Abstract
The bactericidal activity of moxifloxacin alone and in combination with cefepime or piperacillin-tazobactam against clinical isolates of Klebsiella pneumoniae, Enterobacter cloacae, and Acinetobacter baumannii was evaluated by using time-kill methods and antimicrobial concentrations of one-half and one times the MIC. Synergy was observed in 58 to 88% of the strains and resulted in bactericidal activity against 60 to 100% of the strains. Combinations including moxifloxacin demonstrated enhanced bactericidal activity compared with that of either agent tested alone.
Gram-negative pathogens such as the Enterobacteriaceae, Pseudomonas aeruginosa, and Acinetobacter baumannii are commonly associated with intra-abdominal sepsis, pneumonia, and bloodstream infections among critically ill patients (1, 4). Moxifloxacin appears to be appropriate for monotherapy for uncomplicated infections due to its broad antimicrobial spectrum (2). However, for more severe nosocomial infections, moxifloxacin combined with a broad-spectrum β-lactam agent may offer the benefits of possibly synergistic activity and reduced emergence of resistance among gram-negative bacteria while providing adequate anaerobic activity in a simplified regimen. Moxifloxacin's activity against A. baumannii may also make it an attractive agent for use in combination regimens to treat multiresistant strains or to prevent the emergence of resistance (5). However, the activity of moxifloxacin in combination with other agents has not previously been evaluated. The objective of this study was to determine the activity of moxifloxacin alone and in combination with cefepime or piperacillin-tazobactam against clinical isolates of Klebsiella pneumoniae, Enterobacter cloacae, and A. baumannii.
Antibiotic MICs were determined by broth microdilution in cation-supplemented Mueller-Hinton broth (Difco, Sparks, Md.) according to NCCLS methods (7). Five clinical isolates of E. cloacae, seven clinical isolates of K. pneumoniae, and seven clinical isolates of A. baumannii were selected based on MIC results in order to provide a range of susceptibilities to the different antimicrobials tested. This approach was taken to simulate clinical scenarios in which empirical drug therapy would be used against pathogens with unknown antibiotic susceptibilities.
Bactericidal activity was determined in duplicate by using time-kill experiments according to the NCCLS M26-A method (8). Antimicrobial concentrations tested were one-half and one times the previously determined MIC for each bacterial strain. The final inoculum was confirmed at time zero, and viable colonies were counted after 4, 8, and 24 h of incubation at 35°C. All tests were performed in duplicate, but the experiment was repeated and quadruplicate data were used for analysis if discordant results were obtained. Bactericidal activity was defined as a ≥3-log10 decrease in CFU per milliliter, while bacteriostatic activity was defined as a <3-log10 decrease in CFU per milliliter at 8 and 24 h. The lower limit of detection for the time-kill assays was 1.3 log10 CFU/ml. Synergism and antagonism were defined as a ≥100-fold increase and a ≥100-fold decrease, respectively, in bacterial killing at 8 and 24 h with the combination of drugs compared with that achieved with the most active single agent alone. Synergistic and bactericidal activities were evaluated at both 8 and 24 h to account for any drug inactivation or bacterial regrowth occurring by 24 h. Additivity was defined as a <10-fold increase in killing at 8 and 24 h with the combination relative to that achieved with the most active single antimicrobial alone. Areas under the log10 CFU per milliliter-time curves from time zero to 24 h (AUC0-24) were calculated by the linear trapezoidal summation method in order to examine differences between antibiotic combinations in a more quantitative manner. Lower calculated AUC0-24 values represent more rapid and/or more complete bacterial killing than higher AUC0-24 values. A one-way analysis of variance was used with the Tukey test for post hoc analysis to compare AUC0-24 values between groups. Statistical P values of ≤0.05 were considered significant.
The MICs of study drugs for all 19 clinical isolates of E. cloacae, K. pneumoniae, and A. baumannii are listed in Table 1. No clinical strains of E. cloacae with decreased susceptibility to either cefepime or moxifloxacin could be identified for testing in this study. Likewise, no strains of K. pneumoniae with reduced susceptibility to cefepime could be identified. Other strains, particularly A. baumannii strains, were of various combinations of susceptibilities to moxifloxacin and the β-lactams.
