Pharmacokinetic and Pharmacodynamic Concepts Underpinning Total
Intravenous Anesthesia
Thomas W. Schnider, MD
V OLATILE ANESTHETICS often are combined with
intravenous (IV) opioids and other IV anesthetics. Strictly
speaking, total intravenous anesthesia (TIVA) refers to the
exclusive use of IV anesthetics. Although many anesthesiolo-
gists use TIVA to mean anesthesia with propofol; other
hypnotics such as midazolam could be used for TIVA as well.
Propofol is popular because of its nonhypnotic properties.1
Although the pharmacokinetic profile with regard to speed of
onset and offset of drug effect is good compared with other IV
hypnotics, recovery can be slow after long infusions of
propofol. In clinical practice, the concentration of propofol,
in contrast to volatile anesthetics, cannot be measured during
anesthesia. If volumetric or syringe pumps are used, setting an
infusion rate is different from targeting an end-tidal concen-
tration of volatile agent.
With target-controlled infusion (TCI) systems, the pre-
dicted concentration can be used similar to an end-tidal
concentration for better titration. However, this option is not
universally available. Anesthesiologists, who give TIVA,
empirically develop some appreciation of the input-to-effect
relationship of IV anesthetics that is adequate for many
situations. For optimal and rational dosing, an understanding
of some basic pharmacokinetic principles is very helpful,
independent of whether a TCI system or continuous infusion
is being used.
PHARMACODYNAMICS
Every anesthesiologist is familiar with the minimal alveolar
concentration (MAC) concept for volatile anesthetics,2 which
describes the relationship between the concentration of the
anesthetic and its effect. The shape of the curve that shows the
probability that a patient will move in response to skin incision
is sigmoidal. The relationship is time independent, and only at
steady state does the concentration relate unequivocally to the
effect. There is a time delay between the time course of the
inspiratory concentration, end-tidal concentration, and concen-
tration at the effect site. The concept of MAC also is useful
because dosing of different volatile anesthetics can be based on
fractions (percentage) of the MAC value. The concentration
can be measured in exhaled air, and it usually is displayed on
anesthesia machines as absolute concentration and the MAC
equivalent.
When administering IV anesthetics, many anesthesiologists
are not as used to titrating based on concentrations as they are
with volatile anesthetics. Although the plasma or effect site
concentration of propofol at which 50% of patients are awake
(Cp50) has been compared with the MACawake of, for example,
desflurane,3 anesthetics uncommonly are dosed in terms of a
fraction of some Cp50. For example, the relative potency of
Journal of Cardiothoracic and Vascular Anesthesia, Vol 29, No S1 (June), 2015: pp S7–S10
alfentanil, fentanyl, and remifentanil for respiratory depression
is 1:1:40,4,5 which is about the same for analgesic endpoints.
With TCI systems, it is possible to titrate IV anesthetics to
a target concentration; dosing is similar to choosing an end-
tidal concentration of a volatile anesthetic. Anesthesiologists
who do not have access to these TCI systems should be
particularly aware that, as described subsequently, infusion
rate correlates well with concentration only in steady-state
conditions.
PHARMACOKINETICS
Correlation Between Infusion Rate and Concentration
IV anesthetics classically are given either by bolus (eg, 100
μg fentanyl) or as a constant infusion (eg, 6 mg/kg/h propofol).
With a constant infusion, the drug concentration of propofol
increases and asymptotically approaches the steady-state con-
centration. It takes about 15 minutes for propofol to reach 80%
of the steady-state concentration. If propofol is titrated by
changing the infusion rate frequently, the correlation between
the “set-point” and the concentration at the effect site is poor
(Fig 1). To achieve a constant concentration of propofol
rapidly, specific infusion schemes are recommended.6 With
the combination of bolus (or high infusion rates) and decreas-
ing infusion rates, a relatively stable concentration can be
achieved. When titrating to an effect by changing the infusion
rate, it takes considerable time until the new effect level
correlates with the new infusion rate. However, if infusing
remifentanil, the steady-state concentration is approached
rapidly because of its pharmacokinetic properties, although
for clinical purposes this is still too slow if a sudden painful
stimulus has to be controlled. With an effect site TCI system, a
new effect site concentration can be reached as fast as is
pharmacokinetically possible. If TCI systems are unavailable,
From the Division of Anesthesiology, Intensive Care, Rescue and
Pain Medicine, Cantonal Hospital St. Gallen, St. Gallen, Switzerland.
Supported by Mylan Specialty, LP (Canonsburg, PA), which pro-
vided funding for editorial assistance provided by StemScientific
(Lyndhurst, NJ).
