Pharmacokinetic Studies in Pregnancy Michael J. Avram , Michael J. Avram PII: DOI: Reference:
S0146-0005(20)30008-2 https://doi.org/10.1016/j.semperi.2020.151227 YSPER 151227
To appear in:
Seminars in Perinatology
Please cite this article as: Michael J. Avram , Michael J. Avram , Pharmacokinetic Studies in Pregnancy, Seminars in Perinatology (2020), doi: https://doi.org/10.1016/j.semperi.2020.151227
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2020 Published by Elsevier Inc.
Pharmacokinetic Studies in Pregnancy Michael J. Avram, Ph.D. Associate Professor, Department of Anesthesiology Northwestern University Feinberg School of Medicine Chicago, Illinois
Please address correspondence to Michael J. Avram, Ph.D. Department of Anesthesiology Northwestern University Feinberg School of Medicine 330 E. Superior Street, Morton 4-660 Chicago, Illinois 60611 e-mail: [email protected]
Office telephone: 312-908-0636
Draft of 11/15/19
Abstract The effects of the many biochemical and physiologic changes of pregnancy on the doseresponse relationship of drugs administered to pregnant women are poorly understood. The doseresponse relationship is affected by pharmacokinetics, or what the body does to a drug, and pharmacodynamics, or what a drug does to the body. Insights into the potential effects of the changes of pregnancy on one aspect of the dose-response relationship of a drug can be obtained by studying the pharmacokinetics of the drug in the various stages of pregnancy and the postpartum period. There are several available approaches to studying pharmacokinetic changes in pregnancy. Single trough screening studies can provide qualitative estimates of elimination clearance, which with the dosing rate determines the steady-state drug concentration, throughout pregnancy and into the postpartum period. Population pharmacokinetic studies such as two stage pharmacokinetic studies and studies using a nonlinear mixed effects pharmacokinetic modeling approach can characterize pharmacokinetic changes more rigorously.
There is a pressing need for studies of the pharmacokinetics of drugs used in pregnant women, who have been poorly represented in drug studies. The origin of the historical under-representation of females of all ages in drug studies has been attributed to concern about unnecessarily exposing women of childbearing potential to drugs that followed the thalidomide-induced phocomelia tragedy in the 1960s.1 While recognition of sex as an important variable in health research by major organizations such as the Institute of Medicine in 1991 stimulated the inclusion of women in clinical trials,1, 2 research to identify appropriate dosing, safety, and effectiveness of medications in pregnant women has only recently begun to be prioritized.3 Unfortunately, the effects of pregnancy on drug disposition remain poorly understood due to lingering concerns about the exposure of the unborn fetus to medications. As a result, standard doses of drugs prescribed to pregnant women could be subtherapeutic or toxic, if they are prescribed at all.4, 5 To identify safe and effective dosing of a drug for the pregnant woman, it is necessary to understand the effect of the biochemical and physiologic changes in pregnancy on the dose-response relationship of the drug. Such studies are urgently needed6 and optimal dosing information derived from them must be translated it into “the right dose for the right condition in the right patient.”7 The Obstetric-Fetal Pharmacology Research Centers (OPRC) Program was established by the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) in 2004 to enhance the understanding of dose-response relationships in pregnant women. Research supported by the OPRC has included clinical studies in pregnant women that identify the longitudinal changes of the dose-response relationship in pregnancy.4 The goal of this report is to review pharmacologic concepts that form the basis of optimal drug prescribing and their application to pharmacotherapy, beginning with a consideration of the dose-
response relationship. It then reviews several approaches to studying the pharmacokinetics of a drug in the various stages of pregnancy and the postpartum period. The Dose-Response Relationship The earliest statement of the dose-response relationship was provided by Philippus Aureolus Theophrastus Bombastus von Hohenheim (1492 – 1541), who is perhaps better known as Paracelsus. In his Third Defense, this remarkable Swiss physician and alchemist said, “… what is there that is not poison, all things are poison and nothing (is) without poison. Solely the dose determines that a thing is not a poison.”8 A contemporary statement of the dose-response relationship has been taught in graduate school as Tatum’s (presumably Arthur Lawrie Tatum, 1884 – 19559) Dictum, “The dose of a drug is enough.” Every physician aspires to prescribe their patients a dose that will produce the therapeutic effect while avoiding toxicity. What determines the dose that is enough? How and why would a dose that is enough in an individual when they are not pregnant be too much or too little when they are pregnant? The dose-response relationship is determined by pharmacokinetics (drug absorption, distribution, and elimination or what the body does to a drug) and pharmacodynamics (the relationship between drug concentration and the effect it produces or what the drug does to a body)10 (Figure 1). Biochemical and physiologic changes that occur in pregnancy could affect the pharmacokinetics or pharmacodynamics of a drug and thus affect its dose-response relationship. This review will focus on the pharmacokinetic aspects of the dose-response relationship. The interested reader is referred for detailed presentations of general and specific aspects of clinical pharmacology to the authoritative textbook on pharmacokinetics, Principles of Clinical Pharmacology, written and edited by Arthur J. Atkinson, Jr. and colleagues.11
