TOXICOKINETICS AND METABOLISM

Michael Kastner Æ Ute Rahm

Irmela Baumann-Wilschke Æ Akintunde Bello

Ralf Stahlmann

Concentrations of the des-F(6)-quinolone garenoxacin in plasmaand joint cartilage of immature rats

Received: 3 June 2003 / Accepted: 19 August 2003 / Published online: 6 December 2003� Springer-Verlag 2003

Abstract Garenoxacin is a des-F(6)-quinolone with abroad antibacterial spectrum, which has previously beenshown to exhibit low chondrotoxicity in juvenile dogscompared with several other quinolones. A study wasperformed to determine whether the low chondrotoxic-ity observed in immature rats following garenoxacintreatment could be explained by poor penetration intocartilage tissue. Garenoxacin was orally administered toimmature (4- to 5-week-old) Wistar rats as a singledose—or as doses given on 5 consecutive days—of0 (vehicle), 200, 400, and 600 mg/kg (n=5 per doselevel). Additional groups of rats were orally dosed with600 mg/kg ofloxacin and ciprofloxacin. One knee jointof each animal (24 h after the last dose) was studiedhistologically after staining with Toluidine blue. Thepharmacokinetics of garenoxacin in plasma (200, 400,and 600 mg/kg) and in knee joint cartilage (200 and600 mg/kg) was assessed in separate groups of rats(n=55 per dose level). Concentrations of garenoxacin inplasma and cartilage were measured using an HPLCmethod. No signs of chondrotoxicity were observed inthe immature rats treated with garenoxacin or cipro-floxacin for 5 days at the doses investigated in this study.However, ofloxacin was found to induce cartilage lesionsthat were typical of those seen for this quinolone. Sys-temic exposure to garenoxacin increased as a function ofdose. Across dose and study day, mean garenoxacinplasma maximum concentration (Cmax) and area under

the concentration–time curve (AUCtau) values were inthe range 12–26 mg/l and 33–133 mg·h/l, respectively.Garenoxacin Cmax and AUC were similar on days 1 and5, within each dose, indicating the absence of accumu-lation or reduction in the systemic exposure. Valuesdetermined for Tmax (0.25–1.0 h) and T1/2 (3.8–6.4 h) ofgarenoxacin in plasma did not vary with dose or studyday. Although peak garenoxacin concentrations in car-tilage were between equal levels to and 2.5-fold of thosefound in plasma, the observed ratios were somewhatlower than those reported for other quinolones, e.g.ofloxacin or sparfloxacin. Since garenoxacin appeared tobe well absorbed following oral administration andconcentrations in cartilage tended to be higher thanthose in plasma, it is unlikely that the low chondrotox-icity in comparison with other quinolones is explainedby differences in the pharmacokinetics of these com-pounds.

Keywords Quinolones Æ Garenoxacin ÆPharmacokinetics Æ Cartilage Æ Joint histology ÆJuvenile rat

Introduction

Quinolones exhibit toxic effects on immature joint car-tilage (epiphyseal–articular complex) in all animal spe-cies studied. Doses needed to induce cartilage damage injuvenile dogs are in the range of therapeutically useddoses and therefore these drugs are contraindicated inpediatric patients (for review see Stahlmann and Lode2000). This decision has been considered controversialand there are arguments against the restricted pediatricuse of an entire class of valuable antimicrobial agents(Schaad 2000).

Garenoxacin is a des-F(6)-quinolone with broadantibacterial spectrum including anti-pneumococcalactivity (Fig. 1). For many years, the C-6 fluorine wasbelieved to be essential for the enhanced antibacterial

Arch Toxicol (2004) 78: 61–67DOI 10.1007/s00204-003-0514-3

M. Kastner Æ U. Rahm Æ I. Baumann-WilschkeR. Stahlmann (&)Department of Toxicology, Institute of Clinical Pharmacologyand Toxicology, Campus Benjamin Franklin ChariteUniversitatsmedizin Berlin, Garystrasse 5,14195 Berlin, GermanyE-mail: ralf.stahlmann@charite.deTel.: +49-30-84451770Fax: +49-30-84451763

A. BelloDept. Clinical Discovery, Pharmaceutical Research Institute,Bristol-Myers Squibb, Princeton,New Jersey, USA

activity of classical ‘‘fluoroquinolones’’ such as nor-floxacin or ciprofloxacin. From a series of compounds,the des-F(6)-quinolone garenoxacin was selected forclinical development because it exhibited a lower toxicpotential in cellular cytotoxicity assays and a higherdegree of selectivity with respect to inhibition of bacte-rial versus human topoisomerases than BMS-340280, its6-fluorinated counterpart (Lawrence et al. 2000).

