Supplement Article

Optimizing Antibiotic Drug Therapy The Journal of Clinical Pharmacology

2018, 58(S10) S108–S122
in Pediatrics: Current State and Future 
C 2018, The American College of

Clinical Pharmacology
DOI: 10.1002/jcph.1128
Needs

Jennifer Le, PharmD, MAS, BCPS-ID, FIDSA, FCCP, FCSHP1

and John S. Bradley, MD, FAAP, FIDSA, FPIDS2,3

Abstract
The selection of the right antibiotic and right dose necessitates clinicians understand the contribution of pharmacokinetic variability stemming from
age-related physiologic maturation and the pharmacodynamics to optimize drug exposure for clinical response. The complexity of selecting the right
dose arises from the multiplicity of pediatric age groups, from premature neonates to adolescents. Body size and age (which relate to organ function)
must be incorporated to optimize antibiotic dosing in this vulnerable population. In the effort to optimize and individualize drug dosing regimens, clinical
pharmacometrics that incorporate population-based pharmacokinetic modeling, Bayesian estimation, and Monte Carlo simulations are utilized as a
quantitative approach to understanding and predicting the pharmacology and clinical and microbiologic efficacy of antibiotics. In addition, opportunistic
study designs and alternative blood sampling strategies can serve as practical approaches to ensure successful conduct of pediatric studies. This review
article examines relevant literature on optimization of antibiotic pharmacotherapy in pediatric populations published within the last decade. Specific
pediatric antibiotic data, including beta-lactam antibiotics, aminoglycosides, and vancomycin, are critically evaluated.

Keywords
Pharmacokinetic, pharmacodynamic, pediatrics, neonates, antimicrobials, vancomycin, Monte Carlo simulation, beta-lactams, vancomycin, aminoglyco-
sides, antibiotics, Bayesian methods, opportunistic study, ceftaroline

Effective pharmacotherapy in pediatric patients with
infection requires the integration of pertinent data
on the antibiotic, bacterial pathogen, and patient. In
addition to appropriate antibiotic selection based on
the susceptibility of the pathogen and the site of
infection, optimizing antibiotic therapy to achieve the
desired outcome in pediatric patients necessitates the
understanding of growth and development related to
age for both physiologic and immunologic processes
that govern drug absorption, distribution, metabolism,
and elimination, as well as maturation of granulocyte,
lymphocyte, and humoral immune function. Physi-
ologic development may alter the pharmacokinetics
(PK) and impact the pharmacodynamics (PD) of drugs.
In selecting the right antibiotic and dose, clinicians
must understand the contribution of PK variability
stemming from age-related physiologic maturation and
the PD to optimize antibiotic exposure for clinical
response.1 While the immature nature of the immune
system in neonates and infants may potentially affect
independent of year of publication, were selected to
highlight important attributes of pediatric pharmacol-
ogy. Recent articles with new information related to
pediatric pharmacology and specific antibiotics, par-
ticularly focusing on literature within the past decade,
were critically reviewed and are presented in this review.
Pharmacokinetics: Alterations Stemming
From Growth and Development
The two key factors stemming from physiologic mat-
uration that contribute to PK alterations in pediatric
patients are body size and organ function (which is
defined by age). The complexity of PK variability arises
1 SkaggsSchool of Pharmacy and Pharmaceutical Sciences, University of
California at San Diego, La Jolla, CA, USA
2 Department of Pediatrics, Division of Infectious Diseases, University of
California at San Diego, La Jolla, CA, USA
3 Rady Children’s Hospital San Diego, San Diego, CA, USA

the PD of antibiotics, this article focuses on the effect of
PKPD alterations stemming from physiologic changes
that occur with age on antibiotic therapy. A PubMed
search for publications up to November 2017was con-
ducted using key terms relevant to pediatric PKPD and
specific antibiotics reviewed in this article. Key articles,
Submitted for publication 29 November 2017; accepted 1 March 2018.
Corresponding Author:
Jennifer Le, PharmD, University of California San Diego, Skaggs School of
Pharmacy and Pharmaceutical Sciences, 9500 Gilman Drive, MC 0714, La
Jolla, CA 92093-0714
Email: jenle@ucsd.edu
Le and Bradley S109

