ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, Oct. 1976, p. 657-663 Copyright 0 1976 American Society for Microbiology
Vol. 10, No. 4 Printed in U.S.A.
Comparison of the Antibacterial Activity of Nine Cephalosporins Against Enterobacteriaceae and Nonfermentative Gram-Negative Bacilli LUDO VERBIST
Department of Microbiology, St. Rafael University Hospital, Leuven, Belgium Received for publication 19 July 1976
The in vitro antibacterial activity of nine cephalosporins (cephalothin, cephaloridine, cephalexin, cefazolin, cefamandole, cefuroxime, cefatrizine, cefoxitin, and cefazaflur) was determined against 344 strains ofEnterobacteriaceae and 99 nonfermentative gram-negative bacilli. Cefamandole, cefazaflur, and cefuroxime were the most active cephalosporins against the Enterobacteriaceae (with the exception of Serratia marcescens). However, cefoxitin was the only cephalosporin that inhibited all 30 S. marcescens strains in a concentration of 16 ,g/ml and was by far the most active compound against selected cephalothin-resistant strains of Escherichia coli, Klebsiella, and Proteus mirabilis. Acinetobacter spp. were inhibited best by cefuroxime, but none of the cephalosporins had appreciable activity against the Pseudomonas spp.
Infections with gram-negative bacilli constitute a major problem, especially among hospitalized patients. An increasing number of these organisms are resistant to most of the currently used antibiotics. Newly developed cephalosporins may offer a safe alternative treatment. A serious drawback to the first clinically
available cephalosporins, cephalothin, cephaloridine, cephalexin, and cefazolin, is that they have only slight activity against Enterobacter spp., Providencia spp., and indole-positive Proteus spp., and Serratia marcescens and Pseudomonas aeruginosa are completely resistant to these antibiotics. However, recently a number of new cephalosporins have been synthesized that seem to have a markedly broader spectrum against gram-negative bacilli. Among the most promising are cefamandole (2, 4, 5, 7, 10), cefuroxime (5, 11, 12), cefatrizine (6, 7, 13, 15), cefoxitin (3, 7-9, 11, 14), and cefazaflur (SKF 59962) (1). In this study the antibacterial activities of these newer cephalosporins against clinical isolates of 344 Enterobacteriaceae and 99 nonfermentative gram-negative bacilli were compared with each other and with that of the older cephalosporins. MATERIALS AND METHODS
Clinical isolates of Enterobacteriaceae and nonfermentative gram-negative bacilli were collected during a 2-month period in St. Raphael, the teaching hospital of the University of Leuven. After identification by standard criteria the following num-
bers of strains were selected for study: 60 Escherichia coli, 55 Klebsiella pneumoniae, 16 Klebsiella oxytoca, 56 Enterobacter spp., 15 Citrobacter freundii, 10 Citrobacter diversus, 62 Proteus mirabilis, 30 Proteus vulgaris, 14 Proteus morganii, 11 Proteus rettgeri, 15 Providencia spp., 30 S. marcescens, 25 Acinetobacter spp., 67 P. aeruginosa, and 7 Pseudomonas spp. The susceptibility of the strains was determined by an agar dilution technique. Overnight cultures in Trypticase soy broth were diluted in sterile Trypticase soy broth and inoculated with an automatic multipoint inoculator (16) to deliver an inoculum of 104 to 105 colony-forming units. Serial twofold dilutions of the cephalosporins were prepared in sterile distilled water. After the addition of one part of each dilution to nine parts of melted and cooled Diagnostic Sensitivity Test agar (DST Oxoid), the final concentration in the petri dishes ranged from 128 to 0.25 ug/ml. The plates were used on the day of preparation or within 24 h. Pure drugs were obtained from the manufacturers: cephalothin, cephaloridine, cefazolin, cephalexin, and cefamandole from Eli Lilly & Co.; cefuroxime from Glaxo; cefatrizine from Bristol Laboratories; cefoxitin from Merck, Sharp and Dohme; and cefazaflur (SKF 59962) from RIT (Belgium). Control drug-free plates were similarly inoculated. For Proteus strains, p-nitrophenyl-38-D-glucoside in a final concentration of 50 ,ug/ml was added to the antibiotic plates and to control plates to avoid swarming. After incubation at 37°C for 18 h, the minimal inhibitory concentration (MIC) was recorded as the lowest concentration of antibiotic that completely inhibited growth; the presence of three or less colonies was disregarded. Clinical susceptibility to an antibiotic was arbitrarily defined as inhibition of strains at or below half the peak levels of antibiotics achieved after the
657
658
ANTIMICROB. AGENTS CHEMOTHER.
VERBIST
usual dose and way of administration in clinical use studies with volunteers. The following MICs considered as limits of susceptibility: cephalothin, 8 ,ug/ml; cephaloridine, 16 gg/ml; cephalexin, 16 ,ug/ml; cefazolin, 16 ug/ml; cefamandole, 16 ,ug/ ml; cefuroxime, 16 ,ug/ml; cefatrizine, 8 jig/ml; cefoxitin, 16 ,g/ml; and cefazaflur, 16 ug/ml. For the calculation of the geometric means of inhibition, strains resistant to 128 ,ug/ml were considered as having an MIC of 256 ,ug/ml.
or in were
RESULTS The percentage of strains susceptible to the different cephalosporins, according to the criteria defined above, is summarized in Table 1. The species have been divided into three groups based upon their susceptibility pattern to cephalothin and cephaloridine. Group 1 comprises the species of which the majority of isolates are susceptible to these antibiotics, namely, Escherichia coli, Klebsiella spp., P. mirabilis, and C. diversus. Group 2 contains the species with only a minority of isolates susceptible to cephalothin and cephaloridine, namely, Enterobacter spp., indole-positive Proteus spp., Providencia spp., Acinetobacter spp., and C. freundii. Group 3 includes Pseudomonas spp. and S. marcescens, which are uniformly and highly resistant to these drugs.
In group 1 there were only minor differences in the percentage of susceptibility to the different cephalosporins at the critical concentrations. Cefoxitin inhibited 99.5% ofthe strains at 16 ,g/ml. Most other cephalosporins inhibited from 98.0 to 92.6% ofthe isolates, except cephaloridine (85.2%) and cephalothin (78.8%). The species of group 2 were clearly more susceptible to the newer cephalosporins than to cephalothin, cephaloridine, cephalexin, and cefazolin. The highest percentages of susceptibility were found toward cefamandole (83.1%), cefuroxime (74.7%), and cefazaflur (67.1%). Cefoxitin inhibited 55.4% of the strains and cefatrizine (42.2%) inhibited a percentage still substantially higher than that of the older cephalosporins (from 10.8 to 31.9%). Cefoxitin had a low score against Enterobacter spp. (18% susceptible) and against C. freundii (20%), but was the most active compound against the indole-positive Proteus and Providencia spp. Cefazaflur and cefatrizine were only moderately active against the Acinetobacter spp. The Pseudomonas strains remained resistant to the newer cephalosporins. In contrast all 30 isolates of S. marcescens were susceptible to 16 ,ug of cefoxitin per ml, and 13 isolates were also susceptible to 16 ,ug of cefamandole per ml. A total
TABLE 1. Susceptibility of gram-negative bacilli to different cephalosporins 9% of strains inhibited at an MIC (Ag/ml) of at least: Species
Group 1 E. coli K. pneumoniae K. oxytoca C. diversus P. mirabilis Total
test 60 55 16 10 62
203
Total
56 15 30 11 14 15 25
166
|
16
16
16
16
(CL?')
