British Journal of Anaesthesia 1998; 81: 495–501
CLINICAL INVESTIGATIONS
Sevoflurane pharmacokinetics: effect of cardiac output†
J. F. A. HENDRICKX, A. A. J. VAN ZUNDERT AND A. M. DE WOLF
Summary
Sevoflurane uptake (Vsevo) can be predicted by
the square root of time model or the fourcompartment model. However, Vsevo and the
effect of cardiac output on anaesthetic uptake have not been quantified clinically. After
obtaining IRB approval and informed consent,
34 adult patients received closed-circuit anaesthesia with sevoflurane for 1 h. The end-expired
sevoflurane concentration was maintained at
2.6% by infusion of liquid sevoflurane into the
breathing system. In a subgroup of 12 patients,
cardiac output was measured every 5 min by
thermodilution (CO group). The effect of patient
characteristics (age, height, weight, body surface area) and cardiac output on Vsevo were determined, and Vsevo was compared with the
theoretical models. In the CO group, measured
cardiac output was used in the formulae of
these
models.
A
two-exponential
curve
described average Vsevo well: Vsevo (ml liquid)
01.62(1–e2.3t)18.1(1–e0.0089t), r2 0.999.
There was no correlation between Vsevo and
patient characteristics, except that Vsevo was
greater in patients with a greater cardiac output
(r20.36) and cardiac index (r20.35). The rate of
sevoflurane uptake decreased less than predicted
by the square root of time and four-compartment
models, even when measured cardiac output was
used in the formulae. These findings confirm that
the square root of time and four-compartment
models do not accurately predict anaesthetic
uptake. In addition, uptake of sevoflurane cannot
be predicted by patient characteristics but was
higher in patients with a higher cardiac output. (Br.
J. Anaesth. 1998; 81: 495–501).
Keywords: anaesthetics volatile, sevoflurane; pharmacokinetics, sevoflurane; equipment, breathing systems
Theoretically, whole-body uptake of potent inhaled
anaesthetics can be predicted from organ volume,
organ blood flow and solubility of the anaesthetic in
the organ.1 After measurement of the solubility of
these agents in various tissues and organs,2 computer
simulations were used to study anaesthetic uptake,
leading to the development of uptake models such as
the square root of time (SqRT) model1 and the fourcompartment (4C) model.3 However, animal data
indicate that the theoretical tissue time constants
may not accurately describe organ uptake.4 In addition, clinical data show that desflurane and isoflurane
uptake are different from predictions based on these
models.5–8 It has been speculated that the discrepancies between the clinical observations and SqRT and
4C models are the result of factors such as changes in
cardiac output (CO) with surgical stimulation or
cutaneous and visceral losses.9 However, the effect of
CO on anaesthetic uptake has not been evaluated in
humans. In addition, sevoflurane uptake (Vsevo) has
not been quantified in humans.
Therefore, we have determined Vsevo using closedcircuit anaesthesia and investigated the relationship
between Vsevo and CO, cardiac index (CI) and other
patient characteristics. We also compared Vsevo with
uptake predicted by the SqRT and 4C models using
both calculated and measured CO in the formulae of
these models.
Patients and methods
After obtaining approval from the hospital’s
Biomedical Institutional Review Board and informed
consent, we studied 34 adult ASA I–III patients
undergoing a variety of peripheral and abdominal
procedures under general anaesthesia with sevoflurane in oxygen. Patient age, sex, height and weight
were obtained, and body surface area (BSA) was calculated.
Premedication was adjusted to patient needs. After
preoxygenation, anaesthesia was induced with
propofol 2.5–3 mg kg1 or etomidate 0.3 mg kg1.
Tracheal intubation was facilitated by succinylcholine 1 mg kg1 or vecuronium 0.08 mg kg1.