from the multiple pediatric age groups, from premature Table 1. Maturational Physiologic Changes Affecting Drug Pharmacoki-
neonates to adolescents. “Pediatric patients” denotes all netics in Pediatric Patients13
pediatric age groups from birth to 18 years, including Age Group
premature neonate (born before 37 weeks’ gestation), Reaching Adult
neonates (within first month of life), infants (1 to 12 Variable Neonatea Valueb
months), children (1 to 12 years), and adolescents (13 Absorption
to 18 years). The neonatal and infant periods com- Gastric pH ↑ Child
prise the most dramatic maturational developments Gastric and intestinal emptying ↑ Child
that affect a drug’s PK. While age-related physiologic time
Biliary function, pancreatic ↓ Infant
maturation continues into childhood, the most promi-
function, and gut microbial
nent differences, compared with adults, are evident in colonization
neonates and infants. Table 1 describes the maturational Intramuscular absorption ↓ Child
processes that affect a drug’s absorption, distribution, Skin permeability and ↑ Child
metabolism, and renal elimination in pediatric patients percutaneous absorption
Distribution
compared with adults. The age group in which the
Total body water and ↑ Late childhood
maturational process reaches adult values are also extracellular water
provided. Total body fat and muscle ↓ Child
mass
Total plasma binding proteins ↓ Early childhood
Hepatic distribution and elimination
Body Size CYP1A2 ↓ Early childhood
Drug dosing in pediatric patients should be scaled to CYP2C9 ↓ (30% of adult Infant, with peak
body size, expressed as milligram per kilogram. Total activity) activity in
body weight is a common and appropriate measure to childhood
use for drug dosing in normal-sized children. However, CYP2C19 ↓ (30% of adult Infant, with peak
activity) activity in
pediatric drug doses cannot be normalized directly childhood
from an adult dose using total body weight (ie, adult CYP2D6 ↓ (30% of adult Infant
dose in milligram per kilogram) when pediatric dosing activity)
data are unavailable. This is because drug elimination, CYP2E1 No data, with ↓ Late childhood
particularly via the renal route, is nonlinear to weight.2 in infant
CYP3A4 ↓ (30%-40% of ↑ in early
Two contrasting consequences arise from this nonlinear adult activity) childhood, with
relationship: the underestimation of drug clearance progressive decline
(CL) in children and its overestimation in adults.2 CYP3A7103 ↑ Low to
While total body weight is an inadequate body size undetectable in
descriptor to apply in pediatric drug dosing that is early childhood
Uridine 5-diphosphate ↓ Early childhood
derived from adult data, a child’s body size can be glucuronyltransferase
referenced to a 70-kg adult using allometric scaling— N-acetyl transferase 2 ↓ Early childhood
a coefficient of 1 for volume distribution (Vd; weight1 ) Methyltransferase ↑ Infant
and 0.75 for CL (weight0.75 ).3 This allometric approach Sulfotransferase ↓ Adolescent, with ↑
to the prediction of drug CL has been applied in in infant and child
Renal elimination
children older than 5 years, including children with Glomerular filtration ↓ Infant
cystic fibrosis (CF) and, possibly, obesity (albeit other Tubular secretion and ↓ Infant
factors play a role in children who are obese).4–6 reabsorption
The exponent of 0.75 has been used to allometrically a
↓ = decreased, ↑ = increased activity compared with adults.
scale weight to describe the maturation of glomerular b
Pediatric age groups include premature neonate (born before 37 weeks
filtration rate (GFR) in pediatric patients, including gestation), neonates (within first month of life), infants (1 to 12 months),
neonates.7 The 0.75 allometric scaling was superior to children (1 to 12 years), and adolescents (13 to 18 years).
body surface area (which uses an exponent of 0.67)
in measuring GFR, liver and kidney volumes, and
cardiac output.2 However, for neonates and infants, this Age
exponent may not be applicable for predicting the CL Similar to weight, age is another significant factor to
of some drugs, especially those that have elimination integrate when optimizing drug dosing in neonates,
outside of glomerular filtration.8 Ideally, to optimize infants, and young children because it correlates well
age-specific dosing across the pediatric age spectrum, with organ function responsible for the maturation of
the exact value used for allometric scaling should be CL, both renal and hepatic. In the healthy pediatric
derived for each individual drug through PK studies. population, age defines organ function. The presence
S110
of diseases, such as CF, may alter organ function.
Relative to body mass, the size and weight of organs,
specifically the kidneys and liver, which are responsible
for drug CL, are several-fold greater in younger
children compared with adults. Clearance accounts
for both metabolism and elimination of a drug from
the body and largely determines the maintenance
dose. Renal elimination, including reduced GFR,
tubular secretion, and reabsorption, is immature in
neonates and infants. Because of this, the CL of
antibiotics that undergo elimination primarily via
GFR (including certain beta-lactams, aminoglycosides,
and vancomycin) are decreased in neonates and infants
relative to older pediatric age groups. This may lead to
increased drug exposure in neonates and infants and
potentially increases the risk for toxicity, particularly
nephrotoxicity for aminoglycosides, if the dosing is not
adjusted.9
Within the first 2 weeks of life, a rapid surge in renal
function is observed that is due to an increase in renal
blood flow.10,11 This marks the most prominent change
within the first year of life and occurs nonlinearly with
age during this interval. The nonlinear relationship be-
tween age and renal function follows an evident plateau
of drug dosing, expressed as milligram per kilogram.
Consequently, increasing drug dosage is necessary up
to a certain age, when dosage then decreases to that
appropriate for adults.
An in-depth discussion of age (usually referred as
postnatal age or chronologic age) is crucial with respect
to neonates because nephrogenesis, which is respon-
sible for the increase in the GFR in utero, requires
completion of 36 weeks’ gestation.10,11 By definition,
gestational (or menstrual) age is the time between the
first day of the last normal menstrual period and
birth; postnatal age accounts for the duration of life
after birth; and postmenstrual age incorporates both
gestational and postnatal age. For premature neonates
born before 36 weeks’ gestation, a drug’s PK may
depend on more than postnatal (or chronologic) age
because nephrogenesis has not occurred. The use of
postnatal age compared with postmenstrual age as an
indicator for the maturation of drug elimination may,
therefore, be suboptimal for premature neonates.3 In
fact, postmenstrual age was more useful than post-
natal age in describing GFR in premature and full-
term neonates.7 In a large pediatric study of pooled
data from 923 participants ranging from 22 weeks’
postmenstrual age to 31 years, allometric scaling was
applied to predict and compare GFR between pediatric
participants and adults. Interestingly, half of the adult
GFR (ie, approximately 60 mL/min per 70 kg) was
reached at 48 weeks of postmenstrual age and 90% of
the adult GFR, at 1 year of age.7 Age becomes less
influential in late childhood and adolescence.
The Journal of Clinical Pharmacology / Vol 58 No S10 2018
Maturation of hepatic pathways that affect drugs
undergoing metabolic transformation or elimination
occurs at diverse rates at different ages, ranging
from the first months to years of life. The body’s
primary organ for drug metabolism, the liver, produces
microsomal cytochrome P450 enzymes at birth.
The activity of most of these enzymes is reduced
during infancy, with maturation occurring from
infancy into late childhood, depending on the specific
enzyme (Table 1).2 Consequently, the metabolic CL,
particularly the glucuronidation reaction, is lower in
neonates and infants than in adults. Two exceptions
to the underdeveloped liver function in neonates and
infants are methyltransferase and sulfotransferase,
respectively, for which enhanced activity is observed.
The different rates of hepatic maturation, along with
the complexity of the hepatic CL of drugs that may
use different mechanisms at varying degrees, precludes
accurate prediction of drug dosages. Enzymatic matu-
ration has been described using the sigmoidal Hill equa-
tion and coefficients that incorporate postmenstrual
age and the maturation half-time (TM50 ).2 In neonates
and infants, calculations based on postmenstrual age
are more accurate than calculations based on other
types of age because it accounts for maturation that
occurs during gestation.
In addition to renal and hepatic functions that
contribute to a drug’s metabolism and elimination (ie,
CL) and therefore dosing, age also influences drug
distribution through body water or fat composition and
plasma protein binding. Body composition and protein
binding change with age. Drug distribution, which is
represented by Vd, equates to the total amount of
drug distributed within the blood and tissues. Drugs
that remain in the circulation and distribute mainly
in the extracellular fluid compartment (eg, hydrophilic
drugs, including aminoglycosides and linezolid) usually
possess a relatively low Vd. In contrast, drugs that
distribute intracellularly or well in tissues generally have
a relatively high Vd. Because physiologic maturation
that occurs with age alters body composition and
plasma protein binding, the Vd of drugs in pediatric
patients changes accordingly (Table 1). For example,
the augmented total body and extracellular water con-
tent evident in neonates and young infants translates
to enhanced Vd of aminoglycosides and linezolid and,
consequently, the need for increased loading doses
(milligram per kilogram) to achieve adequate drug
exposure (ie, usually by peak concentration).9,12 The
reverse is observed for highly lipophilic drugs such as
clindamycin.13
Another important factor that affects drug distri-
bution is plasma protein binding. The decrease in
both the concentration and affinity of plasma binding
proteins such as albumin and alpha-1-acid glycoprotein
Le and Bradley
in neonates, infants, and young children alters the distri-
bution of drugs, resulting in an increase in the unbound
moiety (which is the free, pharmacologically active
drug) (Table 1). For example, vancomycin demonstrates
partial binding to albumin, with a reported range of
30% to 60% in adults. Compared with adults, the de-
crease in albumin binding evident in pediatric patients
results in a significant increase in the unbound van-
comycin fraction, to approximately 80%.14 In addition,
a recent study showed that the Vd of clindamycin
in children who are non-obese and children who are
obese was proportional to concentrations of alpha-1-
acid glycoprotein and albumin because it binds highly
to both proteins.15
Plasma membrane transporters mediate the absorp-
tion, distribution, and excretion of drugs. The activities
of these transporters are affected by age. Data on
the ontogeny of these transporters and their clinical
relevance are limited. The two major transporters are
adenosine triphosphate (ATP)-binding cassette (ABC)
and the solute carrier (SLC) transporters. In general,
the expression of drug transporters appears low during
the fetal and neonatal periods, but augments from 7
years of age onward.16 Data on age-related maturation
from infancy to early childhood are currently limited.
One of the most studied transporters is ABCB1,
which supports renal CL of daptomycin. Interestingly,
the development of ABCB1 is also dependent on the
organ type, with stable expression in the intestines of
adults and pediatric patients, low hepatic expression in
neonates and infants until 12 months of age, and low
renal expression until 3 to 6 months of age.16,17 The
renal CL of daptomycin is higher in neonates and in-
fants than in adults due to the enhancement of ABCB1
expression that mediates renal tubular secretion despite
their immature GFR.16
Population-Based Pharmacokinetic
Modeling and Bayesian Estimation
While extrapolating pharmacologic data from adults
may produce meaningful predictions in older chil-
dren and adolescents, age-related maturational changes
in neonates, infants, young children, and individuals
with pediatric-predominant diseases (including CF and
Kawasaki disease) require PKPD studies be conducted
directly in these pediatric groups to serve the unmet
needs.18 In the effort to optimize and individualize
drug dosing regimens, pharmacometrics is utilized as
a quantitative approach to understand and predict the
PK and PD of a drug. Pharmacometric modeling,
specifically population-based PK (PopPK), allows us to
integrate different factors (or covariates) to ultimately
characterize a population average, with intersubject and
intrasubject variabilities of PK parameters, including
S111
Vd and CL, that determine drug dosing.19 The incorpo-
ration of covariates, particularly weight and age, as ex-
plained above, is integral to optimizing the use of drugs
in pediatric patients because this approach can address
the wide age spectrum across the diverse pediatric age
groups.20 The use and acceptance of PopPK (which
