Pharmacometrics

The Journal of Clinical Pharmacology

Population Pharmacokinetics and Target 2015, 55(6) 639646
2015, The American College of
Attainment Analysis of Moxioxacin in Clinical Pharmacology
DOI: 10.1002/jcph.464
Patients With Diabetic Foot Infections

Sebastian G. Wicha1, Thomas Haak, MD, PhD2, Karl Zink, MD2,

Frieder Kees, PhD3, Charlotte Kloft, PhD1, and Martin G. Kees, MD1,4

Abstract
The objective of this study was to provide a pharmacokinetic/pharmacodynamic (PK/PD) analysis of moxioxacin in patients with diabetic foot
infections (DFI). The plasma concentration-time courses were determined in 50 DFI patients on day 1 and 3 after intravenous moxioxacin 400 mg
once-daily. A two-compartment population pharmacokinetic model was developed, identifying as covariates total body weight on central and
peripheral volume of distribution (V1, V2) and ideal body weight on clearance (CL), respectively. For a 70 kg patient V1 was 68.1 L (interindividual
variability, CV: 27.4%), V2 44.6 L, and CL 12.1 L/h (25.6%). Simulations were performed to calculate the probability of target attainment (PTA) for
Gram-positive and Gram-negative pathogens with fAUC/MIC targets of
30 and
100, respectively. PTA was 0.681 for susceptible (MIC 0.5 mg/L
according to EUCAST) Gram-positive, but <0.25 for Gram-negative pathogens with MIC
0.25 mg/L. With the exception of the rst 24 hours of
therapy, obesity affected PTA only marginally. Pharmacokinetic parameters in DFI patients were similar to those reported for healthy volunteers,
indicating the appropriateness of the standard dose of moxioxacin. Overall clinical efcacy has been shown previously, but PTA is limited in a
subpopulation infected with formally susceptible Gram-negative pathogens close to the EUCAST breakpoint.

Keywords
complicated skin and skin structure infections, uoroquinolones, tissue concentrations, pharmacokinetic/pharmacodynamic analysis