(CLD)
(CLX)
(CZL)
(CMD) (CFU)
73 69 87 70 92
90 78 81 80 89
93 100 94 90 92
98 100 87 90 95
100 98 94 90 98
78.8
85.2
94.6
96.6
11 7 10 18 0 13 16
16 7 10 27 7 13 24
25 7 37 27 7 33 24
37 47 23 27 7 53 24
10.8
15.1
24.7
31.9
Group 2 Enterobacter spp. C. freundii P. vulgaris P. rettgeri P. morganii Providencia spp. Acinetobacter spp.
16
8
98.0 95 80 77 91 100 100 44
83.1
95 100 94 90 97 96.6
86 93 43 55 86 100 64
(CFO)
16 (CZF)
90 95 87 90 92
98 100 100 100 100
92 100 93 80 97
92.6 54 53 37 36 64 47 24
74.7
Group 3 0 0 0 0 3 43 30 S. marcescens 0 0 0 0 0 0 P. aeruginosa 67 0 0 0 0 0 0 7 Pseudomonas spp. a CLT, Cephalothin; CLD, cephaloridine; CLX, cephalexin; CZL, cefazolin; cefuroxime; CTZ, cefatrizine; CFO, cefoxitin; CZF, cefazaflur.
8 (CTZ)
45.2 0 0 0
16
99.5 18 20 100 64 100 100 52
55.4 100 0 0
96.6 74 80 80 55
50 93 25 67.1 3 0 0
CMD, cefamandole; CFU,
VOL. 10, 1976
ANTIBACTERIAL ACTIVITY OF CEPHALOSPORINS
of 18 of the 30 S. marcescens isolates were resistant to 100 ug of carbenicillin per ml. Cefamandole inhibited 10 of the 12 strains susceptible to carbenicillin, but only 3 of the 18 strains resistant to carbenicillin. The distribution of the susceptibility to the different cephalosporins, expressed as the cumulative percentage of strains inhibited at increasing concentrations, is shown in Tables 2 and 3. Table 2 summarizes the susceptibility pattern of E. coli, Klebsiella, P. mirabilis, Enterobacter spp., and the indole-positive Proteus and Providencia spp. Cefamandole and cefazaflur inhibited a large percentage of the E. coli, Klebsiella, and P. mirabilis strains already at 0.25 and 0.5 ,ug/ml, but the remaining isolates were inhibited over a wide range of MIC values. Cefoxitin, in contrast, inhibited the strains from a concentration of 1 ,.g/ml, but generally over a shorter range of MIC values. Cefamandole and to a lesser extent cefazaflur also inhibited part of the strains of the Enterobacter spp. and the indole-positive Proteus spp. and Providencia spp. at the lowest concentrations. Cefuroxime was the most active cephalosporin against the Acinetobacter spp., followed by cefoxitin and cefamandole (Table 3). Cefoxitin was clearly the most active compound against S. marcescens, and cefamandole was active in part of these strains. However, against the Pseudomonas spp. there was scarcely any activity of the cephalosporins. Table 4 shows the geometric means of the MIC values of the different cephalosporins. The newer cephalosporins had clearly lower mean MICs than did the older ones. Against the species of group 1, the lowest mean MICs were obtained with cefamandole (0.8 ,ug/ml), cefazaflur (0.9 ,ig/ml), and cefuroxime (1.5 jig/ml). Against the species of group 2, the same cephalosporins had the lowest mean MIC values: cefamandole, 3.1 jig/ml; cefuroxime, 7.8 ,ug/ml; and cefazaflur, 10.1 ,g/ml. Cefoxitin took only the fourth place as a whole (14.1 jig/ml) due to its weaker activity against Enterobacter spp. and C. freundii. However, against the indolepositive Proteus spp. and the Providencia spp., cefoxitin was the second best, with a mean MIC of 3.1 jig/ml, cefamandole being most effective with a mean MIC of 2.3 jig/ml. The activity of cefoxitin against S. marcescens, with a mean MIC of 10.4 ,g/ml, was similar to the activity of this antibiotic against the species of group 2 (namely, 10.1 ,ug/ml). Since the activity of all cephalosporins was tested simultaneously on the same inoculum of each species, a direct comparison between the MICs of each cephalosporin and the MIC of cephalothin as a refer-
659
ence was possible for each isolate. Therefore, the linear regression lines of the log MICs between each cephalosporin and cephalothin were calculated for these species, whereby more than half of the isolates were at least susceptible to 128 ,ug of cephalothin per ml (resistance to 128 ,ug/ml was considered as susceptibility to 256 ,ug/ml for convenience in the calculations). In the other species, with more than half of the strains resistant to 128 ,ug of cephalothin per ml, the lack of an exact end point of inhibition made the calculation of regression lines less valuable. Table 5 shows the correlation coefficients for E. coli, Klebsiella spp., P. mirabilis, Enterobacter spp., C. freundii, P. vulgaris, and Providencia spp. between the log MICs of each cephalosporin and cephalothin. A perfect correlation (coefficient r = 1.0) could not be expected. Therefore, the statistical difference was calculated between the observed r values and a correlation coefficient of at least 0.90. The correlation coefficients of cephaloridine, cefazolin, cephalexin (with the exception of the Klebsiella spp.), and cefatrizine (except the Enterobacter spp.) toward cephalothin were not statistically different from r = 0.90. This means that higher or lower MIC values with one of these cephalosporins corresponded very well with higher or lower MICs with cephalothin. In other words, isolates resistant to cephalothin were as a rule less susceptible or even resistant to these other cephalosporins. In contrast, especially cefoxitin and to a lesser extent cefuroxime had very poor correlation coefficients with cephalothin. Cefamandole took a place in between, and cefazaflur tended rather to a parallel susceptibility with cephalothin. Table 6 shows the geometric mean MICs of the different cephalosporins according to the susceptibility of the isolates to cephalothin at a concentration of 16 ,g/ml. The ratio between the geometric mean MIC of the susceptible isolates and of the cephalothin-resistant isolates reflects the degree of cross-resistance or crosssusceptibility with cephalothin. This ratio was 1:26 for cephalothin in the species of group 1 (E. coli, Klebsiella spp., and P. mirabilis). The ratios for cefazolin, cephaloridine, cefamandole, cefatrizine, and cefazaflur were very similar and varied from 1:15 to 1:28. In the species of group 2 (Enterobacter, C. freundii, indole-positive Proteus, Providencia, and Acinetobacter) the ratio between the geometric mean MIC of susceptible and resistant isolates was 1:15 for cephalothin and between 1:8 and 1:19 for cephalexin, cefazaflur, cephaloridine, cefatrizine, and cefazolin. The susceptibility to cefoxitin
660
ANTIMICROB. AGENTS CHEMOTHZR.
