In vitro activity of ciprofloxacin, levofloxacin, and trovafloxacin, alone and in combination with β-lactams, against clinical isolates of Pseudomonas aeruginosa, Stenotrophomonas maltophilia, and Burkholderia cepacia

In vitro activity of ciprofloxacin, levofloxacin, and trovafloxacin, alone and in combination with β-lactams, against clinical isolates of Pseudomonas aeruginosa, Stenotrophomonas maltophilia, and Burkholderia cepacia

ANTIMICROBIAL SUSCEPTIBILITY STUDIES In Vitro Activity of Ciprofloxacin, Levofloxacin, and Trovafloxacin, Alone and in Combination with b-Lactams, ag...

155KB Sizes 0 Downloads 29 Views

ANTIMICROBIAL SUSCEPTIBILITY STUDIES

In Vitro Activity of Ciprofloxacin, Levofloxacin, and Trovafloxacin, Alone and in Combination with b-Lactams, against Clinical Isolates of Pseudomonas aeruginosa, Stenotrophomonas maltophilia, and Burkholderia cepacia Henry D. Isenberg, Phyllis Alperstein, and Kenneth France

We tested three fluoroquinolones (ciprofloxacin, levofloxacin, and trovafloxacin), each combined with each of four b-lactams (cefoperazone, ceftriaxone, imipenem, and meropenem) for synergy against clinical isolates of nosocomial strains of Pseudomonas aeruginosa, Stenotrophomonas maltophilia, and Burkholderia cepacia. The ciprofloxacin–b-lactam combinations showed synergy against none or only a small fraction (7 to 10%) of the P. aeruginosa and B. cepacia isolates. Ciprofloxacin-cefoperazone, -ceftriaxone, and -meropenem were synergic against 50%, 25%, and 30% of the S. maltophilia isolates, respectively. Among the levofloxacin combinations, only those with cefoperazone and imipenem showed significant

synergy, and this only against B. cepacia (50% and 30%, respectively). Trovafloxacin-cefoperazone and -imipenem showed modest synergy against P. aeruginosa (23% and 27%, respectively), as did trovafloxacin-cefoperazone and -ceftriaxone against B. cepacia (30%). The trovafloxacinimipenem combination was synergic against all isolates of B. cepacia. Because of their synergy, the following combinations may be useful in the nosocomial setting: trovafloxacincefoperazone or -imipenem against P. aeruginosa; ciprofloxacin-cefoperazone, -ceftriaxone, or -meropenem against S. maltophilia; levofloxacin-cefoperazone and trovafloxacinimipenem against B. cepacia. © 1999 Elsevier Science Inc.

INTRODUCTION

have been isolated from a variety of sources, including hand-washing sinks, flower vases, disinfectant solutions, soaps, and irrigation fluids. Because of their ubiquity, they have become important agents of nosocomial infections that range from uncomplicated skin infections to fulminating sepsis and pneumonia. B. cepacia frequently is found in lung abscesses of cystic fibrosis patients who have died of lung failure (Gilligan 1995). Because they are unusually resistant to acceptable doses of many antibiotics (Gilligan 1995; von Graevenitz 1995), nosocomial infections by P. aeruginosa, S. maltophilia, and B. cepacia frequently are difficult to

The motile Gram-negative bacilli Pseudomonas aeruginosa, Stenotrophomonas (formerly Xanthomonas) maltophilia, and Burkholderia cepacia have a singular ability to thrive in aqueous environments. In hospitals, they From the Division of Microbiology, Long Island Jewish Medical Center, Long Island Campus for Albert Einstein College of Medicine, New Hyde Park, New York, USA. Address reprint requests to Dr. Henry D. Isenberg, Chief Emeritus/Consultant, Division of Microbiology, Long Island Jewish Medical Center, New Hyde Park, NY 11042. Received 29 July 1998; accepted 18 September 1998.

DIAGN MICROBIOL INFECT DIS 1999;33:81–86 © 1999 Elsevier Science Inc. All rights reserved. 655 Avenue of the Americas, New York, NY 10010

0732-8893/99/$–see front matter PII S0732-8893(98)00126-6

H.D. Isenberg et al.