Diabetic foot infections (DFI) can be caused by a wide
range of Gram-positive and Gram-negative pathogens
including anaerobes, and are frequently polymicrobial.1
Antimicrobial regimens with an accordingly broad
spectrum are therefore required. Depending on clinical
presentation and the course of disease, switching to or even
starting with oral therapy is feasible and advantageous.1
Compared with other uoroquinolones such as cipro-
oxacin and levooxacin, moxioxacin has an improved
activity against Gram-positive pathogens and clinically
relevant activity against anaerobes.2 It is approved as a
once-daily at dosing regimen both for intravenous and
oral administration.3 These features make it a suitable
alternative for both empiric and targeted therapy.
In the case of DFI, all patients share the dening
primary disease, but may be more heterogenous with
respect to drug distribution and metabolism: particularly,
both progressing renal disease and obesity are frequent
comorbidities of diabetes mellitus, and could negatively
affect the reliability of the standard regimen with regard to
efcacy and safety.
The aim of this study was to describe the pharmacoki-
netics of intravenous moxioxacin in patients with DFI.
Population pharmacokinetic modeling and Monte Carlo
simulations were used to characterize the pharmacokinet-
ic variability and its impact on the probability to attain
relevant PK/PD targets. Additionally, moxioxacin
concentrations in tissue samples obtained during wound
debridement were determined and related to plasma
concentrations.
Methods
Study Design and Sampling Schedule
The open-label, non-randomized, monocentric pharma-
cokinetic study in patients was approved by the
responsible Ethics Committee of the Landesrztekammer
1
Department of Clinical Pharmacy and Biochemistry, Institute of
Pharmacy, Freie Universitaet Berlin, Berlin, Germany
2
Diabetes Center Mergentheim, Bad Mergentheim, Germany
3
Department of Pharmacology, University of Regensburg, Regens-
burg, Germany
4
Department of Anesthesiology and Intensive Care, Charit Universi-
ttsmedizin Berlin, Berlin, Germany
Submitted for publication 10 December 2014; accepted 15 January
2015.
Corresponding Author:
Martin G. Kees, MD, Department of Anesthesiology and Intensive
Care, Charit Universittsmedizin Berlin, Hindenburgdamm 30,
12200 Berlin, Germany
Email: martin.kees@charite.de,martin.kees@web.de
640
Hessen (Ref. FF63/2006) and the Federal Institute for
Drugs and Medical Devices, Bonn, Germany (EudraCT-
No. 2006-003955-20). Adult patients of both sexes with a
diagnosis of diabetes mellitus (type 1 or 2) hospitalized
for infected foot ulcer grade IIV according to the Wagner
classication (0 no ulceration, V gangrenous in-
volvement of the whole foot) were eligible for inclusion
into the study. Infection was diagnosed as a combination
of signs and symptoms of systemic and local inamma-
tion (elevated C-reactive protein, leucocyte count, or
erythrocyte sedimentation rate; purulent exudate, pain or
insensitivity, edema, erythema). Patients were excluded if
any of the following conditions was present: septic shock,
necessity of additional antimicrobial therapy besides
moxioxacin, life-threatening disease with expected
survival <48 hours, immunosuppression or neutropenia,
epilepsy or psychosis, pregnancy (or absence of a safe
method of contraception) or lactation, treatment with
moxioxacin within the previous 30 days, contraindica-
tions to moxioxacin as mentioned in the summary of
product characteristics (particularly hypersensitivity to
uoroquinolones, a history of uoroquinolone-associated
tendinopathy, comedications or conditions inducing QTc-
prolongation or favoring arrhythmias, impaired liver
function Child-Pugh C). After giving informed consent,
eligible patients were enrolled in the study and treated
with moxioxacin 400 mg i.v. once daily, infused over 1
hour. Venous blood was sampled on days 1 and 3
immediately prior to administration (0 hour), at the end of
infusion (1 hour), and between 24, 610, and 1824
hours after start of infusion. In patients with tissue lesions
and the necessity for surgical wound debridement,
samples of skin, soft tissue, necrotic tissue and/or, bone
were collected on day 3. All samples were stored at
20 C until analysis.
Bioanalytical Procedure
Total concentrations of moxioxacin were determined in
