Vol. 37, No. 10ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, OCt. 1993, p. 2132-21380066-4804/93/102132-07$02.00/0Copyright © 1993, American Society for Microbiology

Influence of Rifampin on Fleroxacin PharmacokineticsJ. SCHRENZEL,l* P. DAYER,2 T. LEEMANN,2 E. WEIDEKAMM,3 R. PORTMANN,3 AND D. P. LEW'

Division of Infectious Diseases' and Division of Clinical Pharmacology,2 Geneva University Hospital,Geneva, and Pharma Research Department, F. Hoffmann-La Roche, Ltd., Basel,3 Switzerland

Received 8 February 1993/Returned for modification 2 May 1993/Accepted 30 July 1993

Staphylococcus aureus infections have been successfully treated in animal models with the combination offleroxacin and rifampin. We studied the influence of rifampin, a potent cytochrome P-450 inducer, on thepharmacokinetics and biotransformation of fleroxacin in 14 healthy young male volunteers. Subjects weregiven 400 mg of fleroxacin orally once a day for 3 days to reach steady state. After a wash-out period of 2 days,the same subjects received 600 mg of rifampin orally once daily for 7 days. On days 5 to 7 of rifampintreatment, 400 mg of fleroxacin was again administered once daily. Concentrations of fleroxacin as well as itstwo major urinary metabolites, N-demethyl- and N-oxide-fleroxacin, in plasma and urine were determined byreverse-phase high-performance liquid chromatography. The extent of hepatic enzyme induction by rifampinwas confirmed by a significant increase of 6-13-hydroxycortisol urinary output from 160.8 ± 41.4 to 544.8 ±120.7 ,ug/4 h. There were no significant changes in the peak fleroxacin concentration in plasma (6.3 ± 1.2versus 6.2 ± 1.9 mg/liter), time to maximum concentration of fleroxacin in plasma (1.1 ± 0.9 versus 1.3 ± 1.1h), or renal clearance (58.3 ± 16.4 versus 61.9 ± 19.2 mVmin). The area under the curve AUC (71.4 ± 15.8versus 62.2 ± 13.7 mg. h/liter) and the terminal half-life of fleroxacin (11.4 ± 2.2 versus 9.2 + 1.1 h) decreased(P < 0.05), while the total plasma clearance increased from 97.7 ± 21.6 to 112.3 ± 25.8 ml/min (P < 0.01).Despite being statistically significant, this 15% increase in total plasma clearance does not appear to beclinically relevant. Metabolic clearance by N demethylation was increased (6.9 ± 2.4 versus 12.5 ± 3.2 ml/min;P < 0.01), whereas clearance by N oxidation did not change (5.8 ± 1.1 versus 5.8 ± 1.5 ml/min). Fleroxacinelimination was slightly increased (about 15%) through induction of metabolic clearance to N-demethyl-fleroxacin. Since fleroxacin levels remained above the MIC for 90%o of the tested isolates of methicillin-susceptible S. aureus for at least 24 h, dose adjustment does not appear necessary, at least for short-termtreatments.

Fluoroquinolones, including fleroxacin, have recentlyemerged as interesting antibacterial agents in the therapy ofdeep-seated infections. This is in part attributable to theirantibacterial spectra, the extent of their oral absorption, andtheir high rate of penetration into biological fluids and tissues(45).Rifampin has long been used as an adjunctive treatment of

staphylococcal infections because of its antibacterial effectson staphylococci and its extensive distribution into tissues.However, rifampin monotherapy almost invariably leads tofailure due to development of resistance. The combined useof both families of antibacterial agents, fluoroquinolones andrifampin, has been advocated for the treatment of seriousstaphylococcal infections to avoid the emergence of resistantbacteria, especially in long-term treatments (3, 14-16, 22, 23,25, 26, 28, 35, 38, 44, 48).The combination of fleroxacin plus rifampin has been used

successfully in animal models, in particular to cure chronicstaphylococcal infections (8, 28). Fleroxacin is virtuallycompletely absorbed; its absolute bioavailability after oraladministration amounts to 100%. It is eliminated primarilyby renal clearance (CLR), about 60 to 70% of a dose beingrecovered unchanged in the urine within 96 h. The terminalhalf-life (t112,3) reaches approximately 13 h. Two majormetabolites in urine have been identified, N-oxide-fleroxacinand N-demethyl-fleroxacin. They amount to approximately 5to 10% and 7%, respectively, in healthy subjects (47).

Rifampin is metabolized by the hepatic microsomal en-zyme system (1, 43) and is known to induce many enzymatic

* Corresponding author.

catabolic pathways (20). By acting on the hepatic mixed-function oxidase system in humans (including the cyto-chrome P-450 III A CYP3A isoenzymes) (34), it enhances theelimination of a large number of drugs (2). Because the twometabolites of fleroxacin are formed by oxidative processesin the liver, an influence of rifampin on fleroxacin kineticsmust be considered. Therefore, it was important to assessthe influence of rifampin on steady-state fleroxacin pharma-cokinetics and to determine its effect under controlled con-ditions. In order to estimate the efficiency of enzymaticinduction by rifampin, the urinary output of 6-3-hydroxy-cortisol was measured. This physiologic nonconjugated me-tabolite of cortisol is synthesized mainly in the endoplasmicreticulum of the liver and poorly in the adrenal cortex andplacenta. It is produced by an isoenzyme of the cytochromeP-450 superfamily (probably CYP3A4) and is therefore anaccurate marker to control environmental induction (17, 46).The results of the present study will be useful to assess thedosages necessary for clinical studies applying this drugcombination in the therapy of staphylococcus infections.