Address reprint requests to Thomas W. Schnider, MD, Klink für
Anästhesiologie, Intensiv-, Rettungs- und Schmerzmedizin Kantonsspi-
tal St. Gallen, Rorschacher Strasse 95, CH-9007 St. Gallen,
Switzerland. E-mail: thomas.schnider@kssg.ch
© 2015 Elsevier Inc. All rights reserved.
1053-0770/2601-0001$36.00/0
http://dx.doi.org/10.1053/j.jvca.2015.01.018
Key words: pharmacokinetics, pharmacodynamics, drug distribu-
tion, target-controlled infusion, TCI, drug interaction
S7
S8
0 ke0 are insufficient descriptors of the pharmacokinetic profile of
Fig 1. (A) The time course of the effect site concentration when
the infusion rate (dotted line) changes at 15, 25, 35, 45, and 55
minutes. Within these time intervals of constant infusion rate, the
concentration is increasing and decreasing, approaching the steady-
state concentration. (B) Correlation between the infusion rate and
the effect site concentration for the simulated scenario. Each circle
represents the predicted effect site concentration sampled every
minute for each infusion rate. The concentration is correlated with
the infusion rate only at steady state; the correlation is poor if the
infusion rate changes frequently.
an initial brief fast infusion (or bolus) hastens the onset of the
drug effect.
Onset of Drug Effect
After a bolus dose, the concentration at the effect site
increases because of the concentration gradient between the
plasma and the effect site concentration. The bigger the dose,
the faster the onset of the effect site concentration. Anesthesi-
ologists make clinical use of this fact, particularly with muscle
relaxants (eg, the recommended dose for rapid-sequence
induction of rocuronium is double the dose for normal
induction). Despite a faster onset with bigger doses, the time
to the peak effect site concentration (tpeak) is drug specific and
dose independent.7 With a higher dose and higher peak
concentration (although tpeak is unchanged), drug effects would
be prolonged, contributing to a potentially greater drug effect
and possibly more side effects (eg, hemodynamic side effects).
When comparing the “speed of onset” of 2 drugs, it would not
SCHNIDER
make sense to compare the onset times of 2 drugs given in
doses that are not equipotent because the onset times are dose
dependent.
The pharmacokinetic model–based description of the transfer
of drug from the plasma to the effect site involves the transfer
constant “ke0.” It often is argued that a drug with a higher ke0
(ie, a shorter t½ke0) will have a more rapid onset than a drug
with a lower ke0, but this is guaranteed to be true only if the
plasma pharmacokinetic profiles are identical. The overall
pharmacokinetic profile (plasma kinetics and plasma effect site
transfer), not only t½ke0, is responsible for the time course of
the effect site concentration. Shafer and Varvel8 demonstrated
with computer simulations that single parameters such as t½ or
anesthetics. They calculated the time to peak effect site
concentration for fentanyl, alfentanil, and sufentanil as 3.6
minutes, 1.4 minutes, and 5.6 minutes. The time to peak effect
site concentration of remifentanil is age dependent (1.22
minutes for a 20-year-old and 2.26 minutes for an 80-year-old).9
Recovery From Drug Effect
The half-life (t½) of a drug is still a popular parameter.
Volume of distribution and clearance also can be estimated
with basic pharmacokinetic analysis, and t½ can be calculated
from these 2 parameters. In anesthesia, in contrast to other
medical specialties, the initial distribution of the drug is also of
interest. Knowing that the terminal t½ for fentanyl is several
hours is not very clinically useful. For all drugs given intra-
venously, it has been observed that time to the first 50%
decrease of drug concentration is less than the time for the next
50%. This difference is due to the distribution of the drug into
the different body compartments. The pharmacokinetic abstrac-
tion of this distribution is a multicompartmental model.
Nevertheless, it takes a defined time for the concentration to
decrease by a defined percentage (ie, 50%, 60%, or 70%).
Instead of defining the decrease of the concentration by
multiples of the half-life (eg, 2
t½ ¼ 75%, 3
t½ ¼
87.5%), determining decrement times is more meaningful.
For many drugs, what is primarily reported is the terminal t½.
Although there is some redistribution after the cessation of
administration of a drug, the terminal t½ can have some
significance outside of anesthesia (and intensive care).
During an infusion, drug is accumulated in the peripheral
compartments, and the initial decrease of the concentration
after the infusion has been stopped becomes less with time.
When the duration of an infusion increases (the context), it
takes longer for the concentration to decrease by 50%. This
idea was formalized by Shafer and Varvel,8 and Hughes et al10
subsequently coined the term “context-sensitive half-time.”
The clinically relevant decrease in concentration is not
necessarily a 50% decrease. The difference between the concen-
tration for adequate effect and the concentration for adequate
recovery is clinically more meaningful. This difference between
the 2 concentrations depends on the steepness of the slope of the
concentration to effect relationship. Shafer and Varvel8 showed
families of recovery curves, and the concept was explored further
later as the “relevant effect site concentration decrement
time”11,12 based on the fictitious drug “Duzitol.”