Initial studies indicated that garenoxacin also exhibitsa comparatively low chondrotoxic potential when givenfor 7 days to juvenile dogs (Nagai et al. 2002), but inroutine toxicological studies in rats treatment withgarenoxacin for 1 or 3 months resulted in signs ofchondrotoxicity (cited in Stahlmann 2002). A study injuvenile rats was conducted in which a 5-day treatmentwas used, a regimen that corresponds to a potentialtherapeutic course. The pharmacokinetics of garenoxa-cin in the plasma and knee joint cartilage of the treatedanimals was studied in order to determine whether thelow chondrotoxicity could be explained by the poorpenetration of this drug into articular tissue.

Material and methods

Animals

Adult male and female Wistar rats (Hsd Cpb:WU) were purchasedfrom Charles River (Sulzfeld, Germany) and kept under specificpathogen-free (spf) conditions in Macrolon cages at a room tem-perature of 23±1�C, a relative humidity of 50±5% and a constantlight/dark schedule (light from 0700 to 1900 hours). The immaturerats (male and female) used for this study were bred at our institute.

Quinolone treatment

Garenoxacin doses of 200, 400, and 600 mg/kg were adminis-tered to groups of immature (4 to 5 weeks of age) Wistar rats(n=5 per dose level) either as a single dose or as once-dailydoses for 5 consecutive days by oral intubation. Additionalgroups of immature rats were similarly treated with 600-mgdoses of ofloxacin or ciprofloxacin. For garenoxacin treatment,the drug as supplied by Bristol Myers Squibb [mesylate salt; lotno. 807 T-3(A)], was suspended in a 2% starch solution(Mondamin). Doses were adjusted to the free base of garenox-acin. The freshly prepared suspension was administered by gas-tric intubation in a volume of 10 ml/kg body weight. Fortreatment with ofloxacin and ciprofloxacin, commercially avail-able fluoroquinolone-containing tablets Tarivid 200 (Hoechst,Frankfurt/Main, Germany) and Ciprobay 500 (Bayer Vital,

Leverkusen, Germany) were suspended in 2% starch solution(Mondamin) and administered by gastric intubation in a volumeof 10 ml/kg body weight.

Pharmacokinetic analysis

Garenoxacin doses of 200, 400, and 600 mg/kg were alsoadministered to satellite groups (n=55 per dose group) of juve-nile Wistar rats (4–5 weeks old) once or once-daily for 5 days byoral intubation. Animals were decapitated and blood samplescollected with haematocrit capillaries coated with sodium hepa-rinate at 0.25, 0.5, 0.75, 1.0, 1.5, 2, 3, 4, 6, 8 and 24 hours post-dose. The blood samples were centrifuged and the separatedplasma was stored at )20�C until analysis by high performanceliquid chromatography (HPLC). Cartilage samples of 14–29 mgwere scraped from each of the two femur heads of each of therats in the 200 and 600 mg/kg dose groups at the above-men-tioned times post-dose. Each of these was washed in Hanks�balanced salt solution, dried on fine tissue paper, and weighedseparately.