VERBIST
TABLE 2. Activity ofdifferent cephalosporins against E. coli, Klebsiella spp., P. mirabilis, Enterobacter spp., and indole-positive Proteus and Providencia spp. % of strains inhibited at an MIC (AgIm1) of: Determination 0.25
0.5
1
60 E. coli Cephalothin
Cephaloridine Cephalexin Cefazolin Cefamandole Cefuroxime Cefatrizine Cefoxitin Cefazaflur
40
67
8
35
61
2
4
8
16
32
64
128
>128
3 58
73 83 85 98 95 93 90 95 92
90 90 93 98 100 95 93 98 92
90 92 93 98
93 97 93 98
93 100 93 100
100 100
100 93 98 94
98 100 98
98
100
63 68 5 52 5 78
82 83 65 75 47 86
45 70 48 90 92 88 85 90 88
18
49 38 17 59 68 92 66 83 75
61 61 85 75 79 96 76 94 90
73 68 99 93 92 97 93 99 95
89 79 99 97 97 99 94 100 98
96 90 99 99 99 99 99
99 99 100 99 99 99 99
99 99
100 100
99 99 99 99
100 100 100 100
98
98
100
85 47 15 79 92 92 90 97 89
90 85 65 90 92 95 92 100 90
92 85 90 94 97 97 92
92 89 92 95 98 97 95
95 92 97 97 100 97 95
97 92 97 100
98 95 98
100 100 100
97 95
97 95
100 100
92
97
97
98
100
4 5 7
25 16 25 .37 95 86 62 18 74
32 21 50 43 95 89 68 21 84
45 27 54 50 100 98 87 50 93
54 30 75 57
100 100 100 100
100 93 71 98
100 100 100
14 13 29 27 89 66 51 94 73
23 16 41 44 97 76 71 96 84
36 23 59 61 99 96
100
71 Klebsiella spp.
Cephalothin Cephaloridine Cephalexin Cefazolin Cefamandole Cefuroxime Cefatrizine Cefoxitin Cefazaflur
37
52 4 25
43
62
48 59 63 58 34 70
85 77 15 2 44
35 6 2 13 89 92 85 61 84
62 P. mirabilis
Cephalothin Cephaloridine Cephalexin Cefazolin Cefamandole Cefuroxime Cefatrizine Cefoxitin Cefazaflur
58 18 11
56 Enterobacter spp.
Cephalothin Cephaloridine Cephalexin Cefazolin Cefamandole Cefuroxime Cefatrizine Cefoxitin Cefazaflur
7
:16
4 57 2 4
7
24
3 3
3 3
6 29 7 4 4
7 44 19 7 19 19
16 66 18 18 2 34
20 86 54 45 11 49
11 14 12 21 91 77 54 16 62
4 6 4 10 49 33 23 64 34
6 6 6 10 60 37 24 70 41
10 7 10 16 86 53 44 93 53
70 Indole-positive Proteus and Providencia spp.
Cephalothin Cephaloridine Cephalexin Cefazolin Cefamandole Cefuroxime Cefatrizine Cefoxitin Cefazaflur
13 6 4
11
94 96 90
37 66 84 91 99 99 99 100 97
100 100 100 100 100 100 100 100
ANTIBACTERIAL ACTIVITY OF CEPHALOSPORINS
VOL. 10, 1976
661
TABLE 3. Activity of different cephalosporins against Acinetobacter spp., S. marcescens, and Pseudomonas spp. % of strains inhibited at an MIC (j,g/ml) of: Determination 1
25 Acinetobacter Cephalothin Cephaloridine Cephalexin Cefazolin Cefamandole Cefuroxime Cefatrizine Cefoxitin Cefazaflur 30 S. marcescens Cephalothin Cephaloridine Cephalexin Cefazolin Cefamandole Cefuroxime Cefatrizine Cefoxitin Cefazaflur
8 8
2
4
16
20 20
12 24 24 12
24 36 24 32 4
8
16
32
64
128
>128
16 24 24 20 32 44 24 36 13
20 24 24 24 44 64 28 52 25
24 40 28 28 76 80 36 80 29
24 52 36 36 84 84 68 80 42
40 64 48 52 92 96 80 92
100 100 100 100 100 100 100 100
63
100
3
10
10
17 3
43 3
73 23 3
90 53 3
93 90 13
100 100 100 100 100 100 100
3
7
13
37
100 3
3
7
10
100
3
4
4
3
3
7
74 Pseudomonas spp.
Cephalothin Cephaloridine Cephalexin Cefazolin Cefamandole Cefuroxime Cefatrizine Cefoxitin Cefazaflur was by far the most independent from resistance to cephalothin and was of an equal degree
in the species of group 1 (ratio of 1:3) and those of group 2 (ratio of 1:2). Cefamandole showed a definite lowered activity against cephalothinresistant isolates ofthe group 1 species (ratio of 1:25), but much less against those of group 2 (ratio of 1:4). Susceptibility to cefuroxime was relatively independent of resistance to cephalothin. This was somewhat more apparent among the species of group 2 (ratio of 1:5) than among those of group 1 (ratio of 1:11). DISCUSSION
The comparative value of the nine cephalosporins simultaneously tested in vitro against the same clinical isolates of gram-negative bacilli may be evaluated according to two different criteria. A first criterion is based on the percentage of strains that have MIC values above a critical level that is achievable in vivo after the administration of the usual dose by
3
1
1
4
4
3 8
1
1
12
100 100 100 100 100 100 100 100 100
the usual route of administration and above which the possibility of a therapeutic success becomes highly improbable. A second criterion is based on the geometric mean MIC of the different cephalosporins for a species or a group ofspecies. It takes into account the entire range of individual MIC values from the very susceptible to the very resistant isolate within a species, and it accentuates more the therapeutic margin between the critical inhibition concentration and the mean MIC of the species or group of species. With regard to the first criterion, the differences between the cephalosporins in the percentage of resistant isolates were rather small in the species of group 1, except for cephaloridine and cephalothin, but they were more pronounced in the species of group 2. As a whole cefamandole inhibited the highest number of isolates at the critical level, followed in decreasing order by cefazaflur < cefuroxime < cefoxitin < cefatrizine < cefazolin < cephalexin < cephaloridine < cephalothin.
662
ANTIMICROB. AGENTS CHEMOTHER.
VERBIST
TABLE 4. Geometric mean MICs of different cephalosporins Groupe1
No. tested
CLTa
Geometric mean MIC (AgIml) of: CLX CZL i CMD f CFU CTZ
CLD
4.3 6.6 5.7
7.6 4.3
6.5
6.5
4.2
4.5
56 15 30 11 14 15 25
166
E. coli K. pneumoniae K. oxytoca C. diversus P. mirabilis
8.7 4.6 4.2 7.5 2.2
203
0.7
2.9
1.5
1.3
CZF
1.9
2.1 1.9 1.9
3.2 2.1 1.2
3.5
2.6
2.5
0.5
0.7
1.2
1.3
0.8 0.8 0.7 1.5 1.0
2.3
0.8
1.5
1.7
2.0
0.9
52.5 48.5 45.2 60.1 115.9 26.6 73.5
45.8 25.4 44.2 38.6 95.1 14.1 84.4
1.8 2.4 8.2 2.3 0.7 1.0 20.5
6.8 3.8 18.8 5.8 11.9 1.7 12.5
13.0 12.1 15.6 13.2
11.3 16.0 33.8
68.9 30.6 2.1 9.1 6.9 1.4 17.9
6.7 6.1 5.8 9.1 19.5 4.8 76.1
54.2
43.8
3.1
7.8
15.5
14.1
10.1
24.3 77.0 233.4 >256 222.9 10.4 S. marcescens 30 >256 >256 253.4 226.1 >256 240.9 243.1 >256 >256 67 >256 P. aeruginosa 156.0 >256 >256 190.2 210.0 >256 128.0 7 >256 Pseudomonas spp. a Abbreviations are the same as those given in the footnote to Table 1.
217.8 >256 >256
Total
60 55 16 10 62
CFO
Group 2
Ente9obacter spp. C. freundii P. vulgaris P. rettgeri P. morganii Providencia spp. Acinetobacter spp.
