In vitro activity of newer β-lactam agents in combination with amikacin against Pseudomonas aeruginosa, Klebsiella pneumoniae, and Serratia marcescens

In vitro activity of newer β-lactam agents in combination with amikacin against Pseudomonas aeruginosa, Klebsiella pneumoniae, and Serratia marcescens

DIAGNMICROBIOLINFECTDIS 1983;1:287-293 287 In Vitro Activity of Newer [3-Lactam Agents in Combination with Amikacin Against Pseudomonas aeruginosa, ...

433KB Sizes 0 Downloads 24 Views



In Vitro Activity of Newer [3-Lactam Agents in Combination with Amikacin Against Pseudomonas aeruginosa, Klebsiella pneumoniae, and Serratia marcescens Richard D. Meyer and Karen Pasiecznik

The in vitro activity of cefoperezone, ceflazidime, ceftizoxime, moxalactam, and N-formimidoyl thienamycin was evaluated alone and in combination with amikacin to assess possible synergistic activity against isolates of amikacin-resistant Pseudomonas aeruginosa and multidrugresistant Serratia marcescens and Klebsiella pneumoniae susceptible to amikacin (one S. marcescens isolate was also resistant to amikacin). The checkerboard agar dilution method was used. Ceftazidime and thienamycin followed by moxalactam and cefoperazone were the most active agents versus the P. aeruginosa alone and in combination testing. Ceftazidime, moxa|actam, and thienamycin showed the greatest activity against S. marcescens~ and all agents except cefoperazone were active against K. pneumoniae. The finding of synergy or partial synergy in combination testing was found in the majority with all three genera, including levels below the breakpoint for both amikacin and the p-lactam agents. This wide in vitro activity indicates that clinical evaluation of these agents in treatment of multidrug-resistant infections is warranted. INTRODUCTION Increasing resistance of nosocomial isolates of Pseudomonas aeruginosa and Enterobacteriaceae to cephalosporins and aminoglycosides is well appreciated. The advent of newer [3-1actam derivatives that have shown considerable activity against certain isolates represents one attempt to circumvent this problem. These agents include cefoperazone (Hinkle et al., 1980; McNamara et al., 1982), ceftazidime (Neu & Labthavikul, 1982), ceftizoxime (Kamimura et al., 1979; McNamara et al., 1982), moxalactam (Jones et al., 1960), and N-formimidoyl thienamycin (McNamara et al., 1982; Tally et al., 1960). Most but not all isolates are inhibited by one or more of these agents, particularly thienamycin; however, amikacin is also an active agent in vitro (McNamara et al., 1982). Amikacin, which has considerable activity against most isolates resistant to gentamicin and tobramycin by virtue of its stability vis-a-vis most plasmid-mediated enzymatic inactivation, has remained the most active available aminoglycoside since its introduction in 1976 (Meyer, 1961). Resistance of isolates to amikacin has arisen

From the Medical and Research Services, Veterans Administration Wadsworth Medical Center; the Department of Medicine, UCLA School of Medicine: and the Department of Medicine, Cedars-Sinai Medical Center-UCLA School of Medicine, Los Angeles, California. Address reprint requests to: Richard D. Meyer, M.D., Director, Division of Infectious Diseases, Department of Medicine, Cedars-Sinai Medical Center, Los Angeles, CA 90048. Received March 4, 1983; revised and accepted July 18, 1983.

© 1983 Elsevier Science Publishing Co., Inc.