(This work was presented in part at the 32nd InterscienceConference on Antimicrobial Agents and Chemotherapy,Anaheim, Calif., 11 to 14 October 1992 [39].)

MATERIALS AND METHODSVolunteers. Thirteen healthy male volunteers (age range,

22 to 36 years) participated in this investigation. Subjectswere admitted to the University Hospital of Geneva,Geneva, Switzerland, after ethics committee approval andinformed written consent were obtained. The age, weight,serum creatinine levels, and creatinine clearance (9) of these

2132

on June 23, 2018 by guesthttp://aac.asm

.org/D

ownloaded from

INFLUENCE OF RIFAMPIN ON FLEROXACIN PHARMACOKINETICS 2133

i1 7 1 8 1 9 110 11

Rifampin

Fleroxa(

121 131141

,in

SAMPLING

< PERIOD A > PERIOD B

FIG. 1. Protocol of the study. Rifampin was at 600 mg/day;fleroxacin, was at 400 mg/day; for the wash-out period, no drug wasadministered on days 4 and 5 (also on days 13 and 14). Samplingrefers to blood and urine samples. Period A, fleroxacin administeredalone; period B, fleroxacin administered after rifampin induction.

subjects were (mean ± standard deviation) 27.4 + 4.0 years,75.9 + 8.5 kg, 96.6 + 12.5 ,umol/liter, and 130 ± 31.2 ml/min,respectively.

Exclusion criteria were known or suspected hypersensi-tivity to nalidixic acid, quinolones, or rifampin; clinicallyrelevant deviation from normal in the physical examinationor in the routine laboratory tests; and any use of drugs within2 weeks before the study.Antacids as well as other drugs were strictly avoided.

Alcohol consumption and smoking (more than 10 cigarettesper day) were prohibited. Only one volunteer was a moder-ate smoker (1 to 3 cigarettes per day), and he refrained fromsmoking throughout the study. Medical histories, physicalexaminations, and results to a panel of laboratory tests (renaland liver chemistries, complete blood cell and plateletcounts, and urinalysis) were obtained before the study andon days 3, 10, and 12 of the study. Whenever volunteerscame for blood sampling, they were systematically ques-tioned about the occurrence of symptoms such as nausea,vomiting, abdominal pain, dizziness, sleep disorders, andpruritus. The symptoms were graded as discrete, moderate,or severe, and, if any were reported, their reversal wasactively checked.Study design. This was an open-label, nonrandomized,

within-subject comparative study. Two 3-day active treat-ment periods were programmed (fleroxacin alone and flerox-acin plus rifampin) to achieve steady-state conditions. Eachwas followed by a 2-day wash-out period. For hepaticenzyme induction, a 7-day run-in period with rifampin wasallowed to (32, 36, 43) overlap with the second 3-dayfleroxacin treatment period. The overall study design dia-gram (Fig. 1) demonstrates the 14-day time schedule of thistrial.

In the course of period A (fleroxacin alone), every volun-teer received, after an overnight fast, one oral dose of 400 mgof fleroxacin daily for 3 days (days 1 to 3). After a 2-daywash-out period (days 4 to 5), a daily morning 600-mg doseof rifampin was given for 7 days (days 6 to 12). Rifampin wastaken alone for 4 days; during the last 3 days (days 10 to 12),rifampin was taken 3 h after the dose of fleroxacin, 400 mgorally, with the volunteer still fasting to avoid any interfer-ence during the absorption. Overnight fasting and a mini-mum of 150 ml of water for drug intake were required.Volunteers were allowed to stop fasting 3 h after fleroxacinintake.

Blood and urine samples for assays were taken at steadystate (days 3 and 12). Sampling schedules were identical forboth periods. Blood was collected (5-ml Vacutainer tube [F.Hoffmann-La Roche, Basel, Switzerland] containing sodiumfluoride and potassium oxalate) before intake and at 0.5, 1,1.5, 2, 3, 4, 8, 12, 24, 48, and 72 h after fleroxacin intake.After centrifugation (1,000 x g for 15 min), plasma wastransferred into polyethylene-stoppered glass tubes coveredwith aluminum foil to prevent photodegradation and thenstored at -20°C until assay.

Urine samples were collected over the following 6 periodsafter the third fleroxacin dose: 0 to 4, 4 to 8, 8 to 12, 12 to 24,24 to 48, and 48 to 72 h. At the end of each time period, theportions were carefully mixed. pH and total volume wererecorded, and an aliquot (10 ml) was transferred to a tubecovered with aluminum foil and frozen at -20°C until assay.During the first urine collection period of days 3 and 12, afurther aliquot (10 ml) was -kept for a 6-,-hydroxycortisolassay as a control measure of hepatic enzyme induction.