HPLC analysis

Plasma samples (100 ll) were deproteinized in duplicate with200 ll of a mixture of 0.5 ml perchloric acid in 200 ml aceto-nitrile, centrifuged, and the supernatant diluted with 400 ll ormore of (isocratic) HPLC buffer (84% 5 mM tetrabutylammo-nium dihydrogen phosphate (Sigma, Deisendorf, Germany), 16%acetonitrile (Merck, Darmstadt, Germany) adjusted to pH 3.6with citric acid). Individual cartilage samples, as described above,were each weighed, extracted twice with 800 ll 0.1 N phosphoricacid on a roller at 8�C for 24 h under light exclusion. After afinal vortexing and centrifugation, the supernatants of these wereused directly as HPLC samples. Chromatography was performedusing a Shimadzu RF-535 fluorescence HPLC monitor (excita-tion wavelength 280 nm, emission wavelength 406 nm), aNucleosil-100-5 C18 guard column (30·4 mm, particle size30–40 lm) and a Nucleosil-100-5 C18 column (125·4 mm, par-ticle size 5 lM). The flow rate was 1 ml/min, all sample volumeswere 20 ll, and all samples were injected manually. Absorbanceamplitudes were calculated via a Shimadzu C-R4AX Chromat-opac. The detection limit here was 0.02 mg/kg. Data werecollected and analysed by Microplate Manager PC Software(Bio-Rad, Munich, Germany). The linear range of the garenox-acin standard calibration curve used was 0.02–7.6 mg/kg forplasma and cartilage (Fig. 2). Precision within series was 3.5%for plasma and 9.8% for cartilage. For further details seeStahlmann et al. 1998.

Histology

For histological examination, knee joints were prepared within24 h after dosing and fixed in formalin (10%), decalcified in ethy-lenediaminetetraacetic acid solution (10%, pH 7.4), dehydrated inan alcohol series and embedded in paraffin. A series of 40–50 slices(6 lm) were prepared from the predilection sites of one knee jointof each animal and stained with an aqueous 1% solution ofToluidine blue (Merck). For further details see Stahlmann et al.1998.

Pharmacokinetics

Noncompartmental pharmacokinetic analysis of plasma concen-tration-time data was performed using WinNonlin (version 1.5,1984–1998, Pharsight Corporation, Mountain View, CA, USA).The following pharmacokinetic parameters were determined:

Fig. 1 Structural formula of garenoxacin, a des-F(6)-quinolone.Note the lack of a fluorine atom at position 6 (arrow)

62

1. Cmax (mg/l or mg/kg), the maximum observed drug concentra-tion in plasma or cartilage2. Tmax (h), the time at which Cmax was observed3. T1/2 (h), the apparent terminal elimination half-life4. AUCtau (mg·h/l), the area under the plasma concentration–time

curve from zero time to the time of the next dose.

Results

The mean body weight of the rats ranged from 65 to80 g at start of treatment, and 88 to 100 g at the end oftreatment. Treatment with garenoxacin, ofloxacin orciprofloxacin had no significant effect on body weightdevelopment of the rats. Five rats were used for eachdosage, varying from all male to all female, but no sex-dependence of plasma and/or tissue concentrationscould be inferred in these immature rats.

Knee joint histology

No joint cartilage lesions were detected histologically inknee joints from immature rats after treatment withgarenoxacin at single doses of 200, 400 or 600 mg/kg(n=15) or after multiple treatment with garenoxacin atthe same doses given for 5 days from postnatal day 32 today 36 (n=39). An example is shown in Fig. 3A. Simi-larly, we observed no joint cartilage lesions in the kneejoints from seven immature rats after treatment with600 mg ciprofloxacin/kg given for 5 days from postnatalday 32 to day 36 (data not shown).

Treatment with ofloxacin, however, caused typicallesions in articular cartilage in five of seven rats undercorresponding conditions, as shown in Fig. 3B, con-firming earlier results from our laboratory. Besides a lossof proteoglycans, as indicated by reduced stainability,joint cartilage is characterized by a loss of chondrocytes,demasking of collagen fibres and typical cleft formation.

Plasma pharmacokinetics

Following the administration of garenoxacin at doses of200, 400, and 600 mg/kg, garenoxacin was rapidly ab-sorbed from the gastrointestinal tract of the immaturerats. Concentrations reached their maximum about0.75–1 h after dosing on day 1, and 0.25–0.5 h on day 5.Plasma concentration–time profiles obtained upon asingle or multiple 200-mg/kg doses demonstrated asteady decline in garenoxacin concentrations from Cmax

levels following achievement of peak levels of 12 and12.5 mg/l on days 1 and 5, respectively. However,at the 400- and 600-mg/kg doses, secondary peaks in

Fig. 2 Calibration curve using garenoxacin at concentrations of20–7660 ng/ml in HPLC buffer. Excel plot after a view of theStandard Curve Report editing in the software used here(Microplate Manager, Bio-Rad). AUC values shown on theordinate (between 0 and 1.2·107) indicate mean values ±SD oftriplicate measurements

