Pharmacodynamic (PK/PD) Integration

Edited by:
Valentina Meucci, Models
University of Pisa, Italy Longfei Zhang 1,2,3, Hongbing Xie 2, Yongqiang Wang 2, Hongjuan Wang 2, Jianhe Hu 2,3* and
Reviewed by: Gaiping Zhang 1*
Pierre-Louis Toutain,
Ecole Nationale
1
Postdoctoral Research Station, Henan Agriculture University, Zhengzhou, China, 2 College of Animal Science and
Vétérinaire de Veterinary
Toulouse, France Medicine, Henan Institute of Science and Technology, Xinxiang, China, 3 Postdoctoral Research Base, Henan Institute of
Ted Whittem, Science and Technology, Xinxiang, China
James Cook University, Australia
*Correspondence:
Jianhe Hu Pharmacokinetic/pharmacodynamic (PK/PD) integration models are used to
[email protected]; investigate the antimicrobial activity characteristics of drugs targeting pathogenic
[email protected]
Gaiping Zhang bacteria through comprehensive analysis of the interactions between PK and PD
[email protected] parameters. PK/PD models have been widely applied in the development of new
drugs, optimization of the dosage regimen, and prevention and treatment of drug-
Specialty section:
This article was submitted to resistant bacteria. In PK/PD analysis, minimal inhibitory concentration (MIC) is the
Veterinary Pharmacology and most commonly applied PD parameter. However, accurately determining MIC is
Toxicology, a section of the
journal Frontiers in challenging and this can influence the therapeutic effect. Therefore, it is necessary to
Veterinary Science optimize PD indices to generate more rational results. Researchers have attempted
Received: 23 January 2022 to optimize PD parameters using mutant prevention concentration (MPC)-based
Accepted: 24 February
2022 PK/PD models, multiple PD parameter-based PK/PD models, kill rate-based PK/PD
Published: 24 March 2022

Citation:
models, and others. In this review, we discuss progress on PD parameters for PK/PD
REVIEW
Zhang L, Xie H, Wang Y, models to provide a valuable reference for drug development, determining
published:24 March 2022 the
Wang H,
Hu J and Zhang G (2022)
dosage regimen, and preventing drug-resistant mutations. doi:10.3389/fvets.2022.860472
Pharmacodynamic Parameters Keywords: kill rate, mutant prevention concentration, PK/PD integration model, PD parameter, time-kill curve,
of MIC, multi-drug resistance, dosage regimen
Pharmacokinetic/
Pharmacodynamic (PK/PD)
Integration Models. Front. Vet.
Sci. 9:860472. doi:
10.3389/fvets.2022.860472
INTRODUCTION

Pharmacod The excessive use of antibiotics has promoted the emergence and selection of multi-drug-resistant
(MDR) pathogenic bacteria. MDR cause infections that threaten the health of humans, livestock,

ynamic and wild animals. Some Gram-negative (G−) bacteria harboring the mobilized colistin resistance
(mcr) gene are resistant to colistin, and this is leading to fewer and fewer antibiotics available for

Parameters clinical treatment of MDR infections (1, 2). Indeed, in the future there may be no drugs available
to treat bacterial infections caused by MDR pathogens. The main approaches to address this
of include the development of new drugs, optimization of dosage regimen, revision of susceptible
breakpoints, and drug combinations. Pharmacokinetic/pharmacodynamic (PK/PD) models can

Pharmacok comprehensively investigate interactions between hosts, pathogens, and drugs, and the obtained
PK/PD parameters can predict the clinical antibacterial efficacy (3, 4). Thus, PK/PD integration is an

inetic/ effective and practical method for addressing MDR bacterial infections, and it has been widely
applied to optimize the dosage regimen and reset the bacterial susceptible breakpoints.
Theoretically, the appropriate application of PK/PD integration has potential to improve the
outcomes on recovering antimicrobial activity, extending the usage life of an antimicrobial, and
preventing the emergence and spread of resistant bacteria (5–9).

Frontiers in Veterinary Science | www.frontiersin.org 1 March 2022 | Volume 10 | Article 860472

Zhang et al. PD Parameters of PK/PD Model

In PK/PD integration emergence of resistant mutations in bacterial infections (13–15). (a) The determination of MIC is
models, the most commonly an all-or-nothing result; a drug will exhibit antibacterial activity when its concentration is above
applied PK/PD parameters the MIC, and have no antibacterial effect when its concentration is below the MIC. This does not
are based on minimal reflect the real antibacterial activity of low drug concentrations against pathogens because
inhibitory concentration different types of drugs have different antibacterial characteristics based on their antibacterial
(MIC). MIC is a standard in mechanisms. (b) MIC is typically determined using the double dilution method, which may result
vitro value for a given in estimated values that are higher than real values. Administration of a dosage regimen based on
antibacterial against a MIC may result in drug residues and ecological damage after a long time. (c) MIC is determined
specific microorganism after using static drug concentrations, which does not reflect the dynamic antibacterial activities of
a fixed time period (16–18 drugs against pathogens, such as changes in kill rate and growth rate in the presence of different
h). In other words, MIC is a drug concentrations over different time periods. (d) The bacterial cell density is typically ∼10 5
net result when the growth CFU/mL for MIC determination. Although this density not representative of in vivo infectious
rate of microorganisms (the situations (the bacterial number may exceed 108 CFU/mL in serious infection), the mutant
population of inoculum) frequency (the natural mutant occurrence frequency is ∼10 −6) may largely increase when amount
equal to the kill rate of a of bacteria is exposed to antibacterial drugs, such as fluoroquinolones. These limitations may
given drug at the fixed result in the emergence of MDR bacteria. Thus, developing new drugs and designing dosage
concentration which regimens based only on MIC for PK/PD simulation is not sufficient. Therefore, it is necessary to
different from the dynamic explore PD parameters to more accurately investigate the antibacterial activities of drugs against
character of drug pathogens.
concentrations in vivo. So, In order to overcome the limitations of MIC-based PK/PD integration in clinical treatment,
MIC is a hybrid and researchers have conducted numerous studies to optimize PD parameters, including mutant
contextual PD variable due prevention concentration (MPC)-based PK/PD models, multiple PD parameter-based PK/PD
to its nature and the models, and kill rate-based PK/PD models. Therefore, this review firstly covers the basic methods,
conditions for obtaining it concepts, and mathematical models of MIC-based PK/PD integration. And then introduces other
(10, 11). The test medium PD parameter-based PK/PD integration approaches. This work provides a valuable guidance for
play an important influence optimizing the dosage regimen, developing new drugs, and preventing the emergence of MDR
on the numerical value of a bacteria.
MIC because of a possible
large matrix effect. In
standard test, the Mueller BASIC METHODS OF PK/PD INTEGRATION
Hinton Broth (MHB) was
chosen because it ensures PK/PD integration models include in vitro, ex vivo, and in vivo PK/PD models. For in vitro PK/PD
the rapid growth of bacteria. studies, peristaltic pump and hollow fiber models are the most commonly used methods for
But the growth rate can be simulating PK processes of drugs in hosts. The peristaltic pump model was applied to simulate
much slower in vivo which one-compartment, two-compartment model, and multiple-compartment models. The simplest
will inevitably change the peristaltic pump model consists of a storage compartment (blank medium), a central
value of the MIC because compartment (drug, bacteria, and medium), and an elimination compartment (waste medium)
the difference of the growth connected by rubber pipe, and the PK characteristics (elimination rate) of drugs in hosts are
rate, the duration of the simulated by a peristaltic pump. A magnetic bar is applied to mix the medium, and a thermostat is
incubations, and the size of used to ensure the optimal growth temperature for pathogens. After measuring the drug
the starting inoculum. For concentration and bacterial cell density, the antibacterial effect and PK/PD parameters are
example, the macrolides calculated and simulated to guide the design of the dosage regimen (16–21). The hollow fiber
have a large difference model consists of thousands of hollow fiber tubes that simulate multiple compartments (22– 29).
between in vitro MIC and The in vitro PK/PD model is simple, economical, and easy to operate, and it can directly describe
effective in vivo plasma the dynamic interaction between drugs and pathogens.
concentrations because of a The ex vivo PK/PD model is used to investigate the antibacterial activities of hosts, drugs, and
very significant matrix effect bacteria in drugcontaining body fluids rather than artificial medium. The most commonly applied
which has been body fluids are plasma, serum, and tissue cage fluid (TCF). After collecting these samples at
demonstrated for all different timepoints following drug administration, drug concentrations and timekill curves are
veterinary macrolides and determined, and a mathematical equation is applied to analyze the relationships between PK/PD
well explained by the fact parameters and antibacterial effect. TCF is a type of extracellular fluid that is typically targeted
that the MHB promotes the during bacterial infection. Thus, TCF is an ideal medium for ex vivo PK/PD integration studies. TCF
overexpression of efflux can be acquired through surgery by implanting a tissue cage between muscle and skin. After ∼4
pump (12). The test method weeks, granulation tissue surrounds the tissue cage and produces TCF. The TCF model has been
of MIC may also result in widely applied in pig, rabbit, cattle, camel, goat, and other animals (30–39). Uterine fluid (40) and
treatment failure and intestinal juice (41) have also been used for ex vivo PK/PD studies. However, although drug

