Gemifloxacin Is Efficacious against Penicillin-Resistant and Quinolone-Resistant Pneumococci in Experimental Meningitis

In experimental rabbit meningitis, gemifloxacin penetrated inflamedmeninges well (22 to 33%) and produced excellent bactericidalactivity (change in log₁₀ [ ${Delta}$ log₁₀] CFU/ml/h, -0.68 ± 0.30[mean and standard deviation]), even superior to that of thestandard regimen of ceftriaxone plus vancomycin (-0.49 ±0.09 ${Delta}$ log₁₀ CFU/ml/h), in the treatment of meningitis due toa penicillin-resistant pneumococcal strain (MIC, 4 mg/liter).Even against a penicillin- and quinolone-resistant strain, gemifloxacinshowed good bactericidal activity (-0.48 ± 0.16 ${Delta}$ log₁₀CFU/ml/h). The excellent antibacterial activity of gemifloxacinwas also confirmed by time-kill assays over 8 h in vitro.

INTRODUCTION

Top

Introduction

Before the emergence of penicillin-resistant pneumococci, penicillinwas usually the first-line antibiotic in the treatment of pneumococcalinfections. The global increase of resistant pneumococci hasjeopardized the treatment of pneumococcal infections (3). Additionalresistance to cephalosporins has further limited therapeuticoptions for penicillin-resistant isolates. Despite the limitationsof the actual therapeutic modalities, ß-lactam antibioticsremain the first-line drugs for pneumococcal diseases, exceptwhen penetration into infected tissues is limited, as is thecase in meningitis. At present, a combination of vancomycinand a cephalosporin is recommended for meningitis due to resistantstrains (3, 10). An alternative regimen based on monotherapywould represent significant progress.

Especially due to its activity against many gram-positive microorganisms,including penicillin-resistant pneumococci, gemifloxacin, anew quinolone, is one of the most interesting candidates [N.Brenwald, M. J. Gill, F. Boswell, and R. Wise, J. Antimicrob.Chemother. 44(Suppl. A):145, 1999; D. Felmingham, M. J. Robbins,C. Dencer, H. Salman, I. Mathias, and G. L. Ridgway, J. Antimicrob.Chemother. 44(Suppl. A):131, 1999]. Little is known about thepenetration of gemifloxacin into inflamed meninges. The aimsof this study were to investigate the kinetics of gemifloxacinin the subarachnoid space and to test its bactericidal activityagainst pneumococci resistant to penicillin and to quinolonesin the rabbit meningitis model. The standard regimen consistedof ceftriaxone combined with vancomycin.

Strains.

Top

Strains.

The pneumococcal strain (WB4) was originally isolated from apatient with pneumonia at the University Hospital of Bern, Bern,Switzerland. The MICs (milligrams per liter) for this strainwere as follows: penicillin, 4; ceftriaxone, 0.5; vancomycin,0.12 to 0.25; trovafloxacin, 0.12; ciprofloxacin, 0.5; and gemifloxacin,0.015. A quinolone-resistant strain was obtained by sequentialexposure of parental strain WB4 to trovafloxacin. High-levelresistance was conferred by point mutations in ParC (Ser79 $->$ Phe)and in GyrA (Ser81 $->$ Phe) (5). The MICs (milligrams per liter)for this strain were as follows: penicillin, 4; ceftriaxone,0.5; vancomycin, 0.12 to 0.25; trovafloxacin, 4; ciprofloxacin,32; and gemifloxacin, 0.5.

Rabbit meningitis model.

Top

Rabbit meningitis model.

