Clinical PK

Concepts in Clinical
Pharmacokinetics
SIXTH EDITION
William J. Spruill, PharmD, FASHP, FCCP

Associate Professor
Department of Clinical and Administrative Sciences
University of Georgia, College of Pharmacy
Athens, Georgia
William E. Wade, PharmD, FASHP, FCCP

Professor
Department of Clinical and Administrative Sciences
University of Georgia, College of Pharmacy
Athens, Georgia
Joseph T. DiPiro, PharmD

Executive Dean
South Carolina College of Pharmacy
Charleston, South Carolina
Robert A. Blouin, PharmD

Professor and Dean
UNC Eshelman School of Pharmacy
The University of North Carolina at Chapel Hill
Chapel Hill, North Carolina
Jane M. Pruemer, PharmD, BCOP, FASHP

Professor of Clinical Pharmacy Practice
James L. Winkle College of Pharmacy
University of Cincinnati
Cincinnati, Ohio
American Society of Health-System Pharmacists®

Bethesda, Maryland
Purchased by nova southeastern university, Todd Puccio
From: ASHP eBooks (digital.ashp.org)
Any correspondence regarding this publication should be sent to the publisher, American Society of Health-System
Pharmacists, 7272 Wisconsin Avenue, Bethesda, MD 20814, attention: Special Publishing.
The information presented herein reflects the opinions of the contributors and advisors. It should not be interpreted as an
official policy of ASHP or as an endorsement of any product.
Because of ongoing research and improvements in technology, the information and its applications contained in this text
are constantly evolving and are subject to the professional judgment and interpretation of the practitioner due to the
uniqueness of a clinical situation. The editors and ASHP have made reasonable efforts to ensure the accuracy and
appropriateness of the information presented in this document. However, any user of this information is advised that the
editors and ASHP are not responsible for the continued currency of the information, for any errors or omissions, and/or
for any consequences arising from the use of the information in the document in any and all practice settings. Any reader
of this document is cautioned that ASHP makes no representation, guarantee, or warranty, express or implied, as to the
accuracy and appropriateness of the information contained in this document and specifically disclaims any liability to any
party for the accuracy and/or completeness of the material or for any damages arising out of the use or non-use of any of
the information contained in this document.
Director, Special Publishing: Jack Bruggeman

Acquisitions Editor: Robin Coleman
Editorial Project Manager: Ruth Bloom
Production Manager: Johnna Hershey
Cover & Page Design: David Wade
Library of Congress Cataloging - in - Publication Data

Spruill, William J., author.
Concepts in clinical pharmacokinetics / authors, William J. Spruill, William E. Wade, Joseph T. DiPiro. - - Sixth edition.
p. ; cm.
Preceded by Concepts in clinical pharmacokinetics / Joseph T. DiPiro ... [et al.]. Includes bibliographical references and
index.
ISBN 978 - 1 - 58528 - 387 - 3
I. Wade, William E., author. II. DiPiro, Joseph T., author. III. American Society of Health-System Pharmacists. IV. Title.
[DNLM: 1. Pharmacokinetics - - Programmed Instruction. 2. Pharmaceutical Preparations - - administration & dosage
- - Programmed Instruction. QV 18.2]
RM301.5
615'.7 - - dc23
2013041917
© 2014, American Society of Health-System Pharmacists, Inc. All rights reserved.
No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical,
including photocopying, microfilming, and recording, or by any information storage and retrieval system, without written
permission from the American Society of Health-System Pharmacists.
ASHP is a service mark of the American Society of Health-System Pharmacists, Inc.; registered in the U.S. Patent and
Trademark Office.
ISBN: 978-1-58528-387-3
10 9 8 7 6 5 4 3 2 1

Table of Contents
Preface .................................................................................................................................................. v
Acknowledgments............................................................................................................................. vii
A Note from the Authors on Using This Edition..................................................................... vii
Abbreviations..................................................................................................................................... ix
Lessons and Practice Sets

Lesson 1. Introduction to Pharmacokinetics and Pharmacodynamics................. 1
Lesson 2. Basic Pharmacokinetics ...................................................................................... 21
Lesson 3. Half-Life, Elimination Rate, and AUC.............................................................. 31
Practice Set 1 .................................................................................................................................. 45
Lesson 4. Intravenous Bolus Administration, Multiple Drug
Administration, and Steady-State Average Concentrations................... 49
Lesson 5. Relationships of Pharmacokinetic Parameters and
Intravenous Intermittent and Continuous Infusions................................ 65
Lesson 6. Two-Compartment Models................................................................................. 81
Practice Set 2 .................................................................................................................................. 93
Lesson 7. Biopharmaceutics: Absorption ........................................................................ 99
Lesson 8. Drug Distribution and Protein Binding......................................................... 115
Lesson 9. Drug Elimination Processes............................................................................... 127
Lesson 10. Nonlinear Processes.............................................................................................. 149
Lesson 11. Pharmacokinetic Variation and Model-Independent
Relationships............................................................................................................ 159
Practice Set 3................................................................................................................................... 175
Lesson 12. Aminoglycosides .................................................................................................... 179
Lesson 13. Vancomycin............................................................................................................... 203
Lesson 14. Theophylline............................................................................................................. 221
Lesson 15. Phenytoin and Digoxin......................................................................................... 231
Appendix A. Basic and Drug-Specific Pharmacokinetic Equations............................. 249
Appendix B. Supplemental Problems...................................................................................... 259
Appendix C. Glossary .................................................................................................................... 267
Index .................................................................................................................................................... 271
iii
Preface
For students just entering the world of pharmacy or seasoned practitioners, the
study of pharmacokinetics and the mathematical equations required for drug
dosing can be quite intimidating. Combined with the terminology of the science,
this may make a course in pharmacokinetics a considerable challenge. In this sixth
edition of Concepts in Clinical Pharmacokinetics, we continue to focus on the funda-
mental pharmacokinetic concepts. These concepts, along with the mathematical
equations, are broken down to their simplest forms, and a step-by-step approach is
adopted to explain the “how to” of the discipline. We believe that such an approach
allows the student to gain greater comprehension of the subject matter, which
allows adaptation of concepts to specific clinical drug dosing situations.
Pharmacokinetic concepts are further illustrated by application to clinical
dosing cases, including aminoglycosides, vancomycin, theophylline, digoxin, and
phenytoin. These cases are designed to show the easily understandable, step-
by-step approach for performing appropriate clinical dosing consults. All cases
provide the complete mathematical solutions for each calculation, allowing
readers to “check their math.” Equations are explained in detail, and all similar
equations used throughout the text are cross-referenced to the basic concept. In
addition there is a valuable appendix containing basic and drug-specific pharma-
cokinetic equations.
This edition expands on several concepts including proper estimation of renal
function, extended-interval aminoglycoside dosing, pharmacogenomic effects on
drug metabolism, a phenytoin “cheat sheet” to help you through the calculations
maze, and new vancomycin cases based on higher desired vancomycin levels and
trough-only dose estimations. As with past editions the reader will find numerous
clinical correlates throughout the text to further highlight specific clinical or math-
ematical explanations.
The goal for this edition, as with the previous five editions, remains the same—
to provide the student or practitioner with the concepts and clinical applications
needed for a better understanding of this complicated, yet vital, subject.
William J. Spruill
William E. Wade
Joseph T. DiPiro
Robert A. Blouin
Jane M. Pruemer
February 2014
v
Acknowledgments
We are indebted to our wives Paula Spruill, Theresa Wade, and Cecily DiPiro for
their love and patience during the preparation of this sixth edition.
A Note from the Authors on Using This Edition
This book teaches the basic biopharmaceutic concepts, mathematical models, and
clinical applications needed to determine such values as dose, interval, steady-
state concentration, etc. Specific conceptual and mathematical formulas are
combined to solve more complex dosing situations. Eleven chapters contain a
practice quiz to chart your progress, and there are three practice sets of questions
with answers. The last four chapters are completely devoted to clinical cases that
fully explain, step-by-step, how to dose several drugs that generally require serum
drug concentrations. We strongly encourage you to attempt to solve these cases
without looking at the step-by-step answers, and then when finished, check to see
if you got them right.
—WS, WW, JD
A complete online course based on this book with four enrollment options is
available through the University of Georgia Center for Continuing Education.
To learn more go to:
http://www.georgiacenter.uga.edu/courses/healthcare-pharmacy
vii
Abbreviations
α : distribution rate constant for two-compartment model

AUC : area under plasma drug concentration versus time curve
AUMC : area under the (drug concentration × time) versus time (moment) curve
β : terminal elimination rate constant
C : concentration
C average steady-state concentration
C0, C1, C2 initial (just after infusion), first, second concentrations
Cin concentration in blood on entering organ
Clast last measured concentration
Cmax maximum concentration
Cmax1, Cmax2 first, second maximum concentrations
Cmax (steady state) steady-state maximum concentration
Cmin (steady state) steady-state minimum concentration
Cmin minimum concentration
Cout concentration in blood on leaving organ
Cpeak peak concentration
Css steady-state concentration
Ct concentration at time t
Ctrough trough concentration
Cl : clearance
Clb biliary clearance
Clh hepatic (liver) clearance
Cli intrinsic clearance
Clm clearance by metabolism (mainly liver)
Clother organs clearance by other organs
ClP→mX formation clearance for a given metabolite X
ClP→m 1 fractional clearance of parent drug (P ) to form metabolite 1 (m1 )
Clr renal clearance
Clt total body clearance
conc : concentration
Δ : change in
E : extraction ratio
continued on next page
ix
Abbreviations
x
e : base of natural logarithm S : salt form of drug

F : fraction of drug absorbed that reaches systemic SST : serum separator tube
circulation (bioavailability) τ : dosing interval
Fm1 fraction of m1 formed from a single dose of the t : time (after dose)
parent drug
t ′ time after end of infusion (t ′ = τ – t for trough
Fp fraction of unbound drug in plasma concentration)
Ft fraction of unbound drug in tissue t ′′ time (duration) of loading infusion
GFR : glomerular filtration rate t0 time zero
GI : gastrointestinal T½ half-life
K : elimination rate constant t 90% time required to reach 90% of steady-state
K0 rate of drug infusion concentration
K12 rate constant for transfer of drug from V : volume; volume of distribution
compartment 1 to compartment 2 Varea volume of distribution by area
K21 rate constant for transfer of drug from Vc volume of central compartment
compartment 2 to compartment 1
Vextrap extrapolated volume of distribution
Ka absorption rate constant
Vp plasma volume
Km Michaelis–Menten constant (drug concentration
Vss steady-state volume of distribution
at which elimination rate = ½ Vmax)
Vt tissue volume
λ : terminal elimination rate constant
Vmax maximum rate of the elimination process
m1, m2, m3 : metabolites 1, 2, and 3
m1, u, m2, u, m3, u : amount of m1, m2, or m3 excreted in the X : amount of drug
urine X0 dose (or initial dose) of drug
MRT: mean residence time X1, X2 amount of drug at different times
n : number of doses Xc amount of drug in central compartment
Q : bloodflow Xd daily dose of drug
Qh hepatic bloodflow Xp amount of drug in peripheral compartment

LESSON 1
Introduction to
Pharmacokinetics
and Pharmacodynamics
OBJECTIVES
After completing Lesson 1, you should be able to:
1. Define and differentiate between pharmacokinetics and clinical pharmacokinetics.
2. Define pharmacodynamics and relate it to pharmacokinetics.
3. Describe the concept of the therapeutic concentration range.
4. Identify factors that cause interpatient variability in drug disposition and drug
response.
5. Describe situations in which routine clinical pharmacokinetic monitoring would be
advantageous.
6. List the assumptions made about drug distribution patterns in both one- and
two-compartment models.
7. Represent graphically the typical natural log of plasma drug concentration versus time
curve for a one-compartment model after an intravenous dose.
Pharmacokinetics is currently defined as the study of the time course of drug

absorption, distribution, metabolism, and excretion. Clinical pharmacokinetics is
the application of pharmacokinetic principles to the safe and effective therapeutic
management of drugs in an individual patient.
Primary goals of clinical pharmacokinetics include enhancing efficacy and
decreasing toxicity of a patient’s drug therapy. The development of strong corre-
lations between drug concentrations and their pharmacologic responses has
enabled clinicians to apply pharmacokinetic principles to actual patient situations.
A drug’s effect is often related to its concentration at the site of action, so it
would be useful to monitor this concentration. Receptor sites of drugs are gener-
ally inaccessible to our observations or are widely distributed in the body, and
therefore direct measurement of drug concentrations at these sites is not practical.
For example, the receptor sites for digoxin are thought to be within the myocar-
dium. Obviously we cannot directly sample drug concentration in this tissue.
However, we can measure drug concentration in the blood or plasma, urine, saliva,
and other easily sampled fluids (Figure 1-1). Kinetic homogeneity describes the
predictable relationship between plasma drug concentration and concentration at
the receptor site where a given drug produces its therapeutic effect (Figure 1-2).
Changes in the plasma drug concentration reflect changes in drug concentrations
1
Concepts in Clinical Pharmacokinetics
2
Sample Removed for Drug Concentration

Determination
FIGURE 1-1. FIGURE 1-3.

Blood is the fluid most often sampled for drug concentration Drug concentration versus time.
determination.
the central nervous system in Parkinson’s disease or
at the receptor site, as well as in other tissues. As bone in osteomyelitis). This assumption, however,
the concentration of drug in plasma increases, the may not be true for all drugs.
concentration of drug in most tissues will increase
proportionally.
Similarly, if the plasma concentration of a drug Clinical Correlate
is decreasing, the concentration in tissues will also Drugs concentrate in some tissues because of
decrease. Figure 1-3 is a simplified plot of the drug physical or chemical properties. Examples include
concentration versus time profile after an intrave- digoxin, which concentrates in the myocardium,
nous drug dose and illustrates this concept. and lipid-soluble drugs, such as benzodiazepines,
The property of kinetic homogeneity is important which concentrate in fat.
for the assumptions made in clinical pharmaco-
kinetics. It is the foundation on which all therapeutic
and toxic plasma drug concentrations are established. Basic Pharmacodynamic Concepts
That is, when studying concentrations of a drug in
Pharmacodynamics refers to the relationship
plasma, we assume that these plasma concentrations
between drug concentration at the site of action
directly relate to concentrations in tissues where the
and the resulting effect, including the time course
disease process is to be modified by the drug (e.g.,
and intensity of therapeutic and adverse effects.
The effect of a drug present at the site of action is
determined by that drug’s binding with a receptor.
Receptors may be present on neurons in the central
nervous system (i.e., opiate receptors) to depress
pain sensation, on cardiac muscle to affect the
intensity of contraction, or even within bacteria to
disrupt maintenance of the bacterial cell wall.
For most drugs, the concentration at the site of
the receptor determines the intensity of a drug’s
effect (Figure 1-4). However, other factors affect
drug response as well. Density of receptors on the
cell surface, the mechanism by which a signal is
transmitted into the cell by second messengers
FIGURE 1-2.
(substances within the cell), or regulatory factors
Relationship of plasma to tissue drug concentrations.
that control gene translation and protein produc-

Lesson 1 | Introduction to Pharmacokinetics and Pharmacodynamics
3
FIGURE 1-4.
Relationship of drug concentration to drug effect at the
receptor site.
tion may influence drug effect. This multilevel

regulation results in variation of sensitivity to drug
effect from one individual to another and also deter-
mines enhancement of, or tolerance to, drug effects.
In the simplest examples of drug effect, there is
a relationship between the concentration of drug
at the receptor site and the pharmacologic effect. If
enough concentrations are tested, a maximum effect
(Emax) can be determined (Figure 1-5). When the
logarithm of concentration is plotted versus effect
FIGURE 1-5.
(Figure 1-5), one can see that there is a concentration Relationship of drug concentration at the receptor site to effect
below which no effect is observed and a concentra- (as a percentage of maximal effect).
tion above which no greater effect is achieved.
One way of comparing drug potency is by the For some drugs, the effectiveness can decrease
concentration at which 50% of the maximum effect with continued use. This is referred to as toler-
is achieved. This is referred to as the 50% effective ance. Tolerance may be caused by pharmacokinetic
concentration or EC50. When two drugs are tested factors, such as increased drug metabolism, that
in the same individual, the drug with a lower EC50 decrease the concentrations achieved with a given
would be considered more potent. This means that dose. There can also be pharmacodynamic toler-
a lesser amount of a more potent drug is needed to ance, which occurs when the same concentration
achieve the same effect as a less potent drug. at the receptor site results in a reduced effect with
The EC50 does not, however, indicate other impor- repeated exposure. An example of drug tolerance
tant determinants of drug response, such as the is the use of opiates in the management of chronic
duration of effect. Duration of effect is determined pain. It is not uncommon to find these patients
by a complex set of factors, including the time that requiring increased doses of the opiate over time.
a drug is engaged on the receptor as well as intra- Tolerance can be described in terms of the dose–
cellular signaling and gene regulation. response curve, as shown in Figure 1-6.

4
and application of the resulting concentration

data to develop safe and effective drug regimens.
If performed properly, this process allows for the
achievement of therapeutic concentrations of a
drug more rapidly and safely than can be attained
with empiric dose changes. Together with observa-
tions of the drug’s clinical effects, it should provide
the safest approach to optimal drug therapy.
The usefulness of plasma drug concentration
data is based on the concept that pharmacologic
response is closely related to drug concentration
at the site of action. For certain drugs, studies in
patients have provided information on the plasma
concentration range that is safe and effective in
FIGURE 1-6. treating specific diseases—the therapeutic range
Demonstration of tolerance to drug effect with repeated dosing. (Figure 1-7). Within this therapeutic range, the
desired effects of the drug are observed. Below it,
To assess the effect that a drug regimen is likely there is greater probability that the therapeutic
to have, the clinician should consider pharmaco- benefits are not realized; above it, toxic effects may
kinetic and pharmacodynamic factors. Both are occur.
important in determining a drug’s effect. No absolute boundaries divide subtherapeutic,
therapeutic, and toxic drug concentrations. A gray
area usually exists for most drugs in which these
Clinical Correlate concentrations overlap due to variability in indi-
vidual patient response.
Tolerance can occur with many commonly
used drugs. One example is the hemodynamic
tolerance that occurs with continued use of
organic nitrates, such as nitroglycerin. For this
drug, tolerance can be reversed by interspersing
drug-free intervals with chronic drug use.
Clinical Correlate
One way to compare potency between two drugs
that are in the same pharmacologic class is to
compare EC50. The drug with a lower EC50 is
considered more potent.
FIGURE 1-7.
Relationship between drug concentration and drug effects for a
Therapeutic Drug Monitoring hypothetical drug.
Source: Adapted with permission from Evans WE, editor.
Therapeutic drug monitoring is defined as the General principles of applied pharmacokinetics. In:
use of assay procedures for determination of drug Applied Pharmacokinetics, 3rd ed. Vancouver, WA: Applied
concentrations in plasma, and the interpretation Therapeutics; 1992. pp.1–3.

5

Example of variability in plasma drug concentration among When pharmacologic effects relate to plasma drug
subjects given the same drug dose. concentrations, the latter can be used to predict the former.
Numerous pharmacokinetic characteristics

of a drug may result in variability in the plasma • The drug’s desired pharmacologic effects
concentration achieved with a given dose when cannot be assessed readily by other simple
administered to various patients (Figure 1-8). This means (e.g., blood pressure measurement
interpatient variability is primarily attributed to for antihypertensives).
one or more of the following: The value of therapeutic drug monitoring is
• Variations in drug absorption limited in situations in which:
• Variations in drug distribution • There is no well-defined therapeutic plasma
concentration range.
• Differences in an individual’s ability to metab-
olize and eliminate the drug (e.g., genetics) • The formation of pharmacologically active
metabolites of a drug complicates the appli-
• Disease states (renal or hepatic insuffi-
cation of plasma drug concentration data to
ciency) or physiologic states (e.g., extremes
clinical effect unless metabolite concentra-
of age, obesity) that alter drug absorption,
tions are also considered.
distribution, or elimination
• Toxic effects may occur at unexpectedly
• Drug interactions
low drug concentrations as well as at high
Therapeutic monitoring using drug concentration concentrations.
data is valuable when:
• There are no significant consequences asso-
• A good correlation exists between the ciated with too high or too low levels.
pharmacologic response and plasma concen-
Theophylline is an excellent example of a drug in
tration. Over at least a limited concentration
which significant interpatient variability in pharma-
range, the intensity of pharmacologic effects
cokinetic properties exists. This is important from a
should increase with plasma concentration.
clinical standpoint as subtle changes in serum concen-
This relationship allows us to predict phar-
trations may result in marked changes in drug response.
macologic effects with changing plasma
Figure 1-10 shows the relationship between theoph-
drug concentrations (Figure 1-9).
ylline concentration (x-axis, on a logarithmic scale)
• Wide intersubject variation in plasma drug and its pharmacologic effect (changes in pulmonary
concentrations results from a given dose. function [y-axis]). This figure illustrates that as the
• The drug has a narrow therapeutic index concentration of theophylline increases, so does the
(i.e., the therapeutic concentration is close intensity of the response for some patients. Wide
to the toxic concentration). interpatient variability is also shown.

6
FIGURE 1-11.
FIGURE 1-10. Process for reaching dosage decisions with therapeutic drug
Relationship between plasma theophylline concentration and monitoring.
change in forced expiratory volume (FEV) in asthmatic patients.
Source: Reproduced with permission from Mitenko PA, Ogilvie TABLE 1-1. Therapeutic Ranges for Commonly
RI. Rational intravenous doses of theophylline. N Engl J Med Used Drugs
1973;289:600–3. Copyright 1973, Massachusetts Medical Society.
Drug Range
Figure 1-11 outlines the process clinicians may Digoxin 0.5–2.0 ng/mL
choose to follow in making drug dosing decisions Lidocaine 1.5–5.0 mg/L
Lithium 0.6–1.4 mEq/L
by using therapeutic drug monitoring. Figure 1-12
Phenobarbital 15–40 mg/L
shows the relationship of pharmacokinetic and
Phenytoin 10–20 mg/L
pharmacodynamic factors.
Quinidine 2–5 mg/L
Examples of therapeutic ranges for commonly Cyclosporine 150–400 ng/mL
used drugs are shown in Table 1-1. Valproic acid 50–100 mg/L
As can be seen in this table, most drug concen- Carbamazepine 4–12 mcg/L
trations are expressed as a unit of mass per volume. Ethosuximide 40–100 mg/L
Primidone 5–12 mg/L
Source: Adapted with permission from Bauer LA. Clinical
Clinical Correlate pharmacokinetics and pharmacodynamics. In: DiPiro JT, Talbert RL,
Yee GC, et al., editors. Pharmacotherapy: a Pathophysiologic Approach,
A drug’s effect may also be determined by 8th ed. New York: McGraw-Hill; http://Accesspharmacy.com
the amount of time that the drug is present
at the site of action. An example is with beta-
lactam antimicrobials. The rate of bacterial Pharmacokinetic Models
killing by beta-lactams (the bacterial cell would
be considered the site of action) is usually The handling of a drug by the body can be very
determined by the length of time that the drug complex, as several processes (such as absorption,
concentration remains above the minimal distribution, metabolism, and elimination) work
concentration that inhibits bacterial growth. to alter drug concentrations in tissues and fluids.
Simplifications of body processes are necessary to

7
FIGURE 1-12.
Relationship of
pharmacokinetics
and
pharmacodynamics
and factors that
affect each.
predict a drug’s behavior in the body. One way to To construct a compartmental model as a
make these simplifications is to apply mathematical representation of the body, simplifications of body
principles to the various processes. structures are made. Organs and tissues in which
To apply mathematical principles, a model of the drug distribution is similar are grouped into one
body must be selected. A basic type of model used compartment. For example, distribution into adipose
in pharmacokinetics is the compartmental model. tissue differs from distribution into renal tissue
Compartmental models are categorized by the for most drugs. Therefore, these tissues may be in
number of compartments needed to describe the different compartments. The highly perfused organs
drug’s behavior in the body. There are one-compart- (e.g., heart, liver, and kidneys) often have similar
ment, two-compartment, and multicompartment drug distribution patterns, so these areas may be
models. The compartments do not represent a considered as one compartment. The compartment
specific tissue or fluid but may represent a group of that includes blood (plasma), heart, lungs, liver, and
similar tissues or fluids. These models can be used kidneys is usually referred to as the central compart-
to predict the time course of drug concentrations in ment or the highly blood-perfused compartment
the body (Figure 1-13). (Figure 1-14). The other compartment that includes
fat tissue, muscle tissue, and cerebrospinal fluid is the
Compartmental models are termed determin-
peripheral compartment, which is less well perfused
istic because the observed drug concentrations
than the central compartment.
determine the type of compartmental model
required to describe the pharmacokinetics of the
drug. This concept will become evident when we
examine one- and two-compartment models.

Simple compartmental model. Typical organ groups for central and peripheral compartments.

8
Another simplification of body processes

concerns the expression of changes in the amount of
drug in the body over time. These changes with time
are known as rates. The elimination rate describes
the change in the amount of drug in the body due to
drug elimination over time. Most pharmacokinetic
models assume that elimination does not change
over time.
The value of any model is determined by how FIGURE 1-15.
well it predicts drug concentrations in fluids and One-compartment model.
tissues. Generally, it is best to use the simplest model
that accurately predicts changes in drug concen- This model is the simplest because there is only
trations over time. If a one-compartment model is one compartment. All body tissues and fluids are
sufficient to predict plasma drug concentrations considered a part of this compartment. Further-
(and those concentrations are of most interest to more, it is assumed that after a dose of drug is
us), then a more complex (two-compartment or administered, it distributes instantaneously to all
more) model is not needed. However, more complex body areas. Common abbreviations are shown in
models are often required to predict tissue drug Figure 1-15.
concentrations. Some drugs do not distribute instantaneously to
all parts of the body, however, even after intravenous
bolus administration. Intravenous bolus dosing means
Clinical Correlate administering a dose of drug over a very short time
period. A common distribution pattern is for the
Drugs that do not extensively distribute into
drug to distribute rapidly in the bloodstream and
extravascular tissues, such as aminoglycosides,
to the highly perfused organs, such as the liver and
are generally well described by one-compartment
models. Extent of distribution is partly determined kidneys. Then, at a slower rate, the drug distributes to
by the chemistry of the agents. Aminoglycosides other body tissues. This pattern of drug distribution
are polar molecules, so their distribution is may be represented by a two-compartment model.
limited primarily to extracellular water. Drugs Drug moves back and forth between these compart-
extensively distributed in tissue (such as ments to maintain equilibrium (Figure 1-16).
lipophilic drugs like the benzodiazepines) or that Figure 1-17 simplifies the difference between
have extensive intracellular uptake may be better one- and two-compartment models. Again, the
described by the more complex models. one-compartment model assumes that the drug is
distributed to tissues very rapidly after intravenous
administration.
Compartmental Models
The one-compartment model is the most frequently
used model in clinical practice. In structuring
the model, a visual representation is helpful. The
compartment is represented by an enclosed square
or rectangle, and rates of drug transfer are repre-
sented by straight arrows (Figure 1-15). The arrow
pointing into the box simply indicates that drug is
put into that compartment; the arrow pointing out
FIGURE 1-16.
of the box indicates that drug is leaving the compart- Compartmental model representing transfer of drug to and
ment. from central and peripheral compartments.

9
FIGURE 1-18.
Two-compartment model.
Clinical Correlate
Digoxin, particularly when given
intravenously, is an example of
a drug that is well described by
two-compartment pharmacokinetics.
After an intravenous dose is
administered, plasma concentrations
rise and then rapidly decline as drug
distributes out of plasma and into
muscle tissue. After equilibration
between drug in tissue and plasma,
plasma concentrations decline less
rapidly (Figure 1-19). The plasma
would be the central compartment,
and muscle tissue would be the
FIGURE 1-17. peripheral compartment.
Drug distribution in one- and two-compartment models.
The two-compartment model can be

represented as in Figure 1-18, where:
X0 = dose of drug
X1 = amount of drug in central compartment
X2 = amount of drug in peripheral
compartment
K = elimination rate constant of drug from
central compartment to outside the
body
K12 = elimination rate constant of drug from
central compartment to peripheral
compartment
K21 = elimination rate constant of drug from
peripheral compartment to central FIGURE 1-19.
compartment Plasma concentrations of digoxin after an intravenous dose.

10
Volume of Distribution When V is many times the volume of the body, the
Until now, we have spoken of the amount of drug (X) drug concentrations in some tissues should be much
in a compartment. If we also consider the volume greater than those in plasma. The smallest volume in
of the compartment, we can describe the concept which a drug may distribute is the plasma volume.
of drug concentration. Drug concentration in the To illustrate the concept of volume of distribu-
compartment is defined as the amount of drug in a tion, let us first imagine the body as a tank filled
given volume, such as mg/L: with fluid as the body is primarily composed of
water. To calculate the volume of the tank, we
amount of drug in body X can place a known quantity of substance into it
=
1-1 concentration =
volume in which drug V and then measure its concentration in the fluid
is distributed (Figure 1-20). If the amount of substance (X) and
Volume of distribution (usually expressed as V, Vd, or the resulting concentration (C) is known, then the
VD) is an important indicator of the extent of drug distri- volume of distribution (V) can be calculated using
bution into body fluids and tissues. V relates the amount the simplified equations:
of drug in the body (X) to the measured concentration in X X
the plasma (C). Thus, V is the volume required to account =X VC=
or C =or V
V C
for all of the drug in the body if the concentrations in all
tissues are the same as the plasma concentration: X = amount of drug in body
amount of drug V = volume of distribution
volume of distribution = C = concentration in the plasma
concentration
A large volume of distribution usually indicates that As with other pharmacokinetic parameters, volume
the drug distributes extensively into body tissues of distribution can vary considerably from one
and fluids. Conversely, a small volume of distribu- person to another because of differences in physi-
tion often indicates limited drug distribution. ology or disease states. Something to note: The dose
Volume of distribution indicates the extent of distri- of a drug (X0) and the amount of drug in the body
bution but not the tissues or fluids into which the drug (X) are essentially the same thing because all of the
distributes. Two drugs can have the same volume of dose goes into the body.
distribution, but one may distribute primarily into In this example, important assumptions have
muscle tissues, whereas the other may concentrate in been made: that instantaneous distribution occurs
adipose tissues. Approximate volumes of distribution and that it occurs equally throughout the tank.
for some commonly used drugs are shown in Table 1-2. In the closed tank, there is no elimination. This
example is analogous to a one-compartment model
TABLE 1-2. Approximate Volumes of Distribution
of the body after intravenous bolus administration.
of Commonly Used Drugs
However, there is one complicating factor—during
Drug Volume of Distribution (L/kg)
Amlodipine 16.0 ± 4
Ganciclovir 1.1 ± 0.2
Ketorolac 0.21 ± 0.04
Lansoprazole 0.35 ± 0.05
Montelukast 0.15 ± 0.02
Sildenafil 1.2 ± 0.3
Valsartan 0.23 ± 0.09
Source: Brunton LL, Lazo JS, Parker KL (editors). The Pharmacologic
Basis of Therapeutics, 11th edition. New York: McGraw-Hill; 2006. pp.
FIGURE 1-20.
1798, 1829, 1839, 1840, 1851, 1872, 1883.
The volume of a tank can be determined from the amount of
substance added and the resulting concentration.

11
Elimination
FIGURE 1-21.
Drug elimination complicates the determination of the volume
of the body from drug concentrations.
the entire time that the drug is in the body, elimina- FIGURE 1-22.
tion is taking place. So, if we consider the body as One-compartment model.
a tank with an open outlet valve, the concentration
used to calculate the volume of the tank would be
constantly changing (Figure 1-21). Clinical Correlate
We can use the relationship given in Equation 1-1
Drugs that have extensive distribution outside
for volume, amount of drug administered, and resulting
of plasma appear to have a large volume of
concentration to estimate a drug’s volume of distribu- distribution. Examples include digoxin, diltiazem,
tion in a patient. If we give a known dose of a drug and imipramine, labetalol, metoprolol, meperidine,
determine the concentration of that drug achieved in and nortriptyline.
the plasma, we can calculate a volume of distribution.
However, the concentration used for this estimation
must take into account changes resulting from drug Plasma Drug Concentration Versus
elimination, as discussed in Lessons 3 and 9. Time Curves
For example:
With the one-compartment model (Figure 1-22),
If 100 mg of drug X is administered intrave- if we continuously measure the concentration of a
nously and the plasma concentration is determined drug in the plasma after an intravenous bolus dose
to be 5 mg/L just after the dose is given, then: and then plot these plasma drug concentrations
dose X 0 100 mg against the times they are obtained, the curve shown
=
volume of distribution = = = 20 L in Figure 1-23 would result. Note that this plot is a
(V ) resulting C 5 mg/L
(V ) concentration curve and that the plasma concentration is highest
just after the dose is administered at time zero (t0).
Clinical Correlate
The volume of distribution is easily approximated
for many drugs. For example, if the first 80-mg
dose of gentamicin is administered intravenously
and results in a peak plasma concentration of
8 mg/L, volume of distribution would be
calculated as follows:
dose X 0 80 mg
=
volume of distribution = = = 10 L
(V ) resulting C 8 mg/L FIGURE 1-23.
(V ) concentration
Typical plasma drug concentration versus time curve for a
one-compartment model.

12
FIGURE 1-24.
Plasma drug concentrations determined at specific time points. FIGURE 1-26.
With a simple one-compartment, intravenous bolus model, a
plot of the natural log of plasma concentration versus time
Because of cost limitations and patient conve-
results in a straight line.
nience in clinical situations, only a small number of
plasma samples are usually obtained for measuring
A straight line is obtained from the natural log of
drug concentrations (Figure 1-24). From these
plasma drug concentration versus time plot only for
known values, one is able to predict plasma drug
drugs that follow first-order elimination processes
concentrations at times when no samples are avail-
and exhibit one-compartment distribution. First-
able (Figure 1-25). In clinical situations, it is rare to
order elimination occurs when the amount of drug
collect more than two samples after a dose.
eliminated from the body in a specific time is depen-
The prediction of drug concentrations based on dent on the amount of drug in the body at that time.
known concentrations can be subject to multiple This concept is explained further in Lesson 2.
sources of error. However, if we realize the assump-
An alternative to calculating the natural log
tions used to make the predictions, some errors can
values is to plot the actual concentration and time
be avoided. These assumptions are pointed out as
values on semilogarithmic (or semilog) paper
we review the one-compartment system.
(Figure 1-27), a special graph paper that auto-
From a mathematical standpoint, the prediction matically adjusts for the logarithmic relationship by
of plasma concentrations is easier if we know that altering the distance between lines on the y-axis. The
the concentrations are all on a straight line rather lines on the y-axis are not evenly spaced but rather
than a curve. This conversion can be accomplished
for most drugs by plotting the natural logarithm (ln)
of the plasma drug concentration versus time. The
plot of a curve (Figure 1-25) is, in effect, converted
to a straight line by using the natural log of the
plasma drug concentration (Figure 1-26).
FIGURE 1-25.
Plasma drug concentrations can be predicted for times when they
were not determined. Concentrations on the line drawn through FIGURE 1-27.
the measured concentrations are predicted concentrations. Paper with one log-scale axis is called semilog paper.

13
are logarithmically related within each log cycle (or

multiple of 10). So when the actual values of plasma
drug concentrations are plotted against the time
values, a straight line results. The x-axis has evenly
spaced lines; there is no logarithmic conversion of
those values. (The term semilogarithmic indicates
that only one axis is converted.) The numbers on
the y-axis may be used to represent 0.1 through 1,
1 through 10, 10 through 100, or any series with a
10-fold difference in the range of values. FIGURE 1-28.
When plasma concentration versus time points fall on a
straight line, concentrations at various times can be predicted
Clinical Correlate simply by picking a time point and matching concentration on
the line at that time.
Semilog graph paper can be found via google.com
Math Principle
If a series of plasma concentration versus time The log of a number is the power to which a given
points are known and plotted on semilog paper, base number must be raised to equal that number.
a straight line can be drawn through the points With natural logarithms, the base is 2.718. For
by visual inspection or, more accurately, by linear example, the natural logarithm of 8.0 is x, where
regression techniques. Linear regression is a math- 2.718x = 8.0 and x = 2.08. Natural logarithms are
ematical method used to determine the line that best used because they relate to natural processes such
represents a set of plotted points. From this line, we as drug elimination, radioactive decay, and bacterial
can predict plasma drug concentrations at times for growth. Instead of 2.718 to indicate the base of the
which no measurements are available (Figure 1-28).
Clinical Correlate
For a typical patient, plasma concentrations
resulting from an 80-mg dose of gentamicin
may be as shown in Table 1-3. The plasma
concentrations plotted on linear and
semilogarithmic graph paper are shown in
Figure 1-29. With the semilog paper, it is
easier to predict what the gentamicin plasma
concentration would be 10 hours after the dose is
administered.
TABLE 1-3. Time Course of Plasma Gentamicin

Concentration
Concentration (mg/L) Time after Dose (hours)
6 1
4.4 2
2.4 4
0.70 8 FIGURE 1-29.
Predicting plasma drug concentrations with semilog scale.

14
natural log function, the abbreviation e is used. Also, estimated data. These are the ln key and the ex key.
instead of writing natural logarithm of 8.0, we shall Certain calculators do not have the ex key. Instead,
use the abbreviation ln 8.0. they will have an ln key and an INV key or a 2nd key.
Natural logarithms can be related to common Pressing the INV key or the 2nd key and then the ln
logarithms (base 10 logarithms) as follows: key will give ex values.
log base e
log base 10 =
2.303 Clinically Important Equations
Identified in This Chapter
Using the Calculator with Natural Log
and Exponential Keys 1. C = X / V
There are two major keys that will be used to calcu- 2. V = X / C

late pharmacokinetic values from either known or

15
REVIEW QUESTIONS
1-1. The study of the time course of drug absorp- 1-6. The EC50 refers to the drug concentration at
tion, distribution, metabolism, and excretion which:
is called: A. one-half the maximum response is
A. pharmacodynamics. achieved.
B. drug concentration. B. the maximal effect is achieved.
C. pharmacokinetics. C. tolerance is likely to be observed.
D. kinetic homogeneity.
1-7. The therapeutic range is the range of plasma
1-2. The application of pharmacokinetic prin- drug concentrations that clearly defines optimal
ciples to the safe and effective therapeutic drug therapy and where toxic effects cannot
management of drugs in an individual patient occur.
is known as: A. True
A. pharmacodynamics. B. False
B. clinical pharmacokinetics.
1-8. Therapeutic drug concentration moni-
1-3. Because we cannot practically measure drug toring with plasma drug concentration data
concentration in specific tissues, we measure assumes that pharmacologic response is
it in the plasma and assume that this concen- related to the drug concentration in plasma.
tration is the same as that in tissue. A. True
A. True B. False
B. False
1-9. One factor that may result in variability in
1-4. Pharmacodynamics refers to the relation- plasma drug concentrations after the same
ship of drug: drug dose is given to different patients
includes variations in:
A. dose to drug concentration in plasma.
A. drug absorption.
B. dose to drug concentration at the
receptor site. B. the EC50 of the drug.
C. concentrations to drug effect.
1-10. An example of a situation that would not
D. dose to drug effect. support therapeutic drug monitoring with
plasma drug concentrations would be one
1-5. Drug pharmacodynamics are affected by in which:
the drug concentration at the site of the
A. a wide variation in plasma drug concen-
receptor, density of receptors on the target
trations is achieved in different patients
cell surface, mechanism by which a signal is
given a standard drug dose.
transmitted into the cell by second messen-
gers, and regulatory factors that control B. the toxic plasma concentration is many
gene translation and protein production. times the therapeutic concentration
range.
A. True
C. correlation between a drug’s plasma
B. False
concentration and therapeutic response
is positive.

16
1-11. For a drug with a narrow therapeutic index, 1-17. For a drug that has first-order elimination
the plasma concentration required for ther- and follows a one-compartment model,
apeutic effects is near the concentration which of the following plots would result in
that produces toxic effects. a curved line?
A. True A. plasma concentration versus time
B. False B. natural log of plasma concentration
versus time
1-12. Highly perfused organs and blood comprise
what is usually known as the peripheral 1-18. A drug that follows a one-compartment
compartment. model is given as an intravenous injection,
A. True and the following plasma concentrations
are determined at the times indicated:
B. False
Plasma Concentration Time after Dose
1-13. The most commonly used model in clinical
(mg/L) (hours)
pharmacokinetic situations is the:
81 1
A. one-compartment model.
67 2
B. two-compartment model.
55 3
C. multicompartment model.
1-14. Instantaneous distribution to most body Using semilog graph paper, determine the
tissues and fluids is assumed in which of the approximate concentration in plasma at
following models? 6 hours after the dose.
A. one-compartment model A. 18 mg/L
B. two-compartment model B. 30 mg/L
C. multicompartment model C. < 1 mg/L
1-15. The amount of drug per unit of volume is

defined as the: ANSWERS
A. volume of distribution.
B. concentration. 1-1. A. Incorrect answer. Pharmacodynamics
C. rate. deals with the relationship between the
drug concentration at the site of action
1-16. If 3 g of a drug are added and distributed and the resulting effect.
throughout a tank and the resulting concen- B. Incorrect answer. Drug concentrations
tration is 0.15 g/L, calculate the volume of in plasma and tissues result from phar-
the tank. macokinetic processes.
A. 10 L C. CORRECT ANSWER
B. 20 L D. Incorrect answer. Kinetic homogeneity
C. 30 L describes the relationship between
D. 200 L plasma drug concentration and concen-
tration at a receptor or site of action.

17
1-2. A. Incorrect answer. Pharmacodynamics 1-8. A. CORRECT ANSWER

alone is not sufficient for effective ther- B. Incorrect answer. This statement is the
apeutic management, as it does not basic assumption underlying the use of
account for absorption, distribution, plasma drug concentrations.
metabolism, and excretion.
B. CORRECT ANSWER 1-9. A. CORRECT ANSWER
B. Incorrect answer. The EC50 is a way of
1-3. A. Incorrect answer. The plasma drug comparing drug potency. The EC50 is
concentration is not the same as that the concentration at which 50% of the
in the tissue but rather is related to the maximum effect of the drug is achieved.
tissue concentration by the volume of
1-10. A. Incorrect answer. A wide variation in
distribution (V). Plasma drug concentra-
plasma drug concentrations would be a
tions are commonly used because blood,
good justification for therapeutic drug
being readily accessible via venipuncture,
level monitoring.
is the body fluid most often collected for
drug measurement. B. CORRECT ANSWER. When the toxic
plasma concentration is much greater
B. CORRECT ANSWER
than the therapeutic concentration
range, there is less need for drug level
1-4. A, B. Incorrect answers. These statements
monitoring.
are definitions of pharmacokinetics.
C. Incorrect answer. A positive correlation
C. CORRECT ANSWER
between concentration and response
D. Incorrect answer. This statement refers makes therapeutic drug level moni-
to the effect of pharmacokinetic and toring more useful.
pharmacodynamic processes.
1-11. A. CORRECT ANSWER. For a drug with a
1-5. A. CORRECT ANSWER narrow therapeutic index, the plasma
B. Incorrect answer concentration required for therapeutic
effects is near the concentration that
1-6. A. CORRECT ANSWER produces toxic effects. The dosage of
B. Incorrect answer. The “50” in EC50 refers such a drug must be chosen carefully.
to 50% of the maximal effect. B. Incorrect answer
C. Incorrect answer. The term EC50 refers
1-12. A. Incorrect answer. The peripheral
to pharmacologic effect and not to toler-
compartment is generally made up
ance.
of less well-perfused tissues, such as
1-7. A. Incorrect answer. Although the thera- muscle and fat.
peutic range of a drug describes a range B. CORRECT ANSWER
of plasma drug concentrations generally
1-13. A. CORRECT ANSWER
considered safe and effective in a patient
population, no absolute boundaries B. Incorrect answer. Although a two-
divide subtherapeutic, therapeutic, and compartment model is often used,
toxic drug concentrations for an indi- it is not used as commonly as a one-
vidual patient. Both pharmacodynamic compartment model.
and pharmacokinetic factors influence a C. Incorrect answer. Multicompartment
patient’s response. models are used occasionally for
research purposes but are not normally
B. CORRECT ANSWER
used in clinical pharmacokinetics.

18
1-14. A. CORRECT ANSWER 1-16. A, C, D. Incorrect answers. A math error must

B. Incorrect answer. In a two-compartment have been made. The answer can be
model, it is assumed that drug distribu- found by dividing 3 g by 0.15 g/L.
tion to some tissues proceeds at a lower B. CORRECT ANSWER
rate than for other tissues.
C. Incorrect answer. In a multicompart- 1-17. A. CORRECT ANSWER
ment model, it is also assumed that drug B. Incorrect answer. This plot would be a
distribution to some tissues proceeds at straight line (see Figure 1-29).
a lower rate than for other tissues.
1-18. A, C. Incorrect answers. These results might
1-15. A. Incorrect answer. The volume of distri- have been determined if linear graph
bution refers to the dose over the paper was used or if the points were
resulting concentration. plotted incorrectly.
B. CORRECT ANSWER B. CORRECT ANSWER
C. Incorrect answer. The amount per unit
of volume is a static value and would not
change over time; therefore, it would
not be considered a rate.

19
Discussion Points
D1. An H2-receptor antagonist is given to control D3. The models shown in Figure 1-31 both well
gastric pH and prevent stress bleeding. The represent actual plasma concentrations of
following gastric pHs were observed when a drug after a dose. Which one should be
steady-state concentrations of the drug preferred to predict plasma levels? Provide
were achieved. What are the Emax and EC50 of a justification for your answer.
this drug?
Plasma Concentration (mg/L) Resulting pH

0.25 1.0
0.5 1.0
1 1.4
2 2.6
3 3.8
4 4.8
5 4.8
D2. The relationship shown in Figure 1-30 is

observed from a clinical study. What are
some of the likely reasons for this result?
FIGURE 1-31.
Models for predicting plasma drug concentrations over time.
D4. Would you expect a large drug molecule that

does not cross physiologic membranes very
well and is not lipid soluble to have a rela-
tively high or low volume of distribution?
Explain your answer.
D5. When plotting plasma drug concentration

(y-axis) versus time (x-axis), what are the
FIGURE 1-30. advantages of using a natural log scale for
Pharmacologic response versus drug plasma concentration. the y-axis rather than a linear scale?

LESSON 2
Basic Pharmacokinetics
OBJECTIVES
1. Define the concept of apparent volume of distribution and use an appropriate
mathematical equation to calculate this parameter.
2. Identify the components of body fluids that make up extracellular and intracellular
fluids and know the percentage of each component.
3. Describe the difference between whole blood, plasma, and serum.
4. Define drug clearance.
5. Describe the difference between first- and zero-order elimination and how each
appears graphically.
To examine the concept of volume of distribution further, let’s return to our

example of the body as a tank described in Lesson 1. We assumed that no drug was
being removed from the tank while we were determining volume. In reality, drug
concentration in the body is constantly changing, primarily due to elimination.
This flux makes it more difficult to calculate the volume in which a drug distributes.
One way to calculate the apparent volume of drug distribution in the body is
to measure the plasma concentration immediately after intravenous administra-
tion before elimination has had a significant effect. The concentration just after
intravenous administration (at time zero, t0) is abbreviated as C0 (Figure 2-1). The
volume of distribution can be calculated using the equation:
amount of drug
administered (dose) X 0 (mg)
=
volume of distribution = or V (L)
initial drug C 0 (mg/L)
concentration
(See Equation 1-1.)

C0 can be determined from a direct measurement or estimated by back-
extrapolation from concentrations determined at any time after the dose. If
two concentrations have been determined, a line containing the two values and
extending through the y-axis can be drawn on semilog paper. The point where that
line crosses the y-axis gives an estimate of C0. Both the direct measurement and
back-extrapolation approaches assume that the drug distributes instantaneously
into a single homogeneous compartment.
21
22
that its distribution is limited to extracellular fluid,

as that is the approximate volume of extracellular
fluid in the body. If a drug has a volume of distribu-
tion of about 40 L, the drug may be distributing into
all body water because a 70-kg person has approxi-
mately 40 L of body water (70 kg × 60%). If the volume
of distribution is much greater than 40–50 L, the drug
probably is being concentrated in tissue outside the
FIGURE 2-1. plasma and interstitial fluid.
Concentration resulting immediately after an intravenous
injection of a drug is referred to as C0. If a drug distributes extensively into tissues,
the volume of distribution calculated from plasma
concentrations could be much higher than the
The volume of distribution is an important
actual physiologic volume in which it distributes.
parameter for determining proper drug dosing regi-
For example, by measuring plasma concentrations,
mens. Often referred to as the apparent volume of
it appears that digoxin distributes in approximately
distribution, it does not have an exact physiologic
440 L in an adult. Because digoxin binds extensively
significance, but it can indicate the extent of drug
to muscle tissue, plasma levels are fairly low relative
distribution and aid in determination of dosage
to concentrations in muscle tissue. For other drugs,
requirements. Generally, dosing is proportional to
tissue concentrations may not be as high as the
the volume of distribution. For example, the larger
plasma concentration, so it may appear that these
the volume of distribution, the larger a dose must be
drugs distribute into a relatively small volume.
to achieve a desired target concentration.
It is also important to distinguish among blood,
To understand how distribution occurs, you
plasma, and serum. Blood refers to the fluid portion
must have a basic understanding of body fluids
in combination with formed elements (white cells,
and tissues (Figure 2-2). The fluid portion (water)
red cells, and platelets). Plasma refers only to the
in an adult makes up approximately 60% of total
fluid portion of blood (including soluble proteins
body weight and is composed of intracellular fluid
but not formed elements). When the soluble protein
(35%) and extracellular fluid (25%). Extracellular
fibrinogen is removed from plasma, the remaining
fluid is made up of plasma (4%) and interstitial fluid
product is serum (Figure 2-3).
(21%). Interstitial fluid surrounds cells outside the
vascular system. These percentages vary somewhat These differences in biologic fluids must be
in a child. recognized when considering reported drug concen-
trations. The plasma concentration of a drug may be
If a drug has a volume of distribution of approxi-
much less than the whole blood concentration if the
mately 15–18 L in a 70-kg person, we might assume
drug is preferentially sequestered by red blood cells.

Fluid distribution in an adult. Relationship of whole blood, plasma, and serum.

Lesson 2 | Basic Pharmacokinetics
23
Clinical Correlate Initial

concentration
Resulting
concentration
Elimination
Most drug concentrations are measured using over time
plasma or serum that usually generate similar
values. It is more relevant to use plasma or
serum than whole blood measurements to Volume from
estimate drug concentrations at the site of effect. which drug is
removed over time
However, some drugs, such as antimalarials, are
extensively taken up by red blood cells. In these FIGURE 2-5.
situations, whole blood concentrations would be Clearance may be viewed as the volume of plasma from which
more relevant, although they are not commonly drug is totally removed over a specified period.
used in clinical practice.
Drugs can be cleared from the body by many
different mechanisms, pathways, or organs, including
Clearance hepatic biotransformation and renal and biliary
excretion. Total body clearance of a drug is the sum of
Another important parameter in pharmacokinetics all the clearances by various mechanisms.
is clearance. Clearance is a measure of the removal of
drug from the body. Plasma drug concentrations are
affected by the rate at which drug is administered,
2-1 Clt = Clr + Clm + Clb + Clother
the volume in which it distributes, and its clearance.
A drug’s clearance and the volume of distribution where
determine its half-life. The concept of half-life and its
Clt = total body clearance (from all
relevant equations are discussed in Lesson 3.
mechanisms, where t refers to total);
Clearance (expressed as volume/time)
Clr = renal clearance (through renal
describes the removal of drug from a volume of
excretion);
plasma in a given unit of time (drug loss from the
body). Clearance does not indicate the amount Clm = clearance by liver metabolism or
of drug being removed. It indicates the volume biotransformation;
of plasma (or blood) from which the drug is Clb = biliary clearance (through biliary
completely removed, or cleared, in a given time excretion); and
period. Figures 2-4 and 2-5 represent two ways Clother = clearance by all other routes (gastro-
of thinking about drug clearance. In Figure 2-4, the intestinal tract, pulmonary, etc.).
amount of drug (the number of dots) decreases but
fills the same volume, resulting in a lower concen- For an agent removed primarily by the kidneys,
tration. Another way of viewing the same decrease renal clearance (Clr) makes up most of the total body
would be to calculate the volume that would be clearance. For a drug primarily metabolized by the
drug-free if the concentration were held constant. liver, hepatic clearance (Clm) is most important.
A good way to understand clearance is to
consider a single well-perfused organ that elimi-
Initial Resulting
concentration concentration nates drug. Blood flow through the organ is referred
Elimination to as Q (mL/minute) as seen in Figure 2-6, where
over time Cin is the drug concentration in the blood entering
the organ and Cout is the drug concentration in the
exiting blood. If the organ eliminates some of the
drug, Cin is greater than Cout.
FIGURE 2-4.
Decrease in drug concentration due to drug clearance.

24
TABLE 2-1. Rating of Extraction Ratios

Extraction Ratio (E ) Rating
> 0.7 High
0.3–0.7 Intermediate
< 0.3 Low
FIGURE 2-6. Clearance may also be a useful parameter for

Model for organ clearance of a drug. constructing dosage recommendations in clinical
situations. It is an index of the capacity for drug
We can measure an organ’s ability to remove a removal by the body’s organs.
drug by relating Cin and Cout. This extraction ratio (E) is:
C in − C out Clinical Correlate
E=
C in
Blood flow and the extraction ratio will determine
This ratio must be a fraction between zero and one. a drug’s clearance. Propranolol is a drug that is
Organs that are very efficient at eliminating a drug eliminated exclusively by hepatic metabolism.
will have an extraction ratio approaching one (i.e., The extraction ratio for propranolol is greater than
100% extraction). Table 2-1 is used as a general 0.9, so most of the drug presented to the liver is
guide. removed by one pass through the liver. Therefore,
The drug clearance of any organ is determined clearance is approximately equal to liver blood
by blood flow and the extraction ratio: flow (Cl = Q × E: when E ~ 1.0, Cl ~ Q). One
indication of the high extraction ratio is the
organ clearance = blood flow × extraction ratio relatively high oral dose of propranolol compared
with the intravenous dose; an oral dose is 10–20
or: times the equivalent intravenous dose. The
difference reflects the amount of drug removed
C −C by first-pass metabolism after absorption from
2-2 Clorgan =
Q × in out or Clorgan =
QE
C in the gastrointestinal tract and before entry into the
general circulation.
If an organ is very efficient in removing drug (i.e.,
extraction ratio near one) but blood flow is low, clear-
ance will also be low. Also, if an organ is inefficient in The average clearances of some commonly used
removing drug (i.e., extraction ratio close to zero) drugs are shown in Table 2-3. These values can
even if blood flow is high, clearance would again be vary considerably between individuals and may be
low. See Table 2-2. altered by disease.
The equations noted previously are not used
routinely in clinical drug monitoring, but they
TABLE 2-2. Effect on Clearance
describe the concept of drug clearance. Examina-
tion of a single well-perfused organ to understand Blood Flow (Q ) Clearance (Cl)
clearance is a noncompartmental approach; no Extraction Ratio (E ) (L/hour) (L/hour)
assumptions about the number of compartments High (0.7–1.0) Low Low
have to be made. Therefore, clearance is said to be
Low (< 0.3) High Low
a model-independent parameter. Clearance also
can be related to the model-dependent param- High (0.7–1.0) High High
eters volume of distribution and elimination rate Low (< 0.3) Low Low
(discussed in Lesson 3).

25
TABLE 2-3. Average Clearances of Common Drugs TABLE 2-4. First-Order Elimination
Amlodipine 5.9 ± 1.5 mL/min/kg Amount of Drug
Time after Drug Amount of Eliminated over Fraction of Drug
Ganciclovir 3.4 ± 0.5 mL/min/kg Administration Drug in Body Preceding Hour Eliminated over
(hours) (mg) (mg) Preceding Hour
Ketorolac 0.50 ± 0.15 mL/min/kg
0 1000 — —
Lansoprazole 6.23 ± 1.60 mL/min/kg
1 880 120 0.12
Montelukast 0.70 ± 0.17 mL/min/kg
2 774 106 0.12
Sildenafil 6.0 ± 1.1 mL/min/kg 3 681 93 0.12
Valsartan 0.49 ± 0.09 mL/min/kg 4 599 82 0.12
5 527 72 0.12
Source: Brunton LL, Lazo JS, Parker KL (editors), The
Pharmacologic Basis of Therapeutics, 11th edition. New York: 6 464 63 0.12
McGraw-Hill; 2006. pp. 1798, 1829, 1839, 1840, 1851, 1872, 7 408 56 0.12
1883.
The actual amount of drug eliminated is different
for each fixed time period depending on the initial
amount in the body, but the fraction removed is the
First-Order and Zero-Order Elimination
same, so this elimination is first order. Because the
The simplest example of drug elimination in a one- elimination of this drug (like most drugs) occurs by
compartment model is a single intravenous bolus a first-order process, the amount of drug eliminated
dose of a drug. It is first assumed that: decreases as the concentration in plasma decreases.
The actual fraction of drug eliminated over any given
1. Distribution and equilibration to all tissues
time (in this case 12%) depends on the drug itself and
and fluids occurs instantaneously so a one-
the individual patient’s capacity to eliminate the drug.
compartment model applies.
On the other hand, with zero-order elimination,
2. Elimination is first order. the amount of drug eliminated does not change with
Most drugs are eliminated by a first-order the amount or concentration of drug in the body,
process, and the concept of first-order elimination but the fraction removed varies (Figure 2-7). For
must be understood. With first-order elimination, example, if 1000 mg of a drug is administered and
the amount of drug eliminated in a set amount of the drug follows zero-order elimination, we might
time is directly proportional to the amount of drug observe the patterns in Table 2-5.
in the body. The amount of drug eliminated over a
certain time period increases as the amount of drug
in the body increases; likewise, the amount of drug
eliminated per unit of time decreases as the amount
of drug in the body decreases.
With the first-order elimination process,
although the amount of drug eliminated may change
with the amount of drug in the body, the fraction
of a drug in the body eliminated over a given time
remains constant. In practical terms, the fraction or
percentage of drug being removed is the same with
either high or low drug concentrations. For example, FIGURE 2-7.
Zero- versus first-order elimination. The size of the arrow
if 1000 mg of a drug is administered and the drug represents the amount of drug eliminated over a unit of
follows first-order elimination, we might observe the time. Percentages are the fraction of the initial drug amount
patterns in Table 2-4. remaining in the body.

26
TABLE 2-5. Zero-Order Elimination

Clinical Correlate
Fraction of
Amount of Drug Drug Elimi- Most antimicrobial agents (e.g.,
Time after Drug Eliminated over nated over
Administration Amount of Drug Preceding Hour Preceding aminoglycosides, cephalosporins, and
(hours) in Body (mg) (mg) Hour vancomycin) display first-order elimination
when administered in usual doses. The
0 1000 — —
pharmacokinetic parameters for these drugs are
1 850 150 0.15
not affected by the size of the dose given. As
2 700 150 0.18 the dose and drug concentrations increase, the
3 550 150 0.21 amount of drug eliminated per hour increases
4 400 150 0.27 while the fraction of drug removed remains the
5 250 150 0.38 same.
Some drugs (e.g., phenytoin), when given in
high doses, display zero-order elimination. Zero-
Now that we have examined zero- and first-
order elimination occurs when the body’s ability
order elimination, let’s return to our simple
to eliminate a drug has reached its maximum
one-compartment, intravenous bolus situation.
capability (i.e., all transporters are being used).
If the plasma drug concentration is continuously
As the dose and drug concentrations increase,
measured and plotted against time after admin-
the amount of drug eliminated per hour does
istration of an intravenous dose of a drug with
not increase, and the fraction of drug removed
first-order elimination, the plasma concentra-
declines.
tion curve shown in Figure 2-8 would result. To
predict concentrations at times when we did not
collect samples, we must linearize the plot by using
semilog paper (Figure 2-9).

Plasma drug concentration versus time after an intravenous As in Figure 2-8, but with a log scale y-axis.
(bolus) drug dose, assuming a one-compartment model with
first-order elimination (linear y-scale).

27
For a drug with first-order elimination, the

natural log of plasma concentration versus time
plot is a straight line. Conversely, plots with zero-
order elimination would be as shown in Figure
A 2-10. Note that for a drug with zero-order elimina-
tion, the plot of the plasma concentration versus
time is linear (plot A in Figure 2-10), whereas on
semilog paper (representing the natural log of
plasma concentration versus time) it is a curve
(bottom plot in Figure 2-10). If the natural log of
a plasma drug concentration versus time plot is
linear, it generally can be assumed that the drug
B follows first-order elimination.
FIGURE 2-10.
Plasma drug concentrations versus time after an intravenous
(bolus) drug dose, assuming a one-compartment model with
zero-order elimination (A, linear plot; B, log plot).

28
REVIEW QUESTIONS
2-1. The volume of distribution equals _________ 2-5. Plasma refers only to the fluid portion of
divided by initial drug concentration: blood, including soluble proteins but not
A.
clearance formed elements.
B. initial drug concentration A.
True
C.
half-life B.
False
D.
dose
2-6. The units for clearance are:
2-2. A dose of 1000 mg of a drug is administered A.
concentration/half-life.
to a patient, and the following concentrations B.
dose/volume.
result at the indicated times below. Assume a C.
half-life/dose.
one-compartment model.
D.
volume/time.
(mg/L) (hours) 2-7. Total body clearance is the sum of clearance
100 2 by the kidneys, liver, and other routes of
elimination.
67 4
45 6
A.
True
B.
False
An estimate of the volume of distribution
2-8. To determine drug clearance, we must first
would be:
determine whether a drug best fits a one- or
A. 10 L. two-compartment model.
B. 22.2 L. A.
True
C. 6.7 L. B.
False
D. 5 L.
2-9. With a drug that follows first-order elimina-
2-3. If a drug is poorly distributed to tissues, its tion, the amount of drug eliminated per unit
apparent volume of distribution is probably: time:
A.
large. A. remains constant while the fraction of
B.
small. drug eliminated decreases.
B. decreases while the fraction of drug
2-4. For the body fluid compartments below, eliminated remains constant.
rank them from the lowest volume to the
highest, in a typical 70-kg person.
A. Plasma < extracellular fluid < intracel-
ANSWERS
lular fluid < total body water 2-1. A, B, C. Incorrect answers
B. Extracellular fluid < intracellular fluid < D. CORRECT ANSWER. You can determine
plasma < total body water the correct answer from the units in
C. Intracellular fluid < extracellular fluid < the numerator and denominator. They
plasma < total body water should cancel to yield a volume unit.
D. Total body water < plasma < intracel- Grams divided by grams per liter would
lular fluid < extracellular fluid leave you with liter as the unit. The

29
volume is therefore determined from 2-4. A. CORRECT ANSWER. See Figure 2-2.
the dose, or amount of drug given, and Plasma would be 2.8 L, extracellular
the resulting initial concentration. fluid would be 18 L, intracellular fluid
would be 25 L, and total body water
2-2. A. Incorrect answer. You may have used would be 42 L.
100 mg/L as the initial concentration. B, C, D. Incorrect answers
B. Incorrect answer. You may have used an
incorrect initial concentration.
B. Incorrect answer
C. CORRECT ANSWER. To find the initial
concentration, plot the given plasma 2-6. A, B, C. Incorrect answers. The units for clear-
concentration and time values on ance are volume/time.
semilog paper, connect the points, and D. CORRECT ANSWER
read the value of the y-axis (concentra-
tion) when x (time) = 0. This should be 2-7.
A. CORRECT ANSWER. Total body clear-
150 mg/L. You can then determine the ance can be determined as the sum of
volume of distribution using the equa- individual clearances from all organs or
tion volume of distribution = dose/ routes of elimination.
initial concentration. B. Incorrect answer
Incorrect answer. You may have used
D. 2-8. A. Incorrect answer
an incorrect initial concentration, or
B. CORRECT ANSWER. It is not necessary
you may have used linear graph paper
to specify a model to determine drug
instead of semilog paper.
clearance.
2-3. A. Incorrect answer. Drug concentrations
2-9. A. Incorrect answer
are generally measured in plasma. When
B. CORRECT ANSWER. With first-order
drug distributes poorly into tissues,
elimination, the amount of drug elimi-
the plasma level will be increased.
nated in any time period is determined
Examining the equation volume of distri-
by the amount of drug present at the
bution = dose/initial concentration, as
start. Although the amount of drug elim-
the initial concentration increases, the
inated in successive time periods may
volume will decrease.
decrease, the fraction of the initial drug
B. CORRECT ANSWER that is eliminated remains constant.

30
Discussion Points
D1. Drug Y is given by an intravenous injection D3. Explain how a person who weighs 70 kg can
and plasma concentrations are then deter- have a volume of distribution for a drug of
mined as follows: 700 L.
Concentration
D4. For drug X, individual organ clearances have
Time after Injection (hours) (mg/L)
been determined as follows:
0 12
1 9.8 Renal clearance 180 mL/minute
2 7.9 Hepatic clearance 22 mL/minute
3 6.4 Pulmonary clearance 5.2 mL/minute
4 5.2
5 4.2 How would you describe the clearance of
6 3.4 drug X?
7 2.8
8 2.2

Is this drug eliminated by a first- or zero-
order process? Defend your answer.
D2. Which of the following patient scenarios is

associated with a smaller volume of distri-
bution?
A. Dose = 500 mg and initial serum concen-

tration is 40 mg/L
B. Dose = 20 mg and initial serum concen-
tration is 1.5 mg/L

LESSON 3
Half-Life, Elimination Rate,
and AUC
OBJECTIVES
1. Calculate the elimination rate constant given a natural log (ln) of plasma drug
concentration versus time curve.
2. Define half-life.
3. Calculate a drug’s half-life given a natural log of plasma drug concentration versus
time curve.
4. Define the relationship between half-life and elimination rate constant.
5. Calculate a drug’s half-life given its elimination rate constant.
6. Define drug clearance and relate it to the area under the plasma drug concentration
curve and drug dose.
7. Calculate a drug’s volume of distribution, concentration at time zero, and area under
the plasma concentration versus time curve (AUC), given plasma concentration data
after an intravenous bolus drug dose.
In Lesson 2, we learned that for most drugs (those following first-order elimina-
tion) a straight line can describe the change in natural log of plasma concentration
over time. Recognizing this relationship, we can now develop mathematical
methods to predict drug concentrations.
Whenever you have a straight line such as that in Figure 3-1, the line is defined
by the equation:
Y = mX + b
where m is the slope of the line and b is the intercept of the y-axis. If you know the
slope and the y-intercept, you can find the value of Y for any given X value.
As the value for the y-intercept may be obtained easily by visual inspection, the
only part of the equation that must be calculated is the slope of the line. A slope
is calculated from the change in the y-axis (the vertical change) divided by the
change in the x-axis (the horizontal change), as in Figure 3-2:
DY Y −Y
slope = or slope = 2 1
DX X2 − X1
where D means “change in.”
31
32

Straight-line plot showing slope and y-intercept. The slope of the natural log of plasma concentration versus
time curve can be determined if two plasma concentrations
The slope is obtained by selecting two different and their corresponding times are known.
points on the line and calculating the difference
between their values. We can apply these same Note that the slope is calculated using ln C1 – ln C0
mathematical principles to the natural log of plasma or ln (C1/C0) and not ln (C1 – C0). The latter would
concentration versus time plot (Figure 3-3). The give an incorrect result. A negative slope indicates
slope is the change in the natural log of plasma that the natural log of concentration declines with
concentrations divided by the change in time increasing time.
between the concentrations: When the natural log of drug concentration is
plotted versus time and a straight line results, as
D In conc In C 1 − In C 0 in the previous example, the slope of that line indi-
slope = or slope =
D time t1 − t0 cates the rate of drug elimination. A steeper slope
(Figure 3-4, top graph) indicates a faster rate of
If, for example, 10 mg/L is the first concentra- elimination than does a flatter slope (Figure 3-4,
tion (C0) drawn immediately after administration bottom graph). For first-order processes, the rate
(t0 = 0 hour) and 1 mg/L is the second concentration of elimination (expressed as the fraction of drug in
(C1) drawn 2.5 hours after administration (t1 = 2.5 the body removed over a unit of time) is the same at
hours): high or low concentrations and is therefore called
an elimination rate constant.
In C 1 − In C 0
slope =
t1 − t0
In 1− In 10 0 − 2.303
= =
2.5 hr − 0 hr 2.5 hr
= −0.92 hr −1
FIGURE 3-2.
The slope of a straight line can be determined from any two FIGURE 3-4.
points on the line. A steeper slope (top) indicates a faster rate of elimination.

Lesson 3 | Half-Life, Elimination Rate, and AUC
33
Therefore, when drug elimination is first Drug concentrations can be predicted using these
order, the negative slope of the natural log of drug mathematical methods instead of the previously
concentration versus time plot equals the drug’s described graphical methods. With mathematical
elimination rate constant: methods, our predictions of drug concentrations
over time are more accurate. So, if the negative slope
slope = – elimination rate constant
of the natural log of drug concentration versus time
or: plot equals the elimination rate constant, our equa-
tion for the line:
– slope = elimination rate constant
Y = mX + b
One must carefully examine the mathematical differ-
ences in positive and negative slope and elimination becomes:
rate constant (K) values as they apply to various
ln (drug concentration) = (– elimination rate constant
dosing equations. The slope value from two plasma
× time) + ln y-intercept
drug concentrations is always a negative number;
however, K can be used in either its positive or nega- To simplify our terminology here, let:
tive form by simple application of one of the rules of
ln C = natural log of drug concentration,
logarithms: Log [A/B] = Log A – Log B.
K = elimination rate constant, and
Remember that the elimination rate constant is
the fraction of drug removed over a unit of time. If t = time after dose.
the elimination rate constant is 0.25 hr–1, then 25% Also, we shall call the y-intercept “ln C0,’’ the drug
of the drug remaining in the body is removed each concentration immediately after a dose is adminis-
hour. tered (at time zero, or t0). Therefore, our equation
Because we know that a plot of the natural log of becomes:
drug concentration over time is a straight line for a In C − In C 0
drug following first-order elimination, we can predict In C = ( − K × t ) + In C 0 or − K =
t
drug concentrations for any time after the dose if we
know the equation for this line. Remember that all This last equation is valuable in therapeutic drug
straight lines can be defined by: monitoring. If two plasma drug concentrations and
the time between them are known, then the elimi-
Y = mX + b
nation rate can be calculated. If one plasma drug
As shown in Figure 3-3: concentration and the elimination rate are known,
Y axis = natural log of drug concentration in then the plasma concentration at any later time can
plasma be calculated. Note that this equation above can
also be expressed to solve for K as a positive value
X axis = time after dose
as shown below:
m = slope of line, or negative elimination
rate constant In C 0 − In C
K =
b = intercept on natural log of plasma t
drug concentration axis (y-intercept)
Note C0 and C have changed locations. Last, either
Now, when we convert to our new terms: version of this equation can now be re-written in a
ln drug concentration = (– elimination rate constant calculator friendly version by applying the log rule,
× time) + ln concentration at y-intercept Log [A/B] = Log A – Log B, yielding
If we know the slope of the line and the intercept In[C 0 /C ] In[C /C 1 ]
of the y-axis, we can predict the natural log of drug =K = or K
t t
concentration at any time after a dose.

34
It is a property of logarithms that:

Clinical Correlate
C1
In C 1 − In C 0 = In
The concepts presented in this lesson can be C0
used to predict plasma concentrations in some
situations. For example, if a patient with renal Then, using numbers from Figure 3-6:
dysfunction received a dose of vancomycin and
plasma concentrations were determined 24 C1  5 mg/L 
In In 
and 48 hours after the dose, then two plasma C0  12.3 mg/L 
concentrations could be plotted on semilog paper
−K = =
t1 − t0 6 hr − 0 hr
to determine when the concentration would reach
10 mg/L (Figure 3-5). This would be approximately So, –K = –0.15 hr –1, or K = 0.15 hr –1.
88 hours after the infusion. This information can
be used to determine when the next dose should In this case, the elimination rate constant is
be given. 0.15 hr –1. This means that 15% of the drug remaining
in the body is removed each hour, so an initial plasma
concentration of 10 mg/L will decrease 15% (0.15 ×
10 mg/L = 1.5 mg/L) to 8.5 mg/L by the end of the
Elimination Rate Constant first hour. By the end of the second hour, the concen-
As stated in the previous section, the elimination tration will be 7.2 mg/L, a 15% reduction from
rate constant (K ) represents the fraction of drug 8.5 mg/L (0.15 × 8.5 mg/L = 1.3).
removed per unit of time and has units of reciprocal The equation ln C = –K t + ln C0 is important
time (e.g., minute–1, hour –1, and day –1). These units because it allows the estimation of the concentra-
are evident from examination of the calculation of tion at any given time. Remember that it is in the
K. For example, in Figure 3-6, C0 is the first plasma form of an equation for a line, Y = mX + b. Remem-
drug concentration measured just after the dose is bering the rule of logarithms that ln Xp = P ln X, if we
given, and C1 is the second plasma drug concentra- take the antilog of each part of this equation, we get:
tion measured at a later time (t1). From our previous
discussion, we know that the equation for this line 3-2 C = C 0 e −Kt
(y = mX + b) is:
where:
In C1 = −K t + In C0 C = plasma drug concentration at time = t,
Furthermore, we know that the slope of the line C0 = plasma drug concentration at time = 0,
equals –K, and we can calculate this slope: K = elimination rate constant,
t = time after dose, and
In C − In C 0
3-1 slope = − K = 1 e = base of the natural log (approximately
t1 − t0
2.718).
e–Kt = percent or fraction remaining after
time (t)
In “plain English,” this equation is saying that a

concentration at some time (C) is equal to some
previous concentration (C0) “time” e–Kt , the frac-
tion of C0 remaining after t hours. To determine the
e–Kt portion on a calculator, enter the value for –Kt
and then the function for ex, or inverse natural log,
or raise 2.718 to the power of the value of –Kt. The
FIGURE 3-5.
Predicting plasma drug concentrations. antilog of a number is equal to e (or 2.718) raised to

35
Half-Life
Another important parameter that relates to the rate
of drug elimination is half-life (T½). The half-life is
the time necessary for the concentration of drug in
the plasma to decrease by one-half. A drug’s half-life
is often related to its duration of action and also may
indicate when another dose should be given.
One way to estimate the half-life is to visually
FIGURE 3-6. examine the natural log of plasma drug concentra-
Determination of a line (log scale) from two known plasma tion versus time plot and note the time required for
drug concentrations. the plasma concentration to decrease by one-half.
For example, in Figure 3-7, the decrease from 10 to
a power equal to that number. The preceding equa- 5 mg/L takes approximately 1.5 hours. It also takes
tion can also be used to predict the concentration at 1.5 hours for the concentration to decrease from 5.0
any time, given an initial concentration of C0 and an to 2.5 mg/L, from 7.0 to 3.5 mg/L, etc. At any point,
elimination rate of K. the decrease in concentration by one-half takes
approximately 1.5 hours, even when the decrease is
from a concentration as low as 0.05 to 0.025 mg/L.
Clinical Correlate Thus, the half-life can be estimated to be 1.5 hours.
If we know that the plasma drug concentration There is another way to estimate the half-life
just after a gentamicin dose is 8 mg/L and the from two known concentrations. Because the half-
patient’s elimination rate constant is 0.25 hr –1, life is the time for a concentration to decrease by
we can predict what the concentration will be one-half, T½ can be estimated by halving the initial
8 hours later: concentration, then taking one-half of that concentra-
tion to get a second concentration, and so on until the
C = C0 e–Kt final concentration is reached. The number of halves
where: required to reach the final desired concentration,
divided into the time between the two concentra-
C0 = 8 mg/L, tions, is the estimated half-life. For example, the
K = 0.25 hr –1, and
t = 8 hours.
–1(8 hr)
Cat 8 hr = 8 mg/L × e–0.25 hr
= 8 mg/L (0.135)
= 1.1 mg/L
Note that the term e–Kt indicates the fraction of

the initial dose of drug that remains in the body
at time t ; 0.135 (or 13.5%) remains in the body
8 hours after the initial dose in this example.
Conversely, the term 1 – e–Kt would indicate the FIGURE 3-7.
percent or fraction excreted after time (t ) . Half-life can be determined from the natural log of plasma
concentration versus time plot.

36
following two concentrations were determined at TABLE 3-1. Example of Half-Life

the times stated after a dose was administered:
Time after Peak
C (mg/L) 8.0 4.0 Concentration (half-life) Plasma Concentration (mg/L)
t (hour) 0 6 0 100
2 50
Because the concentration drops from 8.0 to 4.0 mg/L 4 25
in 6 hours, the half-life is 6 hours. Consider a case
6 12.5
in which concentrations and times were as follows:
8 6.25
C (mg/L) 12.0 3.0 10 3.125
t (hour) 8 12
The concentration drops from 12 mg/L to 3 mg/L in By definition, the concentration (C) at the time
4 hours. To get from 12 to 3 requires a halving of 12 (t) equal to the half-life (T½) is half the original
to 6 and a halving of 6 to 3, representing two half- concentration (C0). Therefore, at one half-life, the
lives in 4 hours, or one half-life of 2 hours. concentration is half of what it was initially. So we
can say that at t =T½ , C = ½C0. For simplicity, let’s
The half-life and the elimination rate constant
assume that C0 = 1. Therefore:
express the same idea. They indicate how quickly
a drug is removed from the plasma and, therefore, In 0.5=
C 0 In C 0 − K (T 1 )
2
how often a dose has to be administered.

If the half-life and peak plasma concentration of = In 1− K (T
In 0.5 1 )
2
a drug are known, then the plasma drug concentra-
tion at any time can be estimated. For example, if Transforming this equation algebraically gives:
the peak plasma concentration is 100 mg/L after
an intravenous dose of a drug with a 2-hour half- K (T 1=
2 ) In 1− In 2
1
life, then the concentration will be 50 mg/L 2 hours

0 − ( −0.693)
after the peak concentration (a decrease by half). T 1
2 =
At 4 hours after the peak concentration, it will have K
decreased by half again, to 25 mg/L, and so on as 0.693
shown in Table 3-1. T 1
2 =
K
Half-life may be mathematically calculated with
the following equation: and
0.693
0.693 K =
3-3 T 1
2 = T 12
K
The equation represents the important relation- Therefore, the half-life can be determined if we
ship between the half-life and the elimination rate know the elimination rate constant and, conversely,
constant shown by mathematical manipulation. We the elimination rate constant can be determined if
already know that: we know the half-life. This relationship between the
half-life and elimination rate constant is important
In C = C0 −Kt in determining drug dosages and dosing intervals.