Total
1.2 1.5
5.5
1.7 2.2 4.2 3.5 2.5
5.0
5.5
79.0 40.3 84.4 56.4 210.0 55.7 108.4
115.9 73.5 99.3 53.0 190.2 73.5 48.5
81.2
90.1
3.5
Group 3
TABLE 5. Correlation coefficients of the linear regression between the log MICs of the different cephalosporins and the log MICs of cephalothin No. tested Species ested
Species
CLDa
Correlation coefficients (r) with: CZL CMD CFU CTZ
CLX
E. coli 60 0.87 0.88 0.79" 0.79" 0.89 Klebsiella spp. 71 0.92 0.72" 0.90 0.91 0.69" P. mirabilis 0.88 62 0.93 0.92 0.83 0.92 0.44b 0.88 Enterobacter spp. 56 0.87 0.53" C. freundii 0.49" 15 0.77 0.84 0.91 0.73 0.67" P. vulgaris 30 0.90 0.83 0.87 0.43b 0.10" Providencia spp. 15 0.92 0.83 0.97 0.19" a Abbreviations are the same as those given in the footnote to Table 1. b Correlation coefficient is statistically different from r = 0.90 at P < 0.005.
0.95 0.92 0.93 0.64 0.76 0.82 0.92
CFO
CZF
0.70"
0.87 0.92 0.95
0.50" 0.42" 0.18" 0.34b 0.14b 0.02b
0.62" 0.84
0.64" 0.66
TABLE 6. Geometric mean MIC of the different cephalosporins according to the susceptibility ofgroup I and group 2 microbial species to cephalothin Geometric mean MIC (,ug/ml) of: Susceptibility to 16 ug Susceptibility mg No of cephalothin per ml * CL? CLD CLX CZL CMD CFU CTZ CFO CZF
Species of group 1 Susceptible Resistant
203 183
3.26
3.64
4.52
1.75
0.60
1.77
0.62
84.44
90.50
31.99
25.99
14.92
1.14 12.55
1.25
20
31.99
5.27
17.21
1:26
1:25
1:7
1:15
1:25
1:11
1:25
1:3
1:28
8.59 130.60
13.86 133.94
9.68 77.95
3.90
73.08
1.07 3.89
2.14 10.23
2.20 23.48
7.44 16.14
1.73 14.65
1:5
1:11
1:2
1:8
Ratio
Species of group 2 Susceptible Resistant
166 29 137
Ratio 1:15 1:10 1:8 1:19 1:4 a Abbreviations are the same as those given in the footnote to Table 1.
ANTIBACTERIAL ACTIVITY OF CEPHALOSPORINS
VOL. 10, 1976
With regard to the second criterion and ranged in the order of increasing geometric mean MICs, the classification was: cefamandole < cefazaflur and cefuroxime < cefatrizine and cefoxitin < cefazolin < cephalexin, cephalothin, and cephaloridine. Cefoxitin is known to be less active against Enterobacter spp. than against the other Enterobacteriaceae, but one of the most active cephalosporins against S. marcescens (3, 7, 8, 14). In the present study an even 100% of the 30 isolates ofS. marcescens were found susceptible to 16 ,tg of cefoxitin per ml. Furthermore, cefoxitin is a cephamycin structurally different from the other cephalosporins. This chemical difference seems to be very important, as reflected by its lack of correlation with cephalothin susceptibility (cf. Table 5) and its high activity against cephalothin-resistant isolates in the species of group 1 that are as a group susceptible to the older cephalosporins. From the in vitro results of this study it may be concluded that the newer generation of cephalosporins has several advantages over the older ones. First, they present a broader spectrum against gram-negative bacilli, extending the activity to indole-positive Proteus spp., Providencia spp., Enterobacter spp., and C. freundii. This is true especially for cefamandole, cefazaflur, cefuroxime, and cefoxitin (with the exception of Enterobacter spp. and C. freundii) but less for cefatrizine. Furthermore they inhibit gram-negative bacilli at appreciable lower MIC values, preserving activity against strains that have a low degree of resistance to the older cephalosporins. Finally, cefoxitin seems to open a new and independent class of cephalosporins that preserves largely its activity against isolates highly resistant to the other cephalosporins and that will be appreciated as a safe treatment for infection with S. marcescens. ACKNOWLEDGMENT The technical assistance of Jos Vandeven is gratefully
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14. 15.
acknowledged. LITERATURE CITED 1. Actor, P., J. V. Uri, J. R. Guarini, I. Zajac, L. Phillips, C. S. Sachs, R. M. Demarinis, J. R. E. Hoover, and J. A. Weisbach. 1975. A new parenteral cephalosporin. SKF 59962: in vitro and in vivo antibacterial activity
16.
663
and serum levels in experimental animals. J. Antibiot. 28:471-476. Bodey, G. P., and S. Weaver. 1976. In vitro studies of cefamandole. Antimicrob. Agents Chemother. 9:452457. Brumfitt, W., J. Kosmidis, J. M. T. Hamilton-Miller, and J. N. G. Gilchrist. 1974. Cefoxitin and cephalothin: antimicrobial activity, human pharmacokinetics, and toxicology. Antimicrob. Agents Chemother. 6:290-299. Eykyn, S., C. Jenkins, A. King, and I. Phillips. 1973. Antibacterial activity of cefamandole, a new cephalosporin antibiotic compared with that of cepahloridine, cephalothin, and cephalexin. Antimicrob. Agents Chemother. 3:657-661. Eykyn, S., C. Jenkins, A. King, and I. Phillips. 1976. Antibacterial activity of cefuroxime, a new cephalosporin antibiotic, compared with that of cephaloridine, cephalothin, and cefamandole. Antimicrob. Agents Chemother. 9:690-695. Leitner, F., R. E. Buck, M. Misiek, T. A. Pursiano, and E. Price. 1975. BL-S 640, a cephalosporin with a broad spectrum of antibacterial activity: properties in vitro. Antimicrob. Agents Chemother. 7:298-305. Lewis, R. P., R. D. Meyer, and L. L. Kraus. 1976. Antibacterial activity of selected beta-lactam and aminoglycoside antibiotics against cephalothin-resistant Enterobacteriaceae. Antimicrob. Agents Chemother. 9:780-786. Moellering, R. C., M. Dray, and L. J. Kunz. 1974. Susceptibility of bacteria to cefoxitin and cephalothin. Antimicrob. Agents Chemother. 6:320-323. Neu, H. C. 1974. Cefoxitin, a semisynthetic cephamycin antibiotic antibacterial spectrum and resistance to hydrolysis by gram-negative beta-lactamases. Antimicrob. Agents Chemother. 6:170-176. Neu, H. C. 1974. Cefamandole, a cephalosporin antibiotic with an unusually wide spectrum of activity. Antimicrob. Agents Chemother. 6:177-182. Norrby, R., J. E. Brorson, and S. Seeberg. 1976. Comparative study of the in vitro antibacterial activity of cefoxitin, cefuroxime, and cephaloridine. Antimicrob. Agents Chemother. 9:506-510. O'Callaghan, C. H., R. B. Sykes, A. Griffiths, and J. E. Thornton. 1976. Cefuroxime, a new cephalosporin antibiotic: activity in vitro. Antimicrob. Agents Chemother. 9:511-519. Overturf, G. D., R. L. Ressler, P. B. Marengo, and J. Wilkins. 1975. In vitro evaluation of BLS640, a new oral cephalosporin antibiotic. Antimicrob. Agents Chemother. 8:305-310. Wallick, H., and D. Hendlin. 1974. Cefoxitin, a semisynthetic cephamycin antibiotic: susceptibility studies. Antimicrob. Agents Chemother. 5:25-32. Watanakunakorn, C., T. Bannister, and C. Glotzbecker. 1975. Susceptibility of clinical isolates of Enterobacteriaceae to BL-S640, a new oral cephalosporin. Antimicrob. Agents Chemother. 7:381-385. Watt, P. R., L. Jeffries, and S. A. Price. 1966. An automatic multipoint inoculator for Petri-dishes, p. 125-128. In B. M. Gibbs and F. A. Skinner (ed.), Identification methods for microbiologists, part A. Academic Press Inc., London.