Fleroxacin assay. Concentrations of fleroxacin and its twomain metabolites, N-oxide-fleroxacin and N-demethyl-fleroxacin, were determined by means of a reverse-phasehigh-performance liquid chromatography (HPLC) method(11). The mobile phase was 5 mM tetrabutylammoniumhydrogen sulfate-methanol (18:7 [vol/vol]), and separationwas performed on a Toyo Soda ODS-120 T 5 ,um column.Concentrations of fleroxacin in plasma and urine weredetected fluorimetrically (excitation at 290 nm, emission at450 nm). In plasma samples, the limit of quantification was20 p,g/liter (RSD, 2.5%), with a linearity of >0.99 over arange of 0.02 to 10 mg/liter and an interassay precision of 2.5to 4.8% over that concentration range. The sensitivity forurinary fleroxacin was 1 mg/liter (RSD, 0.98%), and theinterassay precision was 0.98 to 4.5% over a concentrationrange from 1 to 200 mg/liter (linearity, >0.99 over thatrange). The limit of quantification for fleroxacin urinarymetabolites was 0.5 mg/liter (RSD, 5%), with an interassayprecision of 3.2 to 4.2% over a concentration range from 1 to40 mg/liter (linearity, >0.99 over that range).

6- -Hydroxycortisol assay. 6-13-Hydroxycortisol urinaryoutput is highly individual (17) and has a circadian rhythm(50). Therefore, each volunteer was his own control, andsampling schedules were identical for both periods. Urinesamples were collected following initial micturition, from 7a.m. to 11 a.m. The volume was determined, and, aftermixing, an aliquot was frozen at -20°C until analyses couldbe performed. We used an immunoenzymatic reaction of acompetitive type with 6-,-hydroxycortisol peroxidase as aconjugate and orthophenylenediamine as a chromogenicsubstrate (51). Urine samples and calibrating solutions weremeasured by enzyme-linked immunosorbent assay (ELISA)(Stabiligen, Nancy, France). The limit of quantification was50 pg/ml, with a linearity superior to 0.99 over the range of 50to 1,000 pg/ml. The intra-assay coefficient of variationthroughout the range of quantifiable values was no greaterthan 10%.

Pharmacokinetic analysis. Visual inspection of the loglinearized data showed monophasic fleroxacin concentrationdecline in all subjects. Monophasic elimination with first-order absorption was therefore assumed, and a biexponen-tial model was fitted to the untransformed data with weight-ing proportional to the inverse of the predicted values. Themaximum concentration in plasma, the time to reach maxi-mum plasma concentration, and the minimum concentrationin plasma (24 h after the drug intake) were determined atsteady state; the area under the concentration-time curve

Days 1 | 2 3 4 5 6

Fleroxacin

SSAMPLING

VOL. 37, 1993

on June 23, 2018 by guesthttp://aac.asm

.org/D

ownloaded from

ANTIMICROB. AGENTS CHEMOTHER.

(AUC) from 0 to 24 h after fleroxacin ingestion (AUCbetween each dose) was determined by the linear trapezoidalrule. The tl2, was calculated by In2/,B, where 13 is the rateconstant of elimination calculated by linear regression of theterminal concentration-time curve. The AUC extrapolatedto infinity was estimated by the trapezoidal rule from timezero to the last time of measurement and extrapolated toinfinity by use of 1B. The total plasma clearance (CL) offleroxacin was calculated by dose/AUC from 0 to 24 h. Theclearance of formation of each metabolite was estimatedfrom the following equation, assuming that the metabolite iseliminated only by excretion via the kidney: CLf = CL xFe0_24, where Feo_24 is the fraction of fleroxacin excreted asa urine metabolite over a 24-h period. The CLR of fleroxacinwas estimated by the amount of unchanged fleroxacin inurine collected from 0 to 24 h, divided by the mean concen-tration of fleroxacin in plasma.

Statistical analysis. For the statistical comparison betweenperiods A and B, a Wilcoxon signed rank test was used (levelof significance, al = 0.05).

RESULTS

Tolerance. One volunteer was excluded because of amoderate allergic reaction (diffuse urticarial rash and nau-sea) 30 min after ingestion of the first dose of fleroxacin. Thisadverse effect disappeared spontaneously within 1 h. Other-wise, the tolerance of the drug combination was good; inparticular, no effects on vital signs were observed. Minormanifestations (mostly gastrointestinal complaints such asnausea [5], epigastric discomfort [2], vomiting [1], anddiarrhea [1]) in nine volunteers were reported. Two volun-teers complained of increasing photosensitivity, with anunusually rapid suntan, and two noted dizziness on the firstday of the study. This fairly high frequency of minor adverseevents may be explained, partly at least, by the systematicquestioning of the volunteers as required by the protocol.Complete blood cell count, urea, serum creatinine, alanineaminotransferase, aspartate aminotransferase, and bilirubinwere within normal ranges for all volunteers. Marginalelevation of alkaline phosphatase levels during the lastcontrol may have been related to the use of rifampin.