52 Vanderbilt Avenue, New York. NY 10017



R.D. Meyer and K. Pasiecznik

in a variable number of patients treated and has involved decreased uptake more frequently than enzymatic inactivation (Meyer, 1977, 1981}. These isolates may also represent therapeutic challenges. Thus, consideration of the use of a [3-1actam agent in combination with amikacin in empirical use to avert possible resistance to one or the other agents is one approach; the possible synergistic in vitro effects are another appealing feature of combined therapy that is generally accepted (Jones and Packer, 1982; Kurtz et al., 1981). The present study was undertaken to determine the in vitro activity of selected newer [3-1actam agents alone and in combination with amikacin; isolates selected were amikacin-resistant Pseudomonas aeruginosa and Serratia marcescens and Klebsiella pneumoniae resistant to one or more of the [3-1actam agents and/or amikacin. MATERIALS AND METHODS

Isolates Thirty different clinical isolates (14 of P. aeruginosa, 8 of K. pneumoniae, and 8 of S. marcescens) obtained between 1975 and 1982 were used. Each isolate of P. aeruginasa was resistant to gentamicin by standard disk testing, with a zone of inhibition of 12 m m or less to a 10-~Lggentamicin disk (Bauer et al., 1966), and had a minimum inhibitory concentration (MIC) of 16 p.g/ml or more in agar dilution testing (Ericsson & Sherris, 1971). Further choice for testing was made by selecting isolates with MICs of 32 p.g/ml or more to amikacin in agar dilution testing. Immunotyping of and analysis for aminoglycoside-inactivating enzymes in representative amikacin-resistant isolates of P. aeruginosa during that period showed several immunotypes and no amikacin-inactivating enzymes, respectively (Meyer, 1977; unpublished observations). All isolates of K. pneumoniae and S. marcescens were resistant to cephalothin, with MICs of 32 p.g/ml or more in agar dilution testing (Ericsson & Sherris, 1971; McNamara et el., 1982); most were also resistant to cefamandole, cefoxitin, and gentamicin (McNamara et al., 1982). One amikacin-resistant isolate of S. marcescens is known to have a 6'-aminoglycoside-acetylating enzyme (Meyer, 1977). The isolates of S. marcescens and K. pneumoniae comprised a variety of serotypes (Lewis et al., 1977; Meyer et el., 1976).

Media Approximately 5 x 104 organisms from an overnight culture grown at 37°C in Mueller-Hinton broth were inoculated with a replicating device (Steers et al., 1959) onto media consisting of Mueller-Hinton broth solidified with 1.5% agar (Difco) and containing 5% laked defibrinated sheep blood in the agar plate dilution method recommended by the International Collaborative Study of the World Health Organization (Ericsson and Sherris, 1971). Antibiotics Antibiotic dilutions were prepared to contain amikacin, cefoperazone, ceftazidime, ceftizoxime, moxalactam, or thienamycin alone in twofold dilutions with a final concentration in agar of 128-0.125 ~g/ml. Amikacin sulfate was supplied by Bristol Laboratories, Syracuse, NY; cefoperazone by Pfizer, Inc., Groton, CT; ceftazidime by Glaxo, Inc., Fort Lauderdale, FL; ceftizoxime by Smith Kline and French Company, Philadelphia, PA: moxalactam by Eli Lilly and Company, Indianapolis, IN; and thienamycin by Merck Sharp and Dohme Company, Inc., West Point, PA. The following

Amikacin and ~-Lactams


MICs (Izg/ml) in agar with single agents were considered to indicate susceptibility: amikacin, 16 or less; cefoparazone, ceftazidime, ceftizoxime, and moxalactam, 32 or less; and thienamycin, 8 or less (Dr. Gary Calandra, Merck Sharp and Dohme Company, Inc., personal communication).