Pharmacokinetics As an introduction to the fundamental concepts of pharmacokinetics, two seminal studies of the pharmacokinetics of thiopental will be considered, one here in the Pharmacokinetics section and one later in the Drug Accumulation section. These studies provide important insights into the distribution of a typical lipophilic molecule. For several decades the barbiturate thiopental was administered by intravenous injection to produce anesthesia. It has a very rapid onset of effect and a very brief duration of action that was initially thought to be due to its irreversible removal from the body by hepatic metabolism12 and later thought to be due to uptake by fat, which is faster than its metabolism.13 Price and colleagues demonstrated that its hypnotic effects are actually terminated by redistribution from the brain, its site of action, to pharmacologically unaffected tissues, such as muscle.14 This is nicely illustrated in Figure 2, which shows that drug concentrations in the blood, into which it has been infused (the central pool), decrease precipitously as a result of distribution to highly perfused, rapidly equilibrating tissues of the viscera, including the heart, kidney, splanchnic area, and brain. Slower distribution occurs to the less well perfused but high capacity lean tissues and to the poorly perfused adipose tissues (fat). The brisk distribution of thiopental from the blood to the brain produces the rapid onset of its effect. Once drug concentrations in the tissues of the viscera and brain reach equilibrium with those in the blood (the peak visceral concentration occurs at approximately 1 minute in Figure 2), they begin to decline as those in the blood (pool) continue to decrease. Meanwhile, the concentrations in the lean and fat tissues continue to rise at the expense of the viscera and brain. This redistribution of drug from the brain and other well perfused viscera to the lean tissues and fat ends the effect of this “ultra-short acting” drug on the brain. After drug concentrations in the lean tissues have reached equilibrium with those in the blood (the peak lean concentration occurs at approximately 32 minutes in Figure 2), concentrations in the lean tissues begin to decrease while those in the brain and viscera further decline and the concentrations in adipose tissues continue to rise.
These insights into drug distribution were based on tissue drug concentration measurements. Although tissue drug concentrations are not typically measured in contemporary clinical pharmacokinetic studies, plasma drug concentrations are. When blood samples are obtained frequently and serially after drug administration, the dynamic processes of distribution and redistribution described by Price and colleagues are reflected in the exponential phases of plasma thiopental concentration versus time relationships15 (Figure 3). A very rapid decrease in plasma drug concentration is observed during the distribution phase, which begins the moment the drug is administered intravenously. Within a matter of minutes, the transition into the redistribution phase, during which drug concentrations decrease more slowly, is evident. The drug redistribution phase lasts several hours before transitioning into a slower, terminal, elimination phase. Because the body is in distributional equilibrium during the elimination phase, the concentration of drug in plasma changes only as a result of the irreversible removal, or elimination, of drug from the body. A drug and its metabolites, if any, are typically eliminated from the body by hepatic metabolism, renal excretion, or both. This disposition (i.e., distribution and elimination) of a drug can be described by a mathematical model that characterizes the plasma drug concentration versus time relationship using a three compartment pharmacokinetic model (Figure 4). In a three-compartment pharmacokinetic model, the central or initial volume of distribution (VC) is the volume in which a drug is assumed to mix instantaneously before being distributed throughout the remaining volume. From VC, drug is distributed to the rapidly (fast) and slowly (slow) equilibrating volumes of distribution (VF and VS, respectively) by intercompartmental clearance (ClI). Intercompartmental clearances between VC and both the rapidly equilibrating and the slowly equilibrating volumes (ClF and ClS, respectively) are volume-independent estimates of drug transfer that are determined by blood flow and transcapillary permeability. These variables are measured in units of volume per unit time. The volume of distribution at steady-state (VSS) is the sum of the volumes of the central, rapidly equilibrating, and slowly equilibrating compartments.
