staphylococcus aureus (ATCC 25923) and one of Escherichia coli (ATCC 25922), were included to establish the validity of the experiment. By using the abovedescribed techniques, the acceptable limits for the 30 Ag cephalothin disk is a zone range of 25 to 37 mm for S. aureus and a 18 to 23 mm range for E. coli. Minimal inhibitory concentrations (agar dilution method). The agar dilution tests were carried out in Trypticase soy agar containing 0.5% glucose and buffered to ph 6.0. Stock solutions of cephalothin and cefazolin were prepared in sterile water, and 12 '33 twofold dilutions were added to petri dishes, result-

64 ACTOR ET AL. ANTIMICROB. AG. CHEMOTHER. ing in final antibiotic concentrations ranging from 0.1 to 200 yg per ml of agar. Approximately 104 gramnegative or 106 gram-positive organisms were inoculated in duplicate onto the agar with a Steers replicating apparatus (11). After incubation at 37 C for 18 h, the plates were examined, and the minimal quantity of compound capable of inhibiting growth (MIC) was determined. RESULTS MIC. The median MIC, obtained with cefazolin and cephalothin by the agar dilution method are shown in Table 1. A total of 591 isolates were included in this table. Values of 200 Ag/ml or greater were obtained with both antibiotics against indole-positive Proteus, Enterobacter hafniae, Serratia, and Pseudomonas aeruginosa. Cefazolin was found to give activity equal or superior to cephalothin against all of the other gram-negative organisms studied. The MIC values obtained for cephalothin against staphylococcus were approximately twofold better than cefazolin. A marked difference was observed with the Enterobacter species studied, where cefazolin was seen to be from 4 to 20 times more effective under the conditions used in these tests. Enterobacter aerogenes and Enterobacter liquefaciens seemed particularly susceptible to cefazolin activity. Disk diffusion studies. A total of four values were obtained with each of the 30-,gg cephalosporin disks against each of the isolates (duplicate disks on 2 test days). These four zone diameters were then averaged to give a single TABLE 1. Median minimal inhibitory concentrations of cefazolin and cephalothin against bacterial isolates Organism Escherichia coli 184 1.6 6.3 Klebsiella pneumoniae 57 1.6 3.1 Proteus mirabilis 16 6.3 6.3 Proteus sp. (indole posi- 22 200.0 >200.0 tive) Enterobacter cloacae 94 12.5 50.0 E. aerogenes 21 1.6 12.5 E. liquefaciens 10 6.3 125.0 E. hafniae 5 > 200.0 > 200.0 E. agglomerans 6 18.7 75.0 Salmonella sp. 16 1.6 3.1 Shigella sp. 7 3.1 12.5 Serratia sp. 24 > 200.0 > 200.0 Pseudomonas aeruginosa 15 > 200.0 > 200.0 Enterococci 65 12.5 12.5 Staphylococcus sp. 49 0.4 0.2 value for each isolate with each antibiotic. An estimate (by the least squares method) of the linear relationship between log MIC and zone sizes for the 30-.ug disks was then determined (Fig. 1 and 2). Only those isolates showing a Minimal inhibitory No. of concn (Mg/ml) isolates Cefazolin Cephalothin 10.0- -ENT. CLOACAE ENTEROCOCCI. PROTEUS (INDOL NEG.) SALMONELLAI ENT. AEROGENESI. KLEE. O - \~- 1.0 z E. C 0 N 4c w SA S~~~~~~~~TAPH U.- 1I-------I 0 10 15 20 25 30 35 40 AVERAGE ZONE DIAMETER FOR CLINICAL ISOLATES WITH CEFAZOLIN DISC, mm FIG. 1. Relationship of zone diameters (30-ug disk) and minimal inhibitory concentration values obtained with cefazolin against clinical bacterial isolates. ENTEROCOCC E 10.0- ENT. CLOACAE ENT. AEROGENES * - E. COLI SALMONELLA KLEB. * PROTEUS (INDOL NEG.) 2 0 OI 4 I. 1.0- U * <_ ~~~~~~~~~~STAPH.\ 0.1 0 10 15 20 25 30 35 40 AVERAGE ZONE DIAMETER FOR CLINICAL ISOLATES WITH CEPHALOTHIN DISC. mm FIG. 2. Relationship of zone diameters (30-;&g disk) and minimal inhibitory concentration values obtained with cephalothin against clinical bacterial isolates.