integrates nonlinear mixed-effects modeling) continues
to improve our understanding of PKPD in children.21
Furthermore, PopPK can be used to individualize drug
dosing directly in the patient care setting to improve
the precise achievement of a therapeutic goal that min-
imizes drug toxicity while maximizing clinical benefit.22
Bayesian estimation can be incorporated into
PopPK modeling to enhance prediction of PK param-
eters and measures of drug exposure, including drug
concentrations and area-under-time-concentration
curves.23 The use of the Bayesian approach facilitates
derivation of more accurate predictions for PK, as well
as the PD of drugs.24 The Bayesian approach assumes
that covariates of interest are random and introduces
uncertainty, which can be better understood with prior
knowledge of drug behavior in that population. The
“Bayesian prior” is the construction of the probable
PK parameters for a patient, obtained from prior PK
studies in the relevant age-specific population, that can
be applied to that patient. Combining this Bayesian
prior with the patient’s measured drug concentrations,
the estimation and prediction of PopPK parameters
are greatly enhanced. Population-based PK modeling
with Bayesian estimation is accepted by the US Food
and Drug Administration as an approach to drug
evaluation in pediatric patients.25
Pharmacodynamics
Evaluation of the PD allows us to understand the rela-
tionship between drug exposure at the site of infection
and the onset, intensity, and duration of pharmacologic
effect on the pathogen.26 Along with the PK, the
PD property of an antimicrobial agent can be used
to maximize clinical response and prevent toxicities
related to therapy. To treat or cure an infection, the
antibiotic exposure should be maximized to inhibit (or
eradicate) the pathogen and suppress the development
of resistance.27 The PD exposure target must integrate
a feature for pathogen susceptibility known as the min-
imum inhibitory concentration, or MIC.28 In addition
to the MIC, the drug concentration of interest, which
should reflect the active unbound moiety, can be the
peak, trough, or average concentrations or exposure
over a period of time (AUC over 24 hours).
The exposure-response relationship for an antimi-
crobial agent is more predictable in adults than in
pediatric patients. A standard antibiotic dose will gen-
erally provide a baseline exposure to elicit a predictable
S112
clinical response in adults. However, one standard dose
cannot be applied in pediatric patients because sub-
stantial variability exists in exposure to an antibiotic
across the pediatric age spectrum. Once the age-specific
exposure is known in pediatric patients to achieve an
equivalent exposure to that demonstrated in adults,
the exposure that yields the desired microbiologic and
clinical cure in adults is then assumed to be similar
in that age group. Notably, the age-specific exposure
in pediatric patients, particularly in neonates due to
their immature immune function, may differ from the
exposure needed to achieve microbiologic and clin-
ical cure in adults; in other words, the drug expo-
sure required in the immune-compromised neonate for
cure may be greater than that associated with cure in
adults. Currently, no prospective data exist in neonates
to validate the exposures required for microbiologic
and clinical cure. In addition, underlying genetic and
clinical factors that may affect response may also be
explored to optimize therapy in both pediatric patients
and adults.
An important component of PD for successful treat-
ment of an infection is antibiotic penetration. The
degree of antibiotic penetration must be accounted for
to ensure adequate drug exposure at the site of infection
to maximize therapeutic response. The MIC provided
by clinical microbiology depicts susceptibility of a
pathogen based on achievable serum concentration. In
the scenario where a pathogen is reported resistant
to a certain antibiotic, that same antibiotic may still
successfully treat the infection if the resistance (assessed
by the actual MIC) is not profound and complete and
adequate drug exposure can be achieved (eg, treating a
urinary tract infection using a drug that highly concen-
trates in the kidneys). In contrast, even if an antibiotic
is reported as susceptible, therapeutic response may
not be achieved if drug exposure is inadequate in
specific tissue sites, particularly for serious infections,
including meningitis and pneumonia, or undrained
abscesses.
Appropriate management of central nervous system
(CNS) infections requires that antibiotics penetrate the
blood–brain barrier to attain concentrations at the site
of infection for eradication of the infecting pathogen.
Independent of age, the physicochemical property of
a drug is the main factor that determines drug distri-
bution into the CNS. In fact, small-sized and highly
lipophilic drugs with low protein binding penetrate the
blood–brain barrier well. The distribution pathways
that are affected by growth and development, and the
presence of meningeal inflammation, are also factors
that affect overall CNS drug penetration. Antibiotics
with good CNS penetration are provided in Table 2. 29
Penetration of these antibiotics is enhanced when the
meninges are inflamed.
The Journal of Clinical Pharmacology / Vol 58 No S10 2018
Table 2. Antibiotics With Predictably Good Penetration Into the
Central Nervous System
Antibiotic
Penicillins, especially aqueous formulation
Aminopenicillins, particularly ampicillin and parenteral (not oral)
amoxicillina
First- and second-generation cephalosporins (excluding cefuroxime)
Third- and fourth-generation cephalosporins (ceftriaxone, cefotaxime,
ceftazidime, cefixime, and cefepime)b
Carbapenems, including meropenem and imipenemb
Aztreonam
Chloramphenicol
Sulfamethoxazole/trimethoprim
Fosfomycin
Metronidazole
Rifampin
Antituberculosis agents (ie, isoniazid and pyrazinamide)b
Daptomycin
Fluoroquinolones
Tetracyclines
a
Nafcillin and piperacillin have been studied in pediatric patients but penetra-
tion is erratic.
b
Studies have documented penetration specifically in children.
For pneumonia caused by extracellular pathogens,
epithelial lining fluid (ELF) has been used as the target
site to determine adequacy of drug exposure. Currently,
the literature on ELF-to-plasma ratios is limited to a
small number of antibiotics, all of which were evaluated
in adults, and comprises beta-lactams (ranging from
0.21 to 1.04, depending on the specific beta-lactam),
vancomycin (0.18 to 0.50), aminoglycosides (highly
variable hysteresis depending on sampling time and
arising from the hydrophilic property of these antibi-
otics), and fluoroquinolones (>1.0, which represents
extensive lung penetration).30 The presence of hystere-
sis on a concentration-effect plot implies a delay in time
between measured concentration and response, as well
as an associated lower peak concentration. Notably,
these studies of ELF do not account for redistribution
of drugs from the ELF back to plasma because only
specific time points after dose administration (rather
than over the entire dosing interval) were evaluated.
Nonetheless, for all of these distinct infection types,
adequate antibiotic exposure implies the need to use
free drug concentrations in PKPD studies to account
for antibiotic penetration at the site of infection.
Monte Carlo Simulation
To identify the best dosing regimen that achieves
microbiologic and clinical cure (and even resistance
suppression), a PD exposure target or index should be
explicitly defined and incorporated (along with age-
specific PK data) to conduct Monte Carlo simulations.
The PD indices that vary with antibiotic class must
integrate pathogen susceptibility and antibiotic
Le and Bradley
exposure, depicted by free drug peak concentration over
MIC (Cmax /MIC, associated with aminoglycosides),
AUC24 /MIC associated with fluoroquinolones, or
the time that free drug concentrations remain above
the MIC (fT>MIC), associated with beta-lactams.
Currently, the PD exposure targets for antibiotics have
originated primarily from preclinical in vitro models,
with very limited validation in clinical studies.27 Fur-
thermore, these limited human clinical investigations
were conducted retrospectively in adult patients. As
such, there is an evident demand for well-designed
studies in pediatric participants, preferably conducted
prospectively, to validate PD targets of antimicrobial
agents that were derived from preclinical and adult
data. Despite improvement in the number of pediatric
PK studies in recent decades, the need for PD studies in
pediatric participants should be underscored to ensure
the optimization of antibiotic dosing required for the
cure of different infections over the entire pediatric
age spectrum. In addition, protein binding, drug
penetration to the site of infection, and immunologic
competence that changes with age are important
factors that should be evaluated in PD studies.
Clinical Pharmacology Optimizing
Bedside Care of Pediatric Infections
Safe and effective antibiotic use in all pediatric age
groups, including neonates, requires thorough under-
standing and application of the quantitative clinical
pharmacology that encompasses PKPD.20 Pharmacol-
ogists, pharmacists, and physicians must collectively be
equipped to support and integrate PopPK modeling,
Bayesian estimation, and simulation methods in the
clinical decision-making process that is integral to ther-
apeutic drug monitoring and management to optimize
exposure-response and clinical outcome relationships
across the pediatric age spectrum.18
To demonstrate the application of clinical pharma-
cology to the optimization of pharmacotherapy and
its positive impact on the care of pediatric patients
with infections, specific antibacterial agents (including
beta-lactams, aminoglycosides, and vancomycin) are
presented in the subsequent sections. The selection
of these antibiotics was based on the availability of
studies in pediatric patients and on their PD prop-
erties. Antibiotics with unique PD features (including
concentration-dependent, time-dependent, and post-
antibiotic effects) and, more importantly, antibiotics
with a sufficient number of PKPD evaluations in pe-
diatric patients are presented. Table 3 describes the PD
properties of antibiotics, including those with limited
PKPD studies in pediatric patients. Antifungal agents
were excluded from this article, but two reviews have
been published.31,32
S113
Beta-lactams
Because of their long-standing safety profile and their
broad spectrum of antibacterial activity, beta-lactam
antibiotics present the most extensively studied class
evaluated for their PKPD properties in pediatric pa-
tients. Because this class of antibiotics exhibits time-
dependent bactericidal but no post-antibiotic effects,
the use of beta-lactams can be optimized by increasing
the infusion time or the dosing frequency, particularly
to treat pathogens with higher MICs (Table 3). These
two approaches align with maximal use of the time-
dependent PD killing of beta-lactams. Nonetheless,
extending the infusion time (which includes prolonged
and continuous infusions) has been evaluated more
frequently than has increasing dosing frequency (ie,
from every 6 to every 4 hours) for its feasible application
in the patient care setting. This is particularly valuable
in the care of pediatric patients with certain medi-
cal conditions (eg, prematurity, CF, or hematologic/
oncologic disorders) that necessitate the use of multiple
concurrent medications.
To determine the probability of target attainment
(PTA) against Pseudomonas aeruginosa with varying
susceptibilities, Monte Carlo analyses in children
between 1 and 12 years of age were conducted
using different regimens of cefepime, ceftazidime,
imipenem/cilastatin, meropenem, and piperacillin/
tazobactam.33–35 The PD target to achieve bactericidal
activity used in these studies was fT>MIC 40%
for carbapenems and 50% for penicillins and
cephalosporins. For bacteriostatic effect, PD targets
20% fT>MIC for carbapenems and 30% fT>MIC
for penicillins and cephalosporins were used. Collec-
tively from these studies, prolongation of infusion from
30 minutes to 3 hours and continuous infusion signifi-
cantly improved the PTA to >90% to achieve bacterici-
dal activity for all antibiotics that were evaluated.33–35
This is a clinically relevant finding that can be applied
to critically ill children to achieve optimal PTA against
moderately susceptible strains of P aeruginosa.
For less susceptible pathogens, increasing the dose
alone, without optimizing the PD property of fT>MIC
by increasing infusion time, may not always achieve
adequate drug exposure. For example, at one institution
with elevated MICs among the P aeruginosa isolates,
increasing the meropenem dose from 20 mg/kg to 40
mg/kg, administered every 8 hours using 30-minute
infusions, resulted in a low, inadequate PTA of 58%.35
This illustrates that increasing the dose alone without
extending the infusion time of beta-lactams may not be
sufficient to adequately treat serious antibiotic-resistant
infections.
Emerging interest in defining age-related devel-
opmental exposures has prompted population-based
PKPD evaluations in preterm neonates.36–41 To
S114 The Journal of Clinical Pharmacology / Vol 58 No S10 2018

