Synergy Between Cephalosporin and Aminoglycoside

ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, June 1974, P. 571--577 Copyright 0 1974 American Society for Microbiology Vol. 5, No. 6 Printed in U.S.A. Synergy Between Cephalosporin and Aminoglycoside Antibiotics Against Providencia and Proteus PHINEAS J. HYAMS, MICHAEL S. SIMBERKOFF, AND JAMES J. RAHAL, JR. Infectious Disease Division, New York Veterans Administration Hospital and Department of Medicine, New York University School of Medicine, New York, New York 10010 Received for publication 14 January 1974 Clinical isolates of Providencia and Proteus with relative aminoglycoside resistance were tested for susceptibility to combinations of gentamicin or tobramycin with cephalothin or cefazolin. The minimal bactericidal concentration of aminoglycoside for one-third of the strains was reduced by fourfold or more in the presence of one-fourth of the minimal bactericidal concentration of either cephalosporin. This effect was achieved by clinically attainable concentrations of cephalothin or cefazolin. Although Providencia has been an uncommon cause of clinical infections, recent epidemics have been described in hospitalized patients, particularly in burn units (5, 14, 15). We have noted a gradually increasing incidence of clinical isolates of Providencia in our hospital, although serious infections have remained rare. In vitro studies on these strains have indicated relative resistance to all aminoglycoside antibiotics. Most strains have required minimal inhibitory and bactericidal concentrations of 12.5 to 50 gg of gentamicin per ml. A synergistic effect of penicillins or cephalosporins with aminoglycosides has previously been demonstrated against Enterobacteriaceae (3, 4, 10) and Pseudomonas (6, 12, 17). We have therefore studied nine strains of Providencia and seven strains of the related genus, Proteus, which were moderately or highly resistant to gentamicin. Our purpose was to evaluate two cephalosporins, cephalothin and cefazolin, in crossover combination with two aminoglycosides, gentamicin and tobramycin, for possible in vitro synergy against these organisms. MATERIALS AND METHODS Providencia and Proteus strains were collected from the clinical microbiology laboratory on the basis of gentamicin resistance by routine disk sensitivity testing. Tube dilution assays were done in Mueller- Hinton broth as previously described (9). Grid studies were performed so that double dilutions of cephalothin, and cefazolin were combined with double dilutions of gentamicin or tobramycin. This resulted in duplicate minimal inhibitory and bactericidal concentration values for each drug against each strain. Of the 144 such duplications, four differed by two tube dilutions and the remainder differed by one dilution or less. Synergy was defined as a fourfold decline in minimal inhibitory concentrations (MIC) or minimal bactericidal concentration (MBC) of gentamicin or tobramycin after the addition of one-fourth of the MIC or MBC of a cephalosporin antibiotic to each tube. 571 RESULTS The MIC and MBC values of each drug against the nine strains of Providencia and seven strains of Proteus tested are shown in Fig. 1 to 4. Eight of nine Providencia strains were resistant to clinically attainable concentrations of either cephalothin or cefazolin. One was inhibited by 62 sg/ml but required >500 gg/ ml for a bactericidal effect. Three strains were inhibited by 3.1 gg of tobramycin per ml and one by that concentration of gentamicin. However, the MIC and MBC of either