Hepatic enzyme induction. Urinary excretion of 6-,B-hydroxycortisol increased in all the volunteers by a mean ofapproximately 350%, as shown in Fig. 2, demonstratingsuccessful induction of the cytochrome P-450 hepatic micro-somal enzyme system by rifampin.

Pharmacokinetic parameters. The concentration-timecurves obtained either before or after rifampin treatment arepresented in Fig. 3 and 4. The pharmacokinetic parametersfor fleroxacin at the end of periods A and B are presented inTable 1. After rifampin pretreatment, fleroxacin t1/2,3 andAUC from 0 to 24 h decreased significantly (P < 0.05), by 19and 13%, respectively. This was associated with a 15%increase in CL. Pretreatment with rifampin also had asignificant influence on the minimum concentration inplasma, resulting in a 30% decrease (1.4 to 1.0 mg/liter).However, no significant influence on the maximum concen-tration in plasma or the time to maximum concentration inplasma was observed.CLR was not significantly altered, whereas metabolic

clearance increased by 28% (39.4 to 50.4 ml/min) in subjectswho were pretreated with rifampin. This higher metabolicclearance was also reflected by a significant increase inurinary recovery of N-demethyl-fleroxacin, from 6.9 to 12.5ml/min, following rifampin pretreatment. The urinary excre-

800-

600I--,

_6 m

o IV.~0

2 C

Io _

400-

200ol

Period A Period BFIG. 2. Effect of a 7-day rifampin treatment (600 mg/day) on

6-1-hydroxycortisol urinary excretion (during 4 h) as a marker ofhepatic enzyme induction. Period A, fleroxacin alone; period B,fleroxacin after rifampin induction.

tion of unchanged drug and N-oxide metabolite decreased by8 and 15%, respectively, but without reaching a statisticallysignificant difference when expressed as clearance.The cumulative urinary excretion (0 to 24 h) of the native

drug and its two main metabolites accounted for 70% andremained unchanged after rifampin pretreatment.

DISCUSSIONRifampin administered alone is one of the best available

antistaphylococcal agents. It has an extremely low MICagainst S. aureus and is extensively distributed into tissues.Moreover, rifampin is known to concentrate within neutro-phils and kill intracellular organisms (29). This mechanismmay be important in the treatment of purulent infections (5,29). However, this agent is not used as monotherapy be-cause resistance develops rapidly (30, 48, 49). Quinoloneshave been used in several clinical studies as single agents inthe therapy of staphylococcal infections (12, 13, 18, 19, 27).There is also considerable concern about the increasingdevelopment of S. aureus resistance to quinolones (6, 10, 40,42), both during therapy and on the epidemiological level.Therefore, quinolone-rifampin combinations have been ofinterest to clinicians. Theoretically, combined therapy couldimprove the rate of therapeutic success and limit the devel-opment of resistance by staphylococci to either agent.Many experimental studies have been conducted with

such combinations for the therapy of staphylococcal infec-tions. Successful results were observed with different animalmodels. A rat model was used to compare the efficacies ofvarious agents, alone and in combination, for the therapy ofmethicillin-resistant S. aureus chronic osteomyelitis (15, 23).The efficacy of a 3-week combination of a quinolone (eitherciprofloxacin [15, 23] or pefloxacin [15]) and rifampin wasshown to be equivalent (15) or statistically superior (23) tothat of a vancomycin-rifampin combination. In any case, theefficacies of these combination regimens were equivalent tothat of rifampin alone but superior to that of any other singleagent.Kaatz et al. reported that ciprofloxacin, alone or in com-

bination with rifampin, was not better than vancomycinalone in the treatment of a rabbit model of S. aureus

2134 SCHRENZEL ET AL.

on June 23, 2018 by guesthttp://aac.asm

.org/D

ownloaded from

INFLUENCE OF RIFAMPIN ON FLEROXACIN PHARMACOKINETICS

- U - Period A-. - Period B-qz

n = 13*p< 0.05

b~~~~~~~

40 60 80

HoursFIG. 3. Mean fleroxacin concentrations in plasma after the last (third) 400-mg fleroxacin dose: influence of rifampin. Period A, fleroxacin

alone; period B, fleroxacin after rifampin induction.

endocarditis (26). However, they suggested that the additionof rifampin to ciprofloxacin may decrease the frequency atwhich high-level resistance to ciprofloxacin emerges.

Rabbits with intraperitoneal foreign bodies were infected

with S. aureus (3). The efficacy of ciprofloxacin alone (8days) was not different from that in the control group,whereas the addition of rifampin showed significantly betterresults.

Finally, a rat model for subcutaneous foreign-body ab-

scesses was developed. A 6-day therapy of fleroxacin plusrifampin was shown to be equivalent to rifampin alone or incombination with vancomycin but clearly better in terms ofresistance to rifampin (28). A prolonged therapy (21 days)demonstrated a clear advantage after the combination offleroxacin and rifampin versus monotherapies, includingvancomycin (8).