Vol. 10, No. 4 Printed in U.S.A.
Comparison of the Antibacterial Activity of Nine Cephalosporins Against Enterobacteriaceae and Nonfermentative Gram-Negative Bacilli LUDO VERBIST
Department of Microbiology, St. Rafael University Hospital, Leuven, Belgium Received for publication 19 July 1976
The in vitro antibacterial activity of nine cephalosporins (cephalothin, cephaloridine, cephalexin, cefazolin, cefamandole, cefuroxime, cefatrizine, cefoxitin, and cefazaflur) was determined against 344 strains ofEnterobacteriaceae and 99 nonfermentative gram-negative bacilli. Cefamandole, cefazaflur, and cefuroxime were the most active cephalosporins against the Enterobacteriaceae (with the exception of Serratia marcescens). However, cefoxitin was the only cephalosporin that inhibited all 30 S. marcescens strains in a concentration of 16 ,g/ml and was by far the most active compound against selected cephalothin-resistant strains of Escherichia coli, Klebsiella, and Proteus mirabilis. Acinetobacter spp. were inhibited best by cefuroxime, but none of the cephalosporins had appreciable activity against the Pseudomonas spp.
Infections with gram-negative bacilli constitute a major problem, especially among hospitalized patients. An increasing number of these organisms are resistant to most of the currently used antibiotics. Newly developed cephalosporins may offer a safe alternative treatment. A serious drawback to the first clinically
available cephalosporins, cephalothin, cephaloridine, cephalexin, and cefazolin, is that they have only slight activity against Enterobacter spp., Providencia spp., and indole-positive Proteus spp., and Serratia marcescens and Pseudomonas aeruginosa are completely resistant to these antibiotics. However, recently a number of new cephalosporins have been synthesized that seem to have a markedly broader spectrum against gram-negative bacilli. Among the most promising are cefamandole (2, 4, 5, 7, 10), cefuroxime (5, 11, 12), cefatrizine (6, 7, 13, 15), cefoxitin (3, 7-9, 11, 14), and cefazaflur (SKF 59962) (1). In this study the antibacterial activities of these newer cephalosporins against clinical isolates of 344 Enterobacteriaceae and 99 nonfermentative gram-negative bacilli were compared with each other and with that of the older cephalosporins. MATERIALS AND METHODS
Clinical isolates of Enterobacteriaceae and nonfermentative gram-negative bacilli were collected during a 2-month period in St. Raphael, the teaching hospital of the University of Leuven. After identification by standard criteria the following num-
bers of strains were selected for study: 60 Escherichia coli, 55 Klebsiella pneumoniae, 16 Klebsiella oxytoca, 56 Enterobacter spp., 15 Citrobacter freundii, 10 Citrobacter diversus, 62 Proteus mirabilis, 30 Proteus vulgaris, 14 Proteus morganii, 11 Proteus rettgeri, 15 Providencia spp., 30 S. marcescens, 25 Acinetobacter spp., 67 P. aeruginosa, and 7 Pseudomonas spp. The susceptibility of the strains was determined by an agar dilution technique. Overnight cultures in Trypticase soy broth were diluted in sterile Trypticase soy broth and inoculated with an automatic multipoint inoculator (16) to deliver an inoculum of 104 to 105 colony-forming units. Serial twofold dilutions of the cephalosporins were prepared in sterile distilled water. After the addition of one part of each dilution to nine parts of melted and cooled Diagnostic Sensitivity Test agar (DST Oxoid), the final concentration in the petri dishes ranged from 128 to 0.25 ug/ml. The plates were used on the day of preparation or within 24 h. Pure drugs were obtained from the manufacturers: cephalothin, cephaloridine, cefazolin, cephalexin, and cefamandole from Eli Lilly & Co.; cefuroxime from Glaxo; cefatrizine from Bristol Laboratories; cefoxitin from Merck, Sharp and Dohme; and cefazaflur (SKF 59962) from RIT (Belgium). Control drug-free plates were similarly inoculated. For Proteus strains, p-nitrophenyl-38-D-glucoside in a final concentration of 50 ,ug/ml was added to the antibiotic plates and to control plates to avoid swarming. After incubation at 37°C for 18 h, the minimal inhibitory concentration (MIC) was recorded as the lowest concentration of antibiotic that completely inhibited growth; the presence of three or less colonies was disregarded. Clinical susceptibility to an antibiotic was arbitrarily defined as inhibition of strains at or below half the peak levels of antibiotics achieved after the
657
658
ANTIMICROB. AGENTS CHEMOTHER.
VERBIST
usual dose and way of administration in clinical use studies with volunteers. The following MICs considered as limits of susceptibility: cephalothin, 8 ,ug/ml; cephaloridine, 16 gg/ml; cephalexin, 16 ,ug/ml; cefazolin, 16 ug/ml; cefamandole, 16 ,ug/ ml; cefuroxime, 16 ,ug/ml; cefatrizine, 8 jig/ml; cefoxitin, 16 ,g/ml; and cefazaflur, 16 ug/ml. For the calculation of the geometric means of inhibition, strains resistant to 128 ,ug/ml were considered as having an MIC of 256 ,ug/ml.
or in were
RESULTS The percentage of strains susceptible to the different cephalosporins, according to the criteria defined above, is summarized in Table 1. The species have been divided into three groups based upon their susceptibility pattern to cephalothin and cephaloridine. Group 1 comprises the species of which the majority of isolates are susceptible to these antibiotics, namely, Escherichia coli, Klebsiella spp., P. mirabilis, and C. diversus. Group 2 contains the species with only a minority of isolates susceptible to cephalothin and cephaloridine, namely, Enterobacter spp., indole-positive Proteus spp., Providencia spp., Acinetobacter spp., and C. freundii. Group 3 includes Pseudomonas spp. and S. marcescens, which are uniformly and highly resistant to these drugs.
In group 1 there were only minor differences in the percentage of susceptibility to the different cephalosporins at the critical concentrations. Cefoxitin inhibited 99.5% ofthe strains at 16 ,g/ml. Most other cephalosporins inhibited from 98.0 to 92.6% ofthe isolates, except cephaloridine (85.2%) and cephalothin (78.8%). The species of group 2 were clearly more susceptible to the newer cephalosporins than to cephalothin, cephaloridine, cephalexin, and cefazolin. The highest percentages of susceptibility were found toward cefamandole (83.1%), cefuroxime (74.7%), and cefazaflur (67.1%). Cefoxitin inhibited 55.4% of the strains and cefatrizine (42.2%) inhibited a percentage still substantially higher than that of the older cephalosporins (from 10.8 to 31.9%). Cefoxitin had a low score against Enterobacter spp. (18% susceptible) and against C. freundii (20%), but was the most active compound against the indole-positive Proteus and Providencia spp. Cefazaflur and cefatrizine were only moderately active against the Acinetobacter spp. The Pseudomonas strains remained resistant to the newer cephalosporins. In contrast all 30 isolates of S. marcescens were susceptible to 16 ,ug of cefoxitin per ml, and 13 isolates were also susceptible to 16 ,ug of cefamandole per ml. A total
TABLE 1. Susceptibility of gram-negative bacilli to different cephalosporins 9% of strains inhibited at an MIC (Ag/ml) of at least: Species
Group 1 E. coli K. pneumoniae K. oxytoca C. diversus P. mirabilis Total
test 60 55 16 10 62
203
Total
56 15 30 11 14 15 25
166
|
16
16
16
16
(CL?')