plasma and tissue homogenate by reversed-phase HPLC
with uorimetric detection as previously described.4,5
The lower limit of quantication was 0.020 mg/L in
plasma and 0.10 mg/kg in tissue, respectively. Intra- and
interassay imprecision and bias were calculated from co-
analyzed quality control samples of plasma and tissue
spiked with moxioxacin, and were <5% coefcient of
variation (CV) or relative error, respectively.
Population Pharmacokinetic Modeling
NONMEMTM 7.2 (Globomax, ICON Development
Solutions, Ellicott City, Maryland) was utilized for
population PK modeling and was executed via PsN (V
3.4.2). One and two-compartment models were assessed
using ADVAN 1 or ADVAN 3 subroutines in NON-
MEMTM. Interindividual variability was implemented on
the structural parameters:
The Journal of Clinical Pharmacology / Vol 55 No 6 2015
Pk;i uk  ehk;i
Pk,i represents the estimated kth PK parameter for the ith
individual calculated from the population PK parameter
uk of the typical DFI patient whilst hk,i represents the
deviation from the typical PK parameter assuming log-
normal distribution.
The residual variability in an individual patient at each
time point, that is, the difference between individual
model predicted (YPRED,i,j) and the observed moxiox-
acin concentration (YOBS,i,j) for the ith subject at the jth
time point, was estimated by a proportional (ep,i,j) or
combined proportional and additive (ea,i,j) residual
variability model:
YOBS;i;j YPRED;i;j  1 ep;i;j ea;i;j
Age, sex, total body weight, ideal body weight,6
plasma creatinine, and creatinine clearance (estimated
according to Cockcroft-Gault)7 were investigated as
potential covariates of the structural PK parameters.
Covariate relations were implemented by linear or power
models and assessed by statistical signicance (CI95% of
the estimate of the covariate effect did not include zero)
and clinical relevance (effect of covariate on PK
parameter change >20%), as suggested by Gastonguay.8
Goodness-of-t was assessed by plotting observed vs.
population or individual predicted concentrations of
moxioxacin. Further, the conditionally weighted resid-
uals were assessed and a visual predictive check was
performed (n 500).
Simulations and Pharmacodynamic Evaluation
The nal population PK model was implemented in R
(V. 3.0.2, R Foundation for Statistical Computing,
Vienna, Austria) to explore PK/PD relationships by
performing simulations. One thousand PK parameter
sets were randomly sampled from the (parametric) log-
normal distributions of clearance, central and peripheral
volume of distribution and from the non-parametric
weight distribution of the study population; these were
used for simulating 1000 moxioxacin concentration-
time curves. For each scenario, the areas under the
concentration-time curve on the rst day (AUC024) and
on the third day of treatment (AUC4872) were
calculated.
For the PK/PD analysis, the probability of target
attainment (PTA) was calculated as a function of the
minimum inhibitory concentration (MIC). Target values
were chosen from published values for the ratio of the area
under the free concentration-time curve to the MIC
(fAUC/MIC) being indicative of therapeutic success:
30
for Gram-positive and
100 for Gram-negative patho-
gens.9 Protein binding of moxioxacin was assumed to be
Wicha et al
40%, that is, unbound concentration 0.6  total
concentration.3
To assess the impact of the covariates total body
weight (TBW) and ideal body weight (IBW) on the PTA
within the heterogeneous DFI population, the AUC values
of 1000 patients with the lowest BMI (TBW 60 kg, IBW
57 kg), the highest BMI (TBW 162 kg, IBW 78 kg), the
standardized (TBW 70 kg, IBW 70 kg), and the typical
(median) body characteristics within the study population
(TBW 93 kg, IBW 67 kg) were simulated.
Results
A total of 51 patients were enrolled in the study. One
patient withdrew his consent before the rst dose. All
patients had a diagnosis of diabetes mellitus type 2 except
for one patient with type 1. For the pharmacokinetic
study, 50 patients (33 male, 17 female) contributed 408
timed plasma concentration samples (312 samples per
patient). Baseline characteristics were (median, range):
age 68 (3785) years, body weight 93 (60162) kg, height
173 (156198) cm, body mass index (BMI) 29.9
(22.047.8) kg/m2. Twenty ve patients (50%) had a
BMI <30, 22 (44%)