Determination of Synergy Synergism was determined by the checkerboard technique, previously shown to be comparable to other methods (Weinstein et al., 1975), using plates containing twofold dilutions of amikacin of 128-0.125 p,g/ml in combination with each of the ~-lactam agents in concentrations of 128-0.125 t~g/ml. Identical plates lacking the antibiotics were used as controls. Plates were generally prepared the same day, stored overnight at 4°C, and inoculated the following day. Cation Measurements Single lots of Mueller-Hinton broth and agar were used to minimize the effect on aminoglycoside testing (Gilbert etal., 1971). No supplementation of cations was used. Repeated freeze-thaw extraction on the agar medium was performed. Mean magnesium and calcium concentrations of the extracts determined by atomic absorption spectrophotometry were 1.52 and 3.50 mg/dl, respectively. Susceptibility Tests and Interpretations The MIC was recorded as the lowest concentration of antibiotic showing only a haze, one colony, or no growth after overnight incubation (Ericsson and Sherris, 1971). Pseudomonas aeruginosa ATCC 27853 and Escherichia cell ATCC 25922 were used as the reference strains and included in parallel tests. Synergy was defined as a fourfold or greater reduction of the MIC of both of the antibiotics. Partial synergy was defined as fourfold reduction in the MIC of one of the antibiotics but not the other compound (Fu and Nan, 1976). Antagonism was defined as a fourfold increase in the MIC of either compound. RESULTS The MICs for individual agents are listed in Table 1, the effect of the combined agents in Table 2. Geometric mean MICs for combinations are not shown because of the numerous permutations, but the number of isolates inhibited at or below the breakpoints for susceptibility of both agents is indicated.

Pseudomonas aeruginosa Ceftazidime and thienamycin showed the greatest in vitro activity against the 14 isolates selected for resistance to gentamicin and amikacin. Both moxalactam and cefoperazone were active under the breakpoint against half of the isolates; ceftizoxime was relatively inactive against the isolates of P. oeruginosa (Table 1). Synergy or partial synergy was found in all instances in which amikacin was combined with cefoperazone, ceftazidime, ceftizoxime, or moxalactam. When amikacin was used in combination with thienamycin, synergy or partial synergy occurred in the majority of isolates; in the others, the MICs were generally quite low, even if synergy did not occur. Overall, the rank order of activity of ~-lactam agents in combination with amikacin, when considering anticipated breakpoints, was thienamy-

R.D. Meyer and K. Pasiecznik


TABLE 1. Comparative Susceptibilities of Isolates to A m i k a c i n and ~-Lactam Agents Alone Range (p.g/ml)


Geometric mean MICa (~g/ml)

MIC50b (~g/ml)

MICgo~ (~g/ml)

Pseudomonas aeruginosa (14 isolates, amikacin-resistant)

Amikacin Cefoperazone Ceftazidime Ceftizoxime Moxalactam Thienamycin

32-128 8-128 2-16 64->256 8-128 2-64

64.0 39.0 5.1 148.5 35.3 5.9

64 32 4 128 32 4

128 128 16 256 64 16

8 >256 2 32 8 4

128 >256 16 64 32 18

2 >256 2 2 8 4

4 >256 32 128 16 4

Serratia marcescens (8 isolates)

Amikacin Cefoperazone Ceftazidime Ceftizoxime Moxalactam Thienamycin

2-128 128->256 2-16 0.5-64 2-32 2-16

9.5 394.8 4,0 19,0 10.4 5.7

Klebsiella pneumoniae (8 isolates)

Amikacin Cefoperazone Ceffazidime Ceftizoxime Moxalactam Thienamycin

2-4 4->256 0.5-32 0.5-128 2-16 2-4

2.4 279.2 2,6 3.7 8 3.1

Abbreviation: MIC = minimum inhibitory concentration.

°For an MIC of 256 ,g/ml or more. an MIC of 512 u.g/mlwas arbitrarilyused to calculategeometricmean MICs. bAmountnecessary to inhibit 50% of isolates. CAmountnecessary to inhibit 90% of isolates.

cin ~ ceftazidime > cefoperazone/> moxalactam > ceftizoxime (Table 2). The MICs (p,g/ml) for a representative isolate are amikacin, 64; cefoperazone, 8; ceftazidime, 2; ceftizoxime, 64; moxalactam, 16; t h i e n a m y c i n , 8; amikacin and cefoperazone, 4 and 4; a m i k a c i n a n d ceftazidime, 8 a n d 1; amikacin and ceftizoxime, 4 and 16; amikacin and moxalactam, 2 and 16; and amikacin and thienamycin, 0.25 and 4.