Fig. 3A,B Knee joint cartilage from juvenile rats (age 37 days)stained with Toluidine blue. A Treated from day 32 to day 36 withdaily oral doses of 600 mg garenoxacin/kg body weight. No lesionswere found in the immature articular–epiphyseal cartilage complex,and tissue cannot be distinguished from controls. B Treated fromday 32 to day 36 with daily oral doses of 600 mg ofloxacin/kg bodyweight. Note the typical lesions in the immature articular-epiphyseal cartilage complex (arrows)

63

garenoxacin plasma concentrations were observed bet-ween 2 and 4 h after dosing. The pH-dependent solubilityof garenoxacin makes it probable that this phenomenonwas a result of the incomplete garenoxacin dissolution inthe gastrointestinal tract at the higher doses.

The pharmacokinetic parameters determined forgarenoxacin following the administration of single (day1) and multiple doses (day 5) are presented in Table 1.Plots of the composite plasma-versus-time profiles forgarenoxacin on day 1 are presented in Fig. 4. For nom-inal doses increasing in a 1:2:3 ratio (200 to 600 mg/kg),Cmax values increased in a ratio of 1:1.5:1.5 on day 1 and

1:1.5:2.1 on day 5. The corresponding AUC values were1:1.8:3.3 on day 5 (AUCtau). Thus, although values forCmax did not increase in a ratio similar to the doseincrement, values for AUC appeared to increase in amanner similar to that of the dose. The ratios of AUCtau

values on day 5 to those on day 1 were 0.8:1.0, indicatingthat there was no systemic accumulation on repeatedadministration. Values for Tmax and T1/2 appeared to besimilar for the doses of 200, 400 and 600 mg/kg and wereindependent of study day (see Figs 4 and 5).

Table 1 Concentrations ofgarenoxacin in plasma and jointcartilage in 4-week-old ratsfollowing oral administration ofsingle or multiple doses for 1 orfor 5 days. Data are themaximum observed drugconcentration in plasma orcartilage (Cmax, mg/l forplasma, mg/kg for cartilage;means ±SD), the time at whichCmax was observed (Tmax) andratio depicting the quotient of[Cmax for cartilage] and [Cmax

for plasma]. ND Notdetermined

Dose Study day Tissue Cmax Tmax Ratio(mg/kg) (mg/l or mg/kg) (h) (Cmax:Cmax)

200 1 Plasma 12.5±2.2 1.0 2.5Cartilage 30.9±27.9 1.0

5 Plasma 12.0±9.0 0.25 1.5Cartilage 17.5±7.5 0.75

400 1 Plasma 18.5±8.1 0.75 –Cartilage ND ND

5 Plasma 18.1±0.0 0.5 –Cartilage ND ND

600 1 Plasma 18.0±5.0 1.0 2.1Cartilage 37.4±7.7 2.0

5 Plasma 25.5±11.3 0.5 1.0Cartilage 23.7±4.1 1.5

Fig. 4 Cartilage and plasma concentrations of garenoxacin injuvenile rats after single oral treatment with 200 or 600 mg/kg bodyweight (n=5 per time and dose level; mean values ±SD)

Fig. 5 Cartilage and plasma concentrations of garenoxacin injuvenile rats after multiple oral treatment at 200 or 600 mg/kg bodyweight once daily for 5 days (n=5 per time and dose level; meanvalues ±SD)

64

Cartilage pharmacokinetics

Garenoxacin concentrations in cartilage were onlystudied at the 200- and 600-mg/kg doses. Peak concen-trations—as for plasma—were found to occur 0.75–1.0 h after a single dose, and 0.25–1.0 h after all multipledoses, resulting in a somewhat smoother decline (Ta-ble 1). As observed in plasma, administration of thehigher dose (600 mg/kg) resulted in a secondary peak ingarenoxacin concentrations in cartilage. Interestingly,no secondary peaks in garenoxacin concentrations wereseen following multiple doses (Figs. 4,5).