30 and <40, and 3 (6%)
40 kg/
m2. Plasma creatinine was available in 38 patients and
was 1.15 (0.712.70) mg/dL, corresponding to a creati-
nine clearance of 78 (23201) mL/min according to
Cockcroft-Gault. For missing covariates their typical
population value was imputed.
Population Pharmacokinetic Modeling
A two-compartment disposition model with rst-order
processes was superior to a one-compartment model in
terms of reduction of objective function value (DOFV:
219.83) and according to goodness-of-t plots. A
combined additive and proportional residual variability
model improved the predictivity of the model in
comparison to the proportional-only model (DOFV:
13.434). The additional residual variability component
was xed to its nal estimate to increase model stability.
Hence, a two-compartment model with linear elimination
and a combined additive and proportional residual
variability model was suitable to describe the PK of
moxioxacin in DFI patients.
Total body weight had a highly signicant effect on the
central and peripheral volumes of distribution (V1, V2)
(DOFV: 24.476, P< 1e-6); for V2, TBW explained the
variability between patients to such an extent that
(unexplained) interindividual variability on V2 did not
have to be considered anymore (DOFV: 2.01 with vs.
without interindividual variability on V2). For V1, 17.4%
of the interindividual variability was explained by TBW
reducing unexplained variability to 27.5% CV. As the
CI95% of the estimated exponents of the power model
(0.98) included 1.0 (0.6711.287), TBW was included
641
according to allometric scaling theory.10 Individual total
volume of distribution (V1 V2) ranged from 94.3 to
249.1 L in the studied population.
Estimated creatinine clearance had a weak but signi-
cant effect on moxioxacin clearance (CL) (exponent:
0.19; CI95%: 0.0560.324, DOFV: 4.43, P .035), which
was mainly driven by the collinearity of the Cockcroft-
Gault formula with TBW rather than by pure renal
function. Weight standardized creatinine clearance was not
signicant anymore (DOFV:1.937, P .164).11 As IBW
is the more plausible demographic covariate for renal
function in the presence of obesity, IBW was included as a
covariate on CL with an allometric scaling exponent of
0.75 because of physiological plausibility,10 which
improved the model (DOFV:3.696, P .055). Individual
moxioxacin clearance ranged from 6.5 to 19 L/h.
Inspection of the h-distribution (in presence of low v-
shrinkage12) of the nal covariate model revealed no
marked trends for all implemented and available further
covariates. Goodness-of-t, conditional weighted resid-
uals, and visual predictive check plots for the nal model
are shown in Figures 1 and 2, and indicated good
predictive performance. The nal parameter set is
presented in Table 1 indicating good precision for all
parameter estimates.
Pharmacodynamic Evaluation
For the studied population of DFI patients, the PTA was
1.0 for Gram-positive bacteria (fAUC/MIC
30) with a
MIC up to 0.25 mg/L, and 0.68/0.88 after the rst/third
dose for pathogens with a MIC of 0.5 mg/L, which is the
upper limit of the susceptibility range according to
EUCAST (Figure 3). Against Gram-negative bacteria
(fAUC/MIC
100), PTA was
0.9 at MIC 0.125 mg/L,
but <0.25 at higher MIC values. Attainment probabilities
of both targets were qualitatively not different after the
rst or the third dose.
The typical (median) DFI patient (TBW 93 kg, IBW
67 kg) displayed similar exposure to the patient with the
standardized body size characteristics (TBW 70 kg, IBW
70 kg). For the covariate effects on AUC and PTA, little
inuence was seen after multiple dosing (Figure 4). During
steady state, AUC is determined by clearance exclusively,
which is linked to ideal body weight with its rather narrow
distribution. Contrarily, due to the effect of total body
weight on volume of distribution, AUC024 (during the rst
24 hours of therapy) was substantially lower in the scenario
with highly obese patients (Figure 4A). Accordingly, in
this scenario the PTA for the rst 24 hours of therapy
started to decline at MICs one log2-step lower than in non-
obese patients (Figure 4B and C).
Tissue Concentrations
Tissue samples were obtained on day 3 of therapy during
surgical wound debridement in 26 patients in median 3.4
642 The Journal of Clinical Pharmacology / Vol 55 No 6 2015