37
TABLE 3-2. Half-Lives of Common Drugs

Clinical Correlate
Drug Half-Life (hours)
Half-life can be calculated from two plasma Enoxaparin 3.8
concentrations after a dose is given. First, the
Cefazolin 2.2
elimination rate constant (K) is calculated as
shown previously. For example, if a dose of Digoxin 39
gentamicin is administered and a peak plasma Gentamicin 2–3
concentration is 6 mg/L after the infusion is Atorvostatin 20
completed and is 1.5 mg/L 4 hours later, the Lithium 22
elimination rate constant is calculated as follows:
Vancomycin 5.6
In C 1 − In C 0 Levofloxacin 7
K = −
t1 − t0 Source: Brunton LL, Lazo JS, Parker KL (editors), The Pharma-
cologic Basis of Therapeutics, 11th edition. New York: McGraw-
In 1.5 − In 6
= − Hill; 2006. pp. 1800, 1806, 1817, 1821, 1830, 1842, 1843,
4 hr − 0 hr 1883.
−1.39
= −
4 hr
Relationships among Pharmacokinetic
= 0.348 hr −1 Parameters
Then: In previous lessons, we discussed elimination rate,
volume of distribution, and clearance. These impor-
0.693 tant parameters aid in calculating a drug dosage
=
T 12 = 2.0 hr
K regimen. All three relate to how fast a drug effect
will terminate. There are significant relationships
among these parameters, the drug dose, and plasma
drug concentrations. In this lesson, we begin to
explore these relationships so that we can better
Clinical Correlate predict plasma drug concentrations achieved with
The average plasma half-lives of some commonly drug doses.
used drugs are shown in Table 3-2. These may Although clearance is a model-independent
vary considerably between individuals and may pharmacokinetic parameter and is not physi-
be altered by disease. Note that drug effects may ologically dependent only on elimination rate, it is
persist for a period of time longer than would be sometimes useful to relate it to such parameters as
predicted by a drug’s half-life. The greater the the elimination rate constant (K) and the volume of
value of the half-life, the longer the drug stays in distribution (V). Mathematically, systemic clearance
the body. As an example from Table 3-2, half of (Clt ) is related to V and K by:
a dose of vancomycin takes approximately 5.6
hours to be eliminated from the body (no matter Clt /V = K
the size of the dose). Also, half of a dose of
cefazolin is eliminated in approximately 2.2 hours or:
after administration, and so on.
3-4 Clt = V × K

38
Clearance and volume are independent factors With a one-compartment model, first-order elimi-
that together determine K (and T½). Because V has nation, and intravenous drug administration, the
units of volume (milliliters or liters) and clearance AUC can be calculated easily:
has units of volume/time (usually milliliters per
minute), K has units of reciprocal time (minute–1, initial concentration (C 0 )
AUC =
hour–1, or day–1). It is important to understand that elimination rate constant (K )
the elimination rate constant and plasma drug
concentration versus time curve are determined by C0 has units of concentration, usually milligrams
drug clearance and volume of distribution. per liter (mg/L), and K is expressed as reciprocal
Clearance can be related to drug dose by first time (usually hour–1), so the AUC is expressed as
evaluating the plasma drug concentration versus milligrams per liter times hours (mg/L × hr). These
time curve after a dose. In examining this curve units make sense graphically as well because when
(Figure 3-8), we see that there is a definite area we multiply length times width to measure area, the
under the curve, referred to as the area under the product of the axes (concentration in milligrams
plasma drug concentration versus time curve or AUC. per liter, and time in hours) would be expressed as
The AUC is determined by drug clearance and milligrams per liter times hours.
the dose given: AUC can be calculated by computer modeling
of the above AUC equation, or by applying the trap-
dose administered ezoidal rule. The trapezoidal rule method is rarely
3-5 AUC =
drug clearance used, but provides visual means to understand AUC.
If a line is drawn vertically to the x-axis from each
When clearance remains constant, the AUC is measured concentration, a number of smaller areas
directly proportional to the dose administered. If the are described (Figure 3-9). Because we are using
dose doubled, the AUC would also double. Another the determined concentrations rather than their
way to think about this concept is that clearance is natural logs, the plasma drug concentration versus
the parameter relating the AUC to the drug dose. time plot is curved. The tops of the resulting shapes
We usually know the dose of drug being are curved as well, which makes their areas difficult
administered and can determine plasma drug to calculate. The area of each shape can be esti-
concentrations over time. From the plasma concen- mated, however, by drawing a straight line between
trations, the AUC can be estimated and drug adjacent concentrations and calculating the area of
clearance can be determined easily by rearranging the resulting trapezoid (Figure 3-10).
the previous equation to:
dose administered
drug clearance =
AUC

Area under the plasma drug concentration versus time curve. A plasma drug concentration versus time curve can be divided
into a series of trapezoids.

39
Clinical Correlate
The AUC can be used to determine a drug’s
clearance. For an individual patient, when the
same drug dose is given over a period of time
and the volume of distribution remains constant,
changes in clearance can be assessed by
changes in the AUC. For example, a doubling of
the AUC would result if clearance decreased by
FIGURE 3-10. half. For orally administered drugs, this would
Calculation of the area of a trapezoid.
only be true if the fraction of drug absorbed from
the gastrointestinal tract remained constant.
If the time between measurements (and hence
AUC is only rarely used in clinical situations to
the width of the trapezoid) is small, only a slight
determine clearance. It is used more frequently in
error results. These smaller areas can be summed clinical research.
to estimate the AUC, as shown in the following equa-
tion.
 C + C    C + C   To calculate drug clearance, divide drug dose

AUC  2 1  (t 2 − t 1 )  +  3 2  (t 3 − t 2 )  etc.
= by AUC. By knowing how to calculate clearance by
 2    2   the area method, it is not necessary to decide first
which model (i.e., one, two, or more compartments)
To calculate drug clearance, however, we best fits the observed plasma levels.
need the AUC from time zero to infinity, and the
preceding method only estimates the AUC to the
final measured drug concentration.
The terminal part of the AUC is estimated by Clinically Important Equations
dividing the last measured plasma concentration by
the elimination rate constant (Figure 3-11):
C last In C 0 − In C
terminal area = 1. K =
K t
Add the terminal area to the value of AUC from
the preceding equation to find the value of AUC from 2. C = C 0 e −Kt Equation 3-2
zero to infinity.
0.693
3. T 1
2 = Equation 3-3
K
4. Clt = V × K Equation 3-4
FIGURE 3-11.
Terminal area.

40
REVIEW QUESTIONS
3-1. Which of the following is the equation for a 3-4. If two patients receive the same drug and
straight line? the plots in Figure 3-14 result, which patient
A. X = mY + b has the larger elimination rate constant
(faster elimination)?
B. b = mY + X
A. Patient A
C. Y = mX + b
B. Patient B
D. mX +Y = b
3-2. Which of the following would be the slope

(and hence the negative elimination rate
constant) of the straight line in Figure 3-12?
A. ln (C0 – C1)
B. t0 – t1
C. C0 – t0
FIGURE 3-14.
D. (ln C1 – ln C0)/(t1 – t0)
Plasma drug concentration versus time.
3-5. A patient with renal dysfunction received a

dose of vancomycin. Plasma concentrations
were 22 and 15 mg/L at 24 and 48 hours
after infusion, respectively. Plot these two
plasma concentrations on semilog paper
and determine when the concentration
would reach 10 mg/L.
FIGURE 3-12. A. 54 hours
Plasma drug concentration versus time. B. 72 hours
C. 96 hours
3-3. Which of the following is the elimination
D. 128 hours
rate constant for Figure 3-13?
A. –0.173 3-6. Using the equation C = C0e–Kt, determine the
B. 0.52 plasma concentration of a drug 24 hours
C. 0.231 after a peak level of 10 mg/L is observed if
the elimination rate constant is 0.05 hr –1.
D. 0.173
A. 3.01 mg/L
B. 33.2 mg/L
C. 18.1 mg/L
3-7. Which of the following is a proper unit for

the elimination rate constant?
A. minutes
B. mg/minute
C. hr –1
FIGURE 3-13.
D. mg/L

41
3-8. If the elimination rate constant is 0.2 hr –1 3-12. If a drug has an elimination rate constant of
the percent of drug removed per hour is: 0.564 hr–1, what is the half-life?
A. 20%. A. 1.23 hours
B. 1%. B. 0.81 hour
C. 0.1%. C. 1.77 hours
D. 10%.
3-13. To calculate drug clearance by the area
3-9. If the plasma concentration just after a method, it is necessary to first determine
gentamicin dose is 10 mg/L and the patient’s whether the drug best fits a one- or two-
elimination rate constant is 0.15 hr –1, predict compartment model.
what the plasma concentration will be A.
True
8 hours later. B. False
A. 6.0 mg/L
B. 3.0 mg/L 3-14. In the trapezoid shown in Figure 3-16, what
is the area?
C. 1.5 mg/L
A. 150 (mg/L) × hour
D. 1.0 mg/L
B. 300 (mg/L) × hour
3-10. For a drug that has an initial plasma concen- C. 100 (mg/L) × hour
tration of 120 mg/L and a half-life of 3 hours, D. 25 (mg/L) × hour
what would the plasma concentration be
12 hours after the initial concentration?
A. 15 mg/L
B. 112.5
C. 7.5 mg/L
D. 60
3-11. From Figure 3-15, the approximate T½ is:

A. 5 hours.
B. 10 hours.
C. 20 hours. FIGURE 3-16.
Trapezoid.
3-15. If the dose (X0) and AUC are known, the

clearance (area method) is calculated by:
A. AUC/dose.
B. dose/AUC.
C. plasma concentration/AUC.
K/AUC.
D.
FIGURE 3-15.

42
3-16. Using Figure 3-17 and knowing that a 3-3. A. Incorrect answer. The elimination rate
500-mg dose was given intravenously, calcu- constant would not have a negative
late clearance by the area method. value; this is merely the slope of the line.
A. 42 L/hour B. Incorrect answer. You may have added
B. 8.4 L/hour ln 2 and ln 4 rather than subtracted ln 4
from ln 2.
C. 3 L/hour
C. Incorrect answer. You may have used 3
D. 4.2 L/hour
hours in the denominator rather than 4
hours for change in time.
D. CORRECT ANSWER
3-4. A. CORRECT ANSWER. The larger the slope

of the line is (i.e., the steeper the line is),
the larger the elimination rate constant
will be.
B. Incorrect answer. Note that the slope of
the line is smaller (i.e., the line is less
steep).
3-5. A, C, D. Incorrect answers. Be sure your points

FIGURE 3-17. are plotted on paper that has a log scale
Plasma drug concentration versus time. for y (concentration) values. Double-
check the placement of your points.
B. CORRECT ANSWER
ANSWERS
B. Incorrect answer. You may have used Kt
3-1. A, B, D. Incorrect answers
and not –Kt.
C. CORRECT ANSWER. The slope is m, and
C. Incorrect answer. Check the –Kt term,
b is the y-intercept.
which should be –0.05 hr –1 × 24 hours
3-2. A. Incorrect answer. This value should be (or –1.2).
(ln C1 – ln C0) and should be divided by
3-7. A. Incorrect answer. A rate constant is a
the change in time (t1 – t0).
unit change per time expressed as recip-
B. Incorrect answer. The numerator [(ln C1 rocal time units (e.g., minute–1).
– ln C0)] hasn’t been included.
B. Incorrect answer. The elimination rate
C. Incorrect answer constant does not include mass units.
D. CORRECT ANSWER. The slope is the C. CORRECT ANSWER
natural log of change in concentra-
D. Incorrect answer. These are the proper
tion divided by the change in time:
units for concentration.
(ln C1 – ln C0)/(t1 – t0).

43
3-8. A. CORRECT ANSWER 3-14. A. CORRECT ANSWER

B, C, D. Incorrect answers. The elimination rate B. Incorrect answer. You may have neglected
constant is 0.2 hr–1, meaning one fifth (or to divide the sum of 100 plus 50 by 2
20%) per hour. before then multiplying by the width of 2.
C, D. Incorrect answers. Be sure you calcu-
3-9. A, C, D. Incorrect answers lated the height correctly as the average
B. CORRECT ANSWER. C8 hr = C0e–Kt, where of 50 and 100.
C0 = 10 mg/L, K = 0.15 hr–1, and
t = 8 hours. 3-15. A, C, D. Incorrect answers
B. CORRECT ANSWER. Remember, clear-
3-10. A, B, D. Incorrect answers ance has units of volume/time, so the
C. CORRECT ANSWER. C12 hr = C0e–Kt, where units in the equation must result in
C0 = 120 mg/L, K = 0.231 hr–1, and volume/time. Dose/AUC has units of
t = 12 hours. K is calculated from half- mg/(mg/L) × hour, which reduces to
life (K = 0.693/T½). Also, 12 hours L/hour.
represents four half-lives. We would
expect the concentration to decrease 3-16. A, B, C. Incorrect answers
from 120 mg/L to 60 mg/L, then to D. CORRECT ANSWER. To calculate clear-
30 mg/L, 15 mg/L, and finally, to ance, the AUC from time zero to infinity
7.5 mg/L. must be used. The AUC from time zero
to 12 hours can be calculated, and to this
3-11. A, C. Incorrect answers area is added the estimated area from 12
B. CORRECT ANSWER. To find the T½ of hours to infinity. This area is estimated
10 hours, find the interval of time neces- by dividing the drug concentration at 12
sary for the concentration to decrease hours, 1 mg/L, by the elimination rate
from 100 to 50. constant, 0.20 hr–1 (estimated by the
slope of the line between the last two
3-12. A. CORRECT ANSWER points), thereby obtaining an area of
B. Incorrect answer. You may have used 5 (mg/L) × hour from 12 hours to
T½ = K/0.693 rather than T½ = 0.693/K. infinity. Clearance = dose/AUC. Dose is
C. Incorrect answer. You may have used 500 mg. AUC is 119 mg/L, which is the
T½ = 1/0.564 rather than T½ = 0.693/K. sum of 114 mg/L (from 0 to 12 hours)
and 5 mg/L (from 12 hours to infinity).
B. CORRECT ANSWER. When using the
area method, it does not matter if the
drug best fits any particular model.

44
Discussion Points
D-1. Drug X is given by intravenous admin- D-3. Why is the half-life of most drugs the same
istration to two patients. Two plasma at high and low plasma concentrations?
concentrations are then determined, and
the slope of the plasma concentration D-4. The plasma concentration versus time
versus time curve is calculated. Determine curves for two different drugs are exactly
which patient (A or B) has the greater elimi- parallel; however, one of the drugs has
nation rate constant. much higher plasma concentrations. What
can you say about the two drugs’ half-lives?
Patient A Patient B
D-5. For drug X, the AUC determines the inten-
Slope of plasma –0.55 –0.23
sity of drug effect. Explain why a reduction
concentration
versus time curve
of drug clearance by 50% would result in
the same intensity of effect as doubling the
dose.
D-2. Drug X is given to two patients, and two
D-6. Discuss the mathematical consequences of
plasma drug concentrations are then deter-
using a negative versus a positive value for
mined for each patient. Determine which
elimination rate constant (k). How do the
patient has the greater elimination rate
rules of logarithms affect the arrangements
constant.
of the equations used to calculate elimi-
Time after nation rate constant and desired dosing
Dose (hours) Plasma Concentration (mg/L) interval (τ)?
Patient A Patient B
8 22 30
16 5 8

Practice Set 1
The following problems are for your review. Defini- PS1-2. An estimate for the volume of distribution
tions of symbols and key equations are provided here: would be:
X0 = dose administered A. 67 L.
K = elimination rate constant B. 43 L.
V = volume of distribution C. 53 L.
T½ = half-life
t0 = time immediately after drug administration PS1-3. For this same example, the half-life would be:
C0 = concentration of drug in plasma at t0 A. 1.3 hours.

Ct = C0e = concentration of drug in plasma
–Kt B. 1.5 hours.
at any time (t) after drug administration C. 3.1 hours.
AUC = area under plasma concentration versus D. 4.6 hours.
time curve
Clt = total drug clearance from body = dose/AUC The following applies to Questions PS1-4 to PS1-6.
In C 0 − In C 1 An 80-mg dose of drug Y is administered as an
K = intravenous bolus, and the following plasma concen-
t1 − t0
trations result:
QUESTIONS Time after Dose (hours) Plasma Concentration (mg/L)

0 6.7
The following applies to Questions PS1-1 to PS1-3. A 0.5 6.0
1-gram dose of drug X is administered by intravenous 1 5.3
injection, and the following plasma concentrations 2 4.2
result (a one-compartment model is assumed): 4 2.6
8 1.0
Time after Dose (hours) Plasma Concentration (mg/L)
2 15
4 9.5
PS1-4. Using the plasma concentrations at 4 and 8
6 6 hours, K is:
A. 0.118 hr–1.
PS1-1. The plasma concentration at 9 hours after
B. 0.239 hr–1.
the dose estimated from a plot of the points
on semilog graph paper* is: C. 0.478 hr–1.
A. < 0.1 mg/L. D. 0.960 hr–1.
B. 0.7 mg/L. PS1-5. Using the trapezoidal rule, calculate the area
C. 3.1 mg/L. under the curve from 0 to 8 hours (AUC0–8).
D. 4.8 mg/L. Then calculate it from 0 hours to infinity
(∞). Remember: AUC0→8 equals AUC0–8 plus
*Hint: 3-cycle semilog graph paper can often
the area under the curve after 8 hours. This
be found free via an Internet search.
45
46
terminal area is calculated by taking the PS1-6. For this same example, the clearance calcu-
final concentration (at 8 hours) and dividing lated by the area method would be:
by K above.
A. 0.36 L/hour.
A. 4.2 (mg/L) × hour B. 0.78 L/hour.
B. 24.8 (mg/L) × hour C. 1.52 L/hour.
C. 28.9 (mg/L) × hour D. 2.76 L/hour.
D. 53.7 (mg/L) × hour
ANSWERS
PS1-1. A, B. Incorrect answers. You may have used PS1-5. A. Incorrect answer. You may have included
linear graph paper rather than semilog just the area from 8 hours to infinity.
paper. B. Incorrect answer. You may not have
C. CORRECT ANSWER included the area from 8 hours to
Incorrect answer. Be sure your x-scale
D. infinity.
for time is correct and that you extrapo- C. CORRECT ANSWER.
lated the concentration for 9 hours.
For AUC from 0 to 8 hours:
PS1-2. A, C. Incorrect answers
6.7 + 6.0
B. CORRECT ANSWER. First, estimate C0 by (0.5) = 3.18
2
drawing a line back to time = 0 (t0) using
the three plotted points. This should 6.0 + 5.3
(0.5) = 2.82
equal 23 mg/L. Then, V = dose/C0 = 2
1000 mg/23 mg/L = 43 L.
5.3 + 4.2
(1) = 4.75
PS1-3. A. Incorrect answer. You may have calcu- 2
lated the numerator incorrectly. 4.2 + 2.6
(2) = 6.80
B, D. Incorrect answers. You may have used 2
the wrong time interval. 2.6 + 1.0
(4) = 7.20
C. CORRECT ANSWER. Half life = 0.693/K, 2
where: K = (ln 6 – ln 15)/(6 hours
3.18 + 2.82 + 4.75 + 6.80 + =
7.20 24.8 (mg/L) × hr
– 2 hours) = 0.23 hr–1. So, half-life =
0.693/0.23 = 3.1 hours.
For AUC from 8 hours to infinity:
PS1-4. A, C, D. Incorrect answers
1.0 mg/L
B. CORRECT ANSWER =C 8 hr = 4.18 (mg/L) × hr
0.239 hr −1
In 1− In 2.6 0 − 0.96
=K =
8 hr − 4 hr 4 hr Therefore, the AUC from time zero to infinity
equals:
= 0.239 hr −1
24.8 (mg/L) × hr + 4.18 (mg/L)
= × hr 28.93 (mg/L) × hr

Practice Set 1 |
47
Incorrect answer. You may not have

D.
multiplied by 1/2 when calculating the
area from 8 hours to infinity.
PS1-6. A. Incorrect answer. You may have inverted

the formula.
B, C. Incorrect answers
D. CORRECT ANSWER
dose 80 mg
=
clearance =
AUC 28.93 (mg/L) × hr
= 2.76 L/hr

LESSON 4
Intravenous Bolus Administration,

Multiple Drug Administration,
and Steady-State Average
Concentrations
OBJECTIVES
1. Describe the principle of superposition and how it applies to multiple drug dosing.
2. Define steady state and describe how it relates to a drug’s half-life.
3. Calculate the estimated peak plasma concentration after multiple drug dosing (at
steady state).
4. Calculate the estimated trough plasma concentration after multiple drug dosing (at
steady state).
5. Understand the equation for accumulation factor at steady state.
In clinical practice, most pharmacokinetic dosing is performed with one-

compartment, intermittent infusion models at steady state. Using these models,
we can obtain, from population estimates or patient-specific calculation, an elim-
ination rate constant (K) and a dosing interval (τ) based on this K value. Volume
of distribution (V) can likewise be either estimated or calculated from patient-
specific values. So far, our discussion has been limited to a single intravenous
(IV) bolus dose of drug; however, most clinical situations require a therapeutic
effect for time periods extending beyond the effect of one dose. In these situ-
ations, multiple doses of drug are given. The goal is to maintain a therapeutic
effect by keeping the amount of drug in the body, as well as the concentration
of drug in the plasma, within a fairly constant range (the therapeutic range).
In this lesson, we construct equations for predicting drug concentrations after
multiple IV bolus (i.e., IV push) doses. Intermediate equations are used simply to
illustrate the derivation of the final equations that can be applied clinically. Full
understanding of this simpler IV bolus model will aid in the understanding of the
slightly more complicated yet more clinically relevant IV intermittent infusion
equations used later in this book.
49
50
Clinical Correlate
This lesson describes a one-compartment, first-
order, IV bolus pharmacokinetic model. It is used
only to illustrate certain math concepts that will
be further explored with the more commonly
used IV intermittent infusion (i.e., IV piggyback)
models described in Lesson 5. Consequently, read
this IV bolus section only for general conceptual
understanding, knowing that it is seldom applied
clinically. FIGURE 4-2.
Plasma drug concentrations resulting from a second dose.
Intravenous Bolus Dose Model The dosing interval is the time between admin-
Although not used often clinically, the simplest istration of doses. The dosing interval, symbolized
example of multiple dosing is the administration of by the Greek letter tau (τ), is determined by a drug’s
rapid IV doses (IV boluses) of drug at constant time half-life. Rapidly eliminated drugs (i.e., those having
intervals, in which the drug is represented by a one- a short half-life) generally have to be given more
compartment model with first-order elimination frequently (shorter τ) than drugs with a longer
(i.e., one-compartment, first-order model). half-life.
The first dose produces a plasma drug concen- If a drug follows first-order elimination (i.e.,
tration versus time curve like the one in Figure 4-1. the fraction of drug eliminated per unit of time is
C0 is now referred to as Cmax, meaning maximum constant), then plasma drug concentrations after
concentration, to group it with the other peak multiple dosing can be predicted from concen-
concentrations that occur with multiple dosing. trations after a single dose. This method uses the
If a second bolus dose is administered before the principle of superposition, a simple overlay tech-
first dose is completely eliminated, the maximum nique.
concentration after the second dose (Cmax2) will be If the early doses of drug do not affect the
higher than that after the first dose (Cmax1) (Figure pharmacokinetics (e.g., absorption and clearance)
4-2). The second part of the curve will be very similar of subsequent doses, then plasma drug concentra-
to the first curve but will be higher (have a greater tion versus time curves after each dose will look the
concentration) because some drug remains from the same; they will be superimposable. The only differ-
first dose when the second dose is administered. ence is that the actual concentrations may be higher
at later doses because drug has accumulated.
Recall that the y-intercept is called C0 and
the slope of the line is –K. Furthermore, the drug
concentration at any time (Ct) after the first IV bolus
dose is given by:
ln Ct = ln C0 – Kt (See Equation 3-2.)
or
FIGURE 4-1.
Ct = C 0 e −Kt
Plasma drug concentrations after a first dose.

Lesson 4 | IV Bolus Administration, Multiple Drug Administration, & Steady-State Average Concentrations
51
A second IV bolus dose is administered after the

dosing interval (τ), but before the first dose is
completely eliminated. Because Ct = C0e–Kt at any
time (t) after the first dose, it follows that:
C min1 = C max1e −K τ
where Cmin1 is the concentration just before the next

dose is given and τ, the dosing interval, is the time
from Cmax to Cmin.
Cmax2 is the sum of Cmin1 and Cmax1 (Figure 4-3), as
the same dose is given again: FIGURE 4-4.
Increase in Cmax.
=
C max 2 C max1 + C min1
which, by substitution for Cmax2 = Cmax1(1+e−Kτ)e−Kτ.
We showed that: Moreover:
C min1 = C max1e −K τ C= C max 2 + C max1
max 3
so:
−K τ
which, substituting for Cmin2 = Cmax1(1+e−Kτ)e−Kτ+ Cmax1.
C max 2 = C max1+ C max1e This simplifies as follows:
By rearranging, we get: C max1[(1+ e −K τ ) (e −K τ ) + 1]

C max 3 =
C max 2 = C max1(1+ e −K τ ) = C max1[e −K τ + e −2K τ + 1]
A third IV bolus dose can be administered after = C max1[1+ e −K τ + e −2K τ ]

the same dosing interval ( τ ). The plasma drug
concentration versus time profile reveals a further As we can see, a pattern emerges—after any
increase in the maximum concentration immedi- number of dosing intervals, the maximum concen-
ately after the third dose, as shown in Figure 4-4. tration will be:
Just as after the first dose:
C max=
n C max1[1+ e −K τ + e −2K τ +  + e − (n−1)K τ ]
C min2 = C max 2e −K τ
where n is the number of doses given. This equation
can be simplified by mathematical procedures to a
more useful form:
(1− e −n K τ )
C max n = C max1
(1− e −K τ )
where Cmax n is the concentration just after n number

of doses are given. So, if we know Cmax1, the elimina-
tion rate, and the dosing interval, we can predict the
maximum plasma concentration after any number
FIGURE 4-3.
Cmax2 calculation.
(n) of doses.

52
We also know that Cmin n (concentration just For the third IV bolus dose:
before a dose is given) equals Cmax ne−Kτ. Therefore:
−1
(1− e −3(0.05 hr
)8 hr
)
−n K τ
(1− e ) −K τ accumulation factor = −1
C min n = C max1 e (three doses) (1− e − (0.05 hr )8 hr )
(1− e −K τ )
(1− 0.301)
and because Cmax1= X0/V (i.e., dose divided by volume =
(1− 0.670)
of distribution):
X 0 (1− e −n K τ ) −K τ = 2.12
C min n = e
V (1− e −K τ )
Therefore, after two or three doses, the observed
This latter change allows us to calculate Cmin if we peak drug concentration will be 1.67 or 2.12
know the dose and volume of distribution, a likely times the peak concentration after the first dose,
situation in clinical practice. respectively. The concept of accumulation factor is
In each of the preceding equations, the term discussed in more detail in the Accumulation Factor
(1−e−nKτ)/(1−e−Kτ) appears. It is called the accumula- section later in this lesson.
tion factor because it relates drug concentration after These equations will be used later to predict
a single dose to drug concentration after n doses with drug concentrations for given dosage regimens. For
multiple dosing. This factor is a number greater than certain drugs (e.g., aminoglycosides), it is important
1, which indicates how much higher the concentration to predict peak (Cmax) and trough (Cmin) concentra-
will be after n doses compared with the first dose. For tions in various clinical situations.
example, if 100 doses of a certain drug are given to a
patient, where K = 0.05 hr –1 and τ = 8 hours, the accu-
mulation factor is calculated as follows:
Clinical Correlate
−n K τ −100(0.05 hr −1 )8 hr
(1− e ) (1− e )
= −K τ
= 3.03 If a drug has a very short half-life (much less
(1− e )
−1
(1− e − (0.05 hr )8 hr ) than the dosing interval), then the plasma
concentrations resulting from each dose will
This means that the peak (or trough) concentra- be the same and accumulation of drug will not
tion after 100 doses will be 3.03 times the peak (or occur (as shown in Figure 4-5). An example
trough) concentration after the first dose. would be a drug such as gentamicin given
every 8 hours intravenously to a patient whose
The accumulation factor for two or three doses
excellent renal function results in a drug half-
can also be calculated to predict concentrations
life of 1.0–1.5 hours.
before achievement of steady state. Remember:
(1− e −n K τ )
4-1 accumulation factor =
(1− e −K τ )
So for the second IV bolus dose:

−1
(1− e −2(0.05 hr
)8 hr
)
accumulation factor = −1
(two doses) (1− e − (0.05 hr )8 hr )
(1− 0.449)
=
(1− 0.670)
FIGURE 4-5.
= 1.67 Plasma drug concentration versus time.

53
Intravenous Bolus Equations at constant (a shorter half-life), steady state is reached

Steady State sooner than with a lower one (a longer half-life)
(Figure 4-7).
As successive doses of a drug are administered, the Steady state is the point at which the amount
drug begins to accumulate in the body. With first- of drug administered over a dosing interval equals
order elimination, the amount of drug eliminated the amount of drug being eliminated over that same
per unit of time is proportional to the amount of period, and it is totally dependent on the elimina-
drug in the body. Accumulation continues until the tion rate constant. Therefore, when the elimination
rate of elimination approaches the rate of adminis- rate is higher, a greater amount of drug is eliminated
tration: over a given time interval; it then takes a shorter
rate of drug going in = rate of drug going out time for the amount of drug eliminated and the
amount of drug administered to become equivalent
As the rate of drug elimination increases and (and, therefore, achieve steady state). If the half-life
then approaches that of drug administration, the of a drug is known, the time to reach steady state can
maximum (peak) and minimum (trough) concen- be determined. If repeated doses of drug are given
trations increase until equilibrium is reached. After at a fixed interval, then in one half-life the plasma
that point, there will be no additional accumulation; concentrations will reach 50% of those at steady
the maximum and minimum concentrations will state. By the end of the second half-life, the concen-
remain constant with each subsequent dose of drug trations will be 75% of steady state, and so on as
(Figure 4-6). shown in Table 4-1. The plasma concentrations will
When this equilibrium occurs, the maximum increase by progressively smaller increments. For all
(and minimum) drug concentrations are the same practical purposes, steady state will be reached after
for each additional dose given (assuming the same approximately four or five half-lives; the concentra-
dose and dosing interval are used). When the tions at steady state may be abbreviated as Css.
maximum (and minimum) drug concentrations for For a drug such as gentamicin, with a 1- to
successive doses are the same, the amount of drug 4-hour half-life in patients with normal renal func-
eliminated over the dosing interval (rate out) equals tion, steady-state concentration is achieved within
the dose administered (rate in) and the condition of 10–20 hours. For agents with longer half-lives, such
steady state is reached. as digoxin and phenobarbital, however, a week or
Steady state will always be reached after longer may be needed to reach steady state.
repeated drug administration at the same dosing With multiple drug doses (Figure 4-8), steady
interval if the drug follows first-order elimination. state is reached when the drug from the first dose
However, the time required to reach steady state is almost entirely eliminated from the body. At this
varies from drug to drug, depending on the elimi- point, the amount of drug remaining from the first
nation rate constant. With a higher elimination rate dose does not contribute significantly to the total

Multiple-dose drug administration. Steady state is reached sooner with a drug having a shorter
half-life.
54
TABLE 4-1. Percentage of Steady-State TABLE 4-2. Time to Reach Steady State for
Concentration Reached Commonly Used Drugsa
Steady-State Drug Time to Reach Steady State (hours)
Duration of Drug Concentration Enoxaparin 19
Administration (half-lives) Reached (%)
Cefazolin 11
1 50 Digoxin 195
2 75 Gentamicin 10–15
3 87.5 Atorvastatin 100
4 93.75 Lithium 110
5 96.875 Vancomycin 28
6 98.4735 Levofloxacin 35
7 99.25
Calculated from average drug half-lives, Table 3-2.
a
amount of drug in the body. After a single dose, therapeutic effects will begin before steady-state
approximately four or five half-lives are required plasma concentrations are reached. For others (e.g.,
for the body to eliminate the amount of drug equiv- zidovudine or lovastatin), a much longer time period
alent to the one dose. However, at steady state, the than that needed to reach steady state is necessary
amount of drug equivalent to one dose is eliminated for full therapeutic benefits.
over one dosing interval. This apparently faster
elimination is a result of accumulation of drug in
the body. Although the same percentage of drug is Clinical Correlate
eliminated per hour, the greater amount of drug in
Time to achieve steady state is a physiologic
the body at steady state causes a greater amount to
function based solely on the drug’s K or
be eliminated over the same time period. half-life, and the amount of time it takes to
The average times to reach steady state for achieve steady state cannot be increased or
some commonly used drugs are shown in Table decreased. However, administration of a loading
4-2. These values may vary considerably between dose for drugs that take many hours to reach
individuals and may be altered by disease. For some steady state is commonly used to achieve a
drugs (e.g., aspirin, ranitidine, and gentamicin), the concentration within the therapeutic range from
the outset of therapy.
The time to reach steady state is determined by

the drug’s elimination rate constant (K), but what
determines the actual plasma drug concentra-
tions achieved? At steady state, the levels achieved
depend on the drug’s clearance, volume of distribu-
tion, dose, and dosing interval (τ). When equivalent
doses are given, a drug with a low elimination rate
constant and small volume of distribution should
achieve higher steady-state plasma concentrations
than an otherwise similar agent with a high elimina-
FIGURE 4-8. tion rate constant and large volume of distribution.
At steady state, the time required to eliminate one dose of drug In the remainder of this lesson, we examine some
is one dosing interval. aspects of multiple drug dosing.

55
Steady-state concentrations are commonly

increased in two ways:
• Method 1—Increase the drug dose but
maintain the same dosing interval (τ), as
shown in Figure 4-9, which results in wider
fluctuations between the maximum (peak)
and minimum (trough) concentrations after
each dose.
• Method 2—Keep the same dose but give it
FIGURE 4-9.
more frequently, as shown in Figure 4-10,
Dose increase with no change in dosing interval to achieve
which reduces the differences between the higher concentrations.
peak and trough concentrations.
Note that the time to achieve steady state is the
same in both figures.
Accumulation Factor
Equations can describe the plasma concentrations
and pharmacokinetics of a drug at steady state.
Clinical Correlate
Remember, steady state will be reached only after
You may wish to change a patient’s steady-state four or five half-lives.
drug concentrations. For example, the patient Recall that with an IV bolus injection of a drug
is not receiving maximal benefits because the fitting a one-compartment model and first-order
steady-state concentrations are relatively low elimination, the drug concentration at any time (t)
or the steady-state levels are high, causing the after any number of doses (n), not necessarily at
patient to experience toxic effects. Remember steady state, can be described by:
from earlier in this lesson that repeated doses of
drug require approximately four or five half-lives X 0 (1− e −n K τ ) −Kt
C n (t ) = e
to reach steady state. Clinically, this means that V (1− e −K τ )
each time a dose or dosing interval is changed,
four or five half-lives are needed to reach a new
steady state. Of course, a drug with a long half-
accumulation factor
life will require a longer time to achieve the new
steady state than a drug with a relatively short
half-life. For example, Drug A has a half-life of 6 Note the inclusion of the accumulation factor from
hours; if the dose or dosing interval is changed, Equation 4-1 as part of this equation.
steady state will not be reached for 24–30 hours
after the change. If Drug B has a half-life of 3
hours, steady state will be reached in 12–15
hours after a change in the dose or dosing
interval.
In deciding on a specific dosing regimen for

a patient, the goal is to achieve a certain plasma
concentration of drug at steady state. Ideally, peak
FIGURE 4-10.
and trough concentrations will both be within the Dosing interval decrease with no change in dose to achieve
therapeutic range (Figure 4-11). higher concentrations.

56
The equation for Cpeak(n) now becomes the equa-

tion for Cpeak ss and can be written as:
X0  1 
C peak ( steady state) =  (1− e −K τ ) 
V  
We can estimate the minimum or trough concen-

tration at steady state. The trough concentration
FIGURE 4-11. occurs just before the administration of the next
Maintenance of plasma drug concentrations within the dose (at t = τ). In this situation, the general equa-
therapeutic range.
tion for the equation for Cn(t) becomes:
To predict the plasma concentration of a drug X 0  1  −K τ

C trough( steady state) = e
at any time t after n number of doses, we therefore V  (1− e −K τ ) 
need to know four values:
• drug dose (X0),
• volume of distribution (V), Clinical Correlate
• elimination rate constant (K), and In most clinical situations, it is preferable to wait
• dosing interval (τ). until a drug concentration is at steady state before
obtaining serum drug concentrations. Use of
If we wish to predict the steady-state peak steady-state concentrations is more accurate and
concentration immediately after an IV bolus dose, makes the numerous required calculations easier.
where t = 0 and e–0 = 1, the equation above for Cn(t)
becomes:
Note the similarity between the equations for
− nK τ
X 0 (1− e ) Cpeak(steady state) and Ctrough(steady state). The expression for
C peak ( n ) =
V (1− e −K τ ) Ctrough(steady state) simplifies to Cpeak(steady state) times e–Kt.
An almost identical equation (below) can be used
because time after the dose equals zero (t = 0, and to calculate the concentration at any time after the
e–0 = 1). peak. The only difference is that t is replaced by the
As multiple drug doses are administered and time elapsed since the peak level. Therefore:
n becomes sufficiently large (> 4 or 5 doses), n
C(t) = Cpeak(steady state)e–Kt
increases and approaches infinity (abbreviated as
n→∞). The preceding equation can then be simpli- where t is the time after the peak.
fied. As n becomes a large number, e–nKτ approaches
This last relationship is very useful in clinical
e–∞, which approaches zero, so 1 – e–nKτ approaches
pharmacokinetics. It is really the same as an equa-
1. As 1 – e–nKτ approaches 1, the value of this numer-
tion presented earlier. (See Equation 3-2.)
ator becomes 1, and the resultant numerator/
denominator combination is termed the accumula- Ct = C0e–Kt
tion factor at steady state:
The preceding equation, stated in words, means
1 a concentration at any time (Ct) is equal to some
4-2
(1− e −K τ ) previous concentration (C0) multiplied by the frac-
tion (or percent) of that previous concentration
(i.e., e–Kt) remaining after it has been allowed to be
eliminated from the body for a number of hours
represented by t.

57
If two drug concentrations and the time between

them are known, K can be calculated. If one concentra-
tion after a dose (e.g., a peak concentration) and K are
known, then other concentrations at any time after a
dose (but before the next dose) can be estimated.
Average Steady-State Concentration

with Intravenous Bolus Dosing
We now have examined both the maximum and FIGURE 4-12.
Average plasma drug concentration at steady state.
minimum concentrations that occur at steady state.
Another useful parameter in multiple IV dosing situ- It is important to recognize from the equations
ations is the average concentration of drug in the that C at steady state is determined by the clear-
plasma at steady state (C ) (Figure 4-12). Because ance and drug dose (dose/τ). If the dose remains
C is independent of any pharmacokinetic model, it the same [n = a time period such as a day (e.g.,
is helpful to the practicing clinician (model assump- 80 mg every 8 hours [80 × 3] or 120 mg every
tions do not have to be made). C is not an arithmetic 12 hours [120 × 2])] while τ is changed, C would
or geometric mean. remain the same. Also, changes in V or K that are not
Several mathematical methods may be used to related to a change in clearance would not alter C .
calculate the average drug concentration, but only With multiple drug dosing at steady state, changes
one is presented here. A plasma drug concentra- in τ, K, or V (with no change in clearance) would
tion versus time curve, after steady state has been alter the observed peak and trough drug concentra-
achieved with IV dosing, is illustrated in Figure tions but not C .
4-13. By knowing the dose given (X0) and the dosing In dealing with such equations, it is helpful to
interval (τ), we can determine the average concen- remember that the units of measure on both sides
tration if we also know the area under the plasma must be the same. For example, in the equation
drug concentration versus time curve (AUC) over τ. above, C should be in micrograms per milliliter,
Therefore: milligrams per liter, or similar concentration units.
AUC Therefore, the right side of the equation must have
C = the same units, as is the case when:
τ
• dose is in a consistent mass unit, such as
and since:
milligrams,
dose • clearance is in liters per hour or milliliters
AUC = per minute, and
drug clearance
• dosing interval is in hours.
dose
C =
drug clearance × τ
The equation:
dose
4-3 C =
Clt × τ
is very useful, particularly with drugs having a long
half-life, in which the difference between peak and FIGURE 4-13.
trough steady-state levels may not be large. AUC for one dosing interval.

58
So, dose/(Cl × τ) has the following units: Then we insert the known Cpeak, K, X0, and τ
values in the equation for Cpeak. By rearranging the
amount
equation to isolate the only remaining unknown
(volume/time) × time variable, we can then use it to calculate V:
Then, as both hour terms cancel out, we see that
X0  1 
amount per volume (concentration) is left. V =
C peak ( steady state)  (1− e −K τ ) 

Predicting Steady-State Concentration

Now we know the values of all the variables in
The equation for Cpeak(steady state) derived above (and the equation (V, K, Cpeak, X0, and τ) and can use
shown below) is valuable because it allows us to this information to calculate a new Cpeak if we
predict the peak plasma concentration achieved change the dose (e.g., if the previous Cpeak is too
when a drug is given in a specified dose (X0) at a high or too low). For example, if we want the
consistent and repeated interval (τ). To predict peak level to be higher and wish to calculate the
peak concentration at steady state, however, we also required dose to reach this new peak level, we
must have an estimate of the elimination rate (K) can rearrange our equation:
and the volume of distribution (V); therefore, the
V × C peak ( steady state) (1− e −K τ )
X0 =
following equation is used only for IV bolus dosing:
X0  1  and substitute our calculated V and K and the

C peak ( steady state) =
V  (1− e −K τ ) 

desired Cpeak. Or we can choose a new dose (X0) and
calculate the resulting Cpeak by inserting the calcu-
It is possible to estimate a patient’s K and V from lated K and V with τ into the original equation:
published reports of similar patients. For example,
X0  1 
most patients with normal renal function will C peak ( steady state) =  (1− e −K τ ) 
have a gentamicin V of 0.20–0.30 L/kg and a K of V  
0.035–0.2 hr–1. In a clinical setting in which a drug
is administered and plasma concentrations are then Remember that each time we calculate a peak
determined, it is possible to calculate a patient’s plasma level (Cpeak), the trough plasma level also
actual K and V using plasma concentrations. Such can be calculated if we know K and τ:
calculations can be performed as follows.
Ctrough = Cpeake–Kτ
Example 1
If the dosing interval is not changed, new doses
A patient receives 500 mg of drug X intra- and concentrations are directly proportional if
venously every 6 hours until steady state is nothing else changes (i.e., K or V).
reached. Just after the dose is administered,
So,
a blood sample is drawn to determine a peak
plasma concentration. Then, 5 hours later, a C peak ( steady state)new
=X 0 (new) × X 0 (old)
second plasma concentration is determined. C peak ( steady state)old
Using the two plasma concentrations, we first
calculate K, as described previously:
and,
In C peak − In C 5hr X 0 (new)
K = =
C peak ( steady state)new × C steady state( old)
5 hr X 0 (old)

59
Clinically Important Equations

X0  1 
1. C peak ( steady state) =  (1− e −K τ ) 
V  
This equation is developed from the equation in

Lesson 1:
C = X/V
by attaching the steady-state accumulation
factor.
2. Ctrough = Cpeake–K τ
This equation is similar to Equation 3-2 in

Lesson 3.
dose
3. C = Equation 4-3
Clt × τ

60
REVIEW QUESTIONS
4-4. When multiple drug doses are given and
steady state is reached, the amount of drug
eliminated during one dosing interval (τ) is
equal to the drug dose.
A.
True
B. False
4-5. A drug with a relatively small K (long T½ )

takes a longer time to reach steady state
FIGURE 4-14. than a drug with a large K.
A.
True
Note: Refer to Figure 4-14 when answering ques- B. False
tions 4-1 through 4-3.
4-6. If a drug with a T ½ of 12 hours is given
every 6 hours and a peak concentration at
4-1. If Cmax1, K, and τ are 100 mg/L, 0.40 hr–1, and
steady state is 10 mg/L, what will be the
6 hours, respectively, what is the value of
approximate peak concentration just after
Cmax2?
the fifth dose is administered?
A. 109.1 mg/L
A. 5 mg/L
B. 9.1 mg/L
B. 7.5 mg/L
C. 100 mg/L
C. 10 mg/L
D. 110 mg/L
4-7. A 100-mg dose of drug X is given to two
4-2. For the example given in the previous ques- different patients every 8 hours. Which
tion, what is the value of Cmin2? patient (A or B) is likely to achieve higher
A. 9.1 mg/L steady-state plasma concentrations?
B. 43.6 mg/L
Elimination Rate Volume of
C. 85.3 mg/L Patient (hr –1) Distribution (L)
D. 9.9 mg/L A 0.2 10
B 0.4 20
4-3. What is the maximum concentration after

15 doses if the dose (X0) is 800 mg and the
volume of distribution (V) is 20 L? Assume A. Patient A
that τ equals 6 hours and K equals 0.50 hr–1. B. Patient B
A. 42.1 mg/L
4-8. Decreasing the dosing interval while
B. 66.7 mg/L
keeping the dose constant will result in
C. 52.9 mg/L lower steady-state concentrations.
D. 156.8 mg/L A.
True
B. False

61
4-9. Which of the following dosage techniques 4-13. A patient receives an antimicrobial dose of
results in the greatest difference between 400 mg IV every 8 hours. After steady state
maximum (peak) and minimum (trough) is reached, a peak level of 15 mg/L is determined;
concentrations after a dose? the level 4 hours after the peak is 4.5 mg/L. What
A. Small doses given at a short dosing dose is required to attain a peak plasma
interval level of 35 mg/L? (Assume IV bolus drug
administration.)
B. Large doses given at a long dosing
interval A. 400 mg
B. 800 mg
4-10. What is the peak drug X concentration C. 933 mg
attained at steady state if 100 mg is given
D. 3108 mg
by IV injection every 6 hours, the patient’s
K = 0.35 hr–1, and V = 20 L? (Assume a one-
4-14. For the example given in the last question,
compartment distribution.)
when the peak plasma level is 35 mg/L,
A. 3.4 mg/L what will the trough plasma level be?
B. 5.7 mg/L A. 2.3 mg/L
C. 16.3 mg/L B. 3.2 mg/L
D. 41 mg/L C. 4.8 mg/L
D. 32 mg/L
4-11. What would be the trough level for the
example in question 4-10?
A. 0.41 mg/L
ANSWERS
B. 0.7 mg/L
C. 2 mg/L 4-1. A. CORRECT ANSWER
D. 5 mg/L B. Incorrect answer. This value is the
minimum concentration after the first
4-12. A 500-mg dose of drug X is given every 6 dose. Remember to add the value of
hours until steady-state levels are reached. Cmax1, which was 100 mg/L.
At steady state, the AUC for one dosing Incorrect answer. This is the value of
C.
interval is 42 (mg/L) × hour. What is the Cmax1. Cmax2 is calculated as the sum of
average concentration over that dosing Cmax1 and Cmin1.
interval?
Incorrect answer. This is close to the
D.
A. 3.1 mg/L Cmax2 but is actually the steady-state Cmax.
B. 7 mg/L Cmax2 is calculated as the sum of Cmax1 and
C. 12.5 mg/L Cmin1.
D. 22 mg/L

62
4-2. A. Incorrect answer. This is Cmin1. 4-7. A. CORRECT ANSWER. Higher concentra-
B, C. Incorrect answers tions would result with a lower K and V.
D. CORRECT ANSWER. Cmin2 can be found Incorrect answer
B.
from Cmax2 as follows: Cmin2 = Cmax2 (e–Kt), so
Cmin2 = 109.1 mg/L(e–0.4/hr × 6 hr) = 109.1 × 4-8. A. Incorrect answer
0.091 = 9.9 mg/L. B. CORRECT ANSWER. By decreasing the
dosing interval the amount of drug
4-3 A. CORRECT ANSWER. administered per unit of time will
increase and steady-state concentra-
(1− e −n K τ )
C max n = C max1 tions will increase.
1− e −K τ
X 0 800 mg 4-9. A. Incorrect answer. A small dose given
=
C max1 = very frequently results in a smaller
V 20 L change from peak to trough concentra-
Then tions.
B. CORRECT ANSWER
 800 mg   (1− e −15(0.5 hr )6 hr ) 
−1
C max 15 =  
−1
 20 L   (1− e − (0.5 hr )6 hr )  4-10. A, C, D. Incorrect answers
B. CORRECT ANSWER. The peak concen-
C max15 = 42.1 mg/L tration is calculated as follows:
B, C, D. Incorrect answers C peak X 0 /V (1/ [1− e −K τ ])

=
−1
4-4. A. CORRECT ANSWER. When steady state = 100 mg/20 L(1/ [1− e −0.35 hr × 6 hr
])
is reached, the amount of drug elimi-
= 5.7 mg/L
nated over one dosing interval is equal
to the dose.
Incorrect answer
B. 4-11. A, C, D. Incorrect answers
4-5. A. CORRECT ANSWER. The half-life B. CORRECT ANSWER. The trough concen-
directly relates to the time required to tration is calculated as follows:
reach steady state. Approximately five = C peak × e −K τ
C trough
half-lives are required to reach steady
state. A longer half-life (lower K) will
−1
= 5.7 mg/L × e −0.35 hr ×6 hr
mean that more time is required to

reach steady state. = 0.7 mg/L
Incorrect answer
B.
In this case, the elapsed time t is equal
4-6. A, C. Incorrect answers to τ.
B.
CORRECT ANSWER. Administration of 4-12. A, C, D. Incorrect answers
five doses would take 24 hours, which
B.
CORRECT ANSWER. The average plasma
is two drug half-lives. After one half-life,
concentration is determined as follows:
the peak concentration would be 50%
of steady-state concentration; at two =
C AUC/τ
half-lives, it would be 75%. So the peak
concentration just after the fifth dose = 42 (mg/L) × hr/6 hr
would be approximately 7.5 mg/L.
= 7 mg/L

63
4-13. A. Incorrect answer. Giving the same dose 4-14. A, C, D. Incorrect answers
would result in the same peak concen- B.
CORRECT ANSWER. To answer this
tration of 15 mg/L. question, K must first be calculated:
Incorrect answer. Doubling the dose
B.
would result in a doubling of the steady- K =(In C 4 hr − In C peak )/4 =0.3 hr −1
state peak concentration to 30 mg/L.
Then, use the following to calculate the
C. CORRECT ANSWER trough plasma concentration:
Incorrect answer. This dose would result
D.
−1
in a steady-state peak concentration of C trough = C peak × e −K = 35 mg/L × e −0.30 hr ×8 hr
117 mg/L.
= 35 mg/L × 0.091= 3.2 mg/L

64
Discussion Points
D-1. Explain why, for most drugs, the increase in D-4. The peak plasma concentration achieved after
drug plasma concentrations resulting from the first IV dose of drug X is 25 mg/L. The
a single dose will be the same magnitude drug’s half-life is 3.5 hours, and it is adminis-
whether it is the first or tenth dose. tered every 12 hours. What will be the peak
plasma concentration at steady state?
D-2. Explain why the plasma concentrations
(maximum or minimum) remain the same D-5. Discuss why the equations for the IV bolus
for each dose after steady state is reached. model may not be relevant in clinical practice.
D-3. Explain why changing the dose or the dosing D-6. Discuss the advantages and disadvantages
interval does not affect the time to reach of using the one-compartment first-order
steady state. model before steady state is attained.

LESSON 5
Relationships of
Pharmacokinetic Parameters
and Intravenous Intermittent
and Continuous Infusions
OBJECTIVES
1. Explain the relationships of pharmacokinetic parameters and how changes in each
parameter affect the others.
2. Describe the relationship between the rate of continuous intravenous (IV) drug
infusion, drug clearance, and steady-state plasma concentration.
3. Calculate plasma drug concentrations during and after continuous IV infusion.
4. Calculate an appropriate loading dose to achieve therapeutic range at onset of
infusion.
5. Calculate peak and trough concentrations at steady state after intermittent IV
infusions.
Relationships of Pharmacokinetic Parameters

Understanding the relationships of pharmacokinetic parameters is important to
determine what will occur to the plasma concentration versus time curve when
changes in any of the parameters arise. If we administer multiple IV doses of a
drug that exhibits one-compartment, first-order elimination kinetics, we might
find a plasma drug concentration versus time curve that resembles Figure 5-1. A
thorough understanding of the basic components of the IV bolus and continuous
infusion model will lead to a better understanding of the more commonly used
steady-state IV intermittent model equations. Consequently, study this lesson with
the knowledge that many of these equations will later be combined and changed
to yield the final, more commonly used dosing equations as shown in the cases
presented later.
Changes in Elimination Rate Constant

If the dose, the volume of distribution, and the dosing interval (τ) all remain the
same but the elimination rate constant (K) decreases (as with decreasing renal or
65
66

Plasma drug concentrations after multiple intravenous doses. Effect of decreased τ on plasma drug concentrations.
hepatic function), the curve should change as shown allowed less time to eliminate drug before receiving the
in Figure 5-2. With a lower K, we would see that: next dose). Because K (and therefore T½) is the same,
1. Peak and trough concentrations at steady the time to reach steady state remains unchanged.
state are higher than before. Changes in Dose
2. The difference between peak and trough Now, suppose that K, V, and τ remain constant but
levels at steady state is smaller because the the dose (X0) is increased. The plasma concentra-
elimination rate is lower. tion versus time curve shown in Figure 5-4 would
Because K is decreased in this situation, the half- result. The drug concentrations at steady state are
life (T½) is increased and, therefore, the time to higher, but there is no difference in the time required
reach steady state (5 × T½) is also lengthened. This to reach steady state, as it is dependent only on T½
concept is important in designing dosing regimens (and K).
for patients with progressing diseases of the primary With some drugs, it is preferable to give a smaller
organs of drug elimination (kidneys and liver). dose at more frequent intervals; with other drugs,
the reverse is true. The disadvantage of larger, less
Changes in Dosing Interval frequent dosing is that the fluctuation from peak to
For another example, suppose everything, including trough concentrations is greater. Thus, the possi-
the elimination rate, remains constant but the dosing bility of being in a toxic range just after a dose is
interval (τ) is decreased. The resulting plasma drug given and in a subtherapeutic range before the
concentration versus time curve would be similar to next dose is given is also greater. The problem with
that in Figure 5-3. The peak and trough concentra- smaller, more frequent doses is that such adminis-
tions at steady state are increased. Also, the difference tration may not be practical, even though plasma
between peak and trough plasma concentrations concentrations may be within the therapeutic range
at steady state is smaller (only because the body is for a greater portion of the dosing interval.

Effect of decreased K (and therefore increased T½) on plasma Effect of increased dose on plasma drug concentrations.
drug concentrations.

Lesson 5 | Relationships of Pharmacokinetic Parameters & IV Intermittent & Continuous Infusions
67
Changes in Clearance and Volume of

Distribution
A drug’s half-life and elimination rate constant are
determined by its clearance and volume of distribu-
tion (discussed in Lessons 2 and 3). These last two
pharmacokinetic parameters determine the plasma
drug concentrations that result from a dosing
regimen, so changes in clearance or volume of distri-
bution result in changes in steady-state plasma drug
concentrations. FIGURE 5-5.
Effect of changes in volume of distribution on plasma drug
Volume of distribution and clearance may
concentrations.
change independently. However, some disease
states may alter both the clearance and the volume The effect of volume of distribution changes on
of distribution. An example is the effect of renal plasma drug concentrations can be easily estimated
failure on aminoglycoside concentrations. The renal for most drugs. When the volume of distribution
clearance of aminoglycosides decreases in patients increases, assuming there are no other changes,
with renal failure, and the volume of distribution peak steady-state plasma drug concentrations
may increase because of the fluid accumulation that decrease. Conversely, if the volume of distribution
occurs with oliguric renal failure. decreases, peak steady-state plasma drug concen-
There are a number of conditions that may trations increase. This can also be demonstrated by
increase or decrease volume of distribution. The the following equation:
volume of distribution of drugs that distribute C ss = K 0 / KV
primarily in body water increases in patients with
conditions that cause fluid accumulation (e.g., renal
X0 /t
failure, heart failure, liver failure with ascites, and
inflammatory processes such as sepsis). As one
would expect, dehydration results in a decreased where K0 = the rate of drug infusion (or administra-
volume of distribution for drugs of this type. Drugs tion). (Note: This equation is derived in the section
that are highly bound to plasma protein (such as Continuous Infusion later in this lesson.)
phenytoin) have a greater volume of distribution There are a number of conditions that may
when protein binding is decreased by hypoalbu- increase or decrease drug clearance. Agents that
minemia or phenytoin-displacing agents. If fewer change renal blood flow directly affect the clearance
proteins are available for binding, then to main- of drugs excreted by the kidneys. Renal clearance
tain equilibrium with the tissues, free drug moves may decrease when agents that compete for active
from the plasma to the tissues, thus increasing the renal secretion are administered concomitantly
“apparent” volume of distribution. (such as penicillin with probenecid). For drugs that
Changes in the volume of distribution directly are eliminated hepatically, clearance may be altered
affect steady-state plasma drug concentrations. In by drugs or conditions that increase or decrease
general, if the drug dose, dosing interval (τ), and liver blood flow. Some conditions (such as hepatitis
drug clearance are all unchanged but the volume or cirrhosis) also may decrease the capability of
of distribution decreases, there will be greater fluc- liver enzymes to metabolize drugs. Drug clearance
tuation of plasma concentrations with higher peak may increase when organ function improves after
concentrations. Conversely, if the volume of distribu- healing, with concomitant drug administration, or
tion increases, there will be less fluctuation of plasma under conditions that increase organ blood flow or
concentrations with a lower peak (Figure 5-5). the activity of metabolic enzymes.

68
Clinical Correlate
Two conditions that may substantially alter the
volume of distribution are severe traumatic or
burn injuries. Severely traumatized or burned
patients often have a cytokine-induced, systemic
inflammatory response syndrome (SIRS), which
results in decreased plasma proteins (i.e., albumin)
and thus an accumulation of fluid in tissues. An
average-weight person (70 kg) may gain as much
FIGURE 5-6. as 20 kg in fluid over a few days. In comparison
Effect of changes in clearance on plasma drug to the extra fluid, the body has decreased albumin
concentrations. for binding, and with the accumulation of fluid
due to this SIRS, free drug shifts from the plasma
Changes in drug clearance affect steady-state into the extravascular fluid, causing drugs that are
plasma drug concentrations. If the dose, dosing primarily distributed into body water to have an
interval, and the volume of distribution are all increased volume of distribution.
unchanged but clearance increases, plasma drug
concentrations will decrease because the drug is
Continuous Infusion
being removed at a faster rate. Conversely, if clear-
ance decreases, plasma concentrations will increase The remainder of this lesson describes the contin-
because the drug is being removed at a slower rate uous infusion model and then shows how it can
(Figure 5-6). be combined with the IV bolus model, previously
described, to yield the commonly used IV intermit-
This can also be demonstrated by the modifica-
tent infusion model (i.e., IV piggyback). As stated
tion of the equation presented above:
earlier, repeated doses of a drug (i.e., intermittent
C = K / Cl infusions) result in fluctuations in the plasma concen-
ss 0 t
tration over time. For some drugs, maintenance of a
consistent plasma concentration is advantageous
KV because of a desire to achieve a consistent effect.
To maintain consistent plasma drug concentrations,
where K0 = the rate of drug infusion and Clt = total continuous IV infusions are often used. Continuous
body clearance. (Note: This equation is also derived IV infusion can be thought of as the administration
in the section Continuous Infusion.) of small amounts of drug at infinitely small dosing
As with volume of distribution, the effect of intervals. If administration is begun and maintained
changes in clearance on plasma drug concentrations at a constant rate, the plasma drug concentration
can be easily estimated for most drugs. For example, versus time curve in Figure 5-7 will result.
if drug clearance increases by a factor of two, the The plasma concentrations resulting from the
average steady-state plasma drug concentration continuous IV infusion of drug are determined by
decreases by half. Conversely, if drug clearance the rate of drug input (rate of drug infusion, K0),
decreases by half, the average steady-state plasma volume of distribution (V), and drug clearance (Clt ).
drug concentration would increase by a factor of The relationship among these parameters is:
two.
K0
=
Ct (1− e − Kt )
VK

69
drug such as theophylline given by continuous IV

infusion, the average half-life in adults is approxi-
mately 7 hours. Therefore, it will take 35 hours (5 ×
7 hours) to reach approximate steady-state plasma
concentrations.
When steady state is achieved, the factor e–nKt
(see Lesson 4) approaches zero, and thus the factor
(1 – e–Kt ) equals 1, and then:
K0 K0 K
C ss= (1− e −∞=) (1− 0) − 0
Clt Clt Clt
FIGURE 5-7.
Plasma drug concentrations over time with a continuous
intravenous infusion. At steady state, the plasma concentration of
drug is directly proportional to the rate of admin-
where t is the time since the beginning of the drug
istration (assuming clearance is unchanged). If
infusion. This equation shows that the plasma
the infusion is increased, the steady-state plasma
concentration is determined by the rate of drug infu-
concentration (Css) will increase proportionally.
sion (K0) and the clearance of drug from the body
Clearance is the pharmacokinetic parameter that
(remember, VK = Clt ). The equation is used to find a
relates the rate of drug input (dosing or infusion
concentration at a time before steady state is reached.
rate) to plasma concentration. The actual plasma
The term (1 – e–Kt) gives the fraction of steady- concentration attained with a continuous IV infu-
state concentration achieved by time t after the sion of drug depends on the following two factors:
infusion is begun. For example, when t is a very low
number just after an infusion is begun, K0(1 – e–Kt) 1. rate of drug infusion (K0) and
is also very small. When t is very large, (1 – e–Kt) 2. clearance of the drug (Clt).
approaches 1, so K0(1 – e–Kt ) approaches K0 and If we know from previous data that a patient
plasma concentration approaches steady state. receives IV theophylline (or aminophylline), which
Suppose that a drug has a half-life of 8 hours (then has a half-life of 6 hours (K = 0.116 hr–1) and a
K = 0.087 hr–1). Table 5-1 shows how the factor (1 – volume of distribution of 30 L (clearance then
e–Kt) changes with time. When (1 – e–Kt) approaches equals 3.48 L/hour), we can predict the steady-state
1 (at approximately five half-lives), steady-state plasma concentration for a continuous IV theophyl-
concentrations are approximately achieved. line infusion of 40 mg/hour:
In Figure 5-7, steady state is attained where
the horizontal portion of the curve begins. With a K0 K0 40 mg/hr
C ss= (1− e −∞=) (1− 0) − = 11.5 mg/L
Clt Clt 3.48 L/hr
TABLE 5-1. Changes in Factor (1 – e–Kt ) Over Time
Time after Starting Value of Drug Half-Lives If we wish to increase the steady-state theoph-
Infusion (hours) (1 – e–Kt ) Elapsed ylline plasma concentration to 14 mg/L, we would
4 0.29 0.5 use the same equation to determine K0:
8 0.50 1.0
16 0.75 2.0 K0
24 0.88 3.0 14 mg/L =
30 L × 0.116 hr −1
40 0.97 5.0
60 0.99 7.5 K=
0 (14 mg/L)(30 L × 0.116 hr −1=) 48.7 mg/hr

70
If this infusion is continued, the steady-state

concentration would be:
K 0 50 mg/hr
C=
ss = = 11.1 mg/L
Clt 4.5 L/hr
Remember that with a continuous infusion, the

steady-state plasma concentration is determined by
the rate of drug going into the body (K0) and drug
FIGURE 5-8. clearance from the body (Clt ). At steady state, the
Plasma drug concentrations over time with a continuous
amount of drug going into the body per hour equals
intravenous infusion.
the amount of drug being removed per hour.
Or, as concentration and infusion rate are directly
You have learned that it takes approximately
proportional, the following equation may be used to
five drug half-lives to reach steady state. Each time
find a new infusion rate to obtain a desired steady-
state concentration: the infusion rate is changed, five half-lives will be
required to attain a new steady-state concentration.
C ss (desired) For example, for a patient receiving IV theophylline
=
K 0(new) × K 0(original)
C ss (measured) at 20 mg/hour, the steady-state plasma concentra-
tion is 7.5 mg/L, and 25 hours is required to reach
14 mg/L steady state (T½ = 5 hours, K = 0.139 hr–1). If the
= × 40 mg/hr
11.5 mg/L infusion rate is increased to 40 mg/hour, an addi-
tional 25 hours will be required to attain the new
= 48.7 mg/hr
steady-state concentration of 15 mg/L (Figure
With the continuous IV infusion method of drug 5-9). If a dosing rate is changed, it takes one half-
administration, it is sometimes necessary to predict life to reach 50% of the difference between the old
drug plasma concentrations at times other than at concentration and the new, two half-lives to reach
steady state. In the following section, we examine
75% of the difference, three half-lives to reach
some of these situations (Figure 5-8).
87.5%, etc.
The equation predicting plasma concentrations
If we wish to calculate the plasma concentra-
with continuous IV infusion can be used to estimate
plasma drug concentrations at times before steady tion before the new steady state is achieved, we can
state, as stated previously in the lesson: use the factor given before: (1 – e–Kt), where t is the
time after beginning the new infusion rate and the
K0
=
Ct (1− e − Kt ) resulting fraction is the relative “distance” between
VK the old and new steady-state concentrations. For
(Remember, VK = Clt)
For example, if Cl t for a drug is known to be
4.5 L/hour (with K = 0.15 hr–1) and this drug is given
at a rate of 50 mg/hour, then the plasma concentra-
tion 8 hours after starting the infusion would be:
50 mg/hr −1
=C 8 hr (1− e − (0.15 hr )(8 hr ) )
4.5 L/hr
50 mg/hr
= (0.70)
4.5 L/hr FIGURE 5-9.
Changing plasma drug concentrations with increased drug
= 7.8 mg/L infusion rate.
71
the example above (where K = 0.139 hr–1), 8 hours

after the infusion rate is increased:
−1
(1− e −Kt ) =
1− e( −0.139 hr )(8 hr )
= 0.67
So at 8 hours, the concentration would be approxi-
mately two-thirds (67%) of the way between 7.5 FIGURE 5-10.
and 15.0 mg/L (about 12.5 mg/L). Plasma drug concentrations after discontinuation of an
If an infusion is stopped before steady state is intravenous infusion.
reached, the concentration can be determined as immediate effect of a drug is desired in a patient.
follows: In these situations, a loading dose is often admin-
Ct = (K0/Clt )(1−e−Kt ) istered at the initiation of the infusion to achieve
an immediate therapeutic plasma concentration of
where t = the duration of the infusion. the drug. By doing so, a serum concentration within
therapeutic range of the drug is maintained from
Another important situation occurs when a
the outset of therapy. This loading dose is usually
continuous infusion is stopped after steady state
relatively large and may produce immediate thera-
is achieved. To predict plasma drug concentrations
peutic plasma concentrations (Figure 5-11). Note
at some time after the infusion is stopped (Figure
that a loading dose should not be used if substantial
5-10), the concentration at steady state (Css ) is
side effects occur with large doses of the drug. Also,
treated as if it were a peak concentration after an
sometimes clinicians prefer that drugs accumulate
IV injection (C0). In this situation, plasma concentra-
slowly rather than achieve therapeutic concentra-
tions after C0 are predicted by:
tions immediately so that the patient may have
adequate time to develop tolerance to the initial
Ct = (C 0 e − Kt ) (See Equation 3-2.)
side effects (e.g., tricyclic antidepressants).
where t in this case is time after C0, which is the time The desired loading dose for many drugs can be
after the infusion is stopped. derived from the definition of the volume of distri-
In the case of continuous infusions: bution. As shown previously, V = X0/C0 (see Equation
1-1) for a drug described by a one-compartment
Ct = (C ss e − Kt ) model. Rearranging this equation, we see that the
loading dose equals the desired concentration
where t = time after the infusion is stopped. multiplied by the volume of distribution:
If, as in the previous example, K = 0.139 hr–1 and X 0 = C 0( desired )V (See Equation 1-1.)
Css = 15 mg/L, the plasma concentration 12 hours
after discontinuing the infusion would be:
C12 hr = (15 mg/L)e(−0.139 hr
−1
)(12 hr)
= 15 mg/L
(0.19)
= 2.9 mg/L
Loading Dose
As stated previously, after a continuous IV infusion FIGURE 5-11.
of drug is begun, five drug half-lives are needed to Plasma drug concentrations resulting from an intravenous
achieve steady state. In many clinical situations, an loading dose.

72

Plasma drug concentrations over time resulting from a Plasma drug concentrations resulting from an intravenous
continuous intravenous infusion. loading dose given with a continuous infusion.
Note that C0 in this case is equivalent to the desired With a continuous infusion, the plasma concentra-
steady-state concentration. tions are described by:
We know that an IV loading dose produces
plasma concentrations as shown in Figure 5-11, K0
=
Ct (1− e −Kt ′ )
and the continuous infusion produces plasma VK
concentrations as shown in Figure 5-12. If both
the loading dose and the continuous IV infusion are where:
given, the net effect should be a fairly steady plasma
concentration, as depicted by the bold line in Figure t′ = time after beginning infusion,
5-13. Before the constant IV infusion has reached K0 = rate of drug infusion,
steady state, the bolus loading dose has produced V = volume of distribution, and
a nearly steady-state drug concentration, and when K = elimination rate constant.
the drug from the loading dose is almost eliminated,
the constant IV infusion should be approximately When both the injection and infusion are admin-
at steady state. With lidocaine, heparin, and istered together, the plasma concentration after
theophylline, loading doses usually precede their
beginning the regimen is calculated by adding the
continuous IV infusions, providing immediate as
two equations:
well as sustained effects that combine to produce a
steady therapeutic plasma concentration. X 0 − Kt K 0
Previously used equations can be combined to Ct = e + (1− e − Kt' )
V VK
describe the plasma concentration resulting from
a bolus injection with continuous infusion. With
an IV injection, the equation describing the plasma For example, an adult patient is estimated to have
concentration after a dose is: a theophylline half-life of 8 hours (K = 0.087 hr –1)
and a V of 30 L. These estimates are obtained from
X 0 − Kt known information about this patient or from
Ct = e
V published reports of similar patients. If the patient
where: is given a loading dose of 400 mg of theophylline,
t = time after dose, and a continuous infusion of 35 mg/hour is begun
X0 = initial loading dose, at the same time, what will the plasma concentra-
V = volume of distribution, and tion be 24 hours later?
K = elimination rate constant.

73
X 0 −Kt K 0 model combines the approaches just presented for

Ct = e + (1− e −Kt' ) multiple-dose injections and continuous infusions.
V VK
Let’s assume that a drug is given intravenously
400 mg −0.087 hr −1 (24 hr) 35 mg/hr −1
over 1 hour every 8 hours. For the first in a series
= e + −1
(1− e −0.087 hr (24 hr) )
30 L 30 L × 0.087 hr of IV infusions lasting 60 minutes each, the plasma
concentrations will be similar to those observed
400 mg 35 mg/hr during the first 60 minutes of a continuous infusion.
= (0.124) + (0.876)
30 L 30 L × 0.087 hr −1 Then, when the infusion is stopped, plasma concen-
trations will decline in a first-order process, just as
= 1.6 mg/L + 11.7 mg/L = 13.3 mg/L
after IV injections (Figure 5-14).
In clinical practice, drugs such as theophylline The peak (or maximum) plasma concentration
usually are not given by IV bolus injection, not even after the first infusion (Cmax1) is estimated by:
loading doses. Loading doses usually are given as K0
short infusions (often 30–60 minutes). Taking this C=
max 1 (1− e − Kt )
procedure into account, we can further modify the
VK
above equations to predict plasma concentrations. where:
For the loading dose: C = concentration in plasma,
K0 = rate of drug infusion (dose/time of
X 0 /t infusion),
C=
peak ss (1− e −Kt )
VK V = volume of distribution,
where: K = elimination rate constant, and
X0 = dose (in this case, the loading dose), t = time (duration) of infusion.
t = infusion period (e.g., 0.5 hour),
K = elimination rate constant, and This equation was used above to describe plasma
drug concentrations with continuous infusion
V = volume of distribution.
before steady state.
The trough concentration after the first dose
Multiple Intravenous Infusions
(Cmin1) occurs at the end of the dosing interval (τ)
(Intermittent Infusions)
directly before the next dose.
In Lesson 4, we discussed multiple-dose IV bolus
Cmin 1 is calculated by multiplying Cmax 1 by e−K(τ−t)
drug administration. With multiple-dose IV bolus
or Cmin 1 = Cmax 1 e−K(τ−t).
administration, we assumed that the drug was
administered by rapid IV injection. However, rapid
IV injections are often associated with increased
risks of adverse effects.
Therefore, many drugs administered intrave-
nously are infused over a 30- to 60-minute time
period; some drugs may require a longer infu-
sion time. This method of giving multiple doses by
infusion at specified intervals (τ), called intermit-
tent IV infusion, changes the plasma concentration
profile from what would be seen with multiple
rapid IV injections. Therefore, a new model must FIGURE 5-14.
be created to predict plasma drug concentrations Plasma drug concentrations resulting from a short intravenous
after multiple-dose intermittent IV infusions. This infusion.

74
This equation can be rewritten as follows: A practical example for this equation is shown
below to determine the Cpmin or trough concentra-
C min 1 (K 0 / VK )(1− e −Kt )(e −K ( τ−t ) )
= tion of a drug given by intermittent infusion.
By the principle of superposition (see Lesson 4), K0  (1− e −Kt )  −Kt'

5-2 C min (steady state) =  −K τ 
e
Cmax2 can be estimated: VK  (1− e ) 
C= C max 1 + C min 1
max 2
where t′ = τ – t.
K0 K The equation for Cmin(steady state) is very important
= (1− e −Kt ) + 0 (1− e −Kt )(e −K ( τ−t ) )
VK VK in clinical practice. It can be used to predict plasma
concentrations for multiple intermittent IV infu-
K sions of any drug that follows first-order elimination
= 0 (1− e −Kt )(1+ e −K ( τ−t ) )
VK (assuming a one-compartment model). It also can
be used to predict plasma concentrations at any
Cmin 2 can be calculated:
time between Cmax and Cmin, where t′ equals the time
C= C max 1 × e −K ( τ−t ) between the end of the infusion and the determina-
min 2
tion of the plasma concentration. For application
K of this method, refer to cases that include IV inter-
= 0 (1− e −Kt )(1+ e −K ( τ−t ) )(e −K τ )
VK mittent infusions, which will show a step-by-step
process for dose calculations.
K0
= (1− e −Kt )(e −K τ + e −2K τ )
VK
Clinical Correlate
This expansion of the equation can continue as in
Lesson 4 until n number of infusions have been given: Here is one way we can illustrate the relationship
of the equations described in this section:
K 0  (1− e −Kt )(1− e −n K τ )  Suppose a patient with severe renal dysfunction
C max n =   receives a 100-mg dose of gentamicin, and a
VK  (1− e −K τ )  peak concentration, drawn at the end of the
infusion, is reported by the laboratory as 8.0 mg/L.
As n becomes very large, (1– e–nKτ) approaches
No additional doses are administered, and a
1, and the equation becomes:
repeat serum concentration drawn 24 hours
later is reported as 3.0 mg/L. Before we can
K 0  (1− e −Kt ) 
5-1 C max ss =   administer a second dose of gentamicin in
VK  (1− e −K τ )  this patient, we want to wait until the serum
concentration is 1.0 mg/L. How much longer
Then, to determine the concentration at any time
must we wait until this occurs?
(t′) after the peak, the following multiple IV infusion
at steady state equation can be used: The first step in solving this question is to
determine the patient’s K. K can be calculated
K  (1− e −Kt )  −Kt' using the equation below. Instead of using the
C= 0  −K τ 
e
VK  (1− e )  variable Cmin, we will use the variable Ct , which in
this case represents the concentration 24 hours
where t′ = total hours drug was allowed to be eliminated. after the first level is drawn:

75
Ct = C peak e −Kt
Clinically Important Equations
3 mg/L = (8 mg/L)e −K (24 hr) Identified in This Chapter
3 mg/L
= e −K (24 hr)
8 mg/L 1. X 0 = C 0( desired )V
0.375 = e −K (24 hr) This is Equation 1-1 rearranged

In 0.375 = −K (24 hr) X0 is the Loading Dose, sometimes abbreviated
as LD
−0.981 =
−K (24 hr)
0.981
=K K 0  (1− e −Kt ) 
24 hr 2. C max ss =   Equation 5-1
VK  (1− e −K τ ) 
K = 0.041 hr −1
Knowing K, we can calculate the time (t ) required

for the concentration to decrease to 1.0 mg/L. Ct 3. C min ss =
K 0  (1− e −Kt )  −Kt' Equation 5-2
 e
will now be our desired concentration of 1.0 mg/L: VK  (1− e −K τ ) 
Ct = C peak e −Kt where t′ = τ – t

−1 This equation may be rewritten as
1.0 mg/L = (8 mg/L)e( −0.041 hr )t
=
C trough C peak ss × e −K ( τ−t )
1.0 mg/L −1
ss
= e( −0.041 hr )t
8 mg/L
or
0.125 = e( −0.041 hr
−1
)t
C=
min ss C max ss × e −K ( τ−t )
In 0.125 = ( −0.041 hr −1 )t
( −0.041 hr −1 )t
−2.08 =
−2.08
=t
−0.041 hr −1
t = 50.7 hours
Therefore, it will take slightly longer than 2

days after the peak concentration for the serum
concentration to decrease to 1.0 mg/L.

76
REVIEW QUESTIONS
5-1. For a drug regimen, if the elimination rate 5-7. If you double the infusion rate of a drug,
(K) of a drug is reduced while V, X0, and you should expect to see a twofold increase
τ remain constant, the peak and trough in the drug’s steady-state concentration.
concentrations will: Assume that clearance remains constant.
A. increase. A. True
B. decrease. B. False
5-2. A decrease in drug dose will result in lower 5-8. Theophylline is administered to a patient
plasma concentrations at steady state but at 35 mg/hour via a constant IV infusion.
will not change the time to reach steady If the patient has a total body clearance for
state. theophylline of 40 mL/minute, what should
A. True this patient’s steady-state plasma concen-
tration be?
B. False
A. 14.6 mg/L
5-3. Which of the following dosing techniques B. 0.875 mg/L
results in smaller fluctuations between peak C. 0.1 mg/L
and trough plasma levels?
A. small doses very frequently 5-9. With a continuous IV infusion of drug,
B. large doses relatively less frequently the steady-state plasma concentration is
directly proportional to:
5-4. When the volume of distribution increases A. clearance.
(and clearance remains the same), steady- B. volume of distribution.
state plasma concentrations will have more
C. drug infusion rate.
peak-to-trough variation.
D. K.
A. True
B. False 5-10. If a drug is given by continuous IV infu-
sion at a rate of 20 mg/hour and produces
5-5. When drug clearance decreases (while a steady-state plasma concentration of
volume of distribution remains unchanged), 10 mg/L, what infusion rate will result in a
steady-state plasma concentrations will: new Css of 15 mg/L?
A. increase. A. 30 mg/hour
B. decrease. B. 35 mg/hour
5-6. Steady-state plasma concentration is C. 50 mg/hour
approximately reached when the contin- D. 75 mg/hour
uous infusion has been given for at least
how many half-lives of the drug? 5-11. For a continuous infusion, given the equa-
tion C = K0(1 – e–Kt)/Clt , at steady state
A. two
the value for t approaches infinity and e–Kt
B. three approaches infinity.
C. five A. True
D. ten B. False

77
This case applies to Questions 5-12 and 5-13. A 5-15. For the patient in the question above, what
patient is to be started on a continuous infusion of will the peak plasma concentration be at
a drug. To achieve an immediate effect, a loading steady state?
dose is administered over 30 minutes and then the A. 4.6 mg/L
continuous infusion will begin. From a previous
B. 7.4 mg/L
regimen of the same drug, you estimate that the
patient’s K = 0.04 hr–1 and V = 28 L. Assume that C. 12.8 mg/L
none of this drug has been administered in the last D. 22 mg/L
month, so the plasma concentration before therapy
is 0 mg/L. 5-16. For the patient in the previous question,
calculate the trough plasma concentration
5-12. If the Css(desired) is 12 mg/L, what should the at steady state.
loading dose be?
A. 0.54 mg/L
A. 13 mg
B. 1.42 mg/L
B.
42 mg
C. 336 mg C. 1.92 mg/L
D. 1200 mg D. 2.25 mg/L
5-13. What rate of infusion (K0) should result in a

Css of 12 mg/L? ANSWERS
A. 0.2 mg/hour
B. 13.4 mg/hour 5-1. A. CORRECT ANSWER. This can be deter-
C. 600 mg/hour mined by examination of the equation
from Lesson 4:
Refer to this equation when working Questions
5-14 through 5-16: X0  1 
C peak(steady state) =  (1− e −K τ ) 
V  
K0  (1− e −n Kt )  −Kt'
C=  −K τ 
e
VK  (1− e )  B. Incorrect answer
5-2. A. CORRECT ANSWER. The time to reach

5-14. A patient is to be given 100 mg of genta-
steady state is determined by K.
micin IV over 1 hour every 12 hours. If the
patient is assumed to have a K of 0.15 hr–1 B. Incorrect answer
and a V of 15 L, how long will it take to reach
5-3. A. CORRECT ANSWER. Because the time
steady state?
interval would be relatively short, there
A. 3.5 hours would not be as much time for plasma
B. 5 hours concentrations to decline.
C. 17 hours
B. Incorrect answer
D. 23 hours

78
A. Incorrect answer
5-4. 5-9. A, B, D. Incorrect answers
B. CORRECT ANSWER. A larger volume C. CORRECT ANSWER. The steady-state
of distribution will result in the same concentration is directly proportional to
amount of drug distributing in a greater the drug infusion rate.
volume, which would result in a lower
peak-to-trough variation. 5-10. A. CORRECT ANSWER
B, C, D. Incorrect answers
5-5. A. CORRECT ANSWER. When clearance
decreases, plasma concentrations will 5-11. A. Incorrect answer. As t becomes larger,
increase because drug is administered at the term e–Kt becomes smaller, and the
the same rate (dose and dosing interval) term 1 – e–Kt approaches 1.
but is being removed at a lower rate. B. CORRECT ANSWER
B. Incorrect answer

A. Incorrect answer. Only 75% of the
5-6. C. CORRECT ANSWER. The loading dose is
steady-state concentration would be determined by multiplying the desired
reached by two half-lives. concentration (12 mg/L) by the volume
B. Incorrect answer. Only 87.5% of the of distribution:
steady-state concentration would be Css (desired) × V = 12 mg/L × 28 L = 336 mg.
reached by three half-lives.
Note that the units cancel out to yield
C. CORRECT ANSWER. At five half-lives,
milligrams.
approximately 97% of the steady-state
concentration has been reached. 5-13. A, C. Incorrect answers
D. Incorrect answer. This is much longer
B. CORRECT ANSWER. The infusion rate is
than necessary. related to Clt and Css as follows:
5-7. A. CORRECT ANSWER. The changes in the
Css = K0/Clt .
infusion rate will directly affect plasma Clt can be determined by multiplying
concentrations, if other factors remain
V × K = 1.12 L/hour.
constant.
So, rearranging,
B. Incorrect answer

K0 = Css × Clt = 12 mg/L × 1.12 L/hour =
5-8. A. CORRECT ANSWER. The equation Css = 13.4 mg/hour.
K0/Clt should be used. The value for K0
is 35 mg/hour. The value for Clt must 5-14. A, B, C. Incorrect answers
be converted from 40 mL/minute to D. CORRECT ANSWER. One half-life is
2.4 L/hour. calculated as follows:
B, C. Incorrect answers
T½ = 0.693/K.
Steady state is reached by five half-lives,
or 23 hours.