Frontiers in Veterinary Science | www.frontiersin.org 2 March 2022 | Volume 10 | Article 860472

Zhang et al. PD Parameters of PK/PD Model

concentrations are still static comprehensively study the interactions between hosts, drugs, and pathogenic bacteria, and the
when study ex vivo kill results are typically consistent with clinical treatment (50–62).
curves, this model can
enable sampling of the
extracellular fluid at BASIC PK/PD PARAMETERS AND MATHEMATICAL MODELS
different times after dosing
which allow collection of For PK/PD integration, MIC is the most commonly applied PD variable used to calculate PK/PD
samples as the indices such as AUC/MIC (area under the curve of concentration divided by MIC), C max/MIC
concentration changes and (maximum concentration divided by MIC), and %T >MIC (percentage of time that drug
partially consider the concentrations exceed MIC during the dosage interval) (50, 63–67) (Figure 1). Actually, all PK/PD
influence of the host. indices must be calculated with free concentrations and not with total plasma concentrations
Previous in vivo PK/PD because MIC is a free concentration. Based on the closeness of the relationship between PK/PD
models have been used to parameters and antibacterial effect, antibacterial drugs can generally be divided into time-
study interactions between dependent drugs, concentration-dependent, and co-dependent drugs (3, 68–70). The represented
hosts, drugs, and pathogens. PK/PD parameter for time-dependent drugs is %T >MIC. This reflects an antibacterial effect that is
The most commonly used closely related to the duration of the drug concentration above MIC; the longer the duration of
are the tissue cage infection the drug concentration above MIC, the better the antibacterial activity. Time-dependent drugs
model (TCIM) and the target include β-lactam antibiotics and macrolides. However, different members of macrolides have
organ infection model different PK/PD dependencies (71), so, the antibacterial activity should be evaluated prudently.
(TOIM). The TCIM was For concentration-dependent drugs, representative PK/PD parameters are AUC/MIC and C max/MIC.
established by adding These reflect antibacterial effects that are closely related to drug concentrations above MIC; the
pathogens to the tissue cage higher of drug concentration, the better of the antibacterial activity. Well-known concentration-
model, a series of dosage dependent drugs are aminoglycosides and fluoroquinolones. However, for aminoglycosides, the
regimens were subsequently more appropriate PK/PD parameter is AUC/MIC rather than Cmax/MIC, because the Cmax/MIC
administrated for treatment, criterion was historically proposed for rodents and can not reflect the over time process (72, 73).
and TCF samples were For co-dependent drugs, the PK/PD parameter were AUC/MIC or %T >MIC against different
collected for drug bacteria, such as tetracycline and glycopeptides.
concentration measurement To confirm the most appropriate PK/PD parameters and predict the values of PK/PD
and bacterial population parameters for designing the dosage
counting at different
timepoints. After calculating
the PK/PD parameters and
antibacterial effect (the
change in bacterial
population), the relationship
between PK/PD parameters
and antibacterial effect was
fitted and the dosage was
predicted (42–49). The TOIM
approach can simulate
clinical infection by directly
inoculating bacteria into FIGURE 1Antimicrobial
| PK, PD, and PK/PD parameters based on
target organs, then (minimumMICinhibitory concentration) and MPC (mutant
prevention
concentration). The most commonly applied PK/PD are parameters
AUC/MIC
administering drugs at (area under the curve of antibacterial concentration divide MIC) or
different dosages and CAUC/MPC,
/MIC (maximum drug concentration divide /MPC,
MIC) or
and>T
max max
intervals. Finally, animals are MICC(the time of the drug concentration above MIC during the dosage
interval)
or > MPC (the time of the drug concentration above MPC during the dosage
killed at different timepoints
T
interval). MSW (mutant selection window) is the drug concentration
for bacterial population between
MIC and MPC. T(dash area) is the time that the
counting and antibacterial MSW
antibacterial
concentration inside MSW during the dosage
concentration determination interval.
at infected sites. The
obtained PK/PD parameters
and antibacterial effect are regimen, a mathematical model is needed. A commonly applied model for this purpose is the
then analyzed and Sigmoid Emax model (Equation 1). This sigmoid concentration-response model is a strongly
evaluated. These in vivo empirical model. This model is a useful tool to formalize our understanding of experimental data,
PK/PD models can but it has no clear relationship to real physical characteristics of the actual system being studied.

Frontiers in Veterinary Science | www.frontiersin.org 3 March 2022 | Volume 10 | Article 860472