Seven of the gentamicin-resistant isolates were susceptible to amikacin. Of the [3lactam agents, ceftazidime a n d t h i e n a m y c i n showed the greatest in vitro activity, followed by moxalactam and ceftizoxime. Cefoperazone was considerably less active (Table 1). For the amikacin-susceptible isolates, the combination of amikacin with ceftizoxime, moxalactam, or t h i e n a m y c i n led to a partially or completely synergistic effect with all isolates. The c o m b i n a t i o n of cefoperazone and amikacin generally had no greater effect than amikacin alone.

Amikacin and ~-Lactarns


TABLE 2. Interactions Between Amikacin and ~)-Lactam Agents in Vitro Number (%) showing interaction when combined with amikacin Effect


Ceftazidime Ceftizoxime Moxalactam


Pseudomonas aeruginosa {14 isolates,amikacin-resistant) Synergy Partial synergy Indifference Inhibition below breakpoints for both agents

9 (64} 5 (36) 0 13{93)

8 (57} 6 (43} 0 14 (100)

13 (93} 1 {7] 0 11 (69)

8 {57} 6 {43} 13 (93)

2 {14} 7 C50) 5 (36) 14 (100)

Serratia marcesens (8 isolates}"

Synergy Partial synergy Indifference

1 (12} O 7 (88}

7 (88) 0 1 (12)

6 (75) 2 (25} 0

8 (100) 0 0

5 (62) 3 (38} 0

3 (38) 4 (50) 1 (12)

5 (62) 2 (25) 1 (12)

Klebsiello pneumoniae (8 isolates) a

Synergy Partial synergy Indifference

0 2 (25) 6 (75)

2 (25) 4 (50) 2 (25}

2 (25) 2 (25} 4 (50}

aAll isolates, except for one isolate of S. morcescens, susceptible to amikacin alone in vitro.

The one amikacin-resistant isolate showed inhibition by ceftazidime, moxalactam, and thienamycin, a pattern similar to that of the other isolates of S. marcescens. Synergy was produced by amikacin alone and in combination with any of the 6" lactam agents tested.

Kiebsiella pneumoniae Moxalactam and thienamycin showed a uniformly high degree of activity. Ceflazidime and ceftizoxime also showed a high degree of activity, with the exception of one isolate versus each agent. Cefoperazone was generally inactive. Partial and/or complete synergy was seen with a moderate number of isolates tested with amikacin and ceftazidime, ceftizoxime, moxalactam, and thienamycin. When cefoperazone was combined with amikacin, there was generally no reduction in MICs from the very high levels observed when it was tested alone. Overall, the MICs in single-agant testing and in combination studies were quite low (compared to those in isolates of P. aeruginosa and S. marcescens) with amikacin and/or ceftazidime, ceftizoxime, moxalactam, or thienamycin. DISCUSSION These results indicate that several of the newer [3-1actam agents (alone or in combination with amikacin), particularly ceftazidime and thienamycin, are very active in vitro against amikacin-resistant P. aeruginosa. Antagonism is not found. These agents, and to a lesser degree moxalactam and cefoperazone, which also show a high degree of in vitro activity, resulted in reduction of the MICs of amikacin to or below the breakpoint in about half of all instances. Ceftizoxime showed little effect against