82 treat. Bacteria resistant to single antibiotics can sometimes be inhibited by antibiotic combinations with synergic activity. Combinations that are synergic in vitro may also be synergic in the patient. Therefore, testing new antibiotics for synergy in the laboratory may point to clinically useful combinations of agents. The objective of our study was (a) to compare the in vitro activities of an older fluoroquinolone antibiotic, ciprofloxacin, two new fluoroquinolones, levofloxacin and trovafloxacin, and four b-lactam antibiotics against recent clinical isolates of P. aeruginosa, S. maltophilia, and B. cepacia and (b) to test each of the three fluoroquinolones combined with each of the four b-lactams to determine their synergic activity against the same isolates.

MATERIALS AND METHODS Bacterial Strains The bacteria used were selected at random from isolates of P. aeruginosa (n 5 30), S. maltophilia (n 5 20), and B. cepacia (n 5 10) between 1 September 1996 and 1 March 1997 from patients at Long Island Jewish Medical Center, New Hyde Park, NY, USA. Each strain tested came from a different patient. All isolates were identified according to standard methods (Isenberg 1992) and kept frozen at 270°C until use. American Type Culture Collection (ATCC) P. aeruginosa strain 27853 served as the control in all test sequences.

Susceptibility Testing The susceptibilities of the isolates to ciprofloxacin, levofloxacin, trovafloxacin, cefoperazone, ceftriaxone, imipenem, and meropenem were determined by the agar dilution method in accordance with the guidelines of the National Committee for Clinical Laboratory Standards (NCCLS 1993). Cationsupplemented Mueller-Hinton agar (Difco Laboratories, Detroit, MI, USA) was used as the culture medium for all tests. Standard inocula were prepared by suspending colonies in 0.9% NaCl solution to match the density of a 0.5 McFarland barium sulfate suspension and diluting them 1:10. The inocula were delivered by a Steers replicator (Craft Machines, Inc., Chester, PA, USA) to agar plates containing twofold antibiotic dilutions ranging from 256.0 to 0.5 mg/L. A control agar plate without antibiotic accompanied each sequence. All tests were run in duplicate and incubated at 35°C for 18 to 24 h. Results were recorded as MICs. MIC50 and MIC90 were defined as the lowest antibiotic concentration that inhibited the visible growth of 50% and 90% of the isolates, respectively. NCCLS (1997) standard MIC breakpoints

were used to define susceptibility for ciprofloxacin and levofloxacin. Susceptibility to trovafloxacin was defined by FDA-approved MIC breakpoints. (The breakpoints used are given in the footnote to Table 1).

Synergy Testing Each of the three fluoroquinolones combined with each of the four b-lactams was tested for synergy against each of the isolates by the standard agar dilution method described above and the checkerboard technique (Andriole et al. 1971; Dembry et al. 1997). For each test, Mueller-Hinton agar plates were prepared to contain combinations of doubling dilutions of the fluoroquinolone, ranging from 256 to 0.5 mg/L, and each doubling dilution of the b-lactam, which also ranged from 256 to 0.5 mg/L. One agar plate containing no antibiotic served as a positive control. Tests were run in duplicate. The activity of the antibiotic combinations was classified according to the following criteria (Dembry et al. 1997): Synergy. The MIC of the fluoroquinolone in combination was at least four times lower than the MIC of the fluoroquinolone alone, and the MIC of the b-lactam in combination was at least four times lower than the MIC of the b-lactam alone. Addition. The MIC of the fluoroquinolone in combination was at least four times lower than the MIC of the fluoroquinolone alone, and the MIC of the b-lactam in combination was at least two times lower than the MIC of the b-lactam alone; or the MIC of the b-lactam in combination was at least four times lower than the MIC of the b-lactam alone, and the MIC of the fluoroquinolone in combination was at least two times lower than the MIC of the fluoroquinolone alone. Indifference. The MIC of the fluoroquinolone in combination was two times lower than the MIC of the fluoroquinolone alone, and the MIC of the b-lactam in combination was two times lower than the MIC of the b-lactam alone; or the MIC of the fluoroquinolone or the b-lactam in combination was the same as the MIC of that antibiotic alone. Antagonism. The MIC of the fluoroquinolone in combination was two times greater than the MIC of the fluoroquinolone alone, and the MIC of the b-lactam in combination was two times greater than the MIC of the b-lactam alone. Indeterminate. The MIC of the fluoroquinolone or the b-lactam alone was .256 mg/L, and the MIC of the same fluoroquinolone or b-lactam in combination was 128, 256, or .256 mg/L.