Zhang et al. PD Parameters of PK/PD Model

Emax × CN high bacterial cell densities (1010 CFU/mL) that can inhibit the growth of first step mutant bacteria
(74, 75). In clinical infections, the bacterial cell density generally cannot exceed 10 12 CFU/mL,
E = EC
hence twostep mutations rarely occur under natural conditions, but they may easily arise under
(1)
low drug concentrations. Therefore, it is of great significance to prevent the emergence of drug-
resistant bacteria when designing the dosage regimen based on MPC. Meanwhile, a mutant
where Emax is the maximum selection window (MSW) theory was proposed based on MIC and MPC (the corresponding PK/PD
antibacterial effect of a parameters are listed in Figure 1). This theory considers that drug-resistant bacteria are easily
drug, EC50 is the value of induced in the middle part of the MSW (the upper part cannot produce effect because the size of
PK/PD parameters or drug the bacterial population is large, and the lower part may have no effects because of the smaller
concentrations at 50% Emax, bacterial population densities which have just been greatly reduced by previous concentrations;
C is the antibacterial Figure 2). Therefore, a rational dosage regimen should be carefully designed that is not only an
concentration or PK/PD effective bacterial treatment, but that also prevents the emergence and spread of resistant
parameter, and N (Hill bacteria.
coefficient) is the slope of Various studies have applied MPC as a PD variable (76– 81). MPC-based PK/PD integration has
the fitted curves between mainly been applied for fluoroquinolones, because the mechanism of resistant mutations in the
PK/PD parameters or drug MSW theory is consistent with the mutant resistance mechanism of fluoroquinolones that usually
concentrations and occurs gradually through gene point mutations. Cui et al. (79) applied a rabbit TCIM to study the
antibacterial effect. The antibacterial activity of levofloxacin against
correlation coefficient (R2) is Staphylococcus aureus, and analyzed the relationship between AUC 24h/MPC and the generation of
applied to evaluate the drug-resistant bacteria. The results showed that the emergence of resistant mutants could be
relationship between PK/PD inhibited when AUC24h/MPC >25 h. Liang et al. (81) applied three kinds of levofloxacin-resistant S.
parameters and aureus strains (with the same MIC but different MPC) to compare the difference between
antibacterial effect; the AUC24/MIC and AUC24/MPC as PK/PD parameters for inhibiting the generation of drug-resistant
higher of the value of R2, the bacteria. The results showed that generation of drug-resistant mutants could be inhibited when
closer the PK/PD AUC24/MPC values were between 22 and 25 (even for strains with different MPC values). When
parameters are related to applying AUC24/MIC as the target PK/PD parameter, the values were 8-fold different despite having
effect. the same MIC. Therefore,
AUC24/MPC was more suitable than AUC 24/MIC as the target PK/PD parameter to inhibit the
generation of drug-resistant mutations. Zhang et al. (46) carried out a mutation window study on
MPC-BASED
danofloxacin against Actinobacillus pleuoniae using a porcine TCIM, and analyzed the relationship
PK/PD between AUC24h/MPC and the generation of drug-resistant bacteria. The results showed that the
INTEGRATION generation of drug-resistant bacteria could be significantly inhibited when AUC 24h/MPC >18.58 h.
These studies shown that the susceptible or resistant mutant bacteria could be inhibited when the
To cope with the low
value of AUC/MIC or AUC/ MPC were 24 h which just means that the average plasma
bacterial cell density in MIC
concentration over 24 h is equal to the MIC or the MPC.
tests, some researchers
In order to expand the application of the MSW approach, various types of antibacterials have
have explored other PD
been studied. For timedependent drugs, %T> MIC is the PK/PD parameter that best reflects
parameters for serious
antibacterial activity. For MPC-based PK/PD parameters, %T >MPC was applied to prevent
infections (>1010 CFU/mL) for
mutations, which reflects the time percentage of the drug concentration within the MSW during
PK/PD integration. In serious
the interval of dosage administration. Alieva et al. (82) applied an in vitro dynamic model to study
infections, susceptible
the relationship between the residence time (T MSW) of linezolone concentration within the MSW
bacteria may produce a
and the emergence of drug-resistant bacteria. The results showed that the correlation between
natural gene mutation (first-
TMSW and T >MPC and area under the bacterial curve of mutation (AUBC M) was 0.99. Thus, TMSW was
step mutant bacteria) that
an important indicator for predicting the emergence and enrichment of drug-resistant bacteria.
are resistant to drugs. At
Xiong et al. (48) applied a rabbit TCIM to study resistance mutations for cefquinome in S. aureus,
low drug concentrations,
and the results showed that drugresistant bacteria appeared and accumulated when T >MIC 99
susceptible bacteria will
>70% or T >MPC <58%. Zhang et al. (49) applied a pig TCIM to study the MSW of cefquinome
gradually be killed and first
against Escherichia coli, and the results showed that the emergence and enrichment of resistant
step mutant bacteria may
bacteria occurred when T >MIC 99 >25% or T >MPC <50%. In these studies, TCIM was used for dose
reproduce rapidly, which
regimens testing because this model do not allow extrapolation to dose regimens for clinical use,
may result in recurrence of
and the concentration time profiles within tissue cages do not mirror those of extracellular fluid
infection and further
because of the perfusion barriers.
resistant mutations.
Therefore, there was an optimal value to prevent the emergence and spread of drug-resistant
Therefore, a new PD
bacteria when applying MPC-based PK/PD integration studies (83–87). However, the mechanism
variable was proposed for
of drug-resistant bacterial resistance have a variety of ways. Such as except for genetic mutations,
high bacterial cell densities.
the emergence of resistant bacteria may be due to the spread of exogenous plasmid, especially for
MPC is the MIC of drugs for

Frontiers in Veterinary Science | www.frontiersin.org 4 March 2022 | Volume 10 | Article 860472

 Zhang et al. PD Parameters of PK/PD Model

drugs other than bacterial growth rate constant and the death rate under different drug concentrations, and the
fluoroquinolones. Therefore, corresponding equation is applied to analyze the relationship between drug concentration and
the use of MSW theory is antibacterial effect. However, in static models, the drug concentration is constant, whereas
dispute which need more dynamic changes in drug concentration occur in the host. Dynamic models can be conducted by
exploration. peristaltic pump and hollow fiber models to simulate dynamic changes in drugs and antibacterial

5 10
anges of bacterial population and sensitivity under different antibacterialCFU/mL)
concentrations
and highCFU/
in low
W◦, susceptible bacteria. •, natural mutant
♦, first step mutant , multiple(10 mL) MIC,
step mutant bacteria.
bacteria.
MPC, bacteria.
mutant prevention concentration. MSW, minimum
mutant selection window, is the drug concentration between MIC and MPC. In low
will be gradually reduced
drug concentrations
with added from MIC to MPC, and the bacterial sensitivity may be decreased because
of during MSW. In high inoculum, a natural mutant bacteria may exist in the initial population. The total population will
ep resistant mutation
t subpopulations may be increased when the drug concentrations below obviously
MPC. The
decreased
sensitivity
because
may of
nd multiple step mutant bacteria during the

effects in the clinic, and they may more accurately reflect clinical therapeutic effects.
Nolting et al. (88) studied the in vitro bactericidal effect of piperacillin against E. coli and
MULTIPLE PD applied a modified multiparameter Emax model (Equation 2) to describe the bactericidal effect
PARAMETER- (dN/dt, change in bacteria over time). This model was used to investigate the maximum
bactericidal effect (Kmax), the normal growth rate of bacteria (Kgrowth), drug concentration (Ct), the
BASED PK/PD concentration that achieves a 50% maximum bactericidal effect (EC 50), the delayed growth
INTEGRATION constant (Z), and the initial bacterial population (N). This model can dynamically describe changes
in bacterial population over time under different drug concentrations. The results showed that the
To investigate antibacterial
bactericidal effect of piperacillin was closely related to the time of drug administration; the more
activity in more detail, time-
frequent the administration, the better the antibacterial effect. Compared with models base on
kill curves with multiple
MIC alone, this model can provide more detailed and accurate guidance for designing the dosage
endpoints can be applied for
regimen.
PK/PD integration.
Compared with PK/PD
integration based on MIC
alone, this multi-parameter-
dN =kgowth − Kmax × Ct × (1 − e−zt) × N (2) dt EC50 + Ct
based time-kill curve model
can provide detailed and
accurate descriptions of
dynamic changes in Regoes et al. (89) established a multiple PD parameter mathematical model (Equation 3) to study
bacterial growth and death the PK/PD integration of five antibacterials against E. coli. In this model, ψ(a) is the net bacterial
rates under different drug growth rate under different drug concentrations (a), ψ max is the maximum bacterial growth rate in
concentrations. Depending drug-free medium, ψmin is the minimum bacterial net growth rate in drugcontaining medium, N is
on whether drug the Hill coefficient (the slope of the graph of drug concentrations against bacterial net growth
concentrations change, this rate), and zMIC represent the MIC values of different bacteria. The results showed that when MIC
model can be divided into was the same, the larger the value of k, the better the bactericidal effect, and the smaller the
static and dynamic models. value of ψmin, the better the bactericidal effect. In this model, four PD parameters were employed,
Static models are mainly which can more precisely reflect the antibacterial characteristics of drugs to help design a more
conducted by comparing accurate and rational dosage regimen. Foerster et al. (90) also applied this model to study the
changes in the number of antibacterial activities of several types of drugs against five strains of Neisseria gonorrhoeae.
bacteria to obtain the