Table 3. Pharmacodynamic Properties of Antimicrobial Classes in Pediatric Patients104

Antibiotic Class Mechanism Index Dose Optimization

Aminoglycosides
β-lactams
Amphenicols (chloramphenicol)
Fluoroquinolones and quinolones
Lincosamides (clindamycin)
Macrolides and azalides
Nitroimidazoles (metronidazole)
Nitrofurantoin
Oxazolidinones (linezolid)
Rifamycins
Sulfonamides and trimethoprim
Tetracyclines and glycylcyclines
(tigecycline)
Glycopeptides
Concentration-dependent, with
moderate post-antibiotic effect
Time-dependent, with no to minimal
post-antibiotic effect
Time-dependent, with some
post-antibiotic effect against
Gram-positive bacteria
Concentration-dependent, with
moderate post-antibiotic effect
Time-dependent, with moderate to
prolonged post-antibiotic effect
Time-dependent, with moderate to
prolonged post-antibiotic effect
Concentration-dependent, with
moderate post-antibiotic effect
Time-dependent for E coli;
concentration-dependent for S
saprophyticus and E faecium
Time-dependent, with moderate to
prolonged post-antibiotic effect
Concentration-dependent, with
moderate post-antibiotic effect at ³ 16
× MIC
Concentration-dependent, with
moderate post-antibiotic effect
Time-dependent, with moderate to
prolonged post-antibiotic effect
Time-dependent, with moderate to
prolonged post-antibiotic effect
Cmax /MIC, AUC24 /MIC Increase dose to maximize Cmax or
AUC24
fT>MIC Increase duration of infusion or
frequency of administration
AUC24 /MIC Increase dose, frequency of
administration, or duration of infusion
Cmax /MIC, AUC24 /MIC Increase dose to maximize Cmax or
AUC24
AUC24 /MIC Increase dose, frequency of
administration, or duration of infusion
AUC24 /MIC Increase dose, frequency of
administration, or duration of infusion
AUC24 /MIC Increase dose to maximize Cmax or
AUC24
fT>MIC, AUC24 /MIC Increase dose, frequency of
administration, or duration of infusion
AUC24 /MIC, fT>MIC Increase dose, frequency of
administration, or duration of infusion
Cmax /MIC, AUC24 /MIC Increase dose to maximize Cmax or
AUC24
Cmax /MIC, AUC24 /MIC Increase dose to maximize Cmax or
AUC24
AUC24 /MIC Increase dose, frequency of
administration, or duration of infusion
AUC24 /MIC Increase dose, frequency of
administration, or duration of infusion

Abbreviations: AUC24, area under the concentration–time curve over a 24-hour period; Cmax , maximum serum concentration; MIC, minimum inhibitory
concentration; fT>MIC, fraction or percentage of the dosing interval above the MIC.