The pharmacokinetic parameters for garenoxacin incartilage on days 1 and 5 are presented in Tables 1 and2. Plots of the composite cartilage- and plasma-versus-time profiles for garenoxacin on days 1 and 5 are pre-sented in Figs. 4 and 5. The cartilage-to-plasma ratio forgarenoxacin concentration was greater than 1, in almostall cases. The highest ratios were seen at the 600-mg/kgdose and ratios tended to be higher on day 1 than onday 5.

Discussion

Garenoxacin was selected for clinical development froma series of related quinolones because it exhibited lesscytotoxicity than other quinolones in an in vitroscreening test; of special interest is the fact that thisapplies also for the corresponding 6-fluoro-derivative(Lawrence et al. 2000). Furthermore, compared withsome other quinolones, the chondrotoxic potential ofthis drug seems to be rather low. Nagai and co-workersperformed a comparative study with garenoxacin, nor-floxacin and ciprofloxacin. The drugs were given orallyat a daily dose of 50 mg/kg for 1 week to three groups of3-month-old dogs. Blister formation was less pro-nounced with garenoxacin than with the two otheragents. The results confirmed previous findings under

similar conditions but using intravenous injection(Nagai et al. 2002).

In routine toxicological studies performed duringpreclinical development by the manufacturer, garenox-acin, when given orally to rats for 1 or 3 months at dailydoses of 400 mg/kg or greater, induced erosions, blistersand chondrocyte degeneration in joint cartilage; animalswere 6 weeks of age at start of the treatment (cited inStahlmann 2002). Under our study conditions at similardose levels but with a shorter treatment period, cartilagelesions were not induced (Kappel et al. 2002). Oneexplanation for the lack of chondrotoxicity of garenox-acin might be poor penetration of the drug into thecartilage compartment. Therefore, we performed a de-tailed pharmacokinetic analysis of the drug in juvenilerats and measured concentrations in plasma as well as incartilage samples.

Plasma concentrations of the comparator agentsofloxacin and ciprofloxacin in juvenile rats after oraldosing were not determined within the experimentsdescribed in this paper because these data have beenpreviously established in our laboratory under identicalconditions. Those experiments clearly showed thatplasma concentrations of ciprofloxacin were signifi-cantly lower than those of ofloxacin. Peak concentra-tions measured after a single, oral dose of 1200 mg/kgciprofloxacin or ofloxacin were 5.1±2.3 and45.4±12.4 mg/l, respectively (Stahlmann et al. 1997).Similar to ciprofloxacin, plasma concentrations ofsparfloxacin were rather low after oral intubation ofdoses as high as 1800 mg/kg (Stahlmann et al. 1998).

HPLC has been used routinely to study quinoloneconcentrations in plasma and tissue (Liang et al. 2002).Comparing the concentrations of garenoxacin found byHPLC in plasma and cartilage with those of ofloxacin(Lozo et al. 2002) reveals two things. Firstly, the peakconcentrations found in the plasma samples for bothdrugs after oral doses of 100 or 200 mg/kg, respectively,were found to be similar (garenoxacin 12 and ofloxacin 8to 14 mg/kg). Single or multiple dosages of garenoxa-cin—when compared with those of ofloxacin—werefound to yield similar absolute concentrations in carti-lage tissue, all of the values found here being within27.5±10.0 mg/kg (Lozo et al. 2002). Secondly—and wefind this more important here—the peak concentrationratios of cartilage-to-plasma are lower for garenoxacinthan for ofloxacin. Most of these ratios (mainly at thepost-dose time of 45 min) are greater than 1 for bothdrugs. These ratios of the peak concentrations vary forofloxacin between 2 and 3 (on days 1, 3 and 6) and forgarenoxacin between 1.0 and 2.5 (on days 1 and 5) asshown in Table 1. It should, of course, not be over-looked that these two drug studies involved differentdoses and different days of treatment, although weconsider them both to be comparable here.