PHARMACOKINETIC AND PHARMACODYNAMIC CONCEPTS OF TIVA
The clinical consequences of these pharmacokinetic insights
are important. For propofol, the time for the concentration to
decrease by 70% is much longer than the time for the 50%
decrease. If a patient is “overdosed” during anesthesia, the
required relevant decrement might be even greater. If the
anesthesiologist does not recognize this overdosing and does
not stop the propofol infusion early enough, delayed recovery
is the consequence. Most TCI systems calculate the user-
defined effect site decrement time. This pharmacokinetic
model–based information supports the anesthesiologist in
deciding when to stop the infusion. For the pharmacokineti-
cally inclined reader, the concepts of context-sensitive half-
time and relevant effect site decrement time are based on
pseudo–steady-state concentrations (ie, TCI rather than a
constant infusion rate).
PHARMACOKINETIC/PHARMACODYNAMIC INTERACTION
TIVA usually includes at least 2 anesthetics, a hypnotic
(propofol) and an opiate, which interact with each other. Pavlin
et al13 observed in a volunteer study that the concentration of
propofol was higher when administered concomitantly with
alfentanil. This finding of a pharmacokinetic drug interaction
was investigated further and confirmed by Mertens et al.14
These investigators hypothesized that the reduction in heart rate
and cardiac output is the source of this interaction. It is possible
to distinguish between pharmacokinetic and pharmacodynamic
interactions only if drug concentrations and effects are meas-
ured simultaneously. Because concentration measurements are
expensive, drug interaction studies often focus on concentra-
tions predicted by TCI systems and other measured effects
instead.
Guignard et al15 measured the electroencephalographic
parameter bispectal index (BIS) before and after laryngoscopy
at different remifentanil concentrations. The propofol concen-
tration was the same in all subjects and remained constant
during the whole study period. Before induction, the BIS value
was not affected by remifentanil, whereas during laryngoscopy,
the higher target concentrations of remifentanil prevented the
increase in the BIS value observed with lower concentrations.
Bouillon et al16 explored the reaction to a series of increasing
stimuli at different combinations of remifentanil and propofol.
They found that the interaction between propofol and remi-
fentanil is additive with regard to electroencephalogram (EEG)
measures of hypnotic drug effect, although remifentanil in the
clinical range (o8 ng/mL) had little effect on the EEG. To
ablate the response to laryngoscopy, the concentration of
propofol was reduced from 6.62 μg/mL to 2 μg/mL when
combined with 3.4 ng/mL remifentanil. With the hierarchical
interaction model, Bouillon et al16 formalized the idea that the
opiate prevents the painful stimulus from counteracting the
effect of the hypnotic at the cortical level.
The findings from these interaction studies can be applied
clinically during TIVA, particularly if an EEG measure of drug
effect is used. The prestimulus depth of hypnosis is not
predictive of whether or not a patient will respond. At a given
level of hypnosis, the concentration of remifentanil or any other
opiate correlates with the probability of patient response.
Because exposure to high concentrations of remifentanil
S9
throughout an anesthetic procedure does not to seem be
favorable with regard to postoperative pain control,17 high
doses should be avoided, although it may be tempting to use
them for pharmacokinetic reasons. It is rational to use moderate
concentrations of propofol for maintenance together with an
opiate but to increase the concentration of remifentanil toward
the end of anesthesia to enable propofol administration to be
stopped early and safely (see relevant decrement time).
PHARMACOKINETIC/PHARMACODYNAMIC VISUALIZATION
The pharmacokinetic formula that describes the time course of
drug concentrations usually intimidates clinical anesthesiologists.
In addition, many physicians still do not appreciate the full
dynamics of the input-to-concentration (and input-to-effect) rela-
tionship. They still believe in a time-independent (linear) relation-
ship between infusion rate or dose and concentration or effect.
Most TCI systems graphically display the time course of the
plasma and the effect site concentrations. Simply by studying
these displayed predicted concentrations, the astute anesthesi-
ologist will begin to develop a much clearer understanding of
the principles of drug distribution. One advantage of a TCI
system is that it takes care of the intricacies of drug
distribution, which allows the anesthesiologist to focus on
titrating the drug effect.
Where TCI systems are unavailable, display systems that
receive the dosing information from the infusion pumps are
valuable commercially available alternatives. The visualized
time course of the model predicted concentration and effect
seems to improve dosing performance.18 The anesthesiologist
can compare his or her dosing regimen with the predicted
concentrations and adjust the input accordingly.