TABLE 1.
MICs for E. cloacae, K. pneumoniae, and A. baumannii
Strain | MIC(s) (μg/ml)a
|
||
---|---|---|---|
Piperacillin-tazobactamb | Cefepimec | Moxifloxacind | |
E. cloacae | |||
U134 | 2, 4 (S) | 1 (S) | 0.0625 (S) |
B8720 | 4, 4 (S) | 0.25 (S) | 0.0625 (S) |
U155 | 32, 4 (I) | 0.25 (S) | 0.0625 (S) |
U110 | 64, 4 (I) | 0.5 (S) | 0.0625 (S) |
U093 | 128, 4 (R) | 8 (S) | 2 (S) |
K. pneumoniae | |||
B9007 | 2, 4 (S) | 0.0625 (S) | 1 (S) |
B9196 | 4, 4 (S) | 0.25 (S) | 1 (S) |
ICU1 | 4, 4 (S) | 0.125 (S) | 2 (S) |
ICU8 | 8, 4 (S) | 0.25 (S) | 1 (S) |
B9152 | 16, 4 (S) | 0.5 (S) | 16 (R) |
B9095 | 256, 4 (R) | 8 (S) | 4 (I) |
B9127 | 256, 4 (R) | 8 (S) | 4 (I) |
A. baumannii | |||
B8620 | 4, 4 (S) | 8 (S) | 4 (I) |
B6994 | 16, 4 (S) | 2 (S) | 0.0625 (S) |
B8560 | 32, 4 (I) | 16 (I) | 0.25 (S) |
N134 | 32, 4 (I) | 32 (R) | 8 (R) |
B8455 | 64, 4 (I) | 4 (S) | 2 (S) |
50-0532 | 256, 4 (R) | 16 (I) | 1 (S) |
ICU4 | 256, 4 (R) | 32 (R) | 4 (I) |
Susceptibility interpretive criteria are based on NCCLS recommendations. S, susceptible; I, intermediate; R, resistant.
For organisms that were susceptible, MICs were ≤16, 4 μg/ml; for organisms that were intermediate, MICs were 34, 4 to 64, 4 μg/ml; for organisms that were resistant, MICs were ≥128, 4 μg/ml. MICs of piperacillin are listed first.
For organisms that were susceptible, MICs were ≤8 μg/ml; for organisms that were intermediate, MICs were 16 μg/ml; for organisms that were resistant, MICs were ≥32 μg/ml.
For organisms that were susceptible, MICs were ≤2 μg/ml; for organisms that were intermediate, MICs were 4 μg/ml; for organisms that were resistant, MICs were ≥8 μg/ml.
Activities of piperacillin-tazobactam or cefepime in combination with moxifloxacin are summarized in Table 2. Depending on the concentrations tested, the combination of moxifloxacin plus cefepime yielded synergistic activity against 63% of E. cloacae strains, 88% of K. pneumoniae strains, and 63% of A. baumannii strains tested, while the combination of moxifloxacin plus piperacillin-tazobactam was synergistic against 63, 75, and 67% of K. pneumoniae, E. cloacae, and A. baumannii strains, respectively. However, synergistic combinations of moxifloxacin plus cefepime or piperacillin-tazobactam resulted in bactericidal activity against 80, 100, and 60% of K. pneumoniae, E. cloacae, and A. baumannii strains, respectively. The levels of achievement of synergistic and/or bactericidal synergistic activity were similar among K. pneumoniae, E. cloacae, or A. baumannii strains and were irrespective of whether strains had reduced susceptibility to one or both drugs in the combination. Antagonism was not observed with any antimicrobial combination.
TABLE 2.