Frontiers in Veterinary Science | www.frontiersin.org 5 March 2022 | Volume 10 | Article 860472

Zhang et al. PD Parameters of PK/PD Model

concentrations. Compared with multiple PD parameter-based PK/PD

integration, this model can directly analyze the relationship between kill rate
N
and drug concentrations at different time periods. This method can dynamically
analyze changes in antibacterial activity over time, and it has been used to classify antibacterial

(3)

This model also have been

applied to classify bacteria
as susceptible or resistant.
Chauzy et al. (91) studied
the antibacterial activity of
polymyxin B against
susceptible and drug-
resistant Klebsiella
pneumoniae using two
consecutive bactericidal kill
curve experiments and
multiple PD parameterbased
FIGURE 3The| relationship curves between kill rate and concentration
PK/PD integration. The of
concentration-dependent (dotted line) and time-dependent (solid line)
results showed that after drugs.
For concentration-dependent drugs, the kill rate will gradually increase with
two consecutive bactericidal the
concentration added. For time-dependent drugs, the kill rate was
smaller
compared to concentration-dependent drugs and rapidly reach a plateau that
experiments, the bacteria
the value have little change with the concentration
whether appeared induced increased.
mutations could be
determined by analyzing
changes in growth and characteristics (concentration-dependent or time-dependent drugs). Two classical curves are
death rates. depicted in Figure 3 to describe the relationship between kill rate and drug concentration. For
In conclusion, compared time-dependent drugs, the value of kill rate increases with increasing drug concentration, but it is
PK/PD integration based on relatively small compared with concentration-dependent drugs at low concentrations.
MIC alone, multiple PD Furthermore, a maximum value is reached and it no longer increases with increasing drug
parameter-based PK/PD concentration after this point. For concentration-dependent antibacterials, the value of kill rate
models can more accurately increases quickly with increasing drug concentration. This antibacterial characteristic can be
reflect the antibacterial directly applied for drug classification.
effects of drugs against Ferro et al. (99) analyzed the antibacterial activities of a variety of antibacterial drugs against
pathogens, which has two types of fast-growing Mycobacterium strains. For M. abscess, the maximum kill rate occurred
important implications for between 24 and 72 h, and the maximum kill rate was 0.0427 h −1, 0.0231 h−1, and 0.0142 h−1 for
designing a rational dosage amikacin, clarithromycin, and cefoxitin, respectively. For M. fortuitum, the maximum kill rate
regimen to prevent the occurred between 3 and 24 h, and amikacin had the strongest antibacterial activity with a
emergence of resistant maximum bactericidal rate of 0.1933 h −1. Zhang et al. (100) analyzed the relationship between kill
mutant pathogens. rate and drug concentration for doxycycline against Mycoplasma gallisepticum at different time
periods, and the results showed that the optimal time period was 0-48 h (R 2 = 0.986) and the
PK/PD maximum kill rate was
0.11−1 h. Zhang et al. (101) studied the kill rate of cefquinome against A. pleuropneumoniae and
INTEGRATION
analyzed the relationship between kill rate and drug concentrations at different time periods. The
BASED ON KILL results showed that cefquinome exerted timedependent antibacterial activity, the optimal time
RATE period was 0-9 h (R2 = 0.9955), and the maximum kill rate was 0.48 log 10 CFU/mL/h. Maneke et al.
(102) compared the difference in kill rate between cephalexin and kanamycin alone and in
Kill rate is a PD reference
combination against E. coli, S. aureus, Streptococcus agalactiae, Streptococcus dysgalactiae, and
that can be obtained from
Streptococcus uberis. The results showed that the drug combination could increase the kill rate
timekill curves (92–98). It is
more than either drug alone, which can shorten the time required to achieve a bactericidal effect,
the slope of time-kill curves,
and the antibacterial activity was concentration-dependent.
which reflects the
Compared with MIC-based PK/PD integration, the kill rate for based PK/PD integration can
interaction between
reflect dynamic changes in antibacterial activity, which can be widely used in the selection of new
bacterial growth and death
drugs, design of the dosage regimen, and monitoring changes in bacterial sensitivity.
rates under different drug

Frontiers in Veterinary Science | www.frontiersin.org 6 March 2022 | Volume 10 | Article 860472