Another option for predicting the time course of concen-
trations is offline simulation. Commercial and free software can
be downloaded from the Internet for different operating
systems. (An example of a simple program for iOS can be
found at https://itunes.apple.com/ch/app/proposim/id654052287?
mt=8.) With these programs, the time course of the concen-
tration of dosing regimens can be simulated. Simulating real
clinical dosing scenarios has a high educational value for trainees
as well as others. Nowadays the calculating power of smart-
phones is adequate for pharmacokinetic simulation. By manually
entering the dosing history of a real case into such simulation
programs at the bedside, the anesthesiologist can obtain addi-
tional pharmacokinetic information (eg, about recovery times).
CONCLUSIONS
In medicine, dosing usually is based on input; this is
feasible outside of anesthesia because at steady state, concen-
tration correlates well with input. However, in the non–steady
state (eg, in anesthesia), concentration correlates poorly with
input. Dosing of IV anesthetics should be concentration based.
Where TCI systems are unavailable, the anesthesiologist still
can gain insight into the pharmacokinetic behavior of the drug
with display systems or simulation programs. With this
information and some understanding of the basic principles
of pharmacokinetics, TIVA can move “from Impressionism to
Realism.”19
S10
REFERENCES
1. Borgeat A, Wilder-Smith OH, Suter PM: The nonhypnotic
therapeutic applications of propofol. Anesthesiology 80:642-656, 1994
2. Merkel G, Eger EI II: A comparative study of halothane and
halopropane anesthesia including method for determining equipotency.
Anesthesiology 24:346-357, 1963
3. Chortkoff BS, Eger EI II, Crankshaw DP, et al: Concentrations of
desflurane and propofol that suppress response to command in humans.
Anesth Analg 81:737-743, 1995
4. Glass PS, Iselin-Chaves IA, Goodman D, et al: Determination of the
potency of remifentanil compared with alfentanil using ventilatory depres-
sion as the measure of opioid effect. Anesthesiology 90:1556-1563, 1999
5. Mildh LH, Scheinin H, Kirvelä OA: The concentration-effect
relationship of the respiratory depressant effects of alfentanil and
fentanyl. Anesth Analg 93:939-946, 2001
6. Roberts FL, Dixon J, Lewis GT, et al: Induction and maintenance
of propofol anaesthesia: A manual infusion scheme. Anaesthesia 43
(suppl):14-17, 1988
7. Minto CF, Schnider TW, Gregg KM, et al: Using the time of
maximum effect site concentration to combine pharmacokinetics and
pharmacodynamics. Anesthesiology 99:324-333, 2003
8. Shafer SL, Varvel JR: Pharmacokinetics, pharmacodynamics, and
rational opioid selection. Anesthesiology 74:53-63, 1991
9. Minto CF, Schnider TW, Shafer SL: Pharmacokinetics and
pharmacodynamics of remifentanil. II. Model application. Anesthesi-
ology 86:24-33, 1997
10. Hughes MA, Glass PS, Jacobs JR: Context-sensitive half-time in
multicompartment pharmacokinetic models for intravenous anesthetic
drugs. Anesthesiology 76:334-341, 1992
SCHNIDER
11. Shafer SL, Stanski DR: Improving the clinical utility of
anesthetic drug pharmacokinetics. Anesthesiology 76:327-330, 1992
12. Schnider TW, Shafer SL: Evolving clinically useful predictors of
recovery from intravenous anesthetics. Anesthesiology 83:
902-905, 1995
13. Pavlin DJ, Coda B, Shen DD, et al: Effects of combining
propofol and alfentanil on ventilation, analgesia, sedation, and emesis
in human volunteers. Anesthesiology 84:23-37, 1996
14. Mertens MJ, Olofsen E, Burm AGL, et al: Mixed-effects
modeling of the influence of alfentanil on propofol pharmacokinetics.
Anesthesiology 100:795-805, 2004
15. Guignard B, Menigaux C, Dupont X, et al: The effect of
remifentanil on the bispectral index change and hemodynamic
responses after orotracheal intubation. Anesth Analg 90:
161-167, 2000
16. Bouillon TW, Bruhn J, Radulescu L, et al: Pharmacodynamic
interaction between propofol and remifentanil regarding hypnosis,
tolerance of laryngoscopy, bispectral index, and electroencephalo-
graphic approximate entropy. Anesthesiology 100:1353-1372, 2004
17. Guignard B, Bossard AE, Coste C, et al: Acute opioid tolerance:
Intraoperative remifentanil increases postoperative pain and morphine
requirement. Anesthesiology 93:409-417, 2000
18. Syroid ND, Agutter J, Drews FA, et al: Development and
evaluation of a graphical anesthesia drug display. Anesthesiology 96:
565-575, 2002
19. Kennedy RR: Seeing the future of anesthesia drug dosing:
Moving the art of anesthesia from impressionism to realism. Anesth
Analg 111:252-255, 2010