drug against most Providencia was 12.5 to 50,g/ml. All seven Proteus strains studied were indole positive and highly resistant to cephalosporins. Unlike Providencia, however, a clear difference in sensitivity to gentamicin and tobramycin existed in four of the seven isolates. Two strains of Proteus morganii and two of Proteus vulgaris were sensitive to tobramycin and relatively resistant to gentamicin. Two strains of Proteus rettgeri were resistant to both aminoglycosides, and another Proteus morganii isolate was highly resistant to all antibiotics. Figure 1 indicates the effect of cephalothin and cefazolin on the MICs of tobramycin and gentamicin against Providencia. Ninety-one percent of Providencia strains were susceptible to a synergistic effect, demonstrated by a fourfold decline in the MIC of tobramycin or gen-

572 HYAMS, SIMBERKOFF, AND RAHAL ANTIMICROB. AG. CHEMOTHER. E 50,0 _ G + c z PROVIDENCIA - MIC G + C F Z 3. _ ugdmdof CZ 0 15 0 )5 0 15 0 15 0)5 015 0)5 015 0)5 MIC of CZ 125 250 250 125 250 62 250 125 125 log/mi) Bacterioa A 8 C 0 E F 6 H Strain E > 50.0 T+CZ fh AA addedlaocf 0 )5 0 31 0 62 0 125 0 62 0 )5 0 31 0 62 0 62 (lag/rni)) (IC9of CF 250 125 250 500 500 250 250 250 250 BDoctrio Strain A C 0 E F C H _ T + C F 0 cr_ti InE0n11 L 11 9ug/rnotfCZ 0 )5 0 )5 0 )5 0 62 0 )5 0 )5 0 )5 0 15 0 15 lg/mnioftc F 0 31 0 62 0 31 0 125 0 62 0 )5 0 62 0 62 0 62 added too added toot IC/ofnC) 125 125 250 500 250 62 250 125 62 N)IC at CF 250 250 250 500 250 250 250 250 250 Bacter ia) Doctoria) Strain A B C D E F 6 H Strain A B C D E F C H tamicin. This usually required less than 62,ug of cephalosporin per ml, a clinically attainable level of either cephalothin or cefazolin. Figure 2 shows the effect of each combination of antibiotics on the MBC of Providencia strains. A synergistic effect occurred in 89% of the tests, but in several instances 125 to 250,g of a cephalosporin per ml was required. The results of studies carried out with Proteus species are presented in Fig. 3 and 4. Tests for synergy involving tobramycin were not done with four of the seven strains since they were highly sensitive to this antibiotic. Of all tests carried out, a fourfold decline in MIC occurred in 75% and usually required 15 to 62,ug of a cephalosporin per ml (Fig. 3). Figure 4 shows that only 65% of Proteus strains were susceptible to a synergistic effect on the MBC of aminoglycosides and approximately half of these required greater than 62 usg of a cephalosporin per ml to achieve this result. FIG. 1. Effect of cephalothin (CF) and cefazolin (CZ) on the MIC of gentamicin (G) and tobramycin (7) against Providencia. The left bar of each pair indicates the MIC of tobramycin or gentamicin in the absence of added cephalosporin. The right bar shows the MIC of the aminoglycoside in the presence of one-fourth or less of the MIC of either cephalothin or cefazolin. Table 1 summarizes the percent of cephalosporin-aminoglycoside combinations which produced synergy against Providencia and Proteus. Providencia were somewhat more susceptible than Proteus to the combinations tested. The difference in effect on MBC (but not MIC) was significant with a P value of <0.05. If only those strains susceptible to synergy with the use of 62,gg/ml or less of either cephalosporin are considered, a significantly greater number of Providencia strains showed fourfold lower MICs. However, the synergistic effect on MBCs occurred in an equal percentage of Providencia and Proteus isolates (36 and 37%, respectively). Thus, a clinically attainable serum concentration of cephalosporin favorably effected the bactericidal activity of tobramycin and gentamicin in approximately one-third of these relatively resistant organisms. The more highly resistant organisms (those with MBCs of 25 50,ug/ml) were usually susceptible to a synergistic