There is controversy about the possible pharmacokineticinteractions between quinolones and rifampin, since ri-

c)

E

cox

20

EUn

FL

8-

7

6-

5-

4-

3-

2-

1

-U- Period A-0- Period B

n = 13*p< 0.05

-T

V II I Iv I

0 10 20Hours

FIG. 4. Time course of fleroxacin concentrations in plasma within a dosage interval (24 h). The levels were assessed under conditionssimilar to those for Fig. 3. Period A, fleroxacin alone; period B, fleroxacin after rifampin induction. * P < 0.05 pertains to the two rightmostvalues in the figure.

10

cmE

(_cox20*u

cn

CL

0.1

,1,0.001

0 20

2135VOL. 37, 1993

on June 23, 2018 by guesthttp://aac.asm

.org/D

ownloaded from

ANTIMICROB. AGENTS CHEMOTHER.

TABLE 1. Fleroxacin and its main metabolites: summary of pharmacokinetic parameters obtained for 13 volunteersa

Parameters (unit) Period A Period B % StatisticsMean SD Mean SD Difference (Wilcoxon)

Cmax (mg/liter) 6.3 1.2 6.2 1.9 -1 0.44C,mn (mg/liter) 1.4 0.5 1.0 0.4 -31 <0.01bTmax (h) 1.1 0.9 1.3 1.1 25 <0.05bt112P (h) 11.4 2.2 9.2 1.1 -19 0.02bAUC_2A (mg. h/liter) 71.4 15.8 62.2 13.7 -13 <O.O1bAUC0O, (mg. h/liter) 78.5 18.2 66.9 14.7 -15 <0.01bCL (mlmin) 97.7 21.6 112.3 25.8 15 <0.01bCLR (mlmin) 58.3 16.4 61.9 19.2 6 0.38CLNR (mVmin) 39.4 10.4 50.4 9.0 28 0.03bCLf N-demethyl (mlmin) 6.9 2.4 12.5 3.2 81 <O.O1bCLf N-oxide (mlmin) 5.8 1.1 5.8 1.5 0 0.81Fe Fleroxacin (%) 59.2 8.5 54.3 6.1 -8 0.17Fe N-Demethyl (%) 7.2 1.9 11.3 2.5 57 <0.02"Fe N-Oxide (%) 6.0 1.0 5.2 1.0 -15 <0.01bCumulative urinary excretion (%) 72.5 10.3 70.8 6.4 -3 0.81

a Period A, fleroxacin alone; period B, fleroxacin plus rifampin; difference, ratio between periods B and A - 1; Cm,, maximum concentration in plasma; Cmin,minimum concentration in plasma; Tma,x time to reach Cmax; t4/20, terminal half-life; AUC_24, area under the concentration-time curve from 0 to 24 h; AUCO,,AUC from 0 to infinity; CL, total plasma clearance; CLR, renal clearance; CLNR, nonrenal clearance; CLf, formation clearance; Fe, fraction of dose excretedas a urine metabolite from 0 to 24 h.

b P < 0.05 for period B versus period A.

fampin is known to induce enzymes responsible for in-creased hepatic drug catabolism. For ciprofloxacin, a rabbitmodel demonstrated an increased catabolism requiring aquinolone dose adaptation (4). However, no clinically signif-icant change in ciprofloxacin pharmacokinetics in elderlyvolunteers (48) or in elderly patients (7) was observed. Inone of these trials, Weinstein et al. (48) excluded a significantrifampin effect, but the experimental protocol included avery short rifampin induction time (2 days). Different resultswere obtained with a pefloxacin-rifampin combination. Ineight healthy volunteers a significant interaction was ob-served, but dosage modification was not required (24).

In a recent clinical study, Dworkin et al. (16) successfullytreated right-sided S. aureus endocarditis in parenteral drugusers with a combined regimen of ciprofloxacin and ri-fampin. One disadvantage of ciprofloxacin is its rapid elim-ination (tl2,3 = 4 h). Fleroxacin has a more prolongedhalf-life and its levels may be maintained continuously abovethe MIC for methicillin-susceptible staphylococcal strains (1jig/ml) (37) during therapy with a single 400-mg daily dose. Itwas therefore necessary to assess whether coadministrationof rifampin would alter the hepatic clearance of fleroxacinand thus its pharmacokinetic properties.Our study was not designed as a crossover because the

time required for reversion of the increased hepatic enzymeactivity is dependent on each individual and therefore nostandard duration for the restoration of baseline status canbe predicted.The pharmacokinetic parameters of fleroxacin in the initial

period were similar to those previously reported. For ri-fampin, the increased urinary output of 6-,-hydroxycortisolconfirms that each healthy young volunteer underwent asignificant hepatic enzyme induction. The results indicate asignificant pharmacokinetic influence of rifampin on flerox-acin. This interaction may be considered to be moderate,with an increase in CL of 15%. As CLR was not affected, themechanism of interaction seems to be mainly an induction ofmetabolic pathways leading to N-demethyl-fleroxacin.Whereas N oxidation is dependent on flavoproteins (31), Ndemethylation is predominantly mediated by cytochromeP-450 (41). Without any hepatic enzyme induction, cumula-