RESULTS The results obtained with the ATCC control strain were within the expected range for all the tests per-

Synergy of Fluoroquinolone–b-lactam Combinations

83

TABLE 1 In Vitro Activity of Three Fluoroquinolone and Four b-Lactam Antibiotics against Isolates of Three Gram-negative Bacteria MIC (mg/L) Bacterium (n)

Antibiotic

MIC50

MIC90

Range

Susceptibilitya (%)

P. aeruginosa (30)

ciprofloxacin levofloxacin trovafloxacin cefoperazone ceftriaxone imipenem meropenem ciprofloxacin levofloxacin trovafloxacin cefoperazone ceftriaxone imipenem meropenem ciprofloxacin levofloxacin trovafloxacin cefoperazone ceftriaxone imipenem meropenem

#0.5 #0.5 1 32 32 2 #0.5 2 #0.5 #0.5 128 128 .256 128 256 128 256 .256 64 16 8

4 4 8 128 .256 4 2 8 2 2 256 .256 .256 .256 .256 256 .256 .256 .256 128 16

#0.5–8 #0.5–16 #0.5–16 #0.5–.256 1–.256 1–8 #0.5–4 #0.5–16 #0.5–2 #0.5–16 8–.256 32–.256 32–.256 8–.256 2–.256 4–256 4–.256 8–.256 4–.256 2–128 #0.5–.256

80 83 70 37 33 90 100 25 100 90 20 0 0 0 0 0 0 10 0 20 10

S. maltophilia (20)

B. cepacia (10)

NCCLS MIC breakpoints (mg/L) for susceptibility, intermediate susceptibility, and resistance: ciprofloxacin, levofloxacin #1, 2, $4; cefoperazone, #16, 32, $64; ceftriaxone, #8, 16–32, $64; imipenem and meropenem, #4, 8, $16 FDA-approved MIC breakpoints (mg/L) for trovafloxacin, #2, 4, $8.

a

formed in the study. Table 1 presents the in vitro activities of the fluoroquinolones and the b-lactams against the isolates of the three bacteria. All three fluoroquinolones showed comparable modest activity against P. aeruginosa. Levofloxacin and trovafloxacin were more effective against S. maltophilia than was ciprofloxacin. None of the fluoroquinolones was active against B. cepacia. Cefoperazone and ceftriaxone were inactive against the vast majority of the isolates of all three bacteria, as were imipenem and meropenem against S. maltophilia. Imipenem and meropenem were very active against P. aeruginosa. Tables 2– 4 summarize the in vitro activities of each of the fluoroquinolones in combination with each of the b-lactams. The four ciprofloxacin–blactam combinations showed synergy against none or only a small fraction of the P. aeruginosa and B. cepacia isolates. The ciprofloxacin-cefoperazone, -ceftriaxone, and -meropenem combinations, however, were synergic against 50%, 25%, and 30% of the S. maltophilia isolates, respectively. Among the levofloxacin combinations, only those with cefoperazone and imipenem showed significant synergy, and this only against B. cepacia (50% and 30% of isolates, respectively). Trovafloxacin-cefoperazone, -ceftriaxone, and -imipenem showed modest synergy against P. aeruginosa, as did trovafloxacin-cefoperazone and -ceftriaxone against B. cepacia. The trovafloxacin-