79
5-15. A, C, D. Incorrect answers 5-16. A, C, D. Incorrect answers

B. CORRECT ANSWER. At steady state, the B. CORRECT ANSWER. The trough concen-
equation below would be used: tration is calculated from the peak value
as follows:
K 0 (1− e −Kt )
C max (steady state) =
VK (1− e −Kt ) = C peak × e −K ( τ−t )
C trough
(100 mg/hr)(0.139) = 7.4 mg/L × e −0.15 hr

−1
(12 hr −1 hr )
=
(2.25 L/hr)(0.835)
= 7.4 mg/L × 0.192 = 1.42 mg/L
= 7.4 mg/L

80
Discussion Points
D-1. With the continuous IV infusion model D-5. Explain how changing the dosing interval
of drug administration, what two factors (τ) influences the time to reach steady state
determine the steady-state plasma concen- when multiple doses are administered.
tration?
D-6. If clearance is reduced to 25% of the initial
D-2. What is the purpose of administering a rate and all other factors (such as dose,
loading dose of a drug? dosing interval, and volume of distribu-
tion) remain constant, how will steady-state
D-3. What is the following portion of the plasma concentrations change?
multiple-dose equation called, and why is it
called that? D-7. Explain why, for most drugs, the increase in
drug plasma concentrations resulting from
1/(1 – e–K τ ) a single dose will be the same magnitude
D-4. Given the equation below for a drug given whether it is the first or the tenth dose.
by intermittent infusion, what does t′
represent?
K0  (1− e −Kt )  −Kt ′

C min(steady state) =  −K τ 
e
VK  (1− e ) 

LESSON 6
Two-Compartment Models
OBJECTIVES
1. Describe when to use back-extrapolation versus method of residuals.
2. Calculate a residual line.
3. Calculate alpha (α), beta (β), and intercepts A and B for a drug conforming to a
two-compartment model.
4. Describe when to use a monoexponential versus a biexponential equation.
5. Calculate Vc, Varea (also known as V β ), and Vss (using both methods) for a
two-compartment model.
Prior lessons focused on one-compartment models, but many drugs are better
characterized by multicompartment models. In this lesson, we briefly discuss
multicompartment models and present a few applications. Multicompartment
models are not used as frequently as the one-compartment model in therapeutic
drug monitoring, partly because they are more difficult to construct and apply.
Generally, multicompartment models are applied when the natural log of
plasma drug concentration versus time curve is not a straight line after an intra-
venous dose or when the plasma concentration versus time profile cannot be
characterized by a single exponential function (i.e., Ct = C0e–Kt). When the natural
log of plasma drug concentration versus time curve is not a straight line, a multi-
compartment model must be constructed to describe the change in concentration
over time (Figure 6-1).
Of the multicompartment models, the two-compartment model is most
frequently used. This model usually consists of a central compartment of the well-
perfused tissues and a peripheral compartment of less well-perfused tissues (such
as muscle and fat). Figure 6-2 shows a diagram of the two-compartment model
after an intravenous bolus dose, where:
X0 = dose of drug administered
Xc = amount of drug in central compartment
Xp = amount of drug in peripheral compartment
K12 = rate constant for transfer of drug from the central compartment to the
peripheral compartment. (The subscript 12 indicates transfer from
the first [central] to the second [peripheral] compartment.)
81
82
FIGURE 6-1.
Concentration versus
time plot for one- versus
two-compartment (CMPT)
model.
K21 = rate constant for transfer of drug from distribution to organs with high blood flow (central
the peripheral compartment to the compartment) and slower distribution to organs
central compartment. (The subscript with less blood flow (peripheral compartment).
21 indicates transfer from the second After the intravenous injection of a drug that
[peripheral] to the first [central] follows a two-compartment model, the drug concen-
compartment. Note: Both K12 and K21 are trations in all fluids and tissues associated with
called microconstants.) the central compartment decline more rapidly
K10 = first-order elimination rate constant in the distribution phase than during the post-
(similar to the K used previously), indi- distribution phase. After some time, a pseudo-
cating elimination of drug out of the equilibrium is attained between the central compart-
central compartment into urine, feces, etc. ment and the tissues and fluids of the peripheral
A natural log of plasma drug concentration compartment; the plasma drug concentration versus
versus time curve for a two-compartment model time profile is then characterized as a linear process
shows a curvilinear profile—a curved portion when plotted on semilog paper (i.e., terminal or linear
followed by a straight line. This biexponential curve elimination phase). For many drugs (e.g., amino-
can be described by two exponential terms (Figure glycosides), the distribution phase is very short (e.g.,
6-3). The phases of the curve may represent rapid minutes). If plasma concentrations are measured after
this phase is completed, the central compartment can
be ignored and a one-compartment model adequately
represents the plasma concentrations observed. Other
drugs (e.g., vancomycin, digoxin) have a longer distri-
bution phase (hours). If plasma concentrations of
these drugs are determined within the first few hours
after a dose is given, the nonlinear (multiexponential)
decline of drug concentrations must be considered
FIGURE 6-2. when calculating half-life and other parameters.
Graphic representation of a two-compartment model.

Lesson 6 | Two-Compartment Models
83
FIGURE 6-3.
Four stages of drug distribution and elimination after rapid intravenous injection. Points I, II, III, and IV (right) correspond to the points
on the plasma concentration curve (left ). Point I: The injection has just been completed, and drug density in the central compartment
is highest. Drug distribution and elimination have just begun. Point II: Midway through the distribution process, the drug density in
the central compartment is falling rapidly, mainly because of rapid drug distribution out of the central compartment into the peripheral
compartment. The density of drug in the peripheral compartment has not yet reached that of the central compartment. Point III:
Distribution equilibrium has been attained, and drug densities in the central and peripheral compartments are approximately equal.
Drug distribution in both directions continues to take place, but the ratio of drug quantities in the central and peripheral compartments
remains constant. At this point, the major determinant of drug disappearance from the central compartment becomes the elimination
process; previously, drug disappearance was determined mainly by distribution. Point IV: During this elimination phase, the drug is
being “drained” from both compartments out of the body (via the central compartment) at approximately the same rate.
Source: Reprinted with permission from Greenblatt DJ and Shrader RI. Pharmacokinetics in Clinical Practice. Philadelphia, PA:
Saunders; 1985.

84
Calculating Two-Compartment
Clinical Correlate Parameters
Digoxin is a drug that, when administered as In this section, we apply mathematical principles
a short intravenous infusion, is best described
to the two-compartment model to calculate useful
by a two-compartment model. After the drug
pharmacokinetic parameters.
is infused, the distribution phase is apparent
for 4–6 hours (Figure 6-4). Digoxin distributes From discussion of the one-compartment
out of plasma (the central compartment) and model, we know that the elimination rate constant
extensively into muscle tissue (the peripheral (K) is estimated from the slope of the natural log
compartment). After the initial distribution of plasma drug concentration versus time curve.
phase, a pseudoequilibrium in distribution is However, in a two-compartment model, in which
achieved between the central and peripheral that plot is curvilinear, the slope varies, depending
compartments. Because the site of digoxin effect on which portion of the curve is examined (Figure
is in muscle (specifically, the myocardium), 6-5). The slope of the initial portion is determined
the plasma concentrations observed after primarily by the distribution rate, whereas the slope
completion of the distribution phase more of the terminal portion is determined primarily by
accurately reflect concentrations in the tissue the elimination rate.
and pharmacodynamic response. For patients
receiving digoxin, blood should be drawn for The linear (or post-distributive) terminal portion
plasma concentration determination after of this curve may be back-extrapolated to time zero
completion of the distribution phase; thus, trough (t0). The negative slope of this line is referred to as
digoxin concentrations are usually used clinically beta (β), and like K in the one-compartment model,
when monitoring digoxin therapy. β is an elimination rate constant. β is the terminal
Vancomycin is another drug that follows a elimination rate constant, which means it applies
two-compartment model with an initial 2- to after distribution has reached pseudoequilibrium.
4-hour α-distribution phase followed by a The y-intercept of this line (B) is used in various
linear terminal elimination phase. As described equations for two-compartment parameters.
later in the vancomycin cases (see Lesson As in the one-compartment model, a half-life
13), peak vancomycin concentrations must be (the beta half-life) can be calculated from β:
drawn approximately 2 hours after the end of a
vancomycin infusion to avoid obtaining a peak 0.693 1 0.693
concentration during the initial distribution phase =T 12 = T 2
(see Figure 13-3). β K
FIGURE 6-5.
FIGURE 6-4. Plasma drug concentrations with a two-compartment model
Digoxin plasma concentration versus time. after an intravenous bolus dose.

85
Throughout the time that drug is present in the

body, distribution takes place between the central
and peripheral compartments. We can calculate a
rate of distribution using the method of residuals,
which separates the effects of distribution and
elimination. This method estimates the effect of
distribution on the overall plasma concentration
curve and uses the difference between the effect of
elimination and the actual plasma concentrations to
determine the distribution rate.
To apply the method of residuals, we use the back- FIGURE 6-7.
Determination of the residual line.
extrapolated line used to determine β and B (Figure
6-6). If w, x, y, and z are actual, determined concentra-
them to the y-axis. Then, for the first five points, extrap-
tion time points, let w′, x′, y′, and z′ represent points
olated values can be estimated at each time (0.25,
on the new (extrapolated) line at the same times that
0.5, 1.0, 1.5, and 2.0 hours) where the time intersects
the actual concentrations were observed. These newly
the new line (similar to Figure 6-6). Subtracting the
generated points represent the effect of elimina-
extrapolated values from the actual plasma concentra-
tion alone, as if distribution had been instantaneous.
tions yields a new set of residual concentration points,
Subtraction of the extrapolated points from the corre-
similar to those values shown in Table 6-2.
sponding actual points (w – w′, x – x′, etc.) yields a new
set of plasma concentration points for each time point. Plot the residual concentrations (on the same
If we plot these new points, we generate a new line, semilog paper) versus time and draw a straight line
the residual line (Figure 6-7). The negative slope of connecting all of your new points (similar to Figure
the residual line is referred to as alpha (α), and α is the 6-7). Determine that the slope of that plot equals
distribution rate constant for the two-compartment –1.8 hours–1.
system. The y-intercept of the residual line is A.
In C 1 − In C 0 (In 3 − In 18.5)
Let’s proceed through an example, applying the Slope (α ) = = =−1.8 hr −1
t1 − t0 (1.5 hr − 0.5 hr)
method of residuals. Draw the plot for the following
example on semilog graph paper. A dose of drug is (distribution rate)
administered by rapid intravenous injection, and
(See Equation 3-1.)
the concentrations shown in Table 6-1 result.
The last four points form a straight line (similar
to Figure 6-5) so back-extrapolate a line that connects TABLE 6-1. Plasma Drug Concentrations after
Rapid Intravenous Injection
Time after
Dose (hours) Plasma Concentration (mg/L)
0.25
43.0
0.5
32.0
1.0
20.0
1.5
14.0
2.0
11.0
4.0
6.5
8.0 2.8
12.0 1.2
FIGURE 6-6. 16.0 0.52
Method of residuals.

86
TABLE 6-2. Residual Concentration Points

Plasma Concentration (mg/L)
Time after
Dose (hours)
Actual Extrapolated
Residual
0.25 43 14.5 28.5
0.5 32 13.5 18.5
1.0 20 12.3 7.7
1.5 14 11.0 3.0
2.0 11 10.0 1.0
FIGURE 6-8.
Plasma drug concentrations with a one-compartment model
after an intravenous bolus dose (first-order elimination).
When the negative is dropped, this slope equals
The two-compartment model (Figure 6-9) is
α; we observe from the plot that the intercept (A)
the sum of two linear components, representing
of the residual line is 45 mg/L. We also can esti-
distribution and elimination (Figure 6-10), so we
mate β (0.21 hour –1) from the slope of the terminal
can determine drug concentration (C) at any time
straight-line portion.
(t) by adding those two components. In each case, A
In C 1 − In C 0 (In 2.8 − In 6.5) or B is used for C0, and α or β is used for K. Therefore:
Slope (β) = = =− 0.21 hr −1
t1 − t0 (8 hr − 4 hr) Ct = Ae –αt + Be –βt
(elimination rate)
This equation is called a biexponential equation
Inspection of the extrapolated portion yields a value because two exponents are incorporated.
for B (15 mg/L). For the two-compartment model, different
Note that α must be greater than β, indicating volume of distribution parameters exist: the central
that drug removal from plasma by distribution into compartment volume (Vc ), the volume by area (Varea,
tissues proceeds at a greater rate than does drug also known as Vβ), and the steady-state volume of
removal from plasma by eliminating organs (e.g., distribution (Vss). Each of these volumes relates to
kidneys and liver). The initial portion of the plot is different underlying assumptions.
steeper than the terminal portion. As in the one-compartment model, a volume can
be calculated by:
Biexponential Equation and Volumes dose dose

=
Vc =
of Distribution A +B C0
The estimations of A, B, α, and β performed above

are useful for predicting plasma concentrations of
drugs characterized by a two-compartment model.
For a one-compartment model (Figure 6-8), we
know that the plasma concentration (C) at any time
(t) can be described by:
Ct = C0 e–Kt (See Equation 3-2.)
where C0 is the initial concentration and K is the

elimination rate. The equation is called a mono- FIGURE 6-9.
exponential equation because the line is described Plasma drug concentrations with a two-compartment model
by one exponent. after an intravenous bolus dose (first-order elimination).

87
Because different methods can be used to

calculate the various volumes of distribution of a
two-compartment model, you should always specify
the method used. When reading a pharmacokinetic
study, pay particular attention to the method for
calculating the volume of distribution.
Clinical Correlate
FIGURE 6-10. Here is an example of one potential problem
Linear components of a two-exponential (two-compartment) model.
when dealing with drugs exhibiting biexponential
For the two-compartment model, this volume elimination: If plasma concentrations are
determined soon after an intravenous dose is
would be equivalent to the volume of the central
administered (during the distribution phase) and
compartment (Vc). The Vc relates the amount of drug
a one-compartment model is assumed, then the
in the central compartment to the concentration in
patient’s drug half-life would be underestimated
the central compartment. In the two-compartment
and β would be overestimated (Figure 6-11).
model, C0 is equal to the sum of intercepts A and B. Recall that:
If another volume (Varea or Vβ) is determined
from the area under the plasma concentration In C1 − In C 0
versus time curve and the terminal elimination rate Slope (β or K ) =
t1 − t0
constant (β), this volume is related as follows:
dose Cl A steeper slope equals a faster rate of elimination
Varea= Vβ= = resulting in a shorter half-life.
β× AUC β
This calculation is affected by changes in clearance If a terminal half-life is being calculated for drugs
(Cl). The Varea relates the amount of drug in the body such as vancomycin, you must be sure that the
to the concentration of drug in plasma in the post- distribution phase is completed (approximately
absorption and post-distribution phase. 3–4 hours after the dose) before drawing plasma
levels.
A final volume is the volume of distribution
at steady state (Vss). Although it is not affected by
changes in drug elimination or clearance, it is more
difficult to calculate. One way to estimate Vss is to use
the two-compartment microconstants:
K 12
Vss= Vc + Vc
K 21
or it may be estimated by:

A B
dose ( + )
α 2 β2
Vss =
( A + B )2
α β
FIGURE 6-11.
using A, B, α, and β. Biexponential elimination.

88
REVIEW QUESTIONS
6-1. In the two-compartment model, Xp 6-4. Which of the following is the best definition
represents the: of beta (β)?
A. amount of drug in the body. A. initial rate constant of elimination
B. amount of drug in the peripheral B. terminal half-life
compartment. C. average elimination rate constant
C. fraction of the dose distributing to the D. terminal elimination rate constant
peripheral compartment.
D. parenteral drug dose. 6-5. For a two-compartment model, which of the
following is the term for the residual y-inter-
6-2. The plasma drug concentration versus time cept for the terminal portion of the natural-log
curve for a two-compartment model is plasma concentration versus time line?
represented by what type of curve? A
A.
A.
biexponential B
B.
B.
monoexponential C. α (alpha)
D. β (beta)
6-3. With a two-compartment model, K12 repre-
sents the:
6-6. The method of back-extrapolation is used to
A. first portion of the natural log of plasma calculate:
drug concentration versus time curve,
A. the rate constant of drug elimination
where the log concentration rapidly
from the body.
declines.
B. the area under the plasma concentra-
B. elimination rate constant.
tion versus time curve.
C. rate constant for drug transfer from
C. the amount of drug present in specific
compartment 1 (central) to compart-
organs.
ment 2 (peripheral).
D. rate constant for drug transfer from 6-7. Which equation below correctly represents
compartment 2 (peripheral) to compart- the two-compartment model?
ment 1 (central). Ct = A + B(e–Kt)
A.
Ct = Ae–βt + Be–αt
B.
Ct = Ae–αt + Be–βt
C.

89
6-8. The equation describing elimination after

an intravenous bolus dose of a drug char-
ANSWERS
acterized by a two-compartment model
6-1. A, C, D. Incorrect answers
requires two exponential terms.
B. CORRECT ANSWER. X represents
A.
True
amount of drug and p represents the
B.
False peripheral compartment.
6-9. A patient is given a 500-mg dose of drug 6-2. A. CORRECT ANSWER. A two-compart-
by intravenous injection and the following ment model is best represented by a
plasma concentrations result: biexponential curve.
Incorrect answer
B.
(mg/L) (hours) 6-3. A, B, D. Incorrect answers
72.0 0.25 CORRECT ANSWER. K12 represents the
C.
46.0 0.5 rate constant for drug transfer from
33.0 0.75 compartment 1 (central) to compart-
26.3 1.0 ment 2 (peripheral).
20.0 1.5 6-4. A, B, C. Incorrect answers
16.6 2.0 D. CORRECT ANSWER. β is the terminal
12.2 3.0 elimination rate constant.
9.9 4.0
6-5. A. CORRECT ANSWER. The y-intercept
5.0 6.0
associated with the residual portion of
2.7 8.0 the curve (which has a slope of –α) is A.
0.82 12.0
Which answer below is the best estimate for 6-6. A. CORRECT ANSWER. It can be used to
α and β? calculate rate of drug elimination.
A. 0.30 hr–1, 3.17 hr–1 B, C. Incorrect answers
B. 3.17 hr , 0.3 hr
–1 –1
6-7. A, B. Incorrect answers
C. 2.1 hr–1, 0.6 hr–1
C. CORRECT ANSWER. The correct
D. 0.6 hr–1, 2.1 hr–1 equation is: Ct = Ae–αt + Be–βt

90
6-8. A. CORRECT ANSWER. One exponent is β can be determined from the slope of
needed for distribution phase and the other the terminal straight-line portion of the
for elimination or post-distribution phase. plot. For example, the points at 3 and 12
Incorrect answer
B. hours may be selected:
6-9. A. Incorrect answer ∆x In C 2 − In C1

=
slope =
B. CORRECT ANSWER. See table below for ∆y t2 − t1
result (your numbers may vary slightly).
In 0.82 − In 12.2
=
Time Actual – Extrapolated
=
Residual 12 hr − 3 hr
(hours) (mg/L) (mg/L) (mg/L)
−0.198 − 2.5 −2.698
= =
0.25
72.0 27.8 44.2 9 hr 9 hr
0.5
46.0 25.5 20.5
0.75
33.0 24.8 8.2
= −0.30 hr −1
1.0
26.3 22.2 4.1 β = −slope = 0.30 hr −1
1.5
20.0 18.9 1.1
C, D. Incorrect answers
The slope of the residual line is deter-
mined to calculate α. Two residual points
are selected, such as 0.25 and 1.0 hour:
∆x In C 2 − In C1
=
slope =
∆y t2 − t1
In 4.1− In 44.2
=
1.0 hr − 0.25 hr
1.41− 3.79
=
0.75 hr
= −3.17 hr −1
α = −slope = 3.17 hr −1

91
Discussion Points
D-1. Describe situations for which it would be D-3. Discuss the clinical implications of obtaining
better to use a two-compartment model a vancomycin peak concentration during
rather than a one-compartment model. the distribution phase on the resultant
pharmacokinetic values of K and V.
D-2. What is the minimum number of plasma-
concentration data points needed to D-4. Discuss the effect that a two-compartment
calculate parameters for a two-compart- drug (such as digoxin) with an extensive
ment model? peripheral compartment has on the value
for that drug’s apparent volume of
distribution (V).

Practice Set 2
The following problems are for your review. For continuous infusion at steady state:
Definitions of symbols and key equations are
provided here: K0 K0
C=
SS =
K = elimination rate constant VK Clt
C0 = plasma drug concentration just after a
single intravenous injection
e = base for the natural log function = 2.718
QUESTIONS
τ = dosing interval
K0 = rate of dose administration (may be The following applies to Questions PS2-1 to PS2-6
expressed as milligrams per hour in the below: A 60-kg patient is begun on a continuous
sense of a continuous infusion or as drug intravenous infusion of theophylline at 40 mg/hour.
dose divided by infusion time for inter- Forty-eight hours after beginning the infusion, the
mittent infusions) plasma concentration is 12 mg/L.
V = volume of distribution
PS2-1. If we assume that this concentration is at
Cpeak = peak plasma drug concentration at steady state, what is the theophylline clear-
steady state ance?
Ctrough = trough plasma drug concentration at
steady state A. 3.3 L/hour
t = duration of intravenous infusion B. 0.3 L/hour
C. 33 L/hour
For multiple-dose, intermittent, intravenous
D. 198 L/hour
bolus injection at steady state:
X0  1  PS2-2. If the volume of distribution is estimated to

C peak = be 30 L, what is the half-life?
V  1− e −K τ 
A. 1.7 hours
C trough = C peak e −K τ
B. 6.3 hours
C. 13.3 hours
For multiple-dose, intermittent, intravenous
infusion: D. 22.1 hours
K 0  1− e −Kt  PS2-3. As we know V and K, what would the plasma

C peak =   concentration be 10 hours after beginning
VK  1− e −K τ 
the infusion?
C trough = C peak e −K ( τ−t )
A. 3.2 mg/L
B. 4.8 mg/L
For continuous infusion before steady state is
C. 8.1 mg/L
reached:
K0 D. 11.0 mg/L
=
C (1− e −Kt )
VK
93
94
PS2-4. If the infusion is continued for 3 days and PS2-8. After the fifth dose, a peak plasma
then discontinued, what would the plasma concentration (drawn at the end of the
concentration be 12 hours after stopping infusion) is 5 mg/L and the trough concen-
the infusion? tration (drawn right before the sixth dose) is
0.9 mg/L. What is the patient’s actual genta-
A. 1.2 mg/L micin half-life?
B. 3.2 mg/L
A. 1 hour
C. 7.6 mg/L
B. 2 hours
D. 8.1 mg/L
C. 4 hours
PS2-5. If the infusion is continued for 3 days at D. 8 hours
40 mg/hour and the steady-state plasma
concentration is 12 mg/L, what rate of drug PS2-9. What would be the volume of distribution?
infusion would likely result in a concentra- [hint, rearrange Equation 5-1]
tion of 15 mg/L?
A. 7.6 L
A. 46 mg/hour B. 10.2 L
B. 50 mg/hour C. 15.5 L
C. 60 mg/hour D. 22.0 L
D. 80 mg/hour
PS2-10. For this patient, what dose should be admin-
PS2-6. After the increased infusion rate above is istered to reach a new steady-state peak
begun, how long would it take to reach a gentamicin concentration of 8 mg/L?
plasma concentration of 15 mg/L?
A. 64 mg
A. 6.3 hours B. 82 mg
B. 12.6 hours C. 95 mg
C. 18.9 hours D. 128 mg
D. 31.5 hours
The following pertains to Questions PS2-7 to

PS2-10: A 60-kg patient is started on 80 mg of
gentamicin every 6 hours given as 1-hour infusions.
PS2-7. If this patient is assumed to have an average

V of 15 L and a normal gentamicin half-life
of 3 hours, what will be the peak plasma
concentration at steady state?
A. 6.3 mg/L
B. 8.9 mg/L
C. 12.2 mg/L
D. 15.4 mg/L

Practice Set 2 |
95
ANSWERS
PS2-1. A. CORRECT ANSWER. Clt = K0/Css = PS2-4. A, C, D. Incorrect answers
40 mg/hour/12 mg/L = 3.3 L/hour B. CORRECT ANSWER. If the continuous
Incorrect answer. You may have inverted
B. intravenous infusion is continued for
the formula. 3 days, steady state would have been
C, D. Incorrect answers reached, so the plasma concentration
would be 12 mg/L. When the infusion
PS2-2. A, C, D. Incorrect answers is stopped, the declining drug concen-
B. CORRECT ANSWER. First, K can be tration can be described just as after an
calculated from the equation Clt = KV. intravenous injection:
Rearranged: Ct = Css e–Kt
K = Clt /V = 3.3 L/hour/30 L = 0.11 hr –1 where:
Ct = plasma concentration after infu-
Then:
sion has been stopped for t hour,
T½ = 0.693/K = 6.3 hours
Css = steady-state plasma concentra-
PS2-3. A, B, D. Incorrect answers tions from continuous infusion,
and K = elimination rate constant.
C. CORRECT ANSWER. To calculate the
plasma concentration with a continuous So, when t = 12 hours:
infusion before steady state is reached,
−1
the following equation can be used: C 12 hr = (12 mg/L)(e −0.11 hr (12 hr)
)
K0 = (12 mg/L)(0.267)
=
C (1− e −Kt )
VK = 3.2 mg/L
where t = 10 hr. Then: PS2-5. A, C, D. Incorrect answers
B. CORRECT ANSWER. The patient’s
40 mg/hr −1
=C −1
(1− e −0.11 hr (10 hr) ) theophylline clearance equals 3.3 L/hour.
30 L × 0.11 hr Then remember that at steady state:
40 mg/hr Css = K0 /Clt
= (0.667)
30 L × 0.11 hr −1 or, rearranged:
= 8.1 mg/L Css × Clt = K0
If the desired Css equals 15 mg/L, then:
K0 = 15 mg/L × 3.3 L/hour
= 49.5 mg/hour (round to 50)

96
PS2-6. A, B, C. Incorrect answers PS2-8. A, C, D. Incorrect answers

D. CORRECT ANSWER. Whenever the
B. CORRECT ANSWER. The half-life can
infusion rate is changed to a new rate
be calculated from the concentrations
(increased or decreased), it will take
given. Recall that there are two concen-
approximately five half-lives to achieve
trations on a straight line, where K is
a new steady state. So it will take 5 × 6.3
the slope of the line. The slope equals
hours = 31.5 hours.
the change in the y-axis divided by the
change in the x-axis. The time between
PS2-7. A. CORRECT ANSWER. First, recall that the
the end of one infusion and the start of
multiple-dose infusion equation should
the next is 5 hours. Therefore:
be used:
K 0 (1− e −Kt ) ∆y
C peak = K =
−slope =
−
VK (1− e −K τ ) ∆x
where: K0 = 80 mg/1 hour (because the (In 5 mg/L − In 0.9 mg/L)
= −
dose is given over 1 hour). As given, V = 0 − 5 hr
15 L, τ = 6 hours, and T½ = 3 hours. So:
−[1.61− ( −0.11)] 1.72
0.693 = =
=K = 0.231 hr −1 −5 hr 5 hr
3 hr
Then:
= 0.344 hr −1
−1
(80 mg/hr)(1− e −0.231 hr (1 hr)
) Then:
C peak = −1
(15 L × 0.231 hr −1 )(1− e −0.231 hr ( 6 hr)
)
0.693 0.693
(80 mg/hr)(0.206) =
T 12 = = 2.01 hr
= K 0.344 hr −1
(15 L × 0.231 hr −1 )(0.75)
= 6.34 mg/L

Practice Set 2 |
97
PS2-9. A, B, D. Incorrect answers PS2-10. A, B, C. Incorrect answers
C. CORRECT ANSWER. To calculate V, the D. CORRECT ANSWER. To calculate a new

multiple-dose infusion equation (Equa- dose, we would use the same equa-
tion 5-1) can be used, where: tion as above but would now include
the known V and desired Cpeak and then
K 0 (1− e −Kt )
C peak = solve for K0:
VK (1− e −K τ )
and: C peakVK (1− e −K τ )
K0 =
Cpeak = 5 mg/L (1− e −Kt )
K0 = 80 mg/hour
(8 mg/L)(15.5 L)(0.344 hr −1 )(0.873)
K = 0.344 hr–1 =
t = 1 hour (0.291)
τ = 6 hours = 128 mg over 1 hr
By substituting, we get:
So, in practical terms, a 125-mg dose
−0.344 hr −1 (1 hr) would be infused over 1 hour to attain a
(80 mg/hr)(1− e )
5 mg/L = −1 peak of approximately 8 mg/L.
(V × 0.344 hr −1 )(1− e −0.344 hr ( 6 hr)
)
Rearranging gives:
−1
(80 mg/hr)(1− e −0.344 hr (1 hr)
)
V = −1
(5 mg/L × 0.344 hr −1 )(1− e −0.344 hr ( 6 hr)
)
(80 mg/hr)(0.291)
=
(5 mg/L × 0.344 hr −1 )(0.873)
= 15.5 L

LESSON 7
Biopharmaceutics: Absorption
OBJECTIVES
1. Define and understand the factors that comprise the term biopharmaceutics.
2. Describe the effects of the extent and rate of absorption of a drug on plasma
concentrations and area under the curve (AUC).
3. Name factors that can affect a drug’s oral bioavailability and explain the relationship of
bioavailability to drug absorption and AUC.
4. Calculate an F factor for a drug given its intravenous (IV) and oral absorption time
versus concentration AUCs.
5. Use the oral absorption model to calculate pharmacokinetic parameters.
6. Describe the pharmacokinetic differences and clinical utility of controlled-release
products and the several techniques used in formulating controlled-release drugs.
7. Calculate dose and clearance of controlled-release products given plasma
concentration, volume of distribution, and elimination rate constant.
Introduction to Biopharmaceutics
The effect of a drug depends not only on the drug’s characteristics but also on the
nature of the body’s systems. The drug enters the body by some route of adminis-
tration and is subjected to processes such as absorption, distribution, metabolism,
and excretion (Figure 7-1).
The concepts used in pharmacokinetics enable us to understand what
happens to a drug when it enters the body. Unless a drug is given intravenously or
is absorbed cutaneously, it must be absorbed into the systemic circulation to exert
its effect. After entering the systemic circulation, the drug is distributed to various
tissues and fluids. While the drug is distributing into tissues and producing an
effect, the body is working to eliminate the drug and terminate its effect.
A term often used in conjunction with pharmacokinetics is biopharmaceutics,
which is the study of the relationship between the nature and intensity of a drug’s
effects and various drug formulations or administration factors. These factors
include the drug’s chemical nature, inert formulation substances, pharmaceutical
processes used to manufacture the dosage form, and routes of administration.
For an orally administered drug, the absorption process depends on the drug
dissociating from its dosage form, dissolving in body fluids, and then diffusing
99
100
FIGURE 7-1.
Disposition of drug in the body.
across the biologic membrane barriers of the gut absorbed at a faster rate than another similar drug,
wall into the systemic circulation (Figure 7-2). the first drug may produce a higher peak concentra-
Different drugs or different formulations of the tion, which may lead to a clinical effect sooner than
same drug can vary considerably in both the rate the second drug (Figure 7-3).
and extent of absorption. The extent of absorp- When drug absorption is delayed (usually
tion depends on the nature of the drug itself (e.g., through manipulation of the rate of drug release from
its solubility and pKa) as well as the physiologic the formulation), a prolonged or sustained effect can
environment (pH, gastrointestinal [GI] motility, be produced. For certain drugs (e.g., select oral anal-
and muscle vascularity). Most drugs given orally gesics and hypnotics), rapid absorption is preferable.
are not fully absorbed into the systemic circula- For other agents (e.g., antiarrhythmics and broncho-
tion. The difference in absorption rates of drugs has dilators), a slower rate of absorption with a stable
important therapeutic implications. Assuming that effect over a longer time may be desirable.
concentration correlates with effect, if one drug is
FIGURE 7-2.
Processes involved in drug absorption
after oral administration.

Lesson 7 | Biopharmaceutics: Absorption
101

Typical effect of absorption rate on plasma drug Typical effect of different extents of absorption on plasma drug
concentrations. concentrations.
The amount of the drug dose that reaches the In Figure 7-4, a drug is given in a similar dose
systemic circulation determines its bioavailability. (e.g., 100 mg) in two different oral dosage prod-
Factors that can affect a drug’s oral bioavailability ucts (A and B). The AUC for product A is greater
include the drug’s absorption characteristics, drug than that for product B, indicating that the bioavail-
metabolism within the intestinal wall, and hepatic ability of product A is greater than that of product B.
first-pass metabolism of a drug. Therefore, overall When comparing AUCs to assess bioavailability, we
oral bioavailability can be described by the following assume that the clearance of drug with each dosage
equation, which shows the combination of all these form is the same, so differences in the AUC are
factors: directly related to the amount of drug that enters
the systemic circulation. For a specific drug, the AUC
Foral = Fabs × Fgut × Fhepatic is determined by the amount of drug that enters the
systemic circulation and its clearance from circula-
where F is a fraction.
tion.
A product with poor bioavailability is not
In discussing drug absorption and bioavail-
completely absorbed into the systemic circulation
ability, we should recognize that absorption from
or is eliminated by the liver before it reaches the
the GI tract is not always desirable. For some agents,
systemic circulation. Differences in bioavailability
the intended effects are limited to the lumen of the
may be evident between two products (A and B)
GI tract, so absorption may be undesirable. Exam-
containing the same drug but producing different
ples would be anthelmintics and antibiotics, such as
plasma concentrations (Figure 7-4). Although
neomycin, given to decrease gut bacterial counts.
these products may contain the same amount of
drug, their formulations are different (e.g., tablet A term used to express bioavailability is F. It is
and capsule). Different formulations may have a number less than or equal to 1 that indicates the
different absorption characteristics and result in fraction of drug reaching the systemic circulation.
different plasma concentrations. Because product F is often erroneously referred to as the fraction of
B is not absorbed to the same extent as product A, a drug absorbed—it actually represents the frac-
lower plasma concentrations result for product B. tion of a drug that reaches the systemic circulation
and can be affected by not only absorption, but that
The AUC of a plasma drug concentration versus
fraction of drug that escapes both presystemic (i.e.,
time plot reflects the total amount of drug reaching
intestinal wall) and systemic first-pass metabolism.
the systemic circulation. Because bioavailability
For instance, for oral formulations of a drug:
describes the extent of drug eventually reaching
the systemic circulation, comparison of the AUCs of amount of drug reaching systemic circulation
various dosage forms of a drug would compare their F=
total amount of drug
bioavailabilities.

102
Usually, F is determined by comparing the

AUC for the oral dosage form with the AUC for IV
administration of the same dose. The AUC for IV
administration is used because when a drug is given
intravenously, it bypasses absorption. The total
amount of drug goes into the systemic circulation.
A typical value for F is 0.70 for digoxin tablets and
0.80 for digoxin elixir. This indicates that more of
the drug reaches systemic circulation when admin- FIGURE 7-6.
istered as the elixir. The F term gives no indication Typical plasma drug concentration versus time curve resulting
of how fast a drug is absorbed. Proper studies of from an oral formulation.
drug product bioavailability examine both the rate
Continuous measurement of plasma drug
and extent of absorption.
concentrations would probably produce a plot
similar to that shown in Figure 7-6 when plotted
Clinical Correlate using semilog graph paper. By knowing F to be 1
in this example and the drug concentrations over
Absorption of a drug whose bioavailability is time, we can calculate the pharmacokinetic param-
low due to a low F factor is erratic and is more eters of elimination rate (K), volume of distribution
likely to be affected by disease-related changes (V), half-life (T½ ), and total clearance (Clt). In many
in absorption (e.g., rapid GI transit times, short cases, the actual F is not known, so these parameters
bowel syndromes). can be calculated only in terms of their relationship
to F (e.g., Clt/F, V/F). But first, let’s examine the plot
more closely. The initial uphill portion of the graph
Bioavailability indicates drug absorption. Of course, elimination of
We next examine a one-compartment model in which drug also begins as soon as some drug is in the body.
the drug is given orally and absorbed from the GI But in the initial portion of the curve (A), the rate
tract. This example would also apply to intramuscular of drug absorption is greater than the rate of elimi-
administration as the drug must undergo absorption nation so there is an increase in the plasma drug
from the muscle to produce a therapeutic effect. concentration (Figure 7-7). As the amount of drug
in the GI tract (or in the muscle with intramuscular
Let us assume that:
administration) decreases, the rate of absorption
• The drug is 100% absorbed (F = 1). begins to taper off; at point B, the rate of absorp-
• The absorption rate is much greater than tion equals the rate of elimination. On the downhill
the elimination rate. portion of the curve (C), elimination predominates
• Distribution to all tissues and fluids is and absorption is nearly complete.
instantaneous (a one-compartment model). If the drug follows first-order elimination, the
• The drug follows first-order elimination terminal portion of the plasma drug concentration
(Figure 7-5).

First-order elimination. Effects of both absorption and elimination on concentration
versus time curve.
103
can be canceled, leaving us with a unit for volume

of distribution—liters. It is referred to as the Varea.
Once we have the values of volume of distribution
(V ) and elimination rate constant (K ), the total
body clearance (Clt ) can be calculated as follows:
Clt = V × K (See Equation 3-4.)
When drug is absorbed from outside the

FIGURE 7-8. systemic circulation, as with oral and intramus-
Determination of slope (and K) from terminal portion of plasma cular doses, the peak plasma drug concentration
drug concentration curve.
occurs sometime after time zero rather than at time
zero, as with an IV drug injection. The peak plasma
versus time curve should theoretically be a straight concentration occurs at the point at which the
line on semilog graph paper (Figure 7-8). The slope amount eliminated and the amount absorbed are
of the straight-line portion of the curve is related to equal (Figure 7-9).
the elimination rate constant (K). To calculate K or
T ½, we use the techniques described previously,
but the calculations are made from the terminal
Clinical Correlate
portion (straight-line portion) of the curve.
When a drug is absorbed more slowly, such as
In C 2 − In C 1 after an intramuscular injection, it will have a
slope = = −K
t2 − t1 smaller peak concentration and a slightly longer
duration of action than the IV administration of
(See Equation 3-1.) the same drug. Because of the slower absorption
of intramuscularly administered drugs, it will
0.693 take longer to reach peak concentrations
T 1
2 =
K than with IV administration. Consequently, a
therapeutic peak concentration may not be
(See Equation 3-3.)
attained. To obtain a correct peak concentration
For most drugs, absorption after oral admin- time for intramuscularly administered drugs, the
istration is usually nearly complete by 1–2 hours. measurement must be made in the appropriate
After that time, plasma concentrations should reflect time frame. For example, a drug that reaches
the effect of elimination. For sustained-release its peak concentration after 1 hour should not
products, however, significant drug absorption can be sampled after 20 minutes; otherwise, a false
continue for considerably longer than 2 hours. value will be obtained because absorption is not
Let’s calculate the volume of distribution after complete. In addition, because intramuscular
oral or intramuscular administration. This calcula- absorption occurs more slowly, allowing
significant drug elimination to occur before
tion can be performed as follows:
absorption is complete, peak concentrations
amount of drug administered after an intramuscular injection can yield a lower
V =
K × AUC value than that seen with IV administration. The
time to peak is the time corresponding with that
(assuming F = 1), using the trapezoidal rule to calcu- peak concentration. The time required to reach
late the AUC. the peak plasma concentration depends on the
Terms in the equation relative rates of absorption and elimination. A
rapidly absorbed drug has a short time to peak
mg concentration.
−1
hr × (mg/L) × hr

104

Time to peak for oral or intramuscular concentration versus Plasma drug concentration versus time for a typical oral
time curve. formulation.
(the uphill portion), absorption is occurring, but

Oral Absorption Model
Ka cannot be measured directly because the curve
The elimination rate constant has been denoted demonstrates the effects of both absorption and
by the symbol K. The absorption rate constant will elimination.
be represented by Ka. This value indicates the frac- Elimination processes begin immediately after
tion of drug present at the absorption site (usually the drug is given. A steeper uphill portion indicates
the GI tract) that is absorbed per unit of time. The a Ka much greater than K, but visual inspection does
usual measurement of Ka is the percentage of drug not provide an accurate assessment of Ka.
absorbed per unit of time. If Ka is greater than one in One way to calculate Ka is to use the method of
a time unit, almost all of the drug would be absorbed residuals, which estimates the plasma drug concen-
over that time interval. tration plot if absorption were instantaneous and
A high Ka (over 1.0 hr–1) indicates rapid absorp- then uses the difference between the actual and
tion. For this explanation, we will assume that estimated concentrations to determine Ka. We first
first-order absorption or elimination rates do not estimate (by back-extrapolation) the straight-line
change with time. Although the rates do not change, portion of the curve (Figure 7-11). The extrapo-
the amount of drug absorbed or eliminated changes. lated portion represents the effect of elimination
alone—as if absorption had been instantaneous.
Clinical Correlate Let us suppose that A, B, and C are actual

measured concentrations and that A′, B ′, and C ′
Some drug absorption rates (Ka) change when are extrapolated concentrations for the same times
large doses are administered as a single oral (Figure 7-12). Points on the extrapolated line can
dose—the percentage of the total dose absorbed be determined visually from the graph or with the
is smaller with a large dose than with a smaller following equation:
dose of the same drug. Gabapentin (Neurontin),
which is actively absorbed via the gut’s l-amino C = (y-intercept) × e–Kt
acid transport system, is a common example of
this absorption phenomenon. Consequently, the
daily dose must sometimes be given in divided
doses, depending on the total daily dose desired.
With an orally administered drug, K is measured

by the slope of the terminal portion of the plasma
drug concentration versus time curve, the time
when absorption no longer has an appreciable FIGURE 7-11.
effect (Figure 7-10). In the first part of the curve Back-extrapolation.

105
Determination of K and Ka can also be used to

predict the resulting plasma drug concentrations
after an oral drug dose. If K, Ka, and the intercepts
of the back-extrapolated line from the drug elimina-
tion phase (B) and the residual line (A) are known,
the plasma drug concentration (C), which repre-
sents the y-intercept, at any time after a single dose
(t) can be calculated:
FIGURE 7-12. C = B e–Kt – A e–Kat

Back-extrapolated concentrations.
Although this equation is similar to the one for
Subtraction of the actual points on the uphill portion
a single IV injection in a one-compartment model
from the corresponding points on the extrapolated
described previously, it accounts for drug yet to be
line (e.g., A′ – A, B ′ – B, and C ′ – C) will yield a new set of
absorbed (–Ae–Kat).
plasma drug concentrations for each time point. These
values can be plotted with the appropriate times, and Plasma drug concentration can also be calcu-
a line is then drawn that best fits the new points. This lated for any given single dose (X0) when K and Ka
new line is called the residual (Figure 7-13). are known and estimates of the bioavailability (F)
and volume of distribution (V) are available:
The slope of the line for these new points gives
an estimate of the absorption rate. Just as the nega- FX 0K a
tive slope of the terminal portion of the plasma =Ct (e −Kt − e −K at )
concentration curve equals K, the negative slope of
V (K a − K )
the residual line equals Ka.
The technique of residuals attempts to separate Ct = concentration at time t,
the two processes of absorption and elimination. F = bioavailability,
These concepts become important when different Xo = amount of drug given orally,
dosage forms of a drug are evaluated. They can Ka = absorption rate constant,
also be used to evaluate the absorption of different
V = volume of distribution,
brands of the same drug in the same dosage form.
A higher Ka indicates a faster absorption rate. This K = elimination rate constant,
factor is only one component of such evaluations, t = time after dose has been given.
but it is often important to know how rapidly a drug
Just as with multiple IV doses, multiple oral
is made available to the systemic circulation. An
doses result in increasing drug concentrations until
overriding assumption of this technique for calcu-
steady state is reached (Figure 7-14). If K, Ka, V, and
lating Ka is that Ka >>> K.
F are known, the steady-state plasma drug concen-
tration at any time (t) after a dose (X0) is given can
also be calculated:
FX 0K a  1 1 
=Ct  − τ
e − Kt
− −K a τ
e − K at 
V (K a − K )  1 − e K
1− e 
These equations are presented to demonstrate

that plasma drug concentrations after oral doses
FIGURE 7-13. can be predicted, but they are infrequently applied
Residual line. in clinical practice.

106

Plasma drug concentration versus time for a typical oral Typical plasma drug concentration versus time curve at steady
formulation given in multiple doses. state for a controlled-release oral formulation.
Controlled-release formulations incorporate

Controlled-Release Products various techniques to slow drug absorption. These
In our discussions of drug absorption so far, it was techniques include the application of coatings that
assumed that the drug formulations used were delay absorption, the use of slowly dissolving salts
relatively rapidly absorbed from the GI tract into or esters of the parent drug, the use of ion-exchange
the systemic circulation. In fact, many drugs are resins that release drug in either acidic or alkaline
absorbed relatively rapidly from the GI tract. With environments, and the use of gel, wax, or polymeric
rapid drug absorption, a peak plasma concentration matrices. Examples of available drugs in controlled-
of drug is evident soon after drug administration release formulations are shown in Table 7-1.
(often within 1 hour) and plasma concentrations Two features of controlled-release products
may decline relatively soon after dose admin- must be considered in therapeutic drug monitoring:
istration, particularly with drugs having short 1. When multiple doses of a controlled-release
elimination half-lives. When drugs are eliminated drug product are administered, before reaching
rapidly from the plasma, a short dosing interval (e.g., steady state, the difference between peak and
every 6 hours) may be required to maintain plasma trough plasma concentrations is not as great
concentrations within the therapeutic range. as would be evident after multiple doses of
To overcome the problem of frequent dosage rapidly absorbed drug products (Figure 7-16).
administration with drugs having short elimination 2. Because the drug may be absorbed for most
half-lives, products have been devised that release of a dosing interval, an elimination phase may
drugs into the GI tract at a controlled rate. These not be as apparent—that is, the log of plasma
controlled-release products (or sustained-release drug concentration versus time curve may not
products) usually allow for less frequent dosage be linear for any part of the dosing interval.
administration. As opposed to the first-order
absorption that occurs with most rapidly absorbed Because, with controlled-release formulations,
oral drug products, some controlled-release drug the drug may be absorbed continuously from the GI
products approximate zero-order drug absorption. tract over the dosing interval, it may not be possible
With zero-order absorption, the amount of drug to calculate a drug’s half-life.
absorbed in a given time remains constant for much
TABLE 7-1. Examples of Controlled-Release
of the dosing interval. The result of zero-order
Formulations
absorption is a more consistent plasma concentra-
tion (Figure 7-15). Drug Formulation
Many types of controlled-release drug products Potassium chloride Wax matrix tablet
have been produced. Products from different manu- Theophylline Coated pellets in tablet
facturers (e.g., theophylline products) that contain Decongestants Coated pellets in capsule
the same drug entity may have quite different Aspirin Microencapsulation
absorption properties, resulting in different plasma
Nifedipine Osmotic pump
concentration versus time curves.

107
For example, if it is known from previous regi-

mens that a patient has a theophylline half-life of 7
hours (K = 0.1 hr–1) and a volume of distribution of
30 L, what dose of a sustained-release preparation
given every 12 hours will be required to achieve an
average plasma concentration of 12 mg/L? Assume
that the product is 90% absorbed.
FIGURE 7-16. First, theophylline clearance must be estimated
Plasma drug concentrations over time with controlled-release as follows:
and rapid-release products.
Clt = K × V
Clinical Correlate = 0.1 hr

−1
× 30 L
= 3.0 L/hr
The peak and trough concentrations of
controlled-release products generally differ very
Then the known variables can be applied:
little, so plasma drug concentration sampling is
generally done at the approximate midpoint of
any dosing interval to approximate the average X 0 ×F
C=
steady-state concentration. τ× Clt
X 0 × 0.9
Some predictions can be made about plasma 12 mg/L =
12 hr × 3 L/hr
drug concentrations with controlled-release prepa-
rations. For preparations that result in continued Rearranging gives:
release of small drug doses, the plasma drug concen-
tration can be estimated as follows: 3 L/hr × 12 hr × 12 mg/L
X0 =
0.9
dose × fraction reaching = 480 mg given every 12 hr
average steady-state = systemic circulation
plasma concentration dosing interval × clearance (may be rounded to 500 mg)
This same equation could be used to estimate drug

or:
clearance if a steady-state plasma drug concentra-
X 0 ×F tion at the midpoint of a dosing interval is known.
C=
τ× Clt Another feature of sustained-release dosage
products is that the drug dose is directly related
(See Equation 4-3.) to the AUC, just as for rapidly absorbed products.
Given this equation, the dose, the amount entering If rapid- and sustained-release products of the
the systemic circulation, dosing interval, and clear- same drug are absorbed to the same extent, then
ance can be used to predict the average steady-state the resulting AUC at steady state for a similar time
plasma drug concentration. Also, if the average will be equivalent for each product if the same
plasma drug concentration is estimated (deter- daily dosages are given. For example, if 500 mg of
mined approximately halfway through a dosing a sustained-release drug product is given every 12
interval), drug clearance can be determined using hours and 250 mg of a rapidly absorbed formula-
the same formula. Finally, the effect of changing the tion of the same drug is given every 6 hours, the AUC
dose or dosing interval on plasma drug concentra- over 12 hours (two dosing intervals for the rapidly
tion can be estimated. absorbed product) should be the same. Again, the

108
assumption is that the bioavailability (F) is the same

for each product. Clinical Correlate
The AUC after administration of a controlled-
The importance of the absorption rate depends to
release dosage formulation is related to drug some extent on the type of illness being treated.
dosage; the relating factor is drug clearance, as For example, when treating pain, it is usually
discussed previously: desirable to use an analgesic that is rapidly
absorbed (i.e., has a high absorption rate constant)
dose × F so that drug effect may begin as soon as possible.
AUC =
clearance For chronic diseases, such as hypertension, it is
more desirable to have a product that results in a
(See Equation 3-5.) lower absorption rate and more consistent drug
absorption over time so that blood pressure does
or: not change over the dosing interval.
dose × F
clearance =
AUC
So if the AUC, dose administered, and fraction

Clinically Important Equation Identified
reaching the systemic circulation are known, drug in This Chapter
clearance can be estimated.
The considerations for controlled-release X0 × F
C=
dosage forms will become increasingly important as τ × Cl t
more drugs are being formulated into preparations
that can be administered at convenient intervals This is a revision of Equation 4-3 in which F is
(daily or even less frequently). incorporated into the equation.

109
REVIEW QUESTIONS
7-1. Absorption of a drug from a tablet form 7-6. Two generic brands of equal strength of
involves dissolution of the solid dosage a drug (as a tablet) are given orally. Tablet
form into GI fluids and diffusion through A results in an AUC of 300 (mg/L) × hour,
body fluids and membranes. whereas tablet B results in an AUC of 500
A.
True (mg/L) × hour. Which product has the better
B.
False bioavailability?
A. tablet A
7-2. The extent to which a drug is absorbed B. tablet B
partially determines its:
A. elimination rate. 7-7. For Figure 7-7, match the sequence of parts
B. bioavailability. of the plasma concentration curve (i.e.,
C. half-life. A, B, or C) with the following description
D. volume of distribution. sequences: (1) The rate of drug absorption
is less than the rate of excretion. (2) The
7-3. Which of the following statements best rate of drug excretion is less than the rate
describes F? of absorption. (3) The rate of drug excretion
A. rate of absorption of the administered equals the rate of absorption.
drug into the systemic circulation A, C, B
A.
B. amount of administered drug that C, A, B
B.
reaches the systemic circulation C, B, A
C.
C. speed at which the administered drug A, B, C
D.
reaches the systemic circulation
D. fraction of the administered drug that Use the following information for Questions 7-8
reaches the systemic circulation through 7-10: A 500-mg oral dose of drug X is given,
and the following plasma concentrations result:
7-4. If 500 mg of a drug is given orally and
Time after Dose
125 mg is absorbed into the systemic circu- Plasma Concentration (mg/L) (hours)
lation, what is F?
0 0
A.
0.6 4.0 0.5
B.
0.5 7.3 1.0
C.
3 9.0 1.5
D.
0.25 8.4 2.0
4.6 4.0
7-5. A drug given intravenously results in an AUC 2.5 6.0
of 400 (mg/L) × hour. If the same dose of
drug is given orally and the resulting AUC is
300 (mg/L) × hour, what percentage of the
7-8. Calculate the elimination rate constant (K)
oral dose reaches the systemic circulation
and the half-life (T ½ ).
and what is the F value?
K = 0.30 hr–1; T ½ = 2.31 hours
A.
75%, F = 0.75
A.
K = 0.03 hr–1; T ½ = 23.1 hours
B.
33%, F = 0.33
B.
K = 0.50 hr–1; T ½ = 1.41 hour
C.
50%, F = 0.5
C.
K = 0.20 hr–1; T ½ = 3.4 hours
D.
0.5%, F = 0.005
D.

110
7-9. Calculate the AUC (from t = 0→∞). On semilog graph paper, draw the back-
A. 8.33 (mg/L) × hour extrapolated line from the terminal portion
B. 32.33 (mg/L) × hour of the curve. Next, using a ruler, draw a line
C. 40.71 (mg/L) × hour directly upward from the times of 0.25, 0.5,
D. 7.1 (mg/L) × hour 0.75, and 1.0 hour; then estimate, on the
concentration scale (y-axis), the concen-
7-10. Calculate the Varea given the above AUC, K, trations from the extrapolated line at the
and dose. following times after drug administration:
A. 24.5 L 0.25, 0.5, 0.75, and 1.0 hour. Which choice
B. 51.5 L below represents these extrapolated values
from each time point?
C. 409 L
D. 40.9 L A. 16.0, 14.0, 12.0, and 10.0 mg/L
B. 20.4, 18.0, 16.0, and 14.1 mg/L
7-11. If two formulations of the same drug are
tested and product A has a faster absorp- C. 18.4, 17.0, 15.0, and 13.1 mg/L
tion rate than product B, product A will D. 18.4, 16.0, 14.0, and 12.1 mg/L
take a shorter amount of time to reach peak
concentration. For each of the four times used in the previous ques-
tion, calculate the corresponding concentrations for
A.
True
the residual line in Questions 7-13 through 7-16.
B.
False
7-13. Time = 0.25 hour
7-12. Refer to Figure 7-17. The plasma concen-
trations and times observed for several A. 14.6 mg/L
points are as follows: B. 8.7 mg/L
Observed Plasma Time after C. 4.9 mg/L
Concentration (mg/L) Dose (hours) D. 2.4 mg/L
3.8 0.25
7.3 0.5 7-14. Time = 0.5 hour
9.1 0.75 A. 14.6 mg/L
9.7 1.0
B. 8.7 mg/L

C. 4.9 mg/L
D. 2.4 mg/L
7-15. Time = 0.75 hour

A. 14.6 mg/L
B. 8.7 mg/L
C. 4.9 mg/L
D. 2.4 mg/L
7-16. Time = 1.0 hour

A. 14.6 mg/L
B. 8.7 mg/L
C. 4.9 mg/L
FIGURE 7-17. D. 2.4 mg/L
Oral absorption plot.

111
The following applies to Questions 7-17 through

7-20. You wish to begin a patient on a sustained-
ANSWERS
release preparation of drug Y and to maintain an
7-1.
A. CORRECT ANSWER. Drug must
average plasma drug concentration of 15 mg/L.
disintegrate before dissolution.
From published data, you estimate V and K for this
B. Incorrect answer
drug to be 12 L and 0.21 hr–1 in this patient, respec-
tively. A. Incorrect answer. Elimination only
7-2.
occurs after drug is absorbed.
7-17. If the fraction of drug absorbed is assumed B. CORRECT ANSWER
to be 1.0 and the drug is to be given every 8 C. Incorrect answer. Half-life is related to
hours, what dose should be administered? elimination rate and occurs only after
A. 302 mg drug is absorbed.
B. 720 mg D. Incorrect answer. Distribution occurs
only after drug is absorbed.
C. 2400 mg
D. 360 mg 7-3. A, B, C. Incorrect answers
D. CORRECT ANSWER
7-18. After 5 days (assume steady state has been
7-4. A, B, C. Incorrect answers
reached), the mid-dose (average) plasma
drug concentration is 12 mg/L. What is the D. CORRECT ANSWER.
patient’s actual drug clearance? F = mg absorbed/total dose,
A. 780 L F = 125 mg/500 mg = 0.25.
B. 3.15 L
7-5. A. CORRECT ANSWER. Percentage of dose
C. 6.42 L reaching blood stream =
D. 20 L AUCpo dose /AUCIV dose
7-19. What should the new daily dose be to result % = (300 mg/L × hr)/(400 mg/L × hr)
in an average plasma drug concentration of = 75%
15 mg/L? B, C, D. Incorrect answers
A. 720 mg 7-6. A. Incorrect answer

B. 552 mg B. CORRECT ANSWER
C. 378 mg 7-7. A, C, D. Incorrect answers
D. 2772 mg B. CORRECT ANSWER
7-20. Finally, due to potential compliance problems 7-8. A. CORRECT ANSWER.

with a three-times-a-day dosing interval, the
physician wishes to prescribe this drug every In 8.4 mg/L − In 2.5 mg/L
K =
12 hours. What dose administered every 6 hr − 2 hr
12 hours would be necessary to achieve a 1.21
steady-state serum level of 15 mg/L? = = 0.30 hr −1
4 hr
A. 125 mg
B. 227 mg 0.693
=
T 12 = 2.31 hr
C. 468 mg 0.30 hr −1
D. 567 mg B, C, D. Incorrect answers

112
7-9. A, B, D. Incorrect answers 7-10. A, C. Incorrect answers. Did you use

C. CORRECT ANSWER. Remember that the K = 0.3 hr –1?
AUC equals the area under the curve Incorrect answer. Did you use
B.
from time zero to 6 hours plus the area AUC = 40.71 (mg/L) × hour?
under the curve from 6 hours to infinity. D. CORRECT ANSWER.
The AUC (6 hours to infinity) is obtained
by dividing the 6-hour concentration by dose
the elimination rate constant (K) from Varea =
AUC × K
above.
AUC from 0 to 6 hours is: 500 mg
= = 40.93 L
40.71 (mg/L × hr) × 0.30 hr −1
0+4
× 0.5 =
1
2 You can see that this value differs consid-
4 + 7.3 erably from Vextrap; Varea is usually a better
× 0.5 =
2.83 estimate.
2
7.3 + 9 7-11. A. CORRECT ANSWER. Drugs that are
× 0.5 =
4.10 absorbed more quickly also reach a
2
higher peak concentration.
9 + 8.4
× 0.5 =
4.35 Incorrect answer
B.
2
8.4 + 4.6 7-12. A, B, C. Incorrect answers
×2 =13.0
2 D. CORRECT ANSWER. Residual concentra-
tion = back-extrapolated concentration
4.6 + 2.5 minus actual concentration.
×2 =7.1
2
AUC hr1+
AUC0→6 0hr→6=
=1+ +2.83 + 4.10 + 4.35 + 13.0 + 7.1 =7-13. A. CORRECT ANSWER. Residual concentra-
2.83 + 4.10 + 4.35 + 13.0 + 7.1 = tion = back-extrapolated concentration
32.38 (mg/L)
32.38 (mg/L) × hr × hr minus actual concentration.
To this value is added AUC6 hr→∞,which is
calculated: 7-14. A, C, D. Incorrect answers
B. CORRECT ANSWER. Residual concentra-
C 6 hr 2.5 mg/L tion = back-extrapolated concentration
= = 8.33 (mg/L) × hr
K 0.30 hr −1 minus actual concentration.
=
AUC0→∞ AUC0→6 hr + AUC6 hr →∞ 7-15. A, B, D. Incorrect answers
=
32.38 + 8.33 (mg/L) × hr C. CORRECT ANSWER. Residual concentra-
= 40.71 (mg/L) × hr minus actual concentration.

113
7-16. A, B, C. Incorrect answers 7-19. A, B, D. Incorrect answers

D. CORRECT ANSWER. Residual concentra- C. CORRECT ANSWER.
minus actual concentration. X 0 ×F
C=
Clt × τ
B, C, D. Incorrect answers. X 0 × 1.0
15 mg/L =
3.15 L/hr × 8 hr
clearance = VK
Rearranging gives:
12 L × 0.21 hr −1 =
= 2.52 L/hr
3.15 L/hr × 8 hr × 15 mg/L
=X0 = 378 mg
Then, the equations below can be used to 1.0
estimate the dose: (given every 8 hr)
X 0 ×F 7-20. A, B, C. Incorrect answers

C=
Clt × τ D. CORRECT ANSWER.
X 0 × 1.0 3.15 L/hr × 12 hr × 15 mg/L

15 mg/L = =X0 = 567 mg
2.52 L/hr × 8 hr 1.0
(given every 12 hr)
Rearranging gives:
15 mg/L × 2.52 L/hr × 8 hr

X0 = 302.4 mg
1.0
(given every 8 hr)

B. CORRECT ANSWER.
We know the following:
τ = 8 hours
X0 = 302 mg
F = 1.0
C = 12 mg/L
And since:
X 0 ×F
C=
Clt × τ
302 mg × 1.0
12 mg/L =
Clt × 8 hr
Rearranging gives:
302 mg × 1.0
=Clt = 3.15 L/hr
12 mg/L × 8 hr

114
Discussion Points
D-1. 500 mg of Drug X is administered by contin- D-5. Find bioavailability data for at least two
uous infusion every 24 hours. A steady-state different brands of the same drug (brand
serum level is reported as 22 mg/L. versus generic, if possible) and describe the
Assuming F = 1, calculate the clearance for bioavailability comparisons made for each
this drug. product.
D-2. Using the clearance value from discussion D-6. For the drug products researched in discus-
point D-1, calculate a new dose adminis- sion point D-5 above, research the U.S. Food
tered by continuous infusion every 24 hours and Drug Administration’s bioequivalence
that would result in a steady-state serum statement. Can these drugs be generically
level of 30 mg/L. substituted and, if so, what data are used to
support this claim?
D-3. Look up the bioavailability for the tablet and
elixir dosage forms of Lanoxin. Plot repre- D-7. Plot (not to scale) the concentration versus
sentation concentration versus time curves time curves for 100 mg of the following four
for these two products at the same dose. oral formulations of a drug and then rank
Discuss all pharmacokinetic differences (from highest to lowest) their relative peak
observed from these plots. concentrations and AUCs. Describe the effects
of variation in these two factors (Ka and F) on
D-4. For the example above (D-3), discuss poten- the concentration versus time curves.
tial advantages and disadvantages of these
two dosage forms. Also, list specific situ- F = 1, Ka = 1 hr –1
A.
ations in which one dosage form might F = 0.7, Ka = 1 hr –1
B.
be preferred or not preferred in a clinical
F = 1, Ka = 0.4 hr –1
C.
dosing situation.
F = 0.7, Ka = 0.4 hr –1
D.

LESSON 8
Drug Distribution and
Protein Binding
OBJECTIVES
1. Describe the major factors that affect drug distribution.
2. Explain the relative perfusion (i.e., high or low) characteristics of various body
compartments (e.g., kidneys, fat tissue, lungs).
3. Describe the three main proteins that bind various drugs.
4. List the major factors that affect drug protein binding.
5. Describe the dynamic processes involved in drug protein binding.
6. Compare perfusion-limited distribution and permeability-limited distribution.
7. Calculate the volume of distribution based on drug protein binding data.
Once a drug begins to be absorbed, it undergoes various transport processes,

which deliver it to body areas away from the absorption site. These transport
processes are collectively referred to as drug distribution and are evidenced by the
changing concentrations of drug in various body tissues and fluids.
Information concerning the concentration of a drug in body tissues and fluids
is limited to a few instances in time (i.e., we know the precise plasma drug concen-
tration only at the few times that blood samples are drawn). Usually, we measure
only plasma concentrations of drug, recognizing that the drug can be present in
many body tissues.
For most drugs, distribution throughout the body occurs mainly by blood
flow through organs and tissues. However, many factors can affect distribution,
including:
• differing characteristics of body tissues,
• disease states that alter physiology,
• lipid solubility of the drug,
• regional differences in physiologic pH (e.g., stomach and urine), and
• extent of protein binding of the drug.
Body Tissue Characteristics

To understand the distribution of a drug, the characteristics of different tissues
must be considered. Certain organs, such as the heart, lungs, and kidneys, are
highly perfused with blood; fat tissue and bone (not the marrow) are much less
115
116
perfused. Skeletal muscle is intermediate in blood Disease States Affecting Distribution

perfusion. The importance of these differences in
perfusion is that for most drugs the rate of delivery Another major factor affecting drug distribution is
from the circulation to a particular tissue depends the effect of various disease states on body physi-
greatly on the blood flow to that tissue. This is called ology. In several disease states, such as liver, heart,
perfusion-limited distribution. Drugs apparently and renal failure, the cardiac output and/or perfu-
distribute more rapidly to areas with higher blood sion of blood to various tissues is altered. A decrease
flow. If the blood flow rate increases, the distribu- in perfusion to the tissues results in a lower rate of
tion of the drug to the tissue increases. distribution and, therefore, a lower drug concentra-
tion in the affected tissues relative to the plasma
Highly perfused organs rapidly attain drug
drug concentration. When the tissue that receives
concentrations approaching those in the plasma;
poor perfusion is the primary eliminating organ, a
less well-perfused tissues take more time to attain
lower rate of drug elimination results, which then
such concentrations. Furthermore, certain anatomic
may cause drug accumulation in the body.
barriers inhibit distribution, a concept referred to
as permeability-limited distribution. This situation
occurs for polar drugs diffusing across tightly knit Lipid Solubility of the Drug
lipoidal membranes. It is also influenced by the oil/
water partition coefficient and degree of ioniza- The extent of drug distribution in tissues also
tion of a drug. For example, the blood–brain barrier depends on the physiochemical properties of the
limits the amount of drug entering the central drug as well as the physiologic functions of the body.
nervous system from the bloodstream. This limita- A drug that is highly lipid soluble easily penetrates
tion is especially great for highly ionized drugs and most membrane barriers, which are mainly lipid
for those with large molecular weights. based, and distributes extensively to fat tissues.
Drugs that are very polar and therefore hydro-
After a drug begins to distribute to tissue, the
philic (e.g., aminoglycosides) do not distribute well
concentration in tissue increases until it reaches an
into fat tissues. This difference becomes important
equilibrium at which the amounts of drug entering
when determining loading dosage requirements of
and leaving the tissue are the same. The drug concen-
drugs in overweight patients. If total body weight is
tration in a tissue at equilibrium depends on the
used to estimate dosage requirements and the drug
plasma drug concentration and the rate at which drug
does not distribute to adipose tissue, the dose can
distributes into that tissue. In highly perfused organs,
be overestimated.
such as the liver, the distribution rate is relatively high;
for most agents, the drug in that tissue rapidly equili-
brates with the drug in plasma. For tissues in which Clinical Correlate
the distribution rate is lower (e.g., fat), reaching equi-
librium may take much longer (Figure 8-1). In general, volume of distribution is based on
ideal body weight for drugs that do not distribute
well into adipose tissue and on total body weight
for drugs that do. If a drug distributes partially
into fat, an adjusted body weight between the
patient’s actual and ideal body weights is often
used. Vancomycin is one notable exception to
this rule; the patient’s total body weight is usually
used to calculate volume of distribution for
vancomycin.
FIGURE 8-1.
Distribution rates.

Lesson 8 | Drug Distribution and Protein Binding
117
Regional Differences in Physiologic pH From this model, it is evident that the volume
of distribution is dependent on the volume of the
Another factor affecting drug distribution is the plasma (3–5 L), the volume of the tissue, the frac-
different physiologic pHs of various areas of the
tion of unbound drug in the plasma, and the fraction
body. The difference in pH can lead to localization
of unbound drug in the tissue. Changes in any of
of drug in tissues and fluids. A drug that is predomi-
these parameters can influence a drug’s volume of
nantly in its ionized state at physiologic pH (7.4)
does not readily cross membrane barriers and distribution. We use this equation to help us under-
probably has a limited distribution. An example of stand why the volume of distribution of a drug may
this phenomenon is excretion of drugs in breast have changed as a consequence of drug interactions
milk. Only un-ionized drug can pass through lipid or disease states. Usually, changes in the volume of
membrane barriers into breast milk. Alkaline drugs, distribution of a drug can be attributed to altera-
which would be mostly un-ionized at pH 7.4, pass tions in the plasma or tissue protein binding of the
into breast tissue. Once in breast tissue, the alkaline drug. This topic is discussed in the next section,
drugs ionize because breast tissue has an acidic pH; Protein Binding.
therefore, the drugs become trapped in this tissue. The clinical consequence of changes in the
This same phenomenon can occur in the urine. volume of distribution of a drug in an individual
Due to the nature of biologic membranes, drugs patient is obvious. An example of this would be
that are un-ionized (uncharged) and have lipo- the use of drug loading doses. Because the initial
philic (fat-soluble) properties are more likely to plasma concentration of the drug (C0) is primarily
cross most membrane barriers. Several drugs (e.g., dependent on the size of the loading dose and
amphotericin) are formulated in a lipid emulsion the volume of distribution (C0 = loading dose/V),
to deliver the active drug to its intended site while changes in either of these parameters could signifi-
decreasing toxicity to other tissues. cantly alter the C0 achieved. Therefore, one must
carefully consider the loading dose of a drug for a
patient whose volume of distribution is believed to
Physiologic Model be unusual.
It is difficult to conceptualize the effect that the Phenytoin is an example of a drug that can
factors discussed above have on the volume of be used to illustrate the effects of changes in the
distribution of a drug. Many of these factors can be factors that determine volume of distribution. For
incorporated into a relatively simple physiologic a typical 70-kg person, the volume of distribution
model. This model describes the critical compo- for phenytoin is approximately 45 L. Generally, the
nents that influence a drug’s volume of distribution.
unbound fraction of this drug in plasma is approxi-
The equation below represents this physiologic
mately 0.1 (90% bound to albumin). If we assume
model and provides a conceptual perspective of the
that the plasma volume is 5 L, the tissue volume is
volume of distribution:
80 L, and the fraction unbound in tissue is 0.2, we
V = Vp + Vt (Fp /Ft ) can estimate how changes in plasma unbound frac-
tion affect volume of distribution.
where:
V = volume of distribution, V = Vp + Vt (Fp /Ft )
Vp = plasma volume,
Vt = tissue volume, = 5 L + 80 L (0.1/0.2)
Fp = fraction of unbound drug in the plasma, and
Ft = fraction of unbound drug in the tissue. = 45 L

118
If the plasma fraction unbound increases to 0.2,

which is possible for patients with hypoalbumin-
emia, the volume of distribution would change as
shown:
V = Vp + Vt (Fp /Ft )
= 5 L + 80 L (0.2/0.2) FIGURE 8-2.

Free drug is available to interact with receptor sites and exert
= 85 L effects.
So, by changing protein binding in the plasma, the Theoretically, drugs bound to plasma proteins
volume of distribution has almost doubled. are usually not pharmacologically active. To exert
an effect, the drug must dissociate from protein
(Figure 8-2).
Protein Binding
Although only unbound drug distributes freely,
Another factor that influences the distribution of drug binding is rapidly reversible (with few excep-
drugs is binding to tissues (nucleic acids, ligands, tions), so some portion is always available as free
calcified tissues, and adenosine triphosphatase) or drug for distribution. The association and dissociation
proteins (albumins, globulins, alpha-1-acid glyco- process between the bound and unbound states is
protein, and lipoproteins). It is the unbound or very rapid and, we assume, continuous (Figure 8-3).
free portion of a drug that diffuses out of plasma. A drug’s protein-binding characteristics depend
Protein binding in plasma can range from 0 to on its physical and chemical properties. Hydrophobic
99% of the total drug in the plasma and varies with drugs usually associate with plasma proteins. The
different drugs. The extent of protein binding may binding of a drug to plasma proteins will primarily
depend on the presence of other protein-bound be a function of the affinity of the protein for the
drugs and the concentrations of drug and proteins drug. The percentage of protein binding of a drug in
in the plasma. plasma can be determined experimentally as follows:
The usual percentages of binding to plasma
proteins for some commonly used agents are shown [total] − [unbound] × 100
% protein binding =
in Table 8-1. [total]
TABLE 8-1. Protein Binding where [total] is the total plasma drug concentration
(unbound drug + bound drug) and [unbound] refers
Drug Binding (%)
to the unbound or free plasma drug concentration.
Ampicillin 18
Chloramphenicol 53
Digoxin 25
Gentamicin < 10
Lidocaine 70
Phenytoin 89
Vancomycin 30
Source: Shargel L, Yu ABC. Applied Biopharmaceutics and

Pharmacokinetics. 3rd ed. Norwalk, CT: Appleton & Lange; 1993.
pp. 594–95. FIGURE 8-3.
Association and dissociation process.


119
Another way of thinking about the relationship TABLE 8-3. Phenytoin Concentration with Regard to
between free and total drug concentration in the Serum Albumin Concentration
plasma is to consider the fraction of unbound drug
Phenytoin Concentration (mg/L)
in the plasma (Fp). Fp is determined by the following
relationship: Total 15.0
Free (normal) 1.5
[unbound]
Fp = Free (hypoalbuminemia) 3.0
[total]
Although the protein binding of a drug will be patients, a lower total phenytoin concentration may
determined by the affinity of the protein for the be effective in controlling seizures.
drug, it will also be affected by the concentration of
the binding protein. Two frequently used methods
for determining the percentage of protein binding Clinical Correlate
of a drug are equilibrium dialysis and ultrafiltration. For certain drugs that are highly protein bound
Three plasma proteins are primarily respon- and have a narrow therapeutic index, it may
sible for the protein binding of most drugs. They be useful to obtain an unbound plasma drug
are shown in Table 8-2 with their normal plasma concentration rather than a total plasma drug
concentration ranges. concentration. This will more accurately reflect
the true concentration of active drug. An example
Although only the unbound portion of drug
of this is phenytoin.
exerts its pharmacologic effect, most drug assays
measure total drug concentration—both bound and
unbound drug. Therefore, changes in the binding The implications of protein binding are not fully
characteristics of a drug could affect pharmacologic understood. The extent of protein binding does not
response to the drug. For example, the anticonvul- consistently predict tissue distribution or half-life
sant and toxic effects of phenytoin are more closely of highly bound drugs. In other words, because an
related to the concentration of free drug in plasma agent has a high fraction bound to protein does not
than to the concentration of total drug in plasma. mean it achieves poor tissue penetration.
In most patients, the free phenytoin concentration Protein binding must be considered in the
is approximately 10% of the total concentration. interpretation of plasma drug concentration data. A
However, in patients with low serum albumin considerable amount of intra- and interpatient vari-
concentrations, a lower fraction of phenytoin is ability exists in the plasma concentration of binding
bound to protein, and the free portion is up to proteins (albumin and alpha-1-acid glycoprotein) as
20% of the total concentration (Table 8-3). With well as their affinity for a specific drug. A major contrib-
hypoalbuminemia, therefore, a patient with a total utor to this variability is the presence of a disease or
phenytoin concentration of 15 mg/L may experi- altered physiologic state, which can affect the plasma
ence side effects (nystagmus and ataxia) usually concentration or affinity of the binding protein. For
seen at a total concentration of 30 mg/L. In these example, albumin concentrations are decreased with
TABLE 8-2. Plasma Protein Plasma Concentrations

Protein Normal Concentration Type of Drugs Bound Example
Albumin 3.5–4.5 g/L Anionic, cationic Phenytoin

Alpha-1-acid glycoprotein 0.4–1.0 g/L Cationic Lidocaine
Lipoproteins Variable Lipophilic Cyclosporine
Source: Shargel L, Yu ABC. Applied Biopharmaceutics and Pharmacokinetics. 3rd ed. Norwalk, CT: Appleton & Lange; 1993. p. 93.