tive urinary excretion of the major fleroxacin metabolitesremains in a narrow range. Estimates of N-oxide-fleroxacinvary in the literature from 5.7% + 0.8% of the total dose for72 h (n = 6) (21) to 6.3% + 1.7% (n = 12) (41); theN-demethyl derivative represents from 6.6% _ 1.3% (n =30) (33) to 6.9% + 2.0% (n = 6) (21). Our results beforerifampin therapy were similar to those reported previously.Hepatic enzyme induction by rifampin led to a significantincrease in N-demethyl metabolite production, whereasN-oxide-fleroxacin clearance remained unchanged. Thereare presently no available data concerning the influence ofother hepatic metabolism inducers (either smoking or otherdrugs) on fleroxacin pharmacokinetics. It should be empha-sized that despite a statistically significant effect of rifampinon the pharmacokinetics of fleroxacin, the extent of thatinteraction has little clinical importance. It should be notedthat the mean minimum concentration in plasma (24 h after a400-mg fleroxacin dose and upon 600-mg daily rifampintreatment) is around 1.0 mg/liter, which corresponds to theminimal concentration inhibiting 90% of strains of methicil-lin-susceptible S. aureus and Staphylococcus epidermidis.The present results now allow the concomitant adminis-

tration of fleroxacin and rifampin in clinical trials, withoutany a priori dose modification of fleroxacin.

ACKNOWLEDGMENTS

We sincerely thank F. Schrenzel for secretarial support, H.Porchet for providing pharmacokinetic software, and C. Chauffat-Rabere and P. Meier for excellent technical assistance.

This study was made possible by a grant from Hoffmann-LaRoche and a grant from the Swiss National Research Foundation(no. 32-30161.90).

REFERENCES1. Acocella, G. 1983. Pharmacokinetics and metabolism of rifam-

picin in humans. Rev. Infect. Dis. 5(Suppl. 3):S428-S432.2. Baciewicz, A. M., and T. H. Self. 1984. Rifampicin drug inter-

actions. Arch. Intern. Med. 144:1667-1671.3. Bamberger, D. M., M. T. Fields, and B. L. Herndon. 1991.

Efficacies of various antimicrobial agents in treatment of Staph-ylococcus aureus abscesses and correlation with in vitro testsof antimicrobial activity and neutrophil killing. Antimicrob.

2136 SCHRENZEL ET AL.

on June 23, 2018 by guesthttp://aac.asm

.org/D

ownloaded from

INFLUENCE OF RIFAMPIN ON FLEROXACIN PHARMACOKINETICS 2137

Agents Chemother. 35:2335-2339.4. Barnere, S. L., G. W. Kaatz, and S. M. Seo. 1989. Enhanced

elimination of ciprofloxacin after multiple-dose administrationof rifampin to rabbits. Antimicrob. Agents Chemother. 33:589-590.

5. Beam, T. R., Jr. 1979. Sequestration of staphylococci at aninaccessible focus. Lancet ii:227-228.

6. Blumberg, H. M., D. Rimland, D. J. Carroll, P. Terry, and I. K.Wachsmuth. 1991. Rapid development of ciprofloxacin resis-tance in methicillin-susceptible and -resistant Staphylococcusaureus. J. Infect. Dis. 163:1279-1285.

7. Chandler, M. H. H., S. M. Toler, R. P. Rapp, R. R. Muder, andJ. A. Korvick. 1990. Multiple-dose pharmacokinetics of concur-rent oral ciprofloxacin and rifampin therapy in elderly patients.Antimicrob. Agents Chemother. 34:442-447.

8. Chuard, C., M. Herrmann, P. Vaudaux, F. A. Waldvogel, andD. P. Lew. 1991. Successful therapy of experimental chronicforeign-body infection due to methicillin-resistant Staphylococ-cus aureus by antimicrobial combinations. Antimicrob. AgentsChemother. 35:2611-2616.

9. Cockcroft, D. W., and M. H. Gault. 1976. Prediction of creati-nine clearance from serum creatinine. Nephron 16:31-41.

10. Daum, T. E., D. R. Schaberg, M. S. Terpenning, W. S. Sottile,and C. A. Kauffman. 1990. Increasing resistance of Staphylo-coccus aureus to ciprofloxacin. Antimicrob. Agents Chemother.34:1862-1863.

11. Dell, D., C. Partos, and R. Portmann. 1988. The determinationof a new trifluorinated quinolone fleroxacin, its N-demethyl andN-oxide metabolites in plasma and urine by high performanceliquid chromatography detection. J. Liq. Chromatogr. 11:1299-1312.

12. Dellamonica, P., E. Bernard, H. Etesse, R. Garaffo, and H. B.Drugeon. 1989. Evaluation of pefloxacin, ofloxacin and cipro-floxacin in the treatment of thirty-nine cases of chronic osteo-myelitis. Eur. J. Clin. Microbiol. Infect. Dis. 8:1024-1030.

13. Desplaces, N., and J. F. Acar. 1988. New quinolones in thetreatment of joint and bone infections. Rev. Infect. Dis.10(Suppl. 1):S179-S183.

