T H R T E E N T H EDITION
`
`Editors
`KURT J. ISSELBACHER, A.B.,
`M.D.
`Mallinckrodt Professor of Medicine. Harvard Medical
`School; Physician and Director, Cancer Center,
`Massachusetts General Hospital, Boston
`
`EUGENE BRAUNWALD, A.B.,
`M.D., M.A. (Hon.), M.D. (Hon.)
`Hersey Professor of the Theory and Practice of Physic,
`Harvard Medical School ; Chairman, Department of
`Medicine, Brigham and Women's Hospital, Boston
`
`JEAN D. WILSON, M.D.
`Charles Cameron Sprague Distinguished Chair and
`Professor of Internal Medicine; Chief, Division of
`Endocrinology and Metabolism, The University of Texas
`Southwestern Medical Center, Dallas
`
`McGRAW-HILL, Inc.
`Health Professions Division
`
`JOSEPH B. MARTIN~ M.D., Ph.D.,
`F.R.C.P. (C), M.A. (Hon.)
`Professor of Neurology and Chancellor, University of
`California, San Francisco
`
`ANTHONY S. FAUCI, M.D.
`Director, National Institute of Allergy and Infectious
`Diseases; Chief, Laboratory of lmmunoregulation; Director,
`Office of AIDS Research, National Institutes of Health,
`Bethesda
`
`DENNIS L. KASPER, M.D.
`William Ellery Channing Professor of Medicine, Harvard
`Medical School; Chief, Division of Infectious Diseases,
`Beth Israel Hospital; Co-Director, Channing Laboratory,
`Brigham and Women's Hospital, Boston
`
`New York St. Louis San Francisco Colorado Springs Auckland Bogota Caracas Hamburg Lisbon London
`Madrid Mexico Milan Montreal New Delhi Paris San Juan Sao Paulo Singapore Sydney Tokyo Toronto
`
`INTELGENX 1032
`
`

`
`Note: Dr. Fauci's work as editor and author was performed outside
`the scope of his employment as a U.S. government employee. This
`work represents his personal and professional views and not necessarily
`those of the U.S. government.
`
`HARRISON'S
`RINCIPLES OF INTERNAL MEDICINE
`Thirteenth Edition
`
`Copyright !0 1994. 1991. 1987. 1983. 1980, 1977. 1974. 1970, 1966, 1962, 1958 by McGraw(cid:173)
`Hill. Inc. All nghts reserved. Copyright 1954 , 1950 by McGraw-Hill , Inc. All rights reserved.
`Copyright renewed 1978 by Maxwell Myer Wintrobe and George W. Thorn. Printed in the
`Unhcd States of America. Except as pem1ittcd under the United Stutes Copyright Act of 1976,
`no part of this publication may be reproduced or distributed in any form or by any means. or
`stored in a data base or retrieval system. without the prior wrillcn permission of the publisher.
`
`l 2 3 4 5 6 7 8 9 0 DOW DOW
`
`987654
`
`Foreign Language Editions
`CHINESE (T welfrh Edition)- McGraw-Hill Book Company-Singapore,
`© 1994
`FRENCH (Twelfth Edition)- Fiammarion. © 1992
`GERMAN (Tenth Edition)- Schwabc and Company, Ltd., © 1986
`GREEK (Twelfth Edition)- Parissianos, © 1994 (est.)
`ITALIAN (Twelfth Edition)- McGruw-Hill Libri ltalia S.r.l. © 1992
`JAPANESE (Eleventh Edition)- Hirokawa. © 1991
`PORTUGUESE (Twelfth Edition)- Editora Guanabara Koogan, S.A ..
`© 1992
`SPANISH (Twelfth Edition)- McGruw-Hillllnteramericana de Espana,
`© 1992
`
`This book was set in Times Roman by Monotype Composition Company.
`The editors were J. Dereck Jeffers and Stuart D. Boynton. The indexer
`was Irving Tullar; the production supervisor was Roger Kasunic; the
`designer was Marsha Cohen; R. R. Donnelley & Sons Company was
`printer and binder.
`
`Library of Congress Cataloging-in-Publication Data
`
`Harrison's principles of internal medicine-13th ed./editors,
`Kurt J . lsselbacher . . . [el al.]