120
hepatic failure, renal dysfunction, burns, stress/trauma, where:

and in pregnancy. Alpha-1-acid glycoprotein concen- V
= volume of distribution,
trations are increased with myocardial infarction, Vp = plasma volume,
renal failure, arthritis, surgery, or stress/trauma. In Vt = tissue volume,
addition, concomitant administration of a displacer
Fp = fraction of unbound drug in the plasma, and
drug (i.e., an agent that competes with the drug of
Ft = fraction of unbound drug in the tissue.
interest for common protein binding sites) can alter
the protein binding of a drug. Examples of displacer How can a drug’s volume of distribution be
drugs include salicylic acid and valproic acid. altered by the administration of other drugs,
Changes in plasma protein binding of drugs can diseases, or an altered physiologic state? The
have considerable influence on therapeutic or toxic unbound fraction in the plasma and tissue is depen-
effects that result from a drug regimen. Provided dent on both the quantity (concentration) and
below are practical considerations regarding quality (affinity) of the binding proteins; therefore,
plasma protein binding, with examples of specific changes in these parameters can alter the volume of
agents for which these considerations are impor- distribution. Four examples are briefly discussed to
tant to therapeutics. demonstrate the potential consequences of altered
protein binding on a drug’s volume of distribution.
The following questions should be considered
when assessing the clinical importance of protein
binding for a given drug: EXAMPLE 1.
• Does the drug possess a narrow therapeutic Plasma Protein Binding Drug Interaction:
index? Effect of Valproic Acid Administration on
• Is a high fraction of the drug bound to Volume of Distribution of Phenytoin
plasma protein? Assuming that Vp and Vt are unchanged as a
• Which plasma protein is primarily respon- consequence of valproic acid administration,
sible for binding, and does it account for the let’s consider the effect of valproic acid on the
majority of the drug’s binding variability? protein binding of phenytoin. Both phenytoin
and valproic acid are highly protein bound
Answers to these questions will help you establish (approximately 90%) to the same site on the
a basis on which to evaluate the clinical significance plasma albumin molecule. When these drugs
of changes in plasma protein binding due to drug– are administered concomitantly, the protein
drug or drug–disease state interactions. binding of phenytoin is reduced (e.g., from
In addition to having an impact on the interpre- 90% to 80%). This is an example of displace-
tation of a drug’s steady-state plasma concentration ment, or reduction in the protein binding of
data, changes in plasma and tissue protein binding a drug due to competition from another drug
can have a major influence on clearance and volume (i.e., the displacer). In this case, valproic acid
of distribution. The remainder of this lesson discusses has a higher affinity for the plasma protein
the effect that changes in a drug’s protein binding binding site on the albumin molecule and
will have on the apparent volume of distribution of a competitively displaces phenytoin, resulting
drug. The ramifications of altered protein binding on in a higher fraction of unbound phenytoin.
drug clearance are discussed in Lesson 9. What is the consequence of phenytoin having
The consequence of protein binding changes a higher unbound fraction due to plasma
on volume of drug distribution was implied in this protein binding displacement by valproic
equation shown earlier in this lesson: acid? The equation above would predict that
an increase in the unbound fraction in the
V = Vp + Vt (Fp /Ft ) plasma would result in an increase in pheny-

121
toin’s volume of distribution and result in a Drug–drug interactions are not the only way
lower plasma drug concentration: a drug’s apparent volume of distribution can be
altered. In Example 3, we next consider the effect of
Fp ( ↑ ) a disease state (chronic renal failure) on the volume
Vp ( ↔ ) + Vt ( ↔ ) = V (↑)
Ft ( ↔ ) of distribution of phenytoin and digoxin.
EXAMPLE 2. EXAMPLE 3.
Tissue Binding Drug Interaction: Effect of Effect of Disease State on Volume of
Quinidine Administration on Volume of Distribution: Renal Failure and Volume of
Distribution of Digoxin Distribution of Phenytoin
As in Example 1, we will assume that Vp and Vt Assuming that Vp and Vt are unchanged as a
are unchanged as a result of quinidine admin- consequence of renal failure, let’s consider
istration. Digoxin is negligibly bound to plasma the consequences of this disease state on
proteins (approximately 25%), whereas the protein binding of phenytoin. Pheny-
70–90% of quinidine is bound to plasma toin’s plasma protein binding is dependent
albumin and alpha-1-acid glycoprotein. Digoxin on both the quantity and quality of albumin.
normally has a very large apparent volume of Because chronic renal failure reduces
distribution (4–9 L/kg), which suggests exten- albumin concentrations as well as albumin’s
sive tissue distribution. Digoxin is signifi- affinity for phenytoin, it is not surprising
cantly associated with cardiac muscle tissue, that the plasma protein binding of phenytoin
as demonstrated by a 70:1 cardiac muscle to could be reduced from approximately 90%
plasma digoxin concentration ratio, which to 80%. What is the consequence of phenyt-
explains why its volume of distribution exceeds oin’s higher unbound fraction (0.2 [renal
any normal physiologic space. failure] versus 0.1 [normal]) due to renal
When these drugs are administered concomi- failure?
tantly, the tissue binding of digoxin is reduced.
The equation below predicts that an increase
This is also an example of displacement but,
in the unbound fraction in the plasma would
in this case, quinidine has a higher affinity for
result in an increase in the volume of distri-
the tissue protein binding site and displaces
digoxin, resulting in a high unbound fraction bution of phenytoin, which would increase
in the tissue. What are the consequences of the concentration of the active unbound
digoxin having a higher unbound fraction in phenytoin able to cross the blood–brain
the tissue due to quinidine displacement? The barrier. This increase could result in supra-
equation given previously predicts that an therapeutic unbound concentrations, even
increase in the unbound fraction in the tissue when the total concentration is within
would result in a decrease in the volume of normal limits:
distribution of digoxin, thus increasing digox-
in’s plasma drug concentration: Fp ( ↑ )
Fp ( ↔ ) Vp ( ↔ ) + Vt ( ↔ ) = V (↑)
Vp ( ↔ ) + Vt ( ↔ ) = V (↓ ) Ft ( ↔ )
Ft (↑ )

122
EXAMPLE 4. In all of these examples, the volume of distri-

bution of the drug in question was altered as a
Effect of Disease State on Volume of consequence of a drug–drug or drug–disease state
Distribution: Renal Failure and Volume of interaction. Consequently, the calculation of their
Distribution of Digoxin loading dose (X0 = C0V) is influenced by changes in
As in Example 3, we will assume that Vp and a drug’s plasma or tissue protein binding. This must
Vt are unchanged as a consequence of renal be considered in the development of a patient’s
failure. Because digoxin is negligibly bound drug dosing regimen.
to plasma proteins, changes in its concen-
tration should not be of clinical significance.
However, renal failure does reduce the
cardiac muscle-to-plasma digoxin concen-
tration ratio to 30:1. What is the conse-
quence of digoxin’s higher unbound fraction
in the tissue due to renal failure? The equa-
tion below predicts that an increase in the
unbound fraction in the tissue would result
in a decrease in the volume of distribution of
digoxin and may cause an increased plasma
digoxin drug concentration:
Fp ( ↔ )
Vp ( ↔ ) + Vt ( ↔ ) = V (↓ )
Ft (↑ )

123
REVIEW QUESTIONS
8-1. Drugs that are very lipid soluble tend to 8-7. Cationic drugs and weak bases are more
distribute poorly into body tissues. likely to bind to:
A. True A. globulin.
B. False B. alpha-1-acid glycoprotein.
C. lipoprotein.
8-2. Drugs that are predominantly un-ionized at
D. A and C.
physiologic pH (7.4) have a limited distri-
bution when compared to drugs that are
8-8. Anionic drugs and weak acids are more
primarily ionized.
likely to bind to:
A. True
A. albumin.
B. False
B. globulin.
8-3. Drugs are generally less well distributed C. alpha-1-acid glycoprotein.
to highly perfused tissues (compared with D. lipoprotein.
poorly perfused tissues).
A. True 8-9. Predict how the volume of distribution (V)
would change if the unbound fraction of
B. False
phenytoin in plasma decreased from 90%
to 85%. Assume that unbound fraction in
8-4. Estimate the volume of distribution for a
tissues (Ft ) and volumes of plasma (Vp) and
drug when the volume of plasma and tissue
tissues (Vt ) are unchanged.
are 5 and 20 L, respectively, and the fraction
of drug unbound in plasma and tissue are A. increase
both 0.7. B. no change
A. 18.5 L C. decrease
B. 100 L D. cannot be predicted with the
C. 30 L information provided
D. 25 L
8-10. A new drug has a tissue volume (Vt ) of 15 L,
an unbound fraction in plasma (Fp) of 5%,
8-5. The portion of drug that is bound to plasma
and an unbound fraction in tissues (Ft ) of
protein is pharmacologically active.
5%. What will be the resulting volume of
A. True distribution if the plasma volume (Vp) is
B. False reduced from 5 to 4 L?
A. 13 L
8-6. Penetration of drug into tissues is not related
to the extent bound to plasma proteins. B. 19 L
A. True C. 18 L
B. False D. 5 L

124
8-11. How is the volume of distribution (V) of 8-6. A. Incorrect answer

digoxin likely to change if a patient has been B. CORRECT ANSWER
taking both digoxin and quinidine and the
quinidine is discontinued? Assume that 8-7. A, C, D. Incorrect answers
plasma volume (Vp ), tissue volume (Vt ), and B. CORRECT ANSWER
unbound fraction of drug in plasma (Fp) are
unchanged. 8-8. A. CORRECT ANSWER
A. increase B, C, D. Incorrect answers
B. no change
C. decrease
C. CORRECT ANSWER. If fraction bound
D. cannot be predicted with the
decreases, then V will also decrease:
information provided
F 
V= Vp + Vt  p 
ANSWERS  Ft 
If Fp is decreased,

B. CORRECT ANSWER Fp ( ↓ )
Vp ( ↔ ) + Vt ( ↔ ) = V (↓)
Ft ( ↔ )
B. CORRECT ANSWER
8-3. A. Incorrect answer B. CORRECT ANSWER. Solve the equation
B. CORRECT ANSWER using Vp = 5 L, then re-solve using 4 L
and compare:
F 
D. CORRECT ANSWER. V= Vp + Vt  p 
 Ft 
F 
V= Vp + Vt  p 
 0.05 
 Ft  =
5 L + 15 L  =20 L
 0.05 
 0.7 
=
5 L + 20 L  =25 L
 0.7  If Vp is decreased to 4 L,
8-5. A. Incorrect answer  0.05 

B. CORRECT ANSWER V =
4 L + 15 L  =19 L
 0.05 

125
8-11.
A. CORRECT ANSWER. Remember, when When quinidine is discontinued, the
quinidine is administered concomi- unbound fraction of digoxin in the
tantly with digoxin, quinidine competes tissues (Ft ) decreases as the tissue
with digoxin for tissue binding sites binding sites formerly occupied by quin-
and increases the unbound fraction of idine become available.
digoxin in the tissues (Ft). Therefore,
Fp ( ↔ )
assuming Vp and Vt remain unchanged, Vp ( ↔ ) + Vt ( ↔ ) = V (↑ )
the effect of quinidine is shown below: Ft (↓ )
Fp ( ↔ ) Therefore, the volume of distribution

Vp ( ↔ ) + Vt ( ↔ ) = V (↓ ) will increase.
Ft (↑ )


126
Discussion Points
D-1. Describe how knowledge of a drug’s D-4. Draw representative concentration versus
distribution and lipid solubility affect the time curves for: (a) a drug that diffuses into
calculation of a drug’s loading dose. Clini- highly vascularized tissue before equili-
cally, what type of loading dose adjustments brating in all body compartments, and (b)
can be made to account for these factors? a drug that distributes equally well into all
body compartments. Describe how these
D-2. A patient has a total plasma phenytoin curves differ and discuss potential clinical
concentration of 19 mcg/mL with a serum implications.
albumin concentration of only 2.5 g/dL.
Estimate this patient’s bound and unbound D-5. Discuss major physiologic and physio-
phenytoin concentration. chemical factors that affect a drug’s
distribution and comment on how these
D-3. In the same patient as described in discus- factors can affect the pharmacokinetic
sion point D-2, calculate a new total variable apparent volume of distribution.
phenytoin concentration that would yield a
therapeutic unbound phenytoin concentra-
tion.

LESSON 9
Drug Elimination Processes
OBJECTIVES
1. Describe the impact of disease and altered physiologic states on the clearance and
dosing of drugs.
2. Identify the various routes of drug metabolism and excretion.
3. Explain the two general types (Phase I and II) of drug metabolism.
4. Define the methods of hepatic drug metabolism and the approaches used to quantitate
and characterize this metabolism.
5. Describe the effects of a drug’s hepatic extraction ratio on that drug’s removal via the
liver’s first-pass metabolism.
6. Explain the various processes involved in renal elimination (i.e., filtration, secretion,
and reabsorption).
7. Define both the physiologic and mathematical relationship of drug clearance to
glomerular filtration.
Drug Elimination
The liver and kidneys are the two major organs responsible for eliminating drugs
from the body. Although both organs share metabolic and excretory functions,
the liver is principally responsible for metabolism and the kidneys for elimina-
tion. The importance of these organs cannot be overestimated in determining
the magnitude and frequency of drug dosing. Additionally, an appreciation of the
anatomy and physiology of these organs will provide insight into the impact of
disease and altered physiologic states, as well as concomitant drug administra-
tion, on the clearance and dosing of drugs.
The physical and chemical properties of a drug are important in determining
drug disposition. For example, lipophilic drugs (compared with hydrophilic drugs)
tend to be:
• bound to a greater extent to plasma proteins,
• distributed to a greater extent throughout the body, and
• metabolized to a greater extent in the liver.
Hydrophilic drugs, particularly ionized species, tend to have more limited distribu-
tion and more rapid elimination (often by renal excretion).
127
128
Drug elimination from the body can be very

complex. Metabolism (also known as biotransforma-
tion) involves conversion of the administered drug
into another substance. Metabolism can result in the
formation of either an active or inactive metabolite,
which may then be excreted either faster or slower
than the parent compound. The various conse-
quences of hepatic biotransformation include active
drug to inactive metabolite, active drug to active FIGURE 9-1.
metabolite, and inactive drug to active metabolite. Plasma concentration versus time curve for drug X, primarily
Two examples of drugs with active metabolites metabolized by the liver.
are carbamazepine and its active metabolite carba-
mazepine-10, 11-epoxide and prednisone and its be Km and not Kr. This result occurs because the plasma
active metabolite prednisolone. For both drugs, the concentration of metabolite Y is determined by the rate
metabolites formed are active and may contribute of formation from drug X (the slower rate constant).
significantly to the patient’s pharmacologic On the other hand, if Kr is much less than Km,
response. Consequently, the plasma concentration the terminal slope of the plasma metabolite Y
of an active metabolite must be considered in addi- concentration versus time plot will be Kr (Figure
tion to that of the parent compound when predicting 9-3). In this case, the relatively slow renal elimi-
overall pharmacologic response. In addition, many nation of metabolite Y determines the resulting
drugs are actually pro-drugs, which require activa- plasma concentrations. Although the liver is the
tion to their active forms. An example of a pro-drug major organ of drug biotransformation, the intes-
is sulfasalazine, which is a pro-drug that is cleaved tines, kidneys, and lungs may also metabolize some
by colonic bacterial reductases to the antibacte- drugs. Before we can develop the concepts of drug
rial agent sulfapyridine and the anti-inflammatory metabolism, we must first examine the anatomy,
agent 5-aminosalicylic acid. physiology, and fundamental functions of the liver.
Let’s next explore two introductory concepts The adult liver weighs 1400–1600 g and is uniquely
with regard to metabolite pharmacokinetics. situated between the gastrointestinal (GI) tract and
Consider the following situation: the systemic circulation (Figure 9-4).
The basic functional unit of the liver is the liver
kidney
liver→ metabolite Y  lobule (Figure 9-5). The human liver contains
drug X  → excreted Y
Km Kr approximately 50,000–100,000 such lobules. The
liver lobule is constructed around a central vein,
where drug X represents the intravenous bolus
administration of the compound and Km and Kr
represent the elimination rate constants for hepatic
metabolism of drug X and renal excretion of metab-
olite Y, respectively.
Figure 9-1 shows the decline in plasma
concentration of parent drug X (assuming a
one-compartment model after intravenous admin-
istration). Now consider the profile of metabolite
Y after the same dose of drug X (Figure 9-2). If
the excretion rate constant of metabolite Y (Kr) is
much greater than the elimination rate constant of FIGURE 9-2.
Plasma concentrations of metabolite Y (from drug X) when the
drug X (Km), the terminal slope of the natural log of elimination rate constant (Kr ) of metabolite Y is greater than the
concentration of metabolite Y versus time plot will rate constant for metabolism (Km ) of drug X.

Lesson 9 | Drug Elimination Processes
129
FIGURE 9-3.
Plasma concentrations of metabolite Y (from drug X) when the
elimination rate constant (Kr ) of metabolite Y is less than the
rate constant for metabolism (Km ) of drug X.
which empties into the hepatic veins and the vena FIGURE 9-5.
Basic structure of a liver lobule showing the hepatic cellular
cava. Therefore, the hepatic cells (hepatocytes),
plates, blood vessels, bile-collecting system, and lymph flow
which are principally responsible for metabolic system composed of the spaces of Disse and interlobular
functions (including drug metabolism), are exposed lymphatics.
to portal blood. Source: Reproduced with permission from Guyton AC. Textbook of
The liver (ultimately the liver lobule) receives Medical Physiology. 7th ed. Philadelphia, PA: WB Saunders; 1986.
its blood supply from two separate sources: the
portal vein and the hepatic artery. The liver receives After entering the liver, blood flows in the veins
approximately 1100 mL/minute of blood from the and arteries of the portal triads, enters the sinu-
portal vein and 350 mL/minute of blood from the soidal spaces of the liver, and exits via the central
hepatic artery. Consequently, blood flow in a normal hepatic vein. In the sinusoids, the drug is trans-
70-kg adult is approximately 1450 mL/minute. ferred from the blood to the hepatocytes, where it is
metabolized or excreted unchanged into the biliary
system (Figure 9-6).
The liver is involved in numerous functions,
including storage and filtration of blood, secretion
and excretion processes, and metabolism. In clinical
pharmacokinetics, we are primarily interested in
the last role, drug metabolism, and the factors that
influence it. It is generally recognized that wide
interpatient and intrapatient variability exists in the
biotransformation of most drugs. It is also accepted
that changes in liver function may greatly alter the
extent of drug elimination from the body. To appre-
ciate the importance of these functions and patient
factors in the metabolism of a specific drug, it is
necessary to understand the mechanisms involved
in hepatic drug metabolism and the relative ability
of the liver to extract that particular drug from the
blood into the hepatocyte.
FIGURE 9-4.
Portal and hepatic circulations. Hepatic metabolism occurs in two phases called
Source: Reproduced with permission from Guyton AC. Textbook biotransformation and conjugation. During Phase
of Medical Physiology. 7th ed. Philadelphia, PA: WB Saunders; I, biotransformation, drugs undergo oxidation,
1986. reduction, or hydrolysis to become more hydro-

130
enzymes most important in human drug metabo-

lism are CYP1, CYP2, and CYP3. In addition to
the action on specific drug substrates, these iso-
enzymes can also be either induced or inhibited
by other drugs, thus increasing or decreasing the
plasma concentration of the drug they metabolize.
This can have clinical significance for drugs whose
concentration-dependent effects are significantly
affected by enzyme inhibition or induction. Table
9-1 lists common drug-metabolizing isoenzymes
and the drugs most commonly affected, as well as
other drugs that can either inhibit or induce the
drug’s metabolism by this enzyme.
Another point regarding Phase I biotrans-
formation reactions is that select drugs may be
metabolized by more than one cytochrome P450
isoenzyme. An example is tricyclic antidepressants.
FIGURE 9-6.
Representation of drug metabolism and excretion by the
Most of these agents are hydroxylated by CYP2D6;
hepatocyte. Qh = hepatic blood flow. however, N-demethylation is probably mediated by
a combination of CYP2C19, CYP1A2, and CYP3A4.
philic. During Phase II, conjugation, drugs receive Acetaminophen, another example, appears to be
a molecular attachment (i.e., glucuronate) that metabolized by both CYP1A2 and CYP2E1.
facilitates transport within the body. Drugs may be Phase II reactions, also called synthetic (or conju-
subjected to either type of reaction, but commonly gation) reactions, result in very polar compounds
drugs undergo Phase I (i.e., preparatory) reac- that are easily excreted in the urine. Examples of
tions followed by Phase II reactions. The majority drugs that undergo Phase I or Phase II reactions are
of drug–drug or drug–nutrient interactions occurs shown in Table 9-2.
during Phase I (biotransformation) and involves Understanding whether a drug undergoes Phase
either the inhibition or induction of the CYP iso- I or Phase II biotransformation may be helpful in
enzyme involved in the drug’s metabolism. predicting how it will be affected by a certain disease
The major hepatic enzyme system responsible state. For example, liver disease and the aging
for Phase I metabolism is called the cytochrome P450 process appear to reduce the elimination of drugs
enzyme system, which contains many isoenzyme that undergo Phase I metabolism more than those
subclasses with varying activity and specificity in dependent on conjugation (Phase II) reactions. This
Phase I drug metabolism processes. Cytochrome fact raises a significant question: at what point does
P450 isoenzymes are grouped into families liver disease significantly alter Phase I metabolic
according to their genetic similarities. Enzymes processes? No single test can accurately estimate
with greater than 40% of their genes in common liver drug-metabolism capacity. High values for alka-
are considered to be from the same family and line phosphatase, aspartate aminotransferase (AST),
are designated by an Arabic number (e.g., 1, 2, 3), and alanin aminotransferase (ALT) usually indicate
and those enzymes within each family that contain acute cellular damage and not poor liver drug-metab-
greater than 55% common genes are given a sub- olism capacity. On the other hand, abnormal values
family designation using a capital letter (e.g., A, B, that may suggest this are elevated serum bilirubin
C). Finally, those enzymes with greater than 97% concentrations, low serum albumin concentrations,
common genes are further classified with another and a prolonged prothrombin time. The Child-Pugh
Arabic number and often represent a very specific Score, a widely utilized clinical assessment tool for
drug-metabolizing enzyme. The cytochrome P450 liver disease, may also be used to evaluate a patient’s

131
TABLE 9-1. Drug-Metabolizing Enzymes and Selected Inhibitors and Inducers

Isozyme Drug Inhibitors Inducers
CYP1A2 Caffeine, tacrine, theophylline, Cimetidine, ciprofloxacin, erythro- Omeprazole, tobacco,

lidocaine, R-warfarin mycin, fluvoxamine, tacrine, carbamazepine, nafcillin,
zafirlukast broccoli
CYP2B6 Cocaine, ifosfamide, cyclophosphamide Chloramphenicol Phenobarbital, rifampin
CYP2C9 S-warfarin, phenytoin, diclofenac, Amiodarone, fluconazole, lovastatin, Rifampin, phenobarbital,

piroxicam clopidogrel, leflunomide secobarbital
CYP2C19 Diazepam, omeprazole, mephenytoin Fluvoxamine, fluoxetine, omeprazole, Carbamazepine, prednisone,

oxcarbazepine rifampin, phenobarbital
CYP2D6 Codeine, haloperidol, dextromethorphan, Bupropion, cinacalcet, quinidine, Rifampin, dexamethasone
tricyclic antidepressants, phenothiazines, fluoxetine, sertraline, amiodarone,
metoprolol, propranolol, risperidone, propoxyphene
paroxetine, sertraline, venlafaxine
CYP2E1 Acetaminophen, alcohol Disulfiram Isoniazid, alcohol

CYP3A3/4/5/7 Nifedipine, verapamil, cyclosporine, Erythromycin, cimetidine, clari- Carbamazepine, rifampin,
carbamazepine, astemizole, tacrolimus, thromycin, fluvoxamine, fluoxetine, phenytoin, phenobarbital,
midazolam, alfentanil, diazepam, lorata- ketoconazole, itraconazole, isoniazid, St. John’s wort
dine, ifosfamide, cyclophosphamide grapefruit juice, metronidazole,
ritonavir, indinavir
ability to metabolize drugs eliminated by the liver. A Biotransformation

score of 8 to 9 indicates the need to initiate therapy
at moderately decreased initial doses (~25%) for Biotransformation processes are affected by many
drugs primarily (> 60%) metabolized hepatically, factors. The functioning of metabolic enzyme
while a score of > 10 suggests a significant decrease systems may be quite different at the extremes of
(~50%) in initial doses of drugs primarily metabo- age. Historically, neonates were at risk of toxicity
lized by the liver. from chloramphenicol because they do not conju-
gate this drug efficiently. Also, the social habits of a
Membrane transport proteins are membrane-
patient may affect drug elimination. Alcohol use and
spanning substances that facilitate drug transport
smoking may increase hepatic clearance of some
across the intestinal tract, excretion into the bile
drugs by inducing metabolic enzymes. Obviously,
and urine, distribution across the blood–brain
disease states such as cirrhosis and conditions that
barrier and drug uptake into target cells. A major
decrease liver blood flow (e.g., heart failure) signifi-
transport protein is P-glycoprotein. Commonly
cantly affect drug metabolism. Finally, concomitant
utilized agents that are affected by this protein
drug use may affect drug metabolism. Certain drugs,
include clopidogrel, digoxin, diltiazem, glyburide,
such as phenytoin and phenobarbital, may induce
and morphine. Increased or decreased expression
hepatic enzymes, whereas other drugs, such as
of this substance can alter absorption, elimination,
cimetidine and valproic acid, may inhibit them.
and serum concentrations of relevant drugs. Other
membrane-transporter families include organic Even in healthy individuals, in the absence of
anion transporters (OAT family), the organic anion hepatic enzyme inducers or inhibitors, the ability
transporting polypeptides (OATP family), and the to metabolize drugs may vary considerably due to
organic cation transporters (OCT family). individual genetic makeup. Genetic polymorphism

132
TABLE 9-2. Drugs Undergoing Phase I or II Hepatic Clearance

Metabolizing Reactions
Now let’s focus on hepatic drug metabolism and the
Phase I Reactions Examples approaches used to quantitate and characterize this
Oxidation: process. Depending on physical and chemical prop-
Hydroxylation Cyclosporine, ibuprofen, phenytoin, erties, each drug is taken up or extracted by the
acetaminophen (also Phase II) liver to different degrees. Knowledge of the affinity
Dealkylation Diazepam, imipramine, tamoxifen of a drug for extraction by the liver is important in
anticipating the influence of various factors on drug
Deamination Amphetamine, diazepam
metabolism. Generally, drugs are characterized as
Sulfoxidation Chlorpromazine, cimetidine,
possessing a low to high affinity for extraction by
omeprazole
the liver. Briefly, drugs with a low hepatic extraction
Reduction Sulfasalazine, chloramphenicol (< 20%) tend to be more available to the systemic
Hydrolysis Aspirin, carbamazepine, enalapril circulation and have a low systemic clearance. Drugs
with a high hepatic extraction (> 80%) tend to be
Phase II Reactions Examples
less available to the systemic circulation and have
Glucuronidation Acetaminophen (also Phase I), lorazepam, a high systemic clearance. Drugs with extraction
morphine, chloramphenicol ratios between 20 and 80 are termed intermediate-
Methylation Captopril, levodopa, methyldopa extraction drugs. These points will become more
Acetylation Clonazepam, corticosteroids, apparent as we develop a mathematical model to
dapsone, isoniazid, sulfonamides relate a drug’s hepatic clearance to hepatic physi-
ology.
The efficiency of the liver in removing drug from
can affect the individual response to a drug. For the bloodstream is referred to as the extraction ratio
example, approximately one-third of Caucasians (E), the fraction of drug removed during one pass
carry at least one variant allele for the gene that through the liver. The value of E theoretically ranges
encodes CYP2C9 involved in the metabolism of from 0 to 1. With high-extraction drugs, E is closer
warfarin. Presence of this polymorphism increases to 1, and with low-extraction drugs, E is closer to
the anticoagulant effect of warfarin, thus requiring zero. The reader may wish to refer to the discussion
lower warfarin doses. Investigators have also of E in Lesson 2.
shown that two distinct subpopulations have In Lesson 1, we learned that the concentration
varying capacities for drug acetylation (Phase II of drug in the body was dependent on the dose of
reaction) as a result of genetic polymorphism. the drug administered and the volume into which
Fast acetylators have a greater rate of elimination the agent was distributed. This was represented by
for drugs such as isoniazid and hydralazine. For the equation:
slow acetylators, the usual doses of these agents
may result in excessive plasma concentrations X
C=
and, therefore, increased drug toxicities. Further V
discussions regarding genetic alteration of drug
metabolism can be found in Lesson 11. (See Equation 1-1.)

133
In Lesson 2, we further discovered that steady- functions. These models consider three factors:
state plasma drug concentrations are affected by 1. the liver’s innate ability to remove unbound
several variables, including the rate at which a drug drug from plasma irreversibly,
is administered and the drug’s clearance. This rela-
2. the fraction of drug unbound in the blood,
tionship is demonstrated in the following equation:
and
K0 3. hepatic blood flow.
C ss =
Clt One practical and useful model is called the jar,
where: venous equilibrium, or well-stirred model:
Clt = the total body clearance of the drug,
Qh Fp Cli
K0 = the drug infusion rate, and Clh =
Qh + Fp Cli
Css = the steady-state plasma drug concentration.
The factors that determine the extraction ratio where:
and its relationship to overall hepatic clearance can Clh = hepatic drug clearance,
be shown mathematically as: Fp = fraction of free drug in plasma,
Cli Cli = intrinsic clearance, and
E=
Qh + Cli Qh = hepatic blood flow.
Recall that Clh equals Clt for drugs eliminated

(See Equation 2-2.)
only by the liver. Therefore, changes in any of the
where: parameters defined in the previous equation will
Cli = intrinsic clearance and have a considerable impact on Clt and, consequently,
Qh = hepatic blood flow. the steady-state drug plasma concentration produced
by a given dosing regimen. In a normal 70-kg indi-
Because Clh = Qh × E, then: vidual, Qh (portal vein plus hepatic artery blood
flows) should approach 1500 mL/minute. Obviously,
Qh × Cli changes in Qh would change the rate of drug delivery
Clh =
Qh + Cli to the liver and have an impact on Clh . However,
the magnitude of that impact would depend on the
The systemic clearance of a drug relates dosing liver’s ability to extract the drug. Fp is incorporated
rate to a steady-state plasma drug concentration. into the relationship because only free or unbound
The systemic clearance of a drug equals the hepatic drug is available to be metabolized by the hepato-
clearance when the liver is the sole organ respon- cytes. Finally, intrinsic clearance (Cli) represents
sible for elimination. Another way of looking at this the liver’s innate ability to clear unbound drug from
relationship is to remember that clearance terms intracellular water via metabolism or biliary excre-
are additive. Therefore: tion. Changes in Cli should have a profound effect on
hepatic clearance. However, as with Qh , the extent
Clt = Clh + Clr + Clother organs (See Equation 2-1.)
and magnitude of such an effect would depend on the
extraction characteristics of the drug.
and Clt is equal to Clh when Clr and Clother organs are Examination of the equation for the venous equi-
minimal. librium model at the extremes of intrinsic clearance
For a drug that is totally dependent on the liver values provides insight into the influences of hepatic
for its elimination, a number of useful mathemat- blood flow and intrinsic clearance on drug dosing.
ical models show critical relationships between For high intrinsic clearance drugs, Cli is much greater
systemic drug clearance and various physiologic than Qh ; Qh becomes insignificant when compared to

134
Cli . Hepatic clearance of drugs with high extraction TABLE 9-3. High-, Intermediate-, and
ratios (> 0.8) is dependent on hepatic blood flow only. Low-Extraction Drugs
It is not influenced by protein binding or enzymes.
High Intrinsic Intermediate Low Intrinsic
Therefore, when Cli is large, Clh equals Qh, or
Clearance Intrinsic Clearance
hepatic clearance equals hepatic blood flow. Hepatic (Cli >> Qh ) Clearance (Cli << Qh )
clearance is essentially a reflection of the delivery
rate (Qh) of the drug to the liver; changes in blood Propranolol Aspirin Warfarin
flow will produce similar changes in clearance. Lidocaine Quinidine Phenytoin
Consequently, after intravenous administration, the Propoxyphene Desipramine Isoniazid
hepatic clearance of highly extracted compounds
Morphine Theophylline
(e.g., lidocaine and propranolol) is principally
dependent on liver blood flow and independent Meperidine Diazepam
of both free fraction and intrinsic clearance. This Nitroglycerin Procainamide
particular commonly used model is best applied
Isoproterenol Antipyrine
to intravenously administered drugs, as orally
absorbed drugs with high extraction ratios may act Pentazocine Phenobarbital
more like low-extraction drugs. Other models may Verapamil Erythromycin
work better in these cases.
For low intrinsic clearance drugs, Qh is much
greater than Cli . Therefore, the hepatic clearance First-Pass Effect
of compounds with a low extraction ratio (e.g.,
phenytoin) is virtually independent of hepatic blood An important characteristic of drugs having a high
flow. Hepatic clearance for these drugs becomes a extraction ratio (e.g., propranolol) is that, with
reflection of the drug’s intrinsic clearance and the oral administration, a significant amount of drug is
free fraction of drug in the plasma. metabolized before reaching the systemic circula-
Some examples of individual intrinsic clear- tion (Figure 9-7). Drug removal by the liver after
ances are given in Table 9-3. However, there is no absorption is called the first-pass effect. The result
clear-cut division between the classes described; can be that the amount of drug reaching the systemic
additional factors may need to be considered when circulation is considerably less than the dose given.
predicting drug disposition. The first-pass effect becomes obvious when
we examine comparable intravenous and oral
doses of a drug with a high extraction ratio. For
Clinical Correlate propranolol, plasma concentrations achieved after
oral doses of 40–80 mg are equivalent to those
If the liver’s ability to metabolize a drug is
achieved after intravenous doses of 1–2 mg. The
increased, possibly due to enzyme induction, then
difference in required dosage is not explained by
the extraction ratio (E ) is also increased. However,
the magnitude of change in E depends on the low oral absorption but by liver first-pass metab-
initial value of the intrinsic clearance of the drug. olism. Anatomically, the liver receives the blood
supply from the GI tract via the portal vein before
If Cli is small (low intrinsic clearance drug), then E its entrance into the general circulation via the
is initially small. Increasing Cli causes an almost hepatic vein. Therefore, the liver can metabolize
proportional increase in extraction and hepatic
or extract a certain portion of the drug before it
clearance. However, if Cli and E are already high, a
reaches the systemic circulation. Also, enzymes
further increase in intrinsic clearance does not greatly
affect the extraction ratio or hepatic drug clearance. in the gut wall can metabolize the drug before it
reaches the liver.

135
Systemic Circulation
Effects of Disease States and
Metabolism Drug Interactions on Hepatically
Liver
Metabolized Drugs
Gut Lumen
It is important to appreciate the effect that a poten-
tial drug or disease state interaction may have on the
Gut Wall pharmacologic response of a drug that is principally
eliminated by the liver. Therefore, we will consider
the potential impact that changes in Qh, Fp, and Cli
will have on the steady-state concentration of both
total and free drug concentration. Remember, we
will assume that Clt (total body clearance) equals Clh
(hepatic clearance) and that steady-state free drug
concentration is the major determinant of pharma-
Portal Vein cologic response.
When trying to assess clinical implications,
Metabolism
always consider the following:
Feces • route of administration (intravenous versus
oral),
FIGURE 9-7.
Routes of drug disposition with oral drug administration. • extraction ratio (high [> 0.8] versus low
[< 0.2]), and
Because the blood supply draining the GI tract • protein binding (high [> 80%] versus low
passes through the liver first, the fraction of an [< 50%]).
oral dose (F) that reaches the general circulation
(assuming the dose is 100% absorbed across the Qh Fp Cli
gut wall) is given by: Clh =
Qh + Fp Cli
F=1–E
And
Remember, E is the extraction ratio that indicates
K0 K
the efficiency of the organ eliminating a drug. For C ss (total) = or 0
example, if the drug is 100% absorbed across the Clt Clh
gut wall and the liver extracts 70% before it reaches Then, substituting for Css(total)
the systemic circulation, 30% of the dose finally
reaches the bloodstream. Therefore: K0
C ss (free) =
Fp × C ss (total) =
Fp ×
Clh
E = 0.7
F = 1− E where:
= 0.3 Clh = hepatic drug clearance,
Again, F is the fraction of drug reaching the systemic Fp = fraction of drug unbound in plasma, and
circulation. K0 = the drug infusion rate.

136
In the following three examples, we apply the Impact on Css(total)

previously described hepatic extraction equation to Because K0 and Fp are unchanged and Cli is
several cases involving a specific disease state effect reduced by 30%, Css(total) should increase by
on drug interaction. 30%.
Impact on Css(free)
EXAMPLE 1. Because K0 is unchanged and Cli is reduced by
30%, Css(free) should increase by 30%.
Effect of Addition of Enzyme Inhibitor on Consequence
Pharmacologic Response of Theophylline You should anticipate significant side effects
Theophylline (which is metabolized primarily as a consequence of a higher free steady-state
by CYP1A2 of the hepatic cytochrome P450 concentration of theophylline (Figure 9-8).
system) was administered to a patient via a The dosing rate of theophylline should be
constant intravenous infusion and produced reduced by 30% in this example.
a steady-state total plasma concentration of
15 mg/L (therapeutic range, 5–15 mg/L).
Ciprofloxacin, a known inhibitor of the EXAMPLE 2.
hepatic cytochrome P450 enzyme system,
was later added to this patient’s drug dosing Effect of Decreased Protein Binding
regimen. Ciprofloxacin reduces the intrinsic of Phenytoin Due to Renal Failure
clearance of theophylline by 25 to 30%. Phenytoin (which is metabolized primarily
What impact should ciprofloxacin adminis- by CYP2C9/10 of the hepatic cytochrome
tration have on this patient’s pharmacologic P450 mixed function oxidase system) was
response (assume a 30% reduction in clear- administered to a patient by intermittent
ance)? intravenous administration and produced
Considerations a steady-state total plasma concentration of
• Theophylline (in this example) is admin- 15 mg/L (therapeutic range: 10–20 mg/L).
istered via a constant intravenous The patient unexpectedly experienced acute
infusion (K0). renal failure. Renal failure is known to reduce
the plasma protein binding of phenytoin from
• Theophylline has a low extraction ratio.
approximately 90% to about 80% but has
• Theophylline possesses low protein minimal effect on phenytoin’s intrinsic clear-
binding. ance. What impact should renal failure have
Because theophylline has a low extraction on this patient’s pharmacologic response?
ratio and is not extensively bound to proteins,
Clh = Fp × Cli
Css(total)
and
K0 K
C ss (total) = or 0 Css(free)
Clh Fp Cli
Then substituting for Css(total)

K0 K FIGURE 9-8.
C ss (free) =
Fp × C ss (total) Fp × =0 Changes in free and total steady-state plasma theophylline
Fp Cli Cli
concentrations with the addition of cimetidine.

137
Considerations EXAMPLE 3.
• Phenytoin is administered by intermit-
tent intravenous administration. Effects of Increased Protein Binding of
• Phenytoin has a low extraction ratio. Lidocaine Due to Myocardial Infarction
• Phenytoin possesses high protein Lidocaine (which is metabolized primarily
binding. by CYP1A2 of the hepatic cytochrome P450
• Because phenytoin has a low extraction mixed function oxidase system) was admin-
ratio and is extensively bound to proteins, istered to a patient for a life-threatening
Clh = Fp × Cli ventricular arrhythmia via a constant intra-
K0 K venous infusion, producing a steady-state
C ss (total) = or 0
Clh Fp Cli total plasma concentration of 4 mg/L (ther-
apeutic range: 2–6 mg/L). The next day, the
Substituting for Css(total)
patient had a myocardial infarction. Myocar-
K0 K dial infarctions are known to significantly
C ss (free) =
Fp × C ss (total) = Fp × =0
Fp Cli Cli increase the concentration of alpha-1-acid
Impact on Css(total) glycoprotein (a serum globulin) and the
protein binding of drugs associated with it.
Because K0 and Cli are unchanged and Fp is
The protein binding of lidocaine is known to
doubled, Css(total) should decrease by half.
be high and primarily dependent on alpha-
Impact on Css(free)
1-acid glycoprotein. What impact should a
Because K0 and Cli are unchanged, Css(free) myocardial infarction have on this patient’s
should remain unchanged.
pharmacologic response (assuming that the
Consequence myocardial infarction had no effect on hepatic
You should anticipate no significant change in blood flow)?
this patient’s pharmacologic response (despite
a significant drop in phenytoin’s steady-state Considerations
total concentration) because steady-state • Lidocaine is administered via a constant
free drug concentrations remain unchanged intravenous infusion.
(Figure 9-9). However, the total concentration
• Lidocaine has a high extraction ratio.
necessary to achieve this therapeutic unbound
concentration will be less than the normal • Lidocaine possesses high protein binding
reference range for phenytoin. to alpha-1-acid glycoprotein.
Because lidocaine has a high extraction ratio

and binds extensively to alpha-1-acid glyco-
protein, Clh = Qh.
K0 K0
C ss (total) = or
Css(total) Clh Qh
Css(free) Substituting for Css(total)
FIGURE 9-9. K0
Change in total steady-state plasma phenytoin concentration C ss (free) =
Fp × C ss (total) = Fp ×
Qh
due to renal failure.

138
Impact on Css(total) Renal Elimination

Because K0 and Qh are unchanged, Css(total)
As stated previously, drug elimination refers to
should remain unchanged.
metabolism and excretion (Figure 9-11). Some
Impact on Css(free) drugs are primarily excreted unchanged; others
Because K0 and Qh are unchanged and Fp is are extensively metabolized before excretion. The
decreased, Css(free) should decrease, which fraction of drug metabolized is different for various
could result in a reduced pharmacologic agents. The overall elimination rate is the sum of all
response (Figure 9-10). metabolism and excretion processes and is referred
Consequence to as total body elimination:
Because only total (bound and unbound)
lidocaine concentrations can be measured
total body elimination = drug excreted unchanged
clinically, you should anticipate a reduced
+ drug metabolized
pharmacologic response despite similar
Excretion is the process that removes a drug
steady-state total lidocaine concentrations.
from tissues and the circulation. A drug can be
This reduced response may necessitate high
excreted through urine, bile, sweat, expired air,
total lidocaine concentrations and a higher
breast milk, or seminal fluid. The most important
dose to achieve the desired response.
routes of excretion for many drugs and their metab-
olites are the urine and bile. For anesthetic gases,
Clinical Correlate pulmonary excretion can play a significant role.
Excretion may occur for a biotransformed drug
This reduced response is why lidocaine’s dose or for a drug that remains unchanged in the body.
is generally titrated to a clinical response based For example, penicillin G is primarily excreted
on electrocardiogram readings (i.e., decrease in unchanged in the urine. Elimination of this drug is
arrhythmias) rather than dosed to a therapeutic
thus dependent on renal function. Renal excretion
concentration.
is the net effect of three distinct mechanisms within
the kidneys:
These three examples represent how the 1. glomerular filtration,
well-stirred model and knowledge of the pharmaco- 2. tubular secretion, and
kinetic characteristics of a drug can be used to 3. tubular reabsorption.
predict the effect of changes in hepatic blood flow, With glomerular filtration, blood flows into the
protein binding, and intrinsic clearance. These capsule of the glomerulus, and there is a passive
same principles can be used to assess a wide variety diffusion of fluids and solutes across the porous
of clinically relevant situations. glomerular membrane (Figure 9-12). In a healthy
Css(total)
Css(free)
FIGURE 9-10.
Change in free steady-state plasma lidocaine concentrations FIGURE 9-11.
due to myocardial infarction. Drug elimination.

139
Secretion
Most drugs also undergo tubular reabsorption
Distal convoluted
tubule back into the blood. This process occurs passively
Proximal
convoluted in the distal tubules for drugs that are lipid soluble
tubule or not highly ionized. For other agents, it can occur
as an active process and (as with tubular secretion)
Glomerulus
is subject to competition from other agents. An
example of reabsorption is glucose, which normally
undergoes 100% reabsorption in the distal tubules
of the kidneys. With renal dysfunction, glucose
Filtration often is not reabsorbed and may appear in the
Henle’s loop urine. Other examples of agents that are actively
reabsorbed include endogenous substances such as
Reabsorption vitamins, electrolytes, and amino acids.
Tubular reabsorption is dependent on the phys-
ical and chemical properties of the drug and the pH
of the urine. Drugs that are highly ionized in the urine
have less tubular reabsorption; they tend to stay in
FIGURE 9-12. the urine and are excreted. Drugs must be uncharged
Renal nephron. to pass easily through biologic membranes. Tubular
reabsorption of some compounds may also be depen-
adult, up to 130 mL of fluid may cross the glomeruli
dent on urine flow rate. Urea, for example, has a high
per minute (total of both kidneys). Three factors
tubular reabsorption at low urine flow rates and a
influence glomerular filtration:
low tubular reabsorption at high urine flow rates.
1. molecular size,
Because renal clearance is determined by filtra-
2. protein binding, and tion, active secretion, and reabsorption, it is fairly
3. glomerular integrity and total number of complicated. Total renal clearance, Clr , can be deter-
functioning nephrons. mined from the following equation:
Drugs dissolved in the plasma may be filtered Clr = amount excreted in urine(t1→t2 ) /AUC(t→t2 )
across the glomerulus; drugs that are protein bound
or have a molecular weight greater than 60,000 where AUC is the area under the plasma concentra-
are not filtered. Pathophysiologic changes in the tion curve.
kidneys may also alter glomerular filtration. However, because it is not easy to differentiate
Some drugs are actively secreted from the these processes when measuring the amount of
blood into the proximal tubule, which contains drug in the urine, renal clearance is calculated from
urine. These drugs (primarily weak organic acids the ratio of the urine excretion rate to the drug
and some bases) are excreted by carrier-mediated concentration in plasma:
active processes that may be subject to competi- drug excretion rate
tion from other substances in the body due to broad Clr =
drug plasma concentration
specificity of the carriers. For example, probenecid
and penicillin are both actively secreted. If given There are several different methods to calculate
together, probenecid competes with penicillin for renal drug clearance. In one method, the excretion
secretion, so penicillin is secreted less rapidly (it rate of the drug is estimated by determining the
has a longer half-life). This particular relationship drug concentration in a volume of urine collected
can be used in therapeutic situations to extend the over short time periods after drug administration.
duration of penicillin action. This excretion rate is then divided by the plasma

140
concentration of drug entering the kidneys at the tion for the line is Y = mX + b. Then, the line in Figure
midpoint of the urine collection period. To express 9-13 can be defined as:
this as an equation:
clearance = (slope) (GFR) + 0
amount of drug in urine from t 1 to t 2 /(t 2 − t 1 )
Clr = or:
C midpoint
clearance = (slope) (GFR)
where t1 and t2 are the times of starting and stop- However, if a drug is excreted by glomerular
ping the collection, respectively, and C is the plasma filtration as well as some other route (e.g., biliary
concentration at the midpoint of t1 and t2. There- excretion), the relationship illustrated in Figure
fore, overall renal clearance is calculated usually 9-14 could exist. As GFR increases, the clearance of
without differentiating among filtration, secretion, drug increases; but when GFR is zero, clearance is
and reabsorption. This method is commonly used to still greater than zero. In this example, the equation
calculate creatinine clearance when the “amount of for the line is:
drug” is the amount of creatinine that appears in the
clearance = slope (GFR) + y-intercept
urine over 24 hours, t2 – t1 = 24 hours, and Cmidpoint is
the serum creatinine determined at the midpoint of Y = mX + b
the urine collection period.
and we see that when GFR is zero, clearance is the
value of the y-intercept, which is nonrenal clear-
Relationship Between Renal
ance.
Clearance and Glomerular Filtration
This approach has been used to relate the
Rate aminoglycoside elimination rate constant (K) to
If a drug is exclusively eliminated renally and the creatinine clearance. When dosing these agents,
only renal process involved is glomerular filtra- we must consider the individual’s GFR, as reflected
tion, the relationship between total body clearance by creatinine clearance. The relationship observed
and glomerular filtration rate (GFR) is as shown between K and creatinine clearance is shown in
in Figure 9-13. Creatinine clearance is commonly Figure 9-15.1 Therefore, K can be predicted for
used as a measure of GFR. Remember that creati- aminoglycosides (such as gentamicin) based on an
nine undergoes some tubular secretion; therefore, individual’s creatinine clearance.
GFR can sometimes be slightly overestimated. As With the equation for a line, Y = mX + b:
GFR increases, clearance of drug increases. When K = 0.00293 hr –1 × creatinine clearance
GFR is zero, clearance is zero. Recall that the equa- (in mL/minute) + 0.014

Relationship between drug clearance and glomerular filtration Relationship between drug clearance and glomerular filtration
rate for a drug that is exclusively eliminated by glomerular rate for a drug that is eliminated by renal and nonrenal
filtration. processes.

141
or
(140 − age)IBW
CrClfemale = 0.85
72 × SCr
where:
CrCl = creatinine clearance (milliliters per
minute),
FIGURE 9-15. age = patient’s age (years),
Relationship between elimination rate constant and creatinine IBW = ideal body weight (kilograms), and
clearance for aminoglycosides.
SCr = serum creatinine concentration
(milligrams per deciliter).
Clinical Correlate Adjusting this equation for a patient’s body
surface area is not necessary clinically.
Note that drugs that are cleared almost solely by This formula also requires the following patient
renal mechanisms will have a y-intercept of zero data:
or very close to zero. Drugs that have extrarenal
routes of elimination will have larger y-intercepts. • ideal body weight (lean body weight) or
adjusted body weight (AdjBW),
• age,
Determining patient-specific creatinine clearance • sex, and
can be accomplished by either direct measurement
• steady-state serum creatinine concentration.
of the amount of creatinine contained in a 24-hour
urine sample or by estimating this parameter using IBW may be estimated as follows:
standard mathematical equations. Direct measure-
ment is the most accurate of these. When using this IBWmales = 50 kg + 2.3 kg for each inch
9-2
method, creatinine clearance (CrCl) is determined as over 5 feet in height
follows:
IBWfemales = 45.5 kg + 2.3 kg for each inch
UV over 5 feet in height
CrCl =
P × 1440
In obese patients, the use of total body weight
where:
(TBW) overestimates whereas the use of IBW
U = urinary creatinine concentration, underestimates creatinine clearance. In patients
V = volume of urine collected, whose TBW is more than 35% over their IBW,
P = plasma creatinine concentration (taken adjusted body weight (AdjBW) should be used to
at midpoint of urine collection), and estimate creatinine clearance:
1440 = number of minutes in 24 hours.
9-3 AdjBW = IBW + 0.4(TBW−IBW)
Although there are several formulas for esti-
For a patient who weighs less than IBW, the actual
mating creatinine clearance, the Cockcroft–Gault
body weight would be used.
equation is commonly used:
It is important to note that the use of serum creat-
(140 − age)IBW inine values less than 1 mg/dL will greatly elevate
9-1 CrClmale = the calculated creatinine clearance value when using
72 × SCr
Equation 9-1. This is especially true in the elderly. In

142
patients with serum creatinine values of less than Reference

1 mg/dL, it has been recommended to either round
the low serum creatinine value up to 1 mg/dL before 1. Matzke GR, Jameson JJ, Halstenson CE. Genta-
calculating creatinine clearance, or round the final micin distribution in young and elderly patients
calculated creatinine clearance value down. with various degrees of renal function. J Clin
Creatinine clearance, estimated creatinine clear- Pharmacol 1987;27:216–20.
ance, and other GFR estimations, such as the
modification of diet in renal disease (MDRD) equa-
tions, are more fully discussed in Lesson 12.

143
REVIEW QUESTIONS
9-1. The body converts a drug to a less active 9-7. The basic functional unit of the liver is the:
substance by a process called: A.
renal lobule.
A.
phosphorylation. B.
hepatocyte.
B.
hydrogenation. C.
liver cell.
C.
biotransformation. D.
liver lobule.
D.
distransformation.
9-8. The liver receives its blood from the:
9-2. Biotransformation is also known as: A.
portal artery and hepatic vein.
A.
hepatic clearance. B.
portal vein and hepatic artery.
B.
elimination. C.
vena cava and aorta.
C.
renal excretion. D.
portal artery and hepatic artery.
D.
metabolism.
9-9. A drug administered orally must go through
9-3. In total, hepatic elimination encompasses the liver before it is available to the systemic
both the processes of: circulation.
A.
biotransformation and excretion. A.
True
B.
oxidation and glucuronidation. B.
False
C.
hydroxylation and oxidation.
9-10. Because the extraction ratio can maximally
D.
absorption and demethylation.
be 1, the maximum value that hepatic clear-
ance can approach is that of:
9-4. Glucuronidation is a Phase II biotransfor-
mation process. A.
creatinine clearance.
A.
True B. glomerular filtration.
B.
False C. renal blood filtration.
D. hepatic blood flow.
9-5. Biotransformation may be dependent on
factors such as age, 9-11. Intrinsic clearance is the maximal ability of
A.
height, and gender. the liver to eliminate drug in the absence of
any blood flow limitations.
B.
gender, and weight.
A.
True
C.
disease, and genetics.
B.
False
D.
disease, and gender.
9-12. Smoking is known to increase the enzymes
9-6. Which of the following is not a Phase I
responsible for theophylline metabolism (a
reaction?
drug with a low hepatic extraction). Would a
A.
oxidation patient with a history of smoking likely require
B.
glucuronidation a higher, lower, or equivalent theophylline total
C.
reduction daily dose compared to a nonsmoking patient?
D.
hydrolysis A.
lower
B.
higher
C.
equivalent

144
9-13. Heart failure reduces cardiac output and 9-18. Route of administration, extraction ratio,
hepatic blood flow. Consequently, the total and protein binding are all factors that
daily dose of lidocaine may need to be should be considered when trying to assess
decreased in a patient with heart failure the effect of disease states on plasma
who has a myocardial infarction. concentrations of drugs eliminated by the
A.
True liver.
B.
False A.
True
B.
False
9-14. Which of the following types of metabo-
lism do drugs with a high extraction ratio 9-19. Will drugs that inhibit the hepatic cyto-
undergo to a significant extent? chrome P450 system likely increase or
A. first-pass decrease the plasma clearance of theo-
phylline?
B.
zero-order
A.
increase
C.
intraluminal
B.
decrease
D.
nonlinear
9-20. Disease states may increase or decrease
9-15. Significant first-pass metabolism means
drug protein binding.
that much of the drug’s metabolism occurs
before its arrival at the: A.
True
A.
hepatocyte. B.
False
B.
systemic circulation.
9-21. Drug elimination encompasses both:
C.
portal blood.
A.
metabolism and excretion.
D.
liver lobule.
B.
metabolism and biotransformation.
9-16. The liver receives blood supply from the GI C.
absorption and metabolism.
tract via the: D.
metabolism and distribution.
A.
portal vein.
9-22. Two important routes of drug excretion are:
B.
hepatic artery.
A.
hepatic and tubular secretion.
C.
hepatic vein.
B.
biliary and metabolic.
D.
portal artery.
C.
renal and biliary.
9-17. For a drug that is totally absorbed without D.
renal and metabolic.
any presystemic metabolism and then
undergoes hepatic extraction, which of the 9-23. Fluid is filtered across the glomerulus
following is the correct equation for F? through active transport.
F = 1 – Ka
A. A.
True
F = 1 – Fp
B. B.
False
F=1–E
C.
9-24. Tubular secretion most often occurs with
F = 1 – the fraction of the drug absorbed
D. weak organic acids.
A.
True
B.
False

145
9-25. Which of the following statements about

tubular reabsorption is false?
ANSWERS
A.
Tubular reabsorption depends on the
pH of the urine. 9-1. A, B, D. Incorrect answers
B. Highly ionized drugs tend to remain in C.
CORRECT ANSWER
the urine.
C.
Tubular reabsorption can only be an
D.
CORRECT ANSWER
active transport process.
D.
A and C. 9-3. A. CORRECT ANSWER
9-26. Renal clearance can be calculated from the
ratio of which of the following rates to the
drug’s concentration in plasma?
Incorrect answer. Oxidation, reduction,
B.
A.
tubular reabsorption rate
and hydrolysis are examples of Phase I
B.
tubular secretion rate reactions.
C. glomerular filtration rate
D.
excretion rate 9-5. A, B, D. Incorrect answers
C.
CORRECT ANSWER
9-27. For aminoglycoside doses, which of the
following must be calculated to estimate an 9-6. A, C, D. Incorrect answers
individual patient’s drug elimination rate? B.
CORRECT ANSWER
An individual patient’s:
A.
pulmonary clearance. 9-7. A, B, C. Incorrect answers
B.
biliary clearance. D.
CORRECT ANSWER
C. creatinine clearance. 9-8. A, C, D. Incorrect answers
D.
A and C. B.
CORRECT ANSWER
9-28. For aminoglycosides, the terminal elimina- 9-9. A. CORRECT ANSWER
tion rate constant can be estimated from
Incorrect answer
B.
the creatinine clearance using which of the
following equations?
K = 0.00293 hr–1 × (creatinine clearance
A.
D.
CORRECT ANSWER
in mL/minute) + 1.4
K = 0.00293 hr–1 × (creatinine clearance
B. 9-11. A. CORRECT ANSWER
in mL/minute) + 0.014 Incorrect answer
B.
K = 2.93 hr–1 × (creatinine clearance in
C.
mL/minute) 9-12. A, C. Incorrect answers
K = 0.00293 hr–1 + (creatinine clearance
D. B.
CORRECT ANSWER. Smoking raises
in mL/minute) the concentrations of enzymes that
also metabolize theophylline, so more
theophylline would be metabolized,
requiring a higher theophylline dose.

146
9-13. A. CORRECT ANSWER 9-22. A, B, D. Incorrect answers

Incorrect answer
B. C.
CORRECT ANSWER
9-14. A. CORRECT ANSWER 9-23. A. Incorrect answer

Incorrect answer. Zero-order processes
B. B.
CORRECT ANSWER
are not determined by amount of hepatic
extraction. 9-24. A. CORRECT ANSWER
Incorrect answer. Intraluminal metabo-
C. Incorrect answer
B.
lism is independent of hepatic
extraction. 9-25. A. Incorrect answer. Urine pH does affect
tubular reabsorption.
D. Incorrect answer. Nonlinear metabo-
lism involves only saturation of hepatic Incorrect answer. Highly ionized drugs
B.
enzymes. do remain in the urine because ionized
forms of drugs do not cross membranes
9-15. A, C, D. Incorrect answers well.
B.
CORRECT ANSWER C.
CORRECT ANSWER
Incorrect answer
D.
B, C, D. Incorrect answers 9-26. A. Incorrect answer. Tubular reabsorption
rate cannot be directly measured.
9-17. A, B, D. Incorrect answers Incorrect answer. Tubular secretion rate
B.
CORRECT ANSWER. F represents the
C. cannot be directly measured.
fraction of drug that reaches the systemic Incorrect answer. Glomerular filtration
C.
circulation; E is the extraction ratio. rate does not account for tubular secre-
tion or reabsorption.
D.
CORRECT ANSWER
Incorrect answer
B.
9-27. A, B, D. Incorrect answers. Aminoglycosides
9-19. A. Incorrect answer do not undergo hepatic or pulmonary
B.
CORRECT ANSWER. Theophylline is a clearance.
low-extraction drug and its clearance is C.
CORRECT ANSWER
roughly equal to intrinsic hepatic clear-
ance (Cli), so the effect of cytochrome 9-28. A. Incorrect answer. The y-intercept is
P450 enzyme induction is likely to wrong. Aminoglycosides undergo little
decrease intrinsic and overall clearance. if any extrarenal elimination and, there-
fore, the y-intercept value should be
9-20. A. CORRECT ANSWER close to zero.
Incorrect answer
B. B.
CORRECT ANSWER
9-21. A. CORRECT ANSWER Incorrect answer. The answer should
C.
represent the approximate fraction of
Incorrect answer. Biotransformation is a
B.
drug excreted per hour, and this value
type of metabolism.
should be less than one.
Incorrect answer. Absorption is not an
C.
Incorrect answer. The correct answer
D.
elimination process.
should be expressed as a product, not a
Incorrect answer. Distribution is not an
D. sum (i.e., A × B, not A + B).
elimination process.

147
Discussion Points
D-1. Research the metabolism of primidone D-4. Research the various oral fluoroquinolones
and discuss the clinical significance of its to determine which can affect the metabo-
metabolites. Discuss the proper method to lism of theophylline and to what extent.
monitor a patient receiving primidone. Discuss why some of these drugs affect
theophylline and others do not.
D-2. Select several drugs whose prescribing
information indicates that the dose should D-5. Describe several clinical situations in which
be decreased with hepatic impairment. a drug’s ability to compete for renal secre-
Describe the pharmacokinetics of these tion with another drug can be either useful
drugs and discuss why this drug’s dose or harmful.
should be decreased. Finally, indicate specif-
ically how you would go about decreasing D-6. Describe situations in which alteration of
this dose. urine pH with urine acidifier or alkalinizing
agents can be used to enhance the clinical
D-3. Research the pharmacokinetics of carbam- response of other drugs.
azepine and discuss its metabolism when
given alone and when given with other D-7. Look up and compare the various equations
enzyme inhibitors or inducers. Specifically, that can be used to calculate the elimination
how would you begin a patient on carbam- rate constant for gentamicin, tobramycin,
azepine and how would you monitor and and amikacin. Are these equations the same
adjust its dose? or different? Try to explain why they are
either the same or different.

LESSON 10
Nonlinear Processes
OBJECTIVES
1. Describe the relationship of both drug concentration and area under the plasma drug
concentration versus time curve (AUC) to the dose for a nonlinear, zero-order process.
2. Explain the various biopharmaceutic processes that can result in nonlinear
pharmacokinetics.
3. Describe how hepatic enzyme saturation can result in nonlinear pharmacokinetics.
4. Use the Michaelis–Menten model for describing nonlinear pharmacokinetics.
5. Describe Vmax and Km.
6. Use the Michaelis–Menten model to predict plasma drug concentrations.
7. Use the t 90% equation to estimate the time required for 90% of the steady-state
concentration to be reached.
Until now, we have used a major assumption in constructing models for drug pharmaco-
kinetics: drug clearance remains constant with any size dose. This is the case
only when drug elimination processes are first order (as described in previous
lessons). With a first-order elimination process, as the dose of drug increases, the
plasma concentrations observed and the AUC increase proportionally. That is, if
the dose is doubled, the plasma concentration and AUC also double (Figure 10-1).
Because the increase in plasma concentration and AUC is linear with drug
dose in first-order processes, this concept is referred to as linear pharmaco-
kinetics. When these linear relationships are present, they are used to predict drug
dosage. For example, if a 100-mg daily dose of a drug produces a steady-state peak
plasma concentration of 10 mg/L, we know that a 200-mg daily dose will result in
a steady-state plasma concentration of 20 mg/L. (Note that linear does not refer to
the plot of natural log of plasma concentration versus time.)
With some drugs (e.g., phenytoin and aspirin), however, the relationships
of drug dose to plasma concentrations and AUC are not linear. As the drug dose
increases, the peak concentration and the resulting AUC do not increase propor-
tionally (Figure 10-2). Therefore, such drugs are said to follow nonlinear,
zero-order, or dose-dependent pharmacokinetics (i.e., the pharmacokinetics
change with the dose given). Just as with drugs following linear pharmacokinetics,
it is important to predict the plasma drug concentrations of drugs following zero-
order pharmacokinetics. In this lesson, we discuss methods to characterize drugs
that follow nonlinear pharmacokinetics.
149
150

Relationship of AUC to drug dose with first-order elimination, Relationship of AUC to drug dose with dose-dependent
where clearance is not influenced by dose. pharmacokinetics.
Nonlinear pharmacokinetics may refer to several Of course, most elimination processes are
different processes, including absorption, distribu- capable of being saturated if enough drug is
tion, and renal or hepatic elimination (Table 10-1). administered. However, for most drugs, the doses
For example, with nonlinear absorption, the frac- administered do not cause the elimination processes
tion of drug in the gastrointestinal (GI) tract that is to approach their limitations.
absorbed per minute changes with the amount of drug
present. Even though absorption and distribution can
be nonlinear, the term nonlinear pharmacokinetics Clinical Correlate
usually refers to the processes of drug elimination.
Many drugs exhibit mixed-order pharmaco-
When a drug exhibits nonlinear pharmacokinetics, displaying first-order pharmacokinetics
kinetics, usually the processes responsible for drug at low drug concentrations and zero-order
elimination are saturable at therapeutic concen- pharmacokinetics at high concentrations. It is
trations. These elimination processes may include important to know the drug concentration at
renal tubular secretion (as seen with penicillins) and which a drug order switches from first to zero.
hepatic enzyme metabolism (as seen with phenytoin). Phenytoin is an example of a drug that switches
When an elimination process is saturated, any order at therapeutic concentrations, whereas
increase in drug dose results in a disproportionate theophylline does not switch until concentrations
increase in the plasma concentrations achieved reach the toxic range.
because the amount of drug that can be eliminated
over time cannot increase. This situation is contrary For a typical drug having dose-dependent
to first-order linear processes, in which an increase pharmacokinetics, with saturable elimination, the
in drug dosage results in an increase in the amount plasma drug concentration versus time plot after a
of drug eliminated over any given period. dose may appear as shown in Figure 10-3.
TABLE 10-1. Drugs Having Dose- or Time-Dependent Pharmacokinetics

Process Agent Mechanism
Absorption Riboflavin, methotrexate, gabapentin Saturable gut wall transport
Penicillins Saturable decomposition in GI tract
Distribution Methotrexate Saturable transport into and out of tissues
Salicylates Saturable protein binding
Renal elimination Penicillin G Active tubular secretion
Ascorbic acid Active reabsorption
Extrarenal elimination Carbamazepine Enzyme induction
Theophylline, phenytoin Saturable metabolism

Lesson 10 | Nonlinear Processes
151
Vmax is expressed in units of amount per unit of

time (e.g., milligrams per day) and represents the
maximum amount of drug that can be eliminated
in the given time period. For drugs metabolized by
the liver, Vmax can be determined by the quantity or
efficiency of metabolizing enzymes. This param-
eter will vary, depending on the drug and individual
patient.
FIGURE 10-3. Km, the Michaelis constant, is expressed in
Dose-dependent clearance of enzyme-saturable drugs. units of concentration (e.g., mg/L) and is the drug
concentration at which the rate of elimination is
After a large dose is administered, an initial slow half the maximum rate (Vmax). In simplified terms,
elimination phase (clearance decreases with higher Km is the concentration above which saturation of
plasma concentration) is followed by a much more drug metabolism is likely.
rapid elimination at lower concentrations (curve A).
Vmax and Km are related to the plasma drug
However, when a small dose is administered (curve B),
concentration and the rate of drug elimination
the capacity of the elimination process is not reached,
as shown in Figure 10-4. When the plasma drug
and the elimination rate remains constant. At high
concentration is less than Km , the rate of drug
concentrations, the elimination rate approaches that
elimination follows first-order pharmacokinetics.
of a zero-order process (i.e., the amount of drug elim-
In other words, the amount of drug eliminated
inated over a given period remains constant, but the
per hour directly increases with the plasma drug
fraction eliminated changes). At low concentrations,
concentration. When the plasma drug concentra-
the elimination rate approaches that of a first-order
tion is much less than Km , the first-order elimination
process (i.e., the amount of drug eliminated over a
rate constant (K) for drugs with nonlinear pharmaco-
given time changes, but the fraction of drug elimi-
kinetics is approximated by Vmax ; therefore, as Vmax
nated remains constant).
increases (e.g., by hepatic enzyme induction), K
A model that has been used extensively in increases.
biochemistry to describe the kinetics of saturable
With drugs having saturable elimination, as
enzyme systems is known as Michaelis–Menten
plasma drug concentrations increase, drug elimi-
kinetics (for its developers). This system describes
nation approaches its maximum rate. When the
the relationship of an enzyme to the substrate (in
plasma concentration is much greater than Km , the
this case, the drug molecule). In clinical pharmaco-
rate of drug elimination is approximated by Vmax ,
kinetics, it allows prediction of plasma drug concen-
and elimination proceeds at close to a zero-order
trations resulting from administration of drugs with
process.
saturable elimination (e.g., phenytoin).
The equation used to describe Michaelis–
Menten pharmacokinetics is:
−dC VmaxC
drug elimination =
rate =
dt Km +C
where –dC/dt is the rate of drug concentration

decline at time t and is determined by Vmax , the theo-
retical maximum rate of the elimination process. Km
is the drug concentration when the rate of elimi-
nation is half the maximum rate, and C is the total FIGURE 10-4.
Relationship of drug elimination rate to plasma drug
plasma drug concentration.
concentration with saturable elimination.

152
Next, we consider how Vmax and Km can be calcu- Then:

lated and how these determinations may be used to
predict plasma drug concentrations in patients. daily dose = –Km (daily dose/C ) + Vmax
Y (slope) = mX + b (intercept)
Calculation of Vmax, Km, and Plasma
Concentration and Dose So the relationship of the Michaelis–Menten param-
eters, C, and dose can be expressed as a straight line
For drugs that have saturable elimination at the (Figure 10-5). If the straight line can be defined,
plasma concentrations readily achieved with thera- then Vmax and Km can be determined; if Vmax and
peutic doses (e.g., phenytoin), prediction of the plasma Km are known, then the plasma concentrations at
concentrations achieved by a given dose is important. steady state resulting from any given dose can be
For these predictions, it is necessary to estimate Vmax estimated.
and Km . Therefore, we must apply the Michaelis–
To define the line, it is necessary to know the
Menten equation presented earlier in this lesson:
steady-state concentrations achieved at a minimum
−dC V C of two different doses. For example, a patient
= max
dt Km +C receiving 300 mg of phenytoin per day achieved
a steady-state concentration (trough) of 9 mg/L;
The change in drug concentration over time is when the daily dose was increased to 400 mg/day, a
related to the Michaelis–Menten parameters Vmax, steady-state concentration of 16 mg/L was achieved.
Km, and the plasma drug concentration (C ). We The data for this patient can be plotted as shown
know that at steady state (after multiple drug doses) in Figure 10-6. Then a line is drawn between the
the rate of drug loss from the body (milligrams two points, intersecting the y-axis. The y-intercept
removed per day) is equal to the amount of drug equals Vmax (observed to be 700 mg/day), and the
being administered (daily dose). In the Michaelis– slope of the line equals –Km.
Menten equation, –dC/dt indicates the rate of drug
loss from the body; therefore, at steady state:
Calculating Km
−dC VmaxC
= daily
= drug dose
dt Km +C doseinitial − doseincreased
10-2 slope =
−K m =
Now we have an equation that relates Vmax, Km, dose/C initial − dose/C increased
plasma drug concentration, and daily dose (at steady
state). To use this relationship, it is first helpful to 300 mg/day − 400 mg/day
=
transform the equation to a straight-line form:  300 mg/day 400 mg/day 
 9 mg/L − 16 mg/L 
 
VmaxC
10-1 daily dose =
Km +C −100 mg/day
=
33.3 L/day − 25 L/day
daily dose (K m + C ) =
VmaxC
= −12.0 mg/L
daily dose (K m ) + daily dose (C ) =
VmaxC
daily dose=
(C ) VmaxC − daily dose (K m ) So Km equals 12 mg/L.

153
FIGURE 10-5.
Linear plot of the Michaelis–Menten equation.
Calculating Dose
FIGURE 10-6.
Knowing Vmax and Km, we can then predict the dose Plot of patient data using two steady-state plasma phenytoin
necessary to achieve a given steady-state concen- concentrations at two dose levels.
tration or the concentration resulting from a given
dose. If we wish to increase the steady-state plasma See Lesson 15 for examples of how these calcu-
concentration to 20 mg/L, we can use the Michaelis– lations are applied.
Menten equation to predict the necessary dose:
VmaxC Clinical Correlate

dose =
Km +C
When performing this calculation using sodium
(700 mg/day)(20 mg/L) phenytoin or fosphenytoin, be sure to convert
= doses to their phenytoin free-acid equivalent
12 mg/L + 20 mg/L
before substituting these values into the equation.
14,000 mg2 /(day × L) To convert, multiply the daily dose by 0.92 (92%
= free phenytoin). Fosphenytoin injection, although
32 mg/L
containing only 66% phenytoin free acid, is
= 437 mg/day actually labeled in phenytoin sodium equivalents
such that the 0.92 factor also applies to this
Note how units cancel out to yield mg/day. product.
Calculating Steady-State
Concentration from This Km and Dose
The preceding example demonstrates how
If we wish to predict the steady-state plasma concen- plasma drug concentrations and drug dose can be
tration that would result if the dose is increased predicted. However, it also shows that for drugs
to 500 mg/day, we can rearrange the Michaelis– like phenytoin, with saturable elimination, when
Menten equation and solve for C: plasma concentrations are above Km, small dose
increases can result in large increases in the steady-
K m (daily dose)
10-3 C= state plasma concentration.
Vmax − daily dose
When clearance changes with plasma concen-
12 mg/L (500 mg/day) tration, there is no true half-life as with first-order
= elimination. As clearance changes, the elimination
700 mg/day − 500 mg/day
rate changes as does the time to reach steady state.
12 mg/L (500 mg/day) With high doses and high plasma concentrations
= (and resulting lower clearance), the time to reach
200 mg/day
steady state is much longer than with low doses and
= 30 mg/L low plasma concentrations (Figure 10-7). Theoret-

154
When the dose is increased to 400 mg/day:
12 mg/L (50 L)
=t 90% [2.3(700 mg/day) −
(700 mg/day − 300 mg/day)2
0.9(400 mg/day)]
600 mg
= [1610 mg/day − 360 mg/day]
(300 mg/day)2
= (0.0067 day 2 /mg)(1250 mg/day)
= 8.38 days
We can see that as the dose is increased, it takes

FIGURE 10-7. a longer time to reach steady state, drug continues
Time to reach t 90% (represented by arrows) at different daily to accumulate, and the plasma drug concentration
dosages. continues to rise. When this occurs with a drug
ically, if the dose is greater than Vmax , steady state such as phenytoin, toxic effects (e.g., ataxia and
will never be reached. nystagmus) probably will be observed if the high
dosage is given on a regular basis.
Because clearance and half-life are concen-
tration-dependent factors, a traditional time to
steady-state value cannot be calculated. Instead, the
Michaelis–Menten equation can be rearranged to
Clinical Correlate
provide an equation that estimates the time required The t90% equation will provide only a rough
(in days) for 90% of the steady-state concentration estimate of when 90% of steady state has been
to be reached (t90%), as shown below for phenytoin reached, and its accuracy is dependent on the
(where the dose equals the daily dose): Km value used. Other ways to check to see if a
K m (V ) patient is at steady state are to examine two
10-4 t 90% [2.3Vmax − 0.9 dose] levels drawn approximately a week apart. If
(Vmax − daily dose)2
these levels are ±10% of each other, then you
From the previous example, when dose = 300 mg/day, can assume steady state. Additionally, it is
Vmax = 700 mg/day, and Km = 12 mg/L, volume of safe to wait at least 2 weeks (and preferably 4
distribution (V) can be estimated as 0.65 L/kg body weeks) after beginning or changing a dose before
weight, or (0.65 × 77 kg body weight) = 50 L. obtaining new steady-state levels.
12 mg/L (50 L)
t 90% [2.3(700 mg/day) −
(700 mg/day − 300 mg/day)2
0.9(300 mg/day)]
600 mg
= [1610 mg/day − 270 mg/day]
(400 mg/day)2
= (0.00375 day 2 /mg)(1340 mg/day)

= 5.0 days

155
REVIEW QUESTIONS
10-1. Which drug pairs demonstrate nonlinear The following information is for Questions 10-7
pharmacokinetics? to 10-11. A patient, JH, is administered phenytoin
A. phenytoin and aspirin free acid, 300 mg/day for 2 months (assume steady
B. penicillin G and gentamicin state is achieved), and a plasma concentration
C. acetaminophen and sulfonamides determined just before a dose is 10 mg/L. The
D. A and B
phenytoin dose is then changed to 400 mg/day;
10-2. Linear pharmacokinetics means that the 2 months after the dose change, the plasma
plot of plasma drug concentration versus concentration determined just before a dose is
time after a dose is a straight line. 18 mg/L. Assume that the volume of distribution of
A. True phenytoin is 45 L.
B. False
10-7. Calculate Km for this patient.
10-3. When hepatic metabolism becomes satu-
rated, any increase in drug dose will lead A. 12.5 mg/L
to a proportionate increase in the plasma B. 25 mg/L
concentration achieved. C. 37.5 mg/L
A. True D. 10 mg/L
B. False
10-8. For the same patient, JH, determine Vmax.
10-4. When the rate of drug elimination proceeds
A. 123 mg/day
at half the maximum rate, the drug concen-
tration is known as: B. 900 mg/day
A. Vmax C. 500 mg/day
B. Km D. 678 mg/day
C. ½Vmax
D. (Vmax)(C) 10-9. For the case of JH above, plot both concen-
trations on a daily dose/C versus Vmax plot
10-5. At very high concentrations—concentrations and then determine this patient’s Vmax.
much higher than the drug’s Km—drugs are A. approximately 550 mg/day
more likely to exhibit first-order elimination. B. approximately 400 mg/day
A. True C. approximately 675 mg/day
B. False
D. approximately 800 mg/day
10-6. Which of the equations below describes the
form of the Michaelis–Menten equation that 10-10. After the dose of 400 mg/day is begun, how
relates daily drug dose to Vmax, Km, and the long will it take to reach 90% of the steady-
steady-state plasma drug concentration? state plasma concentration?
A. daily dose = –Km(daily dose/C)(Vmax) A. approximately 14 days
B. daily dose = –Km(daily dose/C) + Vmax B. approximately 9 days
C. daily dose = –Km(daily dose × C) + Vmax C. approximately 30 days
D. daily dose = –Km – (daily dose/C) + Vmax D. approximately 90 days

156
10-11. If the patient, JH, misunderstood the dosage 10-5. A. Incorrect answer
instructions and consumed 500 mg/day B. CORRECT ANSWER. At very low concen-
of phenytoin, what steady-state plasma trations, drugs are more likely to exhibit
concentration would result? first-order kinetics because hepatic
A. 29.4 mg/L enzymes are usually not yet saturated,
B. 36.8 mg/L whereas at higher concentrations,
enzymes saturate, making zero-order
C. 27.2 mg/L
kinetics more likely.
D. 19.6 mg/L
B. CORRECT ANSWER
ANSWERS
10-7. A. CORRECT ANSWER. The Km is calculated
10-1. A. CORRECT ANSWER from the slope of the line above:
10-2. A. Incorrect answer dose1 − dose2

slope =
−K m =
B. CORRECT ANSWER. Linear pharmaco- dose1 /C 1 − dose2 /C 2
kinetics means that the AUC and plasma
concentrations achieved are directly
related to the size of the dose adminis- =
tered. Drugs with linear pharmacokinetics  300 mg/day 400 mg/day 
 10 mg/L − 18 mg/L 
may exhibit plasma concentrations  
versus time plots that are not straight
lines, as with multicompartment drugs. −100 mg/day
=
30 L/day − 22 L/day
B. CORRECT ANSWER. There will be a −100 mg/day
=
disproportionate increase in the plasma 8 L/day
concentration achieved because the
amount of drug that can be eliminated
= −12.5 mg/L
over time cannot increase. So Km equals 12.5 mg/L.
10-4. A. Incorrect answer. Vmax is the maximum
B, C, D. Incorrect answers. Use dose pairs of
rate of hepatic metabolism.
300 and 400 and concentration pairs of
B. CORRECT ANSWER 10 and 18 to calculate Km.
C. Incorrect answer. ½Vmax is only one-half
of the maximum hepatic metabolism 10-8. A, C. Incorrect answers. Try again; you prob-
and does not relate Km to Vmax. ably made a math error.
D. Incorrect answer. (Vmax)(C) is only the B. Incorrect answer. Try again, and use
numerator of the Michaelis–Menten either set of dose and concentration
equation. pairs (i.e., 300 and 10 or 400 and 18).

157
D. CORRECT ANSWER. 10-11. A. Incorrect answer. Perhaps you used a

400-mg dose instead of a 500-mg dose.
daily dose = –Km(daily dose/C ) + Vmax
B. CORRECT ANSWER. The steady-state
400 = (–12.5)(400/18) + Vmax plasma concentration resulting from
a daily dose of 500 mg would be esti-
400 = (–12.5)(22.22) + Vmax mated from the line equation as follows:
400 = –277.77 + Vmax daily dose =

−K m (dose/C ) + Vmax
677.77 = Vmax  500 mg/day 

500 mg/day =
−12.5 mg/L   + 670 mg/day
 C 
C. CORRECT ANSWER. See Figure 10-8 Rearranging gives:
for a plot of the daily dose versus daily
dose/C. −170 mg/day 500 mg/day
=
−12.5 mg/L C
500 mg/day
B. CORRECT ANSWER. The time to reach 13.6 L/day =
steady state is calculated by: C
13.6 L/day 1
K m (V ) =
t 90% [2.3Vmax − 0.9 dose] 500 mg/day C
(Vmax − dose)2
1
12.5 mg/L (45 L) 0.0272 L/mg =
= [2.3(670 mg/day) − C
(670 mg/day − 400 mg/day)2
0.9(400 mg/day)] C = 36.8 mg/L
562.5 mg
= [1541 mg/day − 360 mg/day] C, D. Incorrect answers. You may have made
(270 mg/day)2 a simple math error.
= (0.00772 day 2 /mg)(1181 mg/day)
= 9.11 days
FIGURE 10-8.
Daily dose versus daily dose divided by steady-state
concentration.

158
Discussion Points
D-1. When using the Michaelis–Menten equa- D-6. Discuss the patient variables that can
tion, examine what happens when daily affect the pharmacokinetic calculation of a
dose is much lower than Vmax, and when it nonlinear drug when using two plasma drug
exceeds Vmax. concentrations obtained from two different
doses.
D-2. When using the t90% equation, examine what
happens to t90% when dose greatly exceeds D-7. Examine the package insert for Cerebyx®
Vmax. (fosphenytoin) and answer the following
questions:
D-3. Using two steady-state plasma drug concen-
trations and two doses to solve for a new Km, A. What salt is this product?
Vmax, and dose using the Michaelis–Menten
equation, examine the values of Km and B. What percent phenytoin sodium is it?
Vmax obtained using this process. Are these C. What percent phenytoin free acid is it?
values close to the actual patient population
parameters? D. How many milligrams of Cerebyx® is
equal to 100 mg of sodium phenytoin
D-4. Discuss several practical methods to deter- injection?
mine when a nonlinear drug has reached
steady state. E. What therapeutic advantage does this
product offer?
D-5. Examine the time to 90% equation and note
the value of Km that is used in this equation.
Substitute several different phenytoin Km
values based on a range of population values
(i.e., from approximately 1 to 15 mg/L) and
describe the effect this has on your answer.
Based on this observation, what value of Km
would you use when trying to approximate
the t90% for a newly begun dose of phenytoin?