Vecuronium was used to maintain neuromuscular
block in all patients. After further denitrogenation
(end-expired nitrogen concentration 3%), the circle
system was closed and liquid sevoflurane was
injected into the inspiratory limb of the circle system
using a syringe pump with a volumetric accuracy of
2%. An end-expired sevoflurane concentration of
2.6% (1.3 minimal alveolar concentration (MAC))
was achieved as rapidly as possible after the start of
infusion of sevoflurane, and this end-expired concentration was maintained throughout the procedure by
adjusting the infusion rate. Rapid vaporization of
sevoflurane was ensured by heating the metal injection port using warm air, and visual inspection of the
J. F. A. HENDRICKX, MD, A. A. J. VAN ZUNDERT, MD, PHD,
Department of Anaesthesiology, Intensive Care and Pain Therapy,
Catharina Hospital, Eindhoven, The Netherlands. A. M. DE
WOLF*, MD, Department of Anesthesiology, Northwestern
University Medical School, 303 E. Superior St., Suite 360,
Chicago, IL 60611–3053, USA. Accepted for publication: May
26, 1998.
†Presented in part at the ASA annual meeting, San Diego, CA,
October, 1997.
*Correspondence to A.M. de W.
496
injection site revealed that after 5 min all liquid
sevoflurane had vaporized. Oxygen flow was titrated
to maintain the volume of the bellows constant.
Ventilation was controlled and normocapnia was
maintained throughout the procedure. Capnograms
were normal in all patients, ensuring reliable endexpired sevoflurane concentrations. Hypotension,
defined as a decrease in arterial pressure of more
than 25% from baseline, was treated with i.v. fluid
administration and ephedrine 5-mg boluses i.v. After
1 h, the circuit was opened and the study terminated.
Anaesthetic gas concentrations were monitored
with a multi-gas analyser (Datex-Engstrom AS/3
Anesthesia Delivery Unit; Datex, Helsinki, Finland).
The sampled gases (150–200 ml min1) were redirected into the system. ADU anaesthesia machines
(Datex-Engstrom AS/3 Anesthesia Delivery Unit;
Datex, Helsinki, Finland) were used with soda lime
as the carbon dioxide absorber. Leak from the anaesthesia machine and breathing system during controlled mechanical ventilation with a peak inspiratory
pressure of 30 cm H2O was measured each morning
using a self-test. The actual leak for each patient during the study was calculated based on the self-test
and measured peak inspiratory pressure, and was
9–27 ml min1. All individual sevoflurane uptake data
were corrected for this leak. The amount of liquid
anaesthetic needed to load the anaesthesia system
and an estimated functional residual capacity (FRC)
of 2 litre with 2.6% sevoflurane was 1.3 ml of liquid
sevoflurane. This dose is included in all data, curve
fitting procedures and model predictions.
In a subset of 12 patients, CO and CI were
measured by thermodilution after placement of a
pulmonary artery catheter (CO group). CO and CI
were measured just before the start of infusion of
sevoflurane and every 5 min thereafter.
The cumulative sevoflurane dose (total amount of
liquid anaesthetic injected over time) and endexpired anaesthetic concentrations were recorded
every 1 min. Individual uptake curves and the average
uptake curve were fitted to a series of mathematical
models including, but not limited to, one- and twoexponential models (Table Curve 2D Automated
Curve Fitting Software and Sigmaplot, Jandel, San
Rafael, CA, USA). While a large number of equations
accurately describe the uptake curves (r20.999), we
chose one- or two-exponential functions, guided by
goodness of fit (visual inspection and r2 values, as calculated by the program), by simplicity of the function
and by the fact that most biological processes behave
exponentially. Exponential curve fitting yielded
the following equations: Vsevo = a2 + b2 × (1 − e-c2 × t) (oneexponential fit); and Vsevo = a3 + b3 ×(1−e-c3×t)+d3 ×(1−e−e3×t)
(two-exponential fit). With the available data, the five
parameters of the two-exponential fit could not be
determined in all patients without the parameters
becoming interdependent from one another. Because
visual inspection revealed uptake at 3–60 min to be
almost constant, and because the second time constant (112 min) of the two-exponential fit was so long
relative to the duration of measurement (60 min), the
individual uptake curves and the average uptake
curve at 3–60 min were also fitted to a linear model:
Vsevoa1b1t. The parameters of the individually
fitted curves (b1, b2, c2, b3, c3, d3 and e3) were correlated
with patient characteristics and in the CO group with
British Journal of Anaesthesia
CO and CI using linear regression analysis. Vsevo was
compared with uptake predicted by the SqRT and
4C models, using calculated CO (Brody’s formula:
CO0.2weight0.75 litre min1) in the total patient
population and measured CO in the CO group. The
formulae used to calculate uptake according to the
SqRT and 4C models are presented in appendices A
and B. Values are presented as mean (SD).