`p.
`em.
`Includes bibliographical references and index.
`ISBN 0·07 -032370-4 (1-vol. ed.) ; 98.00 -
`ISBN 0-07-911 169-6 (2
`vol. ed. set): 127.00 -ISBN 0-07-032371-2 (bk. 1). -
`ISBN
`0-07-032372-0 (bk. 2)
`1. Internal medicine.
`11. lsselbacher. Kurt J.
`medicine.
`[DNLM: 1. Internal Medicine. WB 115 P957 1994]
`RC46.H333 1994
`616-dc20
`DNLM/DLC
`for Library of Congress
`
`I. Harrison, Tinsley Randolph, 1900-
`Ill. Title: Principles of internal
`
`93-47393
`CIP
`
`INTELGENX 1032
`
`

`
`PART FOUR
`
`CLINICAL P.HARMACOLOGY
`
`Because the drug is not absorbed instantly after oral administration
`and is delivered into the systemic circulation more slowly, much of
`the drug is distributed by the time absorption is complete. Thus
`procainarnide, wbich is almost totally absorbed after oral administra(cid:173)
`tion , can be given as a single 750-mg loading dose with little risk of
`hypotension; in contrast, loading of the drug by the intravenous route
`is more safely accomplished by giving the dose in [ractions of about
`I 00 mg at 5-min intervals to avoid the hypotension that might ensue
`during the distribution phase if the entire loading dose were given as
`a single bolus.
`In contrast, other drugs are distributed slowly to their sites of
`action during the distribution phase. For example, levels of digoxin
`at the receptor site (and its pharmacologic effect) do not reflect plasma
`levels during the distribution phase. Digoxin is transported (or bound)
`to its cardiac receptors more slowly by a process that proceeds
`throughout distribution. Thus plasma levels fall during a distribution
`phase of several hours, while levels at the site of action and
`pharmacologic effect increase. Only at the end of the distribution
`phase, when the drug has reached equilibrium with the receptor, does
`the concentration of digoxin in plasma reflect pharmacologic effect.
`For this reason, there should be a 6- to 8-h wait after administration
`before plasma levels of digoxin are obtained for a guide to therapy.
`Equilibrium phase After distribution has proceeded to the point
`where the concentration of drug in plasma is in dynamic equilibrium
`with that in the tissues outside the vascular compartment, the levels
`in plasma and tissues fall in parallel as the drug is eliminated from
`the body. Thus the equilibrium phase is sometimes also referred to
`as the elimination plwse. Measurement of drug concentration in
`plasma provides the best reflection of drug level in tissues during this
`phase.
`Most drugs are eliminated as a first-order process. During the
`equilibrium phase, a characteristic of the first-order process is that
`the time required for the level of drug in plasma to fall to one-half
`the original value (the half-life, t 111) is the same regardless of which
`
`FlGURE 66-1 Concentr:llions of lidocaine in plasma following the adminis(cid:173)
`trntion of 50 mg imravenou~ly. The half-life of 108 min is computed as the
`time required for levels to fall from uny given value during the equilibrium
`phase (Cp,..a~) to one-half that level. Cp0 is the hypothetical concentration of
`lidocaine in plasma at time zero if equilibrium had been achieved instantly.
`
`Cone
`Lidocaine
`In Plasma
`{J.IgJmL)
`
`5.0
`
`0.1
`
`CpiiOIIl!
`2
`
`Q.Q5L-.....,20~40+-,Jf:JJ~80~,:--,1-!:00:-1J:!,20=-::-'14""0-:-160~1±-:B0""'200b-:-:220!b='240
`Time(min)
`
`66 PRINCIPLES OF DRUG THERAPY
`
`OHN A. OATES I GRANT R. WLLKINSON
`
`-J
`
`QUANTITATIVE DETERMINANTS OF DRUG ACTION
`
`Safe and effective therapy with drugs requires their delivery to target
`tissues in concentrations within the narrow range that yields efficacy
`without toxicity. Optimal precision in achieving concentrations of
`drug within this therapeutic .. window" can be achieved with regimens
`that are based on the kinetics of the drug's availability to target sites.
