Bactericidal Activity and Postantibiotic Effect of Levofloxacin against Anaerobes

Abstract

The bactericidal activity and postantibiotic effect (PAE) of levofloxacin against nine anaerobes were determined. Levofloxacin at concentrations of the MIC and twice the MIC was bactericidal at 24 h to five of nine and nine of nine strains, respectively. The PAE of levofloxacin following a 2-h exposure ranged from 0.06 to 2.88 h.

Several new quinolones which have increased activity against anaerobes have been introduced (7, 8). Levofloxacin was the first of the new agents approved by the Food and Drug Administration. The in vitro activity of levofloxacin against anaerobes is reported to be similar to those of sparfloxacin and grepafloxacin but lower than those of trovafloxacin, clinafloxacin, and gatifloxacin (1, 8). While there is an abundance of literature on levofloxacin, the data available on anaerobes are limited. We performed time-kill and postantibiotic effect (PAE) studies of levofloxacin against nine anaerobic strains to examine the potential usefulness of this agent in anaerobic or mixed anaerobic infections.

(This work was presented in part at poster sessions of the 33rd Annual Meeting of the Infectious Diseases Society of America, San Francisco, Calif., 1995, and the 36th Interscience Conference on Antimicrobial Agents and Chemotherapy, New Orleans, La., 15 to 18 September 1996 [13a]).

Eight clinical strains and one American Type Culture Collection (ATCC) control organism were tested. The clinical isolates were obtained from David Hecht (Chicago, Ill.). The anaerobes were stored at −70°C in skim milk and underwent three subcultures prior to testing. Isolates with various susceptibility patterns were selected for time-kill and PAE assays. Prereduced Wilkins-Chalgren broth (WCB) (Oxoid-Unipath, Ogdensburg, N.Y.) was used for MIC, time-kill, and PAE testing. Centers for Disease Control and Prevention anaerobic blood agar (Remel, Lenexa, Kans.) was used for viable count determinations for the Bacteroides thetaiotaomicron and Peptostreptococcus magnus strains, and Wilkins-Chalgren agar was used for Clostridium perfringens.

Each anaerobe was suspended in 10 ml of thioglycolate broth and incubated overnight at 35°C on a shaker in an anaerobic chamber (Bactron; Sheldon Manufacturing, Cornelius, Oreg.) to obtain log-phase growth. The organisms were diluted with sterile saline until the turbidity matched a 0.5 McFarland standard. Each suspension was further diluted in WCB to obtain final inoculum sizes of approximately 5 × 10⁵ CFU/ml for MIC and time-kill studies and 5 × 10⁶ CFU/ml for the PAE determination. The exact inoculum size was determined via colony counts.

Levofloxacin powder was obtained from the R. W. Johnson Pharmaceutical Research Institute (Raritan, N.J.) and prepared according to the manufacturer’s recommendations. MICs were determined in duplicate by using the microbroth dilution method with WCB (12). The microtiter plates were incubated anaerobically at 35°C and read at 48 h. The MIC was defined as the lowest concentration at which there was no visible growth.

Time-kill assays were performed per National Committee for Clinical Laboratory Standards guidelines (13). Bacterial suspensions were added to test tubes containing prereduced WCB and levofloxacin concentrations of one-half the MIC, the MIC, and twice the MIC. One test tube was used as a control and contained no antimicrobial agent. The inoculum was confirmed at time 0; subsequent viable counts were determined at 2, 4, 6, 8, and 24 h. Sampling for colony counts was done by removing 0.1-ml samples of broth at the specified times. Each sample was serially diluted with sterile saline to produce 10-fold dilutions. Colonies were counted after anaerobic incubation at 35°C for 48 to 72 h. All procedures were performed in duplicate in an anaerobic environment. The rate and extent of killing were determined by plotting log₁₀ viable counts (CFU/milliliter) against time. Bactericidal activity was defined as a ≥3 log₁₀ decrease in CFU/ml, while bacteriostatic activity was defined as a <3 log₁₀ decrease in CFU/ml. The lower limit of detection was 2 log₁₀ CFU/ml.

The PAE of levofloxacin was determined by a method of repeated washing (5). Concentrations of 4.0, 6.0, and 8.0 μg/ml were tested. A tube containing no antimicrobial agent was included as a growth control. Except for experiments with C. perfringens, all tubes were incubated for 2 h on a shaking platform in a 35°C anaerobic incubator. PAE determinations for C. perfringens were performed following a 1-h incubation period. At the end of the exposure period, the antibiotic was removed by washing the tubes three times. The tubes were centrifuged, the supernatant was removed, and the bacterial pellet was resuspended with drug-free WCB. Viable counts were determined for all tubes at this time. The tubes were placed back on a shaking platform, with sampling performed every hour thereafter until the broth became cloudy. All procedures were performed in duplicate in an anaerobic environment. PAE was defined as T − C, where T is the time required for the count in the test culture to increase 1 log₁₀ above the count observed immediately after drug removal and C is the time required for the count in the untreated control to increase 1 log₁₀ above the count observed immediately after drug removal. Regression analyses of the MIC versus duration of PAE and of the ratio of the antibiotic concentration/MIC versus duration of PAE were performed.