Zhang et al. PD Parameters of PK/PD Model

CONCLUSION Scientific Research Program of Henan Institute of Science and Technology (103010620002/004),
Postdoctoral Research Foundation of Henan Institute of Science and Technology, Program for
In conclusion, MIC based Innovative Research Team (in Science and Technology) in University of Henan Province
PK/PD is the most (22IRTSTHN026), and Key Scientific Research Projects of Universities in 2020 (20B230002).
commonly applied method REFERENCES
for evaluating antibacterial
1. Laxminarayan R, Sridhar D, Blaser M, Wang M, Woolhouse M. Achievingglobal targets for antimicrobial resistance.
activities. However, with the
Science. (2016) 353:874– 5. doi: 10.1126/science.aaf9286
emergence and spreading of 2. Liu YY, Wang Y, Walsh TR, Yi LX, Zhang R, Spencer J, et al. Emergenceof plasmid-mediated colistin resistance
resistant mutant bacteria, mechanism MCR-1 in animals and human beings in China: a microbiological and molecular biological study. Lancet
MIC-only-based PK/PD Infect Dis. (2016) 16:161–8. doi: 10.1016/S1473-3099(15)00424-7
integration cannot meet the 3. Rao GG, Landersdorfer CB. Antibiotic pharmacokinetic/pharmacodynamicmodeling: MIC, Pharmacodynamic indices,
and beyond. Int J Antimicrob Agents. (2021) 2021:106368. doi: 10.1016/j.ijantimicag.2021.106368
needs of clinical medicine
4. Rayner CR, Smith PF, Andes D, Andrews K, Derendorf H, FribergLE, et al. Model-Informed drug development for anti-
due to the limitations of infectives: state of the art and future. Clin Pharmacol Therapeutics. (2021) 109:867– 891. doi: 10.1002/cpt.2198
MIC. Therefore, it is 5. Rodríguez-Gascón A, Solinís MÁ, Isla A. The role of PK/PD analysis inthe development and evaluation of
necessary to optimize PD antimicrobials. Pharmaceutics. (2021) 13:833. doi: 10.3390/pharmaceutics13060833
parameters for detailed 6. Luo W, Chen D, Wu M, Li Z, Tao Y, Liu
Q, et al.
analysis of antibacterial
Pharmacokinetics/Pharmacodynamics models of veterinary antimicrobial agents. J Vet Sci. (2019) 20:e40. doi:
activity. Further research on 10.4142/jvs.2019.20.e40
MPC, multiple PD 7. Toutain PL, Pelligand L, Lees P, Bousquet-Mélou A, Ferran AA, Turnidge JD,et al. The
parameters, and kill rate- pharmacokinetic/pharmacodynamic paradigm for antimicrobial drugs in veterinary medicine: recent advances and
based PK/PD integration will critical appraisal. J Vet Pharmacol Ther. (2021) 44:172–200. doi: 10.1111/jvp.12917
8. Schmidt S, Barbour A, Sahre M, Rand KH, Derendorf H. PK/PD: newinsights for antibacterial and antiviral applications.
gradually overcome the Curr Opin Pharmacol. (2008) 8:549–56. doi: 10.1016/j.coph.2008.06.010
limitations of MIC, and 9. Asín-Prieto E, Rodríguez-Gascón A, Isla A. Applications of the pharmacokinetic/pharmacodynamic (PK/PD) analysis of
these could assist optimizing antimicrobial agents. J Infect Chemother. (2015) 21:319–29. doi: 10.1016/j.jiac.2015.02.001
the dosage regimen and 10. Mouton JW, Vinks AA. Pharmacokinetic/pharmacodynamic modelling of antibacterials in vitro and in vivo using
prevent or slow the bacterial growth and kill kinetics. Clin Pharmacokinet. (2005) 44:201–10. doi: 10.2165/00003088-200544020-00005
11. Mouton JW, Vinks AA. Relationship between minimum inhibitory concentration and stationary concentration
emergence of resistant revisited. Clin Pharmacokinet. (2005) 44:767–8. doi: 10.2165/00003088-200544070-00007
mutations. 12. Buyck JM, Plésiat P, Traore H, Vanderbist F, Tulkens PM, Bambeke FV. Increased susceptibility of Pseudomonas
aeruginosa to macrolides and ketolides in eukaryotic cell culture media and biological fluids due to decreased
expression of oprM and increased outer-membrane permeability. Clin Infect Dis. (2012) 55:534–42. doi:
AUTHOR 10.1093/cid/cis473
13. Mouton JW, Muller AE, Canton R, Giske CG, Kahlmeter G, Turnidge
CONTRIBUTIONS J. MIC-based dose adjustment: facts and fables. J Antimicrob Chemother. (2018) 73:564–8. doi: 10.1093/jac/dkx427
14. Mouton J W, Meletiadis J, Voss A, Turnidge J. Variation of MICmeasurements: the contribution of strain and
LZ, HX, YW, and HW laboratory variability to measurement precision. J Antimicrob Chemother. (2018) 73:2374– 9. doi: 10.1093/jac/dkz142
contributed to the 15. Mueller M, de la Pena A, Derendorf H. Issues in pharmacokineticsand pharmacodynamics of anti-infective agents: kill
methodology, software use, curves versus MIC. Antimicrob Agents Chemother. (2004) 48:369–
validation, formal analysis, 77. doi: 10.1128/AAC.48.2.369-377.2004
16. Budha NR, Lee RB, Hurdle JG, Lee RE, Meibohm B. A simple in vitro PK/PD model system to determine time–kill
and preparation of the curves of drugs against Mycobacteria. Tuberculosis. (2009) 89:378–85. doi: 10.1016/j.tube.2009.08.002
manuscript writing— 17. Zinner SH, Lubenko IY, Gilbert D, Simmons K, Zhao X, Drlica K, et al.Emergence of resistant Streptococcus pneumoniae
reviewing and editing. GZ in an in vitro dynamic model that simulates moxifloxacin concentrations inside and outside the mutant selection
and JH contributed to the window: related changes in susceptibility, resistance frequency and bacterial killing. J Antimicrob Chemother. (2003)
52:616– 22. doi: 10.1093/jac/dkg401
manuscript supervision. All
18. Liu P, Rand K H, Obermann B, Derendorf H. Pharmacokineticpharmacodynamic modelling of antibacterial activity of
authors contributed to the cefpodoxime and cefixime in in vitro kinetic models. Int J Antimicrob Agents. (2005) 25:120– 9. doi:
article and approved the 10.1016/j.ijantimicag.2004.09.012
submitted version. 19. Vaddady PK, Trivedi A, Rathi C, Madhura DB, Liu J, Lee RE, et al. Dynamictime-kill curve characterization of
spectinamide antibiotics 1445 and 1599 for the treatment of tuberculosis. Eur J Pharmaceut Sci. (2019) 127:233– 9.
doi: 10.1016/j.ejps.2018.11.006
20. Meletiadis J, Al-Saigh R, Velegraki A, Walsh TJ, Roilides E, Zerva L. Pharmacodynamic effects of simulated standard
FUNDING doses of antifungal drugs against Aspergillus species in a new in vitro pharmacokinetic/pharmacodynamic model.
Antimicrob Agents Chemother. (2012) 56:403–40. doi: 10.1128/AAC.00662-11
This work was supported by
21. Blaser J. In-vitro model for simultaneous simulation of the serum kinetics of two drugs with different half-lives. J
the Postdoctoral Research Antimicrob Chemother. (1985) 15:125– 30. doi: 10.1093/jac/15.suppl_a.125
Foundation of Henan 22. Marchbanks CR, McKiel JR, Gilbert DH, Robillard NJ, Painter B, ZinnerSH, et al. Dose ranging and fractionation of
Province (202002067), Key intravenous ciprofloxacin against Pseudomonas aeruginosa and Staphylococcus aureus in an in vitro model of
Technology Research and infection. Antimicrob Agents Chemother. (1993) 37:1756– 63. doi: 10.1128/AAC.37.9.1756
23. Cadwell JJS. The hollow fiber infection model for antimicrobial pharmacodynamics and pharmacokinetics. Adv
Development Program of
Pharmacoepidem Drug Safety S. (2012) 1:2167–1052. doi: 10.4172/2167-1052.S1-007
Henan Province 24. Gumbo T, Louie A, Deziel MR, Drusano GL. Pharmacodynamic evidencethat ciprofloxacin failure against tuberculosis is
(212102110373), the not due to poor microbial kill but to rapid emergence of resistance. Antimicrob Agents Chemother. (2005) 49:3178–

Frontiers in Veterinary Science | www.frontiersin.org 7 March 2022 | Volume 10 | Article 860472

