Previous Article | Next Article ![]()
Antimicrobial Agents and Chemotherapy, June 2003, p. 1902-1906, Vol. 47, No. 6
0066-4804/03/$08.00+0 DOI: 10.1128/AAC.47.6.1902-1906.2003
Copyright © 2003, American Society for Microbiology. All Rights Reserved.
Infectious Diseases Biology, Pharmacia Corporation, Kalamazoo, Michigan 49007
Received 25 November 2002/ Returned for modification 27 January 2003/ Accepted 26 February 2003
|
|
|---|
|
|
|---|
Linezolid is an antimicrobial agent from the oxazolidinone class and is especially active against multidrug-resistant gram-positive organisms such as methicillin-resistant Staphylococcus aureus, vancomycin-resistant Enterococcus faecalis and Enterococcus faecium, and penicillin-resistant Streptococcus pneumoniae. Linezolid has also been shown to have activity against certain gram-negative and anaerobic bacteria (7, 11, 13, 14, 19, 23). This drug was tested in vitro alone and in combination with 35 antimicrobial agents against multiple bacterial species by using checkerboard broth microdilution plates prepared by a semiautomated method. FIC indices were generated and analyzed for synergism, antagonism, and indifference.
The objective of this study was to determine the effects of the interaction of linezolid when it was combined with other antimicrobial agents and tested against multiple strains of drug-sensitive and -resistant S. aureus, E. faecalis, E. faecium, and S. pneumoniae. Selected strains of Escherichia coli and Klebsiella pneumoniae were tested as well.
(This study was presented in part at the 39th Interscience Conference on Antimicrobial Agents and Chemotherapy, San Francisco, Calif., 26 to 29 September 1999; the 40th Interscience Conference on Antimicrobial Agents and Chemotherapy, Toronto, Ontario, Canada, 17 to 30 September 2000; and the 41st Interscience Conference on Antimicrobial Agents and Chemotherapy, Chicago, Ill., 16 to 19 December 2001.)
|
|
|---|
Antimicrobial agents. The following antibiotic powders were used in the experiments and were obtained from the indicated sources: amoxicillin, ampicillin, bacitracin, cefoxitin, erythromycin, fusidic acid, gentamicin, methicillin, metronidazole, nalidixic acid, norfloxacin, novobiocin, ofloxacin, rifampin, tetracycline, and vancomycin, Sigma Chemical Company (St. Louis, Mo.); ceftazidime, cephalothin, chloramphenicol, clindamycin, and oxacillin, U.S. Pharmacopeia (Rockville, Md.); ampicillin-sulbactam and trovafloxacin, Pfizer (New York, N.Y.); amoxicillin-clavulanic acid, GlaxoSmithKline (Research Triangle Park, N.C.); cefdinir and sparfloxacin, Parke-Davis (Ann Arbor, Mich.); aztreonam and gatifloxacin, Bristol-Myers Squibb (Princeton, N.J.); cefotaxime, Calbiochem (La Jolla, Calif.); imipenem, Merck (Whitehouse Station, N.J.); ciprofloxacin, Miles (Kankakee, Ill.); teicoplanin, Aventis Pharmaceuticals (Bridgewater, N.J.); and cefpodoxime, linezolid, and neomycin, Pharmacia Corporation (Kalamazoo, Mich.).
Susceptibility and FIC testing. Mueller-Hinton broth (Difco Laboratories, Detroit, Mich.) was used for susceptibility testing of S. aureus, enterococci, and enteric bacteria; Mueller-Hinton broth supplemented with 2% lysed horse blood was used for susceptibility testing of S. pneumoniae. The MICs of each drug were determined by broth microdilution according to the standards of the National Committee for Clinical Laboratory Standards (17).