`This chapter deals with the principles of drug elimination and
`distribution that form the basis for loading and maintenance regimens
`for the average patient and considers instances in which elimination
`of the drug is impaired (e.g., renal failure). The basis for optimal
`utilization of plasma level data is also discussed.
`PLASMA LEVELS AFTER A SINGL.E DOSE The levels of lido(cid:173)
`caine in plasma following intravenous administration decline in two
`phases. as illustrated in Fig. 66-1; such a biphasic decline is typical
`for many drugs. Immediately following rapid injection, essentially
`all of the drug is in the plasma compartment. and the high initial plasma
`level renects its confinement to this small volume. Subsequently, the
`drug is transferred into the extravascular compartment, and the period
`of time during which this occurs is referred to as the distribution
`phase. For lidocaine the distribution phase is virtually complete within
`30 min: then a slower rate of fall ensues, referred to as the equilibrium
`phase or elimination phase. During this latter phase, the drug levels
`in plasma and those in the tissues change in parallel.
`Distribution phase Phannacologic events during the distribution
`phase depend on whether the level of drug nt the receptor site is
`similar to that in the plasma. If this is the case, the pharmacologic
`effects, whether favorable or adverse, may be inordinately great
`during this period because of the high initial levels in plasma. For
`example, following a smal l bolus dose (50 mg) of lidocaine,
`antiarrhythmic effects may be evident during the early distribution
`phase but disappear as levels fall below those which are minimally
`effective and even before equilibrium between plasma and tissue is
`reached. Thus larger single doses or multiple small doses must be
`administered to achieve an effect that is sustained into th.e equilibrium
`phase. Toxicity resulting from high levels of some drugs during the
`...distribution phase precludes administration of a single intravenous
`loading dose that wi II achieve therapeutic levels during the equilibrium
`phase. For example, the administration of a loading dose of phenytoin
`as a single intravenous bolus can cause cardiovascular collapse due
`to the high levels during the distribution phase. If a loading dose of
`phenytoin is administered intravenously, it must be given in fractions
`nt intervals sufficient to permit substantial distribution of the prior
`ll~s~ before the next is given (e.g .. 100 mg every 3 to 5 min). Por
`Stnular reasons. the loading dose of many potent drugs that rapidly
`~quilibrate with their receptors is divided into fractional doses for
`mtravenC)us administration.
`After an oral dose that delivers an equivalent amount of drug into
`the systemic circulation, plasma levels during the initial period after
`<~drninistmtion are not as high as after an intravenous bolus dose.
`
`INTELGENX 1032
`
`

`
`394
`
`point on the plasma level curve is chosen as a struting point for the
`measurement. Another characteristic of the first-order process is that
`a semilogarilhmic plot of the concentrations i.n plasma versus time
`during the equilibrium phase is llnear. From such a plot {Fig. 66-1)
`it can be seen that the half-life of lidocaine is 108 min.
`One can calculate what amount of the administered dose remains
`in the body at any multiple of the half-Jife interval following
`administration:
`
`Ellmlno1ion
`(%of max.
`Plasma level
`ofler o single
`do5e)
`
`Number
`of half·livcs
`
`I
`2
`J
`4
`5
`
`Amount of dose
`remaining in the body. %
`50
`25
`12.5
`6.25
`3.125
`
`Amount of dose
`eliminated, %
`
`50
`75
`87.5
`93.75
`96.875
`
`Ln principle , the elimination process never reaches completion. From
`a clinical standpoint, however, elimination is essenrialty complete
`when it has reached 90 percent. Therefore. for practical purposes, a
`first-order elimination proce.u reaches completion qfter 3 to 4 ha(F
`lives.