VOL. 5, 1974 CEFAZOLIN AND CEPHALOTHIN DISK SUSCEPTIBILITY 65 positive zone of inhibition with a MIC within the test range were used in the comparison. Because the data were dominated by a few species of bacteria (i.e., E. coli and Enterobacter cloacae), we felt that a comparison based on species averages rather than on data for individual isolates would yield more representative results. Examination of the slopes obtained with the two antibiotics shows that at equivalent MIC values, the cefazolin 30-,ug disk tended to give a zone diameter somewhat less than that of cephalothin; however, there was no statistical difference in the slopes of the regression lines (-0.0604 versus -0.0801 in logarithms) obtained with the two antibiotics. Figure 3 represents a plot of the log MIC for cefazolin plotted against the species average zone diameters for cephalothin. By using the equation given for the regression line, it can be calculated that the zone diameter of 14 mm for cephalothin will, on the average, correspond to a cefazolin MIC of 4.8,ug/ml. This MIC represents a concentration which is readily attainable in serum of patients treated with cefazolin, even with a 250-mg dose. Determination of susceptibility of isolates. The results of the Bauer-Kirby disk susceptibility tests are shown in Table 2, which shows the susceptibility patterns obtained using several parameters of classification for sensitive, intermediate, and resistant. A comparison of the data obtained with the 30-,gg cefazolin and cephalothin disks, by using 18 mm as the susceptibility cut-off point (SCOP), shows a number of significant differences with the various species of clinical isolates. In general, a much higher percentage of the isolates were found to be susceptible to the cefazolin disk. 10.0-: X 1.0- z 0 U. * ENT. CLOACAE SiENTEROCOCCI ENT. AEROGENES * KLEE E. COLI PROTEUS (INDOL NEG.) SALMONELLA I., lb 1 io is5 io STAPH. AVERAGE ZONE DIAMETER FOR CLINICAL ISOLATES WITH CEPHALOTHIN DISC, mm FIG. 3. Relationship of cephalothin zone diameters (30-jAg disk) and cefazolin minimal inhibitory concentration values obtained against clinical bacterial isolates. Notable examples include: E. coli, 95.7 versus 75.8%; E. aerogenes, 85.7 versus 19.1%; and enterococci, 35.4 versus 13.9%. If the parameters for susceptibility for the cephalothin disk are adjusted so that a zone size of > 14 mm is used for susceptibility classification, then the data obtained more closely approximate that obtained with the cafazolin disk when > 18 mm is used for susceptibility. DISCUSSION The MICs obtained with cefazolin and cephalothin (Table 1) are in general agreement with the values reported in the literature (6-10, 13). Differences observed can, in many cases, be attributed to differences in testing conditions such as inoculum size, media, strains of organisms, incubation time, and tube dilution versus agar dilution. The MIC values obtained with enterococcal and enterobacteral isolates are, in general, lower than those reported by other investigators. Reller et al. (7) have shown that the MIC for cefazolin for isolates of E. coli, Proteus mirabilis, Klebsiella, Salmonella, and Shigella were in close agreement when the broth or agar dilution techniques were employed. Some of their Enterobacter strains showed higher MIC values in broth dilution than did the enterococci. The lower MIC values obtained in our studies may, in part, be due to the use of the agar dilution technique. A marked difference in MIC response to cefazolin was observed with the five Enterobacter species studied. E. aerogenes and E. liquefaciens were found to be inhibited at low cefazolin concentration, whereas E. cloacae and Enterobacter agglomerans were inhibited at higher concentrations. The five strains of E. hafniae were completely resistant (MIC > 200 gg/ml) to both cafazolin and cephalothin. Most investigators do not speciate their Enterobacter isolates; therefore, it is difficult to compare our data with literature values. In addition, the Trypticase soy agar medium employed in our studies is buffered at ph 6.0, a condition that tends to result in lower MIC values with the cephalosporin antibiotics. The differences between the two antibiotics obtained with Enterobacter species are also observed in the disk susceptibility tests (Table 2). In any event, patients infected with strains of Enterobacter found to be susceptible to cefazolin were found to respond to cefazolin treatment (12, Gold et al., J. Infect. Dis., in press). The log of the MIC values for cefazolin and cephalothin were plotted against the average zone diameters for each species of clinical isolate studied (Fig. 1 and 2). Only those species