Elimination clearance (ClE) is a concentration-independent quantitation of the ability of the body to remove drug, and has units of volume per unit time. It can be thought of as the portion of the drug distribution volume from which drug is irreversibly removed per unit time. Unlike ClE, the elimination rate of a drug at any given time is concentration-dependent. It is the product of the ClE and the plasma concentration of the drug at that time. The clearance of a drug by an organ is a function of both blood flow to the organ (Q) and the efficiency with which the clearing organ removes drug from the blood, the extraction ratio (E). It can, therefore, be affected by factors affecting blood flow to the organ or by factors affecting the efficiency of the organ in removing the drug, such as hepatic enzyme induction or inhibition. Thus, hepatic elimination clearance (ClH) can be described as:16 ClH = QH . EH Drugs cleared by the liver can be classified as high (EH > 0.7), intermediate (EH = 0.3 to 0.7), or low (EH < 0.3) EH drugs. Drugs with a high QH have flow-limited hepatic ClE because their EHs are minimally affected by physiologic perturbations. The ClH of drugs with low EHs have capacity-limited clearance because it is not affected by changes in QH. Factors, such as the biochemical and physiologic changes of pregnancy, producing changes in either QH or the EH of drugs with intermediate EHs can affect the ClH of these drugs. Renal clearance is affected by glomerular filtration, tubular secretion, and tubular reabsorption. Glomerular filtration depends on both renal blood flow and the efficiency with which the glomeruli filter the drug (the extraction ratio), which change in pregnancy. In non-pregnant adults, renal blood flow is generally less variable than hepatic blood flow because it is autoregulated and is, therefore, less likely to be rate limiting in drug clearance. Because only unbound drugs are filtered, a slight decrease in the
protein binding of an extensively (> 90%) bound drug that is eliminated by glomerular filtration will increase its glomerular filtration rate. The elimination half-life (t1/2β) is the time required for the amount of drug in the body to decrease by 50%. The t1/2β of a drug is a dependent variable that is related directly to a modelindependent estimate of its volume of distribution (Vdβ) and inversely to its ClE: t1/2β = 0.693 Vdβ/ClE Drug Accumulation and Bioavailability The longer the duration of an intravenous drug infusion, the closer the concentrations in the tissues come to reaching equilibrium with the drug in the blood. As drug accumulates in tissues during continuous administration, they lose their dilutional capacity after termination of drug administration.17 That is, plasma concentrations decrease less rapidly after stopping a drug infusion than they do after rapid intravenous administration. Therefore, for example, the time required to recover from an infusion of thiopental increases with the duration of the infusion (Figure 5). The accumulation of drug as a result of equilibration of drug in the blood with the tissues during infusion or multiple-dose regimens occurs until a steady-state is achieved. At steady-state, all compartments are in equilibrium, no net tissue distribution occurs, and the infusion rate (I) exactly matches the elimination rate. The steady-state plasma drug concentration (CSS) is predictable from these variables: CSS = I/ClE The approaches of plasma drug concentrations to CSS for a continuous constant rate intravenous infusion and an intermittent constant rate oral administration are illustrated in Figure 618. When the
drug is administered by a continuous constant rate intravenous infusion, the plasma drug concentrations increase steadily until they plateau at the CSS after an infusion lasting more than four t1/2βs (called half-times in Figure 6). Continued infusion at the same rate would maintain that CSS unless there is a change in the ClE of the drug. The accumulation of a drug during repeated oral administration at intervals equal to the t1/2β of the drug is illustrated by the plasma drug concentration versus time relationship that oscillates around the smooth increase in drug concentrations produced by a constant infusion at the same rate (Figure 6). After each oral dose, drug concentrations increase from the predose trough during the absorption phase until a maximum concentration is reached (Cmax ) for the dosing interval at time Tmax. After this point the concentrations decrease as a result of redistribution and elimination. Oral dosing achieves a CSS that is the average of the peak and trough concentrations after constant rate (e.g., the same dose administered every t1/2β) oral administration lasting more than four t1/2βs. Like continuous intravenous administration, continued oral administration at the same rate maintains that CSS unless there is a change in the ClE of the drug. As suggested above, administration of drug by any route other than intravenous adds additional complexities to the pharmacokinetics of the drug, such as the rate of drug absorption. An additional complexity is the bioavailability of the drug, F, which is the extent of absorption or the fraction of the dose reaching the systemic circulation after administration. An intravenously administered drug has a bioavailability of 1 because all of the drug administered enters the systemic circulation. The bioavailability of a drug administered by nearly any other route is usually less than 1 because some portion of the dose administered does not enter the systemic circulation. In the case of an orally administered drug, its bioavailability may be affected by incomplete absorption, intestinal metabolism, or first-pass metabolism by the gut wall and liver. First-pass refers to the absorption of drugs from the