`DRUG ACCUMULATION-LOADING AND MAINTENANCE
`DOSES With repeated administration of a drug. the amount in the
`body accumulates if the elimination of the first dose is incomplete when
`[be second dose is given. and both the amount of drug in the body and
`its pharmacologic effect increase with continuing administration until
`they reach a plateau. The accumulation of digoxin administered in
`repeated maintenance doses (without a loading dose) is illustrated in
`Fig. 66-2. Since digoxin's half-life is about 1.6 days in a palient with
`normal renal function, 65 percent of digoxin remains in the body at the
`end of I day. Thus the second dose will raise the amount of digoxin in
`the body (and average plasma level) to 165 percent of that foiJowing the
`first dose. Each subsequent dose will result in greater amounts in the
`body until a steady state is achieved. At this point, drug intake per unit
`of time is the same as the rate of elimination. with the fluctuation
`between peak and trough plasma levels remaining constant. lf the rate
`of drug delivery is subsequently altered, a different and new steady smte
`will be attained. Continuing infusion of a drug at constant rate also wiJJ
`result in progressive accumulalion to a predictable steady state (Fig.
`66-3). In this case. a constant plasma level (Cp,) is achieved which is
`between the peak and trough values attained when the same rate of drug
`delivery is administered in an intem1ittent fashion. For all drugs with
`tirst-order kinetics, the rime required to achieve steady state levels can
`be predicted from the half-life because accumulation also is a first(cid:173)
`order process with a half-life identical to that for elimination. Hence
`
`F IGURE 66-2 1l1c time course of digoxin accumulation when a single daily
`maintenance close is given without a loading dose. Note that accumulation is
`more thnn 90 percelll complete by the end of 4 half-lives.
`
`Amount of Olgo:iln
`In Body
`(%of that
`aller tne tnt dose)
`
`300
`
`200
`
`100
`65
`
`2
`
`h - 1.6 days
`
`!) 6 7 8 9 10 11 12
`nme.Doys
`5
`3
`4
`2
`llme. Mulllples of Holf·llfe
`
`6
`
`7
`
`100
`
`- - -...,100
`
`Constant Infusion
`(occumulolion)
`
`75
`
`50
`
`Accumulotloo
`(% oftne
`plasma level
`at steady sfotel
`25
`
`~-L--~~3--~d~=s~~6~7°
`Time
`(TI"4JJf.~ Qfhalf·flfe)
`
`FIGURE 66·3 The time course of plasma levels of a drug following a single
`intravenous dose compnrcd with those durins a constant intravenous infusion.
`This relationship applies to all drugs that rapidly achieve equilibrium between
`plasma and ti~sucs.
`
`accumulation reaches 90 percent of steady state levels at the end of 3to
`4 half-lives. For digoxin, with u half-life of 1.6 days (with normal
`renal function). accumulation will be practically complete in 5 days,
`Cont.inuous infusion of a drug at a constant rate also will result in
`progressive accurnu tat ion to a steady state with a time course predictable
`from theeliminntion curve forthatdrug (Fig. 66·3).
`When the time required to reach steady state is longer than one
`wishes to wait, desired plasma levels may be achieved more rapidly by
`the administration of a loading dose. Loading entails the administration
`of an an1oumthat will bring the concentration in plasma (at equilibrium)
`to the level present during steady state. lf the desired plasma level ( Cp.J
`is known, the loading dose can be estimated with knowledge of the
`extent of the drug's extravascular distribution at equilibrium, the appar(cid:173)
`ent volume of distribution, or V":
`Loading dose = desired plasma level x volume of distribution
`at steady state
`= CIJ,. X v.,
`Loading may be accomplished by the administration of the loading
`amount as u single dose, or in the case of drugs for which there is
`risk of toxicity if all the drug is introduced into the plasma compartment
`rapidly, the loading amount is administered in a series of fractions or
`the tow I loading amount. Since the accumulation of procainamide to
`90 percent of steady state by infusion would require approximately
`LO h (the t 111 is 3 h), a loading regimen is almost always desirable.
`The load required to suppress an <ln'hythmia, however. varies among
`individuals from 300 to 1000 mg. and rapid intravenous administration
`of the average loading dose causes hypotension during the distribution
`phase in some patients. Therefore. the intravenous loading dose of
`procninamide is given in fractions (e.g .. 100 mg every 5 min) until
`the arrhythmia is controlled or adverse effects such as hypotension
`indicate that no further drug should be given. Dividing the loading
`dose into fractions is appropriate for most drugs that have a low
`thentpcutic index (the therapeutic index is the ratio of toxic dose to
`Ute therapeutic dose). This permits better individualization of the
`loading amount and minimizes adverse effects.