Robotics were used to create and inoculate microdilution checkerboard plates. For each linezolid-drug combination tested, a 96-well deep-well plate (Beckman-Coulter, Fullerton, Calif.) was filled with Mueller-Hinton broth containing 1% Alamar Blue (Trek Diagnostics, Westlake, Ohio) by the Multidrop instrument (Labsystems, Helsinki, Finland). Alamar Blue is a colorimetric redox indicator used as an aid for the visual reading of checkerboard plates (package insert; Trek Diagnostics) and had no effect on organism growth (3). Test drugs were diluted twofold from column 1 to column 7 of each deep-well mother plate by the Biomek 2000 instrument (Beckman-Coulter). Column 8 contained no test drug. Twofold dilutions of linezolid were then added to rows 1 to 7 of each deep-well plate containing a specific test drug. Row 8 contained no linezolid. A total of 25 µl was transferred by the Multimek 96 instrument (Beckman-Coulter) from each well of each mother plate to a daughter plate containing 165 µl of medium. Twelve daughter plates could be created from each mother plate. The Biomek instrument was used to inoculate the daughter plates with 10 µl of a 1:10-diluted culture equal to a 0.5 McFarland standard (approximately 1 x 108 CFU/ml) from bacterial growth on 18- to 24-h-old blood agar plates, for a final organism concentration of approximately 5 x 105 CFU/ml. The plates were incubated at 35°C for 18 ± 2 h.
The plates were read visually by observing a reduction (a color change from blue to purple or pink, indicating organism growth) or no reduction [a blue color, indicating no growth of organism due to inhibition by a drug(s)]. The MIC of each drug was determined. For wells along the growth-no growth interface, FICs were determined by the following formula: (MIC of linezolid in combination/MIC of linezolid alone) + (MIC of drug x in combination/MIC of drug x alone), where x is any of the drugs used in combination with linezolid. The average FIC index was calculated from individual FICs by the formula (FIC1 + FIC2 + . . . FICn)/n, where n is the total number of individual wells per plate for which FICs were calculated. Synergism was defined as an average FIC index
0.5, antagonism was defined as an average FIC index >4.0, and indifference was defined as an average FIC index from >0.5 to 4.0 (2, 10).
Assay reproducibility for synergism was monitored by using a combination of amoxicillin and clavulanic acid with a specific beta-lactamase-positive S. aureus strain (10). Assay reproducibility for antagonism was monitored by using a combination of chloramphenicol and sparfloxacin with a specific vancomycin-resistant E. faecalis strain (18).
|
|
|---|
|
View this table: [in a new window] |
TABLE 1. MICs of the antibacterial agents used in combination for gram-positive organisms
|
|
View this table: [in a new window] |
TABLE 2. Ranges of MICs of antimicrobial agents used in combination for gram-negative organisms
|
|
View this table: [in a new window] |
TABLE 3. Results of checkerboard testing of the activities of linezolid drug combinations against the organismsa
|
FIC index interpretations for the activities of linezolid in combination with 19 of 21 antimicrobials against vancomycin-resistant and -sensitive E. faecium strains predominantly showed indifference. Linezolid plus imipenem was synergistic against a vancomycin-resistant E. faecium strain (FIC index, 0.49), and linezolid plus tetracycline was synergistic against a vancomycin-resistant E. faecium strain (FIC index, 0.5).
FIC interpretations for the activities of linezolid in combination with 21 of 22 antimicrobials against penicillin-resistant, -intermediate, and -sensitive S. pneumoniae strains predominantly showed indifference. Linezolid plus erythromycin was synergistic against a penicillin-intermediate S. pneumoniae strain (FIC index, 0.48). No antagonism was determined.
The activities of linezolid and the other antimicrobial agents from the in vitro checkerboard interactions against E. coli and K. pneumoniae are also summarized in Table 3. Minimum and maximum FIC indices and interpretations for linezolid in combination with 21 of 22 antimicrobials predominantly showed indifference. Linezolid plus sparfloxacin was synergistic against a K. pneumoniae strain (FIC index, 0.46). No antagonism was determined.