Figure 1. Diagnostic plots of the nal PK model for intravenous moxioxacin in 50 patients with diabetic foot infection. Observed moxioxacin
concentrations versus (A) population predicted concentrations, and (B) individual predicted concentrations; solid line: line of identity. Conditional
weighted residuals (CWRES) versus population predicted concentrations (C), and time after rst dose (D).

(2.16.3) hours after moxioxacin infusion. The median
(range) moxioxacin concentrations were 5.2 (2.412.7;
n 23) mg/kg in soft tissue, 2.6 (0.303.7; n 7) mg/kg
in necrotic tissue, 2.8 (0.126.9; n 17) mg/kg in bone,
and 2.6 (1.74.8; n 6) mg/kg in skin, respectively. The
tissue concentrations in relation to the individual
10% observed
10 50% observed
Concentration (mg/L)
concurrent plasma concentrations (median 2.4 mg/L,
range 1.64.2 mg/L; n 26) are depicted in Figure 5.
Discussion
With this study, we provide a population pharmacokinetic
model and pharmacodynamic evaluation of intravenous
moxioxacin in patients suffering from diabetic foot
infections. The pharmacokinetic raw data (with compara-
tively few samples per patient, n 312, from a rather

90% observed large cohort) proved to be a reliable basis for this

8
10% predicted approach.
6 50% predicted As both obesity and renal disease are frequent
90% predicted
comorbidities of diabetes mellitus, and both body weight
4
and renal function are important factors in drug
2 disposition and excretion, one could expect different
pharmacokinetics in this population when compared with
0
0 12 24 48 60 72 healthy volunteers or other patient populations. More-
Time (h) over, as all stages of both conditions are certainly more
Figure 2. Visual predictive check of the nal PK model for
prevalent in diabetic patients but not obligatory, the
moxioxacin 400 mg i.v. once daily in patients with diabetic foot variability of the pharmacokinetic parameters may
infection. Lines represent percentiles of the respective distribution at be more affected than their mean values. Such variability
the center of binning intervals. between patients may render a xed dosing regimen
Wicha et al 643

Table 1. Typical Pharmacokinetic Parameters (u), With Respect to Allometric Covariate Relationships (Ideal Body Weight, IBW; Total Body Weight,
TBW), Unexplained Interindividual Variability (v2), and Residual Variability (s2) Obtained After Intravenous Moxioxacin in Patients With Diabetic
Foot Infection

PK Parameter Estimate RSE (%) Shrinkage (%)

Clearance u1  (IBW/70)0.75 [L/h]

u1 12.1 3.5
v12 0.0633 (25.6% CV) 24.8 4.8
Central Volume of
Distribution u2  (TBW/70)1.0 [L]
u2 68.1 5.1
v22 0.0725 (27.4% CV) 34.8 12
Intercompartimental Clearance u3 [L/h]
u3 20.3 14.5
Peripheral Volume of
Distribution u4  (TBW/70)1.0 [L]
u4 44.6 7.1
Residual Variability Model
s2additive 0.00465 (SD: 0.0682 mg/L) FIX 10
s2proportional 0.034 (18.4% CV) 28.9 10

CV, coefcient of variation; SD, standard deviation; RSE, relative standard error (reported on variance scale for variability parameters).

