Previous Article | Next Article ![]()
Antimicrobial Agents and Chemotherapy, June 2003, p. 1907-1911, Vol. 47, No. 6
0066-4804/03/$08.00+0 DOI: 10.1128/AAC.47.6.1907-1911.2003
Copyright © 2003, American Society for Microbiology. All Rights Reserved.
Experimental Infection Laboratory, Infectious Diseases Service,1 Microbiology Service, Ciutat Sanitària i Universitària de Bellvitge, c/Feixa Llarga s/n, 08907 L'Hospitalet de Llobregat, Barcelona, Spain2
Received 20 September 2002/ Returned for modification 25 November 2002/ Accepted 10 March 2003
|
|
|---|
|
|
|---|
LY333328 is a semisynthetic glycopeptide antibiotic derived from LY264826, which is active in vitro against gram-positive pathogens including methicillin- and amynoglycoside-resistant Staphylococcus aureus; coagulase-negative staphylococci; enterococci, including some enterococci resistant to vancomycin; and susceptible and penicillin- and cephalosporin-resistant Streptococcus pneumoniae (2, 10, 13, 20, 22). LY333328 has been demonstrated to have efficacy in animal models of S. pneumoniae septicemia, vancomycin-resistant enterococcal endocarditis, S. aureus endocarditis, S. aureus catheter-related infections, S. aureus soft tissue infections, and S. aureus foreign body-related skin infections (1, 17). Also, recent work by Gerber et al. (14) showed that the antibiotic has good activity against a penicillin-susceptible strain of S. pneumoniae in an experimental meningitis model. However, there are no data on its efficacy against drug-resistant strains causing meningitis.
The aim of the present study was to determine the efficacy of LY333328, alone and in combination with ceftriaxone, in the therapy of cephalosporin-resistant pneumococcal meningitis and the possible influence of dexamethasone in the efficacy of this treatment.
|
|
|---|
In vitro killing curves.
Killing curve studies were performed with glass tubes containing a final volume of 10 ml. The bacterial strain was grown in cation-adjusted Mueller-Hinton broth with 5% lysed horse blood. The final bacterial inoculum was 5 x 105 CFU/ml. Concentrations of one-half the MIC, the MIC, and two times the MIC of ceftriaxone or LY333328 were studied, as were concentrations of one-half the MIC and the MIC of each drug in combination. Bacterial titers were determined at 0, 6, and 24 h of incubation by serial dilution of samples, which were then plated on agar plates containing 5% sheep blood. No carryover effect was observed. The detection limit was 1 log10 CFU/ml. S. pneumoniae ATCC 49169, Enterococcus faecalis ATCC 29212, and Escherichia coli ATCC 25922 were used as control strains. Synergy was defined as a bactericidal effect of a drug combination with greater than 2-log killing over the level of killing of the most active drug used alone when one of the drugs in the combination was used at a subinhibitory concentration. A bactericidal effect was defined as a decrease in the initial inoculum of
3 log CFU/ml.