Zhang et al. PD Parameters of PK/PD Model

81. doi: pathogens Actinobacillus pleuropneumoniae and Pasteurella multocida. J Vet Pharmacol Ther. (2017) 40:505–16. doi:
10.1128/AAC.49.8.3178- 10.1111/jvp.12385
3181.2005 31. Dorey L, Pelligand L, Cheng Z, Lees P. Pharmacokinetic/pharmacodynamicintegration and modelling of florfenicol for
25. Louie A, Grasso C, Bahniuk N, the pig pneumonia pathogens
Scoy BV, Brown DL, Kulawy R, Actinobacillus pleuropneumoniae and Pasteurella multocida. PLoS ONE. (2017) 12:e0177568. doi:
et al. The combination of 10.1371/journal.pone.0177568
meropenem and levofloxacin 32. Dorey L, Pelligand L, Lees P. Prediction of marbofloxacin dosage for thepig pneumonia pathogens Actinobacillus
is synergistic with respect to pleuropneumoniae and Pasteurella multocida by pharmacokinetic/pharmacodynamic modelling. BMC Vet Res. (2017)
both Pseudomonas 13:1–10. doi: 10.1186/s12917-017-1128-y
aeruginosa kill rate and 33. Aliabadi FS, Ali BH, Landoni MF, Lees P. Pharmacokinetics and PK-PDmodelling of danofloxacin in camel serum and
resistance suppression. tissue cage fluids. Vet J. (2003) 165:104–18. doi: 10.1016/S1090-0233(02)00258-7
Antimicrob Agents 34. Aliabadi FS, Landoni MF, Lees P. Pharmacokinetics (PK), pharmacodynamics (PD), and PK-PD integration of
Chemother. (2010) 54:2646– danofloxacin in sheep biological fluids. Antimicrob Agents Chemother. (2003) 47:626–35. doi: 10.1128/AAC.47.2.626-
54. 10.1128/AAC.00065-10 635.2003
26. Louie A, Heine HS, Kim K, 35. Aliabadi FS, Lees P. Pharmacokinetics and pharmacodynamics of danofloxacin in serum and tissue fluids of goats
Brown DL, Scoy BV, Liu W, et following intravenous and intramuscular administration. Am J Vet Res. (2001) 62:1979–89. doi:
al. Use of an in vitro 10.2460/ajvr.2001.62.1979
pharmacodynamic model to 36. Sidhu P, Rassouli A, Illambas J, Potter T, Pelligand L, Rycroft A, et al. Pharmacokinetic–pharmacodynamic integration
derive a linezolid regimen and modelling of florfenicol in calves. J Vet Pharmacol Ther. (2014)
that optimizes bacterial kill 37:231–42. doi: 10.1111/jvp.12093
and prevents emergence of 37. Li Z, Liu W, Su J, Chu X, Wang C. Pharmacokinetics and ex vivo pharmacodynamics of Minocycline against Salmonella
resistance in Bacillus abortus equi in donkey plasma and tissue cage fluid. Res Vet Sci. (2021) 135:293–
anthracis. Antimicrob Agents 6. doi: 10.1016/j.rvsc.2020.10.003
Chemother. (2008) 52:2486– 38. Zhang BX, Lu XX, Gu XY, Li XH, Gu MX, Zhang N, et al. Pharmacokineticsand ex vivo pharmacodynamics of cefquinome
96. doi: 10.1128/AAC.01439- in porcine serum and tissue
07 cage fluids. Vet J. (2014) 199:399–405. doi: 10.1016/j.tvjl.2013.12.015
27. Jacobsson S, Golparian D, 39. Zhou Q, Zhang G, Wang Q, Liu W, Huang Y, Yu P, et al. Pharmacokinetic/pharmacodynamic modeling of tulathromycin
Oxelbark J, Alirol E, against
Franceschi F, GustafssonTN, Pasteurella multocida in a porcine tissue cage model. Front Pharmacol. (2017) 8:392. doi: 10.3389/fphar.2017.00392
et al. Pharmacodynamic 40. Maan MK, Sattar A, Mi K, Shabbir MAB, Xie S, Xin L, et al. Integration of PK/PD for dose optimization of aditoprim
evaluation of dosing, against
bacterial kill, and resistance Trueperella pyogenes causing endometritis in bovines. Microb Pathog. (2020) 142:104097. doi:
suppression for Zoliflodacin 10.1016/j.micpath.2020.104097
against neisseria 41. Lei Z, Liu Q, Xiong J, Yang B, Yang S, Zhu Q, et al. Pharmacokineticand pharmacodynamic evaluation of marbofloxacin
gonorrhoeae in a dynamic and PK/PD modeling against Escherichia coli in pigs. Front Pharmacol. (2017) 8:542. doi: 10.3389/fphar.2017.00542
hollow fiber infection model. 42. Zhang L, Wu X, Huang Z, Zhang N, Wu Y,
Front Pharmacol. (2021) Cai Q, et al.
12:1225. doi: Pharmacokinetic/pharmacodynamic assessment of cefquinome against Actinobacillus pleuropneumoniae in a piglet
10.3389/fphar.2021.682135 tissue cage infection model. Vet Microbiol. (2018) 219:100–106. doi: 10.1016/j.vetmic.2018.02.027
28. Bhagunde P, Zhang Z, Racine 43. Cao C, Qu Y, Sun M, Qiu Z, Huang X, Huai B, et al. In vivo antimicrobial activity of marbofloxacin against Pasteurella
F, Carr D, Wu J, Young K, et al. multocida in a tissue cage model in calves. Front Microbiol. (2015) 6:759. doi: 10.3389/fmicb.2015.00759
A 44. Zeng QL, Mei X, Su J, Li XH, Xiong WG, Lu Y, et al. Integratedpharmacokinetic–pharmacodynamic (PK/PD) model to
translationalpharmacokinetic evaluate the in vivo antimicrobial activity of Marbofloxacin against Pasteurella multocida in
/pharmacodynamic model to piglets. BMC Vet Res. (2017) 13:1–7. doi: 10.1186/s12917-017-1099-z
characterize bacterial kill in
45. Zhang L, Wu X, Huang Z, Kang Z, Chen Y, Shen X, et al. Pharmacokinetic/pharmacodynamic integration of cefquinome
the presence of imipenem- against
relebactam. Int J Infect Dis.
Pasteurella Multocida in a piglet tissue cage model. J Vet Pharmacol Ther. (2019) 42:60–6. doi: 10.1111/jvp.12705
(2019) 89:55–
46. Zhang L, Kang Z, Yao L, Gu X, Huang Z,
61. doi:
Cai Q, et al.
10.1016/j.ijid.2019.08.026
Pharmacokinetic/pharmacodynamic integration to evaluate the changes in susceptibility of Actinobacillus
29. Sadouki Z, McHugh TD,
pleuropneumoniae after repeated administration of Danofloxacin. Front Microbiol. (2018) 9:2445. doi:
Aarnoutse R, Canseco JO,
10.3389/fmicb.2018.02445
Darlow C, Hope W, et
47. Yao Q, Gao L, Xu T, Chen Y, Yang X, Han M, et al. Amoxicillinadministration regimen and resistance mechanisms of
al.Application of the hollow
Staphylococcus aureus established in tissue cage infection model. Front Microbiol. (2019) 10:1638. doi:
fibre infection model (HFIM)
10.3389/fmicb.2019.01638
in antimicrobial
48. Xiong M, Wu X, Ye X, Zhang L, Zeng S, Huang Z, et al. Relationshipbetween cefquinome PK/PD parameters and
development: a systematic
emergence of resistance of Staphylococcus aureus in rabbit tissue-cage infection model. Front Microbiol. (2016)
review and
7:874. doi: 10.3389/fmicb.2016.00874
recommendations of
49. Zhang B, Gu X, Li Y, Li X, Gu M, Zhang N, et al. In vivo evaluation of mutant selection window of cefquinome against
reporting. J Antimicrob
Escherichia coli in piglet tissue-cage
Chemother. (2021) 76:2252–
model. BMC Vet Res. (2014) 10:1–8. doi: 10.1186/s12917-014-0297-1
9. doi: 10.1093/jac/dkab160
50. Greko C, Finn M, Franklin A, Bengtsson B.
30. Dorey L, Pelligand L, Cheng Z,
Pharmacokinetic/pharmacodynamic relationship of danofloxacin against Mannheimia haemolytica in a tissue-cage
Lees P.
model in calves. J Antimicrob Chemother. (2003) 52:253–57. doi: 10.1093/jac/dkg339
Pharmacokinetic/pharmacod
51. Xiao X, Lan W, Zhao Y, Li R, Liu Y, Liu J, et al. In vivo pharmacokinetic and pharmacodynamic (PK/PD) modeling and
ynamicintegration and
establishment of the PK/PD Cutoff of florfenicol against pasteurella multocida in ducks. Front Microbiol. (2021)
modelling of oxytetracycline
11:3321. doi: 10.3389/fmicb.2020.616685
for the porcine pneumonia

Frontiers in Veterinary Science | www.frontiersin.org 8 March 2022 | Volume 10 | Article 860472