(CLD)
(CLX)
(CZL)
(CMD) (CFU)
73 69 87 70 92
90 78 81 80 89
93 100 94 90 92
98 100 87 90 95
100 98 94 90 98
78.8
85.2
94.6
96.6
11 7 10 18 0 13 16
16 7 10 27 7 13 24
25 7 37 27 7 33 24
37 47 23 27 7 53 24
10.8
15.1
24.7
31.9
Group 2 Enterobacter spp. C. freundii P. vulgaris P. rettgeri P. morganii Providencia spp. Acinetobacter spp.
16
8
98.0 95 80 77 91 100 100 44
83.1
95 100 94 90 97 96.6
86 93 43 55 86 100 64
(CFO)
16 (CZF)
90 95 87 90 92
98 100 100 100 100
92 100 93 80 97
92.6 54 53 37 36 64 47 24
74.7
Group 3 0 0 0 0 3 43 30 S. marcescens 0 0 0 0 0 0 P. aeruginosa 67 0 0 0 0 0 0 7 Pseudomonas spp. a CLT, Cephalothin; CLD, cephaloridine; CLX, cephalexin; CZL, cefazolin; cefuroxime; CTZ, cefatrizine; CFO, cefoxitin; CZF, cefazaflur.
8 (CTZ)
45.2 0 0 0
16
99.5 18 20 100 64 100 100 52
55.4 100 0 0
96.6 74 80 80 55
50 93 25 67.1 3 0 0
CMD, cefamandole; CFU,
VOL. 10, 1976
ANTIBACTERIAL ACTIVITY OF CEPHALOSPORINS
of 18 of the 30 S. marcescens isolates were resistant to 100 ug of carbenicillin per ml. Cefamandole inhibited 10 of the 12 strains susceptible to carbenicillin, but only 3 of the 18 strains resistant to carbenicillin. The distribution of the susceptibility to the different cephalosporins, expressed as the cumulative percentage of strains inhibited at increasing concentrations, is shown in Tables 2 and 3. Table 2 summarizes the susceptibility pattern of E. coli, Klebsiella, P. mirabilis, Enterobacter spp., and the indole-positive Proteus and Providencia spp. Cefamandole and cefazaflur inhibited a large percentage of the E. coli, Klebsiella, and P. mirabilis strains already at 0.25 and 0.5 ,ug/ml, but the remaining isolates were inhibited over a wide range of MIC values. Cefoxitin, in contrast, inhibited the strains from a concentration of 1 ,.g/ml, but generally over a shorter range of MIC values. Cefamandole and to a lesser extent cefazaflur also inhibited part of the strains of the Enterobacter spp. and the indole-positive Proteus spp. and Providencia spp. at the lowest concentrations. Cefuroxime was the most active cephalosporin against the Acinetobacter spp., followed by cefoxitin and cefamandole (Table 3). Cefoxitin was clearly the most active compound against S. marcescens, and cefamandole was active in part of these strains. However, against the Pseudomonas spp. there was scarcely any activity of the cephalosporins. Table 4 shows the geometric means of the MIC values of the different cephalosporins. The newer cephalosporins had clearly lower mean MICs than did the older ones. Against the species of group 1, the lowest mean MICs were obtained with cefamandole (0.8 ,ug/ml), cefazaflur (0.9 ,ig/ml), and cefuroxime (1.5 jig/ml). Against the species of group 2, the same cephalosporins had the lowest mean MIC values: cefamandole, 3.1 jig/ml; cefuroxime, 7.8 ,ug/ml; and cefazaflur, 10.1 ,g/ml. Cefoxitin took only the fourth place as a whole (14.1 jig/ml) due to its weaker activity against Enterobacter spp. and C. freundii. However, against the indolepositive Proteus spp. and the Providencia spp., cefoxitin was the second best, with a mean MIC of 3.1 jig/ml, cefamandole being most effective with a mean MIC of 2.3 jig/ml. The activity of cefoxitin against S. marcescens, with a mean MIC of 10.4 ,g/ml, was similar to the activity of this antibiotic against the species of group 2 (namely, 10.1 ,ug/ml). Since the activity of all cephalosporins was tested simultaneously on the same inoculum of each species, a direct comparison between the MICs of each cephalosporin and the MIC of cephalothin as a refer-
659
ence was possible for each isolate. Therefore, the linear regression lines of the log MICs between each cephalosporin and cephalothin were calculated for these species, whereby more than half of the isolates were at least susceptible to 128 ,ug of cephalothin per ml (resistance to 128 ,ug/ml was considered as susceptibility to 256 ,ug/ml for convenience in the calculations). In the other species, with more than half of the strains resistant to 128 ,ug of cephalothin per ml, the lack of an exact end point of inhibition made the calculation of regression lines less valuable. Table 5 shows the correlation coefficients for E. coli, Klebsiella spp., P. mirabilis, Enterobacter spp., C. freundii, P. vulgaris, and Providencia spp. between the log MICs of each cephalosporin and cephalothin. A perfect correlation (coefficient r = 1.0) could not be expected. Therefore, the statistical difference was calculated between the observed r values and a correlation coefficient of at least 0.90. The correlation coefficients of cephaloridine, cefazolin, cephalexin (with the exception of the Klebsiella spp.), and cefatrizine (except the Enterobacter spp.) toward cephalothin were not statistically different from r = 0.90. This means that higher or lower MIC values with one of these cephalosporins corresponded very well with higher or lower MICs with cephalothin. In other words, isolates resistant to cephalothin were as a rule less susceptible or even resistant to these other cephalosporins. In contrast, especially cefoxitin and to a lesser extent cefuroxime had very poor correlation coefficients with cephalothin. Cefamandole took a place in between, and cefazaflur tended rather to a parallel susceptibility with cephalothin. Table 6 shows the geometric mean MICs of the different cephalosporins according to the susceptibility of the isolates to cephalothin at a concentration of 16 ,g/ml. The ratio between the geometric mean MIC of the susceptible isolates and of the cephalothin-resistant isolates reflects the degree of cross-resistance or crosssusceptibility with cephalothin. This ratio was 1:26 for cephalothin in the species of group 1 (E. coli, Klebsiella spp., and P. mirabilis). The ratios for cefazolin, cephaloridine, cefamandole, cefatrizine, and cefazaflur were very similar and varied from 1:15 to 1:28. In the species of group 2 (Enterobacter, C. freundii, indole-positive Proteus, Providencia, and Acinetobacter) the ratio between the geometric mean MIC of susceptible and resistant isolates was 1:15 for cephalothin and between 1:8 and 1:19 for cephalexin, cefazaflur, cephaloridine, cefatrizine, and cefazolin. The susceptibility to cefoxitin
660
ANTIMICROB. AGENTS CHEMOTHZR.