VOL. 5, 1974 CEPHALOSPORIN-AMINOGLYCOSIDE SYNERGY 573 PROVIDENCIA - MBC >',0.S Ii-,I G + C Z G + CF 3.1 QC) ioq lmi lofc 0 31 0 31 0 IS 0 15 0 250 0 125 0 250 0 250 015IIS /l eofcf 0 31 0 125 0 125 0 31 0 250 0 250 0 250 0 250 0 250 added to > >dd5 ftc NBC of CZ 500 >500 >500 500 >500 >500 >500 >500 500 NBC of C F 500 500 500 500 500 >500 >500 >500 >500 lag/all Cq /0If BoSteraio A B C 0 E F G H Decteriel A B C D E St rolo E >50.0- LA. 00 T+CZ T+ CF 49/l oftcz o 15 0 125 0 15 0 31 0 250 0 62 0 125 0 250 0 62 g/iloft CF 0O15 0125 0125 0125 0125 0250 0125 0 250 0125 N(BCo/f CZl 500 >500 >500 >500 >500 >500 >250 >500 >500 18C/of >500 500 >500 *500 >500 >250 >500 500 BaSterail B C D E F G B cterial A B C D E F G H FIG. 2. Effect of cephalothin (CF) and cefazolin (CZ) on the MBC of gentamicin (G) and tobramycin (7) against Providencia. The left bar of each pair indicates the MBC of tobramycin or gentamicin in the absence of added cephalosporin. The right bar shows the MBC of the aminoglycoside in the presence of one-fourth or less of the MBC of either cephalothin or cefazolin. effect on MIC values, but not on MBCs. In three instances (1.6% of tests performed), the addition of 15 ug of cephalothin per ml to either gentamicin or tobramycin resulted in an antagonistic effect on the MBC only. The incidence of synergy associated with each of the four antibiotics studied is presented in Table 2. Synergy occurred in 75% of tests performed with cephalothin and in 93% tested with cefazolin, as determined by MIC. The difference in effect on MBCs was less pronounced. If synergy by 62,ug/ml or less was considered alone, cefazolin was again more active than cephalothin although at a slightly lower than 95% probability level (0.1 > P > 0.05). DISCUSSION The results of this study indicate that the MIC and MBC of either gentamicin or tobramycin for relatively resistant Providencia or Proteus isolates may be significantly reduced by the addition of cephalothin or cefazolin. Although a bacteriostatic effect is produced by this combination of antibiotics in 60 to 80% of instances, bactericidal activity occurs in only 36 to 37%. Further, when the MBC of gentamicin or tobramycin for a particular strain is 25 /g/ml or greater, a synergistic effect is unlikely to result from clinically attainable concentrations of cephalothin or cefazolin. Nevertheless, since the MIC of gentamicin for a large proportion of Providencia strains tested in our laboratory is 12.5,ug/ml or more, the additional use of a cephalosporin antibiotic may be clinically useful. Cefazolin is slightly more active against Providencia and Proteus than cephalothin, and higher blood levels are attained by usual doses. Our data showed no evident differences in synergistic effect between tobramycin and gentamicin when combined with cephalosporins. Four strains of indole-positive Proteus were c H