adequately design dosing strategies that reflect
maturational development, including nephrogenesis,
both gestational and postnatal ages (resulting in
postmenstrual age) were used in these studies. To
achieve antibiotic microbiologic and clinical efficacy
and account for deficiencies in the immune response
system expected in neonates, higher PD targets of
>60% fT>MIC for meropenem and >75% fT>MIC
for ampicillin and piperacillin were employed in these
studies.
Meropenem was evaluated in preterm and full-
term neonates <2 months old with gestational ages
of 23 to 42 weeks.36,41 Important covariates for CL
were postmenstrual age, weight, and serum creatinine.
Meropenem CL increased in older neonates, defined
by gestational age and chronologic age. Meropenem
20 mg/kg every 8 hours, infused over 30 min-
utes, achieved >90% PTA for its bactericidal effect
against nosocomial Gram-negative pathogens.36 How-
ever, meropenem 40 mg/kg every 8 hours infused over
4 hours was recommended for less susceptible bac-
teria, including P aeruginosa, with MIC of 4 to 8
mcg/mL.36,41
The PKPD of meropenem was evaluated for the
treatment of meningitis and intra-abdominal infec-
tions in neonates.39 For meningitis, meropenem con-
centrations in the cerebrospinal fluid (CSF) that were
measured at various time periods varied from 4.1 to
34.6 µg/mL, which represent an average of 70% (range
of 5% to 148%) penetration. For suspected or compli-
cated intra-abdominal infections, meropenem 20 to 30
mg/kg, depending on gestational age and postnatal age,
was administered every 8 or 12 hours. Most (85% of
200) of these infants experienced clinical therapeutic
success and tolerated meropenem well.38,39
Similar to meropenem, piperacillin/tazobactam was
also studied in premature and term infants 90 days
with suspected or complicated intra-abdominal infec-
tions. The CL of piperacillin improved with increasing
gestational age at birth, but decreased by 60% when
serum creatinine was 1.2 mg/dL.42 Piperacillin 80
to 100 mg/kg infused every 8 hours did not meet
PD efficacy target for P aeruginosa in 70% of in-
fants, implying that further studies of higher doses
or more frequent or more prolonged infusions are
warranted.42
Le and Bradley
A phase 1, single-dose study of doripenem was
conducted in 52 infants up to 12 weeks of postnatal
age.43 Displaying linear PK, neonates <4 weeks had
decreased CL and increased mean AUC exposure. The
PK of doripenem infused at 5 to 8 mg/kg over 1 hour
in these term and preterm infants was similar to those
previously reported for other carbapenems in neonates
and young infants.
Similar to premature infants, children with CF have
altered PK that may necessitate different dosing strate-
gies. Increased CL (eg, aztreonam 100 mL/min in chil-
dren with CF versus 76 mL/min in healthy participants,
P < .01), without a significant change in Vd, has been
reported.4,44 To prevent subtherapeutic dosing that may
result from the increased CL, the PD property of an-
tibiotics should be maximized to optimize dosing in this
population. One study showed significant improvement
in achieving the PTA of 65% fT>MIC with the
use of 4- to 5-hour prolonged or continuous infusions
of ceftazidime 6 grams per day (weight-adjusted for
70 kg). Extended or continuous infusions increased the
PTA for pathogens with elevated MIC, even as high
as 8 to 12 μg/mL.4 This is invaluable in patients with
CF because they usually have a history of previous
exposure to antibiotics for both acute and chronic
respiratory infections.45,46 Likely due to their repeated
antibiotic exposures, reported antibiotic resistance rates
are higher in children with CF.
Ceftaroline is the first beta-lactam antibiotic with
activity against methicillin-resistant Staphylococcus au-
reus (MRSA) and, as such, may play a significant role in
patients with CF. A case report of a ceftaroline-resistant
MRSA strain from a girl with CF was reported after
22 ceftaroline treatment courses using standard adult
dosing that achieved suboptimal antibiotic exposure,
possibly arising from increased CL and Vd in this
patient. This case report prompted a study evaluating
the single-dose PK and safety of ceftaroline in children
with CF for staphylococcal pneumonia. In this phase
1, open-label, single-dose, prospective study using
10 mg/kg (up to 600 mg) in 20 participants between 6
and 18 years of age, PopPK modeling showed that chil-
dren with CF have increased ceftaroline CL compared
with published data from children without CF. As such,
higher dosages are likely to be required in children with
CF to achieve adequate exposure for cure and possibly
delay the emergence of resistance.47
In addition to neonates and children with CF, criti-
cally ill children may have experienced alteration in PK
parameters necessitating changes in dosing strategies.
Increased Vd of certain antibiotics, particularly those
with high water-solubility, has been reported in criti-
cally ill children. For example, the Vd of piperacillin
is 0.51 L/kg in critically ill young children versus
0.28 L/kg in young children who are not critically
S115
ill.33,48 This increased Vd may potentially result from
underlying disease states affecting the interstitial fluid,
where increased extracellular fluids observed in third-
fluid spacing increases the Vd.
The optimal PD target for immune-compromised
neonates and children is not well-defined. However,
one investigator has proposed an aggressive and un-
validated PD target to combat serious P aeruginosa
infections in patients with immune-compromised sta-
tus and maintain free drug concentrations at 100%
fT >6 times the MIC.49,50 Studies in adults have
evaluated a PD target of 100% fT>MIC to im-
prove clinical cure and microbiologic eradication
using cefepime, piperacillin/tazobactam, meropenem,
and ceftobiprole.51,52 Using this PD target in ado-
lescents with febrile neutropenia, a loading dose
of 100 mg/kg followed by continuous infusion of
piperacillin/tazobactam at 350 to 400 mg/kg/day
achieved >99% PTA at an MIC of 4 mg/L.49 The
standard dose of 300 mg/kg/day achieved 85% PTA,
which the authors considered inappropriate for pa-
tients with recent intensive chemotherapy, predicted
prolonged and profound neutropenia, or fever >39°C
because such patients are at increased risk for serious
infections.
In conclusion, studies in the different pediatric age
groups serve as evidence for the use of various dosing
strategies for beta-lactams to enhance their PD target
attainment. Strategies for the optimization of beta-
lactam use consist of frequent dosing and extended
or continuous infusions that may improve PTA in
certain pediatric populations, including individuals
who are critically ill, immunocompromised (including
premature neonates), have CF, or are infected with
pathogens exhibiting high MICs. Outside these specific
medical conditions, the PD-guided dosing strategies,
as compared to traditional dosing, may not provide
additional clinical benefit to patients. Based on a
systematic review of limited studies conducted in
pediatric patients, the routine use of extended or
continuous infusion of beta-lactam antibiotics was not
supported by current evidence, as excessive treatment
failures have not been reported.53 Nonetheless, more
well-designed prospective trials in pediatric patients,
particularly those with underlying diseases that may
benefit from PD-guided dosing strategies, are necessary
to confirm the clinical value of these dosing strategies
to directly affect patient care.
Aminoglycosides
Based on studies in adults using traditional multiple-
daily dosing of tobramycin and gentamicin, clinical
response to these antibiotics is maximized when the
Cmax /MIC ratio is 8 to 10 (Table 3).54,55 This traditional
dosing has been associated with decreased resistance
S116
patterns, but suppression of P aeruginosa requires a
much higher Cmax /MIC target of 30.56,57 To optimize
the aminoglycosides’ PD features and potentially
minimize adaptive resistance, the once-daily (or
extended-interval) dosing strategy consolidates multi-
ple doses into one large daily dose to achieve high peak
concentrations to effectively eradicate certain bacteria,
including those with reduced susceptibility and an
MIC >1 μg/mL.58,59 With their post-antibiotic effect,
once-daily aminoglycoside dosing provides a drug-free
interval where serum concentrations can fall below the
MIC of susceptible organisms, yet retain inhibition of
bacterial growth. As an additional clinical benefit, the
extended-interval strategy may reduce nephrotoxicity
because it allots sufficient time for complete elimination
of the drug prior to re-administration.59,60 While
Cmax /MIC is the PD index used for aminoglycoside
efficacy, the clinical significance of AUC monitoring re-
quires further exploration. Current dose-fractionation
studies reveal that AUC/MIC may be correlated to the
development of P aeruginosa resistance.55,56
The efficacy and safety of extended-interval dosing
was compared to traditional thrice-daily dosing of
tobramycin at 10 or 15 mg/kg/day in a meta-analysis
of four randomized, controlled studies that evaluated
both pediatric and adult participants with CF.61 No
significant differences in clinical outcomes (including
pulmonary function, nutritional status, time to first
pulmonary exacerbation after treatment course, and
ototoxicity) were reported between these dosing strate-
gies. However, the advantage of the extended-interval
dosing strategy was decreased nephrotoxicity, defined
by serum creatinine and urinary renal biomarkers.
Gentamicin, tobramycin and, to some extent,
amikacin have been thoroughly examined in pediatric
patients. While tobramycin and amikacin are the pri-
mary aminoglycosides used to treat pulmonary infec-
tions caused by P aeruginosa in CF, gentamicin has
been more widely used for decades for the treatment of
suspected or proven bacterial sepsis caused by enteric
bacilli in neonates. Currently, this PD-guided, once-
daily strategy is standard of care for adults and has
been evaluated in children with CF, children with febrile
neutropenia, and neonates.55,57,60–62
Children with CF commonly receive multiple
courses of aminoglycoside therapy because they
have pulmonary infections caused by P aeruginosa
throughout their lifetime. As a result, the infecting
bacteria often have decreased antibiotic susceptibility
owing to frequent previous antibiotic exposure.57
Furthermore, adequate antibiotic concentrations are
difficult to achieve in the ELF and mucosal surfaces
in these infected children. In addition, patients with
CF tend to have increased drug Vd and CL, which
necessitates higher antibiotic doses. Collectively,
The Journal of Clinical Pharmacology / Vol 58 No S10 2018
for these reasons, extended-interval dosing provides
a rational strategy to employ in these patients to
maximize clinical benefit.
Pharmacodynamically, once-daily dosing can
achieve antibiotic exposures that are effective
against less susceptible bacteria. In fact, tobramycin
10 mg/kg/day (which achieves Cmax of 20 to 40 μg/mL,
trough concentrations <1 μg/mL, and AUC24 of 60
to 120 mg-h/L) and once-daily amikacin 30 to 35
mg/kg/day (which achieves Cmax of 80 to 120 μg/mL,
trough concentrations <1 μg/mL, and AUC24 of 235 ±
110 mg-h/L) can reach the PD index of Cmax /MIC 10
for less susceptible pathogens, specifically those with
MICs up to 2 μg/mL for tobramycin and 4 μg/mL for
amikacin.57 Because of the PD property of aminogly-
cosides, the routine use of extended-interval dosing is
supported by the Cystic Fibrosis Foundation.63
Gentamicin is bactericidal against Gram-negative
pathogens and has been used widely in neonates for
decades. Multiple elements affect the PK of gentamicin
in neonates, including postconceptional age, weight,
and renal function. The PD index usually targeted in
this population is Cmax of 5 to 10 μg/mL for efficacy
(assuming MIC 1 μg/mL) and trough concentrations
<1 to 2 μg/mL to minimize toxicity.60 The high variabil-
ity and unpredictability of gentamicin’s PK reinforces
the value of individualized dosing and therapeutic
drug monitoring, particularly in premature neonates.64
In fact, a multicenter study demonstrated the rapid
achievement of therapeutic concentrations using an
individualized nomogram-based dosing strategy, with
less need for measuring drug concentrations and dosing
adjustments.65
With an evidently prolonged half-life due largely to
the neonate’s immature renal function and, to a lesser
extent, increased Vd, once- daily dosing is reasonable
and, especially in extreme preemies, dosing every 36 to
48 hours may be necessary.62 Based on a comparative
meta-analysis of 574 neonates who were <28 days of
life, once-daily dosing, as compared with multiple-daily
dosing, improved attainment of peak concentrations
5 μg/mL and trough concentrations <2 μg/mL.60
Efficacy and toxicity, including ototoxicity and nephro-
toxicity (assessed by changes in serum creatinine and
renal biomarkers) were comparable between these dos-
ing strategies. Accordingly, neonates who receive once-
daily gentamicin dosing and for a duration that is
less than 48 to 72 hours may not routinely require
the therapeutic monitoring of drug concentrations or
assessments of renal function for toxicity.
Children may experience febrile neutropenia as
a result of cancer chemotherapy or transplantation
immunosuppressive therapy. Depending on patient
age, gentamicin or tobramycin administered at 6 to
10.5 mg/kg/day achieves Cmax /MIC 10, which is
Le and Bradley
necessary to treat infections caused by P aeruginosa
in these children.66–68 Age-related differences in renal
CL necessitate a higher milligram per kilogram dose of
gentamicin for younger children with febrile neutrope-
nia. In addition, once-daily administration is preferred
to thrice-daily administration to maximize efficacy
and minimize nephrotoxicity and has the concomitant
benefit of less nursing time for drug administration.
With minimal post-antibiotic leukocyte enhancement
because of their neutropenia, loading doses to achieve
therapeutic peak concentrations early in the course of
therapy may be advantageous in these patients.
Aminoglycosides may cause a concerning adverse
effect that limits their use, nephrotoxicity. Prospective
randomized trials have documented the advantage of
once-daily over multiple-daily dosing in reducing the
occurrence of nephrotoxicity.55,69,70 In proximal renal
tubular epithelial cells, aminoglycosides can accumulate
and thereby result in nephrotoxicity. Once-daily dosing
exposes the renal cells to high, intermittent concen-
trations that saturate drug uptake and consequently
lead to an increase in drug excretion.55,60 Further-
more, once-daily dosing facilitates sufficient time for
elimination of the drug to negligible concentrations
before the subsequent dose is administered, thereby
preventing nephrotoxicity. In a study of neonates, once-
daily gentamicin dosing resulted in reduction in the
ratio of urinary N-acetyl-βD-glucosaminidase to cre-
atinine, which appears to be a sensitive marker for
nephrotoxicity.69
The other significant adverse effect of amingoglyco-
sides is ototoxicity, in which free radicals irreversibly
damage the cochlear and vestibular hair cells. The
occurrence of ototoxicity has not been associated with
frequency of aminoglycoside administration.23,55
Vancomycin
Vancomycin exhibits time-dependent bactericidal activ-
ity, with some post-antibiotic effect lasting 0.7 to 2.6
hours for Staphylococcus aureus and 4.3 to 6.5 hours
for Staphylococcus epidermidis.71 Based on adult PKPD
studies, the efficacy surrogate for vancomycin in treat-
ing invasive MRSA infections is achieving AUC/MIC