inappropriate for individual patients, leaving some
patients at risk for therapeutic failure and exposing others
to concentration-dependent adverse effects. In this study,
both body weight/body mass index and renal function
were well distributed throughout the normal and
pathological range. However, renal function had no
inuence on the pharmacokinetics of moxioxacin, and
both population means and variability of pharmacokinetic
parameters were in a similar range to those reported for
healthy volunteers,13 indicating the appropriateness of the
standard dosing regimen. This nding is in line with
studies on moxioxacin in morbidly obese patients or in
patients with renal dysfunction,5,14 which concluded that
Figure 3. Probability of target attainment following moxioxacin
400 mg i.v. once daily in patients with diabetic foot infections along with
the EUCAST breakpoint for fully susceptible isolates (dotted vertical
line).
dose adjustments for moxioxacin are generally unnec-
essary in either condition. The here reported analysis on
the extremes of the weight distribution suggests that a
formal loading dose might have a role in highly obese
patients in order to rapidly achieve equal drug exposure
and target attainment as in non-obese patients. However,
the safety and the actual clinical benet of such a loading
dose must be formally established before it can be
adopted. For this, further data from PK/PD studies with
obese patients are required.
One previous study specically examined the phar-
macokinetics of moxioxacin in patients with DFI both
after oral and intravenous administration, reporting
similar plasma concentrations and indicating near-
complete oral bioavailability.15 In that study, only non-
compartmental analysis and a basic estimation with
regard to the attainment of pharmacodynamic targets were
performed. By the more sophisticated analysis presented
here, additional quantitative data and interpretations are
provided.
Probability of target attainment was satisfactory for
Gram-positive pathogens with MICs below or at the
EUCAST susceptibility breakpoint of 0.5 mg/L for
moxioxacin. Contrarily, our analysis indicated insuf-
cient PTA for Gram-negative pathogens close to0.5 mg/L,
and sufciently high PTA was only attained for pathogens
with MICs 0.125 mg/L. This is not in conict with
demonstrated overall clinical efcacy of moxioxacin in
the treatment of complicated skin and skin structure
infections (cSSSI) including DFI,16,17 as the cumulative
target attainment rate for all pathogens in question is
much higher (and will come close to 100%) when only
few pathogens with high MIC and accordingly low PTA
644
Figure 4. Inuence of body weight on (A) the area under the
concentration-time curve over 24 hours (AUC; shown as median,
interquartiles and 10th/90th percentile of 1000 simulated patients each)
after rst and third dose, (B) probability of target attainment for Gram-
positive (fAUC
30), and (C) Gram-negative pathogens (fAUC
100)
following moxioxacin 400 mg i.v. once daily in patients with diabetic
foot infection (slim: BMI 22 kg m2, TBW 60 kg, IBW 57 kg; standard:
standardized patient with TBW 70 kg, IBW 70 kg; typical: patient with
population medians, BMI 29.9 kg m2, TBW 93 kg, IBW 67 kg; obese:
BMI 47.8 kg m2, TBW 162 kg, IBW 78 kg). BMI, body mass index; TBW,
total body weight; IBW, ideal body weight.
The Journal of Clinical Pharmacology / Vol 55 No 6 2015
are present in a population. In the cited study on DFI, the
MIC50 of the most commonly isolated Gram-negative
pathogen, non-ESBL producing Escherichia coli, was
only 0.03 mg/L.17 Single cases with pathogens with
higher MIC and lower chances of cure would not
signicantly reduce the success rate in the whole
population. Moreover, the applicability of generic PK/
PD target values to cSSSI and DFI may be limited, for
example, because of a partially anaerobic environment
and the importance of surgical interventions.
Therefore, infection specic target values would be
highly desirable. Although the chosen target values are
therefore debatable, it is generally accepted that much
higher fAUC/MIC values are required for Gram-negative
pathogens.9,18 At a rst glance, it seems therefore illogical
that the same susceptibility breakpoint (0.5 mg/L)
applies to both Gram-positives and Gram-negatives.
However, for the denition of a breakpoint further clinical
or microbiological data, the MIC distribution of the wild-
type population and also the imprecision of the fAUC/
MIC target value itself are considered.19 As for the latter,
a certain range is reported in the literature (eg, 3040 for
Gram-positives and 80100 for Gram-negatives),20 and
the precise result of our analysis would have been slightly