Results
Patient characteristics are shown in table 1. In two
patients in the CO group, administration of sevoflurane was discontinued after 30 min because of
hypotension, which was the result of excessive
peripheral vasodilatation. Their uptake data for the
first 30 min are included in the final data analysis. No
patient developed clinical signs of renal dysfunction
after operation.
Within 3–5 min, a constant end-expired sevoflurane concentration of 2.6% was obtained in all
patients (fig. 1). There was significant inter-individual variation in Vsevo, with a coefficient of variation of
14–17 % over time. Sevoflurane cumulative doses
after 5, 15, 30 and 60 min are presented in table 2.
Individual linear, one-exponential and two-exponential curve fit parameters are presented in tables 3–5.
A two-exponential curve was fitted in only 10
patients without the parameters becoming interdependent. The average uptake curve (fig. 2) was also
fitted using linear, one-exponential and two-exponential models, yielding the following equations:
Vsevo (ml liquid)2.040.122t (r20.995) (the linear
fit for 3–60 min); Vsevo (ml liquid)1.3414.29
(1–e0.013t) (r20.993); and Vsevo (ml liquid)01.62
(1– e2.3t)18.1(1– e0.0089t ) (r20.999). Because the
amount of liquid anaesthetic needed to load the
anaesthesia system and an estimated FRC of 2 litre
with 2.6% sevoflurane was 1.3 ml, the a2 intercept of
the one-exponential fit and the sum of the a3 intercept and b3 of the two-exponential fit mainly
reflected the anaesthesia system and FRC wash-in. b1
(the slope of the linear fit) did not correlate with age
(r20.01), height (r2 0.01), weight (r20.08) or BSA
(r20.05) (table 6). The correlations between the
parameters b2, c2, b3, c3, d3 and e3 of the one- and twoexponential fitted curves and patient characteristics,
CO and CI were not better than those between b1
and the same characteristics. The SqRT and 4C
models (based on calculated CO) overestimated
average Vsevo during the initial part of the procedure
(first 36 and 15 min, respectively), and underestimated average Vsevo thereafter (fig. 2). Similar observations were made in individual patients.
In the CO group, measured CO was remarkably
stable (less than 5% variability over time) (fig. 3).
Surgical incision did not affect CO. Average Vsevo in the
CO group was similar to that in all 34 patients (fig. 4).