The cartilage and plasma concentrations of gare-noxacin reveal that for cartilage the concentrations onday 5 tended to be lower than those on day 1. This is themain reason for the differences in the cartilage-to-plas-

Table 2 Concentrations of garenoxacin in plasma in 4-week-oldrats following oral administration of 200, 400, or 600 mg/kg dosesfor 1 or for 5 days. Data are the apparent terminal elimination half-life (T1/2) and the area under the plasma concentration–time curvefrom zero time to the time of the next dose (AUCtau). ND Notdetermined

Dose Study day Tissue T1/2 AUCtau

(mg/kg) (h) (mg·h/l)

200 1 Plasma 6.4 34Cartilage 2.1 67.7

5 Plasma 5.8 33Cartilage 2.7 52.3

400 1 Plasma 5.3 67Cartilage ND

5 Plasma 3.8 58Cartilage ND

600 1 Plasma 4.7 133Cartilage 6.4 189

5 Plasma 4.5 108Cartilage 3.6 117

65

ma ratios between the two days, since plasma concen-trations tended to be similar on days 1 and 5.

For ofloxacin, concentrations in cartilage and plasmawere measured after treatment on days 1, 3, and 6. Forall days and all times of measurement, these ratios variedbetween about 2 and 6 (Lozo et al. 2002). Concentra-tion-to-plasma ratios were found to be even higher(between 5 and 10) for sparfloxacin (Stahlmann et al.1998). Thus, due to a lower cartilage-to-plasma ratio,garenoxacin can be considered to be a quinolone with asomewhat decreased potential to induce lesions of thecartilage. The findings of this study are in agreementwith those previously reported in a poster presentationdescribing the kinetics of garenoxacin in another species(Nagai et al. 2002).

Data on the pharmacokinetic behaviour of the gare-noxacin in pediatric patients are not currently available;however, initial data obtained in healthy adult subjectsindicated that peak plasma concentration and AUCvalues of 5–6 mg/l and 84 mg·h/l, respectively wereachieved following an oral dose of 400 mg (Gajjar et al.2000).

The findings of the present study indicate that thesystemic exposure to garenoxacin in 4-week-old ratsfollowing repeated daily administration increased as afunction of dose and showed no accumulation. Peakconcentrations measured in 4-week-old rats were threeto five times higher at doses of 200–600 mg/kg thanthose seen in humans receiving 400 mg, correspondingto a dose of approximately 5–6 mg/kg. However, AUCvalues in rats only exceeded those seen in humans at the600-mg/kg dose. In this context it is of interest that peakplasma concentrations of more than 75 mg/l were ob-served in an additional study in which 4-day-old ratswere orally given a single garenoxacin dose of either 100or 300 mg/kg (data not shown here). The increasedsystemic exposure to garenoxacin noted in the newbornrats relative to juvenile animals (4 weeks old) was pos-sibly a result of the immature state of the clearancemechanisms (i.e. renal and or metabolic capacity) inthese very young animals. As in the juvenile rats, nohistological changes in knee joint cartilage were ob-served in the 4-day-old rats. However, at an ultrastruc-tural level effects on chondrocytes and tenocytes wereobserved in rats treated with garenoxacin or other qui-nolones (Fassheber et al. 2001).

Data on the use of quinolones in pediatrics aremainly available for three compounds: nalidixic acid,norfloxacin and ciprofloxacin (Hampel et al. 1997). Allof these quinolones demonstrate comparatively lowsystemic exposure or poor tissue penetration (nalidixicacid). Drugs such as pefloxacin, which show highersystemic exposure, are known to be associated withrelatively high incidences of arthropathy or tendopa-thy in humans (Pertuiset et al. 1989; Stahlmann andLode 2000). Although ‘‘chondrotoxicity’’ is consideredto be a ‘‘class effect’’, major differences seem to existwith respect to the risks associated with individualquinolones. In many cases the lack of chondrotoxicity

after oral treatment can be explained by poor bio-availability of the quinolones (Stahlmann and Lode1998, 2000).

In summary, garenoxacin was not found to bechondrotoxic when given to 4-week-old rats at dosesthat lead to peak plasma concentrations higher thanthose expected during therapy. As garenoxacin wasfound to be well absorbed following oral administrationand drug concentrations in cartilage tended to besomewhat higher than in plasma, the low chondrotox-icity seen with garenoxacin in comparison with that ofother quinolones cannot simply be explained by differ-ences in the pharmacokinetic behaviour of this drug.However, it is of interest that garenoxacin does notappear to penetrate into the immature joint cartilagetissue to the same extent as ofloxacin or some otherquinolones.