ratio 400, with the MIC determined by Etest
R
(Biomérieux; Marcy-l’Étoile, France). An AUC/MIC
ratio 400 corresponds to a minimum concentration
of 15 to 20 μg/mL when the MIC is 1 μg/mL in
adults.72,73 However, in infants and children >3 months
of age, this PD index may be achieved with lower
trough concentrations of 8 to 10 μg/mL when the
MIC is 1 μg/mL (evaluated using broth microdilution
and Etest methods).74,75 This crucial observation was
based on two independent PK studies that incorporated
robust Bayesian estimation and Monte Carlo simu-
lations, emphasizing the fact that pediatric response
S117
differs from adult response and adult data may not
always be applicable. Furthermore, there are insuffi-
cient PD data to demonstrate improved outcomes with
attainment of vancomycin AUC/MIC 400 in pediatric
patients and the data come primarily from retrospective
analyses.76–79 Prospective, comparative PD studies in
pediatric patients are necessary to validate the PD
target that so far has been evaluated only in adults.
Optimal vancomycin use requires incorporating
both age-related PK developmental changes and MIC
data. PK studies in pediatric participants suggest lower
trough concentrations will achieve AUC/MIC 400
but PD outcomes data are limited by inconclusive
retrospective studies. In the largest PopPK study in
pediatric participants >3 months (N = 702), the inte-
gration of weight, age, and serum creatinine (which,
in aggregate, reflects developmental renal maturation)
was necessary to estimate vancomycin CL.75 Using
AUC/MIC 400 as the PD target index due to lack of
current pediatric studies, increased vancomycin doses at
60 to 80 mg/kg/day (depending on age, serum creatinine,
and MIC by Etest) was necessary to achieve adequate
PTA. Notably, vancomycin AUC/MIC, compared with
trough concentrations, was a more achievable and,
hence, realistic target in children. Higher doses of 90 to
100 produced excessively high AUCs of 840 to 940 mg-
h/L, with trough concentrations exceeding 20 μg/mL.75
This excessive exposure may potentially place patients
at unnecessary and increased risk of nephrotoxicity, as
concluded by multiple studies in children with trough
concentrations 15 μg/mL, AUC >800 mg-h/L, and
doses 10 mg/kg.80–82
Because vancomycin is eliminated primarily via
GFR with some active tubular secretion, dosing at
15 mg/kg every 8 hours (ie, 45 mg/kg/day) may be most
appropriate empirically for older children with acute re-
nal insufficiency.83,84 Furthermore, children with acute
renal insufficiency generally do not have chronic disease
and, hence, may require close monitoring because re-
covery of renal function may occur during vancomycin
therapy, with markedly improved CL.
Population-based PK studies with Bayesian estima-
tion in preterm and full-term neonates have evaluated
vancomycin doses ranging from 15 to 60 mg/kg/day
administered intermittently every 8 to 24 hours or as
continuous infusions.85–88 Notably, these studies did
not evaluate PTA by AUC/MIC 400, which is the PD
target best linked to successful treatment outcomes in
adults. However, one study evaluated the association
between AUC and trough concentrations in 240
neonates, with 1702 vancomycin concentrations.89 This
PopPK study derived a model for vancomycin CL that
was based on weight, postmenstrual age, and serum cre-
atinine; using Monte Carlo simulations with Bayesian
estimation, trough concentrations ranging from 7 to
S118
11 μg/mL were highly predictive of an AUC24 of >400
μg-h/mL in at least 90% of neonates. Doses to achieve
this PKPD target ranged from 15 to 20 mg/kg every 8 to
12 hours, depending on postmenstrual age and serum
creatinine. The incidence of vancomycin-associated
nephrotoxicity reported in neonates has been low,
ranging from 1% to 9%.90 The high variability of
vancomycin’s PK in preterm and full-term neonates
necessitates therapeutic drug monitoring, preferably
with AUC estimation, to ensure adequate therapeutic
exposure.29
Recent PK data highlighting differences in trough
concentrations required to achieve the PD vancomycin
exposure target of AUC/MIC 400 in children and
adults prompts the need to investigate the optimal
monitoring strategy that can be integrated into clinical
practice.74,75 The minimal PD evidence in children,
coupled with the differences in trough concentrations,
supports the monitoring of vancomycin exposure by
AUC because it is a more relevant and achievable
PD target. Furthermore, AUC targeting and moni-
toring may prevent excessive dosing, frequent dosing
adjustments, and potentially adverse effects. A PopPK
study with Bayesian estimation in 138 pediatric par-
ticipants demonstrated improved the accuracy and
precision of estimating AUC using peak and trough
concentrations compared with trough concentrations
alone.91 However, one study in adults suggests that a
single sample may be adequate to predict appropriate
dosing.92
Current Challenges With Potential
Solutions
Antimicrobial agents are widely used in pediatric pa-
tients, especially neonates and young infants. Despite
extensive use of antibiotics, the PKPD in this popu-
lation are limited, arising from its historical exclusion
from clinical trials. Legislation via the Best Pharma-
ceuticals for Children Act (BPCA) and the Pediatric
Research Equity Act (PREA) support and mandate pe-
diatric research to improve drug product labeling.93 The
PREA legislation authorizes the National Institutes
of Health to implement research programs through
funding clinical trials to study off-patent drugs in all
children, including neonates and young infants. BPCA
provides economic incentives to companies in exchange
for important clinical data that address an important
medical need in pediatric patients. Continued efforts are
necessary to promote the need for PKPD evaluations of
drugs (both generic and new) across the broad pediatric
age spectrum.
While quantitative pharmacology is increasingly em-
ployed to evaluate the PKPD of drugs in pediatric
patients, the off-label use of drugs without PKPD data
The Journal of Clinical Pharmacology / Vol 58 No S10 2018
in the pediatric setting remains unacceptably high. Fur-
thermore, the integration of PKPD into clinical prac-
tice still needs improvement. The link between in vitro
and in vivo PKPD properties and their extrapolation to
the clinical use of drugs in pediatric patients needs to
be communicated clearly to clinicians in a manner that
changes practice.26 One approach may be to address the
quality of PKPD studies by developing an evidence-
based grading scheme that enables clinicians, many
of whom are not pharmacologists, to appreciate the
strength of the evidence.21 In addition, new evidence in
pediatric patients, including those who are obese, often
emerges post-marketing and should be integrated into
clinical practice by updating drug formularies.
Despite the successful use of allometric weight in
pediatric patients to estimate PK parameters, errors
may still occur, especially for drugs undergoing hepatic
metabolic CL. Physiologically based pharmacokinetic
models are emerging to facilitate simulations of PK
profiles under varying physiological conditions.94 These
models integrate data that are traditionally not in-
tegrated into PK modeling, including drug proper-
ties, physiological changes, and biological parameters,
which may differ between individuals due to age and
disease states. The biological and mechanistic basis of
PBPK models closes the gap between tissue concen-
trations and observed therapeutic or toxic effects. A
unique approach that combines the full PBPK and stan-
dard PopPK analyses, which has been termed as “semi-
physiological-based PK,” has been used to characterize
and predict the developmental changes of important
elimination pathways, like GFR, across the human life
span, from neonates to adults. This system-based phar-
macometric methodology permits the development of
one model that simultaneously describes multiple drugs
(eg, aminoglycosides and vancomycin) because they
undergo similar elimination.95
The ethics and logistics of blood sampling (both
by number and volume) in pediatric patients present
a challenge to conducting PKPD studies with enough
samples to accurately estimate PK parameters. Tra-
ditionally, a whole blood volume up to 5 mL and
sampling 5 to 8 times within a dosing interval were
necessary to accurately describe PK parameters in
pediatric patients. Innovative strategies to overcome
this challenge have been attempted and include sparse
sampling, microsampling, dried blood spot technique,
and opportunistic study design.
Sparse sampling is more practical than the tradi-
tional intensive approach and has been increasingly
appearing in published PK studies evaluating pediatric
populations.40,42,75,91,96 Sparse sampling provides one
major advantage: it reduces the number of blood draws
and, when applied in population-based modeling, pro-
vides estimation for both individual and population
Le and Bradley
PK parameters with inter-subject, intra-subject, and
“unexplained” residual variabilities.21 To accommodate
the reduced number of blood samples, a larger partic-
ipant sample size (often >100) must be used to pro-
duce meaningful results in PopPK modeling. Similar
to sparse sampling, microsampling reduces the overall
blood volume needed: whereas sparse sampling reduces
the frequency, microsampling reduces the actual vol-
ume drawn during sampling. In fact, microsampling
requires a volume <50 μL, which can be collected by
skin prick.97
Dried blood spot is a sampling technique that utilizes
an ultra-low volume (ie, 30 μL of whole blood, which
is 20 times lower than traditional venous or arterial
samples) to evaluate the PK of drugs. It is stable and
reproducible on paper filters. In addition, dried blood
spot sampling eliminates the need for centrifugation or
freezing of samples and measurement of drug concen-
tration in whole blood. As such, the advantages of dried
blood spot sampling are low sample volume, minimal
personnel training, no sample processing, room tem-
perature storage, and simple bioanalytical analysis.98
Using this technique, the estimates and precision of
metronidazole PK parameters for premature infants in
one small study were similar between plasma and dried
blood spot samples.99 Of concern though, metronida-
zole concentration in the dried blood spot samples was
15% lower than in the plasma samples. Limited pedi-
atric PK studies currently utilize this technique.99,100
Future studies should explore this sampling strat-
egy and identify class- and drug-specific concentra-
tion differences between dried blood spot and plasma
samples.
The opportunistic study design capitalizes on stan-
dard clinical care to perform clinical research. This
design is perceived as minimal risk, diminishes research-
related costs, and enhances parental consent. While the
opportunistic design does not permit administration
of study drug to patients because the drug is already
prescribed as standard of care, it provides four main
advantages and therefore is attractive to use in pediatric
research. First, this design evaluates a drug that is
already prescribed by clinical indication. Second, it
optimizes the collection for PK samples by coinciding
timing with routine labs to minimize pricking. Third,
for drugs with active therapeutic drug monitoring, data