Bactericidal activities of piperacillin/tazobactam plus moxifloxacin and cefepime plus moxifloxacin against E. cloacae, K. pneumoniae, and A. baumanii strains
Strain | Effect ofa:
|
|||||||
---|---|---|---|---|---|---|---|---|
Piperacillin/tazobactam + moxifloxacin
|
Cefepime + moxifloxacin
|
|||||||
0.5× MIC
|
1× MIC
|
0.5× MIC
|
1× MIC
|
|||||
8 h | 24 h | 8 h | 24 h | 8 h | 24 h | 8 h | 24 h | |
E. cloacae | ||||||||
U134 | A | A | A | S | A | S | A | S |
BS | BS | BS | BC | BS | BC | BC | BC | |
B8720 | S | S | A* | A* | A* | S | A* | A* |
BC | BC | BC | BC | BC | BC | BC | BC | |
U155 | A | A | A* | S | S | A | A* | S |
BS | BS | BC | BC | BS | BS | BC | BC | |
U110 | A | A | A* | A | S | S | A* | S |
BS | BS | BC | BS | BC | BS | BC | BS | |
U093 | A | S | A | A | A | S | A | A |
BS | BS | BS | BS | BS | BS | BS | BS | |
K. pneumoniae | ||||||||
B9007 | A | A | A | A* | A | S | A | A* |
BS | BS | BC | BC | BC | BC | BC | BC | |
B9196 | S | A | A* | S | A | S | A* | A |
BC | BS | BC | BC | BC | BC | BC | BC | |
ICU1 | S | S | A* | A* | S | S | A* | A* |
BC | BC | BC | BC | BC | BC | BC | BC | |
ICU8 | A | A | A* | S | S | S | A* | S |
BS | BS | BC | BC | BC | BC | BC | BC | |
B9152 | A | A | S | A* | S | A | S | S |
BS | BS | BC | BC | BS | BS | BC | BC | |
B9095 | A | S | A* | A* | S | S | A* | A* |
BC | BS | BC | BC | BC | BS | BC | BC | |
B9127 | S | S | A | A* | S | A | A* | A |
BC | BS | BS | BC | BS | BS | BC | BS | |
A. baumannii | ||||||||
B8620 | A | A | S | S | S | A | S | S |
BS | BS | BC | BC | BS | BS | BC | BC | |
B6994 | A | S | A | S | A | A | S | S |
BS | BS | BS | BC | BS | BS | BS | BC | |
B8560 | S | A | A | S | S | A | A | S |
BC | BS | BC | BC | BS | BS | BC | BC | |
N134 | S | S | A | S | S | A | S | S |
BC | BC | BC | BC | BS | BS | BC | BC | |
B8455 | A | S | A* | A* | S | A | A* | A* |
BS | BC | BC | BC | BS | BS | BC | BC | |
50-0532 | S | S | A | S | A | A | A | S |
BS | BC | BS | BC | BS | BS | BS | BC | |
ICU4 | S | A | A | S | A | A | S | S |
BS | BS | BC | BC | BS | BS | BC | BC |
S, synergistic; A, additive; BC, bactericidal; BS, bacteriostatic. *, due to the bactericidal activity of the single agent, synergistic activity could not be detected.
Because the MIC of moxifloxacin for K. pneumoniae strain B9152 was found to be higher than clinically relevant concentrations of the drug, time-kill studies were repeated with moxifloxacin concentrations of one-fourth of the MIC combined with β-lactam concentrations of one and two times the MIC to more accurately reflect clinically achievable drug concentrations. Bactericidal synergistic effects were observed with moxifloxacin concentrations of one-fourth of the MIC with piperacillin-tazobactam at one and two times the MIC as well as with cefepime at two times the MIC (data not shown).
For quantitative evaluation by areas under the time-kill curves, no statistically significant differences in AUC0-24 values were found for any strain when antibiotic combinations were tested at one times the MIC for the organism. When tested at one-half of the MIC, the combination of cefepime plus moxifloxacin resulted in significantly lower AUC0-24 values, indicative of more rapid and/or extensive bacterial killing, for 8 of 19 strains (1 E. cloacae strain, 4 K. pneumoniae strains, and 3 A. baumannii strains), while piperacillin-tazobactam plus moxifloxacin resulted in significantly lower AUC0-24 values for 3 strains (two E. cloacae strains and one K. pneumoniae strain). However, this difference was not significantly different (P = 0.15), and the activities of these two combination regimens were therefore considered to be equivalent.