574 HYAMS, SIMBERKOFF, AND RAHAL ANTIMICROB. AG. CHEMOTHER. E>500 o 500 C-!, gu 24 1 G+CZ a/e;co 15 031 0l1 015 031 015 0125 ItO CZ 500 125 500 62 250 6? 500 Bocteriul StrmiR > 500 z 50.0 z? 12.4 m 3'1 LA- O.78 --' <39 ae - j N T+ CZ g/lmlaof CZ 0 MIC of CZ (A g/ml) 250 Bacterial Strain I L N K I PROTEUS- M C p lig/o; f CF o 62 0 15 0 62 0 125 0 31 0 62 0250 Idded t 2 NIC of CF > 500 62 250 500 500 > 500 >500 ag/im) lecterisi Stroll J K N I 0 P T+CF - 0-0 125 0-0 15 0-0 125 Ag/mlato Cf 0-0 - 0 31 0-0 31 0-0 250 62 5 00 62 2 50 62 5 00 NBCatoft > 500 62 2 50 500 500 >500 >500 FIG. 3. Effect of cephalothin (CF) and cefazolin (CZ) on the MIC of gentamicin (G) and tobramycin (T) against Proteus. The left bar of each pair indicates the MIC of tobramycin or gentamicin in the absence of added cephalosporin. The right bar shows the MIC of the aminoglycoside in the presence of one-fourth or less of the MIC of either cephalothin or cefazolin. Bacterial Strain G+CF 0 P

VOL. 5, 1974 CEPHALOSPORIN-AMINOGLYCOSIDE SYNERGY 575 >>50.0 0 50 0 G +CZ PROTEUS-MBC Go+CF B 3.1 <.39 oail/mliofcz 0 250 0 31 0 250 0 31 0 15 0 31 0 250 aodded to G NBCgomi) >500 >500 >500 12 5 >500 125 > 500 Bocter ioi Stroin -E>50.0- ~'50.0- ~?12.4- m 3.1 0 18 <-)<39 11 11 0 P T+CZ ag/mi of CF added to C NBC o0 CF BAg) gobterial ~~Strais ~~~~ 0 250 0 31 0 125 0 125 0 62 0 250 0 250 > 500 250 > 500 > 500 > 500 > 500 > 500 J L N I 0 P T+CF ag/m"ifcz 0125 015 0250 0-0 31 0-0250 ag/mietfcf ts. 0 31 0 15 0250 0-0 62 0-0250 NBCofaCZ >500 250 >500 125 500 125 > 500 'NBCoCf >500 500 >500 500 500 >500 >500 ag/ mi) Bacterial N N oterial Strain 0 P Strain K L 0 P FIG. 4. Effect of cephalothin (CF) and cefazolin (CZ) on the MBC of gentamicin (G) and tobramycin (T) against Proteus. The left bar of each pair indicates the MBC of tobramycin or gentamicin in the absence of added cephalosporin. The right bar shows the MBC of the aminoglycoside in the presence of one-fourth or less of the MBC of either cephalothin or cefazolin. ag/m

576 HYAMS, SIMBERKOFF, AND RAHAL ANTIMICROB. AG. CHEMOTHER. I TABLE 1. Percent of tested combinations of cephalosporin plus aminoglycoside which produce synergy against Providencia and Proteus Genus tested Tests producing synergy p Tests producing synergy with 62 pg or less of cephalosporin per ml P Providencia 91 (89)a 83 (36) >0.1 ( <0.05)b 0.05 (>0.1)b Proteus 75 (63) 60 (37) a Refers to the percent of tested combinations producing synergy as determined by changes in MIC. Numbers in parenthesis indicate percent of synergy as determined by changes in MBC. bp values were determined by the chi-square test. TABLE 2. Percent of tested combinations of cephalosporin plus aminoglycoside which produce synergy against Providencia and Proteus Tests producing Tests synergy producing synergy with 62 pg/ml or Antibiotic.int p i nless of cephalosporin combination Antibiotic inclu de in Cephalothin 975 (73) >0.05 ( >0.1)" 64 (27) >0.05 (>0.1) Cephazlotin 93 ( (>1) 86 (47) Gentamicin 834(86) >0.1(>0.1) 712(39) >0.1(>0.1) a Refers to the percent of tested combinations producing synergy as determined by changes in MIC. Numbers in parentheses indicate percent producing synergy as determined by changes in MBC. P p values were determined by the chi-square test. quite sensitive to tobramycin despite moderate to high gentamicin resistance. It is possible that gentamicin resistance in these strains is mediated by an acetylating enzyme which does not utilize tobramycin (2). The phenomenbn of cephalosporin-aminoglycoside synergy against Providencia and Proteus is consistent with previous findings of synergy between inhibitors of cell wall metabolism, e.g., penicillins or cephalosporins, and aminoglycosides. Klastersky et al. have recently shown synergy between cephalothin and gentamicin or tobramycin against gentamicinresistant Providencia (11). Our findings confirm their results and further define the bacteriostatic and bactericidal activity of this