imipenem combination was synergic against all 10 isolates of B. cepacia. The combinations with the greatest synergy against P. aeruginosa, S. maltophilia, and B. cepacia, respectively, were trovafloxacin-imipenem (27% of isolates), ciprofloxacin-cefoperazone (50% of isolates), and trovafloxacin-imipenem (100% of isolates). Of the 240 synergy tests run with each fluoroquinolone (4 combinations against each of 60 isolates), 33 (14%) of the ciprofloxacin, 29 (12%) of the levofloxacin, and 46 (19%) of the trovafloxacin tests exhibited synergy. Four of the ciprofloxacin synergic outcomes (all against S. maltophilia), six of the levofloxacin synergic outcomes (two against P. aeruginosa and four against B. cepacia), and five of the trovafloxacin synergic outcomes (all against B. cepacia) were achieved with antibiotic concentrations that were equal to or greater than the NCCLS resistance breakpoint for one or both of the antibiotics in the combination (see footnotes to Tables 2– 4). Among the 720 synergy tests performed in the study, only five instances of antagonism were observed.

DISCUSSION Antibiotic combinations are used for a variety of reasons, one of which is to achieve antibiotic syn-

H.D. Isenberg et al.

84

TABLE 2 In Vitro Activity of Ciprofloxacin in Combination with Four b-Lactams against Isolates of Three Gram-negative Bacteria Bacterium/Antibiotic Combination P. aeruginosa (n 5 30) ciprofloxacin plus cefoperazone ceftriaxone imipenem meropenem S. maltophilia (n 5 20) ciprofloxacin plus cefoperazone ceftriaxone imipenem meropenem B. cepacia (n 5 10) ciprofloxacin plus cefoperazone ceftriaxone imipenem meropenem

Activity of Combination [number (%) of isolates] Synergy

Addition

Indifference

3 (10) 2 (7) 3 (10)

4 (13) 6 (20) 5 (17) 3 (10)

23 (77) 22 (73) 22 (73) 27 (90)

10 (50)a 5 (25) 2 (10)b 6 (30)

4 (20) 7 (35) 4 (20) 3 (15)

6 (30) 8 (40) 14 (70) 10 (50)

1 (10) 2 (20) 3 (30) 3 (30)

6 (60) 4 (40) 5 (50) 6 (60)

1 (10) 1 (10)

Antagonism

Indeterminate

1 (5)

1 (10) 1 (10)

3 (30) 2 (20) 1 (10)

a

Synergy against three isolates achieved at a concentration equal to or greater than the resistance breakpoint for one or both of the antibiotics. b Synergy against one isolate achieved at a concentration equal to or greater than the resistance breakpoint for one or both of the antibiotics.

TABLE 3 In Vitro Activity of Levofloxacin in Combination with Four b-Lactams against Isolates of Three Gram-negative Bacteria Bacterium/Antibiotic Combination P. aeruginosa (n 5 30) levofloxacin plus cefoperazone ceftriaxone imipenem meropenem S. maltophilia (n 5 20) levofloxacin plus cefoperazone ceftriaxone imipenem meropenem B. cepacia (n 5 10) levofloxacin plus cefoperazone ceftriaxone imipenem meropenem

Activity of Combination [number (%) of isolates] Synergy

Addition

Indifference

4 (13) 4 (13)a 3 (10) 1 (3)

8 (27) 6 (20) 8 (27) 2 (7)

18 (60) 20 (67) 19 (63) 27 (90)

3 (15) 1 (5)

3 (15)

14 (70) 19 (95) 19 (95) 18 (90)

1 (5) 2 (10)

5 (50)a 2 (20) 3 (30)a 1 (10)

2 (20) 5 (50) 3 (30) 5 (50)

2 (20) 3 (30) 4 (40) 4 (40)

Antagonism

Indeterminate

1 (10)

a

Synergy against two isolates achieved at a concentration equal to or greater than the resistance breakpoint for one or both of the antibiotics.

ergy, that is, to increase the inhibition or killing of organisms resistant to acceptable doses of single antibiotics. Many studies have been conducted to determine the in vitro activity of fluoroquinolones

combined with other antibiotics. Most of these investigations have been conducted using the checkerboard technique, which is recognized as having two important drawbacks (King et al. 1981). First, it gen-