The MIC, PAE, and extent of killing of levofloxacin for the nine anaerobes tested are shown in Table 1. Representative time-kill curves are presented in Fig. 1. Like other fluoroquinolones, levofloxacin demonstrates concentration-dependent killing. Differences in assay conditions have been shown to affect the bactericidal activity of the fluoroquinolones (10, 16, 17). Killing is not influenced by different media, inoculum sizes, or human sera but is decreased under acidic conditions, in human urine, and in the presence of magnesium and ferrous ions (16). The bactericidal activity of fluoroquinolones against Staphylococcus aureus was delayed, but not reduced when they were tested under anaerobic conditions (17).

TABLE 1.

MIC, time-kill, and PAE results for levofloxacin against nine anaerobes

Organism and strain	MIC (μg/ml)	Mean log10 change (CFU/ml) at 24 h with:		Mean PAE (h) ± standard deviation at levofloxacin concn of:
		MIC	2× MICa	4 μg/ml	6 μg/ml	8 μg/ml
B. fragilis
ATCC 25285	0.5	0.20a	−3.45	2.83 ± 0.03	2.86 ± 0.05	2.61 ± 0.38
2910	0.5	−2.27	−3.80	2.11 ± 0.62	2.27 ± 0.58	1.66 ± 0.94
6547	1.0	−3.51	−4.85	1.90 ± 0.64	2.59 ± 0.01	2.88 ± 0.38
3044	2.0	−4.66	−4.69	0.04 ± 0.44	0.07 ± 0.19	0.13 ± 0.43
3053	4.0	−4.19	−4.15	0.28 ± 0.60	0.78 ± 1.50	1.21 ± 0.17
B. thetaiotaomicron E15	4.0	−3.04	−3.34	0.32 ± 0.05	0.44 ± 0.29	0.61 ± 0.30
P. magnus
3081	1.0	−2.02	−4.14	0.89 ± 0.18	0.93 ± 0.21	0.88 ± 0.62
3035	4.0	−1.91	−4.62	0.52 ± 0.10	0.29 ± 0.04	0.40 ± 0.12
C. perfringens 2849	1.0	−3.20	−3.41	1.94 ± 0.18	2.21 ± 0.02	2.19 ± 0.13

^a2× MIC, twice the MIC. 

^b−0.91 at 6 h. 

FIG. 1.

Time-kill results for levofloxacin against four representative anaerobes. Error bars indicate standard deviation. ●, control; ◊, one-half the MIC; ▴, MIC; ○, twice the MIC. …, lower limit of detection.

The results of our time-kill experiments are in agreement with data available on aerobes. At a concentration of the MIC, levofloxacin was bactericidal for five of nine of the organisms at 24 h. Increasing the concentration to twice the MIC resulted in bactericidal activity against all organisms by 24 h. As a class, the fluoroquinolone antibiotics are characterized by their rapid bactericidal activity. The rate of killing was lower with the anaerobes compared to results obtained with aerobic organisms (2, 10, 16, 17). The mechanism involved in delayed killing under anaerobic conditions is unknown but is thought to be multifactorial (10, 17).

Currently, there is only one published study comparing the time-kill kinetics of levofloxacin and other antibiotics against anaerobes. Spangler et al. examined the bactericidal activity of levofloxacin against 11 anaerobes using a modified time-kill assay with oxyrase solution (14). They found that 90% of the anaerobes were killed at the MIC and 99% were killed at four times the MIC. They noted more-rapid killing at 6 h by levofloxacin, DU-6859a, and clindamycin than by ciprofloxacin, sparfloxacin, piperacillin, piperacillin-tazobactam, imipenem, and metronidazole but did not include specific data.

Literature on the PAE of antibiotics against anaerobes is extremely limited. Against aerobes, levofloxacin has a PAE ranging from 0.5 to 4.5 h (6, 9, 11, 15). Major differences in PAE procedures make comparison of data difficult. In our study, the PAE of levofloxacin following a 2-h exposure ranged from 0.04 to 2.88 h. Regression analyses of the MIC versus PAE and the ratio of the levofloxacin concentration/MIC versus PAE revealed no statistically significant correlation. There was a trend among the anaerobes for which the MICs were low (0.5 to 1 μg/ml) for a longer duration of PAE than that among the anaerobes for which the MICs were high (2 to 4 μg/ml). Levofloxacin was rapidly bactericidal against C. perfringens, such that the antibiotic exposure period had to be reduced to 1 h to prevent complete killing. The concentrations that we tested are lower than serum levels achievable with current dosing recommendations. Steady-state C_max values of levofloxacin for daily doses of 500, 750, and 1,000 mg are reported to be 5.72, 8.6, and 11.8 μg/ml, respectively (3, 4).

In summary, the time-kill kinetics and PAE of levofloxacin against the nine anaerobes demonstrated results similar to those of studies conducted with aerobic organisms. The bactericidal activity of levofloxacin against all organisms was observed at a concentration of twice the MIC. The rate of killing was delayed compared to time-kill curves obtained with aerobic organisms. Our in vitro data indicate that levofloxacin should be effective in the treatment of infections due to susceptible anaerobes. This agent may be useful in the treatment of mild-to-moderate mixed aerobic and anaerobic infections. Results from clinical trials will ultimately determine the role of levofloxacin in the treatment of these infections.