VERBIST
TABLE 2. Activity ofdifferent cephalosporins against E. coli, Klebsiella spp., P. mirabilis, Enterobacter spp., and indole-positive Proteus and Providencia spp. % of strains inhibited at an MIC (AgIm1) of: Determination 0.25
0.5
1
60 E. coli Cephalothin
Cephaloridine Cephalexin Cefazolin Cefamandole Cefuroxime Cefatrizine Cefoxitin Cefazaflur
40
67
8
35
61
2
4
8
16
32
64
128
>128
3 58
73 83 85 98 95 93 90 95 92
90 90 93 98 100 95 93 98 92
90 92 93 98
93 97 93 98
93 100 93 100
100 100
100 93 98 94
98 100 98
98
100
63 68 5 52 5 78
82 83 65 75 47 86
45 70 48 90 92 88 85 90 88
18
49 38 17 59 68 92 66 83 75
61 61 85 75 79 96 76 94 90
73 68 99 93 92 97 93 99 95
89 79 99 97 97 99 94 100 98
96 90 99 99 99 99 99
99 99 100 99 99 99 99
99 99
100 100
99 99 99 99
100 100 100 100
98
98
100
85 47 15 79 92 92 90 97 89
90 85 65 90 92 95 92 100 90
92 85 90 94 97 97 92
92 89 92 95 98 97 95
95 92 97 97 100 97 95
97 92 97 100
98 95 98
100 100 100
97 95
97 95
100 100
92
97
97
98
100
4 5 7
25 16 25 .37 95 86 62 18 74
32 21 50 43 95 89 68 21 84
45 27 54 50 100 98 87 50 93
54 30 75 57
100 100 100 100
100 93 71 98
100 100 100
14 13 29 27 89 66 51 94 73
23 16 41 44 97 76 71 96 84
36 23 59 61 99 96
100
71 Klebsiella spp.
Cephalothin Cephaloridine Cephalexin Cefazolin Cefamandole Cefuroxime Cefatrizine Cefoxitin Cefazaflur
37
52 4 25
43
62
48 59 63 58 34 70
85 77 15 2 44
35 6 2 13 89 92 85 61 84
62 P. mirabilis
Cephalothin Cephaloridine Cephalexin Cefazolin Cefamandole Cefuroxime Cefatrizine Cefoxitin Cefazaflur
58 18 11
56 Enterobacter spp.
Cephalothin Cephaloridine Cephalexin Cefazolin Cefamandole Cefuroxime Cefatrizine Cefoxitin Cefazaflur
7
:16
4 57 2 4
7
24
3 3
3 3
6 29 7 4 4
7 44 19 7 19 19
16 66 18 18 2 34
20 86 54 45 11 49
11 14 12 21 91 77 54 16 62
4 6 4 10 49 33 23 64 34
6 6 6 10 60 37 24 70 41
10 7 10 16 86 53 44 93 53
70 Indole-positive Proteus and Providencia spp.
Cephalothin Cephaloridine Cephalexin Cefazolin Cefamandole Cefuroxime Cefatrizine Cefoxitin Cefazaflur
13 6 4
11
94 96 90
37 66 84 91 99 99 99 100 97
100 100 100 100 100 100 100 100
ANTIBACTERIAL ACTIVITY OF CEPHALOSPORINS
VOL. 10, 1976
661
TABLE 3. Activity of different cephalosporins against Acinetobacter spp., S. marcescens, and Pseudomonas spp. % of strains inhibited at an MIC (j,g/ml) of: Determination 1
25 Acinetobacter Cephalothin Cephaloridine Cephalexin Cefazolin Cefamandole Cefuroxime Cefatrizine Cefoxitin Cefazaflur 30 S. marcescens Cephalothin Cephaloridine Cephalexin Cefazolin Cefamandole Cefuroxime Cefatrizine Cefoxitin Cefazaflur
8 8
2
4
16
20 20
12 24 24 12
24 36 24 32 4
8
16
32
64
128
>128
16 24 24 20 32 44 24 36 13
20 24 24 24 44 64 28 52 25
24 40 28 28 76 80 36 80 29
24 52 36 36 84 84 68 80 42
40 64 48 52 92 96 80 92
100 100 100 100 100 100 100 100
63
100
3
10
10
17 3
43 3
73 23 3
90 53 3
93 90 13
100 100 100 100 100 100 100
3
7
13
37
100 3
3
7
10
100
3
4
4
3
3
7
74 Pseudomonas spp.
Cephalothin Cephaloridine Cephalexin Cefazolin Cefamandole Cefuroxime Cefatrizine Cefoxitin Cefazaflur was by far the most independent from resistance to cephalothin and was of an equal degree
in the species of group 1 (ratio of 1:3) and those of group 2 (ratio of 1:2). Cefamandole showed a definite lowered activity against cephalothinresistant isolates ofthe group 1 species (ratio of 1:25), but much less against those of group 2 (ratio of 1:4). Susceptibility to cefuroxime was relatively independent of resistance to cephalothin. This was somewhat more apparent among the species of group 2 (ratio of 1:5) than among those of group 1 (ratio of 1:11). DISCUSSION
The comparative value of the nine cephalosporins simultaneously tested in vitro against the same clinical isolates of gram-negative bacilli may be evaluated according to two different criteria. A first criterion is based on the percentage of strains that have MIC values above a critical level that is achievable in vivo after the administration of the usual dose by
3
1
1
4
4
3 8
1
1
12
100 100 100 100 100 100 100 100 100
the usual route of administration and above which the possibility of a therapeutic success becomes highly improbable. A second criterion is based on the geometric mean MIC of the different cephalosporins for a species or a group ofspecies. It takes into account the entire range of individual MIC values from the very susceptible to the very resistant isolate within a species, and it accentuates more the therapeutic margin between the critical inhibition concentration and the mean MIC of the species or group of species. With regard to the first criterion, the differences between the cephalosporins in the percentage of resistant isolates were rather small in the species of group 1, except for cephaloridine and cephalothin, but they were more pronounced in the species of group 2. As a whole cefamandole inhibited the highest number of isolates at the critical level, followed in decreasing order by cefazaflur < cefuroxime < cefoxitin < cefatrizine < cefazolin < cephalexin < cephaloridine < cephalothin.
662
ANTIMICROB. AGENTS CHEMOTHER.
VERBIST
TABLE 4. Geometric mean MICs of different cephalosporins Groupe1
No. tested
CLTa
Geometric mean MIC (AgIml) of: CLX CZL i CMD f CFU CTZ
CLD
4.3 6.6 5.7
7.6 4.3
6.5
6.5
4.2
4.5
56 15 30 11 14 15 25
166
E. coli K. pneumoniae K. oxytoca C. diversus P. mirabilis
8.7 4.6 4.2 7.5 2.2
203
0.7
2.9
1.5
1.3
CZF
1.9
2.1 1.9 1.9
3.2 2.1 1.2
3.5
2.6
2.5
0.5
0.7
1.2
1.3
0.8 0.8 0.7 1.5 1.0
2.3
0.8
1.5
1.7
2.0
0.9
52.5 48.5 45.2 60.1 115.9 26.6 73.5
45.8 25.4 44.2 38.6 95.1 14.1 84.4
1.8 2.4 8.2 2.3 0.7 1.0 20.5
6.8 3.8 18.8 5.8 11.9 1.7 12.5
13.0 12.1 15.6 13.2
11.3 16.0 33.8
68.9 30.6 2.1 9.1 6.9 1.4 17.9
6.7 6.1 5.8 9.1 19.5 4.8 76.1
54.2
43.8
3.1
7.8
15.5
14.1
10.1
24.3 77.0 233.4 >256 222.9 10.4 S. marcescens 30 >256 >256 253.4 226.1 >256 240.9 243.1 >256 >256 67 >256 P. aeruginosa 156.0 >256 >256 190.2 210.0 >256 128.0 7 >256 Pseudomonas spp. a Abbreviations are the same as those given in the footnote to Table 1.
217.8 >256 >256
Total
60 55 16 10 62
CFO
Group 2
Ente9obacter spp. C. freundii P. vulgaris P. rettgeri P. morganii Providencia spp. Acinetobacter spp.