intestinal lumen through the gut wall and into the portal circulation, which flows directly to the liver where they may be metabolized before entering the systemic circulation. A high hepatic extraction ratio for a drug reduces its bioavailability. For orally administered drugs, pharmacokinetic equations containing the terms dose or dosing rate must make an adjustment for its bioavailability. For an orally administered dose, the steady-state concentration is determined by the bioavailable dosing rate and the elimination clearance: CSS = ((Dose x F)/ τ)/ClE where τ is the dosing interval (e.g., the t1/2β dosing interval in the example in Figure 6). Bioavailability is determined by comparing the area under the curve of the plasma drug concentration versus time relationship of a dose after administration by a non-intravenous route with that of the dose after intravenous administration. Such studies are often done in the same individuals to whom drug is administered by each route on separate occasions. When a stable isotopically labeled formulation is available for oral or intravenous use, the two routes can be studied simultaneously. This approach eliminates not only interindividual differences in drug disposition by the two routes being compared but also interday differences in drug disposition in the same individuals.19 The Need for Pharmacokinetic Studies in Pregnancy The dose-response relationships of drugs taken by a woman before she becomes pregnant are likely to change when she becomes pregnant. If she has been taking a dose of the drug on a regular basis, that dose may become ineffective or result in new side effects. Changes in the dose-response relationship in pregnancy are due to changes in the pharmacokinetics or the pharmacodynamics of the drug (or both). Much of what is known about the alterations in the dose-response relationship of drugs in pregnancy has focused on changes in the steady-state concentration, which may change due to decreased bioavailability, increased elimination clearance, or both.
Drug distribution throughout the body, including to the site of drug action, occurs as a result of mixing, flow, and diffusion, which are affected by cardiovascular physiology.20, 21 Irreversible removal of drugs from the body (ClE) occurs primarily by enzymatic metabolism and renal elimination, which are affected by factors ranging from blood flow to hormonal enzymatic induction or inhibition. Physiologic changes that occur throughout pregnancy and can affect the pharmacokinetics of drugs include increases in cardiac output, hepatic blood flow, and glomerular filtration rate. Such physiologic changes as well as pregnancy-induced changes in drug metabolizing enzymes, such as increases in CYP3A4, CYP2D6, and UGT1A4 activity and decreases in CYP1A2 activity, have been reviewed elsewhere.22, 23, 24, 25 (Figure 7). The result of many of these changes is an enhanced ClE, with a resulting decrease in both CSS and therapeutic effectiveness as has been reported for antiepileptic agents,23 antiretroviral drugs,26 antidepressants,27 and other drugs.28 Another important element of pharmacokinetics is drug absorption from the gastrointestinal tract. Since many, if not most, drugs are administered orally, the decrease in gastric pH, prolonged gastric emptying, and reduced gastric motility in pregnancy could affect the rate and extent of drug absorption.29 Similarly, increased dermal hydration and increased skin blood flow affects transdermal drug absorption.29 The dynamic physiologic and biochemical changes that occur across pregnancy are associated with the timing of alterations in drug pharmacokinetics. To develop optimal dosing strategies throughout pregnancy and after birth when physiologic changes are reversed, longitudinal studies must be conducted to identify the time course of alterations in the pharmacokinetics and pharmacodynamics of drugs.
Population Pharmacokinetic Studies Population pharmacokinetic approaches are well-suited to identifying changes that occur during pregnancy.30 The goal of population pharmacokinetics is to characterize inter-individual, intraindividual, and inter-occasion variability in drug pharmacokinetics and explain the observed variability. Factors such as patient characteristics, gestational timing, co-morbid diseases, and concomitant medication can affect the pharmacokinetic parameters of a drug. The two most common population pharmacokinetic strategies are the two-stage and the nonlinear mixed effects modeling approaches. In designing a pharmacokinetic study to test hypotheses such as that ClE is increased or decreased in pregnancy, published data describing the pharmacokinetics of the drug in nonpregnant individuals can help plan blood sampling for plasma drug concentration measurements. Sample collection during the absorption, distribution, and elimination phases is critical to careful characterization of pharmacokinetics. For a detailed discussion of not only population pharmacokinetics but also population pharmacodynamics, the interested reader is referred to an excellent series of articles by Mould and Upton.31, 32, 33 Studies for the two-stage approach need to be data rich for all subjects each time they participate (e.g., each trimester and postpartum). Blood samples need to be obtained before drug administration and then frequently during the absorption, distribution, and elimination phases. Each set of the individual subject’s concentration versus time data is first modeled using nonlinear regression to obtain an estimation of pharmacokinetic parameters for each time they were studied (the first stage). Individual parameter estimates for each stage of pregnancy and postpartum are then combined to generate mean population parameters along with their variance and covariance to explain the observed variability (the second stage).