14. Desplaces, N., L. Gutman, J. Carlet, J. Guibert, and J. F. Acar.1986. The new quinolones and their combinations with otheragents for therapy of severe infections. J. Antimicrob.Chemother. 17(Suppl.):25-39.

15. Dworkin, R., G. Modin, S. Kunz, R. Rich, 0. Zak, and M.Sande. 1990. Comparative efficacies of ciprofloxacin, pefloxa-cin, and vancomycin in combination with rifampin in a rat modelof methicillin-resistant Staphylococcus aureus chronic osteo-myelitis. Antimicrob. Agents Chemother. 34:1014-1016.

16. Dworkin, R. J., B. L. Lee, M. A. Sande, and H. F. Chambers.1989. Treatment of right-sided staphylococcus aureus endocar-ditis in intravenous drug users with ciprofloxacin and rifampi-cin. Lancet 2:1071-1073.

17. Ged, C., J. M. Rouillon, L. Pichard, J. Combalbert, N. Bressot,P. Bories, H. Michel, P. Beaune, and P. Maurel. 1989. Theincrease in urinary excretion of 63-hydroxycortisol as a markerof human hepatic cytochrome P-450 III A induction. Br. J.Pharmacol. 28:373-387.

18. Gentry, L. O., and G. G. Rodriguez. 1990. Oral ciprofloxacincompared with parenteral antibiotics in the treatment of osteo-myelitis. Antimicrob. Agents Chemother. 34:40-43.

19. Gentry, L. O., and G. Rodriguez-Gomez. 1991. Ofloxacin versusparenteral therapy for chronic osteomyelitis. Antimicrob.Agents Chemother. 35:538-541.

20. Goodman Gilman, A., T. W. Rall, A. S. Nies, and P. Taylor.1990. The pharmacological basis of therapeutics, 8th ed. Perga-mon Press, Elmsford, N.Y.

21. Griggs, D. J., R. Wise, B. Kirkpatrick, and J. P. Ashby. 1988.The metabolism and pharmacokinetics of fleroxacin in healthysubjects. J. Antimicrob. Chemother. 22(Suppl. D):191-194.

22. Hackbarth, C. J., H. F. Chambers, and M. A. Sande. 1986.Serum bactericidal activity of rifampin in combination withother antimicrobial agents against Staphylococcus aureus. An-timicrob. Agents Chemother. 29:611-613.

23. Henry, N. K., M. S. Rouse, A. L. Whitesell, M. E. McConnell,

and W. R. Wilson. 1987. Treatment of methicillin-resistantStaphylococcus aureus experimental osteomyelitis with cipro-floxacin or vancomycin alone or in combination with rifampicin.Am. J. Med. 82(Suppl. 4A):73-75.

24. Humbert, G., I. Brumpt, G. Montay, A. Le Liboux, A. Frydman,F. Borsa-Lebas, and N. Moore. 1991. Influence of rifampicin onthe pharmacokinetics of pefloxacin. Clin. Pharmacol. Ther.50(6):682-687.

25. Kaatz, G. W., S. L. Barriere, D. R. Schaberg, and R. Fekety.1987. The emergence of resistance to ciprofloxacin duringtreatment of experimental Staphylococcus aureus endocarditis.J. Antimicrob. Chemother. 20:753-758.

26. Kaatz, G. W., S. M. Seo, S. L. Barriere, L. M. Albrecht, andM. L. Rybak. 1989. Ciprofloxacin and rifampin, alone and incombination, for therapy of experimental Staphylococcus au-reus endocarditis. Antimicrob. Agents Chemother. 33:1184-1187.

27. Ketterl, R., T. Beckurts, B. Stubinger, and B. Claudi. 1988. Useof ofloxacin in open fractures and in the treatment of post-traumatic osteomyelitis. J. Antimicrob. Chemother. 22(Suppl.C):159-166.

28. Lucet, J.-C., M. Herrmann, P. Rohner, R. Auckenthaler, F. A.Waldvogel, and D. P. Lew. 1990. Treatment of experimentalforeign body infection caused by methicillin-resistant Staphylo-coccus aureus. Antimicrob. Agents Chemother. 34:2312-2317.

29. Mandell, G. L. 1983. The antimicrobial activity of rifampicin:emphasis on the relation to phagocytes. Rev. Infect. Dis.5(Suppl. 3):S463-S467.

30. Mandell, G. L., and D. R. Moorman. 1980. Treatment ofexperimental staphylococcal infections: effect of rifampin aloneand in combination on development of rifampin resistance.Antimicrob. Agents Chemother. 17:658-662.

31. McManus, M. E., I. Stupans, W. Burgess, J. A. Koenig, P. de laHall, and D. J. Birkett. 1987. Flavin containing monoxygenaseactivity in human liver microsomes. Drug Metab. Dispos.15:256-261.

32. Michot, F., M. Buergi, and J. Buettner. 1970. Rimactan (Ri-fampizin) und Antikoagulantientherapie. Schweiz. Med.Wochenschr. 100:583-584.