on drug concentrations are already measured by the
clinical laboratory and available for use in PK mod-
eling. Finally, the opportunistic study design utilizes
scavenged samples to maximize testing on drawn blood
that would otherwise be left unused.
Scavenged PK sampling is a minimal-risk approach
that can provide meaningful information related to
the development of PK models but cannot inform
dosing recommendations. The utility of scavenged
S119
sampling in providing definitive dosing recommenda-
tions may be drug dependent and needs to be further
explored.42 Opportunistic studies have been used to
extract important findings for the preliminary construct
of phase 1 through 3 trials and enhance the dosing
of currently marketed antimicrobial agents in pediatric
populations.75,101,102
Conclusions
Children of all ages may experience variability in
response to antimicrobial agents, thereby potentially
leading to undesired subtherapeutic or unanticipated
toxic effects when PKPD data are unavailable. This
underscores the importance of thoroughly understand-
ing the age-related physiologic changes that occur with
normal human growth and development. Regulatory
agencies through PREA are requiring PK and efficacy
data to optimize pharmacotherapy in pediatric pop-
ulations, including neonates. While the availability of
PK data in these populations has improved over the
past decades, PKPD outcome assessments remain in-
finitesimal. Rational antibiotic dosing can only improve
with more well-designed PKPD pediatric studies in the
future.
Author Disclosure
Drs Bradley and Le have no personal financial involve-
ment with any company that creates, develops, markets, or
sells antimicrobial agents. Their employer, the University of
California, has contracts with many companies to develop
and test antimicrobial agents. Dr Le has received research
support from NICHD and NIAID (1U54HD090259 and
1K23AI089978) and subagreement from Duke University
(HHSN27500027). Dr Bradley has received research support
from NICHD (1U54HD090259).
References
1. Bradley JS, Garonzik SM, Forrest A, Bhavnani SM. Pharma-
cokinetics, pharmacodynamics, and Monte Carlo simulation:
selecting the best antimicrobial dose to treat an infection.
Pediatr Infect Dis J. 2010;29(11):1043–1046.
2. Anderson BJ, Holford NH. Understanding dosing: children are
small adults, neonates are immature children. Arch Dis Child.
2013;98(9):737–744.
3. Anderson BJ, Holford NH. Mechanism-based concepts of
size and maturity in pharmacokinetics. Annu Rev Pharmacol
Toxicol. 2008;48:303–332.
4. Bulitta JB, Landersdorfer CB, Huttner SJ, et al. Population
pharmacokinetic comparison and pharmacodynamic break-
points of ceftazidime in cystic fibrosis patients and healthy
volunteers. Antimicrob Agents Chemother. 2010;54(3):1275–
1282.
5. Mahmood I. Prediction of drug clearance in children: impact of
allometric exponents, body weight, and age. Ther Drug Monit.
2007;29(3):271–278.
S120
6. Natale S, Bradley J, Nguyen WH, et al. Pediatric obesity: phar-
macokinetic alterations and effects on antimicrobial dosing.
Pharmacotherapy. 2017;37(3):361–378.
7. Rhodin MM, Anderson BJ, Peters AM, et al. Human renal
function maturation: a quantitative description using weight
and postmenstrual age. Pediatr Nephrol. 2009;24(1):67–76.
8. Mahmood I. Prediction of drug clearance in children from
adults: a comparison of several allometric methods. Br J Clin
Pharmacol. 2006;61(5):545–557.
9. Nielsen EI, Sandstrom M, Honore PH, Ewald U, Friberg LE.
Developmental pharmacokinetics of gentamicin in preterm and
term neonates: population modelling of a prospective study.
Clin Pharmacokinet. 2009;48(4):253–263.
10. Kearns GL, Abdel-Rahman SM, Alander SW, Blowey DL,
Leeder JS, Kauffman RE. Developmental pharmacology–drug
disposition, action, and therapy in infants and children. N Engl
J Med. 2003;349(12):1157–1167.
11. van den Anker JN, Schwab M, Kearns GL. Developmental
pharmacokinetics. Handb Exp Pharmacol. 2011;205:51–75.
12. Vinita BP, Nahata MC. Drug dosing in pediatric patients. In:
Murphy JE, ed. Clinical Pharmacokinetics. 5th ed. Bethesda,
MD: American Society of Health-System Pharmacists;
2012.
13. Pai VB, Nahata MC. Dosing in pediatric patients. In: Murphy
JE, ed. Clinical Pharmacokinetics. 6th ed. Bethesda, MD:
American Society of Health-System Pharmacists; 2017:65–76.
14. Oyaert M, Spriet I, Allegaert K, et al. Factors impacting
unbound vancomycin concentrations in different patient popu-
lations. Antimicrob Agents Chemother. 2015;59(11):7073–7079.
15. Smith MJ, Gonzalez D, Goldman JL, et al. Pharmacokinetics
of clindamycin in obese and nonobese children. Antimicrob
Agents Chemother. 2017;61(4).
16. Mooij MG, Nies AT, Knibbe CA, et al. Development of human
membrane transporters: drug disposition and pharmacogenet-
ics. Clin Pharmacokinet. 2016;55(5):507–524.
17. Lam J, Baello S, Iqbal M, et al. The ontogeny of P-glycoprotein
in the developing human blood-brain barrier: implication for
opioid toxicity in neonates. Pediatr Res. 2015;78(4):417–421.
18. Vinks AA, Emoto C, Fukuda T. Modeling and simulation
in pediatric drug therapy: application of pharmacometrics
to define the right dose for children. Clin Pharmacol Ther.
2015;98(3):298–308.
19. Ette EI, Williams PJ. Population pharmacokinetics I:
background, concepts, and models. Ann Pharmacother.
2004;38(10):1702–1706.
20. Samardzic J, Allegaert K, Wilbaux M, Pfister M, van den Anker
JN. Quantitative clinical pharmacology practice for optimal use
of antibiotics during the neonatal period. Expert Opin Drug
Metab Toxicol. 2016;12(4):367–375.
21. Barker CI, Standing JF, Turner MA, McElnay JC, Shar-
land M. Antibiotic dosing in children in Europe: can we
grade the evidence from pharmacokinetic/pharmacodynamic
studies - and when is enough data enough? Curr Opin Infect Dis.
2012;25(3):235–242.
22. Jelliffe RW, Schumitzky A, Van Guilder M, et al. Individualiz-
ing drug dosage regimens: roles of population pharmacokinetic
and dynamic models, Bayesian fitting, and adaptive control.
Ther Drug Monit. 1993;15(5):380–393.
23. Beaubien AR, Desjardins S, Ormsby E, et al. Incidence
of amikacin ototoxicity: a sigmoid function of total drug
exposure independent of plasma levels. Am J Otolaryngol.
1989;10(4):234–243.
24. Lunn DJ. Bayesian analysis of population pharmacoki-
netic/pharmacodynamic models. In: Husmeier D, Dybowski
R, Roberts S, eds. Probabilistic Modelling in Bioinformatics
The Journal of Clinical Pharmacology / Vol 58 No S10 2018
and Medical Informatics. Advanced Information and Knowledge
Processing. London, England: Springer-Verlag; 2005:351–369.
25. Food and Drug Administration. Guidance for industry popula-
tion pharmacokinetics. https://www.fda.gov/downloads/Drugs/
.../Guidances/UCM072137.pdf. Accessed October 3, 2017.
26. Gabrielsson J, Green AR. Quantitative pharmacology or phar-
macokinetic pharmacodynamic integration should be a vital
component in integrative pharmacology. J Pharmacol Exp
Ther. 2009;331(3):767–774.
27. Drusano GL. Pharmacokinetics and pharmacodynamics of
antimicrobials. Clin Infect Dis. 2007;45(1 suppl):S89–S95.
28. Michalopoulos AS, Falagas ME. Colistin: recent data on
pharmacodynamics properties and clinical efficacy in critically
ill patients. Ann Intensive Care. 2011;1(1):30.
29. Sullins AK, Abdel-Rahman SM. Pharmacokinetics of antibac-
terial agents in the CSF of children and adolescents. Pediatr
Drugs. 2013;15(2):93–117.
30. Onufrak NJ, Forrest A, Gonzalez D. Pharmacokinetic and
pharmacodynamic principles of anti-infective dosing. Clin
Ther. 2016;38(9):1930–1947.
31. Autmizguine J, Guptill JT, Cohen-Wolkowiez M, Benjamin
DK Jr, Capparelli EV. Pharmacokinetics and pharmacody-
namics of antifungals in children: clinical implications. Drugs.
2014;74(8):891–909.
32. Stockmann C, Constance JE, Roberts JK, et al. Pharmacoki-
netics and pharmacodynamics of antifungals in children and
their clinical implications. Clin Pharmacokinet. 2014;53(5):429–
454.
33. Cies JJ, Shankar V, Schlichting C, Kuti JL. Population phar-
macokinetics of piperacillin/tazobactam in critically ill young
children. Pediatr Infect Dis J. 2014;33(2):168–173.
34. Courter JD, Kuti JL, Girotto JE, Nicolau DP. Optimizing
bactericidal exposure for beta-lactams using prolonged and
continuous infusions in the pediatric population. Pediatr Blood
Cancer. 2009;53(3):379–385.
35. Ellis JM, Kuti JL, Nicolau DP. Use of Monte Carlo simulation
to assess the pharmacodynamics of beta-lactams against Pseu-
domonas aeruginosa infections in children: a report from the
OPTAMA program. Clin Therapeut. 2005;27(11):1820–1830.
36. Bradley JS, Sauberan JB, Ambrose PG, Bhavnani SM, Ras-
mussen MR, Capparelli EV. Meropenem pharmacokinet-
ics, pharmacodynamics, and Monte Carlo simulation in the
neonate. Pediatr Infect Dis J. 2008;27(9):794–799.
37. Cohen-Wolkowiez M, Ouellet D, Smith PB, et al. Population
pharmacokinetics of metronidazole evaluated using scavenged
samples from preterm infants. Antimicrob Agents Chemother.
2012;56(4):1828–1837.
38. Cohen-Wolkowiez M, Poindexter B, Bidegain M, et al. Safety
and effectiveness of meropenem in infants with suspected or
complicated intra-abdominal infections. Clin Infect Diseases.
2012;55(11):1495–1502.
39. Smith PB, Cohen-Wolkowiez M, Castro LM, et al. Population
pharmacokinetics of meropenem in plasma and cerebrospinal
fluid of infants with suspected or complicated intra-abdominal
infections. Pediatr Infect Dis J. 2011;30(10):844–849.
40. Tremoulet A, Le J, Poindexter B, et al. Population pharmacoki-
netics of ampicillin in neonates using an opportunistic study
design. Antimicrob Agents Chemother. 2014;58(6):3013–3020.
41. van den Anker JN, Pokorna P, Kinzig-Schippers M, et al.
Meropenem pharmacokinetics in the newborn. Antimicrob
Agents Chemother. 2009;53(9):3871–3879.
42. Cohen-Wolkowiez M, Benjamin DK Jr, Ross A, et al.
Population pharmacokinetics of piperacillin using scavenged
samples from preterm infants. Ther Drug Monit.
2012;34(3):312–319.
Le and Bradley
43. Cirillo I, Vaccaro N, Castaneda-Ruiz B, et al. Open-label study
to evaluate the single-dose pharmacokinetics, safety, and toler-
ability of doripenem in infants less than 12 weeks in chrono-
logical age. Antimicrob Agents Chemother. 2015;59(8):4742–
4749.
44. Vinks AA, van Rossem RN, Mathot RA, Heijerman HG, Mou-
ton JW. Pharmacokinetics of aztreonam in healthy subjects and
patients with cystic fibrosis and evaluation of dose-exposure
relationships using Monte Carlo simulation. Antimicrob Agents
Chemother. 2007;51(9):3049–3055.
45. Ashish A, Shaw M, Winstanley C, Ledson MJ, Walshaw
MJ. Increasing resistance of the Liverpool Epidemic Strain
(LES) of Pseudomonas aeruginosa (Psa) to antibiotics in cystic
fibrosis (CF)–a cause for concern? J Cyst Fibros. 2012;11(3):
173–179.
46. Spencker FB, Staber L, Lietz T, Schille R, Rodloff AC. Devel-
opment of resistance in Pseudomonas aeruginosa obtained from
patients with cystic fibrosis at different times. Clin Microbiol
Infect. 2003;9(5):370–379.
47. Le J, Bradley JS, Hingtgen S, et al. Pharmacokinetics of single-
dose ceftaroline fosamil in children with cystic fibrosis. Pediatr
Pulmonol. 2017;52(11):1424–1434.
48. Reed MD, Goldfarb J, Yamashita TS, Lemon E, Blumer
JL. Single-dose pharmacokinetics of piperacillin and tazobac-
tam in infants and children. Antimicrob Agents Chemother.
1994;38(12):2817–2826.