Rabbit model. Use of the animal model was approved by the Ethical Committee for Animal Experiments at the University of Barcelona (Campus de Bellvitge). Experiments with the animal model were performed by an established protocol (9). Five different therapeutic groups consisting of eight rabbits each were formed, as was a control group which was inoculated but not treated. Female New Zealand White rabbits (weight, 2 kg) were anesthetized intramuscularly with 35 mg of ketamine (Ketolar; Parke-Davis, El Prat de Llobregat, Spain) and 5 mg of xylazine (Rompun; Bayer AG, Leverkusen, Germany) per kg of body weight, and an acrylic dental helmet was affixed to each rabbit's calvaria. Twenty-four hours later, the animals were anesthetized again and placed in a stereotaxic frame. A spinal needle was introduced into the cisterna magna, 200 µl of cerebrospinal fluid (CSF) was withdrawn, and 200 µl of 106 CFU of a strain of S. pneumoniae belonging to serotype 23F per ml of saline was instilled into the subarachnoid space. The rabbits were placed back in their cages, and 18 h later the rabbits were again anesthetized with urethane (Sigma Chemical Company, St. Louis, Mo.) at 1.75 g/kg subcutaneously and phenobarbital (Penthotal Sodico; Abbott Laboratories, Madrid, Spain) 5 mg/kg intravenously (i.v.) and again placed in the stereotaxic frame, and a baseline CSF sample was taken. Then, an i.v. dose of 0.25 mg of dexamethasone (Fortecortin; Merck, Mollet del Vallés, Barcelona, Spain) or saline (Suero fisiológico; Braun S.A. Rubí, Barcelona, Spain) was administered, and 10 min later an i.v. dose of either 10 mg of LY333328 (Lilly S.A., Alcobendas, Madrid, Spain) per kg/day or 10 mg of LY333328 plus 100 mg of ceftriaxone (Rocefalin; Roche, Madrid, Spain) per kg/day was administered. The total dose of dexamethasone was 0.25 mg/24 h, which was administered every 12 h over a 26-h period (three doses); LY333328 and ceftriaxone were administered every 24 h (two doses). Therapeutic groups were as follows: LY333328, LY333328 plus dexamethasone, LY333328 plus ceftriaxone, LY333328 plus ceftriaxone and dexamethasone, and a control group. Even though ceftriaxone alone could not be considered a good therapeutic alternative against this strain (5), a group of rabbits was treated with ceftriaxone alone in order to study whether synergy exists between LY333328 and ceftriaxone. Serial CSF samples were taken at 2 (peak), 6, 24 (trough), and 26 (peak) h of treatment. CSF samples were used to determine white blood cell (WBC) counts and lactic acid concentrations, for direct and quantitative bacterial culture, and to determine bacteriostatic and bactericidal activities CSF and the CSF LY333328 concentrations at the trough and peak time points. WBC counts were determined by optical microscopy with a Neubauer chamber after the red blood cells had been lysed with Turk solution (0.2% acetic acid and methylene blue prepared in-house). CSF lactic acid concentrations were determined with a Lactate PAP kit (Biomérieux, S.A., Marcy l'Etoile, France) and by reading with a spectrophotometer. Serial 10-fold dilution cultures were made to determine the bacterial counts at each time point (the detection limit by this method was 102 CFU/ml). A value of 1.9 log CFU/ml was assigned to the first sterile culture, and a value of 0 log CFU/ml was assigned to the subsequent ones. Bactericidal activities in CSF were determined by a microdilution method (18) with cation-adjusted Mueller-Hinton broth (Difco Laboratories, Detroit, Mich.) with 2 to 5% lysed horse blood. Serial twofold dilutions (range, 1:2 to 1:4,096) of CSF samples were prepared, and a concentration of 5 x 105 CFU of the same strain used in the meningitis model per ml was inoculated into each well. After incubation at 35°C for 24 h, the dilution with bacteriostatic activity was the highest dilution without visible turbidity; then, 100 µl of each well without turbidity was subcultured at 35°C for 24 h, and the dilution with bactericidal activity was the highest dilution capable of killing 99.9% of the inoculated bacteria. To avoid interference from antimicrobial agent carryover, the sample was placed onto the plate as a single streak down the center, the sample was allowed to be absorbed into the agar until the plate surface appeared to be dry, and then the inoculum was spread over the plate (7). Aliquots of CSF were carefully frozen and sent to Eli Lilly (Indianapolis, Ind.) for CSF LY333328 level determination by high-performance liquid chromatography (HPLC). LY333328 was assayed by a validated method (15), by which solid-phase extraction followed by HPLC with fluorescence detection was used. The lower limit of quantitation was 0.075 µg/ml, and levels ranged between 0.078 and 2.213 µg/ml. Therapeutic failure was defined as an increase in bacterial concentration of at least 1 log CFU/ml compared with a previous count. A therapy was considered bactericidal when it achieved a reduction of 3 log CFU/ml.