`The size of the loading dose required to achieve the plasma levels
`at steady state also can be determined from the fraclion of drug
`eliminated during the dosage interval and the maintenance dose (in
`the case of intermittent drug administration). For example. if the
`fraction of digoxin eliminated daily is 35 percent and the planned
`maintena11ce do~c is to be 0.25 mg daily, then the loading dose to
`achieve stc:tdy state levels should be I 00/35 times the maintenance
`dose, or approximately 0. 75 mg. Thus
`I 00
`Londing
`=
`dose %or drug eliminated
`per dosage interval
`
`.
`x mamrenance dose
`
`INTELGENX 1032
`
`

`
`I;HAPTER 66 PRINCIPLES 01 DRUG lliERI\P~
`
`395
`
`The frnction of drug eliminated during any dosage interval can be
`determined from a semilogarithmic g<c1ph in which the total amount
`ln the body ut time zero is set at I 00 percent and the fraction remaining
`at the end of one half-life is 50 percent.' Conversely, if Lhe loading
`dose is known, the maintenance dose can be simi larly calculated .
`To cnfculate a loading dose designed to achieve the plasma
`concentration of a known infusion rate at steady state.
`
`_ infusion rate
`. d
`Lo d
`a mg osc -
`k
`
`where k is lhe fractional eliminut ion constant that describes the rate
`of drug elimination.•
`Regat•dless of the size of the loading dose , after maintenance
`tltempy has been given }or J to 4 ltll/f-lives. 1fle amount of drug in
`tltt body is determined by the maimenaru:e close. The independence
`of the plasma levels at sready state from I he load is ill usc rated in Fig.
`66-.3. which indicates thai the elimination of lhe londing dose would
`be practicnlly complete after three to four half-lives.
`DETERMINANTS OF PLASMA LEVELS DURING THE EQUILIB·
`AlUM PHASE An important determinant of the level of drug in
`plasma during the equilibrium phase after ll single dose is the extent
`to which the drug is distributed ours ide the plasma compartment. For
`example, if the distribution of a 3-mg dose or a large macromolecule
`is confined to a plasma volume of 3 L, then rhe concentration in
`plnsma will be l mg/L. However. if a different drug is distributed so
`lhBI 90 percenl of it Jcavcs the plasma compartment, then only 0.3
`mg will remain in the 3-L plasma volume and the concentration in
`plasma will be only 0.1 mg!L. The ttpparent volunw of distribution,
`or V," eKpresses the relationship between the amount of drug in the
`body and the plasma concentration at equilibrium:
`V = amount of drug in body
`"
`plasma concentration
`TI1e nmoum oi drug in the body is expressed as mass (e.g ..
`milligrams) , and the plasma concentration is expressed as mass per
`volume (e.g., milligrams per liter). Thus v., is a hypotheticalvolumc
`into which a quantity of drug would distribute if its concentration in
`tbe entire volume were the same as that in plasma. Although it is nOt
`a rent volume, it is an important coocepr because it determines rhe
`fraction of total drug in the plasma and therefore the fraction available
`to the organs of elimination. An approximation of V., in the equilibrium
`phase cun be obtained by estimating the concenlr.ttion of drug in
`plasma at time zero (Cp0 ) by back-o.xtrapolation of the equilibrium(cid:173)
`phase plot to zero time, ns illustrated in Fig. 66- 1. Then. after
`intravenous admtnistration when the amount in the body at time zero
`is the dose. we bave
`
`V
`
`_dose
`
`., ---Cpo
`For the administration of the large macromolecule mentioned above.
`the measured Cp0 of 1 mg/L after u 3-mg dose indicates a v., that is
`i.\ rcul volume, the plasma vt1lume. This is the exception, however.
`for lhe VJ of most drugs is lurger than plasma volume: many drugs
`'ilre so extensively taken up by cells that tissue levels exceed those in
`plasma. For such drugs, rhe hypothetical Va is large. even grealer
`than lhc volume of body water. For example. Fig. 66-1 indicates that
`t~H! Cpo obtained by extrapololion after adminisrration of 50 mg
`hdocaine is 0.42 mg/L, yielding a V,1 of 11 9 L.