R.D. Meyer and K. Pasiecznik

P. aeruginosa herein and in other studies (Kamimura et al., 1979; McNamara et al., 1982). Other studies have shown increased effects of the newer [3-1actam agents and antipseudomonal penicillins against P. aeruginosa and Enterobacteriaceae (Fass, 1982; Mintz and Drew, 1981; Scribner et al., 1982; Yu et al., 1983). A minority of the strains tested in these studies were multidrug-resistant; the isolates tested, however, showed synergy in vitro (Fass, 1982; Mintz and Drew, 1981; Yu et al., 1983). Studies by Kurtz et al. (1981) showed that moxalactam, cefotaxime, or cefoperazone in combination with amikacin led to synergistic antibacterial activity against P. aeruginosa and S. marcescens in 40-70% of isolates tested. Most of their isolates were susceptible to amikacin, but with amikacin-resistant P. aeruginosa the combination of moxalactam and amikacin was the most active of those tested. These results agree with ours; however, we also found that thienamycin and ceftazidime were more active in vitro than moxalactam. The results of testing of the [3-1actam agents alone and in combination with amikacin against gentamicin-resistant isolates of S. marcescens and K. pneumoniae indicate that ceftazidime, ceftizoxime, moxalactam, and thienamycin are quite active; cefoperazone showed little or no activity. The latter results differ from those of Kurtz et al. (1981), who found a modicum of activity of cefoperazone in combination with amikacin against amikacin-susceptible isolates of S. marcescens. Our results also agree with those of Jones and Packer (1982), who found that the combination of amikacin with cefoperazone, ceftazidime, or moxalactam led to a synergistic effect in vitro with most isolates of Enterobacteriaceae. They selected moderately susceptible isolates with MICs to the ~-lactam agents of 2-32 ~,g/ml. The one amikacin-resistant isolate of S. marcescens studied herein was inhibited by all of the [3-1actam agents alone or in combination except cefoperazone. In four of five instances the MIC of amikacin was lowered synergistically to achievable levels. As we do not know the mechanism of resistance to ~-lactam agents in this multidrugresistant isolate, we do not know the mechanism of interaction between the ~-lactam agent(s) and amikacin. It is important that, as with amikacin-resistant P. aeruginosa, most of the amikacinsusceptible isolates of S. marcescens and K. pneumoniae were inhibited by low concentrations of each of four [3-1actam agents. Whether a single agent would be useful in therapy of infections caused by such isolates is speculative pending prospective clinical trials. The results indicate that broad generalizations can be made about the susceptibility of multidrug isolates, but that specific testing, not only with individual agents but with combinations, is necessary for individual isolates; Yu et al, (1983) also emphasized the need for individual testing with isolates of P. aeruginosa. Amikacin combined with one of the newer [3-1actam agents provides a very broad spectrum of activity against nosocomial multidrug-resistant isolates. However, the clinical utility and possible role in preventing emergence of resistance by use of one or more of the newer [3-1actam agents in combination with amikacin in immunocompromised patients or those with serious infections remain to be shown by controlled clinical studies. The authors thank Jack Coburn for the magnesium and calcium determinations, George Hough for the immunotyping, and Peter Kresel for the analysis for aminoglycoside-inactivating enzymes in isolates of Pseudomonas aeruginosa. This work was supported by a grant from Bristol Laboratories, Syracuse, NY, by the Medical Research Service of the Veterans Administration, and by a BRSG-NIHgrant at Cedars-Sinai Medical Center.