b1 in the CO group was 0.126 (0.021) and did not
correlate with height (r2 0.01), weight (r20.01) or
BSA (r2 0.01) (table 6). A correlation was found
between b1 and age: b10.04220.0025(age)
(r20.35). A higher CO and CI were associated with
a higher b1 (r20.36 and r20.35, respectively) (table
6, fig. 4). The correlations between the parameters of
the one- and two-exponential fitted curves and
patient characteristics, CO and CI were not better
Sevoflurane pharmacokinetics
497
Table 1 Patient characteristics with calculated cardiac output (COc) and cardiac index (CIc), and measured cardiac output (COm) and
cardiac index (CIm). BSABody surface area
Patient
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
Mean
(SD)
Sex
M
F
M
M
F
F
F
F
M
F
F
M
F
M
F
F
M
M
M
F
M
M
M
F
F
M
F
F
F
F
F
F
M
F
Age
(yr)
Height
(cm)
Weight
(kg)
BSA
(m2)
COc
(litre min1)
CIc
(litre min1 m2)
17
65
70
62
52
56
24
28
67
75
63
74
38
64
74
40
50
65
65
63
83
34
66
67
47
68
45
26
53
59
39
24
65
60
54.4
(17.2)
185
158
169
169
165
163
153
168
182
155
166
172
160
170
158
163
186
176
178
158
170
171
174
160
164
174
156
160
156
157
168
169
187
165
167.2
(9.2)
64
62
60
72
86
69
67
73
79
68
60
78
52
69
56
57
70
74
65
71
63
58
74
61
60
86
67
71
75
50
72
65
65
65
67.1
(8.5)
1.85
1.63
1.69
1.82
1.93
1.74
1.65
1.83
2.00
1.67
1.67
1.91
1.53
1.80
1.55
1.60
1.93
1.90
1.81
1.73
1.73
1.68
1.88
1.63
1.65
2.01
1.67
1.74
1.75
1.48
1.82
1.75
1.88
1.71
1.75
(0.13)
4.50
4.42
4.31
4.94
5.65
4.79
4.68
4.99
5.30
4.74
4.31
5.25
3.87
4.79
4.07
4.12
4.84
5.05
4.58
4.89
4.47
4.20
5.05
4.37
4.31
5.65
4.68
4.89
5.10
3.76
4.94
4.58
4.58
4.58
4.68
(0.44)
2.44
2.71
2.55
2.71
2.92
2.74
2.85
2.73
2.65
2.83
2.59
2.75
2.54
2.66
2.62
2.57
2.51
2.66
2.53
2.83
2.58
2.51
2.68
2.67
2.61
2.81
2.81
2.81
2.91
2.54
2.72
2.62
2.44
2.68
2.67
(0.13)
than those between b1 and the same characteristics.
When measured CO data were used in the formulae
of the theoretical models, the SqRT and 4C models
still overestimated average Vsevo during the initial part
of the procedure (first 32 and 18 min, respectively),
and underestimated uptake thereafter (fig. 5); the
same was true for individual patients.
Discussion
A two-exponential curve fitted the average Vsevo very
well for procedures up to 1 h. The relatively long time
constant (112 min) of the second exponential function of the two-exponential curve fit indicated that
after the 3–5-min wash-in period, the rate of sevoflurane uptake over time changed little during the first
hour. In addition, the rate of sevoflurane uptake, or
Vsevo per unit time (ml liquid min1), decreased less
than predicted by the SqRT and 4C models, even
when measured CO was used in the formulae of
these models. The rate of uptake of isoflurane and
desflurane also decreased significantly less than predicted by the SqRT model.5–8 It has been speculated
that the discrepancies between the clinical observations
and SqRT and 4C models are the result of factors such
as changes in CO with surgical stimulation, or cutaneous and visceral losses.9 However, we found that
CO did not fluctuate significantly in this clinical
study using sevoflurane anaesthesia. After substituting measured CO values in the formulae, both mod-
COm
(litre min1)
CIm
(litre min1 m2)
3.64
2.23
3.83
2.10
4.83
4.80
2.41
2.87
6.00
3.14
4.71
3.03
4.18
3.33
2.87
2.20
1.84
1.66
4.61
2.44
3.59
1.79
5.39
4.31
(0.91)
3.15
2.41
(0.53)
els overestimated uptake during the initial part of the
procedure and underestimated it thereafter. Although
a higher CO and CI were associated with higher
sevoflurane uptake, we found that the correlation
between CI and b1 (r20.35) was no better than
the correlation between CO and b1 (r20.36).
Preliminary data for desflurane uptake appear to
confirm these findings.8 Visceral and percutaneous
losses are minimal and probably cannot account for
the differences between our data and the theoretical
models.10 11 Although soda lime absorbs and degrades
Figure 1 End-expired sevoflurane concentration over time in all
34 patients (mean, SD).
498
British Journal of Anaesthesia
Table 2 Cumulative dose of sevoflurane (CDsevo) after 5, 15, 30,
45 and 60 min. †Based on 30-min observations
Table 4 One-exponential fit parameters (Vsevo = a2 + b2 ×(1− e−c2×t ).
†Based on 30-min observations
Patient
No.