Zhang et al. PD Parameters of PK/PD Model

52. Xiao X, Jiang L, Lan W, Jiang Y, spectrum beta-lactamase-producing Escherichia coli in vitro and in vivo in a murine thigh infection model. Pharm Res.
Wang Z. In vivo (2021) 38:27– 35. doi: 10.1007/s11095-020-02977-8
pharmacokinetic/Pharmacod 59. Li Y, Xie M, Zhou J, Lin H, Xiao T, Wu L, et al. Increased antimicrobialactivity of colistin in combination with
ynamic modeling of gamithromycin against pasteurella multocida in a neutropenic murine lung infection model. Front Microbiol. (2020)
Enrofloxacin against 11:2230. doi: 10.3389/fmicb.2020.511356
Escherichia coli in broiler 60. Li X, Chen Y, Xu X, Li Y, Fan Y, Liu X, et al. Pharmacokineticsand pharmacodynamics of nemonoxacin in a neutropenic
chickens. BMC Vet Res. murine lung infection model against streptococcus pneumoniae. Front Pharmacol. (2021) 12:1050. doi:
(2018) 10.3389/fphar.2021.658558
14:1–9. doi: 10.1186/s12917- 61. Watanabe E, Matsumoto K, Ikawa K, Yokoyama Y, Shigemi A, Enoki Y,et al. Pharmacokinetic/pharmacodynamic
018-1698-3 evaluation of teicoplanin against Staphylococcus aureus in a murine thigh infection model. J Glob Antimicrob
53. Xiao X, Pei L, Jiang LJ, Lan Resistance. (2021) 24:83–7. doi: 10.1016/j.jgar.2020.11.014
WX, Xiao JY, Jiang YJ, et al. In 62. Drusano GL. Role of pharmacokinetics in the outcome of infections.Antimicrob Agents Chemother. (1988) 32:289–97.
VivoPharmacokinetic/Pharm doi: 10.1128/aac.32.3.289
acodynamic profiles of 63. Craig WA. Pharmacokinetic/pharmacodynamic parameters: rationale forantibacterial dosing of mice and men. Clin
danofloxacin in rabbits Infect Dis. (1998) 26:1– 10. doi: 10.1086/516284
infected with Salmonella 64. Mouton JW, Dudley MN, Cars O, Derendorf H, Drusano GL. Standardization of pharmacokinetic/pharmacodynamic
typhimurium after oral (PK/PD) terminology for anti-infective drugs. Int J Antimicrob Agents. (2002) 19:355–8. doi: 10.1016/S0924-
administration. Front 8579(02)00031-6
Pharmacol. (2018) 9:391. 65. Mouton JW, Dudley MN, Cars O, Derendorf H, Drusano GL. Standardization of pharmacokinetic/pharmacodynamic
doi: (PK/PD) terminology for anti-infective drugs: an update. J Antimicrob Chemother. (2005) 55:601–7. doi:
10.3389/fphar.2018.00391 10.1093/jac/dki079
54. Zeng D, Sun M, Lin Z, Li M, 66. Lees P, Giraudel J, Landoni MF, Toutain PL. PK–PD integration and PK–PD modelling of nonsteroidal anti-inflammatory
Gehring R, Zeng Z. drugs: principles and applications in veterinary pharmacology. J Vet Pharmacol Ther. (2004) 27:491–502. doi:
Pharmacokineticsand 10.1111/j.1365-2885.2004.00618.x
pharmacodynamics of 67. Meibohm B, Derendorf H. Basic concepts of pharmacokinetic/pharmacodynamic (PK/PD) modelling. Int J Clin
tildipirosin against Pharmacol Ther. (1997) 35:401–13. doi: 10.1097/00004850-199710006-00006
Pasteurella multocida in a 68. Sumi CD, Heffernan AJ, Lipman J, Roberts JA, Sime FB. What antibioticexposures are required to suppress the
murine lung infection model. emergence of resistance for Gramnegative bacteria? A systematic review. Clin Pharmacokinet. (2019) 58:1407– 43.
Front Microbiol. (2018) doi: 10.1007/s40262-019-00791-z
9:1038. doi: 69. Yan L, Xie S, Chen D, Pan Y, Tao Y, Qu W, et al. Pharmacokinetic andpharmacodynamic modeling of cyadox against
10.3389/fmicb.2018.01038 Clostridium perfringens in swine. Sci Rep. (2017) 7:1–11. doi: 10.1038/s41598-017-03970-9
55. Zhang N, Wu Y, Huang Z, Yao 70. Onufrak NJ, Forrest A, Gonzalez D. Pharmacokinetic and pharmacodynamic principles of anti-infective dosing. Clin
L, Zhang L, Cai Q, et al. The Ther. (2016)
PK– 38:1930–47. doi: 10.1016/j.clinthera.2016.06.015
PD relationship and resistance 71. Girard D, Finegan SM, Dunne MW, Lame ME. Enhanced efficacy ofsingle-dose versus multi-dose azithromycin
development of danofloxacin regimens in preclinical infection models. J Antimicrob Chemother. (2005) 56:365–71. doi: 10.1093/jac/dki241
against Mycoplasma 72. Mouton JW, Brown DFJ, Apfalter P, Cantón R, Giske CG, Ivanova M, et al. The role of
gallisepticum in an In Vivo pharmacokinetics/pharmacodynamics in setting clinical MIC breakpoints: the EUCAST approach. Clin Microbiol Infect.
infection model. Front (2012) 18:E37– E45. doi: 10.1111/j.1469-0691.2011.03752.x
Microbiol. (2017) 8:926. 73. Kitamura Y, Yoshida K, Kusama M, Sugiyama Y. A proposal of apharmacokinetic/pharmacodynamic (PK/PD) index map
10.3389/fmicb.2017.00926 for selecting an optimal PK/PD index from conventional indices (AUC/MIC, C max/MIC, and TAM) for antibiotics. Drug
56. Yang Q, Liu X, Zhang C, Yong Metab Pharmacokinet. (2014) 29:455– 62. doi: 10.2133/dmpk.DMPK-14-RG-013
K, Clifton AC, Ding H, et al. 74. Blondeau JM. New concepts in antimicrobial susceptibility testing: themutant prevention concentration and mutant
Pharmacokinetics and selection window approach. Vet Dermatol. (2009) 20:383–96. doi: 10.1111/j.1365-3164.2009.00856.x
pharmacodynamics of 75. Blondeau JM, Hansen G, Metzler K, Hedlin P. The role of PK/PD parameters to avoid selection and increase of
gamithromycin treatment of resistance: mutant prevention concentration. Journal of Chemotherapy. (2004) 16:1–19. doi:
Pasteurella multocida in a 10.1080/1120009X.2004.11782371
murine lung infection model. 76. Drlica K. The mutant selection window and antimicrobial resistance. J Antimicrob Chemother. (2003) 52:11–17. doi:
Front Pharmacol. (2019) 10.1093/jac/dkg269
10:1090. doi: 77. Drlica K, Zhao X. Mutant selection window hypothesis updated. Clin Infect Dis. (2007) 44:681–8. doi: 10.1086/511642
10.3389/fphar.2019.01090 78. Xilin Z, Drlica K. Restricting the selection of antibiotic-resistant mutantbacteria: measurement and potential use of
57. Nakamura R, Ito-Horiyama T, the mutant selection window. J Infect Dis. (2002) 185:561–5. doi: 10.1086/338571
Takemura M, Toba S, 79. Cui J, Liu Y, Wang R, Tong W, Drlica K, Zhao X. The mutant selectionwindow in rabbits infected with Staphylococcus
Matsumoto S,Ikehara T, et al. aureus. J Infect Dis. (2006) 194:1601–8. doi: 10.1086/508752
In vivo pharmacodynamic 80. Lozano-Huntelman NA, Singh N, Valencia A, Valencia A, Mira P, Sakayan M, et al. Evolution of antibiotic cross-
study of cefiderocol, a novel resistance and collateral sensitivity in Staphylococcus epidermidis using the mutant prevention concentration and the
parenteral siderophore mutant selection window. Evol Appl. (2020) 13:808– 23. doi: 10.1111/eva.12903
cephalosporin, in murine 81. Liang B, Bai N, Cai Y, Wang R, Drlica K, Zhao X. Mutant preventionconcentration-based
thigh and lung infection pharmacokinetic/pharmacodynamic indices as dosing targets for suppressing the enrichment of levofloxacin-resistant
models. Antimicrob Agents subpopulations of Staphylococcus aureus. Antimicrob Agents Chemother. (2011) 55:2409–12. doi:
Chemother. (2019) 10.1128/AAC.00975-10
63:e02031– 18. doi: 82. Alieva KN, Strukova EN, Golikova MV, Portnoy YA, Zinner SH, Firsov AA. Time inside the mutant selection window as a
10.1128/AAC.02031-18 predictor of staphylococcal resistance to linezolid. J Antibiot. (2018) 71:514– 21. doi: 10.1038/s41429-017-0016-9
58. Tashiro S, Hayashi M, 83. Firsov, AA, Portnoy YA, Zinner SH. PK/PD-based prediction of “antimutant” antibiotic exposures using in vitro dynamic
Takemura W, Igarashi Y, Liu X, models. In: Antimicrobial Resistance in the 21st Century. Cham: Springer (2018). pp. 643–66.
Mizukami Y,et al.
84. Tängdén T, Martín VR, Felton TW, Nielsen EI, Marchand S, BrüggemannRJ, et al. The role of infection models and
Pharmacokinetics/Pharmaco
PK/PD modelling for optimising care of critically ill patients with severe infections. Intensive Care Med. (2017)
dynamics evaluation of
43:1021–32. doi: 10.1007/s00134-017-4780-6
flomoxef against extended-