When plasma drug concentration data are collected from individuals using a sparse sampling strategy (i.e., one to six drug concentrations are available per subject), parameter estimates for individuals are not possible. However, if the samples are collected in a carefully planned manner from a relatively large number of individuals during the various pharmacokinetic phases, a non-linear mixed effects modeling approach can be used. This approach estimates the population pharmacokinetic parameters and their relationship to covariates from the concentration versus time data and characteristics of interest of all of the participants in the study. The disposition of beta-blockers in pregnancy has been studied using both the two stage modeling approach and non-linear mixed effects modeling. Hebert et al. studied the pharmacokinetics of atenolol in 17 patients in the second and third trimesters of their pregnancy and three months postpartum using a two stage pharmacokinetic approach.34 They found that its renal clearance in the second and third trimesters were 38% and 36% higher than postpartum renal clearance, respectively. and was strongly correlated with creatinine clearance. However, the total oral ClE/F of atenolol did not change during pregnancy. The oral ClE/F of the high EH drug labetalol was studied in 57 women from the twelfth week of pregnancy through 12 weeks postpartum by Fischer et al. using nonlinear mixed effect modeling.35 They reported its ClE/F at 12 and 40 weeks gestational age was 40% and 60% higher than postpartum, respectively. Lean body weight was the only covariate retained in the final model of labetalol ClE/F. Single Trough Screen Studies Elimination clearance determines the CSS achieved by a dose of a drug administered chronically, and the ClE of certain drugs is expected to change in pregnant patients. A simpler study design, the single trough screen, could be used to provide qualitative estimates of ClE throughout pregnancy and into the postpartum period. In the single trough screen, one blood sample is obtained from each
patient at or near the trough during steady-state dosing, shortly before the next dose is administered, for plasma drug concentration measurement. A frequency distribution of trough plasma drug concentrations provides a fairly accurate picture of the variability in trough concentrations, and, by inference, the ClE, in the population of interest. The sample size should be sufficiently large (20 or more patients studied multiple times during pregnancy and postpartum), assay and sampling errors small, and the dosing regimen and sampling times identical for all patients. The relationship of patient characteristics to the trough concentrations can be explored by simple statistical procedures. Plasma drug concentration to dose ratios can be used in lieu of plasma concentrations if the study participants are taking different doses. An example of such a study is one designed to determine how soon lamotrigine ClE changes in pregnancy.36 Blood samples of convenience (rather than trough samples) were collected from 22 participants during 25 pregnancies before conception and every two weeks between 5 and 13 gestational weeks. Increased lamotrigine ClE was present at 5 weeks gestation but the relationships of changes in ClE to serum estradiol concentrations and to gestational week were not strong enough to be used clinically because of the significant interpatient variability that is commonly observed in studies of this design. Physiologically Based Pharmacokinetic Modeling Physiologically based pharmacokinetic (PBPK) modeling has been a rapidly growing field for more than a decade. Such modeling integrates physiological, physicochemical, and drug-dependent data in a complex mathematic model that is used to simulate the absorption, distribution, metabolism, and excretion of a drug under a variety of conditions.37 In so doing, it is able to simulate the drug concentration versus time relationships in arterial and venous blood in vascular beds throughout the body and in a large number of tissues. Physiologically based pharmacokinetic models are becoming an increasingly important part of drug discovery and development in the pharmaceutical industry where
they have been used to do everything from informing study design to facilitating drug approval and labeling.38 A pregnancy module incorporating time-varying physiology of pregnancy using a commercially available simulator was introduced in 2013.39 Published physiologically based pharmacokinetic pregnancy models have been reviewed recently, with a discussion of their many strengths and several limitations as well as identification of needed refinements.40 Application of a physiologically based pharmacokinetic modeling and simulation to predict an optimal dose of darunavir/ritonavir in pregnancy nicely illustrates the utility of such models in the absence of formal pharmacokinetic studies.41 The predicted dose is being tested in patients and the results of the testing will be used to refine the model. Closing Thoughts Conducting studies of the pharmacokinetics and pharmacodynamics of drugs in pregnancy involves clinicians, patients, clinical pharmacologists, analytical chemists, and statisticians. Clinicians are needed guide the selection of diseases and medical treatments to be studied and assess patient characteristics, safety, and effectiveness. Carefully selected patients taking the drug of interest and willing to participate in the study multiple times throughout the course of their pregnancies (e.g., in each of their trimesters) and postpartum (e.g., at two weeks and one or two months postpartum) are essential for longitudinal studies. Clinical pharmacologists help design the studies and analyze and interpret the data. Analytical chemists measure a wide range of drug concentrations in biological matrices with great precision and accuracy. Biostatisticians provide invaluable input into the design, analysis, and interpretation of complex longitudinal studies. The studies described above have varying degrees of difficulty and provide answers with different degrees of completeness. When one wishes to know whether pregnancy affects the pharmacokinetics of a drug, qualitative estimates of ClE throughout pregnancy and into the postpartum