Acknowledgements The authors wish to thank Mrs. Ulla Schwi-kowski and Mrs. Helga Sturje for their photographic assistance.Further thanks go to Dr. Edith Lozo, Dr. Eva-Maria Kappel andDr. Susanne Fassheber for their contributions and to Mrs. BarbaraSteyn for her help in preparing the manuscript. This study wassupported by a grant from Bristol-Myers Squibb.

References

Fassheber S, Kappel EM, Riecke K, Rahm U, Shakibaei M,Stahlmann R (2001) Chondrotoxicity of ofloxacin comparedwith BMS-284756 in juvenile rats. In: Abstracts of 41st Inter-science Conference on Antimicrobial Agents and Chemother-apy (ICAAC), Chicago 16–19 December 2001, Abstract No.2207

Gajjar DA, Grasela DM, Bello A, Ge Z, Christopher L (2000)Safety, tolerability, and pharmacokinetics of BMS-284756, anovel des-F(6)-quinolone, following single oral doses in healthyadult subjects. In: Abstracts of 40th Interscience Conference onAntimicrobial Agents and Chemotherapy (ICAAC), Toronto17–20 September 2000, Poster No. 2259

Hampel B, Hullmann R, Schmidt H (1997) Ciprofloxacin in pedi-atrics: worldwide clinical experience based on compassionateuse—safety report. Pediatr Infect Dis J 16:127–129

Kappel EM, Shakibaei M, Bello A, Stahlmann R (2002) Effectsof the des-F(6)-quinolone garenoxacin (BMS-284756) incomparison to those of ciprofloxacin and ofloxacin, on jointcartilage in immature rats. Antimicrob Agents Chemother46:3320–3322

Lawrence LE, Wu P, Fran L, Gouveia K, Card A, Denbleyker K,Barrett JF (2000) The structure-activity relationship of BMS-284756, a novel des-F(6)-quinolone. In: Abstracts of 40thInterscience Conference on Antimicrobial Agents and Chemo-therapy (ICAAC), Toronto 2000, Poster No. 751

Liang H, Kays MB, Sowinski KM (2002) Separation of levoflox-acin, ciprofloxacin, gatifloxacin, moxifloxacin, trovafloxacinand cinoxacin by high-performance liquid chromatography:application to levofloxacin determination in human plasma.J Chromatogr B 772:53-63

Lozo E, Riecke K, Schwabe R, Vormann J, Stahlmann R (2002)Synergistic effect of ofloxacin and magnesium deficiency onjoint cartilage in immature rats. Antimicrob Agents Chemother46:1755–1759

Nagai A, Miyazaki M, Morita T, Forubo S, Kizawa K, FukumotoH, Sanzen T, Hayakawa H, Kawamura Y (2002) Comparativearticular toxicity of garenoxacin, a novel quinolone antimicro-bial agent, in juvenile beagle dogs. J Toxicol Sci 27:219–228

66

Pertuiset E, Lenoir G, Jehanne M, Douchain F, Guillot M, MenkesCJ (1989) Tolerance articulaire de la pefloxacine et de l�oflox-acine chez les enfants et adolescents atteints de mucoviscidose.Rev Rhum Mal Osteoartic 56:735–740

Schaad UB (2000) Use of the quinolones in pediatrics. In: AndrioleVT (ed) The quinolones, 3rd edn. Academic Press, San Diego,pp 455–475

Stahlmann R (2002) Clinical toxicological aspects of fluoroqui-nolones. Toxicol Lett 127:269–277

Stahlmann R, Lode H (1998) Safety of quinolones [Letter]. Lancet352:1313

Stahlmann R, Lode H (2000) Safety overview. Toxicity, adverseeffects, and drug interactions. In: Andriole VT (ed), The qui-nolones, 3rd edn. Academic Press, San Diego, pp 398–453

Stahlmannn R, Vormann J, Gunther T, Forster C, Zippel U, LozoE, Schwabe R, Kociok K, Shakibaei M, Merker H-J (1997)Effects of quinolones, magnesium deficiency or zinc deficiencyon joint cartilage in rats. Magn Bull 19:7–22

Stahlmann R, Zippel U, Forster C, Schwabe R, Shakibaei M,Merker H-J, Borner K (1998) Chondrotoxicity and toxicoki-netics of sparfloxacin in juvenile rats. Antimicrob AgentsChemother 42:1470–1475

67