Synergy of Fluoroquinolone–b-lactam Combinations

85

TABLE 4 In Vitro Activity of Trovafloxacin in Combination with Four b-Lactams against Isolates of Three Gram-negative Bacteria Bacterium/Antibiotic Combination P. aeruginosa (n 5 30) trovafloxacin plus cefoperazone ceftriaxone imipenem meropenem S. maltophilia (n 5 20) trovafloxacin plus cefoperazone ceftriaxone imipenem meropenem B. cepacia (n 5 10) trovafloxacin plus cefoperazone ceftriaxone imipenem meropenem

Activity of Combination [number (%) of isolates] Synergy

Addition

Indifference

7 (23) 5 (17) 8 (27) 1 (3)

8 (27) 7 (23) 6 (20) 6 (20)

15 (50) 17 (57) 16 (53) 23 (77)

2 (10) 1 (5) 3 (15) 2 (10)

3 (30)a 3 (30)a 10 (100)b 1 (10)

2 (10)

18 (90) 18 (90) 15 (75) 16 (80)

1 (10) 1 (10)

2 (20) 1 (10)

3 (30)

7 (70)

Antagonism

Indeterminate

1 (3)

1 (5) 1 (5)

1 (5)

4 (40) 5 (50)

a

Synergy against two isolates achieved at a concentration equal to or greater than the resistance breakpoint for one or both of the antibiotics. b Synergy against one isolate achieved at a concentration equal to or greater than the resistance breakpoint for one or both of the antibiotics.

erates discontinuous (growth-or-no-growth) data rather than continuous information on antibiotic response, such as might be obtained with a time-kill study. Thus, the checkerboard approach does not permit a precise definition of the effect of antibiotic concentrations below or above the MIC. Second, the checkerboard technique relies on the assumption that inhibition of bacterial growth by each of the antibiotics in a combination follows a linear doseresponse curve, an assumption that is not always valid. Despite these methodologic weaknesses, the checkerboard technique is adequate for a preliminary appraisal of synergy, especially when a detailed assessment is not necessary (e.g., when neither drug is effective when used alone). Generally, studies of the in vitro activity of fluoroquinolones combined with other antibiotics have shown only additive or indifferent activity, with only occasional synergy or antagonism (Neu 1991). Combinations of fluoroquinolones with penicillins or imipenem have exhibited synergy against 20 to 50% of isolates of P. aeruginosa, but fluoroquinoloneaminoglycoside combinations generally have not been synergic against this species (Neu 1992). Ciprofloxacin has shown synergy against P. aeruginosa when combined with azlocillin, aztreonam, or ceftazidime (Bustamante et al. 1990). Little has been published on the activity of levofloxacin combined with other antibiotics against P. aeruginosa, S. maltophilia, or B. cepacia. No synergy or antagonism was ob-

served when levofloxacin was combined with zidovudine in vitro against P. aeruginosa (Lewin et al. 1990). Using a standard microdilution broth method and the checkerboard technique, Dembry and associates (1997) tested combinations of ciprofloxacin or trovafloxacin and ampicillin-sulbactam or gentamicin against P. aeruginosa (n 5 25) and S. maltophilia (n 5 25). Only one instance of synergy was noted: the ciprofloxacin-gentamicin combination had a synergic effect on one isolate of S. maltophilia. In our study, significant synergy was observed with a number of fluoroquinolone–b-lactam combinations, and a complete lack of synergy was rare (ciprofloxacinmeropenem showed no synergy against P. aeruginosa, ciprofloxacin-cefoperazone, and -imipenem were not synergic against B. cepacia, and levofloxacin-imipenem failed to show synergy against S. maltophilia). Using checkerboard titration to determine fractional inhibitory concentration indices, Visalli and colleagues (1997), found that trovafloxacin-ceftazadime and trovafloxacin-imipenem had moderate synergy (28% and 23%, respectively) against P. aeruginosa (n 5 60) and that trovafloxacin-ceftazidime had significant synergy (89%) against S. maltophilia (n 5 36). The trovafloxacin-imipenem combination, however, was synergic against only 2 of 32 isolates of B. cepacia. In our study, by contrast,