LESSON 11
Pharmacokinetic Variation
and Model-Independent
Relationships
OBJECTIVES
1. Identify the various sources of pharmacokinetic variation.
2. Explain how the various sources of pharmacokinetic variation affect pharmacokinetic
parameters.
3. Describe how to apply pharmacokinetic variation in a clinical setting.
4. Name the potential sources of error in the collection and assay of drug samples.
5. Explain the clinical importance of correct sample collection, storage, and assay.
6. Describe ways to avoid or minimize errors in the collection and assay of drug samples.
7. Explain the basic concepts and calculations of the model-independent
pharmacokinetic parameters of total body clearance, mean residence time (MRT),
volume of distribution at steady state, and formation clearance.
Sources of Pharmacokinetic Variation

An important reason for pharmacokinetic drug monitoring is that a drug’s effect
may vary considerably among individuals given the same dose. These differences
in drug effect are sometimes related to differences in pharmacokinetics. Some
factors that may affect drug pharmacokinetics are discussed below. However, irre-
spective of pharmacokinetics, drug effects may vary among individuals because of
differences in drug sensitivity.
Age
At extremes of age, major organ functions may be considerably reduced compared
with those of healthy young adults. In neonates (particularly if premature) and
the elderly, renal function and the capacity for renal drug excretion may be greatly
reduced. Neonates and the elderly are also more likely to have reduced hepatic
function. Renal function declines at a rate of approximately 1 mL/minute/year
after the age of 40 years. In the neonate, renal function rapidly progresses in
infancy to equal or exceed that of adults. Pediatric patients may have an increased
rate of clearance because a child’s drug metabolism rate is increased compared to
adults. When dosing a drug for a child, the drug may need to be administered more
frequently.
159
160
Other changes also occur with aging. Compared person although the dosing interval may need to
with adults, the neonate has a higher proportion of be increased. Alternatively, smaller doses could be
body mass made up of water and a lower proportion administered over a shorter dosing interval. When
of body fat. The elderly are likely to have a lower the volume of distribution is altered, the dosing
proportion of body water and lean tissue (Figure interval can often remain the same but the dose
11-1). Both of these changes—organ function and administered should change in proportion to the
body makeup—affect the disposition of drugs and change in volume of distribution.
how they are used. Reduced function of the organs
of drug elimination generally requires that doses
of drugs eliminated by the affected organ be given Clinical Correlate
less frequently. With alterations in body water or fat
When adjusting a dose of a drug that follows
content, the dose of drugs that distribute into those first-order elimination, if you do not change
tissues must be altered. For drugs that distribute the dosing interval, then the new dose can
into body water, the neonatal dose may be larger be calculated using various simple ratio and
per kilogram of body weight than in an adult. proportion techniques. For example, if gentamicin
peak and trough serum drug concentrations (in a
Disease States
patient receiving 120 mg every 12 hours) were 9
Drug disposition is altered in many disease states, and 2.3 mcg/mL, respectively, then a new dose
but the most common examples involve the kidneys can be calculated: “if 120 mg gives a peak of 9,
and liver, as they are the major organs of drug elimi- then X mg will give a desired peak of 6,” yielding
nation. In patients with major organ dysfunction, an answer of 80 mg every 12 hours. Likewise,
drug clearance decreases and, subsequently, drug one can check to see if this trough would be
half-life lengthens. Some diseases, such as renal acceptable with this new dose: “if 120 mg gives a
failure or cirrhosis, may even result in fluid reten- trough of 2.3, then 80 mg will give a trough of X,”
tion and an increased volume of drug distribution. yielding an answer of 1.5 mcg/mL.
Alterations in drug clearance and volume
of distribution require adjustments in the dose
administered and/or the dosing interval. For most EXAMPLE
drugs, when clearance is decreased but the volume
of distribution is relatively unchanged, the dose Effect of Volume of Distribution and
administered may be similar to that in a healthy Impaired Renal/Hepatic Function on
Drug Dose
A 23-year-old male experienced a major
traumatic injury from a motor vehicle acci-
dent. On the third day after injury, his renal
function is determined to be good (creati-
nine clearance = 120 mL/minute), and his
weight has increased from 63 kg on admis-
sion to 83 kg. Note that fluid accumulation
(as evidenced by weight gain) is an expected
result of traumatic injury. He is treated with
gentamicin for gram-negative bacteremia.
n initial gentamicin dose of 100 mg is
A
given over 1 hour, and a peak concentra-
tion of 2.5 mg/L is determined. Four hours
FIGURE 11-1. after the peak, the plasma concentration is
Effect of age on body composition. determined to be 0.6 mg/L, and the elimina-

Lesson 11 | Pharmacokinetic Variation & Model-Independent Relationships
161
tion rate constant and the volume of distri- Genetic Factors, Pharmacogenetics, and
bution are determined to be 0.36 hr–1 and Pharmacogenomics
33.6 L, respectively. This volume of 33.6 L Interpatient variability in drug response may result
equals 0.40 L/kg compared to a typical V of from genetically determined differences in metab-
0.2–0.3 L/kg. In this case, the patient’s genta- olism, distribution, and target proteins of drugs.
micin elimination rate constant is similar to Pharmacogenetics is the study of genetic varia-
that found in people with normal renal functions that lead to interpatient variations in drug
tion, but the volume of distribution is much response. This concept is often used interchange-
greater. To maintain a peak plasma genta- ably with pharmacogenomics. In the strictest sense,
micin concentration of 6–8 mg/L, a much pharmacogenetics refers to monogenetic variants
larger dose would have to be administered at in drug response while pharmacogenomics refers
a dosing interval of 6 or 8 hours. Using the to the entire spectrum of genes that interacts to
multiple-dose infusion equation from Lesson determine drug safety and efficacy. The goals of
5 (see Equation 5-1), we would find that a these two areas of study are to optimize drug
dose as high as 220 mg given every 6 hours therapy and limit drug toxicity based on an indi-
would be necessary to achieve the desired
vidual’s genetic profile. Information gained from
plasma concentrations.
studies in these areas will enable clinicians to use
On the other hand, we would expect patients genetic tests to select a drug, drug dose, and treat-
with impaired renal function to have a lower ment duration that will have the greatest likelihood
creatinine clearance and, therefore, a smaller
for achieving therapeutic outcomes with the least
elimination rate constant compared to
potential for adverse effects in a given patient based
patients with normal renal function. A smaller
on DNA profiles. Much work has already been done
than normal elimination rate constant would
in the area of cancer treatment, and information is
produce a longer half-life and would require
emerging in the areas of cardiology, neurology, and
an increase in the dosage interval.
infectious disease.
I n patients with both impaired renal func-
Genetic variations commonly occur either as rare
tion and abnormal volume of distribution
defects or polymorphisms. Rare mutations occur in
values, the dose and dosing interval should
less than 1% of the population while polymorphisms
be adjusted accordingly.
occur in at least 1% of humans. To date, polymor-
J ust as renal dysfunction may alter the phisms in drug-metabolizing enzymes are the most
dosage requirement for drugs eliminated documented examples of genetic variants that result
renally, hepatic dysfunction alters the dosage in altered drug response and toxicity. We will briefly
requirement for hepatically metabolized or discuss two examples of polymorphic metabolizing
excreted drugs. For example, the daily dose of enzymes and corresponding drugs whose plasma
theophylline must be reduced in patients with concentrations and pharmacologic effect may be
liver dysfunction. With this agent, however, a altered as a result of genetic variation: CYP2C9 and
consistent plasma concentration (as opposed warfarin, and CYP2C19 and clopidogrel.
to a large difference in peak and trough
CYP2C9 is a polymorphic isoenzyme that
plasma concentrations) is desired. There-
metabolizes warfarin, phenytoin, and tolbutamide.
fore, with liver dysfunction, smaller doses of
The S-isomer of warfarin is metabolized by this
theophylline than usual are generally admin-
isoenzyme, and genetic alterations can result in
istered but at the usual dosage intervals (two
significant reductions in clearance necessitating
to four times daily). For a continuous intra-
substantial dose reductions. On the other hand,
venous infusion, the infusion rate must be
ultrarapid metabolizers of CYP2C9 require higher
reduced.
doses of warfarin.

162
Another factor to consider with warfarin morphism has been observed in some patients
metabolism is its target enzyme vitamin K oxido- associated with decreased expression of P-glyco-
reductase or VKOR. Warfarin inhibits VKOR, thereby protein (a drug transporter in the duodenum). In
preventing carboxylation of vitamin K–dependent these patients, the bioavailability of P-glycoprotein
clotting factors II, VII, IX, and X. Genetic alterations substrates, such as digoxin, is greatly increased;
in VKOR can result in rare cases of warfarin resis- therefore, a decrease in dose may be required.1
tance in which carriers of these mutations require Many variants have also been observed in the
extremely high warfarin doses, or actually may cytochrome P450 enzyme system. These variations
cause a lack of response to warfarin at any dose. can cause different responses to drugs metabo-
Specifically, the VKORC1 genotype in combination lized by the CYP450 enzyme system. For example,
with CYP2C0 genotype explains approximately poor CYP2D6 metabolizers have been found to
30% of the interpatient variability in warfarin doses have elevated fluoxetine levels, and poor CYP2C19
commonly encountered in clinical practice. metabolizers were found to have an increased inci-
Patients who are intermediate metabolizers dence of fluoxetine adverse effects.2
or poor metabolizers of CYP2C19 may experience Pharmacogenomic research is still in progress.
a reduced response to clopidogrel and potentially Many sequence variations have currently been
require higher doses or alternative antiplatelet observed, but there are countless polymorphisms
therapy for adequate clinical outcomes. The reason left to be discovered. Currently, more than 100 drugs
is that clopidogrel is a pro-drug that must undergo contain references to pharmacogenetic informa-
conversion via CYP2C19 to its active form. tion in their approved labeling, and guidelines for
There are many other examples of differences in the use of genetic information in drug prescribing
response to drugs and adverse drug reactions due are beginning to emerge, including those from
to variation in a patient’s genetic sequence. These the Clinical Pharmacogenetics Implementation
sequence variations can affect enzymes respon- Consortium. These guidelines are available through
sible for drug metabolism, drug targets, and drug the Pharmacogenomics Knowledge Base website
transporters, all of which will lead to deviation in (www.PharmGKB.org).
absorption, distribution, metabolism, and elimi-
nation. With isoniazid, for example, there are two Obesity
distinct subsets of the population with differences Obesity alters drug pharmacokinetics. Because
in isoniazid elimination (Figure 11-2). The elimina- obesity is common in our society, it is an important
tion of isoniazid is said to exhibit a bimodal pattern. source of pharmacokinetic variation. With obesity,
This difference in clearance is caused by geneti- the ratio of body fat to lean tissue is greater than
cally controlled differences in hepatic microsomal in non-obese patients. Fat tissue contains less water
enzyme production. Likewise, genetic differences than lean tissue, so the amount of body water per
in drug elimination also have been observed for kilogram of total body weight is less in the obese
hydralazine, warfarin, and phenylbutazone. Poly- person than in the non-obese person.
For some drugs, alterations in body makeup
that accompany obesity require changes in drug
dosages. Drugs that are lipophilic (such as thio-
pental) and distribute well into fat tissues must
often be given in larger doses to achieve the desired
effects. Drugs that distribute primarily in extra-
cellular fluids (such as the aminoglycosides) may be
given in higher absolute doses to the obese person,
FIGURE 11-2. but the overall milligram per kilogram dose will be
Bimodal distribution for isoniazid half-life. lower. The morbidly obese person who is twice ideal

163
body weight will have an aminoglycoside volume of gel barrier that separates the fluid portion of blood
distribution that is approximately 1.4 times greater from the solid portion. After collection, the blood
than a person of ideal body weight.3 is first allowed to clot, which takes approximately
30 minutes, and is then centrifuged for at least 15
Other Factors minutes to separate the solid components of the
Many other factors may affect drug pharmacokinetics, blood (blood cells, fibrin, fibrinogen, etc.) from the
including pregnancy and drug interactions. Specific fluid component. This fluid component is called
changes in pharmacokinetics during pregnancy serum. If whole blood is centrifuged before it clots,
include increased renal drug clearance, alterations then only the blood cells are separated from the
in volume of distribution, and changes in plasma fluid component, which is called plasma (Figure
protein binding. Another example of an effect 11-3).
on pharmacokinetics is the histamine-2 blocker, Most assays of therapeutically monitored drugs
cimetidine, which inhibits the hepatic enzymes are performed on serum, hence the term serum drug
that metabolize theophylline, thereby decreasing concentration. However, the operations manual for
theophylline clearance. When evaluating drug specific assay instruments often indicate whether
pharmacokinetics in an individual patient, the clini- a particular drug may be tested using plasma or
cian must consider the many factors that may cause serum. Most instruments allow the use of either
variations from the expected results. serum or plasma.
The assay should be performed within 24 hours
Potential Sources of Error in the of sample collection. If this is not possible, refrig-
Collection and Assay of Biologic erate the sample (at 2–6°C) until the assay can be
Samples performed.
If plastic or glass SSTs are used, it is important
Pharmacokinetic calculations depend greatly on the to ensure that the drug to be assayed is not affected
validity of the reported drug concentration from (i.e., absorbed or adsorbed) by the polymeric gel
a biologic sample (e.g., blood, serum, or plasma). barrier used to separate the plasma from the cells.
Using incorrect concentration values to calculate The composition of this barrier depends on the
dosages can result in subtherapeutic or suprathera- brand of SST used. The barriers are usually made of
peutic (i.e., toxic) drug concentrations. Inaccurate
concentration values can result from incorrect drug
sampling or assay procedures. To ensure that drug
concentrations are valid, several factors should be
considered:
• proper laboratory sample collection and
handling,
• physiochemical factors affecting assay
accuracy,
• proper laboratory instrument calibration
and controls check, and
• proper drug administration and sample
timing.
Sample Collection and Handling

To measure drug concentrations, whole blood is
usually collected in a blood collection tube called FIGURE 11-3.
a serum separator tube (SST). The SST contains a Blood products.

164
acrylic, silicon, or polyester polymers, and although

they are generally chemically inert, they can absorb Clinical Correlate
or adsorb the drug being tested.
Use caution when interpreting a serum drug
The degree of absorption or adsorption depends
concentration reported as < X (e.g., < 0.5 mg/L).
on the hydrophilicity of the drug, the type of barrier This is not the same value as X, but instead
used in the SST, the amount of contact time, and means that the sample has no detectible drug
the volume of the plasma sample. For example, concentration above the assay’s lower limit of
decreases ranging from 6 to 64% were reported sensitivity. Consequently, this value could be 0,
in measured concentrations of phenytoin, pheno- and it could have been 0 for many hours. One
barbital, lidocaine, quinidine, and carbamazepine cannot reliably use this value to calculate
when plasma was stored in Vacutainer SST collec- patient-specific K values.
tion tubes.4 Adsorption or absorption of drug by
the SST barrier is of particular concern when small
sample volumes are used (e.g., in pediatric patients) Upper Limit of Drug Detection
or prolonged storage times are required. The upper limit of drug detection indicates the
highest drug concentration that can be accurately
Physicochemical Factors Affecting Assay measured. Plasma drug concentrations above the
Accuracy upper limit will often be reported as higher than
Most commercially available drug assay methods this value. If this occurs, assay parameters can be
are immunoassays that use an antibody specific adjusted to increase the dilution volume of the
for binding sites only on the drug to be assayed. plasma sample, thus allowing higher drug concen-
Various detection methods, such as fluorescence trations to be measured.
polarization with instruments such as Abbott
Diagnostics’ TDX/FLEX/ADX and other instru- Assay Interference
ments, are used to quantitate the amount of drug Assay interferences are generally categorized as
present. These assays can detect the presence of cross-reactivity and physiologic interferences. The
drug at very low concentrations (i.e., microgram or degree of cross-reactivity with other structur-
nanogram amounts). ally similar compounds is called assay specificity.
Several assay-specific factors listed in the assay Cross-reactivity is a function of the specificity of
kit package insert can aid in the clinical interpreta- the antibody used to bind to the drug. Often, this
tion of a plasma drug concentration. antibody will also at least partially bind to other
compounds that are structurally related to the
Lower Limit of Drug Detection desired analyte, such as metabolites and chemical
The lower limit of drug detection indicates the analogues of the analyte.
lowest drug concentration that the assay can reli- For example, gentamicin assays cross-react
ably report. This is a function of the particular assay with the seldom-used aminoglycoside netilmicin,
instrument and is called assay sensitivity. Assay and amikacin assays cross-react with kanamycin;
sensitivity is the lowest measurable drug concen- however, gentamicin and tobramycin assays do not
tration that can be distinguished from zero with generally cross-react. In addition, patients with
95% confidence. Plasma drug concentrations lower impaired renal function who are receiving vanco-
than this concentration should be reported as less mycin have been shown to accumulate a vancomycin
than this value. metabolite called vancomycin crystalline degrada-

165
tion product 1 (CDP-1). CDP-1 can cross-react with (control values) should be tested for each drug
older assays for vancomycin; however, many newer assay per working shift. If these control values are
assay methodologies have reduced this assay inter- out of range, as defined per individual laboratory
ference to an acceptable amount. Clinicians must standards, then this assay should be recalibrated
still be aware of this potential for cross-reactivity. and measurement of the patient’s plasma drug
Physiologic substances in the patient’s sample may concentration repeated.
also interfere with the assay. Examples include
excess amounts of bilirubin, hemoglobin (i.e., hemo- Drug Administration and Sample Timing
lyzed sample), protein, and triglycerides. Plasma To accurately assess drug concentration data and
drug concentrations are usually not affected by such make dosing recommendations, it is important to
interferences. be aware of administration and sampling factors
that may affect the reported drug concentra-
tions. First, drug administration times should be
Clinical Correlate documented, noting any deviations from the recom-
mended dosing schedule. Second, sampling times
Ask your laboratory’s clinical chemistry should be carefully noted so that adjustments can
department for copies of the assay kit package be made in dosage calculations if necessary. Third,
inserts for all drugs that they assay in-house. it is important to note any other medications the
These inserts will provide useful information, patient is receiving. Occasionally, a patient’s sample
such as the upper and lower limits of assay will contain two drugs, one of which can inactivate
sensitivity, as well as interfering and cross- the other, particularly if both drugs are infused
reacting substances. concomitantly, the sample is taken while the inter-
fering drug is infusing, or a sample containing both
drugs is stored at room temperature for a prolonged
Instrument Calibration and Controls period.
Check A good example is the in vitro inactivation of
Each drug assay should be calibrated to establish the aminoglycosides by penicillins. Cephalosporins
a linear relationship between drug concentration have not been shown to inactivate aminoglycosides.5
and the instrument’s detection method. Calibration Penicillins and aminoglycosides form a chemical
of the assay is an automatic process of measuring complex that is not detected by commercially avail-
and plotting different known drug concentra-
tions (i.e., calibrators) based on the instrument’s
method of detection and measurement. Most assay
instruments use some type of spectrophotometric
measurement unit, such as fluorescence polariza-
tion with Abbott’s TDX instrument. Figure 11-4
is a plot of drug concentration versus the instru-
ment’s detection measure (polarization), showing
the linear relationship between concentration and
polarization. Note that the calibration curve is not
linear at very high and very low drug concentra-
tions.
Once this calibration plot or curve is stored in
the instrument’s software, other unknown drug
concentrations (i.e., patient samples) can be accu-
rately determined from this plot. As a quality control FIGURE 11-4.
check, at least two different known concentrations Drug concentration versus net polarization.

166
able drug assays.6 This in vitro inactivation results Throughout this self-instructional course, we
in a falsely low plasma drug concentration report, have emphasized the mathematical relationships
which can in turn result in an unnecessary dosage of specific pharmacokinetic compartmental models
increase. Quantitatively, the aminoglycoside concen- (e.g., one- or two-compartment model after an
tration can decline to less than 10% of its original intravenous bolus or oral dose administration). This
concentration within 24 hours of the beginning of lesson reviews several pharmacokinetic param-
this reaction. These reactions are time and tempera- eters that are derived without the assumption of a
ture dependent. Refrigerating the sample slows the specific model.
inactivation process, and freezing the sample stops The primary purpose of rigorous pharmaco-
it completely. To avoid the inactivation process, it kinetic data analysis, compartmental or model-
is important to adjust the administration times to independent, is to determine the pharmacokinetic
avoid concomitant infusions of aminoglycosides parameters useful in dosing drugs for patients.
and penicillins. Drawing a plasma aminoglycoside Consequently, multiple plasma drug concentra-
concentration during infusion of the penicillin tions are obtained at specific time points in healthy
should be avoided as well. and diseased persons to assess a drug’s population
pharmacokinetic parameters. In clinical practice, it
may be difficult to obtain multiple plasma samples
Clinical Correlate
after the first dose to determine a patient’s pharma-
When assaying the concentration of an cokinetic parameters. Consequently, clinicians use
aminoglycoside from a patient who is population parameters from the literature to make
concomitantly receiving a penicillin, the individual patient dosage calculations.
laboratory must perform the assay immediately Model-independent pharmacokinetic data
or freeze the sample. Freezing the sample analysis provides the opportunity to obtain
instantly stops the in vitro inactivation of the pharmacokinetic values that do not depend on a
aminoglycoside by the penicillin, whereas compartmental model. Total body clearance, mean
refrigerating the sample only slows down residence time (MRT), volume of distribution at
this degradation reaction. If the assay is not
steady state, and formation clearance are four of the
performed immediately, the aminoglycoside level
most frequently used model-independent param-
from the assay will be lower than the patient’s
actual serum aminoglycoside level. eters and are the focus of this section.
The use of model-independent data analysis
techniques to generate model-independent param-
eters offers several advantages over traditional
Model-Independent Relationships compartmental approaches. First, it is not neces-
sary to assume a compartmental model. Many drugs
Until now, we have used a major assumption in possess complex distribution patterns requiring
constructing models for drug pharmacokinetics: two, three, or more exponential terms to describe
that drug clearance remains constant with any size their elimination. As the number of exponential
dose. Drug clearance remains constant for small or terms increases, a compartmental analysis requires
large doses when drug elimination processes are more intensive blood sampling and rigorous data
first order (as described in previous lessons). With calculations. Second, several drugs (e.g., gentamicin)
a first-order elimination process, as the dose of can be described by one, two, or more distribution
drug increases, the plasma concentrations observed compartments, depending on the characteristics
and the area under the plasma drug concentration of the patients evaluated or the aggressiveness of
versus time curve (AUC) increase proportionally. the blood sampling. Therefore, a compartmental
That is, if the dose is doubled, the plasma concen- approach would require that pharmacokinetic
tration and AUC also double. parameters be obtained for each distribution

167
pattern, making it difficult to compare one data set

to another. Third, calculations are generally easier
with model-independent relationships and do not
require a computer with sophisticated software.
One drawback of using model-independent
parameters is the inability to visualize or predict
plasma concentration versus time profiles. This
may result in the loss of specific information that
provides important insight regarding drug disposi- FIGURE 11-5.
tion. Concentration versus time profile after a single intravenous
dose. AUC = area under the plasma drug concentration versus
Like compartmental pharmacokinetic data time curve.
analysis, the main purpose of assessing plasma
concentration versus time data with model- where:
independent relationships is to determine useful
X0 = drug dose, and
pharmacokinetic parameters. These parameters are
usually, but not always, obtained from serial plasma AUC0→∞ = area under the concentration versus
concentration determinations after a single intrave- time curve from time zero to infinity.
nous bolus or oral dose of a drug. This is a model-independent relationship because
In practice, total body clearance and apparent calculations do not depend on a specific compart-
volume of distribution are the two most important mental model. In other words, we only need the
pharmacokinetic parameters because they facilitate dose and the AUC0→∞ to calculate total body clear-
the calculation of maintenance and loading dose ance. Because the dose is known, a determination of
regimens, respectively. Understanding the effect the AUC→∞ is all that is needed.
that disease, altered physiologic state, or drug–drug As you can see from Figure 11-6, the trape-
interaction may have on these pharmacokinetic zoidal rule applies only to drugs whose clearance is
parameters is important in applying these princi- constant with respect to dose and does not apply to
ples to clinical practice. AUC and area under the first drugs whose clearance is nonlinear. Remember, the
moment curve (AUMC) are two tools used to calcu- trapezoidal rule is a model-independent approach
late most model-independent parameters. AUC and used to directly calculate the AUC of the drug from
AUMC are discussed in the next section. time zero to the time point that coincides with the
last measured plasma concentration value (tlast).
Total Body Clearance However, because AUC must include all of the area
Total body clearance (Clt) is the most important from zero to infinity after a single intravenous
pharmacokinetic parameter because it relates bolus dose, an estimate of the area between tlast and
the dosing rate of a drug to its steady-state
concentration. It is usually used to calculate a main-
tenance-dosing regimen. An estimate of Clt for a
drug is usually obtained after a single intravenous
bolus dose (Figure 11-5). Total body clearance is
calculated with the following equation:
X0
Clt =
AUC0→∞
FIGURE 11-6.
Concentration × time versus time curve. AUMC = area under
(See Equation 3-5.)
the first moment curve.

168
infinity is needed. This terminal area can be easily

obtained by the following equation:
C last
terminal area =
λ
where:
C last = last measured plasma concentration, and
λ =terminal elimination rate constant.
Two key assumptions in estimating this

terminal AUC are that you have a reliable estimate
of the terminal elimination rate constant (i.e., slope)
and that this value remains constant between tlast FIGURE 11-7.
and infinity. Concentration × time versus time curve with trapezoids.
To determine several model-independent rela-
tionships, such as MRT and volume of distribution Consequently, the terminal area, which includes the
at steady state, it is important to understand how portion of the curve from tlast to infinity, must be
to calculate the AUMC. The AUMC is the area under estimated. Assuming the terminal elimination slope
the drug concentration versus time versus time curve. remains constant over this time period, the terminal
The AUMC is generated with the AUC data from the area is calculated with the following equation:
concentration versus time profile for a single intra-
venous bolus dose (see Figure 11-5). (C last × t last ) C last
=
terminal area + 2
To calculate AUMC after a single intravenous λ λ
bolus dose of a drug, it is necessary to collect serial where:
drug plasma concentrations over time, determine Clast = last observed plasma concentration,
concentration × time for each plasma concentra-
tlast = time of the last observed plasma
tion, and plot these values versus time on graph
concentration, and
paper (see Figure 11-6).
λ = terminal elimination rate constant from
As you can see from Figure 11-6, the shape
the concentration versus time curve. λ is
of the [concentration × time] versus time curve
used here (instead of K) to indicate that
is very different from the drug plasma concen-
this represents elimination in a model-
tration (C) versus time (t) plot used to calculate
independent or noncompartmental analysis.
AUC. The trapezoidal rule can be used to calculate
AUMC. Following a plot as in Figure 11-6, a series of
straight lines can be drawn from the concentration Mean Residence Time
× time point to its accompanying time value on the MRT is defined as the average time intact drug mole-
x-axis, forming individual trapezoids (Figure 11-7). cules transit or reside in the body. For a population
The area of each trapezoid is calculated with the of drug molecules, individual molecules spend
following equation: different times within the body. Following the princi-
(C 2 × t 2 ) + (C 1 × t 1 ) ples of statistical probability, specific drug molecules
=
area of trapezoid (t 2 − t 1 ) may be eliminated quickly, whereas others may
2
remain in the body much longer. Consequently, a
The sum of all of the trapezoidal areas yields distribution of transit times can be characterized by
an estimate of the AUMC from time zero to the last a mean value. In other words, elimination of a drug
observed time point. As in calculating AUC, it is can be thought of as a random process. Residence
important to obtain AUMC from time zero to infinity. time reflects how long a particular drug molecule

169
remains or resides in the body. The MRT reflects a patient’s drug dosing regimen, but is useful when
the overall behavior of a large number of drug assessing the impact that a specific drug treatment,
molecules. This parameter is not used frequently in disease, or altered physiologic state may have on a
clinical practice to monitor patients. However, it is specific metabolic pathway of a drug. The following
useful when comparing the effect of disease, altered equations are used to calculate the formation clear-
physiologic state, or drug–drug interaction on the ance of a drug:
pharmacokinetics of a specific drug. MRT can be
calculated with the following equation: ClP →m1 = Fm 1 Clt
where:
AUMC0→∞
MRT = ClP→m1 = fractional clearance of the parent
AUC0→∞
drug (P) to form metabolite 1 (m1),
Volume of Distribution at Steady State Fm1 = fraction of metabolite m1 formed from
a single dose of the parent drug, and
Volume of distribution at steady state (Vss) is a
parameter that relates total amount of drug in the Clt = total body clearance.
body to a particular plasma concentration after
Or:
a single dose. This parameter is not affected by
changes in drug elimination or clearance, making it  m  X 0 
a useful tool in assessing the effect disease, altered ClP →m 1 =  1,u  
physiologic state, or drug–drug interaction may  X 0  AUC0→∞ 
have on the volume of distribution of a drug. Vss
m1,u
was calculated previously but was only applicable =
to a drug fitting a two-compartment model. The AUC0→∞
following equation for Vss does not depend on the
model used to describe drug distribution or elimi- where:
nation from the body: m1,u = amount of metabolite m1 excreted in
the urine.
=
Vss MRT × Clt
For example, if a drug is metabolized by three
And since: separate enzyme systems, each producing a unique
metabolite, what effect would the addition of a
AUMC0→∞ X0
MRT = and Clt = known hepatic enzyme inducer have on the indi-
AUC0→∞ AUC0→∞ vidual metabolic pathways? Figure 11-8 provides a
visual perspective of this situation.
then:
X 0 × AUMC0→∞
Vss =
(AUC0→∞ )2
Formation Clearance
Formation clearance (ClP→mX) is a model-indepen-
dent parameter that provides a meaningful estimate
of the portion of the total body clearance that is
accounted for by production of a specific metabo-
lite. Formation clearance is analogous to systemic FIGURE 11-8.
and renal clearance of a drug and refers to the Metabolic pathways for a parent drug, where m1 = metabolite 1,
m1,u = amount of m1 excreted in the urine, m2 = metabolite 2,
formation of metabolites in the course of drug elim- m2,u = amount of m2 excreted in the urine, m3 = metabolite 3,
ination. This parameter is not used to individualize and m3,u = amount of m3 excreted in the urine.

170
To simplify this example, we will assume that data would suggest that the enzyme(s) responsible
systemic clearance equals hepatic clearance, these for the formation of m2 and m3 was significantly
three metabolic pathways account for 100% of the increased by the enzyme inducer, whereas the
hepatic clearance of the drug, the metabolite is enzyme(s) responsible for the formation of m1 was
rapidly secreted unchanged in the urine, and the unaffected. The preceding example demonstrates
dose is equal to 100 mg. Table 11-1 shows the effect the value of formation clearance versus the more
of an enzyme inducer on each metabolic pathway traditional approach of calculating the percentage
portrayed in Figure 11-8 as shown by changes in of a drug dose excreted as a specific metabolite.
the percentage of drug dose excreted in the urine
for each metabolite and formation clearance. References
As Table 11-1 shows, the administration of an 1. Evans WE, McLeod HL. Pharmacogenomics—
enzyme inducer substantially increased the systemic Drug disposition, drug targets, and side effects. N
clearance of this drug, from 25 to 75 mL/minute. Engl J Med 2003;348(6):538–47.
However, the change in the percentage of the dose
2. Mancana D, Kerewin RW. Role of pharmaco-
excreted as a specific metabolite does not exactly
genomics in individualising treatment with SSRIs.
reflect the change in formation clearance values.
CNS Drugs 2003:17(3):143–51.
The percentage of dose excreted in the urine for
m1 was reduced threefold, but no change in the 3. Bauer LA, Blouin RA, Griffin WO, et al. Amikacin
formation clearance was observed. This means that pharmacokinetics in morbidly obese patients.
the enzyme inducer had no effect on the enzyme Am J Hosp Pharm 1980; 37:519–22.
responsible for producing m1. 4. Dasgupta A, Dean R, Saldana S, et al. Absorption
On the other hand, treatment with the enzyme of therapeutic drugs by barrier gels in serum
inducer produced only a 1.3-fold increase in the separator blood collection devices. Am J Clin
percentage of the dose excreted in the urine for m2 Pathol 1994;101:456–61.
but a fourfold increase in its formation clearance. 5. Spruill WJ, McCall CY, Francisco GE. In vitro inac-
Finally, the percentage of dose excreted in urine tivation of tobramycin by cephalosporins. Am J
for m3 was unchanged despite a threefold increase Hosp Pharm 1985;42:2506–9.
in its formation clearance. Because the formation 6. Riff LF, Jackson GG. Laboratory and clinical condi-
clearance of a drug to a metabolite reflects more tions for gentamicin inactivation by carbenicillin.
accurately the activity of that specific enzyme, the Arch Intern Med 1972;130:887–91.
TABLE 11-1. Changes in Formation Clearance of Three Metabolites as a Result of Enzyme Induction
Percentage of Dose Formation Clearance

Metabolite Excreted in Urine (mL/minute)
Control (Clt = 25 mL/minute) m 1 20.0 5.0
m 2 50.0 12.5
m 3 30.0 7.5
Enzyme induction (Clt = 75 mL/minute) m 1 6.7 5.0
m 2 63.3 47.5
m 3 30.0 22.5

171
REVIEW QUESTIONS
11-1. The proportion of total body weight that is 11-6. Assay cross-reactivity refers to diminished
water is lowest in: assay performance caused by:
A.
healthy adults. A. physiologic substances found in some
B.
neonates. patients’ plasma that directly affect the
assay itself.
C.
elderly.
B. structurally related drug compounds or
D.
teenagers.
metabolites for which the assay method
11-2. With dysfunction of the major organs of measures as if they were the desired
drug elimination (kidneys and liver), drug assay compound.
clearance, volume of distribution, and drug C. an in vitro inactivation of one drug by
plasma protein binding may be affected. another drug that is also present in the
A.
True patient’s plasma.
B. False D. none of the above.
11-3. For drugs that distribute primarily in extra- Indicate Yes or No for Questions 11-7 through
cellular fluid, a dose for an obese person 11-10:
should be calculated using total body weight. Yes = the accuracy of the drug concentrations
A. True is of concern and should be redrawn.
B. False No = the accuracy of the drug concentrations
is not of particular concern.
11-4. The fluid portion of a sample of whole
blood allowed to clot for 30 minutes before 11-7. A gentamicin concentration from a sample
centrifugation is called: stored at controlled room temperature
and assayed 24 hours after it was collected
A. serum.
from a patient receiving both ampicillin and
B. plasma. gentamicin.
C. serous fluid. A. Yes
D. citrated blood. B. No
11-5. The fluid portion of whole blood centrifuged 11-8. A plasma tobramycin concentration from a
before clot formation is called: sample stored at controlled room temper-
A. serum. ature and assayed 24 hours after it was
B. plasma. collected from a patient receiving both
C. serous fluid. tobramycin and ceftazidime.
D. citrated blood. A. Yes
B. No

172
11-9. A plasma gentamicin concentration from a 11-2. A. CORRECT ANSWER. Major organ
sample stored in a freezer until assayed 12 dysfunction can affect most pharmaco-
hours after it was collected from a patient kinetic parameters.
receiving both ampicillin and gentamicin. B. Incorrect answer
A. Yes
B. No 11-3. A. Incorrect answer
B. CORRECT ANSWER. The proportion
11-10. A plasma gentamicin concentration from of fat tissue that is extracellular fluid
a sample assayed immediately after it was is less than in lean tissue, but the drug
collected from a patient receiving both will still distribute somewhat in the
piperacillin and gentamicin. adipose extracellular fluid.
A. Yes
B. No
B. Incorrect answer. Plasma contains
11-11. The trapezoidal rule can be used to calculate clotting factors.
AUC for model-independent relationships. C. Incorrect answer. Serous fluid is a
A. True natural body fluid and is not centri-
fuged to remove cellular components.
B. False
D. Incorrect answer. Citrated blood
11-12. The ratio of AUMC→∞ to AUC→∞ is called: contains citrate additives that keep the
A. trapezoidal rule. blood from clotting.
B. total body clearance. 11-5. A. Incorrect answer. Serum does not
C. mean residence time. contain clotting factors.
D. formation clearance. B. CORRECT ANSWER
C. Incorrect answer. Serous fluid is a
11-13. Which statement(s) is/are false about the
natural body fluid and is not centri-
calculation of formation clearance (ClP→mX)?
fuged to remove cellular components.
Formation clearance can be used to
calculate the: D. Incorrect answer. Citrated blood
contains citrate additives that keep the
A. clearance rate of individual metabolites
blood from clotting.
of a drug.
B. mean residence time. 11-6. A. Incorrect answer. This is assay interfer-
C. total body clearance of a drug that has ence.
multiple metabolites. B. CORRECT ANSWER
D. A and C. C. Incorrect answer. Assay cross-reactivity
does not involve assay measurement of
inactivated products that result from
ANSWERS some physiochemical process.
D. Incorrect answer
11-7. A. CORRECT ANSWER. Ampicillin will
C. CORRECT ANSWER. The proportion of
inactivate gentamicin in vitro.
the body that is water is greatest in the
neonate and lowest in the elderly. B. Incorrect answer

173
11-8. A. Incorrect answer 11-12. A. Incorrect answer. Trapezoidal rule is a

B. CORRECT ANSWER. Aminoglycosides method to calculate AUC.
are not inactivated by cephalosporins, B. Incorrect answer. Total body clearance is
just penicillins. X0/AUC0→∞.
C. CORRECT ANSWER
D. Incorrect answer. Formation clearance
B. CORRECT ANSWER. Freezing this is a calculation of metabolite clearance.
sample will stop inactivation from
occurring. 11-13. A, C, D. Incorrect answers
B. CORRECT ANSWER. MRT is calculated
using AUC and AUMC.
B. CORRECT ANSWER. Inactivation does
not have time to occur if you assay the
sample immediately.
11-11. A. CORRECT ANSWER. The trapezoidal

rule is a model-independent method for
AUC calculation.
B. Incorrect answer

174
Discussion Points
D-1. Write a pharmacy protocol to ensure proper D-2. Try to get a package insert from your labora-
serum drug concentration collection and tory on any therapeutically monitored drug.
assay. Describe the type of information found.
Specifically, how are the issues of assay
sensitivity, specificity, and cross-reactivity
noted?

Practice Set 3
Definitions of symbols and key equations are: Before proceeding to the questions below, on
AUC = area under plasma concentration versus linear graph paper, plot the plasma drug concentra-
time curve tion versus time data for the two formulations.
F = fraction of drug reaching systemic
circulation
Clt = total drug clearance from body =
QUESTIONS
dose/AUC
Ka = absorption rate constant PS3-1. What is the AUC0–12 for the oral tablet
hr
formulation (using the trapezoidal method)?
The following applies to Questions PS3-1 to PS3-4.
The relative bioavailabilities of two dosage forms (a A. 7.25 (mg/L) × hour
sustained-release tablet and an oral solution) of oral B. 71.25(mg/L) × hour
morphine sulfate are being compared. The following C. 62.34 (mg/L) × hour
plasma drug concentrations were obtained after 30 mg D. 64.51 (mg/L) × hour
of each was administered:
PS3-2. What is the AUC0–12 hr for the oral solution
Concentration (mg/L)
formulation (using the trapezoidal method)?
Time after Sustained-Release
Dose (hours) Tablet Oral Solution A. 6.25 (mg/L) × hour
0 0 0 B. 71.25 (mg/L) × hour
0.5 2.26 19.80
C. 47.42 (mg/L) × hour
1.0 4.57 18.31
D. 64.51 (mg/L) × hour
1.5 5.27 15.49
2.0 6.73 12.37 PS3-3. What are the peak plasma drug concentra-
3.0 7.36 9.62 tions for the oral tablet and oral suspension,
4.0 7.40 6.85 respectively?
5.0 7.50 5.52 A. 7.5 and 19.8 mg/L
6.0 6.24 3.10 B. 6.5 and 18.9 mg/L
8.0 4.70 1.79
C. 4.5 and 12.5 mg/L
12.0 2.76 1.58
D. 19.5 and 7.9 mg/L

PS3-4. Which product has greater bioavailability?
A. Oral solution
B. Oral tablet
Purchased by nova southeastern university, Todd Puccio 175

176
The following applies to Question PS3-5. A single Before proceeding to Question PS3-5, plot the
oral dose (500 mg) of a sustained-release procain- concentration versus time data on semilog graph
amide tablet was given, and the following plasma paper.
drug concentrations were determined:
PS3-5. What is the absorption rate constant (Ka)
Time after Plasma Drug Concentration of this formulation (using the method of
Dose (hours) (mg/L)
residuals)?
0 0
0.1 0.60 A. 2.1 hr–1
0.25 1.35 B. 1.2 hr–1
0.5 2.34
C. 2.5 hr–1
0.75 3.05
1.0 3.55 D. 3.2 hr–1
2.0 4.33
4.0 3.89
8.0 2.41
12.0 1.49

Practice Set 3 |
177
ANSWERS
PS3-1. A, B, D. Incorrect answers PS3-2. A, C, D. Incorrect answers
C. CORRECT ANSWER. Using the equation B. CORRECT ANSWER. Using the equation
found in Figure 3-10, found in Figure 3-10,
(2.26 + 0)(0.5 − 0) (0 + 19.80)(0.5 − 0)

= 0.56 = 4.95
2 2
(4.57 + 2.26)(1− 0.5) (18.31+ 19.80)(1− 0.5)
= 1.71 = 9.53
2 2
(5.27 + 4.57)(1.5 − 1) (15.49 + 18.31)(1.5 − 1)
= 2.46 = 8.45
2 2
(6.73 + 5.27)(2 − 1.5) (12.37 + 15.49)(2 − 1.5)
= 3.00 = 6.97
2 2
(7.36 + 6.73)(3 − 2) (9.62 + 12.37)(3 − 2)
= 7.05 = 10.99
2 2
(7.40 + 7.36)(4 − 3) (6.85 + 9.62)(4 − 3)
= 7.38 = 8.24
2 2
(7.50 + 7.40)(5 − 4) (5.52 + 6.85)(5 − 4)
= 7.45 = 6.18
2 2
(6.24 + 7.50)(6 − 5) (3.10 + 5.52)(6 − 5)
= 6.87 = 4.31
2 2
(4.70 + 6.24)(8 − 6) (1.79 + 3.10 )(8 − 6)
= 10.94 = 4.89
2 2
(2.76 + 4.70)(12 − 8) (1.59 + 1.79)(12 − 8)
= 14.92 = 6.74
2 2
+ ____ +____
= 62.34 (mg/L) × hr = 71.25 (mg/L) × hr

178
PS3-3. A. CORRECT ANSWER. Observe peak PS3-5. A, C, D. Incorrect answers

concentration from AUC plot. B. CORRECT ANSWER. First, the data
B, C, D. Incorrect answers points at 4, 8, and 12 hours are on the
straight-line terminal portion of the plot
PS3-4. A. CORRECT ANSWER. The oral tablet and, therefore, are not used to calculate
has a smaller AUC and, therefore, has the residual line. Next, the terminal,
a lower bioavailability than the oral straight-line portion of the graph is back-
solution. extrapolated to the y-axis. For each time
B. Incorrect answer at which a concentration was actually
determined, the concentration corre-
sponding to the back-extrapolated line
is noted (extrapolated concentration).
The residual is the remainder of the
actual concentration subtracted from
the extrapolated concentration. The Ka
is the negative slope of the natural log of
the residual concentration versus time
curve. We can choose any two residual
points to determine the slope, but it is
usually best to select the points most
widely separated by time. Therefore,
 ∆y  In 5.57 − In 0.57
Ka =
− =
 ∆x  0.1 hr − 2.0 hr
 1.72 − ( −0.56) 
= − 
 −1.9 hr 
= 1.20 hr −1

LESSON 12
Aminoglycosides
This lesson reviews the various renal function assessment equations followed by
cases on the appropriate dosing of aminoglycosides.
Renal Function Assessment

The National Kidney Foundation (NKF) stages chronic kidney disease (CKD) into five
stages based on glomerular filtration rate (GFR) reported in units of mL/min/1.73 m2
body surface area (BSA) as shown in Table 12-1. Recently, the NKF has recom-
mended the use of the modified diet in renal disease (MDRD) equation (MDRDEQ)
to estimate a patient’s GFR as a screening tool for early detection of CKD (http://
www.kidney.org/professionals/KDOQI/gfr.cfm).
GFR has historically been measured by renal clearance of substances that
are 100% excreted via glomerular filtration with no renal tubular reabsorption
or secretion, such as inulin clearance, and more recently by renal clearance of
radio-labeled 125I-iothalamate, which has now become the gold standard for GFR
measurements. However, actual GFR measurements are not commonly done in
clinical practice. Instead, clinicians rely on more easily performed estimations of
GFR, such as measured creatinine clearance (CrCl), estimated CrCl, and now esti-
mated GFR via the MDRD equation.
CrCl measurement requires a 24-hour collection of urine as shown in the
formula below:
UV
CrCl =
P × 1440
where:
U = urinary creatinine concentration,
V = volume of urine collected,
P = plasma creatinine concentration (taken at midpoint of
urine collection), and
1440 = number of minutes in 24 hours.
Due to the cumbersome nature of direct measurements of either GFR or CrCl, renal
function is most commonly estimated with the Cockcroft–Gault creatinine clearance
equation (CGEQ) or the MDRD4revised estimated GFR equation. The various versions
of the MDRDEQ were developed in a sample that consisted primarily of patients
with some degree of CKD and as such may not accurately assess renal function in
non-CKD patients.1 Table 12-2 shows several versions of both of these equations,
including the most commonly used CGEQ using ideal or adjusted body weight.
179
180
TABLE 12-1. Stages of Chronic Kidney Disease

Stage 1 Kidney damage with normal or increased GFR > 90 mL/min/1.73 m2
Stage 2 Kidney damage with mildly decreased GFR 60–89
Stage 3 Moderately decreased GFR 30–59
Stage 4 Severely decreased GFR 15–29
Stage 5 Kidney failure < 15
GFR, glomerular filtration rate.
Some controversy now exists as to which demographic biases, especially age. The CGEQ was
equation, CGEQ or MDRDEQ, is best to use for derived from a simple general linear multiple regres-
renal adjustments of drug doses. Complicating sion analysis such that each factor (age, weight,
this issue is the recent conversion to new global serum creatinine value) in the equation is linearly
serum creatinine assay standards that result expressed for the entire tested range of the values.
in a more accurate measurement of creati- Conversely, the MDRDEQs were correlated using
nine (yielding a value 10% to 20% lower than log transformed values and then re-expressed as a
older assays).2 Newer measurements can now be multiplicative linear model that now contains expo-
reported two places past the decimal (i.e., 1.68 mg/dL). nents for the variables of age and serum creatinine
The MDRD4revised equation is the recommended and, therefore, produce a geometric relationship
version of the MDRD equation for use with these across the range of values for each variable tested.
new assay standards. Figure 12-1 is a plot of the age component for
Much research has been done to determine both equations, showing that the MDRDEQ calcu-
which equation is the best; however, it appears lates a much smaller decline in GFR from age 40
that these equations are so dissimilar in their to 80 years than does the CGEQ. Therefore, the
formulation that meaningful comparisons are diffi- MDRDEQ may not predict age-related declines in
cult to perform and are subject to various patient renal function in the elderly as well as the CGEQ.3
TABLE 12-2. Equations Used to Estimate Creatinine Clearance (CrCl) or Glomerular Filtration Rate (GFR)
Cockcroft–Gault estimation of CrCl
Original form used total weight with no BSA adjustment TBW(0.85 if female)(140 – age)/(72 × Cr)
Cockcroft–Gault equation most commonly recommended, (IBW or AdjBW*)(0.85 if female)(140 – age)/(72 × Cr)
using IBW or AdjBW*
Modified diet in renal disease (MDRD) equations
MDRD 6 variable equation with UUN 198 (Cr–0.858 × age–0.167) × BUN–0.293 × UUN–0.249 [× 1.178 if black and ×
0.822 if female]
MDRD 6 variable equation with albumin 170 (Cr–0.999 × age–0.176) × BUN–0.170 × albumin–0.318 [× 1.178 if black and
× 0.822 if female]
MDRD original 4 variable equation 186 (Cr–1.154 × age–0.203) [× 1.212 if black and × 0.742 if female]
MDRD revised 4 variable equation with new 175 (Cr–1.154 × age–0.203) [× 1.212 if black and × 0.742 if female]
creatinine assay standards
AdjBW, adjusted body weight; BSA, body surface area; BUN, blood urea nitrogen; Cr, creatinine; IBW, ideal body weight; TBW, total body weight;
UUN, urine urea nitrogen.
* AdjBW = IBW + 0.4(TBW – IBW)

Lesson 12 | Aminoglycosides
181
2. Most drug manufacturers address dosing

adjustments in patients with renal
impairment by providing tables showing
CrCl versus drug dose, with instructions to
adjust dose based on patients’ estimated
CrCl calculated using the CGEQ.
3. If your laboratory is reporting serum
creatinine values that are calibrated to the
Cockcroft–Gault new IDMS-traceable creatinine assay, be
aware that these newer serum creatinine
assays will report values to two places past
the decimal and will produce values that are
lower by as much as 10% to 20% (i.e.,
~0.3 mg/dL) compared with older assay
methods. This, in turn, will result in a CGEQ
estimation of CrCl that is slightly higher than
that using the older creatinine calibration
techniques. Consequently, a clinical
adjustment may be needed when applying
FIGURE 12-1.
A plot of age components of modified diet in renal disease
a manufacturer’s dosing adjustment data
(MDRD) equation and Cockcroft–Gault equation, showing that based on older creatinine assays to this
the MDRD calculates a much smaller decline in GFR from age newer reported creatinine value.
40 to 80 years than does the Cockcroft–Gault.
Considerations when using MDRD equations
to adjust doses:
Disregarding this ongoing controversy, it is
1. Note that the MDRD equations are
reasonable to use the CGEQ to adjust drug dosing
predictors of GFR and not CrCl, and,
according to manufacturer’s dosing tables that were
therefore, their use as replacements of
developed using this same CGEQ. By extension, most
CGEQ estimates for CrCl for drug dosing
pharmacokinetic population values for the elimina- adjustments will require an individualized
tion rate constant (K) were also developed from assessment of many dosing situations.
regression analyses of drug clearance versus CrCl
2. The MDRD equations were initially validated
via the CGEQ, and thus more closely match existing in patients with chronic kidney disease and
drug dosing tables. may not be as easily generalized to other
subsets of patients, including the elderly, as
is the more commonly used CGEQ.
Clinical Correlate
3. The MDRD equation may or may not
Considerations when using the Cockcroft– actually be more accurate than the CGEQ
Gault equation to adjust drug doses in for any given patient; however, it can
declining renal function: be more easily calculated and reported
using routine clinical chemistry analyzer
1. Use either IBW or an adjusted body weight software, as it does not require the
in the formula and do not make any patient’s weight or height.
further BSA adjustments as this formula 4. There is no useful conversion factor to
contains a body size factor of weight/72, convert MDRD GFR to the equivalent CGEQ
which is sufficient to adjust the result for value because of the differences in the
the patient’s body size or BSA. Value, in regression models used. Sometimes the
units of mL/min, can now be assumed to MDRD value will be higher and sometimes
approximate a patient’s GFR expressed in the CGEQ values will be higher.
units of mL/min/1.73 m2.

182
Aminoglycoside Dosing or
(0.85)(140 − age)IBW
Individualization of aminoglycoside dosing regi- CrClfemale mL/min =
mens is important to optimize efficacy while 72 × SCr
minimizing potential toxicity. Cases 1–4 outline where:
traditional dosing methods of individualized dosing,
CrCl = creatinine clearance (mL/min per
and Cases 5 through 7 focus on the extended
1.73 m2 BSA),
interval administration of aminoglycosides.
age = patient’s age (years),
Because the currently available intravenous
aminoglycosides (gentamicin, tobramycin, and IBW = ideal body weight (kilograms) or
amikacin) exhibit similar pharmacokinetics, case adjusted body weight (AdjBW) in obese
discussions of one aminoglycoside can be extrapo- patients, and
lated to any other. Although amikacin has the same SCr = serum creatinine concentration
pharmacokinetic profile as other aminoglycosides, (milligrams per deciliter).
it requires doses and target concentrations approx-
imately two to four times as high as the other
Calculate Ideal Body Weight or
aminoglycosides.
Adjusted Body Weight
Because creatinine is produced by muscle metabo-
Several key points should be reviewed before
lism (and not by fat), we must use the patient’s IBW
beginning these cases. Aminoglycosides are
or AdjBW when estimating creatinine clearance.
excreted unchanged by renal glomerular filtra-
The IBW for adult males can be estimated by:
tion. The elimination, therefore, is proportional to
a patient’s GFR, which can be estimated by deter-
IBW = 50 kg + 2.3 kg for each inch over 5 feet
mining CrCl.
in height (See Equation 9-2.)
The IBW for adult females is:
Estimation of Elimination Rate
Constant (K) and Volume of IBW = 45.5 kg + 2.3 kg for each inch over 5 feet
Distribution (V) for All Aminoglycosides in height (See Equation 9-2.)
In obese patients, the use of total body weight
Calculate Estimated Creatinine Clearance in the CGEQ overestimates creatinine clearance
Sometimes it is impractical or impossible to collect calculations, and the use of IBW underestimates
a 24-hour urine specimen; CrCl must then be esti- calculation of this variable. Consequently, an AdjBW
mated from serum creatinine. Although there are should be used. If a patient’s actual body weight is
several formulas for estimating CrCl, we use the ≥ 30% above his or her IBW, then the AdjBW must
Cockcroft–Gault equation4: be used to calculate CrCl5:
AdjBW = IBW + 0.4(TBW – IBW) (See Equation 9-3.)
(140 − age)IBW
CrClmale mL/min =
72 × SCr For a patient who weighs less than IBW, the actual
body weight would be used in the CGEQ to calculate
(See Equation 9-1.) CrCl.

183
Or, for example, one commonly used regression

Clinical Correlate equation is:
Close examination of the Cockcroft–Gault slope (K ) = 0.00293 (CrCl) + 0.014

equation reveals that serum creatinine values where K is the elimination rate constant for aminogly-
less than 1 mg/dL greatly elevate the calculated cosides (population estimate). This equation will be
CrCl value. This is especially true for elderly used throughout the lesson for all aminoglycosides.
patients for whom unrealistically high CrCl These estimates also involve the appropriate calcu-
values may be calculated using this equation. lations of CrCl and IBW. Because many small-sample
The elderly often have reduced muscle mass studies have been done to estimate K and V, there are
as a fraction of TBW, and so may generate many different estimates for both. We shall use:
less creatinine than a younger patient of
similar weight. In this population, the serum 12-1 K = 0.00293 (CrCl) + 0.014
creatinine value may not be an appropriate
indicator of the patient’s true renal function. It 12-2 V = 0.24 L/kg (IBW)
has been recommended to either round the low
serum creatinine values up to 1 mg/dL before
calculating CrCl, or round the final calculated Clinical Correlate
CrCl value down to ~100 mL/minute or less.
Although these recommended adjustments yield Approximately 10% of a given aminoglycoside
more accurate estimations, they still add error dose distributes into adipose tissue. When
calculating aminoglycoside volume of distribution,
to the original CrCl calculation.
one can use either IBW or an AdjBWAG formula
that reflects this 10% of distribution into adipose
tissue as follows:
Calculate Estimated Elimination Rate
and Volume of Distribution 12-3 AdjBWAG = IBW + 0.1(TBW – IBW)
To calculate an initial maintenance dose and dosing Note that this adjustment is less than that used
interval using traditional dosing methods, we must when calculating AdjBW for use in the creatinine
use population estimates for the elimination rate clearance formula. In clinical practice, the
constant (K) and the volume of distribution (V). AdjBWAG is rarely used.
Population estimates of K are derived from small
studies that correlate an aminoglycoside’s clearance
(and hence K) to the patient’s CrCl (Figure 12-2).
Creatinine clearance and aminoglycoside clearance
are not equal; some amount of aminoglycoside is
eliminated by organs other than the kidneys. When
creatinine clearance is zero, the aminoglycoside
clearance is still approximately 0.014 mL/minute,
reflecting this nonrenal clearance and, perhaps,
some active tubular secretion.
The equation for the line of best fit through
these points can be used to estimate an elimina-
tion rate constant (K) for this sample of patients, as
shown here:
FIGURE 12-2.
Y = mX + b Aminoglycoside clearance versus creatinine clearance.

184
Clinical Correlate CASE 1

The values for K and V represent population A 66-year-old white female, SG, is hospitalized
estimates of the elimination rate and volume of for lysis of adhesions secondary to previous
distribution, respectively, based on statistical bowel resection. Before surgery, you are asked
averages with relatively large standard to begin this patient on an aminoglycoside. Other
pertinent patient data include: height, 5' 8'';
deviations. For this reason, it is important
weight, 52 kg; and serum creatinine, 1.02 mg/dL.
to obtain peaks and troughs after the initial
dosing regimen is established to properly adjust
the dosage and dosing interval based on the Problem 1A. Calculate an appropriate amino-
individual patient’s specific pharmacokinetic glycoside maintenance dose and loading dose,
data. including the most appropriate dosing interval, for
patient SG. Assume a desired Cpeak of 6 mg/L and a
Ctrough of 1 mg/L.
Population values of K and V for the aminoglyco-
Clinical Correlate sides should be used to estimate maintenance doses.
A patient usually will receive a loading dose over
Some clinicians would use the above K and V 1 hour when therapy is initiated because a loading
equations as initial population estimates for all dose quickly brings aminoglycoside plasma concentra-
aminoglycosides; however, other clinicians may tion close to the desired therapeutic concentration. If a
use equations with slightly different numbers loading dose is not given, the patient’s aminoglycoside
based on regression equations derived from concentration will not reach the desired concentration
similar studies. Additionally, most clinicians will until steady state is achieved—in five drug half-lives.
use the same estimates for all aminoglycosides; Lessons 4 and 5 describe the mathematical
however, some will use slightly different models used for various multiple dose, intravenous
equations for each aminoglycoside. We will use drug dosing situations. For aminoglycosides, which
the above equation for all aminoglycosides as are usually given intravenously over 30–60 minutes at
they are all excreted renally via the exact same regular (i.e., intermittent) intervals, Lesson 5 describes
mechanism. the appropriate dosing equation. A quick review of
Lessons 4 and 5 may help you understand the deriva-
tion of these equations.
Desired Aminoglycoside Briefly, this equation is arrived at by taking the
Plasma Concentrations equation from Lesson 4 for a single intravenous
The traditional ranges for desired aminoglycoside bolus dose and adding the appropriate factors for
plasma concentrations are a peak of 4–10 mg/L the following:
for gentamicin and tobramycin and 15–30 mg/L • multiple doses,
for amikacin, and a trough of 1–2 mg/L for genta- • simultaneous drug administration and drug
micin and tobramycin and 5–10 mg/L for amikacin.5 elimination,
Attainment of adequate peak concentrations is • drug administration over 30–60 minutes
related to efficacy for some infections, and a low instead of an intravenous bolus, and
trough concentration may minimize the risk of • attainment of steady state (for simplicity,
nephrotoxicity and ototoxicity. 1 hour is assumed for the drug administra-
tion time).

185
The equation is: However, because SG’s actual body weight of 52 kg

is less than her IBW, we should use her actual body
K 0 (1− e −Kt ) weight in the calculation of CrCl and in the estima-
C peak (steady state) =
VK (1− e −K τ ) tion of her volume of distribution (V).
Estimated CrCl (via the Cockcroft–Gault
(See Equation 5-1.) equation) is:
where: (IBW*)(140 − age)0.85
CrClfemale mL/min =
Cpeak(steady state) = desired peak concentration at 72 × SCr
steady state (milligrams per liter),
(52)(140 − 66)0.85
K0 = drug infusion rate (also main- =
tenance dose you are trying to 72 × 1.02
calculate, in milligrams per hour),
= 45.54 mL/min
V = volume of distribution (population
*the lesser of IBW or actual body weight
estimate for aminoglycosides, in *the lesser of IBW or actual body weight
liters), Furthermore:
K = elimination rate constant (popula-
tion estimate for aminoglycosides, estimated K = 0.00293 (CrCl) + 0.014
in reciprocal hours),
= 0.00293 (45) + 0.014
t = infusion time (hours), and
= 0.146 hr−1
τ = desired or most appropriate dosing
interval (hours). estimated T½ = 0.693/K (See Equation 3-3.)
To solve this equation, we must: = 0.693/0.146 hr−1
1. Determine creatinine clearance (CrCl).
= 4.75 hours
2. Insert population estimates for V and K.
3. Choose a desired Cpeak, based on clinical and estimated V = 0.24 L/kg (weight)
microbiologic data.
4. Determine our infusion time in hours.
the lesser of IBW
5. Calculate an appropriate dosing interval or actual weight
(τ), as shown below.
6. Determine K0 (maintenance dose, in
milligrams per hour).
= (0.24)(52)
= 12.48 L, rounded to 12.5 L
To calculate an initial maintenance dose and
dosing interval, we use the population estimates of Cpeak (desired) = 6 mg/L
K and V calculated from Equations 12-1 and 12-2:
Ctrough (desired) = 1 mg/L
K = 0.00293 hr –1 × CrCl (in mL/minute) + 0.014
V = 0.24 L/kg × IBW Estimation of Best Dosing Interval (τ)

For patient SG, we can calculate the IBW: The choice of dosing interval influences the Cpeak and
Ctrough eventually obtained as well as the magnitude
IBW = 45.5 + 2.3 kg per inch over 60 inches of the fluctuations in Cpeak and Ctrough. The equation
below is used to determine the most appropriate
= 45.5 + 2.3(8) = 63.9 kg dosing interval (τ) that will yield the desired Cpeak

186
and Ctrough. As can be seen, this calculation is driven Step 5. Further rearrange Step 4 by considering
by the patient’s elimination rate constant (K) and the rule of logarithms:
the Cpeak and Ctrough desired:
log a – log b = log (a/b )
1 Therefore:
=
12-4 τ (In C trough (desired) − In C peak (desired) ) + t
−K 1  trough 
=τ In   +t
−K  peak 
where t is the duration of the infusion in hours. This
equation can be used to evaluate several different
The equation in either Steps 4 or 5 can be used to
Cpeak and Ctrough combinations to find an appropriate
calculate the dosing interval (τ). You may find the
dosing interval.
equation in Step 5 easier to enter into a hand-held
calculator.
Derivation of Above Dosing Interval Calculating the dosing interval with both equa-
Equation tions (Steps 4 and 5) will serve as an added arithmetic
check, because both methods should give the same
The above dosing interval equation comes from
answer.
a simple rearrangement of the equation for K as
shown below.
For a concentration versus time curve (following Clinical Correlate
first-order elimination), the terminal slope equals –K
and: Some people guess at the best dosing interval
and gloat when they get it correct without
Y2 − Y1 In C trough − In C peak calculating it. However, if they guess wrong, they
=
−K =
X2 − X1 τ −t will not get the desired trough concentration, and
they will have to guess again.
This equation can be rearranged to easily calcu-
late τ as shown below:
Step 1. Rearrange Equation 12-4 and then solve
for : Calculation of Best Dosing Interval (τ)
( −K )(=
τ − t ) In C trough − In C peak for Patient SG
For patient SG, the calculation of the dosing interval
Step 2. Divide both sides by –K: (τ, in hours) proceeds as follows if we want a Cpeak
of 6 mg/L and a Ctrough of 1 mg/L:
In C trough − In C peak
τ −t =
−K 1
=τ (In C trough (desired) − C peak (desired) ) + t
Step 3. Transpose t to the right-hand side of the −K
equation:
1
In C trough − In C peak = (In 1 mg/L − In 6 mg/L) + 1 hr
=τ +t −0.146
−K
=
( −6.849)(0 − 1.79) + 1 hr
Step 4. Rearrange:
=
( −6.849)( − 1.79) + 1 hr
1
=τ (In C trough − In C peak ) + t = 13.26 hours
−K

187
At this point, we know that the best dosing which would be given over 1 hour for an infusion
interval to obtain our desired Cpeak and Ctrough concen- rate (K0) of 66.7 mg/hour.
trations is 13.26 hours. In practice, this number Therefore, patient SG should receive 66.7 mg
would be rounded down to 12 hours. every 12 hours. In practice, the dose would be
rounded to 70 mg every 12 hours. This amount is
Calculation of Maintenance Dose for SG the initial estimated maintenance dose that would
be given until Cpeak and Ctrough results are obtained.
Next, we must determine the maintenance dose to Because we rounded the dose up from the calcu-
be given at our desired interval of 12 hours. Note lated value of 66.7 to 70 mg, the actual Cpeak(steady state)
that in this example we are calculating the main- is slightly higher than our desired value of 6 mg/L.
tenance dose first and will use it to calculate the This actual Cpeak(steady state) can be determined via a
proper loading dose. simple ratio:
Once K and V have been estimated, the desired Cpeak
and Ctrough concentrations determined, and τ calcu- actual (rounded) dose
desired level × =
actual peak
lated, these values can be substituted in our general calculated dose
equation and solved for K0 (maintenance dose):
For this patient, the actual Cpeak(steady state) is
K 0 (1− e −Kt )
C peak (steady state) = calculated as follows:
VK (1− e −K τ )
70 mg
(See Equation 5-1.) 6 mg/L × =
6.3 mg/L
66.7 mg
where:
Cpeak(steady state) =
desired peak drug concentration at Problem 1B. Calculate the Ctrough concentration
steady state (milligrams per liter),
expected from the dose of 70 mg every 12 hours for
K0 = drug infusion rate (also maintenance patient SG.
dose you are trying to calculate, in
The answer to this problem requires the use of
milligrams per hour),
another equation:
V = volume of distribution (population esti-
mate for aminoglycosides, in liters), C = C0e–Kt (See Equation 3-2.)
K
= elimination rate constant (popula-
tion estimate for aminoglycosides, where:
in reciprocal hours), C = drug concentration at time t,
t = duration of infusion (hours), and C0 = drug concentration at time zero or some
τ = desired or most appropriate dosing earlier time, and
interval (hours). e–Kt = fraction of original or previous concen-
Then: tration remaining at time t.
−1 This general equation can be rewritten to show

K 0 (1− e −0.146 hr (1 hr)
) the calculation of patient SG’s Ctrough concentration
6 mg/L = −1
0.146 hr −1(12.5)(1− e −0.146 hr (12 hr)
) after she receives her dose of 70 mg every 12 hours:
K 0 (0.1358) C trough (steady state) = C peak (steady state) e − Kt ′

=
(1.825)(0.8265)
= K 0 (0.09) (See Equation 3-2.)
where t′ = τ – time of infusion (t), or the change in
66.7 mg = K 0 time from the first concentration to the second.

188
In this case, we are saying that Ctrough equals

Cpeak multiplied by the fraction of Cpeak remaining (as CASE 2
described by e–Kt′ ) after elimination has occurred for t
In Case 1, we showed how to calculate an
hours (i.e., 11 hours). As shown in Figure 12-3, because
appropriate maintenance dose and dosing
the peak concentration occurs at the end of the 1-hour interval. For this case, we use the data presented
infusion, t′ in this equation is always τ (dosing interval) in Case 1 and continue treating patient SG.
minus t (duration of infusion).
You should understand how t, τ, and t′ differ.
In patient SG’s case, the estimated Ctrough would be: Problem 2A. Calculate an appropriate loading dose
to approximate a plasma concentration of 6 mg/L.
C trough (steady state) = C peak (steady state) e −Kt ′
There are several methods to calculate a loading
−1 dose, and two are presented. Because one method
= (6.3 mg/L)e − (0.146 hr )(12 hr −1 hr)
requires estimation of the maintenance dose first,
= (6.3 mg/L)e − (0.146 hr
−1
)(11 hr) the loading dose is determined after the mainte-
nance dose and dosing interval are calculated. In
= (6.3 mg/L)e −1.606 clinical practice, the loading and maintenance doses
would be calculated at the same time.
= (6.3 mg/L)(0.200) Like all drugs given at the same maintenance
= 1.26 mg/L; round to 1.3 mg/L dose via intermittent administration, amino-
glycosides will not reach the desired steady-state
This calculation tells us that 70 mg every 12 hours therapeutic concentration for three (87.5% of
will give an estimated Cpeak of 6.3 mg/L and an esti- steady state) to five (96.9% of steady state) drug
mated Ctrough of 1.3 mg/L. Remember that we actually half-lives. Therefore, subtherapeutic concentrations
picked a desired Ctrough of 1 mg/L, but we also short- may exist for 1–2 days of therapy in patients with
ened the desired dosing interval from 13.2 to 12 longer half-lives. Figure 12-4A shows a plasma
hours, making the estimated Ctrough higher than the drug concentration versus time simulation for an
initial desired Ctrough. If, based on clinical judgment, a aminoglycoside given at the same dose six times.
lower Ctrough is desired, the dose can be recalculated In Lesson 3, we learned that a patient’s drug
with a longer dosing interval, such as 16 hours. half-life is dependent on the elimination rate
In patient SG’s case, if a Ctrough close to 2 mg/L had
been attained, it would have been because we chose a
dosing interval shorter than that recommended by our
dosing interval calculation. Therefore, we would need
to reexamine the rounding of our dosing interval and
would probably round it up from 13.2 to 16 or 18 hours.
FIGURE 12-4.
A. Drug accumulation to steady state without a loading
FIGURE 12-3. dose. B. Concentration versus time simulation for the same
Hours of elimination after drug peaks. aminoglycoside dose preceded by a loading dose.

189
constant (K). Mathematically, T½ equals 0.693/K

and, vice versa, K equals 0.693/T½. Thus, time to Clinical Correlate
reach steady state is dependent on the elimination
rate constant (K) for a given patient. In Lesson 5, we calculated the loading dose of a
drug administered by intravenous push,
X0 = C0(desired)V. This equation assumes a rapid
Clinical Correlate infusion of a drug. Because aminoglycosides are
infused over 30 minutes to an hour, the equation
This is an interesting and conflicting concept. below must be used to calculate a loading dose
Reaching a steady-state drug concentration to account for the amount of drug eliminated over
depends only on the patient’s elimination rate the infusion period. The term (1 – e–Kt ) represents
(K). Steady state occurs in three to five drug the fraction remaining after (t ), the time of
half-lives. The time to steady state cannot be infusion.
shortened with a loading dose infusion. However,
a loading dose infusion can produce a plasma
drug concentration approximately equal to the This equation can be rearranged to isolate K0 on
eventual steady-state concentration (see Figure one side of the equation:
12-4B). That is, a loading dose infusion will
quickly bring the patient’s drug concentration C peak (steady state) (VK )
K0 =
to a concentration that approximates the (1− e − Kt )
concentration at steady state. In addition, any
time the dose or dosing interval is changed, it will
Patient SG’s loading dose infusion can then be
take another three to five half-lives to reach a
calculated:
new steady-state concentration. After changing a
dosing regimen, remember to allow enough time 6 mg/L (desired peak) can be used
to reach a new steady-state concentration before instead of actual peak (6.3 mg/L)
repeating plasma drug concentrations.
(6 mg/L)(12.5 L)(0.146 hr −1 )
Calculating a Loading Dose loading dose = −1
(1− e − (0.146 hr )(1 hr )
)
The loading dose infusion can be calculated from the
formula for an intermittent infusion not at steady 10.95
=
state as shown in Lesson 5: 0.1358
K0 = 80.6 mg
C peak (steady=
state) (1− e − Kt )
VK
By this method, the loading dose infusion can be
where: determined before the maintenance dose is calcu-
Cpeak(steady state) = desired peak drug concentration at lated, but only with a complicated equation.
steady state, Another, easier loading dose formula that
K0 = loading dose (in mg) to be infused ÷ requires calculation of the maintenance dose first is
duration of infusion (in hours), shown below:
estimate, in liters), maintenance dose
K = elimination rate constant
(population estimate, in reciprocal
hours), and K0
12-5 loading dose =
t = duration of infusion (1 hour). (1− e −K τ )

190
where: For patient SG, the loading dose should be:
K0 = estimated maintenance dose, K0

loading dose =
1/(1 – e ) = accumulation factor at steady
–Kτ
1− e −K τ
state (see Equation 4-2), and
66.7 mg/hr
τ = dosing interval at which estimated 66.7 mg (actual dose = −1
maintenance dose is given.

calculated) can be 1− e − (0.146 hr )(12 hr )
used instead of 70 mg
With this loading dose formula you are, in essence, 66.7 mg/hr
(actual dose given) =
multiplying the desired maintenance dose by a 0.826
factor (the accumulation factor) representing the
= 80.8 mg
sum of the fraction of doses that have accumulated
at steady state. This factor describes how much the Both loading dose formulas will give approxi-
concentration will be increased at steady state. mately the same number. However, some prefer
These two formulas are derivations of each the loading dose equation that requires the main-
other, as shown below. Begin with our general tenance dose to be calculated first because it
formula and rearrange it to solve for K0: is simple. Patient SG should receive a loading
dose of 80 mg (rounded) followed by a mainte-
K 0 (1− e −Kt ) nance dose of 70 mg every 12 hours. The 80-mg
C peak(steady state) =
VK (1− e −K τ ) loading dose should give an approximate Cp of
6 mg/L. Based on the estimated parameters, steady
C peak(steady state) (VK ) (1− e −K τ ) state should be attained in three to five half-lives
K0 =
(1− e −Kt ) (3 × 4.75 = 14.25 hours; 5 × 4.75 = 23.75 hours).
(See Equation 3-3.)
This corresponds to the circled portion of the

next equation. CASE 3
To continue with patient SG from Cases 1 and
The first numerator/denominator combination in
2, blood was drawn for drug concentration
the above equation is also found in the equation for assessment around the fourth dose (i.e.,
the loading dose: approximately 5 minutes before dose was due
C peak(steady state) (VK ) and immediately after the 1-hour dose infusion).
loading dose = Cpeak and Ctrough were determined as follows:
(1− e −Kt )
• 7:55 am: Ctrough was 0.4 mg/L (before fourth dose).
Therefore, the right-hand term of this loading dose • 8–9 am: 70-mg dose was infused over 1 hour.
equation can be substituted into the general equa- • 9 am: Cpeak was 4.6 mg/L (after fourth dose).
tion for K0 (Step 1 below) and then rearranged Although a peak and trough was ordered, a trough
(Step 2 below) to then yield our other loading dose and peak was actually drawn. In this example, the
formula: trough level was taken just before the fourth dose
was given, and the peak level was obtained just
Step 1: K 0 = (loading dose)(1− e −K τ ) after the fourth dose was given. This procedure is
normal and appropriate if the concentrations are at
K0 steady state. Figure 12-5 illustrates that, at steady
Step 2: (loading dose) =
(1− e −K τ ) state, Ctrough from a trough and peak is equal to the

191
Problem 3A. Are patient SG’s concentrations of

4.6 mg/L (peak) and 0.4 mg/L (trough) at steady state?
Patient SG’s concentrations were determined around
the fourth dose, meaning 36 hours after her first
dose. To determine whether these serum values are
steady-state concentrations, we must use Cpeak and
Ctrough to calculate SG’s actual K and T½. These calcu-
FIGURE 12-5.
lations are done in Problem 3C, and the results are
Peak and trough concentrations at steady state.
0.222 hr –1 for K and 3.12 hours for T½. Five half-lives
Ctrough from a peak and trough because all Ctrough would equal 15.6 hours (3.12 × 5), which is less than
and all Cpeak values are the same. We know that if the 36 hours elapsed. Therefore, these concentra-
we measured a Ctrough after the Cpeak, it would equal tions are considered to be at steady state. If the drug
the Ctrough before the Cpeak. This is not true before is not at steady state, the pre-dose Ctrough would be less
steady state is reached. In this case, therefore, than the post-dose Ctrough and would overestimate K.
when a peak and trough is ordered, the literal
interpretation would be: Problem 3B. How can you determine when to
order drug concentration samples so they are likely
1. Give the infusion from 8 to 9 am.
to be at steady state?
2. Draw a sample to determine Cpeak at You want to determine Cpeak and Ctrough after the patient
approximately 9 am.
is at steady state. Therefore, you must draw blood
3. Wait until the end of SG’s 12-hour dosing samples three to five drug half-lives after the first
interval (approximately 7:55 pm) to draw a dose. You must estimate the patient’s K and T½ using
sample to determine Ctrough. population estimates as in Case 1 and then multiply
In practice, this method is too cumbersome the T½ by three to five. As shown in Lesson 4, after
for pharmacy, nursing, and laboratory staff, so three half-lives, concentrations are 87.5% of steady
usually a trough and peak is drawn if steady state state, whereas after five half-lives, they are 96.9%
has been attained. of steady state. Use judgment when choosing three,
four, or five half-lives to calculate time to steady state.
It is recommended that Cpeak be measured either
at the end of a 1-hour infusion, 30 minutes after Plasma concentration sampling should be scheduled
the end of a 30-minute infusion, or 1 hour after an to follow the dose that achieves steady state.
intramuscular injection. Infusing aminoglycosides For patient SG, the estimated K and T ½ (from
over 1 hour allows simpler pharmacokinetic Case 1, Problem 1A) were 0.146 hr–1 and 4.75
calculations in that the duration of infusion (t ) is hours, respectively. Therefore, steady state would
1 hour and the infusion rate (K0) is simply the dose be reached in 23.75 (5 × 4.75) hours. You could then
given. Remember that K0 is expressed as milligrams schedule Cpeak and Ctrough determinations at the next
per hour. So, if the drug is infused over 30 minutes dose after 24 hours have elapsed.
(0.5 hour), then K0 = dose (mg)/0.5 (hour).
Clinical Correlate
Clinical Correlate
By calculating the patient’s actual elimination rate
An actual peak and trough, as opposed to a (K ) and volume of distribution (V ), pharmacists
trough and peak, is sometimes ordered on the can more accurately predict patient-specific
first (i.e., not at steady state) dose of a drug to pharmacokinetic data, thereby optimizing patient
estimate volume of distribution and K in a patient care. Once patient-specific parameters are known,
whose drug half-life is quite long. (See Lesson it is important not to continue to use population
13, Vancomycin, Case Two, for a case on this.) estimates to adjust dosages or dosage intervals.

192
Problem 3C. Adjust patient SG’s dosing regimen, Either form of this equation may be used to
based on Cpeak and Ctrough concentrations, to obtain calculate K, as follows:
the desired Cpeak of 6 mg/L and Ctrough of 1 mg/L.
Adjustment of patient SG’s dose involves using the In 0.4 mg/L − In 4.6 mg/L
K = −
measured drug concentrations to calculate an actual 12 hr − 1 hr
K and V and then substituting these new values for −0.916 − 1.52
our initial estimates of K and V, in Equations 3-2, = −
5-1, and 12-4, used in Case 1. The formula for K
11
below comes from a rearrangement of the general −2.44
equation used to calculate the slope of the natural =
11
log of plasma drug concentration versus time line
as described in Case 1. Remember that because = 0.222
concentration decreases with time, the slope (and
hence, –K) is a negative number. Therefore, K = 0.222 hr –1, compared to 0.146 hr –1,
which was our estimate, or:
Calculation of SG’s Actual Elimination
 0.4 mg/L 
Rate (K) In 
 4.6 mg/L 
To calculate K, the equation is: K = −
12 hr − 1 hr
In C trough − In C peak In (0.0869)
K = − = −
τ −t 11 hr
(See Equation 3-1.)
−2.442
where: = −
11 hr
K = elimination rate constant (in reciprocal
hours), = 0.222 hr −1
Ctrough = measured trough concentration (0.4 mg/L),
Patient SG’s actual K of 0.222 hr–1 is greater than
Cpeak = measured peak concentration the estimated value of 0.146 hr –1, so her elimination
(4.6 mg/L), probably was greater than estimated. Her actual
τ = dosing interval at the time concentra- drug half-life (T½ ) is 3.12 hours, shorter than the
tions are obtained (12 hours), and population-estimated T½ of 4.75 hours:
t = duration of infusion (1 hour).
T½ = 0.693/K (See Equation 3-3.)
Again, remembering a rule of logarithms:
ln a – ln b = ln (a/b ) = 0.693/0.222 hr –1
we can simplify this equation for hand-held calculators: = 3.12 hours
C 
In  trough  The formula for K above can also be used to
C 
K = − 
peak  calculate the slope, –K, for any two points on the
τ −t natural log of plasma drug concentration versus
time line. For instance, suppose that instead of a
Cpeak, patient SG had a concentration measured
This equation version is more calculator-friendly at 11 am (2 hours after Cpeak). This concentration

193
was 2.95 mg/L. You can still calculate her K value as where:
follows (Figure 12-6): t = the duration of the infusion and
In 0.4 mg/L − In 2.95 mg/L τ = dosing interval at the time
K = − concentrations are obtained.
12 hr − 1 hr − 2 hr
−0.916 − 1.0818 In this case, some of the variables substituted in
= − this equation are different from those used when
9
initially estimating a dose. The changes from Case 1
−1.9978 are shown here in bold type:
= −
9 Cpeak(steady state) = Cpeak measured at steady state,
= 0.222 K0 = maintenance dose infused
Note that this K value is the same as the one at time Cpeak and Ctrough were
calculated with the measured Cpeak and Ctrough concen- measured,
trations. V = patient’s actual volume of
distribution that you are trying
Calculation of SG’s Actual Volume of to determine based on Cpeak and
Distribution (V ) Ctrough values,
Patient SG’s actual volume of distribution (V) is calcu- K = elimination rate constant calculated
lated with the equation from Case 1. Use the actual from patient’s Cpeak and Ctrough values,
Cpeak and Ctrough values, dose, and dosing interval. t = duration of infusion (hours), and
K 0 (1− e −Kt ) τ = patient’s dosing interval at time
C peak(steady state) = Cpeak and Ctrough were measured.
VK (1− e −K τ )
In patient SG’s case, she received a maintenance
This equation can also be rearranged to isolate V dose of 70 mg every 12 hours, with subsequent
on one side of the equation if the reader so prefers, Cpeak and Ctrough concentrations of 4.6 and 0.4 mg/L,
although it is not normally necessary. respectively. Her K value from these concentrations
was 0.222 hr –1.
K 0 (1− e −Kt )
V = If we substitute these values into the previous
C peak(steady state)K (1− e −K τ )
equation, we can solve for patient SG’s actual V:
(See Equation 5-1.) −1

(70 mg/hr)(1− e − (0.222 hr )(1 hr)
)
4.6 mg/L = −1
(V )(0.222 hr −1 )(1− e − (0.222 hr )(12 hr)
)
(315)(0.199)
4.6 mg/L =
(V )(0.930)
V = 14.7 L
compared to 12.5 L that we estimated.

Patient SG’s V value of 14.7 L (which equals
0.28 L/kg IBW) is larger than estimated and would
FIGURE 12-6. tend to make her actual Cpeak and Ctrough lower than
K from any two points. estimated.