period can be obtained by a single trough screening study, which is relatively easy to conduct and requires a moderate number of subjects. If a qualitative change in the pharmacokinetics of a drug is discovered in a single trough screening study, a follow-up population pharmacokinetic study could be conducted to characterize the pharmacokinetic change more rigorously. A two stage pharmacokinetic study can be conducted in ten to twenty individuals but requires the collection of a relatively large number of samples over up to 24 hours from each individual on each occasion, which may necessitate admission to a Clinical Research Unit. A nonlinear mixed effects pharmacokinetic modeling approach requires fewer blood samples be collected from each individual using a sparse sampling strategy on each occasion, but often necessitates the study of up to 100 or more subjects. Physiologic pharmacokinetic modeling can be used to facilitate the efficient design of population pharmacokinetic studies. Pariente et al. conducted a systematic review of 198 studies of the pharmacokinetics of 121 medications in pregnancy.42 They found that pharmacokinetic changes during pregnancy were often consistently observed among studies, including the increased ClE along with the resultant decrease in drug exposure for a given dose of many drugs. What they often found missing was the very important consideration of whether a dose adjustment was necessary to compensate for the observed pharmacokinetics changes, which is, presumably, the point of conducting such studies. When suggesting a dose adjustment to be tested in a prospective clinical trial, changes in not only dose but also dosing strategies should be considered. For example, if an increase in dose is recommended to offset an increase in ClE, dosing the drug more frequently may be warranted to avoid substantial increases in peak concentrations that could be associated with adverse effects not seen pre-pregnancy. Many of the physiologic changes of pregnancy, including those that may lead to an increase in dose requirements, return to normal after delivery over variable lengths of time. Studying drug
disposition is imperative not only during pregnancy but also when their physiology has returned to the non-pregnant baseline to identify when the increased dose of pregnancy is no longer needed.
Figure 1. The dose-response relationship is determined by pharmacokinetics, or what the body does to a drug, and pharmacodynamics, or what a drug does to the body. Modified from Sheiner and Tozer.10
Figure 2. Price’s physiologic pharmacokinetic mode of thiopental distribution from the pool (intravascular space) to the viscera (which includes the brain, where the drug produces its hypnotic effect), lean tissue, and, ultimately, fat after rapid intravenous administration. Reprinted from Price et al.14 with permission.
Figure 3. The disposition of a thiopental viewed from the plasma after rapid intravenous administration, illustrating the rapid distributional phase, the redistribution phase, and the terminal elimination phase. Modified from Fragen and Avram.15
Figure 4. The three compartment mammillary pharmacokinetic model in which drug enters the central compartment (volume VC) and immediately begins equilibrating with the rapidly equilibrating (fast) peripheral compartment (volume VF) and the slowly equilibrating (slow) peripheral compartment (volume VS) by their corresponding intercompartmental clearances (ClF and ClS). As the distribution processes begin, the body also begins eliminating the drug by elimination clearance (ClE).
Figure 5. The rates at which thiopental leaves the brain by redistribution to indifferent tissues after rapid intravenous administration (Rapid Inj) and after 20 minute, 60 minute, and 120 minutes infusions. Reprinted from Price17 with permission.
Figure 6. The approach of plasma drug concentrations to the steady-state concentration (CSS) for a continuous constant rate intravenous infusion and intermittent constant rate oral administration. Reprinted from Buxton18 with permission.
Figure 7. Physiologic changes in pregnancy that can affect the absorption, distribution, and elimination (metabolism and excretion) of drugs. CL = elimination clearance, F = bioavailability, AUC = area under the drug concentration vs. time relationship, t1/2 = elimination half-life, ln2 = 0.693, Vd = modelindependent volume of distribution, Css = steady-state concentration, τ = dosing interval, Tmax = time to maximum plasma concentration (Cmax) after administration of a dose, CYPs = cytochrome P450 enzymes, UGTs = uridine 5'-diphospho-glucuronosyltransferases, and GFR = glomerular filtration rate. Reprinted with from Tomson et al.23 permission.