The meningitis model, originally described by Dacey and Sande(6), was slightly modified. The experimental protocol was acceptedby the Veterinäramt des Kantons Bern. Young New ZealandWhite rabbits weighing 2 to 2.5 kg were anesthetized by intramuscularinjections of ketamine (30 mg/kg of body weight) and xylazine(15 mg/kg) and were immobilized in stereotaxic frames for inductionof meningitis and cerebrospinal fluid (CSF) samplings. An inoculumcontaining approximately 10⁵ CFU of either a penicillin-resistantstrain (WB4) or a penicillin- and quinolone-resistant pneumococcusof serotype 6 was directly injected into the cisterna magna.A long-acting anesthetic (ethylcarbamate-urethane, 3.5 g/rabbit)was injected subcutaneously, and animals were returned to theircages. Fourteen hours later, a catheter was introduced intothe femoral artery for serum sampling, and the cisterna magnawas punctured again for periodic CSF samplings before and 1,2, 4, 6, and 8 h after initiation of therapy. Antibiotics wereadministered through a peripheral ear vein as bolus injectionsat the following concentrations: gemifloxacin, 15 mg/kg forthe penicillin-resistant strain and 30 mg/kg for the quinolone-resistantstrain; ceftriaxone, 125 mg/kg; and vancomycin, 20 mg/kg. Ceftriaxoneand gemifloxacin were injected once at 0 h and vancomycin wasinjected at 0 and 4 h as described by Friedland et al. (7),Gerber et al. (9), and Cottagnoud et al. (4). Untreated controlsreceived saline. All antibiotics and anesthetic drugs were commerciallypurchased, except for gemifloxacin, which was kindly providedby Glaxo SmithKline Company.

Bacterial titers were measured by using 10-fold serial dilutionsof CSF samples plated on blood agar plates containing 5% sheepblood and incubated overnight at 37°C. In parallel, 20-µlundiluted CSF samples were plated (limit of detection, 50 CFU/ml).A comparison between different dilutions of CSF was used toexclude significant carryover effects during therapy. The antimicrobialactivities of the regimens during the 8-h treatment were calculatedby linear regression analysis and expressed as a change in thelog₁₀ ( ${Delta}$ log₁₀) CFU per milliliter per hour and as the killingrate over 8 h. A value of 1.7 (log₁₀ limit of detection) wasassigned to the first sterile CSF sample, and a value of 0 wasassigned to any following sterile sample. The results are expressedas means and standard deviations. Statistical significance wasdetermined by the Newman-Keuls test.

Measurement of antibiotic levels in CSF.

Top

Measurement of antibiotic levels...

Antibiotic concentrations in CSF were determined by the agardiffusion method. Standard curves were determined for salinewith 5% rabbit serum in order to mimic the CSF protein concentration(12). Bacillus subtilis (ATCC 6633) was used as a test strainfor gemifloxacin (14). The intra- and interday variabilitiesof this method were less than 10%. The limit of detection forgemifloxacin was 0.05 mg/liter.

In vitro assays.

Top

In vitro assays.

The pneumococcal strains (penicillin-resistant strain WB4 andthe quinolone-resistant strain) were grown in C+Y medium (11)to an optical density at 590 nm of 0.3 and then diluted 40-foldto 10⁶ CFU/ml, corresponding to the CSF bacterial titer in rabbitsbefore the initiation of therapy. Antibiotics were added atconcentrations corresponding to 1, 5, and 10 times the MIC (theMICs were 0.015 and 0.5 mg/liter for the penicillin-resistantand the quinolone-resistant strains, respectively). Bacterialtiters were determined at 0, 2, 4, 6, and 8 h by serial dilutionof samples, plating on agar plates containing 5% sheep blood,and incubation at 37°C for 24 h. Experiments were performedin triplicate, and the results are expressed as means and standarddeviations.

Figure 1 shows the kinetics for gemifloxacin after a singledose of 15 mg/kg. The gemifloxacin concentration in serum peakedat a mean of 5.9 mg/liter and declined slowly to 0.45 mg/liter8 h later. With 320 mg of gemifloxacin per os in healthy volunteers,an area under the curve (AUC) in serum over 24 h of 10.3 mg· h/liter was reached, comparable to the serum AUC from0 to 8 h (AUC_0-8) of 12.2 mg · h/liter observed in rabbits,although the peak level in serum was higher in the animals (5.9versus 2.3 mg/liter) (8). The highest level in CSF was 0.7 mg/literand decreased progressively to 0.12 mg/liter by the end of thetreatment period. CSF penetration of gemifloxacin was calculatedby comparison of serum and CSF AUCs for each rabbit (SystatSoftware; SSPP Inc., Evanston, Ill.). In our rabbit model, thepenetration of gemifloxacin into the subarachnoid space wasabout 22% ± 6% after one injection of 15 mg of gemifloxacin/kg.The CSF drug concentration/MIC ratios ranged between 46 and8.