Total
1.2 1.5
5.5
1.7 2.2 4.2 3.5 2.5
5.0
5.5
79.0 40.3 84.4 56.4 210.0 55.7 108.4
115.9 73.5 99.3 53.0 190.2 73.5 48.5
81.2
90.1
3.5
Group 3
TABLE 5. Correlation coefficients of the linear regression between the log MICs of the different cephalosporins and the log MICs of cephalothin No. tested Species ested
Species
CLDa
Correlation coefficients (r) with: CZL CMD CFU CTZ
CLX
E. coli 60 0.87 0.88 0.79" 0.79" 0.89 Klebsiella spp. 71 0.92 0.72" 0.90 0.91 0.69" P. mirabilis 0.88 62 0.93 0.92 0.83 0.92 0.44b 0.88 Enterobacter spp. 56 0.87 0.53" C. freundii 0.49" 15 0.77 0.84 0.91 0.73 0.67" P. vulgaris 30 0.90 0.83 0.87 0.43b 0.10" Providencia spp. 15 0.92 0.83 0.97 0.19" a Abbreviations are the same as those given in the footnote to Table 1. b Correlation coefficient is statistically different from r = 0.90 at P < 0.005.
0.95 0.92 0.93 0.64 0.76 0.82 0.92
CFO
CZF
0.70"
0.87 0.92 0.95
0.50" 0.42" 0.18" 0.34b 0.14b 0.02b
0.62" 0.84
0.64" 0.66
TABLE 6. Geometric mean MIC of the different cephalosporins according to the susceptibility ofgroup I and group 2 microbial species to cephalothin Geometric mean MIC (,ug/ml) of: Susceptibility to 16 ug Susceptibility mg No of cephalothin per ml * CL? CLD CLX CZL CMD CFU CTZ CFO CZF
Species of group 1 Susceptible Resistant
203 183
3.26
3.64
4.52
1.75
0.60
1.77
0.62
84.44
90.50
31.99
25.99
14.92
1.14 12.55
1.25
20
31.99
5.27
17.21
1:26
1:25
1:7
1:15
1:25
1:11
1:25
1:3
1:28
8.59 130.60
13.86 133.94
9.68 77.95
3.90
73.08
1.07 3.89
2.14 10.23
2.20 23.48
7.44 16.14
1.73 14.65
1:5
1:11
1:2
1:8
Ratio
Species of group 2 Susceptible Resistant
166 29 137
Ratio 1:15 1:10 1:8 1:19 1:4 a Abbreviations are the same as those given in the footnote to Table 1.
ANTIBACTERIAL ACTIVITY OF CEPHALOSPORINS
VOL. 10, 1976
With regard to the second criterion and ranged in the order of increasing geometric mean MICs, the classification was: cefamandole < cefazaflur and cefuroxime < cefatrizine and cefoxitin < cefazolin < cephalexin, cephalothin, and cephaloridine. Cefoxitin is known to be less active against Enterobacter spp. than against the other Enterobacteriaceae, but one of the most active cephalosporins against S. marcescens (3, 7, 8, 14). In the present study an even 100% of the 30 isolates ofS. marcescens were found susceptible to 16 ,tg of cefoxitin per ml. Furthermore, cefoxitin is a cephamycin structurally different from the other cephalosporins. This chemical difference seems to be very important, as reflected by its lack of correlation with cephalothin susceptibility (cf. Table 5) and its high activity against cephalothin-resistant isolates in the species of group 1 that are as a group susceptible to the older cephalosporins. From the in vitro results of this study it may be concluded that the newer generation of cephalosporins has several advantages over the older ones. First, they present a broader spectrum against gram-negative bacilli, extending the activity to indole-positive Proteus spp., Providencia spp., Enterobacter spp., and C. freundii. This is true especially for cefamandole, cefazaflur, cefuroxime, and cefoxitin (with the exception of Enterobacter spp. and C. freundii) but less for cefatrizine. Furthermore they inhibit gram-negative bacilli at appreciable lower MIC values, preserving activity against strains that have a low degree of resistance to the older cephalosporins. Finally, cefoxitin seems to open a new and independent class of cephalosporins that preserves largely its activity against isolates highly resistant to the other cephalosporins and that will be appreciated as a safe treatment for infection with S. marcescens. ACKNOWLEDGMENT The technical assistance of Jos Vandeven is gratefully
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14. 15.
acknowledged. LITERATURE CITED 1. Actor, P., J. V. Uri, J. R. Guarini, I. Zajac, L. Phillips, C. S. Sachs, R. M. Demarinis, J. R. E. Hoover, and J. A. Weisbach. 1975. A new parenteral cephalosporin. SKF 59962: in vitro and in vivo antibacterial activity
16.
663
and serum levels in experimental animals. J. Antibiot. 28:471-476. Bodey, G. P., and S. Weaver. 1976. In vitro studies of cefamandole. Antimicrob. Agents Chemother. 9:452457. Brumfitt, W., J. Kosmidis, J. M. T. Hamilton-Miller, and J. N. G. Gilchrist. 1974. Cefoxitin and cephalothin: antimicrobial activity, human pharmacokinetics, and toxicology. Antimicrob. Agents Chemother. 6:290-299. Eykyn, S., C. Jenkins, A. King, and I. Phillips. 1973. Antibacterial activity of cefamandole, a new cephalosporin antibiotic compared with that of cepahloridine, cephalothin, and cephalexin. Antimicrob. Agents Chemother. 3:657-661. Eykyn, S., C. Jenkins, A. King, and I. Phillips. 1976. Antibacterial activity of cefuroxime, a new cephalosporin antibiotic, compared with that of cephaloridine, cephalothin, and cefamandole. Antimicrob. Agents Chemother. 9:690-695. Leitner, F., R. E. Buck, M. Misiek, T. A. Pursiano, and E. Price. 1975. BL-S 640, a cephalosporin with a broad spectrum of antibacterial activity: properties in vitro. Antimicrob. Agents Chemother. 7:298-305. Lewis, R. P., R. D. Meyer, and L. L. Kraus. 1976. Antibacterial activity of selected beta-lactam and aminoglycoside antibiotics against cephalothin-resistant Enterobacteriaceae. Antimicrob. Agents Chemother. 9:780-786. Moellering, R. C., M. Dray, and L. J. Kunz. 1974. Susceptibility of bacteria to cefoxitin and cephalothin. Antimicrob. Agents Chemother. 6:320-323. Neu, H. C. 1974. Cefoxitin, a semisynthetic cephamycin antibiotic antibacterial spectrum and resistance to hydrolysis by gram-negative beta-lactamases. Antimicrob. Agents Chemother. 6:170-176. Neu, H. C. 1974. Cefamandole, a cephalosporin antibiotic with an unusually wide spectrum of activity. Antimicrob. Agents Chemother. 6:177-182. Norrby, R., J. E. Brorson, and S. Seeberg. 1976. Comparative study of the in vitro antibacterial activity of cefoxitin, cefuroxime, and cephaloridine. Antimicrob. Agents Chemother. 9:506-510. O'Callaghan, C. H., R. B. Sykes, A. Griffiths, and J. E. Thornton. 1976. Cefuroxime, a new cephalosporin antibiotic: activity in vitro. Antimicrob. Agents Chemother. 9:511-519. Overturf, G. D., R. L. Ressler, P. B. Marengo, and J. Wilkins. 1975. In vitro evaluation of BLS640, a new oral cephalosporin antibiotic. Antimicrob. Agents Chemother. 8:305-310. Wallick, H., and D. Hendlin. 1974. Cefoxitin, a semisynthetic cephamycin antibiotic: susceptibility studies. Antimicrob. Agents Chemother. 5:25-32. Watanakunakorn, C., T. Bannister, and C. Glotzbecker. 1975. Susceptibility of clinical isolates of Enterobacteriaceae to BL-S640, a new oral cephalosporin. Antimicrob. Agents Chemother. 7:381-385. Watt, P. R., L. Jeffries, and S. A. Price. 1966. An automatic multipoint inoculator for Petri-dishes, p. 125-128. In B. M. Gibbs and F. A. Skinner (ed.), Identification methods for microbiologists, part A. Academic Press Inc., London.