Disclosures Michael J. Avram, Ph.D., is the Assistant Editor-in-Chief of Anesthesiology and on the Editorial Board of Clinical Pharmacology & Therapeutics. He is a co-investigator on several NIH grants, including NICHD U54 HD085601 “Optimizing Medication Management for Mothers with Depression during Pregnancy (OPTI-MOM)” CoPIs: Katherine Wisner, M.D., M.S., Catherine Stika, M.D., Alfred George, M.D.
1. Uhl K, Parekh A, Kweder S. Females in clinical studies: Where are we going? Clin Pharmacol Ther 2007;81:600-602. 2 . Institute of Medicine (US) Committee on Understanding the Biology of Sex and Gender Differences. Wizemann TM, Pardue ML (eds). Exploring the Biological Contributions to Human Health: Does Sex Matter? Washington, D.C., National Academies Press, 2001. 3. Ballantyne A: Pregnant women can finally expect better. Hastings Cent Rep 2019;49:10-11. 4.
Zajicek A, Giacoia GP. Obstetric clinical pharmacology: Coming of age. Clin Pharmacol Ther 2007;81:481-482.
5. Anger GJ, Piquette-Miller M. Pharmacokinetic studies in pregnant women Clin Pharmacol Ther 2008;83:184-187. 6. McCormack SA, Best BM. Obstetric pharmacokinetic dosing studies are urgently needed. Front Pediatr 2014;2:9. 7. Kearns GL, Ritschel WA, Wilson JT, Spielberg SP. Clinical pharmacology: A discipline called to action for maternal and child health. Clin Pharmacol Ther 2007;81:463-468. 8. Deichmann WB, Henschler D, Holmstedt B, Keil G. What is there that is not poison? A study of the Third Defense by Paracelsus. Arch Toxicol 1986;58:207-213. 9. Tatum AL. Notes on the question of drug toxicities. Wis Med J 1952;51:780. 10. Sheiner LB, Tozer TN. Clinical pharmacokinetics: The use of plasma concentrations of drugs. In Melmon KL, Morrelli HF, Editors, Clinical Pharmacology: Basic Principles in Therapeutics. New York, Macmillan Publishing Company, Inc., 1978, PP 71 to 109. 11. Atkinson AJ Jr., Huang S-M, Lertora JJL, Markey SP (eds). Principles of Clinical Pharmacology, 3rd Edition, New York: Academic Press; 2012.
12. Brodie BB, Mark LC, Papper EM, et al. The fate of thiopental in man and a method for its estimation in biological material. J Pharmacol Exp Ther 1950;98:85-96. 13. Brodie BB, Bernstein E, Mark LC. The role of body fat in limiting the duration of action of thiopental. J Pharmacol Exp Ther 1952;105:421-426. 14. Price HL, Kovnat PJ, Safer JN, Conner EH, Price ML. The uptake of thiopental by body tissues and its relation to the duration of narcosis. Clin Pharmacol Ther 1960;1:16-22. 15. Fragen RJ, Avram MJ. Barbiturates. In Miller RD, Ed. Anesthesia, 5th Edition, New York, Churchill Livingstone, 2000, PP. 209-227. 16. Wilkinson GR, Shand DG. A physiological approach to hepatic drug clearance. Clin Pharmacol Ther 1975;18:377-390. 17. Price HL. A dynamic concept of the distribution of thiopental in the human body. Anesthesiology 1960;21:40-45. 18. Buxton ILO. Pharmacokinetics and Pharmacodynamics: The Dynamics of Drug Absorption, Distribution, Action, and Elimination. In Brunton LL, Lazo JS, Parker KL (eds) Goodman & Gilman’s The Pharmacological Basis of Therapeutics, Eleventh Edition Chicago, McGraw-Hill Medical Publishing Division, 2006, PP. 1-39. 19. Strong JM, Dutcher JS, Lee WK, Atkinson AJ Jr. Absolute bioavailability in man of Nacetylprocainamide determined by a novel stable isotope method. Clin Pharmacol Ther 1975;18:613-622. 20. Riggs DS. The Mathematical Approach to Physiological Problems: A Critical Primer. Baltimore, Williams & Wilkins Co., 1963.