CDsevo 5 CDsevo 15
(ml)
(ml)
CDsevo 30
(ml)
CDsevo 45 CDsevo 60
(ml)
(ml)
Patient
No.
a2
(ml)
b2
(ml)
c2
(min1)
r2
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19†
20
21
22
23†
24
25
26
27
28
29
30
31
32
33
34
2.6
2.2
2.7
2.9
2.6
2.7
3.5
3.1
2.2
2.3
2.3
1.8
2.6
2.1
2.0
2.4
2.6
2.3
2.2
2.1
2.1
2.7
2.2
2.1
1.9
2.1
2.4
2.8
2.4
2.0
2.5
2.4
2.5
2.0
3.6
3.5
4.2
5.3
4.2
4.1
5.2
4.4
3.8
3.6
3.7
3.7
3.4
3.4
3.7
4.0
5.0
4.1
3.2
3.1
3.6
3.5
3.7
3.6
3.3
3.5
5.0
4.8
4.7
3.2
4.8
3.6
3.7
3.5
4.9
5.3
5.9
7.2
6.3
5.7
7.4
6.9
5.9
5.8
5.5
5.8
4.9
5.3
6.2
5.8
7.6
6.2
5.0
4.7
6.0
4.7
5.3
5.6
4.9
5.0
7.5
6.4
6.4
5.1
6.8
4.8
5.3
5.4
6.2
6.6
7.5
9.3
7.9
7.0
9.1
8.9
7.7
7.9
6.6
8.4
6.3
6.6
8.3
7.8
10.3
7.8
7.3
8.3
8.9
11.3
9.5
8.7
11.2
11.3
9.4
9.9
7.7
10.2
7.6
7.8
10.2
9.6
12.5
9.2
6.0
8.1
6.2
7.4
10.1
7.3
6.9
6.4
6.3
9.8
7.8
8.8
6.5
8.8
6.1
6.6
6.8
8.2
7.9
7.9
12.1
9.4
11.2
7.7
10.4
7.3
7.8
7.9
Mean
(SD)
2.4
(0.4)
3.9
(0.6)
5.8
(0.8)
7.5
(1.2)
9.1
(1.5)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19†
20
21
22
23†
24
25
26
27
28
29
30
31
32
33
34
Mean
(SD)
1.5
1.3
1.5
1.6
1.5
1.5
1.9
1.8
1.3
1.3
1.2
0.7
1.4
1.2
1.4
1.3
1.3
1.1
1.0
1.3
1.1
1.7
1.0
1.1
1.2
1.3
1.2
1.6
1.6
1.1
1.3
1.6
1.5
1.0
1.33
(0.26)
8.5
11.9
11.8
15.1
12.5
10.4
13.7
33.2
17.0
51.9
7.6
33.5
10.5
11.0
10.5
20.2
24.0
11.9
6.8
14.1
24.1
16.7
6.2
10.0
12.6
11.8
18.7
9.5
31.0
11.8
13.5
10.2
8.4
10.4
15.62
(9.56)
0.018
0.014
0.016
0.016
0.016
0.018
0.017
0.006
0.011
0.003
0.028
0.006
0.014
0.015
0.014
0.008
0.011
0.019
0.028
0.009
0.008
0.007
0.040
0.020
0.012
0.013
0.014
0.025
0.006
0.014
0.018
0.013
0.021
0.018
0.015
(0.007)
0.979
0.990
0.989
0.988
0.990
0.984
0.984
0.989
0.993
0.994
0.989
0.997
0.983
0.993
0.983
0.993
0.996
0.995
0.961
0.989
0.996
0.977
0.968
0.993
0.988
0.988
0.996
0.984
0.989
0.992
0.993
0.979
0.982
0.994
0.987
(0.008)
Table 3 Linear fit parameters (Vsevoa1+b1t) based on 3–60-min
observations. †Based on 3–30-min observations
Patient
No.