combination. In addition, the present data suggest that cefazolin may be more effective than cephalothin in producing synergy with aminoglycosides. Recent reports have suggested that the combination of cephalothin and gentamicin is potentially more nephrotoxic than either drug alone (1, 7, 8, 13). Since cefazolin has shown somewhat greater nephrotoxic potential than cephalothin in experimental studies (16), renal function should be carefully monitored during the administration of any cephalosporin-aminoglycoside combination. ACKNOWLEDGMENTS This study was supported by a grant from Eli Lilly and Co. We thank Martha Kuepper for valuable technical assistance. LITERATURE CITED 1. Bobrow, S. N., E. Jaffe, and R. C. Young. 1972. Anuria and acute tubular necrosis associated with gentamicin and cephalothin. J. Amer. Med. Ass. 222:1546-1547. 2. Brzezinska, M., R. Benveniste, J. Davies, P. J. L. Daniels, and J. Weinstein. 1972. Gentamicin resistance in strains of Pseudomonas aeruginosa mediated by enzymatic N-acetylation of the doexystreptamine moiety. Biochemistry 11:761-765. 3. Bulger, R. J. 1967. In vitro effectiveness of kanamycin and kanamycin/cephalothin against Klebsiella. Amer. Intern. Med. 67:523-532. 4. Bulger, R. J., and U. Roosen-Runge. 1969. Bactericidal activity of the ampicillin/kanamycin combination against Escherichia coli, Enterobacter-Klebsiella and Proteus. Amer. J. Med. Sci. 258:7-13. 5. Curreri, P. W., H. M. Bruck, R. B. Lindberg, A. D. Mason, Jr., and B. A. Pruitt, Jr. 1973. Providencia stuartii sepsie: a new challenge in the treatment of thermal injury. Ann. Surg. 177:133-138. 6. Eickhoff, T. C. 1969. In vitro effect of carbenicillin combined with gentamicin or polymyxin B against Pseudomonas aeruginosa. Appl. Microiol. 18:469-473. 7. Fillastre, J. P., R. Laumonier, G. Humbert, D. Dubois, J. Metayer, A. Delpech, J. Leroy, and M. Robert. 1973. Acute renal failure associated with combined gentamicin and cephalothin therapy. Brit. Med. J. 2:396-397. 8. Hansten, P. D. 1973. Cephalothin, gentamicin, colistin

VOL. 5, 1974 CEPHALOSPORIN-AMINOGLYCOSIDE SYNERGY 577 hazards. J. Amer. Med. Ass. 223:1158. 9. Hyams, P. J., M. S. Simberkoff, and J. J. Rahal, Jr. 1973. In vitro bactericidal effectiveness of four aminoglycoside antibiotics. Antimicrob. Ag. Chemother. 3:87-94. 10. Klastersky, J., R. Cappel, G. Swings, and L. Vandenborre. 1971. Bacteriological and clinical activity of the ampicillin/gentamicin and cephalothin/gentamicin combinations. Amer. J. Med. Sci. 262:283-289. 11. Klastersky, J., A. Henri, and L. Vandenborre. 1973. Antimicrobial activity of tobramycin and gentamicin used in combination with cephalothin and carbenicillin. Amer. J. Med. Sci. 266:13-21. 12. Klastersky, J., G. Swings, and D. Daneau. 1970. Antimicrobial activity of the carbenicillin/gentamicin combination against gram-negative bacilli. Amer. J. Med. Sci. 260:373-380. 13. Kleinknecht, D., D. Ganeval, and D. Droz. 1973. Acute renal failure after high doses of gentamicin and cephalothin. Lancet 1:1129. 14. Milstoc, M., and P. Steinberg. 1973. Fatal septicemia due to Providence group bacilli. J. Amer. Geriat. Soc. 21:159-163. 15. Morris, A. H. 1973. Nebulizer contamination in a burn unit. Amer. Rev. Resp. Dis. 107:802-808. 16. Silverblatt, F., W. 0. Harrison, and M. Turck. 1973. Nephrotoxicity of cephalosporin antibiotics in experimental animals. J. Infect. Dis. 128:S367-S372. 17. Sonne, M., and E. Jawetz. 1969. Combined action of carbenicillin and gentamicin on Pseudomonas aeruginosa in vitro. Appl. Microbiol. 17:893-896.