`, _Sillce elimination is performed largely by !he kidneys and liver.
`11 IS U~ct'ul to consider !he elimination of drugs according to the
`('/fJarlmce concept. For example. in the kidney. regardless of the
`
`extent to which removal of drug is determined by filtration, secretion,
`or reabsorption. the net result is a reduction of the concentmtion of
`drug in plasma as it passes through the organ. The extent to which
`lhe concentration is reduced is expres~d as the e.rtr(tction ratio, or
`E, which is constant as long as fir-<;1-order elimination occurs.
`E =c. - C.
`c.
`
`where C, = arterial plasma concentmtjon
`C, = venous plasma concentration
`If lhe extraction is complete, E = I . lf the 1otal plasma flow to the
`kidneys is Q (mUminl. the total volume of plasma from which drug
`is completely removed in a unit time (clearance from the body, Cl)
`is determined as
`
`Cl,..,1 = QE
`tf the renal extraction ratio of penicillin is 0.5 and renal plasma now
`is 680 mUmin , then penicillin' s renal clearance is 340 mUmin. If
`the extracrion ratio is high. as is !he case for renal extraction of
`nminohippuratc or hepatic extraction of propranolol, then clearance
`is u function of organ blood flow_ 2
`Clearance from !he body is the sum of clearance from all organs
`of elimination and is the best measure of the efficiency of the
`elimination processes. If a drug is removed by both the kidney and
`liver. then
`
`Cl = C1,.,>1 + Cl,.,"'•••
`'T'hus, if penicillin is eliminated by both renal clearance (340 mL/
`min) and hepatic cl.earancc (36 mL/rnin) in o normal individual, total
`clearance is 376 mUmin. If renal clearance is reduced to half, total
`clearance is 170 + 36 or 206 mUmjn. Ln anuria, total clearance
`equals hepatic clearance.
`Only the drug in the vascular compartment can be cleared during
`each passage through an organ. To ascertdin the effect of a given
`plasma clearance by one or more organs on the. rate of removal of
`drug from the body. the clearnnce must be related w the volume of
`"plasma equivalents' ' ttl be cleared, that is, the volume of d]stribution.
`If the volume of distribution is 10 Land clearance is I :Umin. chen
`one-tenth of the drug in the body is elim.inaled per minule. Th.is
`fraction. CIIVJ, is known as nfractiono/ elimination consmm and is
`designated as k:
`
`k = Cl
`v,,
`If the fraction k is mulriplied by the totaJ amoum of drug in lhe
`body, the actual rate of elimination al any given time can be
`determined:
`
`Rote of elimination = k X amount in body = CICp
`Tllis is rhe general equation for all firsl-order processes and
`expresses lhe fact that rate is prop01tional to the declining quantiry
`in a first-order process.
`Since half-life is a temporal expression of the exponential first(cid:173)
`order process, half-life (f 11J can be related to k as follows:
`
`0.693
`r,n = --k--- -
`
`CI
`Because k = - - --
`
`then
`
`1112 =
`
`1 ~itnt~1ivcty. the frucuon of drugloSI from w body during a dosage in1ervat can be
`""'Ct'!mll~d nongrijphitalty from lhl> equation:
`Fruction or drug loll from body = I -
`• ''
`Vptue£ f'(l[ ~ 11 <'lin be obluin~d frum ij 1uble of nmur~l exroncntlal functiom. or by u
`~(,~C\Ilntor, wher~ k = (0.693/t111 ) Is the fru~lional cllmintJiilln constnnt (<,!~scribed in
`next section) nnd 1 is the time inrcrvut after drug administrntlun,
`
`As shown in the section on drug dosage in renal failure. the linear
`relationship of k to creatinine clearance makes k a useful parameter
`
`~ Wh~n rlrug is pre~enL iu the funned otemMI> of blood. I hen cnlculullon of extraction
`aml clearance frum blood is more physiologically me11ningful thnn from th~ ptnsm3,
`
`INTELGENX 1032
`
`

`
`396
`
`Plllll I'OUII CLINil.Al I'IIARMACoJI.Of'Y
`
`upon which to estimate changes in drug elimination with reduction
`in creatinine clearance in renaJ insufficiency. Half-life is not linearly
`related to clearance.