66 ACTOR ET AL. ANTIMICROB. AG. CHEMOTHER. TABLE 2. Percentage of bacterial isolates responding to cefazolin and cephalothin antibiotic disks No. of Cefazolin Diska Cephalothin Diska Cephalothin Disk" Organisms Isolates R R S R S Escherichia coli 184 0.6 3.6 95.7 6.6 17.6 75.8 4.9 1.6 93.5 Enterobacter aerogenes 21 14.3 0 85.7 57.1 23.8 19.1 14.3 23.8 61.9 E. cloacae 94 72.3 11.7 16.0 84.0 13.8 2.1 67.0 12.8 20.2 E. liquefaciens 10 50.0 0 50.0 70.0 20.0 10.0 40.0 20.0 40.0 E. hafniae 5 80.0 0 20.0 80.0 0 20.0 80.0 0 20.0 E. agglomerans 6 83.3 16.7 0 83.3 16.7 0 50.0 33.3 16.7 Proteus, indole negative 16 6.3 0 93.8 6.3 0 93.8 6.3 0 93.8 Proteus, indole positive 22 68.2 4.6 27.3 68.2 13.6 18.2 68.2 0 31.8 Klebsiella 57 0 3.5 96.5 0 8.8 91.2 0 0 100.0 Enterococci 65 13.6 50.1 35.4 29.2 57.0 13.9 3.1 10.8 86.2 Salmonella 16 6.3 12.5 81.3 18.8 0 81.3 6.3 6.3 87.6 Shigella 7 0 14.3 85.7 14.3 0 85.0 14.3 0 85.8 Staphylococcus 49 0 0 100.0 0 0 100.0 0 0 100.0 a Resistant (R), < 14 mm; intermediate (I), 15 to I br, <10 mm; I 11 to 13 mm; S, >14 mm. where positive values were obtained for 10 or more isolates were used to generate the regression lines. The regression lines as calculated by the method of least squares are similar to values previously reported for these antibiotics (5, 13). The zone diameter values obtained with cefazolin are apparently smaller than those obtained with cephalothin at equivalent MIC determinations. This difference in zone size may be due to a difference in diffusion of the two antibiotics. A cefazolin 30-,ug susceptibility disk could employ the same parameters as cephalothin in the Bauer-Kirby disk susceptibility test: susceptible, > 18 mm; intermediate, 15-17 mm; resistant, < 14 mm. A zone size of 18 mm for cephalothin would correspond to a MIC of 10.3 jig/ml on the regression line. Similarly, an 18-mm zone for cefazolin would indicate a MIC of 4.8 jg/ml, a level easily reached in the serum of men dosed intramuscularly with levels as low as 250 mg (6). By using the cefazolin disk with a susceptibility cut-off point of 18 mm, it was seen that 95.7% of the E. coli isolates were susceptible to cefazolin, whereas only 75.8% responded to cephalothin (Table 2). Large differences in susceptibility to the two antibiotics also were observed with E. aerogenes (85.7 versus 19.1%) and Enterococci (35.4 versus 13.9%). With E. coli and E. aerogenes isolates, cefazolin showed a four- to eightfold lower MIC; however, with Enterococci, there was no difference in the median MIC values for the two antibiotics. It should be noted that many of the enterococci fell into the intermediate category with both antibiotics. Cefazolin has been shown to be a useful antibiotic for treatment of enterococcal 17 mm; susceptible (S), >18 mm. infections in humans, and the disk data would tend to support this finding (Gold et al., J. Infect. Dis., in press). Isenberg et al. (4) have reported on a marked difference in response with enterococci with 30-gAg cephalothin and cephacetrile disks. Although cephacetrile consistently gave larger zone diameters in the Bauer-Kirby system, these differences rarely resulted in a change in the susceptibility category of an organism. The cephalothin response (zone diameter measurements) did not predict the cephacetrile response. The responses of individual isolates toward cephaloridine and cephaloglycine also were found to be different. As a result of these findings, we questioned the wisdom of using a single representative of a group of antibiotic agents to measure susceptibility toward all derivatives. A similar conclusion has been expressed by Wick and Preston (13), who worked with three heterocyclic thiomethyl cephalosporins, including cefazolin. They reasoned