different if we had used other values. The same would
have been true if a different degree of protein binding had
been assumed, or if individual free concentrations had
actually been determined rather than calculated, intro-
ducing a further source of interindividual variability into
the model. However, the general shape of the PTA-vs.-
MIC curve would still depend on (1) the order of
magnitude of the fAUC/MIC target, which is at least twice
the value for Gram-negatives compared to Gram-
positives and (2) the pathogen MIC which varies by
several powers of two. The results of our analysis should
be interpreted with respect to the above-mentioned
aspects, and further data on the PK/PD relationship in
DFI are desirable. Until further evidence is available, we
recommend clinicians to be watchful in those rather rare
cases when Gram-negative pathogens with a moxiox-
acin MIC of
0.25 mg/L are isolated.
Concentration values in perinecrotic wound tissue (in
mg/g) in DFI have been reported previously to be fairly
equivalent to concurrent plasma concentrations (in
mg/L).15 Our data basically conrm this result, and agree
as well with data on the penetration of moxioxacin into
bone obtained during hip replacement surgery.21 Howev-
er, concentrations determined in tissue homogenate
should be interpreted with caution, most importantly
because no distinction can be made between the intra- and
the extracellular space.21,22 Particularly, high concen-
trations of uoroquinolones in tissue homogenate could
be due to enrichment within the cells, whereas most
bacterial infections reside in the extracellular space.
Therefore, the only rm conclusion we would draw from
Wicha et al
Figure 5. Individual and median (horizontal bar) ratios of tissue (mg/g)
to plasma concentrations (mg/L) of moxioxacin in tissue obtained
during surgical wound debridement in patients with diabetic foot
infections 2.16.3 hours after the third dose of moxioxacin 400 mg i.v.
once daily.
the tissue data is that the concentrations of moxioxacin
observed at the site of infection do not preclude good
efcacy as observed in clinical studies. In contrast to
sampling tissue homogenates, microdialysis is deemed
the method of choice to determine antimicrobial concen-
trations at the target site, that is, in the interstitial uid,23
and has also been applied to assess the tissue pharmaco-
kinetics of ciprooxacin, levooxacin, or moxioxacin in
patients with soft tissue infections.2426 For all three
drugs, interstitial concentrations were reported to be
roughly equal to unbound plasma concentrations, without
evidence of major differences between diabetic or non-
diabetic patients, or between the inamed site and a
non-affected control site. Accordingly, the tissue phar-
macokinetics of these agents with standard doses is
considered to be suitable for treatment of soft tissue
infections. An additional advantage of microdialysis is
that only the unbound, active fraction of the drug is
determined, thus eliminating a further variable in the
relationship between plasma concentrations and antibac-
terial effect. Thus, this technique theoretically offers the
opportunity to establish infection- and site-specicPK/PD
target values, which could be more predictive of clinical
outcome than values based on plasma concentrations.
However, such attempts are hindered so far by the
typically small sample size of clinical microdialysis
studies, and the high variability observed, which might
reect actual interindividual differences (and thus differ-
ent efcacy), but also the technical challenges of the
method.
To summarize, we developed a population pharma-
cokinetic model of moxioxacin for patients with
diabetic foot infection. Despite patient heterogeneity
with regard to body weight and renal function,
pharmacokinetic parameters, their variability, and drug
645
exposure were similar to those of healthy volunteers and
other patient populations, indicating the pharmacokinet-
ic reliability of the standard dose of moxioxacin in this
condition. Clinical efcacy has been shown previously,
but probability of target attainment is limited for
formally susceptible Gram-negative pathogens close to
the EUCAST breakpoint, which deserves further
attention.
Acknowledgments
Parts of the results have been presented at the 50th Interscience
Conference on Antimicrobial Agents and Chemotherapy
(September 1215, 2010, Boston, USA; poster A1683) and
at the 24th European Congress of Clinical Microbiology and
Infectious Diseases (May 1013, 2014, Barcelona, Spain; poster
P1745).
Declaration of Conicting Interests
SGW received a travelling grant, MGK and FK travelling grants
and speaker honoraria from Bayer Vital GmbH. The other
authors have no conicting interests to declare.
Funding