Frontiers in Veterinary Science | www.frontiersin.org 9 March 2022 | Volume 10 | Article 860472

Zhang et al. PD Parameters of PK/PD Model

85. Li Y, Feng B, Gu X, Yang, D, experimental trick to discriminate between PK/PD models with distinct heterogeneous sub-populations versus
Zeng Z, et al. Correlation of homogenous population with adaptive resistance. Antimicrob Agents Chemother. (2020) 64:e00788–20. doi:
PK/PD indices withresistance 10.1128/AAC.00788-20
selection for cefquinome 92. Nielsen EI, Friberg LE. Pharmacokinetic-pharmacodynamic modeling of antibacterial drugs. Pharmacol Rev. (2013)
against Staphylococcus 65:1053– 90. doi: 10.1124/pr.111.005769
aureus in an in vitro model. 93. Mi K, Zhou K, Sun L, Hou Y, Ma W, Xu X, et al. Applicationof semi-mechanistic pharmacokinetic and pharmacodynamic
Front Microbiol. (2016) model in antimicrobial resistance. Pharmaceutics. (2022)
7:466. doi: 14:246. doi: 10.3390/pharmaceutics14020246
10.3389/fmicb.2016.00466 94. Grégoire N, Raherison S, Grignon C, Comets E, Marliat M, Ploy M,et al. Semimechanistic pharmacokinetic-
86. Yang F, Chen P, Wang H, Xing, pharmacodynamic model with adaptation development for time-kill experiments of ciprofloxacin against
X, Wang S, et al. Comparative Pseudomonas aeruginosa. Antimicrob Agents Chemother. (2010) 54:2379– 84. doi: 10.1128/AAC.01478-08
minimalinhibitory and 95. Sy SK, Zhuang L, Xia H, Beaudoin ME, Schuck VJ, Nichols WW,et al. A mathematical model-based analysis of the time-
mutant prevention kill kinetics of ceftazidime/avibactam against Pseudomonas aeruginosa. J Antimicrob Chemother. (2018) 73:1295–
concentration of eight 304. doi: 10.1093/jac/dkx537
antimicrobial agents against 96. van Os W, Zeitlinger M. Predicting antimicrobial activity at the target site: pharmacokinetic/pharmacodynamic indices
Klebsiella pneumoniae. versus time–kill approaches. Antibiotics. (2021) 10:1485. doi: 10.3390/antibiotics10121485
Microbial Drug Resistance. 97. Kristoffersson AN, Bissantz C, Okujava R, Haldimann A, Walter I, Shi T, et al.A novel mechanism-based
(2021) 28:229–35. doi: pharmacokinetic–pharmacodynamic (PKPD) model describing ceftazidime/avibactam efficacy against β-
10.1089/mdr.2021.0228 lactamaseproducing Gram-negative bacteria. J Antimicrob Chemother. (2020) 75:400– 8. doi: 10.1093/jac/dkz440
87. Blondeau JM, Fitch SD. 98. Thorsted A, Tano E, Kaivonen K, Sjölin J, Friberg LE, Nielsen EI. Extension of pharmacokinetic/pharmacodynamic
Mutant prevention and (PK/PD) time-kill studies to include LPS/endotoxin release from E. coli exposed to cefuroxime. Antimicrob Agents
minimum Chemother. (2020) 64:e02070–19. doi: 10.1128/AAC.02070-19
inhibitoryconcentration drug 99. Ferro B E, van Ingen J, Wattenberg M, van Soolingen D, Mouton JW. Time–kill kinetics of antibiotics active against
values for enrofloxacin, rapidly growing mycobacteria. J Antimicrob Chemother. (2015) 70:811–7. doi: 10.1093/jac/dku431
ceftiofur, florfenicol, 100. Zhang N, Gu X, Ye X, Wu X, Zhang B, Zhang L, et al. The PK/PD interactionsof doxycycline against Mycoplasma
tilmicosin and tulathromycin gallisepticum. Front Microbiol. (2016) 7:653. doi: 10.3389/fmicb.2016.00653
tested against swine 101. Zhang L, Xie H, Wang H, Ding H, Zhang G, Hu J. Kill rate and evaluation of ex vivo PK/PD integration of cefquinome
pathogens Actinobacillus against Actinobacillus pleuropneumoniae. Front Vet Sci. (2021)
pleuropneumoniae, 8:751957. doi: 10.3389/fvets.2021.751957
Pasteurella multocida and 102. Maneke E, Pridmore A, Goby L, Lang I. Kill rate of mastitis pathogens by a combination of cefalexin and kanamycin. J
Streptococcus suis. PLoS Appl Microbiol. (2011) 110:184–90. doi: 10.1111/j.1365-2672.2010. 04876.x
ONE. (2019) 14:e0210154.
doi:
Conflict of Interest: The authors declare that the research was conducted in the absence of any commercial or financial
10.1371/journal.pone.02101
relationships that could be construed as a potential conflict of interest.
54
88. Nolting A, Costa TD, Rand
KH, Derendorf, H. Publisher’s Note: All claims expressed in this article are solely those of the authors and do not necessarily represent those
Pharmacokineticpharmacody of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated
namic modeling of the in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
antibiotic effect of
piperacillin in vitro. Pharm Copyright © 2022 Zhang, Xie, Wang, Wang, Hu and Zhang. This is an open-access article distributed under the terms of the
Res. (1996) 13:91–96. doi: Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided
10.1023/A:1016085402278 the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in
89. Regoes RR, Wiuff C, Zappala accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with
RM, Garner KN, Baquero F, these terms.
Levin BR.Pharmacodynamic
functions: a multiparameter
approach to the design of
antibiotic treatment
regimens. Antimicrob Agents
Chemother. (2004)
48:3670–76. doi:
10.1128/AAC.48.10.3670-
3676.2004
90. Foerster S, Unemo M,
Hathaway LJ, Low N, Althaus
CL. Time-killcurve analysis
and pharmacodynamic
modelling for in vitro
evaluation of antimicrobials
against Neisseria
gonorrhoeae. BMC
Microbiol. (2016) 16:1– 11.
doi: 10.1186/s12866-016-
0838-9
91. Chauzy A, Ih H, Jacobs M,
Marchand S, Grégoire N,
Couet W, et al. Sequential
time-kill: a simple

Frontiers in Veterinary Science | www.frontiersin.org 10 March 2022 | Volume 10 | Article 860472