`The imponant relationship
`
`0.693V"
`Ita = _C_I_
`
`indicntcs clearly the dependency of half-life, a measure of rate of
`elimination , on the two physiologically independent variables of
`volume of distribution and clearance, which expresses the efficiency
`of elimination. Thus half-life is shortened when phenobarbital induces
`the enzymes responsible for hepatic cleurance of a drug, and half-life
`is lengthened when a drug's renal clearance is auenuated in renal
`failure. Also, the half-life of some drugs is shortened when their
`volume of distribution is reduced. If. as in the case of cardiac failure,
`the volume of disuibution is reduced at the same time that clearance
`is reduced, there may be little change in drug half-life to reflect the
`impaired clearance, but steady state plasma levels will be increased,
`as is the case with lidocaine. ln treating patients after an overdose.
`the effects of hemodialysis on a drug's elimination arc dependent on
`its volume of distribution . When the volume of distribution is large,
`as with tricyclic antidepressants (V,1 of desipramine equals more than
`2000 L) , the removal of drug, even with a high-clearance dialyzer.
`proceeds slowly.
`The extent to which a drug is bound to plasma protein also
`determines the fraction extracted by the organ(s) of elimination.
`Altered binding changes the extraction ratio significantly. however.
`only when elimination is limited to the unbound (free) drug in plasma.
`The extent to which binding influences elimination depends on the
`relative affinity of tbe plasma binding versus the affinity of the drug
`for the extraction process. Tbe high affinity of the renal tubular anion
`transport system for many drugs leads to extraction of bound and
`unbound drug, and the efficient process by which the liver removes
`propranolol extracts most of this highly bound drug from blood.
`However, in the case of drugs with low organ extraction ratios, only
`unbound drug is available for elimination.
`STEADY STATE With a constant infusion of drug, the infusion
`rate equals elimination rate at steady state. Therefore,
`
`response should lend to cQnsideration of bioavaiJnbility as a possible
`facror. Calculation of a dosage regimen should be corrected for
`bioavailability:
`
`I d
`0
`ra ose =
`
`Cp •• x Cl x dosage interval
`F
`
`DRUG ELIMINATION THAT IS NOT FIRST-ORDER The e]jmina(cid:173)
`tion of some drugs such as phenytoin. salicylate, and theophylline
`does not follow first-order kinetics when amounts of drug in the body
`are in the therapeutic range. For these drugs, the clearance changes
`us levcJs in the body fall during elimination or after alterations in
`dose. This pattern of elimination is said to be dose-dependem.
`Accordingly, the time for the concentration to fall to one-half becomes
`less as plasma levels fall: thls halving time is not truly a balf-life.
`because the term half-life applies to first-order kinetics and is a
`constant. The elimination of phenytoin is dose-dependent, and when
`very high levels are present (in the toxic range), tbe halving time
`may be longer than 72 h , whereas as the concentration in plasma
`declines, the clearance increases and the concentration in plasma Will
`halve in 20 to 30 h. When n drug is eliminated by first-order kinetics,
`the plasma level at steady state is directly related to the amount of
`the maintenance dose, and a doubling of the dose should lead to
`doubling of the steady state plasma level. However, for drugs with
`dose-dependent kinetics, increases in the dose may be accompanied
`by disproportionate increases in plasma level. Thus. if the daily dose
`of phenytoin is increased [rom 300 to 400 mg, plasma levels rise by
`mof'e than 33 percent. The extent of increase is not predictable because
`of the interpatient variability in the extent ro which clearance deviates
`from first order. Theophylline and salicylaies also are eliminated by
`dose-dependent kinetics, and in child1-en particular caution must be
`taken with the administrotion of salicylates in high doses. Changes
`in dosage regimens for such drugs should always be accompanied
`by survei llance for adverse effects and by measurement of the
`concentration of the drug in plasma after sufficient time to establish
`n new steady state. Ethanol metabolism also is dose-dependent, with
`obvious implications. The mechanisms involved in dose-dependent
`kinetics may include the saturation of the rate-limitjng step in
`metabolism or a feedback inhibition of the rate-limiting enzyme by
`a product of the reaction.