The results of this study demonstrate that combinations of moxifloxacin and either cefepime or piperacillin-tazobactam achieve in vitro synergy in approximately 60 to 70% of the tested E. cloacae and A. baumannii clinical isolates and approximately 75 to 90% of the tested K. pneumoniae isolates. This synergistic activity resulted in bactericidal effects against 60 to 100% of the tested strains. It is of particular interest that this bactericidal activity was consistently observed even when MICs of one or both antibiotics in a combination were very high for bacterial strains. Appropriate selection of antibiotics for empirical treatment of gram-negative infections is often problematic because of the high rates of multidrug resistance seen in many institutions and the potential for choosing an antibiotic regimen to which pathogens may not be fully susceptible (1, 6). The potential to achieve bactericidal activity against these less-susceptible pathogens is a possible advantage favoring the use of potentially synergistic combinations of antibiotics rather than monotherapy.
In this experiment, moxifloxacin demonstrated synergistic and bactericidal activities against clinical isolates when tested at concentrations of one-half and one times the previously determined MIC. However, clinically achievable plasma concentrations at customary dosages would be important considerations in the use of agents intended to provide synergistic effects. Moxifloxacin achieves mean peak plasma concentrations of approximately 6.1 μg/ml with a regimen of 400 mg administered intravenously every 24 h (3). Moxifloxacin could thus be expected to achieve concentrations of up to one times the MICs for organisms with MICs of ≤4 μg/ml and one-half of the MICs for strains with MICs of up to 8 μg/ml but would not be expected to achieve adequate concentrations against strains with MICs of >8 μg/ml. Since concentrations of at least one-half of the MIC are clinically achievable against all strains tested in this study except one (K. pneumoniae B9152), moxifloxacin would have been suitable for use against the clinical isolates used in this experiment with some expectation of achieving synergistic activity as part of a combination regimen. Combination regimens including moxifloxacin would also have been suitable for use against K. pneumoniae B9152, since concentrations of one-fourth of the MIC were achievable against this strain. Based on the results of these in vitro experiments as well as pharmacokinetic considerations, moxifloxacin would be appropriate for patients with infections suspected or documented to be caused by K. pneumoniae, E. cloacae, or A. baumannii for whom therapy with a fluoroquinolone-containing regimen is desired.
Acknowledgments
This investigator-initiated study was supported by a grant from the Bayer Corporation.
Antibiotic powders of moxifloxacin, cefepime, and piperacillin-tazobactam were kindly provided by the Bayer Corporation, West Haven, Conn., Elan Pharmaceuticals, San Diego, Calif., and Wyeth Pharmaceuticals, Philadelphia, Pa., respectively.
REFERENCES
- 1.Anonymous. 1999. National Nosocomial Infections Surveillance (NNIS) System report, data summary from January 1990-May 1999, issued June 1999. Am. J. Infect. Control 27:520-532. [DOI] [PubMed] [Google Scholar]
- 2.Balfour, J. A., and L. R. Wiseman. 1999. Moxifloxacin. Drugs 57:363-374. [DOI] [PubMed] [Google Scholar]
- 3.Bayer Corporation. 2001. Avelox and Avelox I.V. (moxifloxacin) prescribing information. Bayer Corporation, West Haven, Conn.
- 4.Bergogne-Berezin, E., and J. Towner. 1996. Acinetobacter spp. as nosocomial pathogens: microbiological, clinical, and epidemiological features. Clin. Microbiol. Rev. 9:148-165. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Fass, R. J. 1997. In vitro activity of Bay 12-8039, a new 8-methoxyquinolone. Antimicrob. Agents Chemother. 41:1818-1824. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Kollef, M. H., G. Sherman, S. Ward, and V. J. Fraser. 1999. Inadequate antimicrobial treatment of infections: a risk factor for hospital mortality among critically ill patients. Chest 115:462-474. [DOI] [PubMed] [Google Scholar]
- 7.National Committee for Clinical Laboratory Standards. 2000. Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically, 5th ed. Approved standard M7-A5. National Committee for Clinical Laboratory Standards, Wayne, Pa.
- 8.National Committee for Clinical Laboratory Standards. 1999. Methods for determining bactericidal activity of antimicrobial agents. Approved guideline M26-A. National Committee for Clinical Laboratory Standards, Wayne, Pa.