Statistical analysis. A Fisher exact test was used to determine categorical variables. Comparisons of means was done by analysis of variance for independent data, and a level of significance of P < 0.05 was assumed. The Mann-Whitney U Wilcoxon rank sum test was used to compare median bactericidal titers.
|
|
|---|
0.01 µg/ml. Ceftriaxone was also bactericidal at 24 h at a concentration of two times the MIC, equivalent to
4 µg/ml. Combinations of LY333328 and ceftriaxone at concentrations of one-half the MIC of each drug (subinhibitory concentrations) showed bactericidal activity and synergistic effects at 6 and 24 h (Fig. 2 and Table 1).
![]() View larger version (35K): [in a new window] |
FIG. 1. In vitro killing curves for S. pneumoniae with several antibiotic regimens. CRO, ceftriaxone; LY, LY333328; DEX, dexamethasone.
|
|
View this table: [in a new window] |
TABLE 1. In vitro activities of ceftriaxone and LY333328 at 6 and 24 h
|
![]() View larger version (29K): [in a new window] |
FIG. 2. In vitro killing curves for S. pneumoniae with several antibiotic regimens. C, ceftriaxone; LY, LY333328.
|
![]() View larger version (23K): [in a new window] |
FIG. 3. Bacterial concentrations in CSF of rabbits infected with S. pneumoniae and treated with several antibiotic regimens. CRO, ceftriaxone; LY, LY333328; DEX, dexamethasone.
|
|
View this table: [in a new window] |
TABLE 2. Bacteriostatic and bactericidal activities in CSF (BD range) and LY333328 levels in CSF of rabbits with pneumococcal meningitisa
|
For the control group, the mean log CFU per milliliter was stable, with high counts detected in the control group.
Bacteriostatic and bactericidal activities in CSF. The median bacteriostatic and bactericidal activities in the CSF of the various therapeutic groups are shown in Table 2. The bacteriostatic and bactericidal activities were similar in all groups at 2 h (peak; activity range, 1:4 to 1:8) and 24 h (trough; activity, 1:2). The best activity was achieved with the combination of LY333328 and ceftriaxone, with or without dexamethasone, at 26 h, after administration of the second antibiotic dose (activities, 1:32 and 1:16, respectively).
CSF LY333328 concentration. It was not possible to determine LY333328 levels in CSF by the microbiological method because of the large molecular size of LY333328, which makes its diffusion in agar difficult. CSF LY333328 levels were low by HPLC, ranging from 0.09 to 0.33 µmol/ml; they were lower in the groups treated with dexamethasone (Table 2), but the difference was not statistically significant. The volumes of the CSF samples used for drug level determinations were small in most cases.
Inflammatory activity. WBC counts in CSF and lactate and protein concentrations are shown in Table 3. There were no statistically significant differences between groups. The decrease in inflammatory parameters of the greatest magnitude (without statistical significance) was achieved with the combination of LY333328, ceftriaxone, and dexamethasone. The WBC counts were 6,511/mm3 at 0 h and 1,586/mm3 at 26 h, the lactate concentrations were 6.8 mmol/liter at 0 h and 3.0 mmol/liter at 26 h, and the protein concentrations were 1.98 mg/ml at 0 h and 0.8 mg/ml at 26 h.