|
|
|---|
To date, there have been few articles related to studies of the activities of linezolid in combination with other drugs. Di Pentima et al. (5) reported results on the in vitro activities of linezolid in combination with rifampin, vancomycin, and ciprofloxacin against four Flavobacterium meningosepticum isolates in which FIC indices indicated additivity or indifference for these combinations. In that study, FIC indices ranged from 0.74 to 1.5 for linezolid plus rifampin, 0.75 to 1 for linezolid plus vancomycin, and 0.62 to 1 for linezolid plus ciprofloxacin. No antagonism was reported. Hirschl et al. (12) also reported predominant indifference and occurrences of partial synergy and synergy for linezolid in combination with amoxicillin, clarithromycin, and metronidazole against Helicobacter pylori strains. FIC indices ranged from 0.31 to 2.5 for linezolid plus amoxicillin, 0.38 to 2.5 for linezolid plus clarithromycin, and 0.5 to 2.12 for linezolid plus metronidazole. No antagonism was reported. Sisson et al. (20) evaluated the in vivo pharmacokinetics of linezolid in combination with aztreonam and determined that the combination did not alter the disposition of either drug under single-dose conditions. Finally, Allen et al. (1) reported that linezolid in combination with cefepime, doxycycline, quinupristin-dalfopristin, and vancomycin improved or enhanced the killing of isolates of staphylococci and enterococci.
The results of this checkerboard study predominantly showed indifference when linezolid was combined with 35 different antimicrobial agents and tested against drug-sensitive and -resistant organisms. In the evaluation of the activities of 1,380 linezolid-drug combinations against the organisms tested, 1,369 combinations (99.2%) were indifferent. From nine determinations of synergism (0.65%), linezolid plus amoxicillin resulted in three cases of synergism against strains of methicillin-resistant S. aureus. Linezolid in combination with ofloxacin and sparfloxacin resulted in low levels of antagonism (0.15%) against two strains of E. faecalis.
Differences in the results of checkerboard assays for the detection of in vitro synergy between antimicrobial agents have been demonstrated. Mackay et al. (15) tested the same checkerboard on 3 separate days to test reproducibility and found no statistically significant differences. However, they did conclude that all combinations tested showing synergism according to the FIC index at 24 h also showed synergism by time-kill assays at 24 h but that the correlation between synergy at 2 or 5 h according to the FIC index and by the time-kill assay at the same time points was poor. Certain problems have been associated with the checkerboard microdilution method itself, such as variability in the inoculum as a potential source of error as well as the selection of inappropriate antibiotic concentrations (9).
Checkerboard assays are laborious and time-consuming. However, the robotic instruments used in the experiments described here allowed large numbers of strains and drug combinations to be tested. In our experiments, the results for each linezolid-drug combination that initially demonstrated synergism or antagonism were successfully obtained upon retesting, while combinations that showed indifference were not further evaluated. Additionally, inherent assay variability appeared to be kept to a minimum, as determined from the results for the control plates for synergism (100% synergism with amoxicillin-clavulanate plates with the appropriate organism) and control plates for antagonism (100% antagonism with chloramphenicol-ciprofloxacin plates with the appropriate organism). These control plates, which were used during each linezolid-drug combination experiment, were included to ensure the accuracy of the assay.
Also of concern was the interpretation of results from the literature from studies that used the checkerboard method, since numerous definitions have been assigned for synergism, antagonism, additivity, and indifference. Cappelletty and Rybak (4) reported synergism as a fourfold decrease in MICs, and synergism has been defined as an FIC index of
0.5, with "marked synergism" being an eightfold decrease in MICs and FIC indices of
0.25. In our experiments, synergism was defined as an FIC index of
0.5. Antagonism has been defined as an FIC index >1,
1, >2, or >4. From these different values, a result may be defined as indifferent or antagonistic, depending on the FIC index value chosen. In our experiments, antagonism was defined as an FIC index >4. On the basis of information in the literature and the recommended interpretations for FIC indices, it is our belief that the FIC interpretations that we selected adequately categorized the linezolid-drug interactions (2, 10). Even with the variabilities known to be inherently associated with synergy studies, such as the 1-dilution (twofold) variability associated with the performance of serial dilutions in the microdilution plate system, checkerboard methods nonetheless indicate that certain combinations of antibiotics are more useful than others against organisms (6, 8, 16, 21, 22).
The major value of this study is the demonstration that the combination of linezolid with a particular beta-lactam, quinolone, or other antibacterial agent primarily results in indifference and rarely results in antagonism. This is important, since there are limitations in the spectrum of activity of linezolid against certain organisms such as gram-negative and anaerobic bacteria. Animal and clinical studies are needed to establish whether therapy with linezolid in combination with another agent in selected clinical situations such as endocarditis, osteomyelitis, nosocomial and community-acquired pneumonia, and skin infections may be beneficial.
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»