View larger version (20K):
[in this window]
[in a new window]
FIG. 1. Gemifloxacin concentrations in serum and CSF over 8 h after intravenous injection of 15 mg of gemifloxacin/kg. The concentration of gemifloxacin in the CSF remained above the MIC (0.015 mg/liter) during the entire treatment period. As determined by comparison of serum and CSF AUCs, the penetration of gemifloxacin into inflamed meninges was 22%.

A higher dose of gemifloxacin (30 mg/kg) was chosen in orderto maintain levels in CSF above the MIC (0.5 mg/liter for thequinolone-resistant strain) over the entire treatment period.The kinetics for 30 mg of gemifloxacin/kg are shown in Fig.2. In serum, gemifloxacin peaked at 12.2 mg/liter and declinedto 2.35 mg/liter after 8 h. The serum AUC_0-8 was 41 mg ·h/liter. The peak level in CSF was 2.5 mg/liter and the troughlevel was 0.45 mg/liter at 8 h. The CSF drug concentration/MICratios ranged between 5 and 0.9. After one injection of 30 mg/kg,the penetration into the CSF was 33% ± 5%. The betterCSF penetration at the higher dose (30 mg/kg) might be relatedto the drastic increase in the AUC_0-8 with a dose of 30 mg/kg(41 versus 12.2 mg · h/liter) in rabbits and might bedue to an oversaturation of protein binding (protein bindingof gemifloxacin in rabbits, 50 to 60%; Glaxo SmithKline Company,personal communication). Increasing the dose from 320 to 640mg per os in humans led to only a doubling of the AUC (2). Thedoses of ceftriaxone and vancomycin used were standard dosesthat have been used in previous studies with the same model(4, 7, 9) and that correspond to high doses in humans (1, 13).

View larger version (20K):
[in this window]
[in a new window]
FIG. 2. Gemifloxacin concentrations in serum and CSF over 8 h after intravenous injection of 30 mg of gemifloxacin/kg. The concentration of gemifloxacin in the CSF remained above the MIC (0.5 mg/liter) during almost the entire treatment period.

The killing rates for the different treatment groups are summarizedin Table 1. Gemifloxacin produced highly bactericidal activityand managed to sterilize the CSF of 7 out of 11 rabbits after8 h. All animals in the gemifloxacin group survived. It is interestingthat gemifloxacin monotherapy was significantly superior tothe standard regimen of ceftriaxone combined with vancomycinfor the penicillin-resistant strain. On the other hand, thestandard regimen was slightly but not significantly superiorto gemifloxacin monotherapy for the penicillin- and quinolone-resistantmutant.

View this table:
[in this window]
[in a new window]
TABLE 1. Single-drug and combination therapy against penicillin- and quinolone-resistant S. pneumoniae in experimental meningitis

In vitro, gemifloxacin had excellent bactericidal activity againstpenicillin-resistant pneumococci at concentrations above theMIC (5 and 10 times the MIC) in time-kill assays over 8 h. Concentrations5 and 10 times the MIC led to a dose-dependent decrease in theviable cell count over 8 h (3.8 and 5.25 log₁₀ CFU/ml, respectively).Even against the quinolone-resistant strain, gemifloxacin showedgood activity at concentrations above the MIC (3.9 and 4.75log₁₀ CFU/ml at 5 and 10 times the MIC, respectively).

The good penetration of gemifloxacin into the CSF (22 to 33%)and its efficacy in vitro and in our animal model qualify gemifloxacinas a potential therapeutic option for the treatment of meningitis,especially when resistant strains are involved.

ACKNOWLEDGMENTS

This study was supported by a grant from the Glaxo SmithKlineCompany and by the Wander Stiftung.

FOOTNOTES

* Corresponding author. Mailing address: Department of Internal Medicine, Inselspital, 3010 Bern, Switzerland. Phone: 41 31 632 2111. Fax: 41 31 632 3847. E-mail: pcottagn{at}insel.ch.

REFERENCES

Top