This work was funded in part through an unrestricted grant from
Bayer Vital GmbH, Leverkusen, Germany.
References
1. Lipsky BA, Berendt AR, Cornia PB, et al. Infectious Diseases
Society of America clinical practice guideline for the diagnosis and
treatment of diabetic foot infections. Clin Infect Dis. 2012;54(12):
e132e173.
2. Andersson MI, MacGowan AP. Development of the quinolones. J
Antimicrob Chemother. 2003;51(suppl 1):111.
3. Summary of Product CharacteristicsAvelox 400 mg/250 ml
solution for infusion. http://www.medicines.org.uk/emc/medicine/
22827/SPC
4. Beckmann J, Kees F, Schaumburger J, et al. Tissue concentrations
of vancomycin and Moxioxacin in periprosthetic infection in rats.
Acta Orthop. 2007;78(6):766773.
5. Kees MG, Weber S, Kees F, Horbach T. Pharmacokinetics of
moxioxacin in plasma and tissue of morbidly obese patients. J
Antimicrob Chemother. 2011;66(10):23302335.
6. Green B, Duffull SB. What is the best size descriptor to use for
pharmacokinetic studies in the obese. Br J Clin Pharmacol.
2004;58(2):119133.
7. Cockcroft DW, Gault MH. Prediction of creatinine clearance from
serum creatinine. Nephron. 1976;16(1):3141.
8. Gastonguay MR. A full estimation model approach for covariate
effects: inference based on clinical importance estimation precision.
AAPS J. 2004;6(S1):W4354.
9. Wispelwey B. Clinical implications of pharmacokinetics and
pharmacodynamics of uoroquinolones. Clin Infect Dis. 2005;41-
(suppl 2):S127S135.
10. Anderson BJ, Holford NHG. Mechanism-based concepts of size and
maturity in pharmacokinetics. Annu Rev Pharmacol Toxicol.
2008;48(1):303332.
11. Mould DR, Holford NHG, Schellens JHM, et al. Population
pharmacokinetic and adverse event analysis of topotecan in patients
with solid tumors. Clin Pharmacol Ther. 2002;71(5):334348.
646
12. Savic RM, Karlsson MO. Importance of shrinkage in empirical
bayes estimates for diagnostics: problems solutions. AAPS J.
2009;11(3):558569.
13. Stass H, Kubitza D. Pharmacokinetics and elimination of
moxioxacin after oral and intravenous administration in man. J
Antimicrob Chemother. 1999;43(90002):8390.
14. Stass H, Kubitza D, Halabi A, Delesen H. Pharmacokinetics of
moxioxacin, a novel 8-methoxy-quinolone, in patients with renal
dysfunction. Br J Clin Pharmacol. 2002;53(3):232237.
15. Majcher-Peszynska J, Sass M, Schipper S, et al. Pharmacokinetics
and penetration of moxioxacin into infected diabetic foot tissue in
a large diabetic patient cohort. Eur J Clin Pharmacol. 2011;67-
(2):135142.
16. Gyssens IC, Dryden M, Kujath P, et al. A randomized trial of the
efcacy and safety of sequential intravenous/oral moxioxacin
monotherapy versus intravenous piperacillin/tazobactam followed by
oral amoxicillin/clavulanate for complicated skin and skin structure
infections. J Antimicrob Chemother. 2011;66(11):26322642.
17. Schaper NC, Dryden M, Kujath P, et al. Efcacy and safety of IV/PO
moxioxacin and IV piperacillin/tazobactam followed by PO
amoxicillin/clavulanic acid in the treatment of diabetic foot infections:
results of the RELIEF study. Infection. 2013;41(1):175186.
18. Ambrose PG, Bhavnani SM, Owens RC Jr. Clinical pharmacodynam-
ics of quinolones. Infect Dis Clin North Am. 2003;17(3):529543.
19. Mouton JW, Brown DFJ, Apfalter P, et al. The role of
pharmacokinetics/pharmacodynamics in setting clinical MIC
The Journal of Clinical Pharmacology / Vol 55 No 6 2015
breakpoints: the EUCAST approach. Clin Microbiol Infect.
2012;18(3):E37E45.
20. MoxioxacinRationale for the EUCAST clinical beakpoints,
version 2.3. 2007. http://www.eucast.org/leadmin/src/media/
PDFs/EUCAST_les/Rationale_documents/Moxioxacin_ratio-
nale_2.3.pdf
21. Landersdorfer CB, Kinzig M, Hennig FF, et al. Penetration of
moxioxacin into bone evaluated by Monte Carlo simulation.
Antimicrob Agents Chemother. 2009;53(5):20742081.
22. Mouton JW, Theuretzbacher U, Craig WA, Tulkens PM, Derendorf
H, Cars O. Tissue concentrations: do we ever learn. J Antimicrob
Chemother. 2008;61(2):235237.
23. Brunner M, Derendorf H, Mller M. Microdialysis for in vivo
pharmacokinetic/pharmacodynamic characterization of anti-
infective drugs. Curr Opin Pharmacol. 2005;5(5):495499.
24. Mller M, Brunner M, Hollenstein U, et al. Penetration of
ciprooxacin into the interstitial space of inamed foot lesions in
non-insulin-dependent diabetes mellitus patients. Antimicrob
Agents Chemother. 1999;43(8):20562058.
25. Joukhadar C, Stass H, Mller-Zellenberg U, et al. Penetration of
moxioxacin into healthy and inamed subcutaneous adipose
tissues in humans. Antimicrob Agents Chemother. 2003;
47(10):30993103.
26. Bellmann R, Kuchling G, Dehghanyar P, et al. Tissue pharmacoki-
netics of levooxacin in human soft tissue infections. Br J Clin
Pharmacol. 2004;57(5):563568.