|
View this table: [in a new window] |
TABLE 3. Inflammatory parameters in CSF of rabbits with pneumococcal meningitis
|
|
|
|---|
Doses of LY333328 (oritavancin) ranging from 20 to 25 mg/kg have been used in several studies of experimental staphylococcal and enterococcal endocarditis (17, 21). In this study, we used a LY333328 dose of 10 mg/kg, which was found to be the most adequate in the experience of Gerber et al. (14), and we obtained efficacy results similar to those achieved by Gerber et al. Consequently, as would be anticipated, the antibiotic was also very effective in vivo against penicillin-resistant strains. The 10-mg/kg dose of LY333328 was chosen to produce concentrations in serum equivalent to those achieved by the doses intended for use in humans. Using these doses, Gerber et al. (14) found that the maximum concentration in the serum of rabbits was 42.72 ± 11.92 µg/ml. According to data from Eli Lilly obtained from studies with healthy volunteers, doses of up to 3 mg/kg produced concentrations in plasma of approximately 50 µg/ml and were well tolerated. In addition, doses of 3 mg/kg have also been used in the only reported study (M. M. Wasilewski, D. P. Disch, J. M. McGill, H. W. Harris, W. O'Riordan, and M. L. Zeckel, 41st Intersci. Conf. Antimicrob. Agents Chemother., abstr. UL-18, 2001) with humans of which we are aware. Gerber et al. (14) reported rates of penetration of LY333328 into the CSF between 1 and 5%. Although LY333328 has limited penetration through the blood-brain barrier and subsequent low levels in CSF, the extremely low MIC for the infecting strain (0.008 µg/ml) guarantees that levels in CSF will be several times above the MICs. The addition of ceftriaxone improved the results in terms of the reduction in the log CFU per milliliter and better bactericidal activity. It is possible that the rapid decrease in the bacterial concentration produced by LY333328 prevented the manifestation of a synergistic effect of the two drugs. Combinations of beta-lactams and glycopeptides have been suggested to be potentially efficacious against infections caused by S. aureus strains with different antibiotic susceptibilities (8). Moreover, a synergistic effect of ceftriaxone and vancomycin against resistant strains has been found in an experimental model of pneumococcal meningitis. In fact, these combinations are recommended as empirical therapy for pneumococcal meningitis (11, 12). According to our data, LY333328 plus ceftriaxone would be an even better combination in this setting. Given the present levels of cephalosporin resistance, synergism will be not relevant in most cases; however, hypothetically, for very highly cephalosporin-resistant strains, such synergy might be useful, as it would increase the activity of LY333328. CSF LY333328 levels were slightly higher in the two groups in which ceftriaxone was included as part of the treatment. One might speculate that the lytic activity of ceftriaxone might induce a higher level of inflammatory activity and perhaps the passage of larger amounts of LY333328 through the blood-brain barrier. Another explanation for the higher LY333328 levels in ceftriaxone-treated animals could be that ceftriaxone might impair the clearance of LY333328 from the CSF. Unlike the combination of vancomycin plus dexamethasone (4), the combination of LY333328 plus dexamethasone performed well in terms of reducing the bacterial concentration (there were no differences in the bacterial concentration after treatment with LY333328 plus dexamethasone compared with those obtained after treatment with LY333328 alone) and bactericidal activities. Although the combination was bactericidal at 2 h, the two therapeutic failures are intriguing. This fact might be related to the low levels achieved in CSF at the time of the trough concentration. In addition, bacteriostatic activities at the time of regrowth (24 h, which was the time of the trough concentration) were very low (<1:2), and the LY333328 concentrations in all CSF samples were below the level of detection. This trend suggests that the anti-inflammatory activity of dexamethasone may reduce the amount of LY333328 that passes through the blood-brain barrier, resulting in lower levels in CSF. For this reason, dexamethasone and LY333328 should not be used concomitantly. In conclusion, LY333328 might be a good alternative in the therapy of penicillin- and cephalosporin-resistant pneumococcal meningitis. The combination of LY333328 and dexamethasone led to some therapeutic failures, whereas the combination of LY333328 plus ceftriaxone and dexamethasone was effective in all cases. According to this experience, clinical studies may be scheduled.
A.F. was the recipient of a grant from Fundació August Pi i Sunyer. J.M.M. was the recipient of a grant from CIRIT.
|
|
|---|
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»