194
Now that we have calculated the patient’s actual where:

K and V, we need to recalculate the dose and dosing Cpeak(steady state) = desired steady-state Cpeak (6 mg/L);
interval. We must first calculate the new dosing
K0 = drug infusion rate (also adjusted
interval (τ) and then calculate the new dose
maintenance dose you are trying to
1 calculate, in milligrams per hour);
=τ (In C trough (desired) − In C peak (desired) ) + t
−K V = actual volume of distribution deter-
where t is the duration of infusion in hours and K is mined from patient’s measured Cpeak
the actual elimination rate calculated from patient’s and Ctrough values, in liters;
peak and trough values, and not the estimated value K = actual elimination rate constant
of 0.146 hr–1. It is important to note that when we calculated from patient’s measured
calculate our new dosing interval, the values we Cpeak and Ctrough values, in reciprocal
insert into the equation are our desired levels and hours;
not the levels reported by the lab on the dose of t = infusion time, in hours; and
70 mg every 12 hours. Then:
τ = adjusted dosing interval rounded to
1 a practical number.
τ (In 1 mg/L − In 6 mg/L) + 1 hr
−0.222 hr −1
Strikeovers in the following equation show differ-
=
( −4.5)(0 − 1.79) + 1 hr ences from initial estimated maintenance dose
calculations (see Case 1):
=
( −4.5)( −1.79) + 1 hr
= 9.06 hr K0
6 mg/L =
( 12.5 14.7 L)( 0.146 0.222 hr −1 )
(See Equation 12-4.)
 (1− e − 0.146 0.222 hr (1 hr) ) 
−1
This adjusted τ should be compared with
our initial τ estimate of 13.26 hours from Case 1,  − 0.146 0.222 hr −1 ( 12 8 hr)

 (1− e ) 
Problem 1A. Because our real τ is shorter than
previously estimated, Cpeak and Ctrough values less K 0 0.199
than those predicted also would be expected. In 6 mg/L =
(3.263) 0.831
other words, we initially administered a dose every
12 hours when, in actuality, the patient needed a = K 0 (0.0732)
dose every 9 hours.
This calculated dosing interval of 9 hours may K 0 = 81.9, rounded to 80 mg
be rounded down to 8 hours for ease in scheduling.
If 81.9 mg gives a peak of 6 mg/L, then our rounded
Problem 3D. How is patient SG’s adjusted mainte- dose of 80 mg will give a peak of 5.86 mg/L.
nance dose now calculated?
Once again, we shall use the general equation from Problem 3E. If we give 80 mg every 8 hours, what
Case 1 and solve for K0. This time, we shall replace will be our steady-state Ctrough?
the estimates of K and V with the calculated (actual) If we give 81.9 mg exactly every 9 hours, our Ctrough
values and use the adjusted τ value of 8 hours: would be precisely as desired: 1 mg/L. But because
we rounded our dosing interval and adjusted the
K 0 (1− e −K t )
C peak(steady state) = maintenance dose down to practical numbers, we
VK (1− e −K τ ) must calculate the steady-state Ctrough that will result.
(See Equation 5-1.) Our roundings could make our Ctrough too high.

195
This Ctrough calculation is performed similarly to new maintenance dose (K0) = 60 mg every
the one in Case 1, Problem 1B (strikeovers show 12 hours, and
differences from our initial calculations): new trough concentration = 1.1 mg/L.
Ctrough(steady state) = Cpeak(steady state)e –Kt ′
(See Equation 3-2.) Problem 4A. Because patient SG’s Ctrough on 80 mg

every 8 hours is now too high (3.2 mg/L), how long
= 5.86 mg/L e − ( 0.146 )(0.222 hr −1 )( τ −1 hour)  would you wait before beginning the new dose of
  60 mg every 12 hours?
= 5.86 mg/L e − ( 0.146 )(0.222)( 11 7)  Before switching, you must wait for the patient’s
Ctrough to decrease to approximately 1 mg/L. There-
(
= 5.86 e −1.554 ) fore, the dose should be held for some time before
you begin a new lower dose. The formula for calcu-
lating the number of hours to hold the dose is:
= 5.86 (0.211)
= 1.24 mg/L Ctrough (steady state)(desired) = Ctrough (steady state) e–Kt ′
(See Equation 3-2.)

So, in this case, a dose of 80 mg every 8 hours
will give a steady-state Ctrough of 1.24 mg/L, still where t′ is the amount of time to hold the dose after
well below the usual maximum acceptable trough the end of the 8-hour dosing interval.
concentration of 2 mg/L.
This formula is an application of the general
formula described in Case 1:
CASE 4 C = C0e–Kt (See Equation 3-2.)

Four days later, another set of peak and trough which means:
concentrations are obtained. Patient SG has
been receiving 80 mg every 8 hours. However, concentration at a time = previous concentration
her renal function has declined, as seen by an × fraction of dose remaining
increase in serum creatinine from 1.02 mg/dL
at baseline to 1.71 mg/dL today. Cpeak and Ctrough In patient SG’s case:
were determined as follows: –1
1 mg/L = (3.2 mg/L) e–0.151 hr (t ′)
• 7:55 am: Ctrough was 3.2 mg/L.

• 8–9 am: an 80-mg dose was infused over 0.312 mg/L = e–0.151(t ′)
1 hour.
Next, take the natural log of both sides:
• 9:00 am: Cpeak was 9.2 mg/L.
A new adjusted K, τ, V, and maintenance dose ln 0.312 = ln (e–0.151(t ′))
(K0) were calculated using the methods described
in Case 3. These values are shown below; see if –1.163 = –0.151(t ′)
you obtain the same numbers:
7.70 hours = t ′
new K = 0.151 hr –1,
new T ½ = 4.6 hours, Thus, we should hold patient SG’s dose for an
additional 7.7 (round to 8) hours after the next
new V = 11.1 L, Ctrough time and then begin her new dose. The
new τ = 12.6 (rounded to 12 hours), next Ctrough time for this patient would be 3:45 pm,

196
7.75 hours after her last dose (8 am). The Ctrough at at 7:55 am. Therefore, her next Ctrough will occur at
this time, at steady state, would also be expected to approximately 3:45 pm (shortly before the next
be approximately 3.2 mg/L. We would need to hold scheduled dose). We need to hold this dose for an
the regularly scheduled 4 pm dose for 8 hours, until additional 9.2 (round to 9) hours and then begin
12 midnight, at which time we would then begin her our new regimen of 60 mg every 12 hours.
new dose of 60 mg every 12 hours.
The calculation of the time to hold a dose can Extended-Interval
be illustrated (Figure 12-7) by plotting patient
Aminoglycoside Dosing
SG’s Cpeak and Ctrough values on semilog graph paper
and then extending the line connecting them until An alternative method to conventional dosing of
it reaches our desired Ctrough of 1 mg/L. You can aminoglycosides is extended-interval dosing—admin-
then count the hours needed to reach this 1-mg/L istering large doses over extended intervals (24, 36,
concentration and hold the dose accordingly. or 48 hours) based on the patient’s renal function.
Another, and often more practical, way to The theory behind this approach is that administering
estimate the time to hold a patient’s dose is by large doses produces higher peak serum concentra-
examining the half-life. By definition, the drug tions than achieved with conventional dosing and,
concentration decreases by one-half over each thus, increases the peak serum concentration to
half-life. In the following paragraph, we can bacterial minimum inhibitory concentration (MIC)
then estimate how many drug half-lives to wait ratio (Peak/MIC).
for the concentration to approach our desired Additionally, administering drug at an extended
1 mg/L. interval creates an aminoglycoside-free period
For patient SG (trough of 3.2 mg/L and T½ of that reduces accumulation of aminoglycoside in
4.6 hours), the concentration will drop to 1.6 mg/L tissues such as the inner ear and kidney, resulting
(half of 3.2) in one half-life of 4.6 hours and then in decreased drug-related toxicity. It is known that
drop to 0.8 mg/L (half of 1.6) in another 4.6 hours. uptake of aminoglycosides by tissues is a saturable
Therefore, we can hold patient SG’s doses for process. Administering smaller doses at a more
approximately two half-lives (4.6 × 2 = 9.2 hours) frequent interval does not saturate this process and
before beginning our new dose. ultimately leads to higher tissue concentrations than
In patient SG’s case, another dose was given those achieved with extended interval dosing. Thus,
from 8 to 9 am, after the Ctrough of 3.2 was obtained this latter dosing method may actually result in less
toxicity to the patient. This drug-free interval may also
decrease the development of adaptive resistance.
Several characteristics of aminoglycosides as
a class enable these drugs to be administered by
the extended-interval method. Aminoglycosides
demonstrate concentration-dependent bactericidal
action such that as the concentration of the drug in
the serum increases, the rate and extent of bacte-
rial killing increases. Because of this property, it
is suggested that the optimal serum peak amino-
glycoside concentration to bacterial MIC ratio is
> 10:1. It appears that bactericidal activity occurs
in a biphasic fashion; initially, bacteria are killed
at a very rapid rate in a concentration-dependent
manner. After a time frame of approximately two
FIGURE 12-7. hours, the rate of bacterial killing declines, which
Calculation of the time to hold a dose. may be due to bacterial adaptive resistance.

197
Aminoglycosides also exhibit a long post- [Equation 9-3: AdjBW = IBW + (0.4 × (TBW–IBW).]
antibiotic effect (PAE) of approximately 4 to 6 hours. Goal serum peak concentrations from these doses
Post-antibiotic effect is defined as the amount of are 16–24 mcg/mL and 56–64 mcg/mL for genta-
time that drug concentration falls below the MIC micin/tobramycin and amikacin, respectively.
before regrowth of the bacteria resumes.6,7,8 PAE Expected trough levels for all three drugs are
is generally thought to increase with high peak < 1 mcg/mL. Table 12-3 lists recommended doses
concentrations of aminoglycosides. in patients with a reduced CrCl.10
A third characteristic of aminoglycosides that
support extended-interval dosing is a decrease in Method 2
the development of adaptive resistance. Adaptive A second method of extended-interval aminoglycoside
resistance results in decreased efficacy of an anti- dosing consists of administering a dose of 7 mg/kg of
biotic and the emergence of resistant organisms. gentamicin or tobramycin at an interval based on
It is a reversible process if a sufficient drug-free the patient’s CrCl > 60 mL/min, every 24 hours;
interval between doses is allowed.9 40 to 59 mL/min, every 36 hours; 20 to 39 mL/min,
Situations in which extended interval amino- every 48 hours. For patients with a CrCl < 20 mL/min,
glycoside dosing probably should NOT be used it is recommended to monitor serial serum concen-
include pregnancy, ascites or significant third trations and administer a subsequent dose once the
spacing, hemodynamic instability, unstable renal serum level is < 1 mcg/mL. With the first dose, a 6- to
function (CrCl < 20 mL/min), and burns > 20%. 14-hour post infusion serum level is measured and
Numerous methods have been proposed for plotted on the Hartford nomogram (Figure 12-8) to
extended-interval aminoglycoside dosing and moni- determine if the dosage interval should be altered
toring. Several of these methods are presented here. for future doses.8
Method 1 Method 3
The 2013 Sanford Guide to Antimicrobial Therapy 5,10
The American Society of Health-System Pharmacists
recommends that for gentamicin and tobramycin, (ASHP) recommends initial doses of 7 mg/kg genta-
the dose in patients with a CrCl > 80 mL/min is micin or tobramycin and 15 mg/kg amikacin.11 This
5.1 mg/kg (7 mg/kg for seriously ill patients) dose is based on the patient’s IBW unless their actual
every 24 hours, and for amikacin, 15 mg/kg weight is 20% over the ideal weight. In this situation,
every 24 hours. A patient’s IBW is used in these the adjBW should be used. Suggested dosage intervals
calculations unless actual weight exceeds ideal are as follows: for CrCl > 60 mL/min, every 24 hours;
weight by ≥ 30%. In this case, an AdjBW is used for CrCl 40 to 60 mL/min, a single dose; and for CrCl
TABLE 12-3. Recommended Extended-Interval Dosing in Patients with Declining Renal Function10
Creatinine Clearance Gentamicin/Tobramycin Amikacin
80–60 mL/min 4 mg/kg every 24 hours 12 mg/kg every 24 hours

60–40 mL/min 3.5 mg/kg every 24 hours 7.5 mg/kg every 24 hours
40–30 mL/min 2.5 mg/kg every 24 hours 4 mg/kg every 24 hours
30–20 mL/min 4 mg/kg every 48 hours 7.5 mg/kg every 48 hours
20–10 mL/min 3 mg/kg every 48 hours 4 mg/kg every 48 hours
< 10 mL/min 2 mg/kg every 72 hours* 3 mg/kg every 72 hours*
*For patients receiving dialysis, these doses should be administered after dialysis.
Source: The Sanford Guide to Antimicrobial Therapy, 43rd ed. Sperryville, VA: Antimicrobial Therapy, Inc.; 2013. p. 205.

198
is based on a dose of 7 mg/kg, if a smaller dose is

administered, the measured serum level should be
multiplied by a factor equal to 7 divided by the dose
given. For example, if a patient is being treated with
the 5 mg/kg dose, 7 divided by 5 equals 1.4, which
is then multiplied by the measured serum concen-
tration. This product is plotted on the Hartford
nomogram. For amikacin serum concentrations,
plot one-half the measured concentration on the
nomogram. In situations in which the measured, or
adjusted, value falls on one of the three lines, choose
the longer interval for administering future doses.
Body weight used for these dosage calculations is
FIGURE 12-8.
Hartford once-daily aminoglycoside adjustment nomogram the patient’s actual weight. In cases in which the
(gentamicin or tobramycin, 7 mg/kg). patient’s actual weight is > 20% over their IBW,
Source: Reproduced with permission from Nicolau DP, the AdjBW should be calculated and used in deter-
Freeman CD, Belliveau PP, et al. Experience with a once-daily mining the dose. Initial dosing intervals are as
aminoglycoside program administered to 2,184 adult patients. follows: for CrCl > 60 mL/min, every 24 hours; for
Antimicrob Agents Chemother 1995;39(3):650–5. CrCl 40–59 mL/min, every 36 hours; and for CrCl
20–39 mL/min, every 48 hours. If estimated CrCl is
< 20 mL/min, do not use extended interval amino-
< 40 mL/min, a single dose. Obtain a random serum
glycoside dosing.
concentration 6 to 12 hours after the first dose
and plot the results on an established nomogram
such as the Hartford nomogram to determine the Method 5
appropriate dosage interval. For patients receiving An alternative method to those described above
amikacin, divide the serum concentration by 2 and consists of using traditional or conventional dosing
plot this value on the nomogram. The serum concen- equations for calculating extended-interval doses.
trations may also be interpreted in light of MIC data Goal serum peak and trough concentrations utilized
if available. Serum levels should be monitored peri- should be 20 to 30 mcg/mL and < 1 mcg/mL, respec-
odically for patients receiving prolonged therapy tively, for gentamicin and tobramycin. A minimum
(> 7 to 10 days) and in patients with unstable renal dosing interval of 24 hours is selected; increases in this
function. Desired peak serum concentrations with value should be used in cases of declining renal func-
these regimens are a value of 10 to 12 times the MIC tion. The same equations for estimating K, V, dosing
of the infecting organism. Desired trough levels are interval, and calculating the maintenance dose as used
< 1 mcg/mL for gentamicin and tobramycin and in traditional dosing methods are then applied.
< 7 mcg/mL for amikacin.
Method 6
Method 4 A sixth method is one that may be used in the treat-
A modification of the ASHP method consists of ment of patients with cystic fibrosis.12 (pp.12–14) An
administering an initial gentamicin or tobramycin initial dose of 10 mg/kg tobramycin or 20 mg/kg
dose of 7 mg/kg, or 5 mg/kg for urinary track infec- amikacin is administered over a one-hour interval.
tions, or 15 mg/kg amikacin followed by measuring Dosing weight is as described for Method 4. Serum
a serum concentration 6 to 14 hours after the start levels are drawn 1 and 5 hours after the end of the
of the infusion.12 (pp. 7–8) This measured value is then infusion. From these two levels, one may calcu-
plotted on the Hartford nomogram, and the dosing late the patient’s elimination rate and half-life. For
interval is determined. Because this nomogram patients with a half-life between 2 and 4 hours,

199
administer the drug every 24 hours; for serum half- Problem 5B. According to the Sanford Guidelines,
lives > 4 to 6 hours, administer every 36 hours; in critically ill patients, a peak serum concentration
for half-lives > 6 to 8 hours, give every 48 hours. If should be drawn on the first dose of the amino-
calculated half-life is > 8 hours, convert the patient glycoside. The laboratory reports that JK has a
to traditional dosing. If the calculated half-life is peak serum tobramycin level of 23.2 mcg/mL. A serum
< 2 hours, consider changing the patient to tobramycin trough level is also drawn and is reported as 0.8 mcg/mL.
7 mg/kg every 12 hours, or amikacin 15 mg/kg every Based on these peak and trough levels, should JK’s
12 hours. For these latter two regimens, consider tobramycin dose be changed?
monitoring half-life from serum levels to verify that The Sanford Guidelines state that serum peak tobra-
the patient’s half-life remains < 2 hours. mycin levels should be between 16 and 24 mcg/mL
and trough levels should be < 1 mcg/mL. JK’s values
are within these ranges; therefore, no changes in his
CASE 5 dose are necessary at this time.
JK is a 53-year-old male chronic smoker

admitted to the hospital for exacerbation of his
chronic obstructive pulmonary disease (COPD). CASE 6
Chest x-ray demonstrates bilateral lower lobe
infiltration. Based on his past history, he is AM is a 32-year-old male with a soft tissue
suspected of having Pseudomonas aeruginosa infection in need of gentamicin therapy. He is 6'
pneumonia. He is 5' 10'' tall and weighs 170 lbs. 2'' tall and weighs 180 lbs. His current serum
His admission serum creatinine is 1.07 mg/dL. creatinine is 0.85 mg/dL.
Problem 5A. Calculate an extended-interval dose of Problem 6A. Calculate an extended interval genta-
tobramycin for this patient according to the Sanford micin dose for AM using the Hartford nomogram
Guidelines. method.
The first step in solving this problem is to determine Using the Cockcroft–Gault equation we can deter-
JK’s creatinine clearance. Using the Cockcroft–Gault mine that AM’s creatinine clearance is 123 mL/min.
equation, we can determine this to be 82 mL/min.
(140 − age)IBW (140 − age)IBW
CrCl = CrCl =
72 × SCr Note: If actual 72 × SCr
weight is less notice that a serum creatinine
(140 − 53) 72.73 (140 − 32) 82
=
than IBW, use = of < 1 (i.e., 0.85) is rounded
72 × 1.07 actual weight 72 × 1.00 up to 1.00 for calculation
purposes
= 82 mL/min = 123 mL/min
Because Pseudomonas aeruginosa pneumonia is a

serious infection, according to the Guidelines, we According to the Hartford nomogram method, he
should administer 7 mg/kg tobramycin times his should receive 7 mg/kg every 24 hours:
IBW of 72.73 kg every 24 hours. This results in a dose
of 509 mg (round off to 510 mg) every 24 hours. 7 mg/kg × 82 kg = 574 mg (round to 570 mg)

200
Problem 6B. Eleven hours after the beginning of τ = −1/K × (ln trough − ln peak) + t
AM’s therapy, a serum gentamicin level is drawn
and reported as 2.5 mcg/mL. Should AM’s genta- = −1/0.307 × (ln 0.9 − ln 25) + 1
micin therapy be adjusted?
= 12 hours, which we will round up to
Because an 11-hour post-dose level of 2.5 mcg/mL falls 24 hours
within the range for 24-hour dosing, AM’s gentamicin
therapy does not require adjustment at this time. Step 7. Calculate a maintenance dose to give the
desired peak and trough concentrations.
Problem 6C. Calculate an extended-interval genta-
micin dose for this patient using conventional or K 0 (1− e − kt )
C peak(steady state) =
traditional dosing equations. VK (1− e − k τ )
Step 1. Round the CrCl of 123 mL/min to
K 0 (1− e −0.307(1) )
100 mL/min. 25 mcg/mL =
19.7 L × 0.307 −1(1− e −0.307(24) )
Step 2. Estimate the patient’s elimination rate
constant. K 0 = 572 mg q 24 hours (round to 570 mg)
K = 0.00293 x CrCl + 0.014
= 0.00293 (100 mL/min) + 0.014
= 0.307 hr −1
CASE 7
A 72-year-old female, AC, is involved in a motor
Step 3. Estimate the patient’s volume of distribution. vehicle accident resulting in multiple injuries. She
V = 0.24 L/kg IBW undergoes surgical correction of her injuries and
postoperatively is admitted to the intensive care unit
= 0.24 L/kg x 82 kg requiring mechanical ventilation. On hospital day
4, her chest X-ray worsens and sputum cultures
= 19.7 L isolate E. coli sensitive to amikacin. Renal function
has remained stable with a serum creatinine of
Step 4. Choose a desired steady-state peak serum
0.67 mg/dL. She is 5' 4'' tall and weighs 130 lbs.
concentration.
• The recommended range is 20 to 30 mcg/mL.
• For illustration purposes, we will choose Problem 7A. This case represents an example of a
25 mcg/mL. hospital-acquired, or nosocomial, infection. Calcu-
late an appropriate dose of amikacin for AC using
Step 5. Choose a desired steady-state trough serum
the ASHP suggested method.
concentration.
The first step in solving this problem is to calculate
• The recommended value is < 1 mcg/mL.
her CrCl.
• For illustration purposes, we will choose
0.9 mcg/mL. (140 − age)IBW
CrCl = 0.85
Step 6. Calculate a desired dosing interval. 72 × SCr
• Because this method of dosing is extended notice that a serum
(140 − 72)54.6 creatinine of < 1 (i.e.,
interval, a minimum dosing interval of 24 = 0.85
hours should be used. This step is to deter- 72 × 1.00 0.67) is rounded up
to 1.00 for calculation
mine if a patient should receive a dose at a = 44 mL/min purposes
36- or 48-hour interval.

201
According to the ASHP guidelines, she should 4. Cockcroft DW, Gault MH. Prediction of creati-
receive 15 mg/kg as a single dose with a random nine clearance from serum creatinine. Nephron
serum level drawn 6 to 12 hours after this dose. 1976;16(1):31–41.
She receives an initial dose of 15 mg/kg x 54.6 kg = 5. Aminoglycoside once-daily and multiple daily
819 mg (rounded to 820 mg). dosing regimens. In: The Sanford Guide to Anti-
microbial Therapy, 43rd ed. Sperryville, VA:
Problem 7B. Ten hours after receiving her initial Antimicrobial Therapy, Inc.; 2013. p. 109.
dose, a random amikacin level is 14 mcg/mL. Calcu- 6. Maglio D, Nightingale CH, Nicolau DP, et al.
late an appropriate amikacin dosing interval for this Extended interval aminoglycoside dosing:
patient. from concept to clinic. Int J Antimicrob Agents
According to the ASHP guidelines, amikacin levels 2002;19:341–48.
from an extended-interval dose are to be interpreted 7. Freeman CD, Nicolau DP, Belliveau PP, et al. Once-
using an established nomogram. If we use the Hart- daily dosing of aminoglycosides: review and
ford nomogram to make this interpretation, it is recommendations for clinical practice. J Anti-
necessary to divide the reported amikacin level by microb Chemother 1997;39:677–86.
2, and this number is then plotted on the nomogram
8. Nicolau DP, Freeman CD, Belliveau PP, et al. Expe-
(14 mcg/mL divided by 2 = 7 mcg/mL). Plotting
rience with a once-daily aminoglycoside program
this value on the nomogram demonstrates a dosing
administered to 2,184 adult patients. Antimicrob
interval of every 36 hours. Therefore, this patient
Agents Chemother 1995;39(3):650–55.
should receive amikacin 820 mg every 36 hours.
9. Barclay ML, Begg EJ. Aminoglycoside adaptive
resistance: importance for effective dosage regi-
References mens. Drugs 2001;61:713–21.
1. Levey AS, Bosch JP, Lewis JB, et al. A more accu- 10. Dosage of antimicrobial drugs in adult patients
rate method to estimate glomerular filtration rate with renal impairment. In: The Sanford Guide to
from serum creatinine: a new prediction equa- Antimicrobial Therapy, 43rd ed. Sperryville, VA:
tion. Modification of Diet in Renal Disease Study Antimicrobial Therapy, Inc.; 2013. p. 205.
Group. Ann Intern Med 1999;130(6):461–70. 11. Devabhakthuni S. Antibiotic pharmacokinetic
2. Myers GL, Miller WG, Coresh J, et al. Recom- monitoring. In: ASHP New Practitioners
mendations for improving serum creatinine Forum, 2011, p. 3–4. http://www.ashp.org/
measurement: a report from the laboratory DocLibrary/MemberCenter/NPF/2011Pearls/
working group of the national kidney disease Antibiotic-Pharma
education program. Clin Chem 2006;52(1):5–18. 12. The Nebraska Medical Center Pharmacokinetic
3. Spruill WJ, Wade WE, Cobb HH. Estimating Training Packet for Pharmacists. 2012, pp 7–8;
glomerular filtration rate with a modification 12–14. http://www.nebraskamed.com/app_
of diet in renal disease equation: implica- file/pdf/careers/education-program/asp/pk_tr
tions for pharmacy. Am J Health Syst Pharm
2007;64(6):652–60.

202
Discussion Points
D-1. In Case 1, Problem 1A, suppose SG was D-3. Steady-state peak and trough serum concen-
admitted to the hospital with gram-negative trations achieved with the maintenance
pneumonia. How would your maintenance dose you calculated in Discussion Point 1
dose differ in this patient to achieve a Cpeak were reported by the laboratory as peak =
of 8 mg/L and a Ctrough of 1 mg/L? 6.8 mg/L, and trough = 1.8 mg/L. Calculate a
new maintenance dose that will give you the
D-2. How would your loading dose differ for
desired peak and trough concentrations of
patient SG in Discussion Point 1 to achieve 8 mg/L and 1 mg/L, respectively.
an approximate peak plasma concentration
of 8 mg/L? D-4. Calculate an extended-interval amino-
glycoside dose for SG for a diagnosis of
gram-negative pneumonia.

LESSON 13
Vancomycin
In this lesson, Cases 1–4 focus on the antibiotic vancomycin. Before beginning,
however, a few key points about vancomycin should be reviewed. Vancomycin
is usually administered via intermittent intravenous infusions. For systemic
infections, vancomycin is given only by the intravenous route; only this route
is considered in this lesson. To minimize the occurrence of an adverse reaction
called red-man syndrome, vancomycin doses greater than approximately 700 mg
are commonly diluted in a larger volume of fluid (i.e., 250 mL) and infused over
2 hours. Vancomycin is the drug of first choice for serious methicillin-resistant
staphylococci infections and enterococci (group D streptococcus).
Pharmacokinetically, vancomycin is an example of a two-compartment model,
a concept that is discussed in Lesson 6. After intravenous administration, vanco-
mycin displays a pronounced distribution phase (a phase) (Figure 13-1) while
the drug equilibrates between plasma and tissues. During this initial distribution
phase (1–3 hours), plasma drug concentrations are quite high. As the drug distrib-
utes throughout the body, the plasma drug concentration declines rapidly over a
short period. This biexponential elimination curve for vancomycin is an important
consideration especially when evaluating peak plasma vancomycin concentration
determinations. It is important not to obtain plasma drug concentrations during
this initial distribution phase, as inaccurate pharmacokinetic calculations may
result.
Because of vancomycin’s strange initial distribution phase, there is some confu-
sion about the therapeutic values for peak and trough concentrations. Older data
that suggested peak concentrations of 30–40 mg/L are wrong because they were
sampled during this initially high distribution phase. Appropriately sampled peak
concentrations were then suggested to be approximately 18–26 mg/L, whereas
trough concentrations were suggested to be between 5 and 10 mg/L, except for
enterococci, for which vancomycin is only bacteriostatic and required a trough of
between 10 and 15 mg/L.
More recently, a consensus review of vancomycin, Vancomycin Therapeutic
Guidelines: A Summary of Consensus Recommendations from the Infectious Diseases
Society of America, the American Society of Health-System Pharmacists, and the
Society of Infectious Diseases Pharmacists, has presented the evidence for vanco-
mycin toxicity and monitoring.1 They concluded vancomycin efficacy is best
modeled as a total area under the drug concentration-time curve (i.e., concen-
tration-independent killing) versus concentration-dependent older methods of
using peak and trough concentrations to measure efficacy. The ratio of the area
under the serum drug concentration versus time curve (AUC/MIC) is a common
research lab method to measure efficacy. However, most clinical microbiology labs
203
204
sary, measuring both values may be useful to better

determine both initial dose, dosing interval, and
estimated trough in hemodynamically unstable
patients with significant decreases in renal func-
tion, in the elderly, and in those patients receiving
concomitant nephrotoxic drugs.
Population Estimates for Vancomycin

Volume of Distribution (V) and
Elimination Rate Constant (K)
Similar to the aminoglycoside cases in Lesson 12,
there are two commonly used population param-
eters to estimate V and K that can be used to
determine an initial vancomycin dose.
Although vancomycin volume of distribution can
FIGURE 13-1. vary quite widely, a commonly used average volume
Typical plasma concentration versus time curve for of distribution for vancomycin is approximately
vancomycin, demonstrating distribution and elimination
phases. 13-1 0.9 L/kg total body weight (TBW)
cannot measure AUC/MIC for all pathogens so we (Note that, unlike the aminoglycosides, it is recom-
rely on vancomycin trough concentrations that can mended that TBW be used to calculate the volume of
be approximated from AUC/MIC measurements distribution.)
for both efficacy and to decrease development of Vancomycin is eliminated almost entirely by
resistance for organisms with a minimum inhibi- glomerular filtration. Therefore, a reduction in
tory concentration (MIC) < 1. This research has renal function results in a decreased vancomycin
led to a commonly used dosing method based on clearance and an increased half-life. The average
MIC. MIC < 1 mcg/L requires a vancomycin trough vancomycin half-life for a patient with normal renal
of 10 to 20 mg/L; an MIC of 1.0 requires a trough function is approximately 6 hours (K = 0.116 hr–1).
of 15 to 20 mg/L. They further recommend trough One method of determining population estimates
concentrations of 15 to 20 mg/L for complicated for the elimination rate constant (K) based on creat-
infections such as bacteremia, hospital-acquired inine clearance (CrCl) is:
pneumonia, endocarditis, meningitis, and osteo-
13-2 K = 0.00083 hr –1 [CrCl (in mL/minute)] + 0.0044
myelitis. Unfortunately, this consensus paper also
presents evidence for increasing nephrotoxicity This equation, developed by Gary Matzke from
associated with vancomycin trough concentrations regression analysis of vancomycin clearance versus
> 15 mg/L, thus creating a dilemma between effi- creatinine clearance, has units of reciprocal hours,
cacy and toxicity. not milliliters per minute. In this type of equation,
Although routine traditional vancomycin peak units are not supposed to cancel out; rather, they
and trough level monitoring is often not neces- assume the units of the correlated value, K.

Lesson 13 | Vancomycin
205
assume that the plasma concentrations during the

elimination phase are more valuable for therapeutic
drug monitoring than the relatively high, transient
vancomycin concentrations of the distribution
phase (the first 1–2 hours after the infusion). With
this assumption, a one-compartment model can
be used to estimate vancomycin dosage or plasma
concentrations (Figure 13-2). We ignore the distri-
bution phase.
To calculate an initial vancomycin dose, given
the desired plasma concentrations, we use popula-
tion estimates for K and V to solve first for dosing
interval and then dose, in a similar fashion to that
done for the aminoglycosides, using the first-order
one-compartment model equation below:
K = 0.00083 hr–1 [CrCl (in mL/minute)] + 0.0044

FIGURE 13-2.
Plasma concentration versus time curve for vancomycin, showing = 0.00083 (99) + 0.0044
simplification with one-compartment model (dashed line). = 0.087, therefore T½ = 0.693/0.087
= 7.97 hours
CASE 1 V = 0.9 L/kg × Total body weight (TBW)
BW, a 42-year-old man, 5' 10'' tall, weighing = (0.9)(90)
90 kg, is admitted to the hospital for a right
= 81 liters
colectomy because of colon cancer. Initially,
he has normal renal function: serum creatinine
rate of drug administration
concentration of 1.00 mg/dL. He subsequently (mg/hr) % remaining after t hours
developed a postoperative wound infection,
which is treated with surgical drainage and an
intravenous cephalosporin. K 0 (1− e −Kt ) −Kt ′
13-3 C peak(steady state) = e
Culture of the wound fluid reveals methicillin- VK (1− e −K τ )
resistant Staphylococcus aureus, with a
vancomycin MIC < 1. Pharmacy is then consulted τ = desired dosing interval, determined as follows:
for vancomycin dosing for BW.
13-4
Problem 1A. Determine an appropriate dosing 1

= τ In C trough(desired) − In C peak(desired)  + t + t ′
regimen of vancomycin to achieve the desired steady- −K 
state plasma concentrations of 30 mg/L for the peak
(drawn 2 hours after the end of a 2-hour= infusion) 1
In 15 mg/L − In 30 mg/L  + 2 hr + 2 hr
and approximately 15 mg/L for the trough. How −0.087 hr −1 
many doses are required to reach steady state?
= 11.97 hr, rounded up to 12 hr Note: Additional 2 hr =
Several approaches are recommended for the calcu-
extra time from end
lation of vancomycin dosages; one relatively simple (See Equation 5-1 and of infusion until level
method is presented here. With this method, we Equation 12-4.) is drawn. Compare to
Equation 12-4.

206
where: where t´´ is the difference in time between the two

Cpeak(steady state) = desired peak concentration 2 hours plasma concentrations. In this case, t´´ equals τ
after infusion, (12 hours) – t (2 hours) – t´ (2 hours), or 8 hours.
K0 = drug infusion rate (dose/infusion Ctrough = 30 mg/L e(–0.087 hr–1)(8 hr)
time),
= 30 mg/L (0.499)
t = duration of infusion (usually
2 hours for vancomycin), = 14.95 mg/L
K = estimated elimination rate constant So the regimen should result in the desired plasma
(0.087 hr –1), concentrations of 30 mg/L and approximately
V = volume of distribution (population 15 mg/L.
estimate of 0.9 L/kg × 90 kg = 81 L), The number of doses required to attain steady
t´ = time between end of infusion and state can be calculated from the estimated half-life
collection of blood sample (2 hours; and the dosing interval. Steady state is attained
inclusion of t´ is different from the in three to five half-lives. In patient BW’s case, we
calculation for aminoglycosides will use three half-lives and our estimated K of
[see Lesson 12] because sampling 0.087 hr–1 in our calculations as follows:
time for vancomycin is actually at
time to steady state = 5 ×T 1
least 4 hours after the beginning of 2
the infusion), and 0.693
τ = desired dosing interval. T 1
2 =
K
These values are then put into the equation:
time to steady state = 5 ×T 1
2
−1
K 0 (1− e −0.087 hr (2 hr )
) −0.087 hr −1 (2 hr )
C peak(steady state) = −1 e 0.693
(81 L)(0.087 hr −1 )(1− e −0.087 hr (12 hr )
) = 5×
0.087 hr −1
K 0 (0.16)
30 mg/L = (0.84) = 5(7.97 hr)
(81 L)(0.087 hr −1 )(0.648)
= 39.85 hr
(30 mg/L)(81 L)(0.087 hr −1 )(0.648)
K0 = If doses are given every 12 hours, then steady state
(0.16)(0.84)
should be achieved by administration of the fourth
= 1019 mg vancomycin per 1 hr dose (by the end of the third dosing interval).
(to be infused over 2 hours) Remember that doses would be given at 0, 12, 24,
and 36 hours.
Because vancomycin is infused over 2 hours,
Problem 1B. To achieve the desired concentrations
total dose = (1019 mg/hr)(2 hr)
rapidly, a loading dose can be given. Determine an
= 2039 mg (round to 2000 mg) appropriate loading dose for patient BW. Assume that
vancomycin over 2 hr the loading dose will be a 2-hour intravenous infusion.
With this regimen, we can then predict the To estimate a loading dose, we need to know the
vancomycin plasma concentration at the end of the volume of distribution and the elimination rate
dosing interval (trough): constant. Because we do not know the patient-
specific pharmacokinetic values, the population
13-5 Ctrough = Cpeak(steady state)e–Kt´´ estimates can be used (V of 0.9 L/kg TBW [see
Equation 13-1] and K of 0.087 hr –1 as previously
where t'' = τ − t − t' determined [see Equation 13-2]). Then the equation
(See Equation 3-2.) as shown in Lesson 5 describing plasma concen-

207
tration over time with an intravenous infusion is

applied. Note that again we ignore the distribution
phase and assume that a one-compartment model is
adequate (Figure 13-3):
X 0 /t
C peak desired = (1− e −Kt )e −Kt ′
VK
where:
Cpeak(steady state) = desired peak plasma concentration
2 hours after infusion,
X0 = dose (note: X0/t = K0),
t = duration of infusion (2 hours),
(hr)
K = 0.087 hr –1,
V = 81 L, and FIGURE 13-3.
Plasma concentrations over time for a loading dose. Dashed
t´ = time after end of infusion (2 hours). line represents simplification to one-compartment model.
Note that the term e–Kt´ describes the decline in
plasma concentration from the end of the infusion
to some later time (2 hours in this example). Then, Clinical Correlate
insertion of the known values gives:
Close observation of Figure 13-3 confirms that
−1 we are not actually measuring a true peak
( X 0 /2 hr)(1− e −0.087 hr (2 hr ) ) −0.087 hr −1 (2 hr) concentration, as we did for aminoglycosides.
30 mg/L = e
(81 L)(0.087 hr −1 ) We are, rather, measuring a 2-hour postpeak
Note: This 2 is from concentration that places this point on the straight-
( X 0 /2 hr)(0.16)
= (0.84) transposing the 2 line portion of the terminal elimination phase.
(7.05 L/hr) in X0/2 component
indicating that the
(30 mg/L)(7.05 L/hr)(2 hr) loading dose is
X0 = infused over 2 hours Problem 1C. After administration of the loading
(0.84)(0.16)
dose and seven doses (1500 mg each) at 12-hour
= 3147 mg vancomycin, which may be rounded to intervals, plasma vancomycin concentrations are
3000 given over 2 or 3 hours determined to be 42 mg/L (2 hours after the end of
the 2-hour infusion) and 22 mg/L at the end of the
dosing interval. Calculate a new dose for BW (this
Clinical Correlate time using actual patient-specific K and V) to attain
the original target peak and trough concentrations
Note that although the calculated loading dose
(30 mg/L and 15 mg/L, respectively).
is 3147 mg, many clinicians would choose to
either infuse this over more than 2 hours, or The information needed to determine a new dosing
give this loading dose as two doses about 4 to regimen is the same as described in Problem 1A.
6 hours apart to minimize potential vancomycin However, because we now have data about this
infusion reactions such as red-man syndrome. specific patient, we no longer have to rely on popu-
Also, note how this calculated loading dose lation estimates. To begin, we should calculate the
compares to the easier method of using 25 to patient’s vancomycin elimination rate constant,
30 mg/kg TBW. half-life, and volume of distribution from the plasma
concentrations determined.

208
The half-life (T½) can then be calculated:

0.693
T 1
2 =
K
0.693
=
0.08 hr −1
= 8.66 hr
(See Equation 3-3.)
Calculation of V
Note that the elimination rate constant is lower, and
the half-life is greater than originally estimated. Now
the volume of distribution (V) can be estimated with
the multiple-dose infusion equation for steady state:
FIGURE 13-4.
Calculation of elimination rate constant given two plasma K 0 (1− e −Kt ) −Kt ′
concentrations (42 mg/L at 2 hours after the infusion and C peak(steady state) = e
22 mg/L at 10 hours after the end of a 2-hour infusion). VK (1− e −K τ )
Calculation of K (See Equation 13-3.)

First, the elimination rate constant (K) is easily where:
calculated from the slope of the plasma drug Cpeak(steady state) = peak concentration 2 hours after
concentration versus time curve during the elimina- infusion = 42 mg/L,
tion phase (Figure 13-4) (see Lesson 3):
K0 = maintenance dose (1500 mg over
In C 2 − In C 1 2 hours),
K = −
t2 − t1 t = duration of infusion (2 hours),
t´ = time between end of infusion and
In 22 mg/L − In 42 mg/L
= − collection of blood sample (2 hours),
10 hr − 2 hr
K = elimination rate constant (0.08 hr–1),
3.09 − 3.73 V = volume of distribution (to be
= −
8 hr determined), and
= 0.08 hr −1 τ = dosing interval (12 hours).
(See Equation 3-1.) These values are then put into the equation:
−1
(1500 mg/2 hr)(1− e −0.08 hr (2 hr)
) −1
Clinical Correlate C peak(steady state) = −1 −1

−0.08 hr (12 hr)
e −0.08 hr (2 hr)
V (0.08 hr )(1− e )
Be careful when selecting t2 and t1. In the above
example, if dose one is begun at 8 am and infused (1500 mg / 2 hr)(0.148)
42 mg/L = (0.85)
for 2 hours, then the patient would receive the V(0.08 hr -1 )(0.617)
entire dose by 10 am. The peak plasma level
would then be drawn 2 hours later, or 12 noon. Rearranging gives:
Given that the trough concentration will be
attained immediately before dose two (given at (1500 mg/2 hr)(0.148)(0.85)
V =
8 pm), the total time elapsed between the plasma (0.08 hr −1 )(0.617)(42 mg/L)
readings is 8 hours, or (t2 – t1).
= 45.5 L or 0.51 L/kg

209
So the original estimate for the volume of distribu- Resultant Cpeak and Ctrough concentrations for a
tion was higher than the volume determined with 1000-mg every-12-hour dose would be ~32 mg/L
the plasma concentrations. and 14.4 mg/L, respectively.
Calculation of New τ
Clinical Correlate
Before calculating a new maintenance dose, we can
first check to see if we need to use a new dosing One can easily see how this tedious, repetitive
interval, as follows: calculation of dose, dosing interval, and trough
1 concentration can be made much simpler
=τdesired [In C trough(desired) − In C peak(desired) ] + t + t ′ by using various computer and PDA dosing
−K
programs, allowing you to try many different
(See Equation 13-4.) combinations.
where:
t = duration of infusion (2 hours),
t´ = time after end of infusion (2 hours), and CASE 2
τ = dosing interval, calculated as follows:
PS, a 74-year-old woman, 60 kg, 5' 7'' tall,
1 serum creatinine of 2.50 mg/dL, is admitted to
τ (In 15 mg/L − In 30 mg/L) + 2 hr + 2 hr the hospital after sustaining multiple traumatic
−0.08 hr −1
injuries in a motor vehicle accident.
1
= (2.7 − 3.4) + 4 hr She experiences a spiking fever; gram-positive
−0.08 hr −1 cocci, resistant to methicillin but susceptible
= 12.66 hr to vancomycin, are subsequently cultured from
her blood. Her physician consults the pharmacy
for vancomycin dosing and monitoring. A quick
Therefore, our best new dosing interval is approxi-
clinical assessment of this patient indicates that
mately 12 hours.
her renal function is extremely low, meaning
her time to steady state would be many days.
Calculation of New K0 Estimated pharmacokinetic parameters confirm
this assumption: CrCl ~18.7 mL/minute,
K 0 (1− e −Kt ) −Kt ′ estimated K of 0.02, V of 54 liters, T½ of ~35
C peak (steady state) = e
VK (1− e −K τ ) hours, and, therefore, a time to steady-state
−1
calculation of between 104 hours, using three
K 0 (1− e −0.08 hr (2 hr )
) −1
half-lives, and 173 hours, using five half-lives.
30 mg/L = −1 −0.08 hr −1 (12 hr )
e −0.08 hr (2 hr )
(0.08 hr )(45.5 L)(1− e ) Note that there are two opportunities to calculate
K 0 (0.148) patient-specific pharmacokinetic values—after
= (0.847) the first dose or after steady state has been
(0.08 hr −1 )(45.5 L)(0.62) achieved. In this case, because the patient has
such a long half-life, it is decided to calculate
Rearranging gives: these parameters after the first dose, which
allows for subsequent dose adjustments without
(30 mg/L)(0.08 hr −1 )(45.5 L)(0.62) waiting the many days necessary for steady state
K0 =
(0.148)(0.847) to be reached. The reason for calculating this
patient’s elimination rate constant and volume of
= (540 mg/hr)(2-hr infusion) distribution is to predict how often a vancomycin
dose will be needed, when the next dose should
= 540
= mg for 2 hours 1080 mg, be given, and the size of the next dose.
which will round to 1000 mg

210
Problem 2A. Two hours after the end of a 1000-mg The half-life in this case can be calculated:
loading dose administered over 1 hour, the vanco-
0.693
mycin plasma concentration was 29 mg/L; it is T 1
2 =
17.5 mg/L at 35 hours after the end of this infu- K
sion (Figure 13-5). Calculate the vancomycin 0.693
elimination rate constant, half-life, and volume of =
0.015 hr −1
distribution in this patient.
First, we calculate the elimination rate constant (K) = 46.2 hr
and half-life (T½):
(See Equation 3-3.)
K = −slope of natural log of vancomycin
K = −slope of natural log of vancomycin Now the volume of distribution (V) can be esti-
concentration
concentration versustime
versus timeplot
plot
mated, using the simple relationship given below:
In C 2 − In C 1
= − loading dose = plasma concentration achieved
t2 − t1
× volume of distribution
In 17.5 mg/L − In 29 mg/L
= − By rearranging, we get:
35 mg − 2 hr
loading dose
= 0.015 hr −1 V =
plasma concentration achieved
(See Equation 3-1.) 1000 mg
=
29 mg/L
Clinical Correlate = 34.5 L
Remember that the second plasma level was
Note that the patient’s calculated K of 0.015 hr and
taken 35 hours after the end of the infusion, not
V of 34.5 L are both lower than our estimated values
35 hours after the first plasma level. Therefore,
we must account for the 2 hours that elapsed of 0.02 hr and 54 L, respectively.
between the end of the infusion and first plasma Problem 2B. With the information just determined,
level.
calculate when the next vancomycin dose should
be given and what it should be. Assume that the
plasma vancomycin concentration should decline
to 12 mg/L before another dose is given and that
the plasma concentration desired 2 hours after the
infusion is complete is 29 mg/L (i.e., desired Cpeak).
First, we must know the time needed for the plasma
concentration to decline to 12 mg/L. It can easily
be calculated from the known plasma concentra-
tions, the elimination rate constant, and the desired
trough plasma concentration:
Ctrough = Cpeak e–Kt
where:
FIGURE 13-5. Cpeak = observed concentration of 29 mg/L,
Plasma concentrations after loading dose of vancomycin in a
patient with renal impairment (29 mg/L at 2 hours and 17.5 mg/L K = elimination rate constant
at 35 hours after the end of the infusion). (0.015 hr–1), and

211
t = time between observed concentration t´ = time after end of infusion (2 hours),

of 29 mg/L and trough concentration of V = 34.5 L, and
12 mg/L (unknown).
K = 0.015 hr–1.
Then:
12 mg/L = (29 mg/L)(e(–0.015 hr –1)(t ) ) Rearranging to solve for K0:
To solve for t, we can first take the natural log of VK (C peak(steady state) )(1− e −K τ )
K0 =
each side of the equation: (1− e −Kt )(e −Kt ′ )
ln 12 mg/L = ln 29 mg/L (–0.015 hr –1)(t ) (34.5 L)(0.015 hr −1 )(29 mg/L)(1− e −0.015 hr
−1
( 48 hr )
)
= −0.015 hr −1 (1 hr ) −0.015 hr −1 (2 hr )
Rearranging gives: (1− e )(e )
In 15 mg/L − In 29 mg/L 7.96 mg/hr
t= =
−0.015 hr −1 (0.015)(0.970)
= 44.6 hr = 547 mg, rounded to 500 mg
Therefore, at approximately 45 hours after the
plasma concentration of 29 mg/L is observed (or Finally, we must check to see what our trough
~47 hours after the end of the infusion), the next concentration will be after rounding both dose and
vancomycin dose can be given. dosing interval:
Next, we determine dosing interval and mainte- Ctrough = Cpeake–Kt´´ (See Equation 13-5.)
nance dose as follows:
where:
1 t´´ = time between Ctrough and Cpeak
=τdesired [In C trough(desired) − In C peak(desired) ] + t + t ′
−K = τ – t – t´
1 = 48 – 1 – 2 = 45 hr
[In 12 mg/L − In 29 mg/L] + 1 hr + 2 hr
−0.015 hr −1 and:
–1)(45 hr)
1 Ctrough = (29 mg/L)e(–0.015 hr
= [2.70 − 3.4] + 3 hr Assume a 1-hour
−0.015 hr −1 infusion and a = 15.2 mg/L
2-hour time to wait
= 45 hr, rounded up to 48 hr before obtaining
vancomycin peak
concentration
The maintenance dose can then be calculated as
CASE 3
follows: A 60-year-old woman, patient BA (weighing 70 kg),
− Kt
K 0 (1− e ) −Kt ′ is being treated for a hospital-acquired, methicillin-
C peak(steady state) = e resistant, Staphylococcus aureus bacteremia seeding
VK (1− e −K τ ) from an infected sacral decubitus ulcer. MIC values
(See Equation 13-3.) for this organism are < 1 mg/L. Her estimated CrCl is
30 mL/minute. Her physician prescribed an initial
where:
1000-mg vancomycin loading dose followed by
Cpeak(steady state) = concentration 2 hours after end of a maintenance dose of 500 mg (infused over 1
infusion, hour) every 12 hours. Per hospital protocol you
τ = 48 hours, are required to check all vancomycin dosing and
t = duration of infusion (1 hour), recommend changes as needed.

212
Problem 3A. Predict the steady-state Ctrough from K 0 (1− e −Kt ) −Kt ′
this dose, using population average values for K and C peak(steady state) = e
VK (1− e −K τ )
V. How do they compare to the recommended Ctrough
of > 10 mg/L? (See Equation 13-3.)
The equation for a one-compartment, intermittent- where:
infusion drug can be used to solve for Cpeak(steady state) Cpeak(steady state) = concentration that would result
and Ctrough(steady state): from this dose at steady state,
K 0 (1− e −Kt ) −Kt ′ K0 = drug infusion rate (maintenance
C peak(steady state) = e dose per hour),
VK (1− e −K τ )
estimate for vancomycin),
where:
Cpeak(steady state) = peak plasma concentration at tion estimate for vancomycin),
steady state,
t = duration of infusion,
K0 = drug infusion rate (also
t´ = time from end of infusion until
maintenance dose given over 1 hour),
concentration is determined
V = volume of distribution (population (2 hours for peak), and
estimate for vancomycin of 0.9 L/kg
τ = desired or most appropriate dosing
TBW),
interval.
Therefore:
tion estimate for vancomycin), −1
(500 mg/1 hr)(1− e −0.03 hr (1 hr)
) −1
t = infusion time (1 hour in this case), C peak(steady state) = e −0.03 hr (2 hr)
−1 −0.03 hr −1 (12 hr)
τ = patient’s current dosing interval, and
(63 L)(0.03 hr )(1− e )
t´ = time between end of infusion and (500 mg/hr)(0.029)
= (0.94)
collection of blood sample (2 hours). (1.89 L/hr)(0.302)
First, we must calculate patient BA’s K and V
= (264.5 mg/L)(0.096)(0.94)
values for use in this equation. The estimated K
would be: = 23.9 mg/L
K = 0.00083 hr –1 (CrCl) + 0.0044 The peak concentration, in this case, is primarily
(See Equation 13-2.) calculated to continue the math necessary to calcu-
late her trough concentration and can be estimated
= 0.00083 (30) + 0.0044
with the following equation:
= 0.029 hr –1
Ctrough(steady state) = Cpeak(steady state)e–Kt
which we shall round to 0.03 hr for ease of calculation.
–1
Patient BA’s estimated volume of distribution In this case, patient BA’s Ctrough will equal the Cpeak
(V) is calculated from the population estimate of (drawn 2 hours after the 1-hour infusion) multiplied
0.9 L/kg TBW: by the fraction of this Cpeak remaining after elimina-
tion has occurred for t´ hours, which, in this case,
0.9 L/kg × 70 kg = 63 L (See Equation 13-1.) is 9 hours (12-hour dosing interval minus 3 hours).
Now that we have these estimates of K and V, we The patient’s estimated Ctrough(steady state) is calcu-
can calculate the Cpeak and Ctrough values that would be lated as follows:
–1)(9 hr)
obtained with this dose of 500 mg every 12 hours. Ctrough(steady state) = (23.9 mg/L)e(–0.03 hr
By application of the general equation for a one-
= (23.9)(0.76)
compartment, first-order, intermittently infused
drug, we get: = 18.2 mg/L
213
Now that you know the eventual expected steady- Then:

state trough of 18 mg/L, a clinical decision can be (K 0 /2)(1− e −0.03 hr
−1
(2 hr)
) −1
made to either decrease dose slightly or wait until C peak(desired) = −1 −0.03 hr −1 (24 hr)
e −0.03 hr (2 hr)
(63 L)(0.03 hr )(1− e )

patient’s vancomycin level is at steady state and
obtain a steady-state trough concentration. (K 0 /2)(0.058)
28 mg/L = (0.94)
Problem 3B. What vancomycin dose would you (1.89)(0.66)
recommend for patient BA to attain a Cpeak of = (K 0 /2)(0.046)(0.94)
28 mg/L (drawn 2 hours after the end of a 2-hour
infusion) and a Ctrough of 12 mg/L? 28 = (K 0 /2)(0.044)
Note: You can also change infusion time to 28 / 0.044 = (K 0 /2)
2 hours when recommending a new dose.
636 = (K 0 /2)
Using the estimates of K (0.03 hr–1) and V (63 L), we
should first determine the best dosing interval (τ) K 0 = 1272 mg rounded down to 1200 mg
for patient BA: because we rounded our interval
Note: Answer of 636 mg/hr
1 for two infusions = 1272 to every 24 hours.
=τ [In C trough(desired) − In C peak(desired) ] + t + t ′
−K
The expected trough concentration can now be
(See Equation 13-4.) calculated:
where t is the time of infusion and t´ is the time after Ctrough(steady state) = Cpeak(steady state)e–Kt
the end of the infusion. Then:
1 In this case, Ctrough will equal the Cpeak (drawn 2 hours
τ (In 12 mg/L − In 28 mg/L) + 2 hr + 2 hr
−0.03 hr −1 after the 2-hour infusion is complete) multiplied
=
−33.33 hr(2.48 − 3.3) + 4 hr by the fraction of this Cpeak remaining after elimina-
tion has occurred for t hours, which, in this case, is
= 31.3 hr, and we will round down to 24 hr 20 hours (24-hour dosing interval minus 2 hours
minus 2 hours). So:
K 0 (1− e −Kt ) −Kt ′ –1)(20 hr)
C peak(steady state) = e Ctrough(steady state) = (28 mg/L)e(–0.03 hr
VK (1− e −K τ )
= 28 (0.55)
where: = 15.4 mg/L
Cpeak(steady state) = desired peak concentration at Thus, a dose of 1300 mg every 24 hours will yield an
steady state, estimated Cpeak of 28 mg/L and an estimated Ctrough of
15.4 mg/L, which should be adequate in BA’s case.
K0 = drug infusion rate (also maintenance
dose you are trying to calculate),
Clinical Correlate
estimate for vancomycin), Don’t let these equations intimidate you. Try
to develop a step-by-step model to walk you
K = elimination rate constant (popula- through the calculations, such as:
tion estimate for vancomycin),
• Determine patient-specific K and V values. If
t = duration of infusion, these values are not known, use population
t´ = time from end of infusion until estimates.
concentration is determined • Determine the dosing interval.
(2 hours for peak), and • Determine the drug infusion rate (K0).
τ = desired or most appropriate dosing • Check the trough to make sure it is within
interval. your desired range.
214
30 Peak
where:
Plasma Drug Concentration
Cpeak(steady state) = measured steady state peak plasma

concentration (30 mg/L) drawn 2
20 Trough
hours after end of a 1-hour infusion,
(mg/L)
K0 = drug infusion rate (maintenance dose

10 9 hours (τ – 3)
of 500 mg, infused over 1 hour),
V = volume of distribution (unknown),
0 3 4 6 8 10 12 K = elimination rate constant calculated
Hours After Start of Infusion from Cpeak and Ctrough (0.034 hr–1),
FIGURE 13-6. t = infusion time (1 hour),
Time between peak and trough t´ = time from end of infusion until
concentration is determined (2 hours
Problem 3C. Despite your dosing recommendation, for peak) (see Figure 13.6), and
BA continued to receive her original dose of 500 mg
τ = dosing interval at time concentra-
every 12 hours (infused over 1 hour) with resul-
tions are obtained (12 hours).
tant steady-state peak and trough levels of 30 and
22 mg/L, respectively (Figure 13-6). Adjust patient By substituting the above values, we obtain:
BA’s dose, this time using her specific pharmaco- −1
kinetic parameters, to give a Cpeak of approximately

(500 mg/1 hr)(1− e −0.034 hr (1 hr)
) −1
30 mg/L = −1 −0.034 hr −1 (12 hr)
e −0.034 hr (2 hr)
28 mg/L and a Ctrough(steady state) of approximately V (0.034 hr )(1− e )

12 mg/L.
(500 mg/hr)(0.033)
To adjust this patient’s dose, we must first deter- = (0.93)
V (0.034 hr −1 )(0.34)
mine her real K and V values, then calculate a new
dosing interval, and finally solve for a new mainte- V = 42.6 L
nance dose.
Note the differences between the previously
To calculate K, we can use:
estimated K and V of 0.03 hr–1 and 63 L and the
In C trough − In C peak calculated values of 0.034 hr–1 and 42.6 L. The K
K = − values are quite similar; however, the V is much
τ −t −t ′
smaller than originally calculated.
In 22 mg/L − In 30 mg/L Now, to calculate the best dosing interval, once
= −
9 hr again infused over only 1 hour in this case, to get
3.09 − 3.40 a Cpeak of 28 mg/L and a Ctrough(steady state) of approxi-
= mately 12 mg/L, we would use:
9 hr
0.31 1
= =τ [In C trough(desired) − In C peak(desired) ] + t + t ′
9 hr −K
= 0.034 hr −1 1
= [In 12 mg/L − In 28 mg/L] + 1 hr + 2 hr
−0.034 hr −1
(See Equation 3-1.)
1
To calculate V, we can use: = [2.48 − 3.33] + 3 hr
−0.034 hr −1
K 0 (1− e −Kt ) −Kt ′
C peak(steady state) = e = 25 hr, rounded to 24 hr
VK (1− e −K τ )
(See Equation 13-3.) (See Equation 13-4.)

215
The new maintenance dose now can be Ctrough(steady state) = Cpeak(steady state)e–t´´
calculated: (See Equation 13-5.)
K (1− e −Kt ) −Kt ′
= (28.4 mg/L)e (–0.034 hr
–1)(21 hr)
C peak(steady state) = 0 e
VK (1− e −K τ )

= 13.9 mg/L
This new dose of 750 mg every 24 hours based on
where:
patient-specific PK parameters will then give a Cpeak
Cpeak(steady state) = desired peak concentration at of approximately 28 mg/L and a Ctrough of approxi-
steady state (28 mg/L), mately 14 mg/L.
K0 = drug infusion rate (also main-
tenance dose you are trying to Problem 3E. Because patient BA’s Ctrough of
calculate, in milligrams per hour), 22 mg/L is too high from the regimen of 500 mg
every 12 hours, the dose will need to be held for a
V = volume of distribution (42.6 L),
certain amount of time before beginning the new
K
= elimination rate constant calculated dose of 750 mg every 24 hours.
from Cpeak and Ctrough (0.034 hr–1), Before changing to 750 mg every 24 hours, you must
t = infusion time (1 hour), wait for patient BA’s Ctrough of 22 mg/L to decrease
t´ = time from end of infusion until to the desired Ctrough of approximately 12 mg/L. The
concentration is determined formula for calculating the number of hours to hold
(2 hours for peak), and the dose is:
τ = desired or most appropriate dosing Ctrough(desired) = Ctrough(actual)e–Kt (See Equation 3-2.)
interval (24 hours).
where t is the amount of time to hold the dose. This
Then:
−1
formula is an application of the general formula (see
K 0 (1− e −0.034 hr (1 hr)
) −1 Lesson 3) that the concentration at any time equals
C peak(steady state) = e −0.034 hr (2 hr)
−1
(42.6 L)(0.034 hr )(1− e −0.034 hr −1 (24 hr)
) a previous concentration multiplied by the fraction
remaining:
K 0 (0.033)
28 mg/L = (0.93) C = C 0 e–Kt
(1.45)(0.56)
K 0 (0.031) where:
=
0.81
C = drug concentration at time t,
K 0 = 740 mg
C0 = drug concentration at some earlier time or
time zero, and
Being conservative, we would round this dose to e–Kt = fraction of original or previous concentra-
750 mg, which would only slightly raise the actual tion remaining.
peak value from 28 to approximately 28.4 mg/L.
In patient BA’s case:
This calculation is shown below:
12 mg/L = (22 mg/L)e(–0.034 hr–1)(t)
(750 mg/740 mg) × 28 mg/L = 28.4 mg/L
0.55 mg/L = e(–0.034 hr–1)(t)
Problem 3D. Calculate the Ctrough(steady state) for patient
BA if she receives the new dose of 750 mg every Next, take the natural logarithm of both sides:
24 hours. ln 0.55 = ln (e(–0.034 hr–1)(t ) )
We can use the following equation, where t´´ is now
–0.61
= –0.034(t )
the number of hours between the peak and trough
(t´´ = τ – t – t´). Therefore, t´´ = 21 hours. t = 17.9 hr

216
We should hold this patient’s dose for an addi-

tional 18 hours after the next Ctrough and then begin
her new dose. The same equation can be used to
CASE 4
determine the amount of time to hold the dose from A 65-year-old man, patient RK, has a history of
the last Cpeak of 30 mg/L. Again, the general equation subacute bacterial endocarditis secondary to
is: a mitral valve replacement 5 years ago. He is
currently hospitalized for methicillin-resistant
C = C0e–Kt (See Equation 3-2.) Staphylococcus aureus bacteremia. He has been
where: treated with 750 mg of vancomycin every 16 hours
for the last 10 days. His most recent Cpeak was
C = drug concentration at time t (representing
26 mg/L (drawn 2 hours after a 2-hour
here the desired Ctrough of 12 mg/L), vancomycin infusion), and his most recent Ctrough
C0 = drug concentration at some earlier time was 9 mg/L.
(representing here Cpeak of 30 mg/L), and
e–Kt = fraction of previous concentration remaining.
Problem 4. Patient RK’s physician wants to
In patient BA’s case: discharge him and allow a local home infusion
company to administer his vancomycin on a once-a-
12 mg/L = (30 mg/L)e(–0.034 hr –1)(t ) day basis. You are asked to determine if it is possible
0.40 mg/L = e(–0.034 hr –1)(t ) to obtain a Ctrough of > 10 mg/L with a once-a-day
dose. What is your response?
Next, take the natural logarithm of both sides: Before answering this question, we must be sure
we know what the question is asking. Basically, this
ln 0.40 = ln (e(–0.034 hr –1)(t) ) question is asking whether, based on the patient’s
–0.92 = –0.034(t ) pharmacokinetic parameters, a dose can be given to
obtain a satisfactory Cpeak and Ctrough given a dosing
27 hr = t interval of 24 hours.
We should hold this patient’s dose for an additional First, we must determine patient RK’s pharmaco-
27 hours after the Cpeak and then begin her new dose. kinetic parameters based on his Cpeak of 26 mg/L and
Note that you can calculate time to hold using either Ctrough of 9 mg/L. To calculate K, we can use:
Cpeak or Ctrough; both methods give the correct answer, In C − In C peak(measured)
but you must examine where you are in the dosing −K = trough(measured)
τ −t −t ′
versus serum concentration sequence.
(See Equation 3-1.)
A more intuitive method for estimating
the time to hold patient BA’s dose is by exami- where t´ represents 2 hours, the number of hours
nation of the vancomycin half-life. We know after the infusion that the Cpeak was drawn. Then:
that the drug concentration decreases by half In 9 mg/L − In 26 mg/L
over each half-life. We can estimate how many
K = −
12 hr
drug half-lives to wait for her concentration to
approach our desired amount of 12 mg/L as 2.20 − 3.26
= −
follows. For patient BA (Ctrough of 22 mg/L and T½ 12 hr
of 20.3 hours [0.693/0.034]), the concentration
will drop by one-half from 22 to 11 mg/L in one = 0.088 hr −1
half-life of 20 hours. Because a concentration of To calculate V, we can use:
11 mg/L is acceptable, we need to hold only the K 0 (1− e −Kt ) −Kt ′
next scheduled dose for an additional 20 hours C peak(steady state) = e
VK (1− e −K τ )
before beginning the new dose of 750 mg every
24 hours. (See Equation 13-3.)

217
where: By substituting the above values, we obtain:

Cpeak(steady state) = measured peak plasma concentration −1
(26 mg/L),
(K 0 /2)(1− e −0.088 hr (2 hr )
) −1
26 mg/L = −1) −0.088 hr (24 hr) −1 e −0.088 hr (2 hr)
K0 = drug infusion rate (also mainte- (32.3 L)(0.088 hr (1− e )

nance dose of 750 mg), (K 0 /2)(0.16)
V = volume of distribution (unknown), 26 mg/L = (0.84)
(2.84)(0.88)
K = elimination rate constant calculated
from Cpeak and Ctrough (0.088 hr–1), K 0 /2 = 483 mg
t = duration of infusion (2 hours), K 0 = 966 mg, round to 1000 mg, infused over 2 hours
t´ = time from end of infusion until
concentration is determined Finally, we must check to see that our Ctrough concen-
(2 hours for peak), and tration with this dose is acceptable.
τ = dosing interval at time concentra- Ctrough(steady state) = Cpeak(steady state)e–Kt´´
tions are obtained (16 hours).
By substituting the above values, we obtain: (See Equation 13-5.)
−1 In this case, patient RK’s Ctrough will be equal to his
(750 mg/2)(1− e −0.088 hr (2 hr)
) −0.088 hr −1 (2 hr) Cpeak of 26 mg/L multiplied by the fraction of the
26 mg/L = −1 e
V (0.088 hr −1 )(1− e −0.088 hr (16 hr)
) Cpeak remaining after elimination has occurred for
t´´ hours, which, in this case, is 20 hours (24-hour
(750 mg/2)(0.16)
= (0.93) dosing interval minus t [2 hours] minus t´ [2 hours]).
V (0.088 hr −1 )(0.755)
Therefore:
V = 32.3 L –1)(20 hr)
Ctrough(steady state) = (26 mg/L)e(–0.088 hr
Next, we use our general equation to solve for = 4.2 mg/L
K0 (maintenance dose) with our predetermined
24-hour dosing interval: We can conclude that 1000 mg every 24 hours
will not yield a trough concentration > 10 mg/L. In
K 0 (1− e −Kt ) −Kt ′
C peak(steady state) = e fact, the dose would need to be doubled to 2000 mg
VK (1− e −K τ ) every 24 hours to accomplish this, which would, in
where: turn, double the expected peak concentration from
Cpeak(steady state) = desired peak concentration at 26 mg/L to 52 mg/L.
steady state (26 mg/L),
K0 = drug infusion rate (also main- Trough-Only Vancomycin
tenance dose you are trying to Pharmacokinetics
calculate, in milligrams per hour
infused for 2 hours), Because the trough vancomycin concentration has
been shown to be most associated with drug effi-
V = calculated volume of distribution
cacy and decreased development of microorganism
(32.3 L),
resistance and yet also associated with nephro-
K = elimination rate constant calculated toxicity, many practitioners simply use a ratio and
from Cpeak and Ctrough (0.088 hr–1), proportion method of dosing adjustment based
t = duration of infusion time (2 hours), solely on the trough level. For instance, if trough =
t´ = time from end of infusion until 8 on a dose of 750 mg every 12 hours, they simply
concentration is determined double both values and give 1500 mg every 12
(2 hours for peak), and hours, to yield a trough of approximately twice the
τ = dosing interval desired (24 hours). previous value (from 8 to 16 mg/L). Unfortunately,

218
single-trough-level-only dosing methods do not References

allow for calculation of individual patient–specific
values for vancomycin clearance (CLvanco), volume 1. Rybak MJ, Lomaestro BM, Rotscahfer JC, et al.
of distribution (V), elimination rate constant (K), or Vancomycin therapeutic guidelines: a summary
estimated Cpeakss, and therefore, makes a concomi- of consensus recommendations from the Infec-
tant change in dosing interval somewhat of a tious Diseases Society of America, the American
guessing game. Consequently, several methods have Society of Health-System Pharmacists, and the
been devised that attempt to estimate one vanco- Society of Infectious Diseases Pharmacists. Clin
mycin population estimate such as K or Vd and solve Infect Dis. 2009;49(3):325–327.
for the other estimate to obtain a “better” Cpeakss. 2. Ambrose PJ, Winter ME. Vancomycin. In: Winter
Although there are many iterations of this method, ME. Basic Clinical Pharmacokinetics. 4th ed.
these single-trough methods estimate either K or Philadelphia: Lippincott Williams & Wilkins;
Vd and then solve for the other. For instance, some 2004:451–476.
practitioners use the Matzke equation (as shown in
Equation 13-3) to estimate K and solve for V or vice
versa. Another popular and intuitive method is the
Ambrose-Winter method that uses the simple equa-
tion of
Cpeakss = (dose/V ) + Ctroughss
This equation allows you to estimate a peak concen-
tration based on the simple relationship that
Concentration = amount of drug (or dose)/Volume
and then using this value of Cpeakss (but now written
as [dose/V] + Ctroughss) where (dose/V) + Ctroughss is
simply a re-expression of Cpeakss as also shown above.
Although these methods are not as accurate as
having both a “real” peak and trough serum concen-
tration, they are more accurate than using estimates
for both K and Vd and may be adequate in most clin-
ical situations.

219
Discussion Points
D1. In Case 1, Problem 1A, suppose BW is actu- vancomycin IV every 12 hours) is sched-
ally 6' 2'' tall, weighs 106 kg, and has an uled for 8 am on 12/1. Describe in detail the
estimated creatinine clearance of 61 mL/ process of how you determine when serum
minute. How would your maintenance dose levels (and what type of levels) should be
differ to achieve plasma concentrations of obtained. Then write an order as it would
22 mg/L for the peak (2 hours after a 2-hour appear in the Physician’s Order section of
infusion) and approximately 12 mg/L for the patient’s medical record for how serum
the trough? levels should be obtained. This order should
be grammatically correct, include only
D2. Steady-state serum concentrations resulting
approved abbreviations, and provide suffi-
from the maintenance dose you calculated
cient detail that nursing services can easily
in D-1 were reported by the laboratory as:
follow your instructions without having to
peak, 17.8 mg/L and trough, 10.2 mg/L.
contact you for further clarification.
Calculate a new maintenance dose to give
our desired peak and trough concentrations D5. Based on your experience in the provision
of 22 mg/L and approximately 12 mg/L, of direct patient care, design a pharmacy-
respectively. managed vancomycin dosing protocol that
could be used in your practice setting. This
D3. Assume that BA in Case 3 actually received a
protocol should be written from the stand-
vancomycin 1200-mg loading dose followed
point that the pharmacist is providing
by a maintenance dose of 1000 mg (over 2
complete dosing and monitoring of vanco-
hours) every 12 hours. Predict the steady-
mycin in a patient case (instead of simply
state peak and trough levels that would
providing recommendations to a physician
result from this maintenance dose, using
to manage). All steps required to effectively
population average values for K and V.
dose and monitor (including equations used)
D4. Assume that the first dose for a patient (a a patient for whom vancomycin is prescribed
41-year-old female, 5' 6'', 148 lbs, with posi- should be included. Describe in detail how
tive blood cultures for methicillin-resistant you would monitor this drug using serum
Staphylococcus aureus; serum creatinine, concentrations. Write the order for this drug
1.2 mg/dL; white blood cell count, 18,300/ as it would appear in the Physician’s Order
mm3, 10% bands; receiving 1000 mg of section of the patient’s medical record.

LESSON 14
Theophylline
Methylxanthines, including theophylline and aminophylline, have been used in the

management of asthma and chronic obstructive pulmonary disease (COPD) for
more than five decades. With time, the use of these agents has declined as a result
of the advent of alternative therapy, including beta-2 agonists, anticholinergics,
corticosteroids, mast cell stabilizers, leukotriene modifiers, and immunomodula-
tors. Although methylxanthines produce little therapeutic benefits for the patient
with asthma,1 these agents may reduce dyspnea, increase exercise tolerance, and
improve respiratory drive in patients with COPD.2 At the same time, theophylline
is an excellent agent for illustrating pharmacokinetic concepts associated with
the continuous intravenous infusion model. Cases in this lesson focus on patient-
specific dosing of aminophylline and theophylline.
Theophylline typically follows first-order pharmacokinetics in most patients
with serum concentrations within the therapeutic range of 5–15 mg/L. It may
undergo nonlinear, or Michaelis–Menten, pharmacokinetics (see Lesson 10) when
serum concentrations are within this range; however, this is more likely to occur
at concentrations exceeding 15 mg/L.3
Theophylline is eliminated from circulation through hepatic oxidative metabo-
lism (cytochrome P450) and has a low intrinsic clearance (see Lesson 9). Therefore,
total hepatic clearance of theophylline is determined by the intrinsic clearance of
the liver and is not dependent on liver blood flow. Disease states, drugs, and other
factors that may influence theophylline clearance are found in Table 14-1.
Theophylline is usually administered intravenously or orally. When theophyl-
line derivatives are used, the theophylline dose equivalent should be calculated.
For example, aminophylline is 80% theophylline. Therefore, to obtain the theophyl-
line dose equivalent, the aminophylline dose should be multiplied by 0.8.
Many different oral formulations of theophylline are available. Some of these
are rapidly absorbed after administration. Others are designed to slowly release
drug in the gastrointestinal tract for up to 24 hours. The type of oral product used
directly affects pharmacokinetic calculations.
221
222
TABLE 14-1. Factors and Drugs That Alter Theophylline Clearance

Total Body Clearance Clearance Adjustment
Factors (L/kg/hour) (× 0.04 L/kg/hour)
Hepatic disease 0.02 0.5

Acute pulmonary edema 0.02 0.5
Severe chronic obstructive pulmonary disease 0.03 0.8
Heart failure 0.016 0.5
Cor pulmonale 0.028 0.7
Cigarette smoking 0.063 1.6
Former cigarette smoking (quit > 2 years) 0.051 1.2
Marijuana smoking 0.072 1.7
Marijuana and cigarettes 0.09 2.2
Elderly cigarette smokers 0.045 1.1
Clearance Adjustment
Drugs (× 0.04 L/kg/hour)
Cimetidine (after 2 or more days) 0.5–0.7

Oral contraceptives 0.7
Interferon 0.15
Ciprofloxacin 0.7–0.75
Diltiazem 0.8–0.9
Norfloxacin 0.85
Phenytoin 1.35–1.5
Phenobarbital 1.35–1.5
Erythromycin 0.75–0.8
Propranolol 0.5–0.7
Verapamil 0.8–0.9
Rifampin 1.35–1.5
Phenytoin + smoking 1.9
CASE 1 Problem 1A. Calculate an appropriate loading

dose of aminophylline for MA that will result in
MA is a 62-year-old, 73-kg, man with a 30-year a theophylline concentration of 14 mcg/mL.
history of mild COPD that has been satisfactorily To calculate a loading dose of aminophylline requires
controlled with beta-2 agonist, ipratropium, and that we know the desired theophylline plasma
inhaled steroid therapy. However, over the past concentration, the patient’s theophylline volume of
two months, MA has experienced increased distribution, the aminophylline salt equivalent for
difficulty in breathing. His physician wishes to
theophylline, and the fraction of drug administered
admit him to the hospital and initiate intravenous
that reaches the systemic circulation.
aminophylline.

Lesson 14 | Theophylline
223
In this case, the desired plasma theophylline where:

concentration is 14 mcg/mL, the aminophylline salt D = the loading dose (milligrams),
equivalent (S) is 0.8, and the fraction of drug admin- Cpd = the desired concentration (milligrams
istered reaching the systemic circulation (F) is 1. per liter),
The one remaining factor that is necessary to make V = the volume of distribution (liters),
this loading dose calculation is the patient’s theophyl-
S = salt form, and
line volume of distribution (V). This is calculated from
F = bioavailability, which is equal to 1 for
the patient’s weight and the expected volume (in liters
drugs given intravenously.
per kilogram) from published literature:
Substituting known values for these parameters,
14-1 V (L) = weight (kg) × 0.5 L/kg
14 mg/L × (36.5 L)
= 73 kg (0.5 L/kg) D=
0.8
= 36.5 L = 638.75 mg
= 640 mg
Clinical Correlate This 640-mg aminophylline loading dose will
For theophylline, the patient’s actual body produce a serum concentration slightly greater than
weight should be used to calculate the volume 14 mcg/mL.
of distribution unless the patient’s actual weight Note: Remember, aminophylline is a salt form
is more than 50% above his or her ideal body of theophylline and contains approximately 80%
weight. In patients more than 50% above ideal theophylline equivalents.
body weight, volume of distribution should be
calculated using ideal body weight. Problem 1B. The loading dose is to be administered
over a 30-minute interval. An aminophylline main-
tenance infusion is to be started immediately on
Based on the information we now have, we can
completion of the loading dose. Suggest an amino-
calculate an aminophylline loading dose for MA.
phylline infusion rate for MA that will achieve a
The basic loading dose equation can be derived plasma theophylline concentration of 12 mcg/mL.
from the plasma concentration equation we learned
in Lesson 1. STEP A
amount of drug in body The first step in solving this problem is to estimate
concentration =
volume in which drug is distributed MA’s theophylline clearance. This can be accom-
plished by using the following equation:
X
C= 14-3 Cl = (0.04 L/kg/hr) × weight (kg)
V
where:
(See Equation 1-1.) Cl = clearance (L/hour; clearance
We can rewrite this equation to: is based on the patient’s actual
body weight), and
D = C ×V
0.04 L/kg/hour = population estimate found in
Taking into consideration the S and F values for the literature.
aminophylline, we can rewrite the above variation Therefore,
of Equation 1-1 as follows:
Cl = (0.04 L/kg/hr) × weight (kg)
CpdV = (0.04 L/kg/hr) × 73 kg
14-2 D=
SF = 2.92 L/hr

224
STEP B
To solve for a maintenance dose (milligrams per hour),
we can rearrange and slightly modify Equation 4-3 to
Equation 14-4 as follows:
dose
C =
Clt × τ
(See Equation 4-3.)
C pss Clτ
14-4 D=
SF
FIGURE 14-1.
where: Plasma concentrations with a loading dose and continuous
D = the maintenance dose (milligrams per infusion of theophylline or aminophylline.
hour),
C pss = average steady-state concentration • This requires that we know the value for K in
desired (micrograms per milliliter), this patient.
Cl = clearance (liters per hour), Using the equation Cl = K x V, we can use our esti-
mated values for Cl and V to estimate K.
S = salt form,
F = bioavailability, and Cl = K x V
τ = dosing interval, which is 1 hour for a
2.92 L/hr = K x (0.5 L/kg x 73 kg)
continuous intravenous infusion.
After inserting the Cl value calculated in Step A, = K x 36.5 L

S and F values, a dosing interval of one hour, and our
desired serum concentration for C , we can solve for
K = 0.08 hr −1
the maintenance dose: Now, substituting K into our half-life equation, we
12 mcg/mL × 2.92 L/hr × 1 hr can solve for half-life.
D=
0.8 T½ = 0.693/K
= 43.8 mg/hr
= 0.693/0.08
= 44 mg/hr
= 8.7 hours
Figure 14-1 demonstrates the relationship between Steady state will be reached in 5 half-lives.
serum levels achieved with the loading and mainte-
nance doses of theophylline or aminophylline. 5 x 8.7 hours = 43.5 hours
Problem 1C. How long will it take for MA’s theophyl- Problem 1D. MA’s steady-state theophylline serum
line therapy to reach steady state? concentration is 11.6 mcg/mL. Is there any reason
• Steady state is reached once a given dose of a to change his dose at this time?
drug is administered for 5 half-lives of the drug. As long as MA is improving clinically and not expe-
• Half-life is determined by the equation riencing theophylline adverse effects, it would be
T½ = 0.693/K. appropriate to leave his dose as is.