21. Krejcie TC, Wang Z, Avram MJ. Drug-induced hemodynamic perturbations alter the disposition of markers of blood volume, extracellular fluid, and total body water. J Pharmacol Exp Ther 2001;296:922-930. 22. Odutayo A, Hladunewich M. Obstetric nephrology: renal hemodynamic and metabolic physiology in normal pregnancy. Clin J Am Soc Nephrol. 2012;7:2073-2080. 23. Tomson T, Landmark CJ, Battino D. Antiepileptic drug treatment in pregnancy: Changes in drug disposition and their clinical implications. Epilepsia 2013;54:405-414. 24. Feghali M, Venkataramanan R, Caritis S. Pharmacokinetics of drugs in pregnancy. Semin Perinatol 2015;39:512-519. 25. Pinheiro E, Stika C. Drugs in Pregnancy: Pharmacologic and Physiologic Changes that Affect Clinical Care Semin Perinatol 2020 (pending). 26. van der Galiën R, Ter Heine R, Greupink R, et al. Pharmacokinetics of HIV-integrase inhibitors during pregnancy: Mechanisms, clinical implications and knowledge gaps. Clin Pharmacokinet 2019;58:309323. 27. Deligiannidis KM, Byatt N, Freeman MP. Pharmacotherapy for mood disorders in pregnancy: A review of pharmacokinetic changes and clinical recommendations for therapeutic drug monitoring. J Clin Psychopharmacol 2014;34:244-255. 28. Zhao Y, Hebert MF, Venkataramanan R. Basic obstetric pharmacology. Semin Perinatol 2014;38:475-486. 29. Soldin OP, Mattison DR. Sex differences in pharmacokinetics and pharmacodynamics. Clin Pharmacokinet 2009;48:143-157. 30. Sun H, Fadiran EO, Jones CD, et al. Population pharmacokinetics. A regulatory perspective. Clin Pharmacokinet 1999;37:41-58.
31. Mould DR, Upton RN. Basic concepts in population modeling, simulation, and model-based drug development. CPT Pharmacometrics Syst Pharmacol 2012;1:e6. 32. Mould DR, Upton RN. Basic concepts in population modeling, simulation, and model-based drug development-Part 2: Introduction to pharmacokinetic modeling methods. CPT Pharmacometrics Syst Pharmacol 2013;2:e38. 33. Upton RN, Mould DR. Basic concepts in population modeling, simulation, and model-based drug development: Part 3-Introduction to pharmacodynamic modeling methods. CPT Pharmacometrics Syst Pharmacol 2014;3:e88. 34. Hebert MF, Carr DB, Anderson GD, et al. Pharmacokinetics and pharmacodynamics of atenolol during pregnancy and postpartum. J Clin Pharmacol 2005:45:25-33. 35. Fischer JH, Sarto GE, Hardman J, et al. Influence of gestational age and body weight on the pharmacokinetics of labetalol in pregnancy. Clin Pharmacokinet 2014;53:373-383. 36. Karanam A, Pennell PB, French JA, et al. Lamotrigine clearance increases by 5 weeks gestational age: Relationship to estradiol concentrations and gestational age. Ann Neurol 2018;84:556-563. 37. Sager JE, Yu J, Ragueneau-Majlessi I, Isoherranen N. Physiologically based pharmacokinetic (PBPK) modeling and simulation approaches: A systematic review of published models, applications, and model verification. Drug Metab Dispos 2015;43:1823-1837. 38. Jones HM, Chen Y, Gibson C, et al. Physiologically based pharmacokinetic modeling in drug discovery and development: a pharmaceutical industry perspective. Clin Pharmacol Ther 2015;97:247-262. 39. Abduljalil K, Furness P, Johnson TN, Rostami-Hodjegan A, Soltani H. Anatomical, physiological and metabolic changes with gestational age during normal pregnancy: A database for parameters required in physiologically based pharmacokinetic modelling. Clin Pharmacokinet 2012;51 365–396.
40. Dallmann A, Pfister M, van den Anker J, Eissing T. Physiologically based pharmacokinetic modeling in pregnancy: A systematic review of published models. Clin Pharmacol Ther 2018;104:1110-1124. 41. Ke AB, Greupink R, Abduljalil K. Drug dosing in pregnant women: Challenges and opportunities in using physiologically based pharmacokinetic modeling and simulations. CPT Pharmacometrics Syst Pharmacol 2018;7:103-110. 42. Pariente G, Leibson T, Carls A, et al. Pregnancy-associated changes in pharmacokinetics: A systematic review. PLoS Med 2016;13:e1002160.