49. Delvallee M, Mazingue F, Abouchahla W, et al. Optimiza-
tion of continuous infusion of piperacillin-tazobactam in
children with fever and neutropenia. Pediatr Infect Dis J.
2013;32(9):962–964.
50. Kuti JL, Nightingale CH, Quintiliani R, Nicolau D. Phar-
macodynamic profiling of continuously infused piperacillin/
tazobactam against Pseudomonas aeruginosa using Monte
Carlo analysis. Diagn Microbiol Infect Dis. 2002;44(1):51–57.
51. Tam VH, McKinnon PS, Akins RL, Rybak MJ, Dru-
sano GL. Pharmacodynamics of cefepime in patients with
Gram-negative infections. J Antimicrob Chemother. 2002;50(3):
425–428.
52. Zelenitsky SA, Ariano RE, Zhanel GG. Pharmacodynamics
of empirical antibiotic monotherapies for an intensive care
unit (ICU) population based on Canadian surveillance data.
J Antimicrob Chemother. 2011;66(2):343–349.
53. Walker MC, Lam WM, Manasco KB. Continuous and ex-
tended infusions of beta-lactam antibiotics in the pediatric
population. Ann Pharmacother. 2012;46(11):1537–1546.
54. Moore RD, Lietman PS, Smith CR. Clinical response to
aminoglycoside therapy: importance of the ratio of peak con-
centration to minimal inhibitory concentration. J Infect Dis.
1987;155(1):93–99.
55. Pagkalis S, Mantadakis E, Mavros MN, Ammari C, Falagas
ME. Pharmacological considerations for the proper clinical use
of aminoglycosides. Drugs. 2011;71(17):2277–2294.
56. Tam VH, Ledesma KR, Vo G, Kabbara S, Lim TP, Nikolaou
M. Pharmacodynamic modeling of aminoglycosides against
Pseudomonas aeruginosa and Acinetobacter baumannii: iden-
tifying dosing regimens to suppress resistance development.
Antimicrob Agents Chemother. 2008;52(11):3987–3993.
57. Young DC, Zobell JT, Stockmann C, et al. Optimiza-
tion of anti-pseudomonal antibiotics for cystic fibrosis pul-
monary exacerbations: V. Aminoglycosides. Pediatr Pulmonol.
2013;48(11):1047–1061.
58. Downes KJ, Hahn A, Wiles J, Courter JD, Vinks AA.
Dose optimisation of antibiotics in children: application of
pharmacokinetics/pharmacodynamics in paediatrics. Int J An-
timicrob Agents. 2014;43(3):223–230.
S121
59. Drusano GL, Louie A. Optimization of aminoglycoside
therapy. Antimicrob Agents Chemother. 2011;55(6):2528–
2531.
60. Rao SC, Srinivasjois R, Hagan R, Ahmed M. One dose per day
compared to multiple doses per day of gentamicin for treatment
of suspected or proven sepsis in neonates. Cochrane Database
Syst Rev. 2011(11):CD005091.
61. Smyth AR, Bhatt J. Once-daily versus multiple-daily dosing
with intravenous aminoglycosides for cystic fibrosis. Cochrane
Database Syst Rev. 2012;2:CD002009.
62. Mohamed AF, Nielsen EI, Cars O, Friberg LE.
Pharmacokinetic-pharmacodynamic model for gentamicin
and its adaptive resistance with predictions of dosing
schedules in newborn infants. Antimicrob Agents Chemother.
2012;56(1):179–188.
63. Flume PA, Mogayzel PJ Jr, Robinson KA, et al. Cystic fibrosis
pulmonary guidelines: treatment of pulmonary exacerbations.
Am J Respir Crit Care Med. 2009;180(9):802–808.
64. Roberts JK, Stockmann C, Constance JE, et al. Pharmacoki-
netics and pharmacodynamics of antibacterials, antifungals,
and antivirals used most frequently in neonates and infants. Clin
Pharmacokinet. 2014;53(7):581–610.
65. Glover ML, Shaffer CL, Rubino CM, et al. A multicenter
evaluation of gentamicin therapy in the neonatal intensive care
unit. Pharmacotherapy. 2001;21(1):7–10.
66. Dupuis LL, Sung L, Taylor T, et al. Tobramycin pharmacoki-
netics in children with febrile neutropenia undergoing stem cell
transplantation: once-daily versus thrice-daily administration.
Pharmacotherapy. 2004;24(5):564–573.
67. Inparajah M, Wong C, Sibbald C, et al. Once-daily gen-
tamicin dosing in children with febrile neutropenia resulting
from antineoplastic therapy. Pharmacotherapy. 2010;30(1):43–
51.
68. Sung L, Dupuis LL, Bliss B, et al. Randomized controlled trial
of once- versus thrice-daily tobramycin in febrile neutropenic
children undergoing stem cell transplantation. J Natl Cancer
Inst. 2003;95(24):1869–1877.
69. de Alba Romero C, Gomez Castillo E, Manzanares Secades
C, Rodriguez Lopez J, Arreaza Lopez L, Saenz Valiente P.
Once daily gentamicin dosing in neonates. Pediatr Infect Dis J.
1998;17(12):1169–1171.
70. van Lent-Evers NA, Mathot RA, Geus WP, van Hout BA,
Vinks AA. Impact of goal-oriented and model-based clin-
ical pharmacokinetic dosing of aminoglycosides on clini-
cal outcome: a cost-effectiveness analysis. Ther Drug Monit.
1999;21(1):63–73.
71. Lowdin E, Odenholt I, Cars O. In vitro studies of pharmacody-
namic properties of vancomycin against Staphylococcus aureus
and Staphylococcus epidermidis. Antimicrob Agents Chemother.
1998;42(10):2739–2744.
72. Kullar R, Davis SL, Levine DP, Rybak MJ. Impact of
vancomycin exposure on outcomes in patients with methicillin-
resistant Staphylococcus aureus bacteremia: support for
consensus guidelines suggested targets. Clin Infect Dis.
2011;52(8):975–981.
73. Rybak M, Lomaestro B, Rotschafer JC, et al. Therapeutic
monitoring of vancomycin in adult patients: a consensus review
of the American Society of Health-System Pharmacists, the
Infectious Diseases Society of America, and the Society of
Infectious Diseases Pharmacists. Am J Health Syst Pharm.
2009;66(1):82–98.
74. Frymoyer A, Guglielmo BJ, Hersh AL. Desired vancomycin
trough serum concentration for treating invasive methicillin-
resistant Staphylococcal infections. Pediatr Infect Dis J.
2013;32(10):1077–1079.
S122
75. Le J, Bradley JS, Murray W, et al. Improved vancomycin dosing
in children using area under the curve exposure. Pediatr Infect
Dis J. 2013;32(4):e155–163.
76. Hahn A, Frenck RW Jr, Allen-Staat M, Zou Y, Vinks AA.
Evaluation of target attainment of vancomycin area under
the curve in children with methicillin-resistant Staphylococcus
aureus bacteremia. Ther Drug Monit. 2015;37(5):619–625.
77. McNeil JC, Kaplan SL, Vallejo JG. The influence of the route
of antibiotic administration, methicillin susceptibility, van-
comycin duration and serum trough concentration on outcomes
of pediatric Staphylococcus aureus bacteremic osteoarticular
infection. Pediatr Infect Dis J. 2017;36(6):572–577.
78. McNeil JC, Kok EY, Forbes AR, et al. Healthcare-associated
Staphylococcus aureus bacteremia in children: evidence for
reverse vancomycin creep and impact of vancomycin trough
values on outcome. Pediatr Infect Dis J. 2016;35(3):263–268.
79. Welsh KJ, Abbott AN, Lewis EM, et al. Clinical character-
istics, outcomes, and microbiologic features associated with
methicillin-resistant Staphylococcus aureus bacteremia in pe-
diatric patients treated with vancomycin. J Clin Microbiol.
2010;48(3):894–899.
80. Le J, Ny P, Capparelli E, et al. Pharmacodynamic charac-
teristics of nephrotoxicity associated with vancomycin use in
children. J Pediatric Infect Dis Soc. 2015;4(4):e109–116.
81. McKamy S, Hernandez E, Jahng M, Moriwaki T, Deveikis
A, Le J. Incidence and risk factors influencing the develop-
ment of vancomycin nephrotoxicity in children. J Pediatr.
2011;158(3):422–426.
82. Myers AL, Gaedigk A, Dai H, James LP, Jones BL, Neville
KA. Defining risk factors for red man syndrome in children and
adults. Pediatr Infect Dis J. 2012;31(5):464–468.
83. Jacqz-Aigrain E, Zhao W, Sharland M, van den Anker JN. Use
of antibacterial agents in the neonate: 50 years of experience
with vancomycin administration. Semin Fetal Neonatal Med.
2013;18(1):28–34.
84. Le J, Vaida F, Nguyen E, et al. Population-based pharmacoki-
netic modeling of vancomycin in children with renal insuffi-
ciency. J Pharmacol Clin Toxicol. 2014;2(1):1017–1026.
85. Mehrotra N, Tang L, Phelps SJ, Meibohm B. Evalua-
tion of vancomycin dosing regimens in preterm and term
neonates using Monte Carlo simulations. Pharmacotherapy.
2012;32(5):408–419.
86. Patel AD, Anand D, Lucas C, Thomson AH. Continu-
ous infusion of vancomycin in neonates. Arch Dis Child.
2013;98(6):478–479.
87. Zhao W, Kaguelidou F, Biran V, et al. External evalua-
tion of population pharmacokinetic models of vancomycin in
neonates: the transferability of published models to different
clinical settings. Br J Clin Pharmacol. 2013;75(4):1068–1080.
88. Zhao W, Lopez E, Biran V, Durrmeyer X, Fakhoury M, Jacqz-
Aigrain E. Vancomycin continuous infusion in neonates: dosing
optimisation and therapeutic drug monitoring. Arch Dis Child.
2013;98(6):449–453.
89. Frymoyer A, Hersh AL, El-Komy MH, et al. Association
between vancomycin trough concentration and area under
The Journal of Clinical Pharmacology / Vol 58 No S10 2018
the concentration-time curve in neonates. Antimicrob Agents
Chemother. 2014;58(11):6454–6461.
90. Lestner JM, Hill LF, Heath PT, Sharland M. Vancomycin
toxicity in neonates: a review of the evidence. Curr Opin Infect
Dis. 2016;29(3):237–247.
91. Le J, Ngu B, Bradley JS, et al. Vancomycin monitoring
in children using Bayesian estimation. Ther Drug Monit.
2014;36(4):510–518.
92. Neely MN, Youn G, Jones B, et al. Are vancomycin trough
concentrations adequate for optimal dosing? Antimicrob Agents
Chemother. 2014;58(1):309–316.
93. Ren Z, Zajicek A. Review of the Best Pharmaceuticals for
Children Act and the Pediatric Research Equity Act: what can
the obstetric community learn from the pediatric experience?
Semin Perinatol. 2015;39(7):530–531.
94. Espie P, Tytgat D, Sargentini-Maier ML, Poggesi I, Watelet JB.
Physiologically based pharmacokinetics (PBPK). Drug Metab
Rev. 2009;41(3):391–407.
95. De Cock RF, Allegaert K, Brussee JM, et al. Simultaneous
pharmacokinetic modeling of gentamicin, tobramycin and van-
comycin clearance from neonates to adults: towards a semi-
physiological function for maturation in glomerular filtration.
Pharm Res. 2014;31(10):2643–2654.
96. Meibohm B, Laer S, Panetta JC, Barrett JS. Population phar-
macokinetic studies in pediatrics: issues in design and analysis.
AAPS J. 2005;7(2):E475–487.
97. Dorofaeff T, Bandini RM, Lipman J, Ballot DE, Roberts JA,
Parker SL. Uncertainty in antibiotic dosing in critically ill
neonate and pediatric patients: can microsampling provide the
answers? Clin Ther. 2016;38(9):1961–1975.
98. Laughon MM, Benjamin DK, Jr., Capparelli EV, et al.
Innovative clinical trial design for pediatric therapeutics. Expert
Rev Clin Pharmacol. 2011;4(5):643–652.
99. Cohen-Wolkowiez M, Sampson M, Bloom BT, et al. Deter-
mining population and developmental pharmacokinetics of
metronidazole using plasma and dried blood spot samples
from premature infants. Pediatr Infect Dis J. 2013;32(9):956–
961.
100. Suyagh M, Collier PS, Millership JS, et al. Metronidazole
population pharmacokinetics in preterm neonates using dried
blood-spot sampling. Pediatrics. 2011;127(2):e367–374.
101. Tremoulet A, Le J, Poindexter B, et al. Characterization of the
population pharmacokinetics of ampicillin in neonates using
an opportunistic study design. Antimicrob Agents Chemother.
2014;58(6):3013–3020.
102. Wade KC, Wu D, Kaufman DA, et al. Population pharma-
cokinetics of fluconazole in young infants. Antimicrob Agents
Chemother. 2008;52(11):4043–4049.
103. Giebel NL, Shadley JD, McCarver DG, et al. Role of chromatin
structural changes in regulating human CYP3A ontogeny. Drug
Metab Dispos. 2016;44(7):1027–1037.
104. Downes KJ, Hahn A, Wiles J, Courter JD, Vinks AA.
Dose optimisation of antibiotics in children: application of
pharmacokinetics/pharmacodynamics in paediatrics. Int J An-
timicrob Agents. 2014;43(3):223–230.