225
CASE 2 Problem 2C. Calculate an infusion rate of amino-

phylline that will maintain CJ’s serum concentration
at 12 mcg/mL.
CJ is a 48-year-old, 60-kg, woman who is
admitted to the hospital for treatment of severe To determine the infusion rate:
chronic bronchitis. She has a history of cigarette
smoking since age 14 and is currently receiving C pss Clτ
D =
verapamil for high blood pressure. SF

Problem 2A. Estimate CJ’s volume of distribution
and clearance for theophylline. 12 mcg/mL × 2.46 L/hr × 1 hr
D=
To estimate CJ’s volume of distribution (see 0.8
Equation 14-1):
= 36.9 mg/hr
V (L) = weight (kg) × 0.5 L/kg = 37 mg/hr
= 60 kg (0.5 L/kg)
Problem 2D. A steady-state theophylline serum
= 30 L concentration is reported by the lab as 18.2 mcg/mL.
Calculate a new aminophylline maintenance dose to
To estimate CJ’s clearance (see Equation 14-3): achieve a steady-state theophylline serum concen-
tration of 12 mcg/mL.
Cl = (0.04 L/kg/hr) × weight (kg) × (adjustment
factors) STEP A
The first step in calculating a new aminophylline
= (0.04 L/kg/hr) × 60 kg × (0.8) x (1.6) × (0.8) maintenance dose for CJ is to solve for her actual
theophylline clearance. When calculating her initial
= 2.46 L/hr
maintenance dose, we estimated clearance using
(The clearance adjustment factors of 0.8, 1.6, and a population value. Now that we have a measured
0.8 are found in Table 14-1 for severe bronchitis steady-state serum concentration, we can calculate
[severe COPD], cigarette smoking and verapamil, an actual value.
respectively.) To determine CJ’s actual theophylline clearance,
we can rearrange Equation 14-4 in Problem 1B and
Problem 2B. Calculate an aminophylline loading calculate this parameter as follows:
dose for CJ that will achieve an initial plasma
concentration of 12 mcg/mL. The dose will be given C pss Clτ
D =
as an infusion over 30 minutes. SF
To calculate the loading dose:
Rearrange to solve for Cl:
CpdV
D = DSF
SF
Cl =
C pss τ
12 mcg/mL × 30 L where:
D=
0.8 Cl = clearance (liters per hour),
= 450 mg D = maintenance dose (milligrams per hour),

226
S = salt form, where:

F = bioavailability, and K = elimination rate constant (hr –1),
C pss = average steady-state concentration Cl = clearance (liters per hour), and
(micrograms per milliliter). V = volume of distribution (liters).
37 mg/hr × 0.8 × 1 K = 1.63 L/hr/30 L

Cl =
18.2 mcg/mL = 0.054 hr −1
= 1.63 L/hr Next we can determine the time we need to wait
by using the following equation:
Notice that we estimated CJ’s theophylline Cl as
2.46 L/hour, but her actual value is 1.63 L/hr. C = C0e–Kt (See Equation 3-2.)
where:
STEP B
t = time to wait (hours),
Now that we have CJ’s actual theophylline clearance,
C = desired concentration (micrograms per
we can calculate a new maintenance dose that will
milliliter),
give us the desired theophylline serum concentra-
tion of 12 mcg/mL: C0 = current concentration (micrograms per

milliliter), and
C pss Clτ K = elimination rate constant (hr–1).
D =
SF
Therefore:
(See Equation 14-4.) 12 = 18.2e–0.054t
12/18.2 = e–0.054t
12 mcg/mL × 1.63 L/hr × 1 hr
D=
0.8 × 1 0.659 = e–0.054t
= 24.45 mg/hr ln 0.659 = ln e–0.054t
= 25 mg/hr –0.417 = –0.054t

STEP C t = 7.7 hours
Before we can begin this new maintenance dose of CJ would receive the new infusion of
aminophylline in CJ, it is necessary to determine 25 mg/hour starting 8 hours after discontinuing
how long we must hold her current dose until her the previous infusion of 37 mg/hour.
serum theophylline concentration declines to an
acceptable value. We will choose a level of 12 mcg/mL.
To determine how long it will be necessary to wait Clinical Correlate
before starting this new maintenance dose, we need
to determine CJ’s theophylline elimination rate. We The most significant side effects from theophylline
will make the calculation using her actual clearance occur at serum concentrations higher than
value. Using the following formula: 20 mcg/mL. These include nausea, vomiting,
headache, diarrhea, irritability, and insomnia.
Cl At concentrations higher than 35 mcg/mL,
K = major adverse effects include hyperglycemia,
V hypotension, cardiac arrhythmias, seizures, brain
damage, and death.
(See Equation 3-4.)

227
CASE 3 This 260-mg theophylline loading dose will

result in a serum concentration slightly less than
14 mg/L.
SR is a 52-kg patient admitted to the emergency
department with difficulty breathing. He has been Notice in this situation we are using a value of 1
prescribed theophylline on an outpatient basis but for S (theophylline is not in a salt form).
admits his compliance to his medication regimen
Problem 3B. Calculate a theophylline maintenance
is poor. A STAT theophylline level is reported as
3.9 mcg/mL. dose that will maintain SR’s serum concentration at
12 mcg/mL.
Problem 3A. Calculate an appropriate theophylline STEP A

loading dose to give SR a serum level of 14 mcg/mL.
As we saw in Case 1, the first step in solving this
STEP A problem is to estimate SR’s theophylline clearance.
This can be accomplished by using the following
Estimate SR’s theophylline volume of distribution.
equation:
V = weight (kg) x 0.5 L/kg
= 52 kg x 0.5 L/kg 14-3 Cl = (0.04 L/kg/hr) × weight (kg)
= 26 L
where:
STEP B
Cl = clearance (L/hour; clearance
Using SR’s estimated V, calculate an appropriate is based on the patient’s actual
theophylline loading dose. body weight), and
In this situation, we will slightly modify the 0.04 L/kg/hour = population estimate found in
loading dose Equation 14-2 to the following: the literature.
(Cpd-Cpi)V Therefore,
D=
SF
Cl = (0.04 L/kg/hr) × weight (kg)
where: = (0.04 L/kg/hr) × 52 kg
D = the loading dose,
= 2.08 L/hr
Cpd = the desired concentration (milligrams
per liter),
STEP B
Cpi = the initial concentration (milligrams per
liter), To solve for a maintenance dose (milligrams per
V = the volume of distribution (liters), hour), we can rearrange and slightly modify Equa-
tion 4-3 to Equation 14-4 as follows:
S = salt form, and
F = bioavailability, which is equal to 1 for dose
C =
drugs administered intravenously. Clt × τ
(14 mcg/mL − 3.9 mcg/mL) × 26 L
D= (See Equation 4-3.)
1× 1
= 262.6 mg
C pss Clτ
14-4 D=
= 260 mg SF

228
where: STEP B
D = the maintenance dose (milligrams per Calculate the dose to be given orally every 12 hours.
hour),
C pss = average steady-state concentration C pss Clτ
D=
desired (micrograms per milliliter), SF
Cl = clearance (liters per hour),
S = salt form, (See Equation 14-3.)
F = bioavailability, and where:
t = dosing interval, which is 1 hour for a D = the maintenance dose (milligrams),
continuous intravenous infusion. C pss = the desired average steady-state
After inserting the Cl value calculated in Step A, concentration (micrograms per milliliter),
S and F values, a dosing interval of one hour, and our C
l = clearance (liters per hour),
desired serum concentration for C pss , we can solve
for the maintenance dose: τ = dosing interval (hours),
S = salt form
12 mcg/mL × 2.08 L/hr × 1 hr
D= F = bioavailability
1× 1
= 24.96 mg/hr
12 mcg/mL × 2.08 L/hr × 12 hr
D=
= 25 mg/hr 1× 1
Notice that both S and F for theophylline are 1. = 299.5 mg
Problem 3C. SR has been stabilized on his 25 mg/hr = 300 mg

of theophylline regimen, and his steady-state serum
theophylline level is now 13.2 mcg/mL. Calculate SR should receive 300 mg of sustained-release
a dose of sustained-release theophylline for SR to theophylline orally every 12 hours.
maintain a theophylline serum concentration of Note that F for many oral sustained-release
12 mcg/mL. theophylline preparations is 1.
Many theophylline sustained-release formulations
follow the same pharmacokinetic profile as contin-
uous intravenous infusions. Therefore, we can easily References
make a conversion from a continuous infusion to an
oral sustained-release formulation as follows: 1. Kelly HW, Sorkness CA. Asthma. In: DiPiro JT,
Talbert RL, Yee GC, et al., eds. Pharmacotherapy:
STEP A
A Pathophysiologic Approach. 8th ed., www.
Calculate SR’s theophylline clearance. accesspharmacy.com
2. Williams DN, Bourdet SV. Chronic obstructive
DSF
Cl = pulmonary disease. In DiPiro JT, Talbert RL, Yee
C pss τ GC, et al., eds. Pharmacotherapy: A Pathophysi-
(See Equation 14-4.) ologic Approach. 8th ed., www.accesspharmacy.
com
25 mg/hr × 1× 1 3. Wagner JG. Theophylline: pooled Michaelis–
Cl =
13.2 mcg/mL × 1 Menten behavior of theophylline and its
parameters (Vmax and Km) among asthmatic chil-
= 1.89 L/hr dren and adults. Ther Drug Monit 1987;9:11.

229
Discussion Points
D-1. Assume MA in Case 1 is a 64-year-old man (including equations used) required to effec-
with significant liver disease due to alco- tively dose and monitor a patient for which
holism and has severe COPD. How would theophylline/aminophylline is prescribed
these factors affect the maintenance dose should be included. Describe in detail how
you calculated for him? you would monitor this drug using serum
concentrations. Write the order for this drug
D-2. In Case 2, assume CJ does not smoke tobacco,
as it would appear in the Physician’s Order
but is started on ciprofloxacin for her bron-
section of the patient’s medical record.
chitis. How would these changes affect her
aminophylline maintenance dose?
D-5. Assume that a 49-year-old female, 5' 8'' and
D-3. If SR in Case 3 had a serum theophylline level
162 lbs, with a serum creatinine of
of 16.2 mcg/mL as a result of his theophyl-
1.26 mg/dL, and white blood cell count
line continuous infusion maintenance dose,
of 18,300/mm3, is receiving aminophyl-
what dose of oral sustained-release theoph-
line as a continuous infusion in the dose of
ylline administered every 8 hours would
38 mg/hour. This dose was initiated at 8 am
he need to achieve a steady-state average
on 12/1. Describe in detail how you would
plasma concentration of 12 mcg/mL?
determine when a serum level (and what
D-4. Based on your experience in the provision type of level) should be obtained. Then write
of direct patient care, design a pharmacy- an order as it would appear in the Physi-
managed theophylline/aminophylline dosing cian’s Order section of the patient’s medical
protocol that could be used in your prac- record for how the serum level should
tice setting. This protocol should be written be obtained. This order should be gram-
from the standpoint that the pharmacist is matically correct, include only approved
providing complete dosing and monitoring abbreviations, and provide sufficient detail
of theophylline/aminophylline in a patient so that nursing services can easily follow
case (instead of simply providing recommen- your instructions without having to contact
dations to a physician to manage). All steps you for further clarification.

LESSON 15
Phenytoin and Digoxin
Phenytoin
Phenytoin is an anticonvulsant medication used for many types of seizure disor-
ders. Phenytoin is usually administered either orally or intravenously and exhibits
nonlinear, or Michaelis–Menten, kinetics (see Lesson 10). Unlike drugs under-
going first-order elimination (Figure 15-1), the plot of the natural logarithm of
concentration versus time is nonlinear with phenytoin (Figure 15-2). Phenytoin
is 90% protein bound; only the unbound fraction is active. (Note that patients with
low serum albumin concentrations will have a higher unbound, or active, fraction
of phenytoin. This should be factored in when dosing these patients.)
Phenytoin is metabolized by hepatic enzymes that can be saturated with
the drug at concentrations within the therapeutic range. Consequently, as the
phenytoin dose increases, a disproportionately greater increase in plasma concen-
tration is achieved. This enzyme saturation process can be characterized with an
enzyme-substrate model first developed by the biochemists Michaelis and Menten
in 1913. In this metabolic process, drug clearance is constantly changing (in a
nonlinear fashion) as dose changes. Drug clearance decreases as drug concentra-
tion increases (Figures 15-3 and 15-4).
To describe the relationship between concentration and dose, a differential
equation can be written as shown below:
dX Vmax × C ss
=
dt K m + C ss

where:
dX = change in amount of drug;
dt = change in time;
Vmax = maximum amount of drug that can be metabolized per unit time,
usually expressed as milligrams per day;
Km = Michaelis–Menten constant, representing the concentration of
phenytoin at which the rate of this enzyme-saturable hepatic
metabolism is one-half of maximum; and
Css = average steady-state phenytoin concentration.
Next, this differential equation can be re-expressed algebraically by assuming

that we are at steady state and dX/dt is held constant. Then dX/dt, the change in the
231
232

First-order elimination model. First-order elimination model.
amount of drug (X) over time (t), can be expressed

as X0/τ (dose over dosing interval), as shown in the Clinical Correlate
following equation:
Although fosphenytoin is a pro-drug containing
Vmax × C ss only 66% phenytoin free acid, it is correctly
( X 0 /τ )(S ) = prescribed and labeled in units of PE, meaning
K m + C ss
phenytoin sodium equivalents. The commercial
(See Equation 10-1.) fosphenytoin product is packaged to be very
where: similar to phenytoin sodium injection; it contains
150-mg fosphenytoin per 2-mL ampule,
X0/τ = amount of phenytoin free acid divided
providing 100 mg PE (100-mg phenytoin sodium
by dosing interval (which can also be
equivalents). Fosphenytoin is readily transformed
expressed as Xd, meaning daily dose of to phenytoin free acid by various phosphatases
phenytoin free acid) and throughout the body.
S = the salt factor or the fraction of phenytoin
free acid in the salt form used. S equals
0.92 for phenytoin sodium injection and This Michaelis–Menten equation (MME) can then
capsules and 1.0 for phenytoin suspension be rearranged to solve for Css as follows:
and chewable tablets (i.e., the free acid
form of phenytoin), and 0.66 for fospheny- Vmax × C ss
( X 0 /τ )(S ) =
toin injection. The oral bioavailability of K m + C ss
phenytoin is considered to be 100%, so an
F factor is not needed in these calculations. (See Equation 10-1.)

Michaelis–Menten elimination model. Michaelis–Menten elimination model.

Lesson 15 | Phenytoin and Digoxin
233
First, cross-multiply by the denominator:
[(X0 /τ)(S ) × Km] + [(X0/τ)(S ) × Css ] = Vmax × Css

CASE 1
SG, a 42-year-old black man, is admitted
Then transpose [(X0 /τ)(S) × Css] to the right side of to the hospital with status epilepticus that
the equation: was successfully treated with intravenous
lorazepam. His physician has written an order
[(X0 /τ)(S ) × Km ] = [Vmax × Css ] – [(X0/τ)(S ) × Css ] for the pharmacy to calculate and order an IV
phenytoin loading dose, recommend an initial
Factor out Css : oral maintenance dose, and order timing of
plasma concentrations. Other pertinent clinical
[(X0 /τ)(S ) × Km ] = Css [Vmax – (X0/τ)(S )] data include weight, 75 kg; height, 5' 8''; serum
creatinine, 1.2 mg/dL; and serum albumin, 4.8 g/dL.
Transpose [Vmax – (X0/τ)(S)] to the left side of the
equation:
Problem 1A. What intravenous loading dose and
( X /τ )(S )× K m oral maintenance dose would you recommend to
C ss = 0
Vmax − ( X 0 /τ)(S ) achieve and maintain a phenytoin concentration of
approximately 20 mg/L?
The two representations of the MME below can Calculation of the loading dose is not affected by
now be used for phenytoin dosing, as illustrated in the nonlinear pharmacokinetics of multiple-dose
the following cases: phenytoin regimens. The loading dose calculation
Vmax × C ss is based on the patient’s weight, estimated volume
( X 0 /τ )(S ) = of distribution, serum albumin concentration, renal
K m + C ss function assessment, and the salt form (i.e., salt
factor) of phenytoin used. The generally accepted
(See Equation 10-1.) population parameter for phenytoin’s volume of
distribution is 0.65 L/kg of body weight. The loading
( X 0 /τ )(S )× K m dose (X0) formula is:
15-1 C ss =
Vmax − ( X 0 /τ)(S ) V × C desired
X0 =
S
(See Equation 1-1.)
Phenytoin Pharmacokinetic Parameters
where:
In contrast to first-order drugs that use pharmaco-
V
= volume of distribution estimate of 0.65 L/kg
kinetic parameters for K and V, phenytoin dose
[V = 0.65 L/kg (75 kg) = 48.8 L for SG],
calculations use population estimates for Km and
Vmax. A Km population estimate of 4 mg/L and a Vmax Cdesired = concentration desired 1 hour after the end
estimate of 7 mg/kg/day are commonly used. Note, of the infusion (20 mg/L for SG), and
however, that Km can range from 1 to 15 mg/L while S = salt factor (0.92 for injection). Note that
Vmax can range from 3 to > 10 mg/kg/day in selected this dose falls within the empiric loading
patients. Phenytoin’s volume of distribution is dose range of 15–20 mg/kg.
usually estimated as 0.65 L/kg total body weight. Therefore:
Last, phenytoin trough serum drug concentrations (48.8 L)(20 mg/L)
(meaning in this case, at least 8 to 12 hours after X0 =
0.92
last oral dose) are usually used in dosing adjustment
calculations to avoid unpredictable rates and extent = 1060 mg of phenytoin sodium
of drug absorption from various dosage forms. or fosphenytoin PE injection

234
We could then order a dose of 1000 mg of phenytoin

mixed in 100 mL of normal saline given intrave- Clinical Correlate
nously via controlled infusion. The administration
rate for phenytoin sodium injection should not The Phenytoin Cheat Sheet at the end of the
exceed 50 mg/minute to avoid potential cardio- Maintenance Dose Calculations section is a
concise review of the equations and sequencing
vascular toxicity associated with the propylene
for the three dose calculation methods.
glycol diluent of the phenytoin injection. The accu-
racy of this loading dose estimate can be checked by
obtaining a phenytoin plasma drug concentration
approximately 1 hour after the end of the loading Method 1A (Empiric)
dose infusion. Alternatively, we could give this Multiply SG’s weight of 75 kg by 5 mg/kg/day to get
1000 mg of phenytoin sodium as the fosphenytoin an estimated dose of 375 mg of phenytoin free acid
salt (Cerebyx®) also at a dose of 1000 mg PE (i.e, or 408 mg of phenytoin sodium, which would be
phenytoin sodium equivalents) of fosphenytoin at rounded to 400 mg. This dose of 400 mg/day may
a rate of 150 mg/min. Both doses will deliver the be divided into 200 mg twice daily, if necessary, to
same amount (920 mg) of phenytoin free acid. decrease the likelihood of enzyme saturation and
reduce concentration-dependent side effects. This
Clinical Correlate assumes that the patient has an average Km and Vmax.
Phenytoin sodium injection uses a propylene Method 1B (Population Parameters)

glycol base as its vehicle and will precipitate in
Substitute population estimates for Vmax and Km into
most IV fluids. It is most compatible in normal
the MME and solve for the dose as follows:
saline but can even precipitate in this fluid and
clog an existing inline IV filter if one is being
used. Propylene glycol is a cardiotoxic agent and V ×C
X d × S =max ss
can cause various complications, such as bradycardia K m + C ss
and hypotension. An alternative is to use the newer
fosphenytoin injection, which is compatible with (See Equation 10-1.)
many IV fluids and can also be administered safely at
a faster rate (up to 150 mg/minute). Therefore:
525 mg/day × 15 mg/L

Maintenance Dose Calculations X d (0.92) =
4 mg/L + 15 mg/L
Several methods to calculate maintenance dose are
described, with each method requiring more serum 7875 mg2L/day
drug concentrations and yielding more accurate =
19 mg/L
dosing estimates. Phenytoin dosage adjustments
using these methods are more commonly done in = 415 mg/day of phenytoin free acid
the outpatient setting because they require steady-
state concentrations that may take more than 2 415 mg/day
Xd =
weeks to be attained. 0.92
There are two methods to calculate an initial daily = 451 mg/day of the sodium salt of phenytoin,
maintenance dose (Xd): an empiric method and a
method based on estimating the patient’s Vmax and Km. rounded to 450 mg/day

235
where: be derived mathematically by examining the equa-

Vmax = population estimate of maximum rate of tions used to calculate dose for first- and zero-order
drug metabolism (7 mg/kg/day × 75 kg = models. We will start by rearranging two definitions
525 mg/day), in the first-order model:
Km = population estimate of Michaelis–Menten X0
constant (4 mg/L), C=
ss rearranges to X=
0 C ss ×V
V
Css = desired average steady-state plasma
(See Equation 1-1.)
concentration of 15 mg/L, and
And
S = salt factor (0.92 for phenytoin sodium
capsules).
Clt
= =
Clt VK rearranges to V
K
Note how the units in the equation cancel out,
yielding mg/day as the final units. so, by substituting for V:
This calculation of 415 mg of phenytoin free
C ss × Clt
acid is larger than our empiric estimate of 375 mg/day, X0 = or X 0 × K = C ss × Clt
showing that the empiric method of 5 mg/kg/day K
results in a lower value for SG’s initial phenytoin
maintenance dose. In SG’s case, although this would Css × Clt from our first-order equation can be substi-
equal a daily dose of four 100-mg phenytoin sodium tuted for X0/τ in the zero-order equation derived in the
capsules plus one 50-mg phenytoin tablet (which introduction:
is phenytoin free acid), a patient would usually be V ×C
started on 400 mg/day (200 mg BID) and titrated X 0 /τ = max ss
K m + C ss
up to 450 mg if needed based on plasma phenytoin
drug concentrations.
Problem 1B. When would you recommend that Substituting Css × Clt for X0 /τ yields:
steady-state plasma concentrations be drawn?
It is difficult to calculate when multiple dosing Vmax × C ss
Clt × C ss =
with phenytoin will reach steady state because the K m + C ss
time to steady state is concentration dependent.
With drugs that undergo first-order elimination, Solving for Clt:
steady state can be reached in three to five drug
half-lives because this model assumes that clear- (C ss × Clt × K m ) + (Clt × C ss 2 ) = Vmax × C ss
ance and volume of distribution are constant.
However, because of its capacity-limited metabo- C ss × Clt × K m Clt × C ss 2
+ =
Vmax
lism, phenytoin clearance decreases with increasing C ss C ss
concentration. Therefore, the calculation of time to
reach steady state is quite complicated and cannot (Clt × K m ) + (Clt × C ss ) =
Vmax
be based on half-life. In fact, phenytoin does not
have a true half-life; its half-life is dependent on Clt (K m + C ss ) =
Vmax
drug concentration.
The major factor in determining how long it will This equation can now be rearranged to represent
take to attain steady state is the difference between clearance in terms of Vmax and Css as shown:
Vmax and the daily dose. The closer Vmax is to the dose,
the longer it will take to achieve steady state. This Vmax
Clt =
relationship between Vmax and concentration can K m + C ss

236
where:
Clt = clearance of phenytoin;
Clinical Correlate
Vmax = maximum rate of drug metabolism, usually The t90% equation is a very rough estimate of time
expressed as milligrams per day, to 90% of steady state and should be used only
Km = Michaelis–Menten constant, representing as a general guide. The clinician should check
the concentration of phenytoin at which nonsteady-state phenytoin concentrations before
the rate of this enzyme-saturable hepatic this time to avoid serious subtherapeutic or
metabolism is half of maximum, and supratherapeutic concentrations.
Css = average steady-state phenytoin concentration.
Examination of this equation shows that when Css
In patient SG’s case, t90% is calculated as follows:
is very small compared to Km, Clt will approximate
Vmax/Km, a relatively constant value. Therefore, at K m ×V
= t 90% (2.3×Vmax ) − (0.9 × X d )
low concentrations, the metabolism of phenytoin
(Vmax − X d )2 
follows a first-order process. However, as Css
increases to exceed Km, as is usually seen with thera-
peutic concentrations of phenytoin, Clt will decrease (See Equation 10-4.)
and metabolism will convert to zero order. We can where:
calculate an estimate of the time it takes to get to
Km = 4 mg/L,
90% of steady state using the following equation:
Vmax = 7 mg/kg/day × 75 kg (525 mg/day),
K m ×V
=t 90% (2.3×Vmax ) − (0.9 × X d )
(Vmax − X d )2 
Xd = 415 mg/day of phenytoin (free acid), and

V = 0.65 L/kg × 75 kg (48.8 L).
(See Equation 10-4.) Therefore:
where:
4 mg/L × 48.8 L
t90% = estimated number of days to get to 90% of t 90% =
steady state, (525 mg/day − 415 mg/day)2
Xd = daily dose of phenytoin (in mg/day), (2.3 × 525 mg/day) − (0.9 × 415 mg/day)
V = volume of distribution, = 0.016(834)
V max = maximum rate of drug metabolism (in
= 13.34 days
milligrams per day), and
Km = Michaelis–Menten constant. Note how the units cancel out in this equation,
This equation is derived from a complex integration leaving the answer expressed in days, not hours.
of the differential equation describing the difference This equation estimates that it will take SG approx-
between the rate of drug coming in (i.e., the daily imately 14 days for his phenytoin concentration to
dose) and the rate of drug going out of the body. reach steady state with a desired Css concentration
This equation gives us an estimate of when to draw of 15 mg/L.
steady-state plasma concentrations and is based Close inspection of this calculation illustrates
on the assumption that the beginning phenytoin the impact that the denominator—the difference
concentration is zero. In patients such as SG, who of Vmax and daily dose—has on the time it takes to
have previously received a loading dose, t90% may be reach steady state. For example, if we assume that
different, usually shorter, unless the loading dose SG’s daily dose was 500 mg/day, we can re-solve
yielded an initial concentration greater than that the t90% equation with strikeouts showing the
desired, in which case t90% would be even longer. changes.

237
4 mg/L × 48.8 L again using the MME. This method is preferred over
t 90% = using population estimates for both unknowns. For
(525 mg/day − 500 415 mg/day)2
SG, this calculation is as follows:
(2.3 × 525 mg/day) − (0.9 × 500 415 mg/day)
V ×C
4(48.8) X d × S =max ss
2.3(525) − 0.9(500) K m + C ss
525 − 500 
195.2 (See Equation 10-1.)
= (1207.5 − 450)
252 First, rearrange the MME to isolate Vmax :
= 0.312(757.5) ( X d × S )(K m + C ss )
Vmax =
= 236.6 days C ss
This means that, theoretically, it would now take (415 mg/day)(4 mg/L + 6 mg/L)
Vmax =
approximately 237 days for patient SG to reach (6 mg/L)
steady state on a dose of 500 mg/day. Of course, SG
would actually show signs of toxicity long before he (415 mg/day)(10 mg/L)
=
reached steady state, but this illustrates the effect (6 mg/L)
the difference of dose (and Vmax, similarly) has on the
calculation of time to steady state. In fact, if the daily (4150 mg2 /day × L)
=
dose exceeds Vmax, steady state is never achieved. (6 mg/L)
= 691.67 mg/day, rounded to 690 mg/day
CASE 2 (See Equation 10-1.)

For this case, we use the data presented in Case where:
1 and continue treating patient SG. A phenytoin Vmax = calculated estimate of patient’s Vmax,
plasma concentration (free acid) of 6 mg/L is
drawn 18 days after the beginning of therapy. Km = population estimate of 4 mg/L,
Although SG’s seizure frequency has decreased, Xd × S = patient’s daily dose of phenytoin
he is still having occasional seizures, and his (415 mg/day of free acid), and
physician has decided to adjust his dosing regimen Css = reported steady-state concentration
to attain a plasma concentration of 15 mg/L. (6 mg/L).
So SG’s new estimated Vmax is 690 mg/day, which is

larger than the population estimate of 525 mg/day,
Method 2 (One Steady-State Level)
meaning that he has a greater phenytoin clearance
Problem 2A. Calculate an appropriate dosing regimen than first estimated.
to attain our desired concentration of 15 mg/L. We now take the new Vmax and the MME for our
Now that we have one steady-state concentration, we new dose and use it in the MME to solve for Xd as
can calculate a new maintenance dose for SG. This is follows:
done using the basic MME with two unknowns, Vmax V ×C
and Km. We can use a population estimate for one X d × S =max ss
K m + C ss
of the unknowns (usually Km), solve for the other
unknown, and then recalculate the new dose once (See Equation 10-1.)

238
690 mg/day × 15 mg/L K m ×V

Xd ×S = =t 90% (2.3 ×Vmax ) − (0.9 × X d )
4 mg/L + 15 mg/L (Vmax − X d )2 
10,455 mg2 /day × L

= (See Equation 10-4.)
19 mg/L
Therefore:
= 544.74 mg/day
4 mg/L × 48.8 L
t 90% =
Xd (0.92) = 545 mg/day of phenytoin free acid (690 mg/day − 545 mg/day)2
(2.3 × 690 mg/day) − (0.9 × 545 mg/day)
Xd = 587 mg/day of phenytoin sodium,
rounded to 590 mg/day = 10.18 days for new t 90% to be reached
where:
where:
Xd × S = new dose of phenytoin sodium (S = 0.92),
Km = population estimate (4 mg/L),
Css = desired steady-state concentration of
15 mg/L, Vmax = calculated estimate based on one steady-
state concentration (697 mg/day),
Km = population estimate of 4 mg/L, and
Xd = daily dose of 552 mg/day of phenytoin free
Vmax = calculated estimate (690 mg/day).
acid, and
Therefore, SG’s new dose would be 590 mg/day V = population estimate of volume of
as phenytoin sodium capsules in divided doses, distribution (48.8 L).
which is equivalent to 545 mg of phenytoin free
Note that our new t90% is slightly smaller than the
acid. Because this is a large increase in dose, it may
previous estimate because the difference between
saturate the patient’s hepatic enzymes, causing the
Vmax and dose is now greater.
plasma concentration to increase disproportion-
ately. The practitioner may decide to give a lower
dose initially. Again, note how units cancel out in Method 3 (Two Steady-State Levels)
this equation, yielding mg/day.
Problem 2C. Three weeks later, SG’s plasma
phenytoin concentration is 20 mg/L. He is now
Clinical Correlate seizure free, and his physician wants to adjust his dose
to get his plasma concentration back to 15 mg/L. Note
Phenytoin doses are usually increased by that SG states that he has been taking his phenytoin
25–100 mg/day. Because the clinical accuracy exactly as prescribed. What dose would you now
using this first method is not as accurate and recommend to achieve a plasma phenytoin concen-
predictive as those for the aminoglycosides and
tration of 15 mg/L?
theophylline, good clinical judgment is required
when recommending a dose. Now that we have measured two different steady-
state concentrations at two different doses, we can
make an even more accurate dosing change.
Problem 2B. When should a plasma phenytoin As shown in Lesson 14, clearance can be expressed
concentration be drawn? as Xd/Css, resulting in the plot in Figure 15-5.
We can recalculate when SG’s phenytoin concen- We can now plot both steady-state doses (X1
tration will reach steady state on this new dose. of 415 mg/day and X2 of 545 mg/day) on the
Remember, time to 90% of steady state (t90%) is y-axis and both steady-state Xd/Css values (X1/C1 =
dependent on plasma drug concentration. SG’s new 415/6 L/day and X2/C2 = 545/20 L/day) on the
estimate of t90% is calculated as follows: x-axis, thus linearizing these relationships. This

239
FIGURE 15-5.
Relationship of daily dose to clearance.
allows us to express the relationship in the alge-
braic form for a straight line, Y = mX + b, where m,
the slope of the line, equals the negative value of
the patient’s Km (i.e., m = –Km) and the y-intercept is
the patient’s Vmax. For SG, this graph is drawn as in
Figure 15-6.
The slope of the line, which represents –Km, can
now be calculated as follows: FIGURE 15-6.
Relationship of daily dose to the dose divided by steady-state
X − X2 concentration achieved.
−K m =1
X1 X2
−
C1 C 2 where:
Xd × S = either of the doses SG received (415 or
= 545 mg/day of free acid),
415 mg/day 545 mg/day
− Css = steady-state concentration at the dose
6 mg/day 20 mg/day
selected, and
K m = 3.10 mg/L Km = calculated value of 3.10 mg/L.

Using the 415-mg/day dose, Css = 6 mg/L.
Clinical Correlate Vmax =

(415 mg/day)(3.10 mg/L + 6 mg/L)
It is best to use the free acid amount of 6 mg/L
phenytoin, not the sodium salt, when calculating
= 629.42 mg/day
Km in the above equation and in all subsequent
calculations.
Using the 545-mg/day dose, Css = 20 mg/L.
Next, we substitute this new value for Km into (545 mg/day)(3.10 mg/L + 20 mg/L)
the MME and solve for an even more accurate Vmax Vmax =
20 mg/L
than calculated previously, as follows:
= 629.48 mg/day
( X × S )(K m + C ss )
Vmax = d
C ss which we will round to 630 mg/day. Note that either
set of doses and concentrations will give the same
(See Equation 10-1.) Vmax.

240
Finally, we substitute our new Vmax of 630 mg/day

and our calculated Km of 3.10 mg/mL into the MME Clinical Correlate
and solve for Xd as follows:
It is important to dose phenytoin correctly so side
V ×C effects do not occur. Dose-related side effects at
X d × S =max ss serum concentrations greater than 20 mcg/mL include
K m + C ss
nystagmus, whereas concentrations greater than
30 mcg/mL may result in nystagmus and ataxia.
(See Equation 10-1.) Concentrations greater than 40 mcg/mL may produce
ataxia, lethargy, and diminished cognitive function.
(630 mg/day)(15 mg/L) Adverse effects that may occur at therapeutic
X d (0.92) = concentrations include gingival hyperplasia, folate
3.10 mg/L + 15 mg/L
deficiency, peripheral neuropathy, hypertrichosis,
X d (0.92) = 522 mg/day of phenytoin free acid and thickening of facial features.
X d = 567.5 mg/day of phenytoin sodium

Long-Term Phenytoin Monitoring
where: Repeated use of Method 3 (above) allows for
continued dosing adjustments over many years in
Vmax = 630 mg/day of phenytoin free acid, patients maintained on phenytoin; this is especially
useful as children get older and bigger. Any two sets
Km = 3.10 mg/L, of drug concentrations and different dose pairs can
Css = desired average steady-state plasma be used to calculate an adjusted dose, which makes
concentration of 15 mg/L, and for interesting math when a patient has dozens of
concentrations over many years. Compliance should
S = salt factor (0.92 for phenytoin sodium always be assessed before trusting these dosage
capsules). adjustment calculations.
Therefore, we could give SG 560 mg (five 100-mg
phenytoin capsules plus two 30-mg phenytoin Phenytoin Cheat Sheet
capsules) daily in divided doses. This would The Phenytoin Cheat Sheet contains the equations
give slightly less than the calculated amount of and sequencing for the three dosing methods detailed
567.5 mg/day. in the Maintenance Dose Calculations Section above.
Equations are presented for calculating doses when
you have no phenytoin serum drug concentration,
one phenytoin serum drug concentration, and, finally,
two phenytoin serum drug concentrations obtained
on two different doses.

241
PHENYTOIN DOSING CHEAT SHEET

DOSING METHOD 1A: DOSING METHOD 3: TWO STEADY-STATE LEVELS ON TWO
Use 5 mg/kg/day DIFFERENT DOSES
Use after you have two steady-state phenytoin

DOSING METHOD 1B: concentrations from two different phenytoin doses. You
Use population estimates for the Michaelis–Menten can now work Equation 3 to solve for an even better
values for Km of 4 mg/L and Vmax of 7 mg/kg/day and value for Km (shown below).
solve the general MME formula as shown below: Use this better Km value to once again re-solve for a better
Vmax value than used in Method 2. Once you get new Vmax
Equation 1 MME: (i.e., real Km and Vmax), re-solve Equation 1 (MME) again for
dose.
V (C )
X d × S = max ss (desired)
(K m + C ss (desired) )
Equation 3:
Km estimate = 4 mg/L
Vmax estimate = 7 mg/kg/day Use this to solve for real Km :
S = Salt form factor (either 1.0 or 0.92, or 0.66) X = dose
C = concentration
DOSING METHOD 2: ONE STEADY-STATE LEVEL
Use this method as in Equation 2 below after you have one To solve for real Vmax:
steady-state phenytoin serum drug concentration to solve Use Equation 2 again.
for Vmax while still using the population parameter for Km of
4 mg/L. To solve for dose:
After solving for this better value for Vmax, use it plus the old Use Equation 1 (MME) again.
Km value in the MME to re-solve for dose, as shown below.
To solve for Vmax: To solve for real Vmax:
Use Equation 2 once again.
Equation 2:
( X d × S )(real K m + C ss (lab) )
This is simply a rearrangement of MME (Equation 1): real Vmax =
C ss (lab)
( X d × S )(K m + C ss (lab) )
Vmax =
C ss (lab) To solve for dose:
Use Equation 1 again.
To solve for dose:
Use Equation 1 again.
Equation 1 (MME) yet again:
Equation 1 (MME) again:
real Vmax (C ss ( desired) )
better Vmax (C ss (desired) ) best X d × S =
Xd ×S = (real K m + C ss ( desired) )
(population parameter for K m + C ss (desired) )
Km estimate = 4 mg/L real Km (mg/L)
Vmax estimate = as solved for mg/kg/day real Vmax (mg/kg/day)

S = Salt form factor (either 1.0 or 0.92, or 0.66) S = Salt form factor (either 1.0 or 0.92, or 0.66)

242
Digoxin indicated or not effective in a given patient.2 The

current recommended loading dose of digoxin for
Digoxin is an inotropic medication that may be ventricular rate control is 0.25 mg IV every 2 hours
useful in the treatment of patients with heart up to a total of 1.5 mg. For PSVT, the loading dose
failure (HF) to decrease hospitalizations due to this is 8–12 mcg/kg given as follows: one-half of the
disorder. Desired serum concentrations in these dose is given over 5 minutes; 25% is given in 4 to
patients range from 0.5 to 0.9 ng/mL.1 This conser- 8 hours; and the final 25% given 4 to 8 hours later.3
vative target has been associated with a decline in Electrocardiogram monitoring is performed during
the overall incidence of digoxin toxicity. Notice that loading dose administration to assess for toxicity.4
the units of plasma concentrations for digoxin are The recommended maintenance dose of digoxin for
different (nanograms per milliliter) from those of ventricular rate control is 0.125 to 0.375 mg per
other commonly monitored drugs (usually milli- day. With declining renal function, these doses may
grams per liter). be further reduced or administered as alternate day
Digoxin may also be used in the management therapy.3
of arrhythmias such as atrial fibrillation (AF) and Steady-state volume of distribution (Vss) of
atrial flutter. It is effective in controlling heart rate at digoxin is large and extremely variable. Differences
rest in patients with AF and may be used in patients in renal function account for some of the inter-
with concomitant heart failure, left ventricular patient variation.
dysfunction, and a sedentary lifestyle. Digoxin is an
acceptable treatment for slowing a rapid ventricular 15-2 Vss = 4 to 9 L/kg ideal body weight (IBW)
response and improving left ventricular function (average adult, 7 L/kg IBW)4
in patients with acute myocardial infarction and
AF associated with severe left ventricular dysfunc- When calculating an oral digoxin dose, the
tion and HF. In combination with a beta-blocker or bioavailability (F) of the dosage form used must be
nondihydropyridine calcium channel antagonist, considered. For patients with normal oral absorp-
digoxin may be used to control heart rate during tion, digoxin tablets are 50–90% (average, 70%)
exercise in patients with AF, as well as in the preg- absorbed (F = 0.7), and digoxin elixir is 75–85%
nant patient with AF. Digoxin may also be used to absorbed (average, F = 0.8).
terminate paroxysmal supraventricular tachycardia Systemic clearance (Clt) of digoxin can be calcu-
(PSVT).2,3 lated as follows5 :
In patients with normal left ventricular func-
tion, digoxin is less effective for ventricular rate 15-3 Clt = (1.303 × CrCl) + Clm
control than calcium channel blockers or beta-
blockers. However, it may be used in combination where:
with these agents in patients with less than satisfac- Clm = metabolic clearance.
tory ventricular rate control from monotherapy.3
= 40 mL/min for patients with no or mild HF
Doses of digoxin are usually administered
= 20 mL/min for patients with moderate to
orally or intravenously. Loading doses are no
severe HF
longer recommended in HF patients.1 Although
digoxin loading doses have been used extensively Several methods have been proposed for calculating
in patients with AF, atrial flutter, and PSVT, other doses of digoxin that have been described in detail
drugs are more effective and/or have a more rapid elsewhere. Recent investigators have developed
onset of action. Therefore, digoxin loading doses are a nomogram for determining digoxin doses that
rarely needed unless alternative therapy is contra- achieve lower serum concentrations.6

243
CASE 3 We have previously established that the desired

steady-state serum digoxin concentration in the
TS is a 72-year-old, 5' 5'', 125-lb female who patient with HF ranges from 0.5 to 0.9 ng/mL. We
has a diagnosis of HF for which she currently will choose a level of 0.8 ng/mL for TS.
receives a beta-blocker, an angiotensin- To determine systemic clearance, we must first
converting enzyme inhibitor (ACEI), and a diuretic. estimate TS’s creatinine clearance. We use the Cock-
She has been hospitalized three times in the past croft–Gault equation:
year for her HF; at this time, her physician wishes
to initiate digoxin therapy. Her current serum (140 − age)(IBW)
creatinine is 1.00 mg/dL. CrCl(female) = (0.85)
72 × SCr
Problem 3A. Calculate a maintenance dose of (See Equation 9-1.)

digoxin tablets to be given to TS to achieve a satis- where:
factory steady-state serum digoxin level.
CrCl = creatinine clearance, in milliliters per
minute;
Clinical Correlate IBW = ideal body weight, in kilograms; and
Patients with HF usually do not require a loading SCr = serum creatinine, in milligrams per
dose of digoxin before initiating maintenance deciliter.
dose therapy.1
Therefore:
(0.85)(140 − 72)(57 kg)
CrCl(female) =
The relationship between the steady-state plasma 72 × 1.00
concentration, maintenance dose, and total systemic
clearance is shown below: = 46 mL/min
X d × 106 × F Total body clearance of digoxin would be:

15-4 C ss =
Clt × τ Clt = (1.303 × CrCl) + Clm (See Equation
15-3.)
(See Equation 4-3.)
The above equation can be rearranged and = (1.303 × 46 mL/min) + 20 mL/min
written as follows:
= 80 mL/min
C × Cl × τ
X d = ss 6 t The daily maintenance dose required to achieve
10 × F
a steady-state concentration of 0.8 ng/mL would be:
where:
C ss × Clt × τ
Xd = maintenance dose of digoxin, in milligrams Xd =
per day, 106 × F
Css = steady-state plasma concentration, in 0.8 ng/mL × (80 mL/min) × 1440 min
nanograms per milliliter, Xd =
106 ng/mg × 0.7
Clt = total body clearance,
τ = dosing interval, in minutes (1440 minutes =
= 0.13 mg
1 day), (See Equation 15-4.)

106 = conversion from nanograms to milligrams Therefore, TS should receive 0.125 mg of digoxin
(i.e., 106 ng = 1 mg), and daily to achieve a steady-state digoxin concentra-
F = 0.7 for tablets. tion of slightly less than 0.8 ng/mL.

244
Problem 3B. Two months later, TS has a steady- This dose can be achieved by alternating 0.25 mg
state serum digoxin level drawn. The laboratory with 0.125 mg every other day. This would be the
reports this value as 0.52 ng/mL. Although this equivalent of administering 0.1875 mg per day.
value is within the thearapeutic range for HF, TS’s
Problem 3C. TS begins her new digoxin regimen.
physician desires to increase the dose to achieve
Three months later she reports to her physician’s
a slightly higher serum concentration. Calculate a
office complaining of nausea and vomiting. A serum
new maintenance dose for TS that will achieve a
digoxin level (10 hours after her last dose) and
serum concentration of 0.8 ng/mL.
serum creatinine are drawn. Her digoxin concen-
The first step to solving this problem is to calculate tration is 1.6 ng/mL and her serum creatinine has
TS’s actual serum digoxin clearance. We can do this risen to 1.92 mg/dL. Her physician tells JS to hold
as follows: her digoxin for the next two days and come back to
the office to have her serum digoxin concentration
X d × 106 × F
Clt = repeated. Her serum level (48 hours after the last
C ss × τ serum level was drawn) is now 1.1 ng/mL. Calculate
a new digoxin dose that will achieve a steady-state
where: serum concentration of 0.8 ng/mL.
Clt = total body clearance, TS appears to now be experiencing declining renal
Xd = maintenance dose of digoxin, in milligrams function (rise in serum creatinine). To determine a
per day, new dose, we must first calculate her actual Clt . As
before, we can do this by rearranging the following
Css = steady-state plasma concentration, in
equation:
nanograms per milliliter,
τ =
1440 minutes (1 day),
X d × 106 × F
C ss =
106 = conversion from nanograms to milligrams
Clt × τ
(i.e., 106 ng = 1 mg), and (See Equation 15-4.)
F = bioavailability (0.7 for digoxin tablets). to:
Plugging in our values for patient TS: X d × 106 × F
Clt =
C ss × τ
0.125 mg × 106 ng/mg × 0.7
Clt =
0.52 ng/mL × 1440 min where:
Clt = total body clearance,
= 117 mL/min
Xd = maintenance dose of digoxin, in milligrams
From this we see that TS’s total body clearance of per day,
digoxin is slightly higher than the value we calcu- Css = steady-state plasma concentration, in
lated using population estimates. nanograms per milliliter,
Now, we can use this clearance value to calcu- τ =
1440 minutes (1 day),
late a new maintenance dose to achieve our desired 106 = conversion from nanograms to milligrams
serum concentration of 0.8 ng/mL. (i.e., 106 ng = 1 mg), and
C ss × Clt × τ F = bioavailability (0.7 for digoxin tablets).
Xd =
106 × F Plugging in our values for patient TS:
0.8 ng/mL × (117 mL/min) × 1440 min 0.1875 mg × 106 ng/mg × 0.7
Xd = Clt =
106 ng/mg × 0.7 1.6 ng/mL × 1440 min
= 0.19 mg = 57 mL/min

245
In making this determination, it is important that we Now, we can solve for time t:
use the average daily dose TS is receiving as well as
Clevel 2(steady state) = Clevel 1(steady state)e–Kt
the serum value resulting from this dose (and not the
serum value reported 2 days later with doses held). 0.8 ng/mL = 1.1 ng/mL e–0.008t
It is of interest to note a significant decline in digoxin
total body clearance with declining renal function. 0.727 = e–0.008t
Then we use patient TS’s actual Clt to calculate
the appropriate dose to achieve our desired Css of ln 0.727 = ln e–0.008t
0.8 ng/mL.
–0.319 = –0.008t
C × Cl × τ
X d = ss 6 t 39.9 hr = t
10 × F
0.8 ng/mL × (57 mL/min) × 1440 min So we need to wait another 40 hours before we
Xd = begin TS’s new digoxin maintenance dose.
106 ng/mg × 0.7
= 0.094 mg/day
This dose can be achieved by alternating 0.125 mg

CASE 4
with 0.0625 mg (one-half of a 0.125-mg tablet) HK is a 56-year-old, 6' 6'' tall, 200-lb patient with
every other day. HF. He is currently receiving a beta-blocker and
an ACEI. His physician wishes to add digoxin to
Problem 3D. How much longer do we need to wait this regimen. His current serum creatinine is
until we can begin TS’s new digoxin maintenance 1.1 mg/dL.
dose?
Since TS’s latest serum digoxin concentration is
elevated (1.1 ng/mL), we cannot begin her new Problem 4A. Calculate a maintenance dose of digoxin
maintenance dose until this value decreases to tablets that will achieve a steady-state serum concen-
approximately 0.8 ng/mL. To calculate the amount tration of 0.7 ng/mL for HK.
of time that must elapse until this occurs, we can The first step in solving this problem is to determine
use the following equation: HK’s total body clearance for digoxin. To determine
this, we must first estimate his creatinine clearance.
(140 − age)(IBW)
where: CrClmale =
72 × SCr
Clevel 2(steady state) = the serum concentration we desire
before the new maintenance dose is (See Equation 9-1.)
started (0.8 ng/mL), where:
Clevel 1(steady state) = the serum concentration the patient CrCl = creatinine clearance, in milliliters per
currently has (1.1 ng/mL), minute,
K = the elimination rate constant, and IBW = ideal body weight, in kilograms, and
t = the time we must wait until SCr = serum creatinine, in milligrams per
Clevel 2(steady state) is reached. deciliter.
We can calculate K as follows:
Therefore:
In 1.1− In 1.6 (140 − 56)(91 kg)
K = CrClmale =
48 hr 72 × 1.1
= 0.008 hr −1 = 97 mL/min

246
Total body clearance of digoxin would be: In problem 3C, we encountered a similar situation
in which we solved for the time to wait before an
Clt = (1.303 × CrCl) + Clm (See Equation 15-3.) elevated serum concentration declined to an accept-
able value with doses held. In that situation, we had
= (1.303 × 97 mL/min) + 40 mL/min two steady-state serum concentrations to solve for
a K value. In the current problem, we will address
= 166.4 mL/min
how we can estimate a K value and, therefore, time
The daily maintenance dose required to achieve to wait, with only one steady-state serum concen-
a steady-state concentration of 0.7 ng/mL would be: tration available.
The first step to solving this problem is to calculate
C ss × Clt × τ
Xd = HK’s actual serum digoxin clearance. We can do this
106 × F as follows:
0.7 ng/mL × (166.4 mL/min) × 1440 min X d × 106 × F
Xd = Clt =
106 ng/mg × 0.7 C ss × τ
= 0.24 mg where:
Clt = total body clearance;
Xd = maintenance dose of digoxin, in milligrams
Therefore, HK should receive 0.25 mg of digoxin per day;
daily.
Css = steady-state plasma concentration, in
Problem 4B. Suppose HK had to initially receive his nanograms per milliliter;
daily digoxin maintenance dose by IV administra- τ = 1440 minutes (1 day);
tion. Calculate this dose. 106 = conversion from nanograms to milligrams
C ss × Clt × τ (i.e., 106 ng = 1 mg); and
Xd = F = bioavailability (0.7 for digoxin tablets).
106 × F
0.7 ng/mL × (166.4 mL/min) × 1440 min Plugging in our values for patient HK:
Xd =
106 ng/mg × 1 0.25 mg × 106 ng/mg × 0.7
Clt =
= 0.17 mg 1.2 ng/mL × 1440 min
= 101 mL/min
Notice F =1 for intravenously administered drugs.
Step 2 to solving this problem is to use the equation
below to calculate time to wait:
Clinical Correlate
IV administration of digoxin should be given by
slow IV push. This method of administration where:
prevents the propylene glycol contained in this Clevel 2(steady state) = the serum concentration we desire
formulation from causing cardiovascular collapse. before the new maintenance dose is
started (0.7 ng/mL),
Clevel 1(steady state) = the serum concentration the patient
Problem 4C. HK is currently receiving digoxin
currently has (1.2 ng/mL),
0.25 mg orally daily. He has a steady-state serum
K = the elimination rate constant, and
digoxin level reported as 1.2 ng/mL. If all doses
are held, predict how long it will take for his serum t = the time we must wait until
concentrattion to fall to 0.7 ng/mL. Clevel 2(steady state) is reached.

247
To be able to use this equation requires that we So we must wait an additional 57 hours for HK’s
know the value for K. serum digoxin level to drop to 0.7 ng/mL. Before
We can estimate this from the following equa- initiating a new maintenance dose, it would be
tion: prudent to repeat a serum digoxin level to ensure
his elimination rate has not changed during this
Cl waiting period and that his serum concentration is
K =
V an acceptable value.
We can estimate V as 7 L/kg IBW. (See Equation
15-2.) References
V = 7 L/kg x 91 kg 1. Yancy CW, Jessup M, Bozkurt B, et al. 2013
= 637 L ACCF/AHA guidelines for the management of
heart failure: A report of the ACC Foundation/
Now, we can estimate K. AHA Task Force on Practice Guidelines. J Am Coll
K is in units of hr −1 Cardiology 2013 (June). http://content.online-
V is in units of liters jacc.org
2. Anderson JL, Halperin JL, Albert NM, et al.
Cl therefore must be converted to units of liters per Management of patients with atrial fibrillation
hour (L/hr): (A compilation of 2006 ACCF/AHA/ESC and
2011 ACCF/AHA/HRS recommendations): A
101 mL/min x 60 min/hour = 6060 mL/hr report of the American College of Cardiology/
6060 mL/hr divided by 1000 mL/L = 6.06 L/hr American Heart Association Task Force on Prac-
tice Guidelines. Circulation 2013; 127:1916–26.
Cl http://circ.ahajournals.org
K =
V 3. Tisdale JM. Arrhythmias. In: Chisholm-Burns
6.06 L/hr MA, Wells BG, Schwinghammer TL, et al., eds.
= Pharmacotherapy: Principles and Practice, 3rd
637 L
ed. New York: McGraw Hill; 2011, pp. 169–96.
= 0.0095 hr −1 4. Sanoski CA, Bauman JL. The arrhythmias. In:
DiPiro JT, Talbert RL, Yee GC, et al., eds. Phar-
Using the equation:
macotherapy: A Pathophysiologic Approach. 8th
Clevel 2(steady state) = Clevel 1(steady state)e–Kt ed. New York: McGraw Hill; 2011. http://www.
accesspharmacy.com
0.7 ng/mL = 1.2 ng/mL e−0.0095t 5. Koup JR, Jusko WJ, Elwood CM, et al. Digoxin
pharmacokinetics: role of renal failure in
0.583
= e−0.0095t dosage regimen design. Clin Pharmacol Ther
1975;18:9–21.
ln 0.583 = ln e−0.0095t
6. Bauman JJ, DiDomenico RJ, Viana M, et al. A
−0.54 = −0.0095t method of determining the dose of digoxin for
heart failure in the modern era. Arch Intern Med
56.8 hr = t 2006;166:2539–45.

248
Discussion Points
Phenytoin D-5. A 41-year-old female, 5' 6'' and 148 lbs,

presents to the emergency depart-
D-1. Suppose SG in Case 1, Problem 1A was 65 ment with uncontrolled seizures (serum
years old, weighed 95 kg, and had a serum creatinine, 1.2 mg/dL; serum albumin,
albumin level of 2.4 mg/L. What would be 4.3 g/dL; white blood cell count 18,300/mm3,
his oral maintenance dose of phenytoin receiving phenytoin 300 mg daily at home).
based on these changes? Assuming that phenytoin 200 mg every 12
hours orally is initiated at 8 a.m. on 12/1,
D-2. Based on your calculations in Discussion
describe in detail the process for how you
Point 1, calculate a new maintenance dose
determine when serum levels (and what
for SG that would result in a steady-state
type of levels) should be obtained. Then
plasma concentration of 15 mg/L.
write an order as it would appear in the
D-3. The laboratory reports a serum phenytoin Physician’s Order section of the patient’s
concentration of 19 mg/L from the dose you medical record for how serum levels should
calculated in Discussion Point 2. Calculate a be obtained. This order should be gram-
new dose that will result in a serum concen- matically correct, include only approved
tration of 15 mg/L (i.e., use Method 2). abbreviations, and provide sufficient detail
that nursing services can easily follow your
D-4. Based on your experience in the provision instructions without having to contact you
of direct patient care, design a pharmacy- for further clarification.
managed phenytoin dosing protocol that
could be used in your practice setting.
This protocol should be written from Digoxin
the standpoint that the pharmacist is
providing complete dosing and monitoring D-6. Suppose TS’s serum digoxin concentration in
of phenytoin in a patient case (instead of Problem 3B had been 1.1 ng/mL. What main-
simply providing recommendations to a tenance dose would be required to achieve a
physician to manage). All steps required serum concentration of 0.8 ng/mL?
(including equations used) to effectively
D-7. Explain how to administer an appropriate
dose and monitor a patient for whom
digoxin loading dose to a patient with atrial
phenytoin is prescribed should be included.
fibrillation.
Describe in detail how you would monitor
this drug using serum concentrations. Write
the order for this drug as it would appear in
the Physician’s Order section of the patient’s
medical record.

APPENDIX A
Basic and Drug-Specific
Pharmacokinetic Equations
Basic Pharmacokinetic Equations
Equation Showing the Relationship of Drug Concentration (mg/L),

Drug Dose (mg), and Volume of Distribution (Liters)
amount of drug in body
1-1 concentration =
volume in which drug is distributed
X
C=
V
(See p. 10.)
Equation for Calculating Total Body Clearance
2-1 Clt = Clr + Clm + Clb + Clother
(See p. 23.)
Equation for Calculating Organ Clearance of a Drug

C −C
2-2 Clorgan =
Q × in out or Clorgan =
QE
C in
(See p. 24.)
Elimination Rate Constant (K) for First-Order,

One-Compartment Model
In C − In C 2
3-1 slope =−K = 1
t1 − t0
(See p. 34.)
or:

C
In 1
C
−K = 2
t1 − t0
249
250
Concentration at Any Given Time, Based or:

on a Previous Concentration (C0) and K dose administered
drug clearance =
for First-Order, One-Compartment Model AUC
3-2 C = C0e–Kt or:
(See p. 34.) initial concentration (C 0 )
where:
AUC =
elimination rate constant (K )
C = plasma drug concentration at time = t,
C0 = plasma drug concentration at time = 0, Accumulation Factor When Not at
K = elimination rate constant, Steady State for a One-Compartment,
t = time after dose, and First-Order Model
e–Kt
= percent or fraction remaining after
(1− e −nK τ )
time (t). 4-1 accumulation factor =
(1− e −K τ )
Note: used often to calculate Cpmin from Cpmax.
(See p. 52.)
Calculation of T½ from K, or K from T½
for First-Order, One-Compartment Model
Accumulation Factor When at Steady
0.693 State for a One-Compartment,
3-3 T 1 =
2
K First-Order Model
(See p. 36.) 1
4-2
or: (1− e −K τ )
0.693
K =
T 12 (See p. 56.)
Mathematical Relationship Between Calculation of Average Drug

Systemic Clearance (Clt ) to Both V and K Concentration from AUC and Dosing
for First-Order, One-Compartment Model Interval or from Dose/Cl
3-4 Clt /V = K AUC
C =
τ
(See p. 37.)
or: and because:
Clt = V × K or V = Clt /K dose

AUC =
drug clearance
Calculation of Area Under the Plasma dose
Drug Concentration Curve (AUC) and Its C =
drug clearance × τ
Relationship to Both Drug Clearance
(K × V ) and Dose Administered or:
dose administered dose

3-5 AUC = 4-3 C =
drug clearance Clt × τ
(See p. 38.) (See p. 57.)

Appendix A | Basic and Drug-Specific Pharmacokinetic Equations
251
Cockcroft–Gault Equations for Calculation of Km, the Michaelis

Calculating Creatinine Clearance (CrCl) Constant (mg/L), Representing the Drug
in Men and Women Concentration at Which the Rate of
Elimination Is Half the Maximum Rate
(140 − age)IBW (Vmax) for Zero-Order (i.e., Nonlinear) Model
9-1 CrClmale =
72 × SCr
dose − doseincreased
10-2 slope = − K m = initial
(See p. 141.) dose/C initial − dose/C increased
or:
(See p. 152.)
(0.85)(140 − age)IBW
CrClfemale =
72 × SCr Calculating Steady-State Concentration
from Estimates of Km, Vmax, and Dose
Note: IBW = ideal body weight.
(Rearrangement of the MME) for
Zero-Order (i.e., Nonlinear) Model
Equations for Estimating IBW
in Men and Women
K m (daily dose)
IBWmales = 50 kg + 2.3 kg for each inch
9-2
10-3 C=
Vmax − daily dose
over 5 feet in height
(See p. 153.)
(See p. 141.)
IBWfemales = 45.5 kg + 2.3 kg for each inch

over 5 feet in height Aminoglycoside Dosing Equations
Adjusted Body Weight (AdjBW) Equation Calculation of Population Estimates

for Patients Whose Total Body Weight for K Based on CrCl
(TBW) Is More Than 35% Over Their IBW
12-1 K = 0.00293 (CrCl) + 0.014
9-3 AdjBW = IBW + 0.4(TBW – IBW)
(See p. 183.)
(See p. 141.)
Calculation of Population Estimates
Michaelis–Menten Equation (MME) for Volume of Distribution (V ) Based
VmaxC on Body Weight or AdjBWAG
10-1 daily dose =
Km +C
12-2 V = 0.24 L/kg (IBW)
(See p. 152.)
or: (See p. 183.)
daily dose (K m + C ) =
VmaxC or:

daily dose (K m ) + daily dose (C ) =
VmaxC V = 0.24 L/kg AdjBWAG
daily dose=
(C ) VmaxC − daily dose (K m ) where:
Note: relates Vmax , Km , plasma drug concentration, 12-3 AdjBWAG = IBW + 0.1(TBW – IBW)
and daily dose (at steady state) for zero-order (i.e.,
nonlinear) model. (See p. 183.)

252
Calculation of Best Dosing Interval (τ) or:

Based on Desired Peak and Trough Ctrough(steady state) = Cpeak(steady state)e–Kt ′
Concentrations
(See p. 34 and Equation 3-2.)
1
τ= (In C trough (desired) − In C peak (desired) ) + t where t′ = τ – time of infusion (t), or the change in
12-4
−K time from the first concentration to the second.
(See pp. 186 and 194.) Calculation of Loading Dose Based

on Initial Calculated Maintenance Dose
where t is the duration of the infusion in hours.
and Accumulation Factor
Note: should be rounded off to a practical dosing
K0
interval such as every 8 hours, every 12 hours, etc. 12-5 loading dose =
(1− e −K τ )
Calculation of Initial Maintenance (See p. 189.)
Dose (K0) Based on Estimates of where:
K, V, Desired Cpeak, and τ K0 = estimated maintenance dose,
1/(1 – e–Kτ) = accumulation factor at steady state, and
−K t
K 0 (1− e )
5-1 C peak (steady state) = τ = dosing interval at which estimated
VK (1− e −K τ ) maintenance dose is given.
(See p. 74.) Calculation of Patient-Specific

where: (i.e., Actual) K Based on Two Drug
Concentrations and Dosing Interval
Cpeak(steady state) =
desired peak drug concentration at
steady state (milligrams per liter), In C trough − In C peak
3-1 K =
K0 = drug infusion rate (also main- τ −t
tenance dose you are trying to (See p. 34.)
or:
estimate for aminoglycosides, in In C peak − In C trough
−K =
liters), τ −t
K
= elimination rate constant (popula- Remembering a rule of logarithms:
tion estimate for aminoglycosides, ln a – ln b = ln (a/b), we can simplify this
in reciprocal hours), equation for hand-held calculators:
t = duration of infusion (hours), and
C 
τ = desired or most appropriate dosing In  trough 
C 
K =− 
peak 
interval (hours).
τ −t
Calculation of Ctrough Concentration or:
Expected from Dose (K0) and C 
Dosing Interval Used (τ) In  peak 
C 
−K = 
trough 
3-2 C = C0e–Kt τ −t
(See p. 34.) Either equation may be used to calculate K.

253
Calculation of Patient-Specific (i.e., where:

Actual) V Based on Actual K, and Dose Cpeak(steady state) = desired steady-state Cpeak;
(K0), τ, and Two Drug Concentrations K0 = drug infusion rate (also adjusted
maintenance dose you are trying to
5-1
K 0 (1− e −K t ) calculate, in milligrams per hour);
C peak (steady state) =
VK (1− e −K τ ) V = actual volume of distribution deter-
mined from patient’s measured Cpeak
(See p. 74.)
and Ctrough values, in liters;
where: K = actual elimination rate constant
Cpeak(steady state) = Cpeak measured at steady state, calculated from patient’s measured
K0 = maintenance dose infused at time Cpeak and Ctrough values, in reciprocal
Cpeak and Ctrough were measured, hours;
V = patient’s actual volume of t = infusion time, in hours; and
distribution that you are trying to τ = adjusted dosing interval rounded to
determine based on Cpeak and a practical number.
Ctrough values,
K = elimination rate constant calculated Calculation of New Expected
from patient’s Cpeak and Ctrough values, Ctrough(steady state) That Would Result from New
t
= duration of infusion (hours), and Maintenance Dose and Interval Used
τ = patient’s dosing interval at time C trough (steady state) = C peak (steady state) e − Kt ′
Cpeak and Ctrough were measured.
Calculation of Actual (i.e., New) Dosing where K is actual patient-specific K.
Interval Based on Patient-Specific Value
for K Calculation of Time to Hold Dose When
Actual Ctrough from Laboratory Is Too High
1
12-4 τ= (In C trough (desired) − In C peak (desired) ) + t C trough (steady state)(desired) = C trough (steady state) e − Kt ′
−K
where t′ is the amount of time to hold the dose after
(See pp. 186 and 194.) the end of the dosing interval.
where t is the duration of infusion in hours and K is Next, take the natural log of both sides:
the actual elimination rate calculated from patient’s
number = number (t ′) and then simply solve for t ′,
peak and trough values.
which is now not an exponent.
Calculation of Patient-Specific or Average Dose for Gentamicin or

Adjusted Maintenance Dose (K0) Tobramycin When Given as an
Based on Actual Values for K and V Extended-Interval (i.e., Once Daily)
Dose Based on Actual Body Weight
K 0 (1− e −K t )
5-1 C peak (steady state) =
VK (1− e −K τ ) X0 = 5.1 mg/kg actual body weight or
adjusted body weight if IBW exceeds
(See p. 74.) actual weight by ≥ 35%

254
Vancomycin Dosing Equations t = duration of infusion (usually 1 or 2

hours for vancomycin),
K = estimated elimination rate
Calculation of Population Estimate
constant,
for K Based on CrCl
V = estimated volume of distribution,
13-2 K = 0.00083 hr [CrCl (in mL/min)] + 0.0044
–1
t′ = time between end of infusion and
collection of blood sample (usually
(See p. 204.)
2 hours) (inclusion of t′ is different
from the calculation for aminogly-
Calculation of Population Estimate for cosides because sampling time for
Volume of Distribution (V) Based on vancomycin is often at least 4 hours
TBW after the beginning of the infusion),
13-1 V = 0.9 L/kg TBW and
τ = desired dosing interval, as
(See p. 204.) determined above.
Note that, unlike the aminoglycosides, it is
recommended that TBW be used to calculate the Calculation of Ctrough Concentration
volume of distribution. Expected from Dose (K0) and Dosing
Interval Used (τ)
Calculation of Best Dosing Interval (τ)
Based on Desired Peak and Trough 13-5 Ctrough = Cpeak(steady state)e–Kt ′′
Concentrations
1
13-4 τ = In C − In C peak (desired)  + t + t ′ where t′′ is the difference in time between the two
−K  trough (desired)
plasma concentrations.
where: Calculation of Patient-Specific
t = duration of infusion (usually 1 or 2 hours (i.e., Actual) K Based on Two Drug
for vancomycin) and Concentrations and Dosing Interval
t′ = time between end of infusion and collection In C trough − In C peak
of blood sample (usually 2 hours). K =−
τ −t −t′
Calculation of Initial Maintenance (See Equation 3-1.)
Dose (K0) Based on Estimates of
K, V, Desired Cpeak, τ, and t Calculation of Patient-Specific (i.e.,
Actual) V Based on Actual K from Two
K 0 (1− e −K t ) −K t ′ Drug Concentrations, Dose (K0), and τ
13-3 C peak (steady state) = e
VK (1− e −K τ )
K 0 (1− e − K t ) − K t ′
(See p. 205 and Equation 5-1.) 13-3 C peak (steady state) = e
VK (1− e − K τ )
where:
Cpeak(steady state) = desired peak concentration (usually (See p. 205.)
2 hours after end of infusion), where Cpeak(steady state) = measured steady-state peak
K0 = drug infusion rate (dose/infusion plasma concentration drawn 2 hours after end of
time), infusion.

255
Calculation of Actual (i.e., New) Dosing Calculation of Time to Hold Dose When
Interval Based on Patient-Specific Value Actual Ctrough from Laboratory Is Too High
for K
1 Ctrough(desired) = Ctrough(actual)e–Kt
13-4 τ = In C trough (desired) − In C peak (desired)  + t + t ′
−K 
(See p. 205.) where t is the amount of time to hold the dose.
Next, take the natural log of both sides:
Calculation of Patient-Specific number = number (t′) and then simply solve for t′
Maintenance Dose (K0) Based which is now not an exponent.
on Actual Values for K and V
K 0 (1− e −K t ) −K t ′ Theophylline Dosing Equations
13-3 C peak (steady state) = e
VK (1− e −K τ )
(See p. 205.) Equation for Calculating the Volume
where: of Distribution for Theophylline
and Aminophylline
Cpeak(steady state) = desired peak concentration at
steady state, 14-1 V (L) = weight (kg) × 0.5 L/kg
K0 = drug infusion rate (also main-
tenance dose you are trying to (See p. 223.)
V = volume of distribution, Equation for Calculating a Loading Dose
of Theophylline or Aminophylline
K
= elimination rate constant calculated
from Cpeak and Ctrough, CpdV
14-2 D=
t = infusion time (usually 1 or 2 hours), SF
t′ = time from end of infusion until (See p. 223.)
concentration is determined
(usually 2 hours for peak), and
Equation for Calculating Clearance
τ = desired or most appropriate dosing for Theophylline or Aminophylline
interval.
14-3 Cl = (0.04 L/kg/hr) × weight (kg)
Calculation of New Expected
(See p. 223.)
Ctrough(steady state) That Would Result from New
Maintenance Dose and Interval Used
Equation for Calculating a Theophylline
13-5 Ctrough(steady state) = Cpeak(steady state)e –Kt ′′ or Aminophylline Maintenance Dose
C pss Clτ
(See p. 206.) 14-4 D=
SF
where t′′ is now the number of hours between the
peak and trough (t′′ = τ – t – t′). (See p. 224.)

256
Phenytoin Dosing Equations

X 0 /τ(S ) × K m
Calculation of Population Estimate
15-1 C ss =
Vmax − X 0 /τ(S )
for Volume of Distribution (V )
(See p. 233.)
V = 0.65 L/kg
Calculation of Time (in Days) for
Michaelis–Menten Constant, Representing Phenytoin Dosing Regimen to Reach
the Concentration of Phenytoin at Which Approximately 90% of Its Steady-State
the Rate of Enzyme-Saturable Hepatic Concentration
Metabolism Is One-Half of Maximum
K m ×V
(½Vmax) =
10-4 t 90% ( 2.3 × Vmax ) − ( 0.9 × X d ) 
(Vmax − X d )2 
Km = 4 mg/L
(See p. 154.)
where:
Maximum Amount of Drug That
Can Be Metabolized per Unit Time Xd = daily dose of phenytoin (in milligrams
per day),
Vmax = 7 mg/kg/day V = volume of distribution,
Note: usually expressed as mg/day. Vmax = maximum rate of drug metabolism (in
milligrams per day), and
Calculation of Phenytoin Loading Dose Km = Michaelis–Menten constant.
V × C desired
1-1 X0 = Phenytoin Dosing Methods
S
(See p. 10.) Method 1A (Empiric)
where: Use 5 mg/kg/day.
V = volume of distribution estimate of
0.65 L/kg, Method 1B (Population Parameters)
Cdesired = concentration desired 1 hour after the Use population estimates for the Michaelis–Menten
end of the infusion, and values for Km of 4 mg/L and Vmax of 7 mg/kg/day and
S = salt factor. solve the general MME formula as shown below:
Vmax × C ss −desired
Two Representations of Michaelis– X 0 /τ(S ) =
K m + C ss −desired
Menten Equation Used to Calculate
Daily Dose [X0/τ (S)] or Expected Serum
Concentration Css Method 2 (One Steady-State Level)
Vmax × C ss Use this method after you have one steady-state

10-1 X 0 /τ(S ) = phenytoin serum drug concentration to solve for
K m + C ss
Vmax while still using the population parameter for
(See p. 152.) Km.

257
First, to solve for Vmax: get new (i.e., real) Km and Vmax, re-solve the MME
equation for dose.
( X d × S )(K m + C ss −lab )
10-1 Vmax = First, solve for “real” Km. The slope of the line,
C ss −lab
which represents –Km, can now be calculated as
follows:
(See p. 152.)
X − X2
where: 10-2 −K m =1
X1 X2
Vmax = calculated estimate of patient’s Vmax, −
C1 C 2
Km = population estimate of 4 mg/L,
Xd × S = patient’s daily dose of phenytoin free (See p. 152.)
acid, and
where:
Css = reported steady-state concentration.
X = dose (where X is milligrams of free acid) and
Second, after solving for this “better” value
C = concentration.
for Vmax, use it plus the old Km value in the MME to
re-solve for dose, as shown below: Next, we substitute this new value for Km into
the MME and solve for a Vmax as follows:
V ×C
10-1 X d × S =max ss
K m + C ss ( X d × S )(K m + C ss )
10-1 Vmax =
C ss
(See p. 152.)
(See p. 152.)
where:
where:
Xd × S = new dose of phenytoin (either free acid
or salt), Xd × S = either of the doses the patient received,
expressed as free acid,
Css = desired steady-state concentration
(usually 15 mg/L), Css = steady-state concentration at the dose
selected, and
Km = population estimate of 4 mg/L, and
Km = calculated value.
Vmax = calculated estimate from above.
Finally, we substitute our new Vmax value (mg/day)
Method 3 (Two Steady-State Levels) and our calculated Km value (mg/mL) into the MME
Use after you have two steady-state phenytoin and solve for Xd as follows:
concentrations from two different phenytoin doses. V ×C
You can now work another equation to solve for a 10-1 X d × S =max ss
better value for Km (shown below). Then use this K m + C ss
better Km value to once again re-solve for an even
better Vmax value than used in Method 2. Once you (See p. 152.)

258
Digoxin Dosing Equations Equation Showing Relationship

between Steady-State Plasma
Concentration, Maintenance Dose,
Volume of Distribution of Digoxin in
and Total Systemic Clearance
Patients with Normal Renal Function
X d × 106 × F
15-2 Vss = 4 to 9 L/kg IBW (average, 6.7 L/kg IBW) 15-4 C ss =
Clt × τ
(See p. 242.)
(See p. 243.)
Equation for Estimating Total

Systemic Clearance for Digoxin
15-3 Clt = (1.303 × CrCl) + Clm
(See p. 242.)
where:
Clt is expressed as mL/minute,
Clm = 40 mL/minute in patients with no or mild
heart failure, and
= 20 mL/minute in patients with moderate to
severe heart failure.

APPENDIX B
Supplemental Problems*
QUESTIONS
SP1. Antipyrine (a drug used for pharmacologic and pharmacokinetic studies

in the evaluation of hepatic mixed-function oxidase activity) was admin-
istered as a single intravenous (IV) bolus dose (1200 mg). The following
plasma drug concentration and time data were collected:
Time after Plasma Drug

Dose (hours) Concentration (mg/L)
2 26.1
4 24.3
8 20.7
18 14.2
32 8.6
48 4.5

Using semilog graph paper, determine the approximate time after the dose
when the plasma drug concentration falls to 3.0 mg/L.
A. 50 hours
B. 40 hours
C. 70 hours
D. 60 hours
SP2. Using the same data for antipyrine dosing above, estimate the volume of
distribution.
A. 50.2 L
B. 42.9 L
C. 45.3 L
D. 24.9 L
*These problems supplement material presented in Lessons 1–11.
259
260
SP3. Just after an IV dose of antibiotic X, the SP6. An IV bolus dose of antibiotic Q (500 mg)
plasma drug concentration was 7.3 mg/L. was administered to a patient on an every-
Six hours later, the concentration was 2.9 mg/L. 6-hour schedule. Predict the plasma drug
Predict the plasma drug concentration at 10 concentrations at 3 and 6 hours after dosing.
hours after the dose. Assume: (1) a one-compartment model, (2)
A. 2.9 mg/L T ½ = 4.95 hours, (3) Clt = 14.2 L/hour, and
(4) the attainment of steady state.
B. 2.1 mg/L
A. 5.7 and 3.7 mg/L, respectively
C. 1.63 mg/L
B. 7.1 and 9.8 mg/L, respectively
D. 3.1 mg/L
C. 3.7 and 2.9 mg/L, respectively
SP4. The following plasma drug concentration D. 6.9 and 5.7 mg/L, respectively
and time data were obtained after an IV
bolus dose of procainamide (420 mg): SP7. For the same patient, predict the plasma
concentrations at 3 and 6 hours after the
Time after Plasma Drug second dose.
Dose (hours) Concentration (mg/L) A. 6.46 and 5.02 mg/L, respectively
0 3.86 B. 3.32 and 2.19 mg/L, respectively
0.5 3.36
C. 8.12 and 5.78 mg/L, respectively
1.0 3.00
2.0 2.29 D. 4.64 and 3.05 mg/L, respectively
3.0 1.77
SP8. An 80-kg patient receives 500 mg of drug Y
5.0 1.06
intravenously by bolus injection every 6 hours.
7.0 0.63
Assume that V = 0.5 L/kg, and T½= 6.4 hours.
10.0 0.29
Predict the steady-state peak and trough
Calculate clearance by the area method. concentrations.
A. 27.83 L/hour A. 28.6 and 14.7 mg/L, respectively
B. 19.4 L/hour B. 26.2 and 13.7 mg/L, respectively
C. 33.6 L/hour C. 24.3 and 12.9 mg/L, respectively
D. 11.8 L/hour D. 19.8 and 9.6 mg/L, respectively
SP5. What will be the minimum concentration SP9. Calculate the theophylline clearance (Clt)
after the thirteenth IV dose of drug X if for a 52-kg patient receiving a continuous
Cmax equals 100 mg/L after the first dose, IV infusion of aminophylline at 75 mg/hour.
K equals 0.4 hr–1, and τ equals 6 hours? The patient’s steady-state plasma theophy-
(Assume an IV bolus dose model.) lline concentration with this dose rate is
20.2 mg/L. Assume that the patient’s V =
A. 8.98 mg/L
0.45 L/kg. Remember, aminophylline = 80%
B. 9.98 mg/L theophylline.
C. 13.9 mg/L A. 2.97 L/hour
D. 7.36 mg/L B. 3.75 L/hour
C. 3.71 L/hour
D. 2.37 L/hour

Appendix B | Supplemental Problems
261
SP10. The following plasma concentration and Plot the plasma concentration versus time
time data were collected after a single profile on semilog paper. From your graph,
500-mg IV dose of amikacin: determine A, B, α, β, Varea, and Clt (in milli-
liters per minute).
Time after Amikacin
Dose (hours) Concentration (mg/L) A. 3.60 hr–1, 0.41 hr–1, 39.1 L, and 11 L/hr,
2 22.5 respectively
4 18.4 B. 2.60 hr–1, 0.31 hr–1, 29.1 L, and 9 L/hr,
8 12.3 respectively
16 5.6 C. 1.60 hr–1, 0.21 hr–1, 19.1 L, and 7 L/hr,
24 2.5 respectively
36 0.75
D. 4.60 hr–1, 0.35 hr–1, 49.1 L, and 12 L/hr,
48 0.23
respectively
Calculate K, Varea, and Clt for this patient.
SP12. Calculate Varea given the data in Supple-
A. 0.20 hr–1, 28.2 L, and 2.82 L/hour, mental Problem 1. Compare it with the V
respectively calculated (using the back-extrapolation
B. 0.01 hr–1, 1.82 L, and 0.182 L/hour, method) in Supplemental Problem 2.
respectively A. 54.6 L
C. 0.10 hr , 18.2 L, and 1.82 L/hour,
–1
B. 43.6 L
respectively
C. 10.3 L
D. 0.10 hr–1, 182 L, and 18.2 L/hour,
D. 42.96 L
respectively
SP13. An outpatient had been taking 400 mg of
SP11. Seven healthy female subjects were each
phenytoin per day for 1 month and had a
given 1250 mg of an experimental drug
plasma concentration of 6.0 mg/L when
(BB-K8) by IV bolus administration. The
sampled 6 hours after the dose. Because of
drug follows first-order kinetics. The
continued seizures, the dose was increased
following mean plasma concentration and
to 500 mg/day. Four weeks later, the patient
time data were obtained:
was seen in a clinic, and the plasma drug
Time after Mean Plasma Drug concentration 6 hours after the dose was
Dose (hours) Concentration (mg/L) 9.0 mg/L (assume steady state). The physi-
0 116.0 cians asked that the dose be increased to
0.08 108.3 provide a plasma concentration of 12 mg/L
0.17 92.8 6 hours after the dose. What dose would you
0.25 83.3 recommend?
0.50 59.2 A. 552 mg phenytoin free acid/day
0.75 38.2
B. 600 mg phenytoin free acid/day
1.0 30.6
1.5 22.9 C. 652 mg phenytoin free acid/day
2.0 19.7 D. 900 mg phenytoin free acid/day
3.0 13.2
4.0 9.3
5.0 7.3
6.0 5.1
7.0 4.1
8.0 2.8

262
ANSWERS
SP1. A, B, C. Incorrect answers (3.84 mg/L + 3.36 mg/L)
=AUC0→0.5 × (0.5 − 0 hr)
D. CORRECT ANSWER 2
= 1.80 (mg/L) × hr
SP2. A, C, D. Incorrect answers
(3.36 + 3.00)
B. CORRECT ANSWER =
AUC 0.5 →1 × (1=
− 0.5) 1.59 (mg/L) × hr
2
C 0 = 28 mg/L (3.00 + 2.29)
=
AUC1→2 ×=
(2 − 1) 2.65 (mg/L) × hr
2
dose 1200 mg
=
V = = 42.9 L (2.29 + 1.77)
C0 28 mg/L =
AUC ×=
(3 − 2) 2.03 (mg/L) × hr
2→ 3
2
SP3. A, B, D. Incorrect answers (2.29 + 1.77)
=
AUC ×=
(3 − 2) 2.03 (mg/L) × hr
C. CORRECT ANSWER. First, calculate the 2→ 3
2
elimination rate constant (K):
(1.77 + 1.06)
=
AUC 3 →5 ×=
(5 − 3) 2.83 (mg/L) × hr
(In 7.3 − In 2.9) 2
K =
− 0.15 hr −1
=
(0 − 6 hr) (1.06 + 0.63)
=
AUC 5 →7 ×=
(7 − 5) 1.69 (mg/L) × hr
2
Then use these equations:
(0.63 + 0.29)
− Kt =
AUC ×
= (10 − 7) 1.38 (mg/L) × hr
C = C0e 7 →10
2
−1 C 10 hr 0.29 mg/L
C at 10 hr = (7.3 mg/L)e −0.15 hr (10 hr)
=
AUC10→∞ == 1.12 (mg/L) × hr
K 0.26 hr −1
= 1.63 mg/L AUC = 1.80 + 1.59 + 2.65 + 2.03 + 2.83 + 1.69 + 1.38 + 1.12
= 15.09 (mg/L) × hr
SP4. A. CORRECT ANSWER. To calculate clear-
ance by the area method, we need to X0 420 mg
=
Clt = = 27.83 L/hr
know the area under the plasma concen- AUC 15.09 (mg/L) × hr
tration curve (AUC) and the dose (X0).
Therefore, it is first necessary to calcu-
late AUC using the trapezoidal method
as shown below. Note that one way to
indicate an AUC from one time point
to another is as AUC0→0.5, which means
AUC from 0 to 0.5 hour.