that the high serum levels attainable plus the extended activity spectrum argue against a class cephalosporin disk. An analysis of the available data, to obtain the regression line relating the logarithm of the cephazolin MIC and cephalothin disk diameter, can be used in reinterpreting the cephalothin zone diameter data so that it will predict cefazolin activity (Fig. 3). Specifically, it appears that the appropriate cephalothin zone diameters for predicting cefazolin susceptibility should be as follows: > 14 mm, susceptible; 11 to 13 mm, intermediate; < 10 mm, resistant. Thus, an alternative to a separate cefazolin susceptibility disk would be to reinterpret the SCOP for

VOL 5, 1974 CEFAZOLIN AND CEPHALOTHIN DISK SUSCEPTIBILITY 67 cephalothin zone diameters to predict cefazolin activity. It would appear that a SCOP of 14 mm would accomplish this end (Table 2). In almost all cases, a SCOP for cephalothin of 14 mm results in a pattern of response similar to that obtained with a cefazolin disk with a SCOP of 18 mm. The cephalothin disk still falls somewhat short of predicting E. aerogenes response; however, a larger than expected percentage of the enterococci are classified as susceptible to cefazolin. It would appear from these studies that the present class cephalosporin (cephalothin) disk fails to predict adequately cefazolin activity, particularly in the case of E. coli and E. aerogenes. In light of the superior serum levels observed in humans with cefazolin (three to four times that of cephalothin) after intramuscular injection and the improved microbiological activity, a separate cefazolin disk would be justified. An alternate approach to the problem would be to reinterpret the cephalothin class susceptibility disk zone sizes so that they adequately predict cefazolin activity. ACKNOWLEDGMENTS We thank Marie E. Knight and Bernard C. Sekula for their excellent technical assistance. Cefazolin was synthesized by the Fujisawa Pharmaceutical Compnay, Osaka, Japan. LITERATURE CITED 1. Bauer, A. W., W. M. Kirby, J. C. Sherris, and M. Turck. 1966. Antibiotic susceptibility testing by a standardized single disc method. Amer. J. Clin. Pathol. 45:493-496. 2. Department of Health, Education, and Welfare, Food and Drug Administration. 1967. Fed. Regist. 32:13221-13223. 3. Department of Health, Education and Welfare, Food and Drug Administration. 1972. Fed. Regist. 37:20527-20529. 4. Isenberg, H. D., B. G. Painter, J. Sampson-Scherer, and M. Siegel. 1973. Clinical laboratory study of cephacetrile and cephalothin against bacteria recently isolated from clinical specimens. Amer. J. Clin. Pathol. 59:700-705. 5. Matsen, J. M., M. J. H. Koepcke, and P. G. Quie. 1970. Evaluation of the Bauer-Kirby-Sherris-Turck single disc diffusion method of antibiotic susceptibility testing. Antimicrob. Ag. Chemother. 1969, p. 445-453. 6. Nishida, M., T. Matsubara, T. Murakawa, Y. Mine, Y. Yokota, S. Kuwahara, and S. Goto. 1970. In vitro and in vivo evaluation of cefazolin, a new cephalosporin C derivative. Antimicrob. Ag. Chemother. 1969, p. 236-243. 7. Reller, L. B., W. W. Karney, H. N. Beaty, K. K. Holmes, and M. Turck. 1973. Evaluation of cefazolin, a new cephalosporin antibiotic. Antimicrob. Ag. Chemother. 3:488-497. 8. Ries, K., M. E. Levison, and D. Kaye. 1973. Clinical and in vitro evaluation of cefazolin, a new cephalosporin antibiotic. Antimicrob. Ag. Chemother. 3:168-174. 9. Seiga, K., K. Yamaji, K. Miyoshi, and M. Minagawa. 1972. Laboratory and clinical studies on cefazolin, a new derivative of semisynthetic cephalosporin. Int. J. Clin. Pharmacol., Therapy, Tox. 62:135-142. 10. Shibata, K., and M. Fujii. 1971. Clinical studies of cefazolin in the surgical field. Antimicrob. Ag. Chemother. 1970, p. 467-472. 11. Steers, E., E. L. Foltz, B. S. Graves, and J. Riden. 1959. An inocula replicating apparatus for routine testing of bacterial susceptibility to antibiotics. Antibiot. Chemother. 9:307-311. 12. Ulmura, R. 1970. Basic and clinical studies of cefazolin in surgical infections. Kagaku Ryoho 18:724-733. 13. Wick, W. E., and D. A. Preston. 1972. Biological properties of three 3-heterocyclic-thiomethyl cephalosporin antibiotics. Antimicrob. Ag. Chemother. 1:221-234.