`
`INDIVIDUALlZATION OF DRUG THERAPY
`
`Optimal drug therapy requires administration of just the right amount
`of drug for the particular patient-
`too little and efficacy is not likely,
`whereas too large a dose increases the risk of undesirable effects.
`When the desired response is a readily determined clinical effect,
`such ns altered blood pressure or consolation time. then an optimal
`dosage requirement can be achieved 10 an empirical fashion. Dosage
`alterations should, however. involve modest changes in amount (50
`percent) and no more frequently than every two to three half-lives.
`Ln most cases, however, drug therapy must be guided by the concept
`of a "therapeutic window" within which drug concentrations must
`be achieved and maintained. lf this therapeutic window is large, i.e ..
`little dose-relnted toxicity. then maximal efflcacy, should this be
`desired and achievable. may be obtained by administering a supraeffec(cid:173)
`tive dose. Such a strategy is often used for penicillins and many
`beta-adrenoceptor blocking agent~. It is also possible under these
`circumstances to usefully extend the duration of action of the drug.
`especially when it is rapidly cljminated from rhe body. Thus 75 mg
`of captopril will result in reduced blood presurc for up to 12 h. even
`though the elimination half-life of the ACE-inhibitor is about 2 h.
`The therapeutic window for most drugs, however, is much narrower.
`and in certain instances (see Table 66-4), as little as a twofold
`difference distinguishes the. dose (concentration) of drug producing
`the desired response from that eliciting an udverse effecl. ln these
`cu~es , the application of pharmacokinetic principles is critical to
`ac:hlcving the defined therapeutic objective.
`
`Cl
`(voVunit time)
`
`Cp ..
`(amtlvol)
`
`X
`
`fnfusion rate
`(ami/unit time)
`when the units for amount, volume, and time are consistent.
`Thus, if clearance (CJ) is known, the infusion rate to achieve a
`given steady state plasma level c:an be calculated. Estimation of drug
`clearance is discussed in the section on renal disease.
`When the dose is given intermittently instead of by infusion.
`the above relationship between plasma concentration and the dose
`administered ar each dosage iotervai can be expressed as
`Dose = Cp,., x Cl x dosage interval
`
`The average plasma concentration (Cp.,) implies, as seen in Fig.
`66-2. that levels can be higher and lower tJum the average during the
`dosage interval.
`When a drug is given orally, the fraction (F) of the administered
`dose that reaches the systemic circu lation is an expression of the
`drug's bionvailnbility. A reduction in bioav;1ilability may reflect a
`poorly formu lated dosage form that fails to disintegrate or dissolve
`in the gastrointestinal fluids. Regulatory standards have reduced the
`extent of this problem. Drug interactions also can impair absorption
`after oral dosing. Bioavailability also may be reduced due to drug
`metabolism in the gastrointestinal tract ond(or the liver during the
`absorption process, the.first-pass effect. This is a particular problem for
`drugs that are extensively extracted by these organs, and considerable
`interpalient variability often exists in bioavnilabilit.y. Lidocaine for
`the control of <~rrhythmias is not administered orally because of the
`lirst-puss effect. Drugs that are injected intramuscularly also may
`have low bioavailabiJity. e.g., phenytoin. An unexpected drug
`
`INTELGENX 1032
`
`

`
`During long-tenn therapy, the most important phannacokinetic
`factor is the drug's clearance, since this detcnnines the steady state
`plasma concentration. Thus, after an oral dose, and assumjng that
`clearance is constant regardless of the dose,
`F x oral dose
`C
`= dosage rate
`1 ••
`1
`Cl x dosage interval
`clenrancc
`Accordingly, steady st:tte drug levels and. therefore, the intensity of
`I'C)ponse can be adjusted by changing the dosing rate. In most cases
`this is best achteved by changmg the drug dose and maintaining the
`same dosage interval. e.g .• 250 mg every 8 h versus 200 mg every
`8 h-drug levels will change in a proportional fashion, but the relative
`Ouctuation between the maximum and minimum values will remain
`the same. On the other hand, the steady state level may be changed
`by altering the frequency of in