263
SP5. A, C, D. Incorrect answers Note that:

B. CORRECT ANSWER. To determine Cmin 0.693
after the thirteenth dose, first calculate =K = 0.14 hr −1
T 12
Cmax after the thirteenth dose using the
multiple-dose equation: Then:
(1− e −nK τ ) 500 mg

C max(nth dose) = C max(nth dose) C max = −1
(1− e −K τ ) (101.43 L)(1− e −0.14 hr (6 hr)

)
(1− e −13K τ ) = 8.67 mg/L

C max13 = C max1
(1− e −K τ ) From Cmax, the concentration at any time
−1 after a dose can be calculated by:
(1− e( −13)(0.4 hr )( 6 hr)
)
= (100 mg/L)
(1− e( −0.4 hr
−1
)( 6 hr)
) Ct = Cmaxe−Kt
= 110 mg/L
So:
Then: C 3 hr = (8.67 mg/L)(e −0.14 hr

−1
( 6 hr)
)
−K τ ( −0.4 hr −1 )( 6 hr)
=
C min13 C=
max13 e (110 mg/L)e = 5.7 mg/L
= 9.98 mg/L
and:
SP6. A. CORRECT ANSWER. To predict plasma
−1
C 6 hr = (8.67 mg/L)(e −0.14 hr (6 hr)
)
concentrations 3 and 6 hours after a
dose at steady state, we should first esti- = 3.74 mg/L
mate the Cmax (at 0 hour after the dose)
using the steady-state IV equation: B, C, D. Incorrect answers
X0 SP7. A, B, C. Incorrect answers

C max = D. CORRECT ANSWER. The equations used
V (1− e −K τ )
to solve Supplemental Problem 5 can be
So we first need to estimate V and K. V used here, with the number of doses (n)
can be estimated from: equal to 2 rather than 13:
Clt = VK X 0 (1− e −nK τ )

C max(2nd dose) =
V (1− e −K τ )
Then:
−1
Clt 14.2 L/hr (500 mg)(1− e( −2)( −0.14 hr )(6 hr)

)
=
V = = ( −0.14 hr −1 )(6 hr)
K  0.693  (101.43 L)(1− e )
 4.95 hr 
  = 7.06 mg/L
V = 101.43 L Then:
Ct = C max e − Kt
−1
C 3 hr = (7.06 mg/L)(e( −0.14 hr )(6 hr)
)
= 4.64 mg/L

264
and: SP10. A, B, D. Incorrect answers

−1 C. CORRECT ANSWER. First, the data
C 6 hr = (7.06 mg/L)(e( −0.14 hr )(6 hr)
)
should be plotted on semilog graph
= 3.05 mg/L paper to determine if they are linear or
nonlinear. When the points are deter-
mined to make a straight line, any two
SP8. A, C, D. Incorrect answers
may be chosen to calculate K. (It is best,
B. CORRECT ANSWER. First, determine K however, to choose two that are not close
and total V: to each other, such as 2 and 4 hours.) So:
=V 0.5 L/kg × 80 kg = 40 L
∆Y  In 0.75 − In 22.5 
K =
− =
−  0.10 hr −1
=
0.693 0.693 ∆X  36 hr − 2 hr 
=K = = 0.108 hr −1
T 12 6.4 hr Then:
Then use the steady-state multiple-dose 0.693

=
T 12 = 6.93 hr
equation (for IV bolus doses): K
X0  1  To calculate Varea and Clt , we should

C peak =
V  1− e −K τ  first estimate the AUC. With a one-
compartment, first-order model after IV
500 mg administration, the calculation of AUC is
= −1
(40 L)(1− e −0.108 hr ( 6 hr)
) simplified. In this case:
= 26.2 mg/L C0
AUC =
K
C trough = C peak e −K τ
where C0 is determined by Ct = C0e–Kt.
= (26.2 mg/L)e −0.108 hr −1 ( 6 hr) For t = 2 hours:
−1
= 13.7 mg/L 22.5 mg/L = C 0 e −0.10 hr (2 hr)
SP9. A. CORRECT ANSWER. To calculate Then:

clearance, use the relationship:
C 0 = 27.5 mg/L
K 0 75 mg/hr (0.80)
=
Clt = = 2.97 L/hr
C ss 20.2 mg/L and:
27.5 mg/L
B, C, D. Incorrect answers AUC =
0.1 hr −1
= 275 mg/L × hr

265
Then: In 2.8 − In 13.2

β= = −0.31
X0 8 hr − 3 hr
Varea =
AUC × K
= 0.31 hr −1
500 mg
= Then from the residual line, the intercept
(275 mg/L × hr)(0.10 hr −1 ) = A (equals 84 mg/L) and the slope gives α:
= 18.2 L
In 1.9 − In 76.3
X α= = −2.60
Clt = 0 1.5 hr − 0.08 hr
AUC
= 2.60 hr −1
500 mg
= = 1.82 L/hr
(275 mg/L × hr) To calculate Varea and Clt, the AUC must
first be determined. The AUC can be
Note that the use of AUC for calculation estimated using the trapezoidal rule or
of clearance generally produces a more by adding the area of each exponential
accurate estimate than the use of equation:
Clt = K × V. A B
AUC = + = 32.3 + 106.5
SP11. A, C, D. Incorrect answers α β
B. CORRECT ANSWER. A, B, α, and β will be = 138.8 (mg/L) × hr
calculated using residuals. First, back-
extrapolate the terminal (straight-line) Then:
portion of the plot and estimate the
back-extrapolated points. Determine
dose
Varea =
the residual points by subtracting the AUC ×β
back-extrapolated concentrations from
1250 mg
the actual concentrations. = = 29.1 L
138.8 (mg/L) × hr (0.31 hr −1 )
Back
Actual Extrapolated Residual dose
Points Points Points Clt =
AUC
108.3 – 32.0 = 76.3 mg/L
92.8 – 31.0 = 61.8 mg/L 1250 mg
= = 9.0 L/hr
83.3 – 30.0 = 53.3 mg/L 138.8 mg/L
59.2 – 28.0 = 31.2 mg/L
SP12. A, B, C. Incorrect answers
38.2 – 26.0 = 12.2 mg/L
30.6 – 24.0 = 6.6 mg/L
D. CORRECT ANSWER.
22.9 – 21.0 = 1.9 mg/L dose
Varea =
AUC × K
Then plot the residual points on the K = 0.037 hr −1
same graph. From the back-extrapolated
line, the intercept = B (equals 33 mg/L)
and the terminal slope gives β:

266
To calculate the AUC, the C0 must first be SP13. A, C, D. Incorrect answers

estimated from the plot (C0 = 28 mg/L): B. CORRECT ANSWER. Phenytoin follows
Michaelis–Menten (saturable) pharmaco-
(28 + 26.1)(2 − 0) kinetics. To determine Vm and Km, the
=
AUC0→2 = 54.1 (mg/L) × hr
2 daily dose must be plotted (y-axis)
versus the daily dose divided by the
(26.1 + 24.3)(4 − 2)
=
AUC2→4 = 50.4 (mg/L) × hr resulting steady-state concentrations
2 (x-axis). From a plot of the dose (y-axis)
(24.3 + 20.7)(8 − 4) versus dose/concentration (x-axis), the
=
AUC4→8 = 90.0 (mg/L) × hr following are observed:
2
Vm = 1000 mg daily (which is equal to
(20.7 + 14.2)(18 − 8) the y-intercept)
=
AUC8→18 = 174.5 (mg/L) × hr
2 Km = 9.0 mg/L (which equals –slope)
(14.2 + 8.6)(32 − 18) Then:
=AUC18→32 = 159.6 (mg/L) × hr
2 VmC ss
dose =
(8.6 + 4.5)(48 − 32) K m + C ss
=
AUC32→48 = 104.8 (mg/L) × hr
2
where:
C 48 4.5 mg/L C ss = 12 mg/L, the desired concentration:
=
AUC48→∞ = = 121.6 (mg/L) × hr
K 0.037 hr −1
(1000 mg)(12 mg/L)
Then: = = 571 mg/day
9.0 mg/L + 12 mg/L
AUC =54.1+ 50.4 + 90.0 + 174.5 + 159.6 Therefore, the likely daily dose would be
+ 104.8 + 121.6 600 mg/day.
Alternatively, you can use Equation
= 755 (mg/L) × hr
10-2, p. X, to calculate –Km from these
X0 two doses and two levels. Both methods
Varea = should give the same answer.
AUC × K
1200 mg
=
755 (mg/L) × hr (0.037 hr −1 )
= 42.96 L
So, in this case, the two estimates for V

are similar.

APPENDIX C
Glossary
Area under the first moment curve (AUMC)—the area under the first moment
curve (drug concentration × time) versus time (moment) curve, an important
model-independent pharmacokinetic parameter.
Area under the plasma concentration versus time curve (AUC)—the area
formed under the curve when plasma drug concentration is plotted versus
time. Drug clearance is equal to the dose administered divided by AUC.
Bioavailability (F)—the fraction of a given drug dose that reaches the systemic
circulation.
Biopharmaceutics—the study of the relationship between the nature and

intensity of a drug’s biologic effects and various drug formulation or admin-
istration factors, such as the drug’s chemical nature, inert formulation
substances, pharmaceutical processes used to manufacture the dosage form,
and routes of administration.
Clearance—the process of removing a drug from plasma (expressed as volume

of plasma per a given unit of time).
Clinical pharmacokinetics—the application of pharmacokinetic principles

to the safe and effective therapeutic management of drugs in an individual
patient.
Compartmental model—a basic type of model used in pharmacokinetics.

Compartmental models are categorized by the number of compartments
needed to describe the drug’s behavior in the body. There are one-compart-
ment, two-compartment, and multi-compartment models. The compartments
do not represent a specific tissue or fluid but may represent a group of
similar tissues or fluids.
Drug distribution—transport processes that deliver drug to body tissues and

fluids after absorption.
50% effective concentration (EC50)—the concentration at which 50% of the

maximum drug effect is achieved.
Elimination rate constant (K)—a constant representing the fraction of drug

removed per unit of time (in units of reciprocal time, usually hr –1).
267
268
Extraction ratio (E)—the fraction of drug Model-independent parameter—a pharmaco-

removed from plasma by one pass through an kinetic parameter, such as clearance, that can be
organ. This ratio is a number between 1 and 0. calculated without the use of a specific model.
Organs that are very efficient at eliminating a
drug will have an extraction ratio approaching Model-independent pharmacokinetics—
1 (i.e., 100% extraction). pharmacokinetic calculations using param-
eters that do not require the use of specific
First-order elimination—when the amount of compartmental models (e.g., one-compartment,
drug eliminated from the body in a specific two-compartment, etc.).
time is dependent on the amount of drug in the
body at that time. A straight line is obtained Pharmacodynamics—the relationship between
from the natural log of plasma drug concentra- drug concentrations at the site of action and
tion versus time plot only for drugs that follow the resulting effect, including the time course
first-order elimination. and intensity of therapeutic and adverse
effects.
First-pass effect—drug metabolism by the liver
that occurs after absorption but before the Pharmacokinetics—the relationship of drug dose
drug reaches the systemic circulation. to the time course of drug absorption, distribu-
tion, metabolism, and excretion.
Formation clearance (CLP→mX)—a model-
independent parameter that provides a Plasma—the fluid portion of blood (including
meaningful estimate of a drug’s fractional soluble proteins but not formed elements).
metabolic clearance.
Receptor—a structure on the surface of a cell to
Half-life (T½)—the amount of time necessary for a which a drug binds and causes an effect within
plasma drug concentration to decrease by half. the cell.
Kinetic homogeneity—the predictable relation- Serum—the fluid portion of blood that remains
ship between plasma drug concentration and when the soluble protein fibrinogen is removed
concentration at the receptor site. from plasma.
Mean residence time (MRT)—the average time Steady state—the point at which, after multiple
for intact drug molecules to transit or reside in doses, the amount of drug administered over
the body. a dosing interval equals the amount of drug
being eliminated over that same period.
Minimum inhibitory concentration—the lowest
concentration of an antibacterial agent that will Therapeutic drug monitoring—determination of
inhibit the visible growth of a microorganism plasma drug concentrations and clinical data to
after overnight incubation. optimize a patient’s drug therapy.
Model—a simplified mathematical simulation of Therapeutic range—the plasma concentration

physiologic processes used to predict the time range that is effective and safe in treating
course of drug concentrations or effect in the body. specific diseases.

Appendix C | Glossary
269
Tolerance—decreased drug effectiveness with Volume of distribution at steady state (Vss)—a

continued use. parameter that relates total amount of drug in
the body to a particular plasma concentration
Volume of distribution (V)—an important indi- under steady-state conditions.
cator of the extent of drug distribution into
body fluids and tissues, V relates the amount Zero-order elimination—when the amount of drug
of drug in the body to the measured concen- eliminated for each time interval is constant,
tration in the plasma. Thus, V is the volume regardless of the amount of drug in the body.
required to account for all of the drug in the
body if the concentration in all tissues is the
same as the plasma concentration.

INDEX
Page numbers followed by f refer to figures; those followed by t refer to tables.
A
Absorption of drugs, 99-114
and bioavailability, 102-103
and disposition in body, 100f
and drug effects in chronic diseases, 108
and elimination processes, 102, 102f, 103-106
and fraction reaching systemic circulation, 108, 134
and plasma drug concentrations, 101-107, 101f-107f
discussion points on, 114
in oral administration, 104-106, 134
processes involved in, 100f
nonlinear, 150, 150t
of digoxin, 102, 242
review questions and answers on, 109-113
with controlled-release products, 106-108, 107f
with different formulations, 100, 101
zero-order, 106
Absorption rate constant (Ka ), 104-105, 108
Accumulation factor, 250
at steady state, 53-55, 250
for aminoglycosides, 189
definition of, 52
equation for, 52
in intravenous bolus administration, 52-55
peak concentrations and, 52
predicting concentrations before steady state achievement with, 52
Acetaminophen metabolism, 131t, 132t
Acetylation in drug metabolism, 132, 132t
Adaptive resistance to aminoglycosides, 197
Adipose tissue, drug distribution in, 7
Adverse effects, of theophylline, 224, 226
Age-related changes, 159-160
and plasma drug concentration, 5
in body composition, 159-160, 160f
in glomerular filtration rate, 181, 181f
Albumin binding to drugs, 117-118, 119t
drug interactions affecting, 119, 119t
in disease states, 119-120, 121
Alcohol
affecting metabolism of other drugs, 131, 131t
metabolism of, 131t
271
272
Alfentanil metabolism, 131t tobramycin, 182 sensitivity of, 164-165

Alpha-1-acid glycoprotein binding to drugs, desired plasma concentration of, specificity of, 164
119, 119t 184 upper limit of drug detection in, 164
in disease states, 120, 137 extended-interval dosing of, 197,
197t, 198, 198f, 199, 253 Astemizole metabolism, 131t
Alpha negative slope in residual line, 85, 85f,
vancomycin compared to, 204, 205, 206, Atorvastatin, 37t, 54t
86
207 AUC (area under plasma concentration versus
Amikacin
volume of distribution, 182-184, 185, time curve), 38-39, 38f-39f
cross-reactivity in assays, 164
187, 189, 191, 193-194, 200 and bioavailability of drugs, 101
dosing regimens, 182
and ideal body weight, 182, 185, and clearance, 39
and desired plasma concentration, 251 renal, 139
184
calculation of, 193-194 total body, 167, 167f, 168
extended-interval, 197, 197t, 198,
199, 200, 201 Aminophylline, 221-229 and volume of distribution at steady
discussion points on, 229 state, 168, 169
Aminoglycosides, 199-202
dosing equations for, 255 definition of, 267
accumulation factor for predicting
concentrations of, 52 infusion rate for, 223, 225 determination of, 38, 250
amikacin. See Amikacin loading dose of, 222-223, 223f, 225, 227, for one dosing interval, 57, 57f
assay methods for 255 trapezoidal rule in, 38-39, 39f, 167
cross-reactivity in, 164 maintenance dose of, 224-228, 255 discussion, 44
penicillin affecting, 165-166 theophylline dose equivalent, 221, 223 for controlled-release products, 108
case studies on, 184-201 Amiodarone affecting metabolism of other relationship to drug dose, 102, 149, 150f
clearance of, compared to creatinine drugs, 131t in dose-dependent pharmaco-
clearance, 140, 141f, 182, 183f Amlodipine, 10t, 25t kinetics, 149, 150f
desired plasma concentration of, 184 Amphetamine metabolism, 132t review questions and answers, 40-43
discussion points on, 202 Amphotericin in lipid emulsion, 117 sustained-release products and, 107
distribution of, 8, 82, 116 Ampicillin, protein binding of, 118t terminal part of, 39, 39f, 168
in obesity, 162-163 Antibiotics AUMC (area under the moment curve)
dosing equations on, 251-253 aminoglycoside. See Aminoglycosides and total body clearance, 167f, 167-168
dosing interval for, 184-186, 193-194, first-order elimination of, 26 and volume of distribution at steady
252, 253 state, 168, 169
minimum concentration inhibiting
and calculation of time to hold a bacterial growth, 6 definition of, 267
dose, 195-196, 196f, 253 vancomycin. See Vancomycin Average steady-state concentration, 57f,
extended-interval, 196-199, 197t, 57-58
198f, 253 Antimalarial drugs, 23
ideal body weight and, 197-198 Antipyrine, intrinsic clearance of, 134t
dosing regimens for, 182 Apparent volume of distribution, 22
elimination rate constant for, 140, 141f, Area under the moment curve. See AUMC B
182-184, 185, 186, 187, 192-193 Area under the plasma concentration versus
and creatinine clearance rate, 183, time curve. See AUC Back-extrapolation
185, 251 Ascorbic acid, nonlinear pharmacokinetics for concentration just after IV
calculation of, 192-193, 252 of, 150t administration, 21
first-order elimination of, 26 Aspirin in absorption rate constant calculation,
controlled-release formulations of, 106t 104-105, 104f-105f
gentamicin. See Gentamicin
intrinsic clearance of, 134t negative slope beta (β) in, 86, 87, 87f
half-life of, 185, 188, 191, 192, 196,
198-199 metabolism of, 132t Bactericidal activity of aminoglycosides, 196
loading dose of, 184-185, 188f, 189-190, nonlinear pharmacokinetics of, 149 Benzodiazepine concentration and tissue
252 distribution, 8
steady-state concentration of, 54
maintenance dose of, 183, 187-188, Beta-lactam antimicrobials, 6
252-253 Assay methods, 163-165
Beta negative slope in back-extrapolated line,
in case studies, 184-185, 187-189, calibration of instruments in, 165-166 86, 87, 87f
194, 195, 200 cross-reactivity in, 164-165 Biexponential elimination, 82, 87, 87f
peak and trough concentrations of, 160, drug interactions affecting, 165, 166-167 of vancomycin, 203
184, 185-186, 187-188, 190-191, drug concentration in plasma and, 4
191f, 192-196, 199-200 Biexponential equation
interferences in, 164-165 definition of, 86
and penicillin interactions, 165-166
lower limit of drug detection in, 164 volume of distribution and, 86-87
in extended interval dosing, 182,
196, 199-200 physiochemical factors affecting Biliary clearance, 23, 129, 133
prediction of, 52 accuracy of, 164-165
Bilirubin as assay interference, 165
pharmacokinetic variations in obesity, quality control checks in, 165
Bioavailability (F), 102-103
162-163 sample collection and handling in,
163-164 definition of, 267
steady-state concentration of, 187, 188,
sampling times in, 165-166 factors affecting, 100
191, 194-195, 200
of digoxin, 101, 242, 244, 246

Index |
273
of phenytoin, 232
C disease states affecting, 135-138, 160
Biopharmaceutics, 99-114 elimination processes in. See Elimination
absorption and. See Absorption of drugs Caffeine metabolism, 131t processes
definition of, 99, 267 Calculators with natural log and exponential extraction ratio in, 24, 24t, 134t, 132
introduction to, 99-102 keys, 14 formation
Calibration of assay instruments, 165 as model-independent parameter,
Biotransformation, 128, 129, 130, 131-132.
166-169, 169f
See also Metabolism Captopril metabolism, 132t
definition of, 268
liver functions in, 129, 130-131 Carbamazepine
hepatic, 23, 67, 135-136
Blood active metabolite of, 128
disease states and drug interac-
compared to plasma and serum, 22f, affecting metabolism of other drugs, tions affecting, 136-138
22-23, 163 131t
in continuous infusions, 68, 69, 70
definition of, 22, 22f measurement of plasma concentrations,
164 intrinsic, 133-134, 134t
Blood–brain barrier, 116, 121
metabolism of, 131t, 132t model dependent, 24
Blood flow
nonlinear pharmacokinetics of, 150t model independent, 24, 38
and clearance rate, 22f, 22-23, 24t
therapeutic range for, 6t formation clearance in, 166-169,
extraction ratio in, 132-134 169f
and drug distribution, 115, 116 Cefazolin total body clearance in, 166,
hepatic, 24, 129, 129f half-life of, 37, 37t 167-168, 169
and extraction ratio,132-134 steady-state concentration of, 54t of commonly used drugs, 25, 25t
through organ (Q), 23, 24 Central compartment, 7, 7f, 8, 8f, 81-82, 83f, of digoxin, 242, 243, 244-246, 258
84, 86, 87 of phenytoin, 134t, 232f, 235, 236-237,
Body composition
amount of drug in, 9, 83f, 84 239f
age-related changes in, 159-160, 160f
volume of drug distribution in, 86 of theophylline. See Theophylline,
and creatinine clearance, 183
Centrifugation, 163, 163f clearance of
in obesity, 162-163
Cephalosporins, first-order elimination of, 26 organ, 23f, 23-24, 133, 249
Body fluids. See Fluids of body
Child-Pugh score, 130-131 relation to other pharmacokinetic
Body weight parameters
adjusted, 116, 141, 180, 180t, 182, 251 Children
dosage in, 38
and aminoglycoside extended- chloramphenicol toxicity in, 131
volume of distribution and elimina-
interval dosing, 197 pharmacokinetic variations in, 159 tion rate constant in, 38, 103,
and aminoglycoside volume of Chloramphenicol 250
distribution, 183, 251 affecting metabolism of other drugs, renal, 23, 67, 128, 139-140. See also
and creatinine clearance, 141-142, 180t, 131t Kidneys, in drug clearance
180-181, 182, 185 metabolism of, 132t steady-state, 68
and aminoglycoside clearance,
protein binding of, 118t total body (Clt ), 23, 133, 135, 249
182-183
toxicity in neonates, 131 and AUC calculation, 167, 167f,
and theophylline clearance, 223,
168, 169
227 Chlorpromazine metabolism, 132t
and AUMC calculation, 167f,
and volume of distribution, 116 Chronic obstructive pulmonary disease, 168-169
of aminoglycosides, 182-185, 251 theophylline in, 221-228
model-independent relationships,
of theophylline, 223 clearance of, 222t 166, 167-168, 169
of vancomycin, 205 Cimetidine relation to volume of distribution
ideal, 116, 251 affecting metabolism of other drugs, and elimination rate constant,
and aminoglycoside extended- 131, 131t 38, 103, 250
interval dosing, 197-198 theophylline, 131t, 136f, 163, 222t with controlled-release formulation,
and aminoglycoside volume of metabolism of, 132t 101, 107
distribution, 182, 183, 185, Clinical pharmacokinetics, 1, 2, 267
Cinacalcet affecting metabolism of other
251
drugs, 131t Clonazepam metabolism, 132t
and creatinine clearance, 141-142,
180t, 181, 182, 185 Ciprofloxacin affecting metabolism of other Clopidogrel affecting metabolism of other
drugs, 131t, 136, 222t drugs, 131t
estimation of, 141-142, 182, 251
Clarithromycin affecting metabolism of other Cocaine metabolism, 131t
in obesity drugs, 131t Cockcroft–Gault equation, 141, 179-183, 185,
creatinine clearance in, 141, 182
Clearance of drugs, 23-24, 23f-25f 199, 251
pharmacokinetic variations in,
and AUC calculation, 38, 39, 139, 166, compared to MDRD equation, 179-181,
162-163
167, 167f, 168 180t, 181f
percentage of fluid portion, 22
and plasma concentration, 22f, 22-23, for aminoglycosides, 182-183
Breast milk or tissue, drugs in, 117 67-68, 68f for digoxin, 243
Broccoli-drug interactions, 131t biliary, 23, 129, 133 Codeine metabolism, 131t
Bupropion affecting metabolism of other blood flow affecting, 22f, 22-23, 24t Compartmental models, 7-8, 7f-8f
drugs, 131t extraction ratio in, 132 and model-independent relationships,
Burns, volume of drug distribution in, 68 definition of, 267 166, 167

274
definition of, 267 with first-order elimination, 25f, clearance estimation, 107
one compartment. See One-compart- 25t, 25-27, 27f, 82, 87f Cor pulmonale, theophylline clearance in,
ment models interpatient variability in, 5, 6f 222t
two compartment. See Two-compart- in therapeutic range, 4, 4f, 6, 6t, 55-56, Corticosteroids
ment models 56f, 71, 268 affecting metabolism of other drugs,
Concentration of drugs, 1-4 in two-compartment models, 81-82, 82f, 131t
absorption rate affecting, 100, 101f, 85, 87, 87f metabolism of, 132t
104-105, 108 maximum. See Peak concentration of Creatinine clearance, 140-142, 160, 161, 179,
after intravenous loading dose, drugs 182-183, 251
21, 22f minimum. See Trough concentrations and aminoglycoside clearance, 140, 141f,
and clearance, 21-23, 22f, 67-68, 68f of drugs 179, 180t, 182-183, 183f, 185,
and drug effects, 1, 2, 4, 6 monitoring of, 4-6. See also Monitoring 197t, 199
prediction of, 4, 4f of drug levels and aminoglycoside extended-interval
and elimination rate constant, 33, 34, natural log of. See Natural log of drug dosing, 196, 197, 197t, 199
35, 65-66 concentrations and digoxin clearance, 243, 245
and half-life, 35-36 prediction of. See Prediction of drug calculation of, 140-142
concentration
and metabolites 128, 129f direct measurement in, 141
ratio to urine excretion rate, 139 estimated, 141-142, 180t, 181
and percent remaining after time (e-Kt),
35, 58 residual, 85-86, 85f-86f, 104-105, 105f
Cross-reactivity in drug assays, 164-165
and protein binding, 118, 118t, 119 second concentration after injection, 32
Curvilinear plot slope. See Slope of curvilinear
and time profile after drug dose, 2, 2f versus time curve, 2, 2f, 11-13, 11f-13f plot
and bioavailability, 102 Cyclophosphamide metabolism, 131t
at time zero (t0), 21, 22f, 33
calculation of area under, 38-39. Cyclosporine
and tissue concentrations, 1, 2, 2f, 22, See also AUC
115-116, 116f metabolism of, 131t, 132t
for controlled-release products,
and volume of distribution, 10-11, 11f, 106, 106f protein binding of, 119t
67f, 67, 68, 131 therapeutic range for, 6t
for vancomycin, 203, 204f, 205,
assay methods in measurement of, 205f, 208 CYP1, CYP2, and CYP3 isoenzymes in drug
163-165 in continuous infusion, 68, 69f metabolism, 130, 131t
at any time, 33, 35, 51, 57, 74, 249 in intravenous bolus dose model, disease states affecting, 136-138
for aminoglycosides, 187-193, 51, 51f drug interactions affecting, 131t, 136
195-200, 205, 206, 252, 253 in one-compartment model, 11, 11f
genetic factors affecting, 130, 161
for vancomycin, 205, 254 in oral administration 102, 102f,
at first dose in multiple dosing, 50, 50f, 104-105, 106f Cytochrome P450 enzymes in drug meta-
73 bolism, 130, 131t
in short half-life, 53, 53f
disease states affecting, 136-137
at receptor sites, 1, 2, 2f, 3, 3f in two-compartment models, 81,
82, 82f, 84, 84f, 87 drug interactions affecting, 131t, 136
at second dose in multiple dosing, 51,
51f prediction of, 11, 11f genetic factors affecting, 130, 161
at steady-state 53-56. See also Steady- with controlled-release formulations,
state concentration of drugs 106f, 106-108, 107f
compartmental models and, 7, 7f, 8 Conjugation reactions in drug metabolism,
definition of, 10 129, 130 D
dosing interval affecting, 66, 66f Continuous IV infusions, 68-72
Dapsone metabolism, 132t
elimination processes affecting, 128, clearance in, 68, 68f, 69, 70
Dealkylation in drug metabolism, 132t
128f discontinuation affecting plasma drug
concentrations, 71, 71f Deamination in drug metabolism, 132t
first-order, 12, 25f, 25t, 25-27, 82,
86f discussion points on, 80 Decongestants, controlled-release formula-
in saturable elimination, 151, 151f tions of, 106t
infusion rate in, 69, 70, 71, 71f
zero-order, 25f, 25t, 25-26, 26f, 26t Desipramine, intrinsic clearance of, 134t
loading dose in, 71f, 71-72, 72f
50% effective concentration (EC50), 3, Deterministic compartmental models, 7
of theophylline, 72-73, 221, 224, 224f
3f, 267 Dexamethasone affecting metabolism of other
loading dose in, 72, 72f
first concentration after injection (C0), drugs, 131t
prediction of plasma concentration in,
21, 22f Dextromethorphan metabolism, 131t
69-71, 72f
and elimination rate constant, 33, Diazepam
34, 35 review questions and answers on, 76-79
intrinsic clearance of, 134t
and half-life, 35-36 steady-state concentration in, 69, 70, 70f
with loading dose, 71, 72f metabolism of, 131t, 132t
and volume of distribution, 118
Contraceptives, oral, and theophylline Diclofenac metabolism, 131t
in AUC calculation, 39
in one-compartment model, 51 interactions, 222t Digoxin, 242-248
Controlled-release drug products, 106f, absorption and bioavailability of, 102,
fluids sampled for measurement of, 1, 2f
106-108, 107f 242, 244
in Michaelis−Menten kinetics, 151-154
examples of, 106, 106t case studies on, 243-247
in one-compartment models, 7, 8, 8f, 11,
of theophylline, 106, 106t, 107 clearance of, 242, 243, 244-246, 258
11f, 12, 12f

Index |
275
discussion points on, 248 tissue characteristics affecting 115-116, in therapeutic range, 4, 4f
dosing equations, 258 121 maximum, 3, 3f
dosing interval for, 243 volume of, 10-11. See also Volume of prediction of, 4, 4f
distribution
elimination rate constant for, 245, 246 protein-binding affecting, 118f, 118-119
Disulfiram affecting metabolism of other
half-life of, 37t tolerance to, 4, 4f, 269
drugs, 131t
interaction with quinidine, 121
Dosage
loading dose for, 242, 243 adjustment in disease states, 67, 74,
maintenance dose for, 242, 243-247 160-161
adjustment in renal disorders, 242, in kidney disease, 67, 74, 160-161,
E
244, 245 179-181
Edema, pulmonary, theophylline clearance
protein binding of, 118t, 121 in liver dysfunction, 161 in, 222t
in renal failure, 122, 244, 245 and absorption rate, 105 Effective concentration. See 50% effective
steady-state concentration of, 53, 54t, and amount of drug in body, 10, 23 concentration
243-246 in one-compartment model, 8f Elderly
therapeutic range for, 6t, 244 in two-compartment model, 9 creatinine clearance rate in, 183
tissue concentration of, 22 and AUC determination, 38, 39, 149, glomerular filtration rate in, 180, 181
in cardiac muscle, 121 150f
pharmacokinetic variations in, 159, 160
unbound fraction, 121 and clearance, 24, 38
theophylline clearance in, 222t
two-compartment pharmacokinetics of, and volume of distribution, 10, 10t, 116
9, 9f, 84, 84f Elimination processes, 127-147
calculation with Michaelis−Menten
and absorption, 102f, 102-106
volume of distribution, 242, 247, 258 equation, 153-154
in renal failure, 122 and bioavailability of drugs, 102
changes affecting plasma drug
quinidine affecting, 121 concentrations, 66, 66f and slope of straight-line plots, 32f,
32-33, 34
Diltiazem therapeutic drug monitoring for
decisions on, 6, 6f and volume of distribution, 11, 11f, 37
affecting theophylline clearance, 222t
Dose-dependent pharmacokinetics, 149, 150f biexponential, 82, 86-87, 87f
volume of distribution for, 11
of enzyme-saturable drugs, 150-151, of vancomycin, 203
Disease states
151f biotransformation in, 128, 129, 130,
absorption rate of drugs in, 108 131-132
Dose-response curve, 3, 4f
distribution of drugs in, 116, 121-122, discussion points on, 147
160-161 Dosing interval (τ ), 49, 50, 250
and AUC for one interval, 57, 57f excretion in, 138-140
dosage adjustment in, 160-161
and half-life, 51, 54f extraction ratio in, 24, 24t, 132-134
hepatic metabolism of drugs in, 128,
129-132, 160-161 and steady state, 53-54, 55, 55f, 57f, first-order, 12, 25-27. See also First-
57-58 order elimination
protein binding of drugs in, 119-120,
134-138 changes affecting plasma drug concen- genetic factors affecting, 162
renal metabolism of drugs in, 160-161 trations, 66, 66f half-life in, 35-36
steady-state concentration of drugs in, for aminoglycosides, 183-195, 198, in compartmental models, 8, 24, 25-27
136f, 136-138, 138f, 141 200-201, 252, 253 two-compartment, 9, 82, 83f,
and calculation of time to hold a 84-87, 86f-87f
theophylline clearance in, 221, 222t
dose, 196, 196f, 253 in steady-state, 53, 54-55, 58
Distribution of drugs, 115-126 extended-interval, 182, 196-200,
kidney function in, 23, 127, 128,
definition of, 267 197t, 253
138-140
discussion points on, 126 for digoxin, 243
liver function in, 23, 127-135
disease states affecting, 116, 121-122, for phenytoin, 232
maximum elimination rate in, 151-152
160-161 for theophylline, 224, 228
metabolism in, 127-134
in obesity, 162-163 for vancomycin, 254, 255
Michaelis−Menten kinetics in, 151-154
in one-compartment model, 7, 8, 8f, 9f and time to hold a dose, 215-216
nonlinear, 150t, 150-154
in two-compartment model, 7, 8, 9, 9f, calculation of, 204, 205, 254, 255
82, 83f, 84-86 review questions and answers on,
in case studies, 204, 205-207, 209,
143-146
lipid solubility affecting, 116 211-218
saturable, 150t, 150-153, 151f
model-independent relationships in, in intermittent infusions, 73f, 73-74
166-167 total body elimination in, 138
in intravenous bolus administration,
perfusion-limited, 116 52-53 zero-order, 25, 25f, 26, 26f, 26t, 27f, 151
permeability-limited, 116 Drug distribution, 267. See also Distribution and plasma drug concentration,
of drugs 26, 26f
physiologic model of, 117-118
compared to first-order
processes in, 99, 100f Drug effect, 1, 2-3, 3f, 4f
elimination, 25f
protein-binding affecting, 67, 118-121 and absorption rate in chronic illnesses,
definition of, 269
108
regional pH levels affecting, 117 Elimination rate
and duration of drug presence at site of
review questions and answers on, action, 6 definition of, 8
123-125
discussion of, 44

276
review questions and answers, 40-43 for vancomycin, 206, 215, 216, 254 13-1: volume of distribution for vancomycin,
204, 206, 212, 254
Elimination rate constant (K), 9, 33, 34-36, 50, 3-3: half-life calculation, 36, 39, 103, 250
82, 84, 87 for aminoglycosides, 185, 190, 192 13-2: elimination rate constant for vanco-
and absorption rate constant, 104 mycin, 204, 205, 206, 212, 254
for vancomycin, 208, 210
and creatinine clearance rate, 140-141, 13-3: maintenance dose for vancomycin, 205,
3-4: clearance related to volume of distribu- 207-208, 211-216, 218, 254, 255
141f, 161 tion and elimination rate constant, 37,
for aminoglycosides, 199, 183, 184, 13-4: dosing interval for vancomycin, 205,
39, 104, 250
185, 251 209, 211, 213-214, 254, 255
for theophylline, 226
and half-life, 35, 35f, 36, 102, 160, 161, 13-5: trough concentration for vancomycin,
3-5: area under the plasma concentration 206, 211, 213, 215, 217, 254, 255
188, 192
versus time curve, 38, 108, 167, 250
and percent remaining after time (e-Kt), 14-1: volume of distribution for theophylline
4-1: accumulation factor not at steady state, or aminophylline, 223, 225, 255
34, 58
52, 250
and total body clearance, 166, 167 14-2: loading dose for theophylline or
4-2: accumulation factor at steady state, 56, aminophylline, 223, 225, 227, 255
changes affecting plasma drug concen- 250
trations, 65-66, 60f, 102, 102f 14-3: clearance of theophylline or
for aminoglycosides, 190 aminophylline, 223, 225, 227, 228, 255
definition of, 267
4-3: average concentrations from AUC and 14-4: maintenance dose for theophylline or
for aminoglycosides, 140, 141f, 181-183, dosing interval, 57, 59, 107, 108, 250 aminophylline, 224, 225, 226, 227, 228,
185-187, 189, 192-194, 200 for digoxin, 243 255
calculation of, 181, 182-186, 200,
for theophylline, 224, 227 15-1: steady-state concentration for
253
phenytoin, 233
and creatinine clearance rate, 183, 5-1: maintenance dose, 74, 161, 252, 253
184, 185, 251 for aminoglycosides, 185-187, 192, 193, 15-2: steady-state volume of distribution for
194, 253, 254 digoxin, 242, 247, 258
for digoxin, 245, 246
for gentamicin, 253 15-3: systemic clearance of digoxin, 243, 246,
for metabolites, 128, 128f 258
for theophylline, 226 for tobramycin, 253
15-4: steady-state concentration and systemic
for vancomycin, 204, 206-208, 208f, for vancomycin, 205, 254 clearance of digoxin, 243, 244, 246, 258
209-210, 212-214, 217, 218 5-2: trough concentration for intermittent Erythromycin
calculation of, 208, 208f, 209, 210, infusions, 74 affecting metabolism of other drugs,
212-215, 217-218, 254, 255 9-1: Cockcroft–Gault for creatinine clearance, 131t, 222t
in one-compartment model, 8f, 102, 249 141, 251 intrinsic clearance of, 134t
in steady state, 56 for aminoglycosides, 182
Ethosuximide, 6t
relation to clearance and volume of for digoxin, 243
Excretion of drugs, 99, 100f, 127-131, 133,
distribution, 35, 103, 250 9-2: ideal body weight estimation, 141, 182, 138-140
Enalapril metabolism, 132t 251 rate calculation, 139
Enoxaparin, 37t, 54t 9-3: adjusted body weight, 141, 182, 251
Exponential key of calculator, 14
Enteric-coated products, 106 10-1: Michaelis−Menten, 152, 251, 256, 257
Extended-interval aminoglycoside dosing,
Enzymes for phenytoin, 231, 233, 234, 235, 237, 197t, 197-199, 253
239, 240, 251, 256, 257
drug-metabolizing, 130-131, 131t Extracellular fluid, 22, 22f, 160f, 162
10-2: Michaelis constant, 152, 251, 257
inducers affecting formation clearance of Extraction ratio (E), 132-134, 134t, 135-137
metabolites, 169-170, 170t for phenytoin, 239, 257
affecting clearance, 24, 24t
kinetics of, 150, 151, 151f 10-3: Michaelis−Menten rearrangement in and first-pass effect, 134-135
nonlinear model, 153, 251
saturable, 150, 151, 256 definition of, 268
10-4: time required for 90% of steady-state
Equations, 249-258 concentration to be reached, 154, 256
1-1: concentration related to dose and volume for phenytoin, 236, 256
of distribution, 10, 11, 21, 71, 132, 249,
256 12-1: elimination rate constant based on
creatinine clearance, 183, 185, 251
F
for aminophylline, 223
for aminoglycosides, 183, 185, 251 Fat tissue
for phenytoin, 233, 235, 256
12-2: volume of distribution based on ideal age-related changes in, 160, 160f
body weight, 183, 185, 251
2-1: total body clearance, 23, 133, 249 and pharmacokinetic variations in
for aminoglycosides, 183, 185, 251 obesity, 162-163
2-2: organ clearance, 24, 133, 249 12-3: volume of distribution based on drug concentration in, 2, 116, 160
3-1: elimination rate constant, 34, 85, 103, adjusted body weight, 183, 185, 251 of aminoglycosides, 182
249, 252, 254 for aminoglycosides, 183, 185, 251
for aminoglycosides, 192, 252 50% effective concentration (EC50), 3, 4f, 267
12-4: dosing interval 252, 253, 254
for vancomycin, 208, 210, 214, 216, 254 First-order elimination, 25t, 25-27, 25f-27f
for aminoglycosides, 186, 194, 252
and mixed-order pharmacokinetic, 150
3-2: concentration at any given time, 34, 39, for vancomycin, 205, 254
50, 51, 55, 71, 85, 86, 250, 252, 253 and plasma drug concentration, 12, 25t,
12-5: loading dose based on maintenance 25-27, 26f-27, 82, 86f
for aminoglycosides, 187-188, 192, 195,
dose and accumulation factor, 252 in AUC calculation, 38
206, 252
for aminoglycosides, 189-190, 252 straight-line slope of, 33, 34
definition of, 268

Index |
277
in intravenous bolus administration, steady-state concentration of, 53, 54, Highly blood-perfused (central)
50-55 54t, 58 compartment, 7, 7f
in oral administration, 105, 105f volume of distribution, 11, 160-161 Homogeneity, kinetic, 1, 2, 268
nonlinear processes compared to, 231, Glomerular filtration rate, 138-140, 140f Hydrolysis in drug metabolism, 132t
232f, 233 age-related changes in, 179-181, 180t, Hydrophilic drugs
zero-order elimination compared to, 181f elimination of, 127
25f-26f and drug clearance, 140-142, 141f measurement of plasma concentrations,
First-pass effect, 134-135, 268 creatinine clearance as measure of, 164
Fluconazole affecting metabolism of other 140-142, 179-181, 180t Hydrophobic drugs, 118
drugs, 131t factors affecting, 139 Hydroxylation in drug metabolism, 132t
Fluids of body, 22, 22f in chronic renal disease, 179, 180t Hypoalbuminemia
age-related changes in, 160, 160f MDRD equation in estimation of, phenytoin concentration in, 119, 119t
and volume of drug distribution, 10, 159-181
volume of drug distribution in, 118
67, 68 Glucose, tubular reabsorption of, 139
in traumatic or burn injuries, 68
Glucuronidation in drug metabolism, 132t
extracellular, 22, 22f, 160f, 162
Glycoprotein
interstitial, 22, 22f alpha-1-acid glycoprotein binding to I
intracellular, 22, 22f, 160f drugs, 118, 118t, 120
measurement of drug concentration in, in disease states, 120, 137 Ibuprofen metabolism, 132t
1, 2f variations in P-glycoprotein expression, Ideal body weight. See Body weight, ideal
percentage of body weight, 22 162 Ifosfamide metabolism, 131t
Fluoxetine affecting metabolism of other Grapefruit juice affecting drug metabolism, Imipramine metabolism, 132t
drugs, 131t 131t Inactivation process in drug assays, 165-166
Fluvoxamine affecting metabolism of other Indinavir affecting metabolism of other drugs,
drugs, 131t 131t
Food−drug interactions, 130, 131t
Formation clearance
H Infusions
continuous, 68-71
as model-independent parameter, intermittent, 73-74
Half-life (T½), 23, 31, 35f, 35-37, 250
166-167, 169f, 169-170, 170t
and dosing interval, 51, 52, 53 short, 73, 73f
definition of, 268
and elimination rate constant, 37-38, Interactions of drugs, 130, 135-136
Fosphenytoin, 153, 232, 233-234 102, 160-161, 188-189 in assay methods, 165, 166
and peak concentration, 37 hepatic cytochrome P450 enzyme
and steady state concentration, 53, 54, system in, 130, 131t
G 54f, 54t, 55 of theophylline with other drugs, 222t
definition of, 35, 268 cimetidine, 131t, 136f, 163, 222t
Gabapentin, 104, 150t discussion points, 44 protein binding in, 118-120
Ganciclovir, 10t, 25t estimation of, 35f, 35-37, 102 renal clearance in, 67, 139-140
Gastrointestinal tract, absorption of drugs in example of, 36t volume of distribution in, 120-121
99, 100, 100f, 101 in two-compartment models, 82, 84, 87 Interferences in drug assays, 164-165
absorption rate constant in, 104 of aminoglycosides, 185, 188, 191, 192, Interferon affecting theophylline clearance,
and first-pass effect, 134-135 195 222t
in controlled-release products, 106 of commonly used drugs, 37, 37t Intermittent infusions, 73-74
nonlinear, 150 of theophylline, 224 Interstitial fluid, 22, 22f
Genetic factors affecting drug metabolism, of vancomycin, 37, 37t, 87, 204, 206-210, Intracellular fluid, 22, 22f, 160f
131-132, 161-162 216
Intramuscular administration of drugs,
Gentamicin review questions and answers, 40-43 absorption and bioavailability in, 102,
accumulation of, 52 Haloperidol metabolism, 131t 103, 104f
cross-reactivity in assays, 164 Hartford nomogram, 197-198, 198f, 199, 201 and peak plasma concentration, 103,
104f
desired plasma concentration of, 184 Heart failure
and volume of distribution, 102-103
dose adjustment in renal dysfunction, digoxin in, 242, 243-247
74, 160-161 theophylline clearance in, 222t Intravenous administration
dosing regimens for, 182 bolus dosing in, 7, 8, 11, 12f
Hemoglobin as assay interference, 165
extended-interval dosing of, 197, 197t, multiple dosing and drug concentrations
Heparin loading dose in continuous IV
198, 198f, 253 in. See Multiple IV dosing affecting
infusions, 72 drug concentrations
half-life of, 37t, 52, 53 Hepatic drug metabolism. See Liver in drug peak plasma concentration in, 50, 50f,
peak and trough levels of, 160, 184 metabolism 52, 53f, 55f, 56-58
prediction of plasma concentration of, Hepatocytes in drug metabolism and accumulation factor in, 52
13, 13f, 35, 74, 160-161 excretion, 129, 130f, 133
protein binding of, 105t

278
and absorption rate, 103 Leflunomide affecting metabolism of other Logarithms

at steady state, 53, 53f, 55, 56, 58 drugs, 131t and semilog graph paper, 12f, 12-13, 13f,
compared to intramuscular admin- Levodopa metabolism, 132t 21, 34
istration, 103 Levofloxacin, 37t, 54t common, 14
in therapeutic range, 55-56, 56f Lidocaine natural, 13-14, 19 See also Natural log of
prediction of, 53, 58 clearance of, 134, 134t drug concentrations
Intrinsic clearance of drugs, 133-134, 134t loading dose in continuous IV infusions, Loratadine metabolism, 131t
Isoniazid 72 Lorazepam metabolism, 132t
affecting metabolism of other drugs, measurement of plasma concentrations, Lovastatin, 54, 131t
131t 164
bimodal pattern of elimination, 162, metabolism of, 131t
162f protein binding of, 118t, 119t
intrinsic clearance of, 134t in myocardial infarction, 137-138, M
metabolism of, 132t 138f
steady-state plasma concentration in Maintenance dose, 68, 167
Isoproterenol, intrinsic clearance of, 134t
myocardial infarction, 137-138, for aminoglycosides, 183, 198, 252, 253
Itraconazole affecting metabolism of other
138f in case studies, 184-185, 187,
drugs, 131t
therapeutic range for, 6t 188-190, 193-195, 200
Linear pharmacokinetics, 149 for aminophylline or theophylline,
224-228, 255
Linear regression techniques, 13, 13f
J Lipid solubility of drugs affecting distribution,
for digoxin, 242, 244-247
adjustment in renal disorders, 244,
116
Jar model of hepatic clearance, 133 245
Lipophilic drugs
for phenytoin, 233, 234-236, 237, 240,
crossing membrane barriers, 116 241
pharmacokinetics in obesity, 162 for vancomycin, 208-209, 211-215, 217,
K properties of, 127 254, 255
Lipoproteins binding to drugs, 119, 119t initial, 204
Kanamycin cross-reactivity in assays, 164 Lithium total body clearance in calculation of,
Ketoconazole affecting metabolism of other 167
half-life of, 37t
drugs, 131t Marijuana affecting theophylline clearance,
steady-state concentration of, 54t
Ketorolac, 10t, 25t 222t
therapeutic range for, 6t
Kidneys MDRD (Modified Diet in Renal Disease)
Liver in drug metabolism, 23, 67, 127-138 equation, 179-181, 180t, 181f
chronic disease of, 179-181
age-related changes in, 159 Mean residence time, 166, 168-169, 268
stages in, 179-180, 180t
anatomy and physiology in, 129f, Meperidine, intrinsic clearance of, 134t
failure of
129-131, 130f
dose adjustment in, 67, 160-161 Mephenytoin metabolism, 131t
and bioavailability, 101
protein-binding of drugs in, Metabolism of drugs, 127-142
121-122, 138-139, 231 and clearance, 23, 67, 132-134
biotransformation processes in, 128,
volume of drug distribution in, biotransformation in, 131-132 129, 130, 131-132
121-122, 161 blood flow affecting, 23, 129-130, 130f, cytochrome P450 enzyme system in,
function assessment, 179-181 133-134, 137-138 130, 131t, 136
in drug clearance, 23, 67, 127, 128, disease states and drug interactions genetic factors affecting, 131-132,
138-140, affecting, 135-138, 160, 161 161-162
age-related changes in, 159 extraction ratio in, 135-136 liver function in. See Liver in drug
calculation of, 137-138 first-pass effect in, 134-135 metabolism
drug interactions affecting, 67, genetic factors affecting, 131-132 of phenytoin, 131t, 132t, 233, 235, 236
137-138 of theophylline, 131t, 221
of propranolol, 24
glomerular filtration in, 138-140, cimetidine affecting, 131t, 136f,
140f of theophylline, 221, 222t
163, 222t
mechanisms in, 138 Loading dose, 71f, 71-72
phase I, 129-130, 132t
nonlinear processes in, 150t and volume of distribution, 71, 117
phase II, 129, 130, 132t
nephrotoxicity of vancomycin, 217 as short infusion, 73, 73f
preparatory reactions in, 130
tubular reabsorption in, 138, 139, 179 for aminoglycosides, 184, 187, 188, 188f,
189-190, 252 Metabolites, 5, 128, 128f, 129f
tubular secretion, 138, 139, 140, 179
for aminophylline or theophylline, formation clearance of, 128
Kinetic homogeneity, 1, 2, 268 222-223, 223f, 225, 227, 255 as model-independent parameter,
for digoxin, 242-243 160, 160f, 169f, 169-170, 170t
elimination rate constant in, 128,
for phenytoin, 233-234, 236, 256
129f
L for vancomycin, 106, 207, 207f, 210,
pharmacokinetics of, 128
210f, 211
Lansoprazole, 10t, 25t

Index |
279
Methotrexate, nonlinear pharmacokinetics Monoexponential equation, 86 Netilmicin cross-reactivity in assays, 164

of, 150t Montelukast, 10t, 25t Nifedipine, 106t, 131t
Methylation in drug metabolism, 132t Morphine, 132t, 134t Nitroglycerin, 4, 134t
Methyldopa metabolism, 132t Multicompartment models, 7, 81 Nonlinear pharmacokinetics, 149-158
Methylxanthines, 221-229 Multiple IV dosing affecting drug discussion points on, 158
aminophylline. See Aminophylline concentrations, 65, 66f Michaelis−Menten, 151-154. See also
theophylline. See Theophylline accumulation factor in, 52-53, 56-57 Michaelis−Menten pharmacoki-
bolus injections in, 8, 11, 12f netics
Metoprolol metabolism, 131t
discussion points on, 64 of theophylline, 150, 150t, 151, 221
Metronidazole affecting metabolism of other
drugs, 131t first dose in, 50, 50f review questions and answers on,
155-157
Michaelis constant (Km), 152-154, 251 intermittent doses in, 73-75
calculation of, 152 Norfloxacin affecting theophylline clearance,
in one-compartment model with first- 222t
for phenytoin, 231, 232, 232f, 235, 237, order elimination, 50, 55, 65, 66f
241, 256 Nutrient−drug interactions, 130, 131t
peak and trough concentrations in,
Michaelis−Menten pharmacokinetics, 52-53 55, 57, 161
152-154, 221, 251 at steady state, prediction of, 53, 58
dose calculation in, 153-154
drug concentration and elimination rate
in intermittent infusions, 73, 74
O
review questions and answers on, 60-63
in, 152-154 second dose in, 50, 50f Obesity
linear plot of, 152, 153f accumulation factor in, 53 creatinine clearance in, 141, 182
maximum elimination rate in, 152-153 concentration just before next pharmacokinetic variations in, 162-163
of phenytoin, 149, 150-153, 153f, 154, dose, 51, 51f
plasma concentration variability in, 5
231, 232, 232f, 241, 256-257 maximum concentration after, 51f,
clearance in, 231, 232, 232f 51-52 Omeprazole
in case studies, 235, 237 steady-state concentrations in, 53-55 affecting metabolism of other drugs,
131t
linear plot of, 153f average, 57f, 57-58
metabolism of, 131t, 132t
plasma concentration in, 231, 232, superposition principle in, 50, 74
232f One-compartment models, 7, 7f, 8f, 10
third dose in, 51-52
steady-state concentration in, 152, 153f, AUC calculation in, 38
accumulation factor in, 52
153-154 bioavailability in, 102-103
Multiplicative linear model, 180
Microconstants, 82, 87 compared to two-compartment models,
Myocardium 8, 9f, 82f
Midazolam metabolism, 131t
drug concentration in, 1, 2
Minimum inhibitory concentration, 196, 268 elimination processes in, 8f, 26f, 26-27,
infarction of, protein binding of lidocaine 27f, 249
Model, definition of, 268 in, 137-138, 138f
intravenous bolus administration in,
Model-dependent pharmacokinetics, 24 50-54, 65, 66f
Model-independent pharmacokinetics, plasma drug concentration versus time
166-171, 268 curve in, 11, 11f, 12, 12f
advantages and disadvantages of, N absorption and elimination rates
166-167 affecting, 103, 103f
definition of, 268 Nafcillin affecting metabolism of other drugs, with first-order elimination, 26,
131t
discussion points on, 174 26f, 65, 82f, 86f
Natural log of drug concentrations with zero-order elimination, 25,
in clearance, 24, 38
and prediction of concentrations at 26, 26t
formation, 167, 169-170
times after dose, 33
total body, 166, 167-168 vancomycin in, 205, 205f, 207, 207f, 212
base of, 34
in mean residence time, 166, 168-169 Opiates, tolerance to, 3
calculator keys for, 14
in volume of distribution at steady-state, Oral administration of drugs
167, 169 time plot in, 12, 12f absorption in, 99-108
and half-life determination, 35, 35f
review questions and answers on, and bioavailability, 101, 102-103
171-173 and slope of straight-line plot, 32, and fraction reaching systemic
32f, ,33, 34
Modified Diet in Renal Disease (MDRD) circulation, 101, 135
in two-compartment models,
equation, 179-181, 180t, 181f and peak plasma concentration,
81-82, 82f, 84
Monitoring of drug levels, 4-5 103, 104f
with first-order and zero-order
dosage decisions in, 6, 7f and volume of distribution,
elimination, 27, 27f 103-104
elimination rate in, 28 Negative slope of back-extrapolated line, 85f, of controlled-release formulations,
limitations of, 5 86 106, 106f
of phenytoin, long-term, 240 Neonates processes involved in, 100f
sample collection and handling for, chloramphenicol toxicity in, 131 AUC calculation in, 39, 101
163-164 pharmacokinetic variations in, 159, 160 first-pass effect in, 134-135
value of, 5 Nephrotoxicity of vancomycin, 217

280
of theophylline, 21, 228 Pharmacodynamics case studies on, 233-240

plasma drug concentration versus time and tolerance to drug effect, 3 clearance of, 231, 232f, 235-236, 237,
curve in, 102, 102f, 103 basic concepts of, 2-4 238-239, 239f
Oxcarbazepine affecting metabolism of other definition of, 2, 268 intrinsic, 134t
drugs, 131t introduction to, 1-19 daily dose of, 232, 256
Oxidation in drug metabolism, 131t clinically important equations for, and steady-state plasma concentra-
14 tion, 237-239, 239f
discussion points on, 19 increase in, 237
review questions and answers on, discussion points on, 248
P 15-18 dosing equations for, 256-257
relationship to pharmacokinetics, 4, 7, 7f dosing interval for, 232
Parameters in pharmacokinetics
in two-compartment models, 84-86 Pharmacogenomics, 161 extraction ratio for, 134
model-independent, 166-170, 268 Pharmacokinetics loading dose of, 233-234, 256
and drug concentrations in plasma and long-term monitoring of, 240
relationships of, 38-39, 65-80
tissues, 1-2, 2f
maintenance dose of, 233-234
Paroxetine metabolism, 131t basic concepts in, 20-30 adjustment of, 234-237, 241
Peak concentration of drugs discussion points on, 30
measurement of plasma concentrations,
and half-life, 37 review questions and answers on, 164, 231, 233-235, 236, 237-238,
dosing interval affecting, 68 28-29 240
elimination rate constant affecting, 65 clinical, 267 metabolism of, 131t, 132t, 231, 256
in controlled-release products, 106, 106f compartmental models of, 7-9, 7f-9f Michaelis−Menten pharmacokinetics of,
in extended interval dosing, 184-185 definition of, 1, 268 151-154, 153f, 231, 231-232, 232f,
dose-dependent, 149-150, 150t, 150f- 235-236, 256-257
in intermittent infusions, 73
151f clearance in, 231-232, 232f
in intramuscular administration, 102,
102f of enzyme-saturable drugs, in case studies, 235-236
150-151, 151f linear plot of, 153f
in intravenous administration, 51-52,
52f introduction to, 1-19 plasma concentration in 231-232,
discussion points on, 19 232f
and absorption rate, 103 review questions and answers on, nonlinear pharmacokinetics of, 150,
15-18 150t, 231, 233
at steady state, 53, 53f, 55, 56, 58
kinetic homogeneity in, 1, 2, 268 protein binding of, 67, 118t, 119, 119t,
compared to intramuscular admin-
231
istration, 103 linear, 149
albumin serum levels affecting,
in therapeutic range, 55-56, 56f Michaelis−Menten, 151-154. See also 119t
prediction of, 53, 58 Michaelis−Menten pharmacoki-
netics in renal failure, 121, 136-137, 137f
in oral administration, 104, 104f valproic acid affecting, 120-121
of aminoglycosides, 160-161, 184-185, mixed-order, 150
steady-state plasma concentration of,
188, 188f, 192, 197 model-independent, 166-170. See also
231, 235-236, 237-239, 239f,
of vancomycin, 204f-205f, 205-207, Model independent pharmaco-
240-241, 256-257
207f-208f, 209-212, 214, 217, kinetics
and daily dose, 237, 238, 239, 239f
254-255 nonlinear, 149-159
in renal failure, 137, 137f
at steady state, 204-205, 209, relationship to pharmacodynamics, 7, 7f
212-214 supratherapeutic unbound concentra-
relationships among parameters in, tions of, 121
desired, 205-207, 210, 213, 38-39, 65-80. See also Rela-
215-217 tionships of pharmacokinetic therapeutic range for, 6t
measured, 214, 216, 217 parameters volume of distribution, 67, 117, 231,
and tolerance to drug effect, 3 233, 235, 236, 238, 256
volume of distribution affecting, 67
sources of variation in, 159-163 and steady-state concentration,
Penicillin 235, 238
and aminoglycoside interactions, time-dependent, 150t
in renal failure, 121
165-166 Phenobarbital
valproic acid affecting, 120-121
nonlinear pharmacokinetics of, 150,152t affecting metabolism of other drugs,
131t, 131, 222t zero-order elimination of, 26
urinary excretion of, 138, 139
intrinsic clearance of, 134t Piroxicam metabolism, 131t
Pentazocine, intrinsic clearance of, 134t
measurement of plasma concentrations, Plasma
Perfusion-limited distribution, 116 compared to whole blood and serum, 22,
164
Peripheral compartment, 7, 7f, 8, 9, 9f, 81-82, 22f, 163
steady-state concentration of, 53
83f, 84, 85 concentration of drugs in. See Concen-
amount of drug in, 9, 83f, 857 therapeutic range for, 6t
tration of drugs
Permeability-limited distribution, 116 Phenothiazine metabolism, 131t
creatinine concentration in, 179
pH Phenytoin, 231-241, 248
definition of, 268
of body areas affecting drug distribu- adverse effects of, 240
distribution in body, 22, 22f
tion, 117 affecting metabolism of other drugs,
131t, 131, 222t fraction of unbound drug in, 119, 120,
urinary, affecting tubular reabsorption 133, 135
of drugs, 139
Index |
281
protein binding in, 118-120. See also and effects of bound and unbound Renal conditions. See Kidneys
Protein binding of drugs proteins, 118 Residence time, mean, 166, 168-169, 268
volume of, 117, 120 and volume of distribution, 67, 118-12 Residual line, 85-86, 85f-86f, 86t
Polarization, and drug concentration, 165, association and dissociation process in, negative slope of, 85, 85f, 86, 105
165f 118, 118f
Residuals method, 85-86
Polymorphism, genetic, 131-132, 161 clinical importance of, 120
concentration points in, 85, 86t
Population estimates, 166 drug interactions affecting, 120-121
for absorption rate constant, 105
for aminoglycosides, 181, 184, 185, 187, in myocardial infarction, 137-138, 138f
189, 191, 192, 251 Resistance to aminoglycosides, 197
in renal failure, 121-122, 136-137
for digoxin, 244 Riboflavin, nonlinear pharmacokinetics of,
of digoxin, 11t, 121, 122, 258 150t
for phenytoin, 233, 234-235, 237, 238, of commonly used agents, 118, 118t
241, 256 Rifampin affecting metabolism of other drugs,
of digoxin, 118t, 121, 122 131t, 222t
for theophylline, 223, 225, 227 in renal failure, 121-122, 244, 245 Risperidone metabolism, 131t
for vancomycin, 204-207, 212, 213, 218, of phenytoin. See Phenytoin, protein Ritonavir affecting metabolism of other drugs,
254 binding of 131t
Portal circulation, 134, 135f
Post-antibiotic effect (PAE) of
aminoglycosides, 197
Potassium chloride, controlled-release Q S
formulation of, 106t
Potency of drugs, 4 Quality control check in drug assays, 165 St. John’s wort-drug interactions, 131t
Practice sets, 45-47, 93-97, 175-178 Quinidine Salicylates, nonlinear pharmacokinetics of,
Prediction of drug concentration affecting metabolism of other drugs, 150t
131t Sample collection and handling, 163-164, 165
interaction with digoxin, 121 Sampling times for assays, 165
at any given time, 34
intrinsic clearance of, 134t Sanford Guide to Antimicrobial Therapy, 197,
at steady state, 58
measurement of plasma concentrations, 197t, 199
for time curve, 11f, 11-12
164 Saturable elimination, 150, 150t, 151, 151f
in continuous infusions 71
therapeutic range for, 6t Secobarbital affecting metabolism of other
in intermittent infusions, 73-74 drugs, 131t
in linear relationships, 13, 13f, 149 Semilog graph paper, 12f, 12-13, 13f, 27, 34
of gentamicin, 13, 13f, 35, 74 Sensitivity of drug assay, 164, 165
peak and trough levels in, 160 R Sertraline, 131t
relationships of pharmacokinetic
Ranitidine, steady-state concentration of, 54 Serum
parameters in, 38
Rates compared to whole blood and plasma,
with compartmental models, 7 22, 22f, 163
with controlled-release absorption rate constant, 104-105
definition of, 268
preparations,106 elimination rate constant. See
Elimination rate constant drug concentrations in, 22, 164
with Michaelis−Menten pharmaco-
kinetics, 151-154 of drug infusion, 67, 68, 69 separator tube, 163
with semilog scale, 12f, 12-13, 13f, 27, Receptors Short IV infusions in loading dose, 73, 73f
27f, 34 concentration of drugs at, 1, 2-3, 3f Sildenafil, 10t, 25t
Prediction of drug effects, 4, 4f definition of, 268 Slope of curvilinear plot, 84-86, 84f-85f
Prednisone, 128, 131t Red-man syndrome, 207 alpha, 85, 85f
Pregnancy, pharmacokinetic variations in, 163 Relationships of pharmacokinetic parameters, beta, 84f, 85f, 85-86
Primidone, therapeutic range for, 6t 38-39, 38f-39f, 65-80 in back-extrapolated line, 85
Probenecid interaction with penicillin, 139 discussion points on, 80 in initial portion, 86
Procainamide, intrinsic clearance of, 134t in clearance changes, 67-68 in residual line, 85f, 85-86, 86t
Pro-drugs, 128, 232 in continuous infusions, 68-72 in terminal portion, 86
Prolonged-action formulations, 106 in dose changes, 66, 66f Slope of straight-line plot, 31-32, 32f, 33
Propoxyphene, 131t, 134t in dosing interval changes, 66, 66f, 73 and elimination rate, 32f, 32-33, 103
Propranolol in elimination rate constant changes, determination of, 32, 32f
affecting theophylline clearance, 222t 65-66, 67, 72
Specificity of drug assays, 164
clearance of, 24, 134, 134t in intermittent infusions, 73-75
Steady-state concentration of drugs, 53-55,
extraction ratio for, 24, 134 in volume of distribution changes, 67f, 137, 137f, 138, 138f, 251
67-68 accumulation factor in, 56
metabolism of, 24, 131t
model-independent, 166-170 for aminoglycosides, 190
Propylene glycol, 234, 246
of whole blood, plasma and serum, 22, and volume of distribution, 55, 58, 67,
Protein as assay interference, 165 22f 232
Protein binding of drugs, 118-120 review questions and answers on, 76-79 in two-compartment models, 86-87

282
model-independent relationships, aminophylline dose equivalent, 221, 223 and first-order elimination, 25-27, 26f,
166, 169-170 case studies on, 221-228 26t, 27f
average concentration in multiple clearance of 134t, 136, 221, 223-228, and plasma concentration of drug, 1-2,
dosing, 57f, 57-58 255 4-5, 5f, 11-12, 11f-13f, 19f
definition of, 268 cimetidine affecting, 136f, 222t elimination rate constant in, 33,
34-35
disease states affecting, 137-138, 137f- disease states and drugs affecting,
138f 221, 222t in half-life, 35f, 35-36
dosing interval in, 53-54, 55, 55f, 55-56 in case studies, 222-228 natural log of, 31-32, 32f, 33
with controlled-release products, prediction of, 33
drug interactions affecting, 137, 137f
107-108 and zero-order elimination, 25f, 25t
elimination processes in, 53, 54, 54f,
55, 58 continuous IV infusion of, 69, 70, 73, Time-dependent pharmacokinetics, 150t
221, 224, 224f
for commonly used drugs, time to reach, Time zero (t0), 21, 22f, 33
54t, 54-55 loading dose in, 72
in two-compartment models, 84
half-life in, 53, 54, 54f, 54t, 55 controlled-release formulations of, 106,
106t Tissue drug concentrations, 1, 2, 2f, 6-9, 22,
in continuous infusions, 68-71, 69f-70f 116
clearance estimation, 107
with loading dose, 71 and plasma concentration, 117, 117f
discussion points on, 229
in Michaelis−Menten kinetics, 152-153, disease states affecting, 116
153f, 154 dose adjustment in liver dysfunction,
161 fraction of unbound drug in, 117, 120
in oral administration, 105
dosing equations on, 255 Tobacco affecting drug metabolism, 131t
methods increasing, 55, 55f of theophylline, 131t, 222t, 225
dosing interval for, 224, 228
of aminoglycosides, 188, 189, 191,
elimination rate constant for, 226 Tobramycin dosing regimens, 182
194-195, 200
first-order pharmacokinetics of, 221 and desired plasma concentration, 184
of digoxin, 53, 54t, 243-246
half-life of, 224 extended-interval, 197, 197t, 198, 198f,
of phenytoin, 231, 235-236, 237-239, 199, 253
239f, 240-241, 256 infusion rate of, 223, 225
Tolerance to drug effect, 3-4, 4f, 269
and daily dose, 237, 238, 239, 239f interaction with other drugs, 222t
in renal failure, 136-137, 136f-137f Total body elimination, 138
cimetidine, 131t, 136t, 163, 222t
of theophylline, 72, 224, 225, 226, 228 ciprofloxacin, 136 Toxic effects, 4
of chloramphenicol, 131
of vancomycin, 54t, 204-205, 209, oral contraceptives, 222t
212-214 interpatient variability in concentration of vancomycin, 203
prediction of, 58 and response to, 5, 6f Trapezoidal rule
therapeutic range in, 56-57, 57f loading dose of, 222-223, 223f, 225, 227, in AUC calculation, 38-39, 39f, 39, 103,
255 167
time required for 90% to be reached,
154, 154f maintenance dose of, 224-228, 255 in AUMC calculation, 168, 168f
for phenytoin, 236-237, 238, 256 metabolism of, 131t, 221 Trauma, volume of drug distribution in, 68
with controlled-release preparations cimetidine affecting, 131t, 136f, Tricyclic antidepressants, metabolism of, 131t
106, 106f 163, 222t
Triglycerides as assay interference, 165
Straight-line plots, 13, 13f ciprofloxacin affecting, 136
Trough concentrations of drugs
equation for, 32, 33 nonlinear pharmacokinetics of, 150,
dosing interval affecting, 66
150t, 151, 221
slope of, 32f, 32-33 elimination rate constant affecting, 65
and elimination rate, 32f, 32-33, oral administration of, 221, 228
in extended interval dosing, 199-201
103, 103f steady-state concentration of, 69, 70,
224, 225, 226, 228 in intermittent infusions, 73
y-intercept in, 31, 32f
sustained-release of, 106, 228 in intravenous bolus administration,
Sulfasalazine, 128, 132t 52-53, 52f
volume of distribution, 222, 223, 225,
Sulfonamide metabolism, 132t at steady state, 53, 55, 56, 57
226, 227, 255
Sulfoxidation in drug metabolism, 132t in therapeutic range, 55-56, 56f
Therapeutic drug monitoring, 4-6. See also
Superposition principle, 50, 74 Monitoring of drug levels prediction of, 53, 55
Sustained-release formulations, 103, 106, 107 definition of, 4, 268 of aminoglycosides, 160, 184-187, 188,
of theophylline, 106, 228 190-197, 191f, 199, 200
Therapeutic range, 4, 4f, 6, 6t
Synthetic reactions in drug metabolism, 130 of vancomycin, 203-206, 208-216, 214f,
definition of, 268
217-218, 254-255
in continuous IV infusions with loading at steady state, 204-205, 209,
dose, 71 212-214
maintenance of plasma drug concentra-
T tions in, 55-56, 56f
desired, 205, 206, 213, 215, 216,
217-218
Tacrine, 131t Thiopental pharmacokinetic variations in measured, 214, 216, 217-218
obesity, 162 with controlled-release formulations,
Tacrolimus metabolism, 131t
Time after dose 106, 107
Tamoxifen metabolism, 132t
and clearance, 23, 23f Tubular reabsorption of drugs, 138, 139
Theophylline, 5, 6f, 221-229
and drug presence at site of action, 6 Two-compartment models, 7-8, 8f, 81-91
adverse effects of, 224, 226

Index |
283
biexponential curve in, 82, 86, 87, 87f in one-compartment model, 205, 205f, for aminoglycosides, 182-184, 191
biexponential equation in, 86 207, 207f, 212 and ideal body weight, 184, 185,
in two-compartment model, 84, 87, 203, 251
calculation of parameters in, 84-86
204f calculation of, 193-194
compared to one-compartment models,
8, 9f, 81, 82, 84, 86, 87 loading dose for, 106, 207, 207f, 210, for digoxin, 242, 247, 248, 258
210f, 211 in renal failure, 121
concentration versus time plot in, 81, 82,
82f, 84, 84f, 87 maintenance dose for, 208-209, 211-215, quinidine affecting, 121
217, 254, 255
discussion points on, 91 for phenytoin, 67, 117, 231, 233, 235,
initial, 204 236, 238, 256
drug distribution in, 8f, 8-9, 82, 83f,
84-87 peak and trough concentrations of, 204f- and steady-state concentration,
205f, 205-207, 207f-208f, 209-212, 235, 238
stages of, 83f 214, 217, 254-255 in renal failure, 121
volume of, 86-87 at steady state, 204-205, 209, valproic acid affecting, 120-121
elimination process in, 9, 82, 83f, 84-86, 212-214
86f, 87, 87f for theophylline, 222, 223, 225, 226,
desired, 205, 206, 207, 210, 213,
227, 255
review questions and answers on, 88-90 215-217
measured, 214, 216, 217 for vancomycin, 116, 204, 254
vancomycin in, 82, 84, 87, 203, 204f
calculation of, 206, 208, 210, 214,
plasma concentration versus time curve
217
for, 203, 204f, 205, 205f, 208
in case studies, 206-210, 212-215,
prediction of plasma concentrations, 217-218
U 34, 34f
in central compartment, 86
protein binding of, 118t
Unbound drug in plasma, fraction of, 117, in intravenous bolus administration, 58
steady-state concentration of, 54t,
118, 120, 133 204-205, 209, 212-214 in oral or intramuscular administration,
Urea, tubular reabsorption of, 139 103
volume of distribution, 116, 204, 254
Urinary excretion of drugs, 140-142 in traumatic or burn injuries, 68
calculation of, 206, 208, 210, 214,
and metabolites, 169f, 169-170, 170t 217 in two-compartment models, 86
ratio to plasma concentration, 139 in case studies, 206-210, 212-215, of commonly used drugs, 10t
217-218 physiologic model of, 117-118
Variations, pharmacokinetic, 5 protein-binding affecting, 67, 118-121
and validity of samples collected, steady-state, 55, 58, 67
V 163-166
definition of, 268
discussion points on, 174 in two-compartment model, 86
Valproic acid review questions and answers on, model-independent relationships,
affecting metabolism of other drugs, 171-173 166, 168
120, 131 sources of, 159-163
and phenytoin volume of distribution,
Venlafaxine metabolism, 131t
120-121
Verapamil
therapeutic range for, 6t
affecting theophylline clearance, 222t
Valsartan, 10t, 25t
intrinsic clearance of, 134t W
Vancomycin, 203-219
metabolism of, 131t
biexponential elimination of, 203 Warfarin
Volume of distribution (V), 10t, 10-11, 11t, intrinsic clearance of, 134t
case studies on, 205-218
21-22, 54
cross-reactivity in assays, 164-165 metabolism of, 131t, 132
and body weight, 116
crystalline degradation product 1 for aminoglycosides, 184, 185, 251 Weight. See Body weight
(CDP-1), 164-165 Well-stirred model of hepatic clearance, 133,
and clearance, 22, 38, 103, 251
dosing equations for, 254-255 138
and elimination rate constant, 38, 103,
dosing interval for, 254 250
and time to hold a dose, 215-216 and loading dose, 71, 117
calculation of, 204, 205, 254, 255
in case studies, 204, 205-207, 209,
apparent, 21, 67 X
211-218 by area, 86, 103
changes affecting plasma drug x-axis, 31, 33, 38
elimination rate constant for, 204,
206-208, 208f, 209-210, 212-215, concentrations, 67, 68, 68f
217, 218 definition of, 269
calculation of, 208, 208f, 209, 210,
212, 213, 214, 215, 217, 218,
disease states affecting, 116, 121-122,
160-161 Y
254, 255 drug interactions affecting, 119-122
y-axis, 21, 33
first-order elimination of, 26 elimination processes affecting, 11, 11f, intercept of, 33
half-life of, 37, 37t, 87, 204, 206-210, 21
216 in back-extrapolated line, 84
equation for, 11, 21
in one-compartment model, 50
infusion rate of, 206, 212-215, 217 factors affecting, 67, 68, 116-120 in renal clearance, 140, 141, 141f

284
in two-compartment models, 84
of residual line, 85
of straight-line plot, 31, 32f
Z
Zafirlukast affecting metabolism of other
drugs, 131t
Zero-order pharmacokinetics
and mixed-order pharmacokinetics, 150
in absorption, 106
in elimination, 25f, 25-27, 26f, 26t-27t,
151
and plasma drug concentration,
25t, 26, 26f
compared to first-order
elimination, 25f
definition of, 269
Zidovudine steady-state concentration, 54


Clinical PK

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Clinical PK

Hochgeladen von

Copyright:

Verfügbare Formate

Concepts in Clinical

William J. Spruill, PharmD, FASHP, FCCP

William E. Wade, PharmD, FASHP, FCCP

Joseph T. DiPiro, PharmD

Robert A. Blouin, PharmD

Jane M. Pruemer, PharmD, BCOP, FASHP

American Society of Health-System Pharmacists®

Director, Special Publishing: Jack Bruggeman

Library of Congress Cataloging - in - Publication Data

© 2014, American Society of Health-System Pharmacists, Inc. All rights reserved.

Purchased by nova southeastern university, Todd Puccio

Lessons and Practice Sets

A Note from the Authors on Using This Edition

α : distribution rate constant for two-compartment model

e : base of natural logarithm S : salt form of drug

Purchased by nova southeastern university, Todd Puccio

Pharmacokinetics is currently defined as the study of the time course of drug

Sample Removed for Drug Concentration

FIGURE 1-1. FIGURE 1-3.

Purchased by nova southeastern university, Todd Puccio

tion may influence drug effect. This multilevel

Purchased by nova southeastern university, Todd Puccio

and application of the resulting concentration

Purchased by nova southeastern university, Todd Puccio

FIGURE 1-8. FIGURE 1-9.

Numerous pharmacokinetic characteristics

Purchased by nova southeastern university, Todd Puccio

Purchased by nova southeastern university, Todd Puccio

FIGURE 1-13. FIGURE 1-14.

Purchased by nova southeastern university, Todd Puccio

Another simplification of body processes

Purchased by nova southeastern university, Todd Puccio

The two-compartment model can be

Purchased by nova southeastern university, Todd Puccio

Purchased by nova southeastern university, Todd Puccio

Purchased by nova southeastern university, Todd Puccio

Purchased by nova southeastern university, Todd Puccio

are logarithmically related within each log cycle (or

TABLE 1-3. Time Course of Plasma Gentamicin

Purchased by nova southeastern university, Todd Puccio

There are two major keys that will be used to calcu- 2. V = X / C

Purchased by nova southeastern university, Todd Puccio

Purchased by nova southeastern university, Todd Puccio

1-15. The amount of drug per unit of volume is

Purchased by nova southeastern university, Todd Puccio

1-2. A. Incorrect answer. Pharmacodynamics 1-8. A. CORRECT ANSWER

Purchased by nova southeastern university, Todd Puccio

1-14. A. CORRECT ANSWER 1-16. A, C, D. Incorrect answers. A math error must

Purchased by nova southeastern university, Todd Puccio

Plasma Concentration (mg/L) Resulting pH

D2. The relationship shown in Figure 1-30 is

D4. Would you expect a large drug molecule that

D5. When plotting plasma drug concentration

Purchased by nova southeastern university, Todd Puccio

To examine the concept of volume of distribution further, let’s return to our

(See Equation 1-1.)

that its distribution is limited to extracellular fluid,

FIGURE 2-2. FIGURE 2-3.

Purchased by nova southeastern university, Todd Puccio

Clinical Correlate Initial

Purchased by nova southeastern university, Todd Puccio

TABLE 2-1. Rating of Extraction Ratios

FIGURE 2-6. Clearance may also be a useful parameter for