33. Nakashima, M., M. Kanamaru, T. Uematsu, A. Takiguchi, A.Mizuno, T. Itaya, F. Kawahara, T. Oole, S. Salto, H. Uchida,and K. Masuzawa. 1988. Clinical pharmacokinetics and toler-ance of fleroxacin in healthy male volunteers. J. Antimicrob.Chemother. 22(Suppl. D):133-144.

34. Nebert, D. W., D. R. Nelson, M. J. Coon, R. W. Estabrook, R.Feyerelsen, Y. Fujil-Kuriyama, F. J. Gonzalez, F. P. Guengerich,I. C. Gunsalus, E. F. Johnson, J. C. Loper, R. Sato, M. R.Waterman, and D. J. Waxman. 1991. The P450 superfamily:update on new sequences, gene mapping, and recommendednomenclature. DNA Cell Biol. 10:1-14.

35. Neu, H. C. 1989. Synergy of fluoroquinolones with other anti-microbial agents. Rev. Infect. Dis. ll(Suppl. 5):S1025-S1035.

36. O'Reilly, R. A. 1975. Interaction of chronic daily warfarintherapy and rifampicin. Ann. Intern. Med. 83:506-508.

37. Rohner, P., M. Peebo, D. P. Lew, R. Auckenthaler, and J. C.Pechere. 1992. Comparative in-vitro activity of new quinolonesagainst clinical isolates and resistant mutants. J. Antimicrob.Chemother. 29:41-48.

38. Rubin, J., G. Stoehr, V. L. Yu, R. R. Muder, A. Matador, andD. B. Kamerer. 1989. Efficacy of oral ciprofloxacin plus rifam-picin for treatment of malignant external otitis. Arch. Otolaryn-gol. Head Neck Surg. 1165:1063-1069.

39. Schrenzel, J., P. Dayer, E. Weidekamm, R. Portmann, and D. P.Low. 1992. Program Abstr. 32nd Intersci. Conf. Antimicrob.Agents Chemother., abstr. 1468.

40. Shalit, I., S. A. Berger, A. Gorea, and A. Frimerman. 1989.Widespread quinolone resistance among methicillin-resistantStaphylococcus aureus isolates in a general hospital. Antimi-crob. Agents Chemother. 33:593-594.

41. Soergel, F., R. Seelmann, K. Naber, R. Metz, and P. Muth. 1988.Metabolism of fleroxacin in man. J. Antimicrob. Chemother.22(Suppl. D):169-178.

42. Trucksis, M., D. C. Hooper, and J. S. Wolfson. 1991. Emerging

VOL. 37, 1993

on June 23, 2018 by guesthttp://aac.asm

.org/D

ownloaded from

2138 SCHRENZEL ET AL.

resistance to fluoroquinolones in staphylococci: an alert. Ann.Intern. Med. 114:424-426.

43. Venkatesan, K. 1992. Pharmacokinetic drug interactions withrifampicin. Clin. Pharmacokinet. 22(1):47-65.

44. Waldvogel, F. A. 1989. Use of quinolones for the treatment ofosteomyelitis and septic arthritis. Rev. Infect. Dis. 11(Suppl.5):S1259-1263.

45. Waldvogel, F. A. 1989. Treatment of osteomyelitis and septicarthritis with quinolone antimicrobial agents, p. 177-186. InJ. S. Wolfson and D. C. Hooper (ed.), Quinolone antimicrobialagents. American Society for Microbiology, Washington, D.C.

46. Waxman, D. J., C. Attisano, F. P. Guengerich, and D. P.Lapenson. 1988. Human liver microsomal steroid metabolism:identification of the major microsomal hormone 61-hydroxylasecytochrome P-450 enzyme. Arch. Biochem. Biophys. 263:424-436.

47. Weidekamm, E., R. Portmann, K. Suter, C. Partos, D. Dell, and

ANTIMICROB. AGENTS CHEMOTHER.

P. Lucker. 1988. Single- and multiple-dose pharmacokinetics offieroxacin, a trifluorinated quinolone, in humans. Antimicrob.Agents Chemother. 31:1909-1914.

48. Weinstein, M. P., R. G. Deeter, K. A. Swanson, and J. S. Gross.1991. Crossover assessment of serum bactericidal activity andpharmacokinetics of ciprofloxacin alone and in combinationin healthy elderly volunteers. Antimicrob. Agents Chemother.35:2352-2358.

49. Zak, O., W. M. Scheld, and M. A. Sande. 1983. Rifampicin inexperimental endocarditis due to Staphylococcus aureus inrabbits. Rev. Infect. Dis. 5(Suppl. 3):S481-S490.

50. Zhiri, A., M. Wellman-Bednawska, and G. Siest. 1986. ELISA of6,-hydroxycortisol in human urine: diurnal variations and ef-fects of antiepileptic therapy. Clin. Chim. Acta 157:267-276.

51. Zhiri, A., M. Wellman-Bednawska, and G. Siest. 1987. Le6,-hydroxycortisol: un reflet non invasif des enzymes du me-tabolisme des m6dicaments. Pathol. Biol. 35:1087-1093.

on June 23, 2018 by guesthttp://aac.asm

.org/D

ownloaded from