A m i k a c i n a n d ~-Lactams


REFERENCES Bauer AW, Kirby WMM, Sherris JC, Turck M (1966) Antibiotic susceptibility testing by a standardized single disk method. Am J Pathol 45:493. Ericsson HM, Sherris JC (1971) Antibiotic sensitivity testing: Report of an international collaborative study. Acta Pathol Microbiol Scand Sect B Suppl 217:9. Fass RJ (1982) Comparative in vitro activities of 13 -lactam-tobramycin combinations against Pseudomonas aeruginosa and multidrug-resistant gram-negative enteric bacilli. Antimicrob Agents Chemother 21:1003. Fu KP, Nan HC (1976} In vitro synergistic effect of netilmicin, a new amiuoglycoside antibiotic. Antimicrob Agents Chemother 10:511. Gilbert DN, Kutscher E, Ireland R, Barnett JA, Sanford JP (1971) Effect of the concentrations of magnesium and calcium on the in vitro susceptibility of Pseudomonas aeruginosa to gentamicin. ] Infect Dis 124:$37. Hinkle AM, LeBlanc, BM, Bodey GP (1980) In vitro evaluation of cefoperazone. Antimicrob Agents Chemother 17:423. Jones RN, Fuchs PC, Sommers HM, Gavan TL, Barry AL, Gerlach EH (1980) Moxalactam (LY127935}, a new semisynthetic 1-oxa-~-lactam antibiotic with remarkable antimicrobial activity: In vitro comparison with cefamandole and tobramyciu. Antimicrob Agents Chemother 17:750. Jones RN, Packer RR (1982) Antimicrobial activity of amikacin combinations against Enterobacteriaceae moderately susceptible to third-generation cephalosporins. Antimicrob Agents Chemother 22:985. Kamimura T, Matsumoto Y, Okada N, Mine Y, Nishida M, Goto S, Kuwahara S (1979) Ceftizoxime (FK 749), a new parenteral cephalosporin: In vitro and in rive antibacterial activities. Antimicrob Agents Chemother 16:540. Kurtz TO, Winston DJ, Bruckner DA, Martin WJ (1981) Comparative in vitro synergistic activity of new beta-lactam antimicrobial agents and amikacin against Pseudomonas aeruginosa and Serratia marcescens. Antimicrob Agents Chemother 18:642. Lewis RP, Meyer RD, Finegold SM {1977} Amikacin therapy of patients with gentamicin-resistant gram-negative bacillary infections. Am J Med (Suppl) 62:142. McNamara B, Meyer RD, Pasiecznik KA (1982} In vitro susceptibility of cephalothin-resistant Enterobacteriaceae and Pesudomonas aeruginosa to amikacin and selected new B-lactam agents. Antirnicrob Agents Chemother 21:753. Meyer RD {1977) Patterns and mechanisms of emergence of resistance to amikacin. ] Infect Dis 138:449. Meyer RD (1981} Drugs five years later: Amikacin. Ann Intern Med 95:328. Meyer RD, Lewis RP, Halter J, White M (1976} Gentamicin-resistant Pseudomonas aeruginosa and Serratia marcescens in a general hospital. Lancet 1:580. Mintz L, Drew WL (1981) Comparative synergistic activity of cefoperazone, cefotaxime, moxalactam, and carbenicillin, combined with tobramycin, against Pseudomonas aeruginosa. Antimicrob Agents Chemother 19:332. Neu HC, Labthavikul P (1982} Antibacterial activity and B-lactamase stability of ceftazidime, an aminothiazolyl aephalosporin potentially active against Pseudomonas aeruginosa. Antimicrob Agents Chemother 21:11. Scribner RK, Marks MI, Weber AH, Tarpay MM, Welch DF (1982} Activity of various 13-1actams and aminoglycosides, alone and in combination, against isolates of Pseudomonas aeruginosa from patients with cystic fibrosis. Antimicrob Agents Chemother 21:939. Steers E, Foltz EL, Graves BS (1959) An inocula replicating apparatus for routine testing of bacterial susceptibility to antibiotics. Antibiot Chemother 9:307. Tally FP, Jacobus NV, Gorbach SL {1980} In vitro activity of N-formimidoyl thienamycin (MK 0787}. Antimicrob Agents Chemother 18:642. Weinstein RJ, Young LS, Hewitt WL (1975) Comparison of methods for assessing in vitro antibiotic synergism against Pseudomonas and Serratia. ] Clin Lab Med 86:853. Yu PKW, Edson RS, Washington JA II, Hermans PE {1983) Bactericidal and synergistic activity of moxalactam alone and in combination with gentamicin against Pseudomonas aeruginosa. Antimicrob Agents Chemother 23:179.