86 trovafloxacin-imipenem had a synergic effect on all 10 of the B. cepacia isolates. Such extreme divergence in the results of studies of combinations of fluoroquinolones with other antibiotics is not uncommon (Neu 1992) and probably is attributable less to methodologic differences than to variability in the bacterial strains themselves. Indeed, it has been suggested that a fluoroquinolone should be combined with an antibiotic from another class not to achieve synergy but to inhibit bacteria inadequately curtailed by the fluoroquinolone alone (Bustamante et al. 1990). Infections with P. aeruginosa, S. maltophilia, and B. cepacia, however, often are unusually resistant to antibiotic therapy, and, consequently, a potential for synergy may be more important in their treatment than it is in the treatment of infections by less resistant species.

H.D. Isenberg et al. We conclude from our findings that, because of their synergy, the following combinations may be useful in the nosocomial setting: trovafloxacincefoperazone or -imipenem against P. aeruginosa; ciprofloxacin-cefoperazone, -ceftriaxone, or -meropenem against S. maltophilia; and levofloxacin- or trovafloxacin- cefoperazone, -ceftriaxone, or -imipenem against B. cepacia. Animal studies and clinical trials are needed to confirm the relevance of these combinations.

Supported by an independent medical grant from Pfizer Inc., U.S. Pharmaceuticals, New York, NY.

REFERENCES Andriole VT (1971) Synergy of carbenicillin and gentamicin in experimental infection with Pseudomonas. J Infect Dis 124(suppl):S46–S55. Bustamante CI, Wharton RC, Wade JC (1990) In vitro activity of ciprofloxacin in combination with ceftazidime, aztreonam, and azlocillin against multiresistant isolates of Pseudomonas aeruginosa. Antimicrob Agents Chemother 34:1814–1815. Dembry LM, Farrel PA, Orcutt DR, Gerrity LA, Andriole VT (1997) In-vitro activity of trovafloxacin against sensitive and resistant aerobic bacteria using the standard microdilution broth method and Etest. J Antimicrob Chemother 39(suppl B):35–42. Gilligan PH (1995) Pseudomonas and Burkholderia. In: Manual of Clinical Microbiology, 6th Edition. Eds, Murray PR, Baron EJ, Pfaller MA, Tenover FC, and Yolken RH. Washington, DC: American Society for Microbiology, pp. 509–519. Isenberg HD (Ed) (1992) Clinical Microbiology Procedures Handbook. Washington, DC: American Society for Microbiology. King TC, Schlessinger D, Krogstad DJ (1981) The assessment of antimicrobial combinations. Rev Infect Dis 3:627–633. Lewin CS, Allen RA, Amyes SGB (1990) Antibacterial ac-

tivity of fluoroquinolones in combinations with zidovudine. J Med Microbiol 33:127–131. National Committee for Clinical Laboratory Standards (1993) Methods for Dilution Antimicrobial Susceptibility Tests for Bacteria That Grow Aerobically—Third Edition. Approved Standard M7–A3. Villanova, PA: NCCLS. National Committee for Clinical Laboratory Standards (1997) Performance Standards for Antimicrobial Susceptibility Testing. M100–S7. Wayne, PA: NCCLS. Neu HC (1991) Synergy and antagonism of combinations of quinolones. Europ J Clin Microbiol Infect Disease 10: 255–261. Neu HC (1992) Quinolone antimicrobial agents. Annu Rev Med 43:465–486. Visalli MA, Bajaksouzian S, Jacobs MR, Appelbaum PC (1997) Comparative activity of trovafloxacin, alone and in combination with other agents, against gramnegative nonfermentative rods. Antimicrob Agents Chemother 41:1475–1481. von Graevenitz A (1995) Acinetobacter, Alcaligenes, Moraxella, and other nonfermentative gram-negative bacteria. In: Manual of Clinical Microbiology, 6th Edition. Eds, Murray PR, Baron EJ, Pfaller MA, Tenover FC, and Yolken RH. Washington, DC: American Society for Microbiology, pp. 520–532.