a1
(ml)
b1
(ml min1)
r2
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19†
20
21
22
23†
24
25
26
27
28
29
30
31
32
33
34
Mean
(SD)
2.301
1.956
2.345
2.663
2.393
2.376
3.079
2.247
1.854
1.545
2.385
1.107
2.092
1.828
1.473
1.800
2.101
2.091
1.503
1.690
1.519
2.051
1.662
1.987
1.782
1.860
2.168
2.884
2.089
1.673
2.371
2.180
2.414
1.853
2.039
(0.416)
0.086
0.106
0.114
0.149
0.123
0.105
0.136
0.151
0.128
0.141
0.094
0.158
0.093
0.105
0.152
0.130
0.180
0.126
0.117
0.096
0.145
0.090
0.132
0.110
0.102
0.100
0.169
0.111
0.150
0.107
0.142
0.087
0.093
0.108
0.122
(0.025)
0.994
0.990
0.992
0.985
0.990
0.990
0.992
0.998
0.995
0.999
0.967
0.997
0.996
0.991
0.995
0.997
0.995
0.986
0.994
0.998
0.997
0.997
0.993
0.980
0.994
0.994
0.991
0.979
0.996
0.990
0.986
0.997
0.985
0.986
0.991
(0.007)
Table 5 Two-exponential fit parameters (Vsevo = a3 + b3 ×(1− e−e3×t) +d3
×(1 − e − e3×t )
Patient
No.
a3
(ml)
b3
(ml)
c3
min1)
d3
(ml)
e3
(min1)
r2
1
3
5
6
7
13
14
31
32
33
Mean
(SD)
0.1
0.0
0.0
0.5
0.0
0.0
0.0
0.0
0.0
0.0
0.07
(0.16)
2.1
1.9
1.8
2.0
2.6
1.9
1.4
1.6
2.0
1.9
1.91
(0.33)
0.647
2.183
2.257
0.796
0.893
0.762
2.657
2.875
1.918
1.294
1.628
(0.847)
16.7
14.6
14.9
15.3
20.3
23.5
12.8
14.9
18.0
10.3
16.12
(3.76)
0.00618
0.01099
0.01184
0.00909
0.00877
0.00451
0.01177
0.01475
0.00577
0.01379
0.00975
(0.00347)
0.999
0.999
0.998
0.996
0.997
0.999
0.999
0.998
0.999
0.996
0.9980
(0.0011)
sevoflurane, we calculated that approximately
0.2 ml h1 of liquid sevoflurane would be absorbed
and degraded under our study conditions, which is
approximately 2% of the total uptake of sevoflurane
after 1 h in this study.12 Sevoflurane undergoes
biodegradation, but this amount is only a fraction
(approximately 3%) of total body uptake.13
There are other indications that the SqRT and 4C
models may not be accurate, as recent observations
suggest that some of the underlying assumptions
used in the uptake models may be incorrect or
incomplete. The SqRT and 4C models assume that
arterial partial pressures equal alveolar partial pres-
Sevoflurane pharmacokinetics
499
Figure 2 Cumulative sevoflurane dose (mean, SD ml liquid) in all
34 patients and cumulative dose, as predicted by the square root of
time (SqRT) and four-compartment (4C) models, using
calculated cardiac output in the formulae of these models.
sures.1 3 This may be the basis of significant error,
because a significant arterial to end-expired gradient
for isoflurane has been described in humans, caused
by deadspace ventilation, shunting and diffusion limitations.14 The arterial to end-expired partial pressure
ratio was 0.66 after end-expired isoflurane partial
pressure had been stable for at least 15 min.14 In
addition, the theoretical models calculate whole body
anaesthetic uptake based on the uptake of each
individual organ, which can be determined from the
volume of the organ, tissue solubility of the agent and
organ perfusion.1 3 Assuming organ uptake is an
exponential process when arterial concentration is
constant, a time constant can be calculated for each
organ or tissue group. However, brain time constants
of desflurane, isoflurane and halothane in rabbits
obtained in vivo differ significantly from the theoretically calculated values.4 Tissue–gas partition coefficients determined on tissue homogenates from
cadavers15 may not represent solubility of an anaesthetic in an organ in vivo. Reported values for tissue–gas partition coefficients for isoflurane differ by
up to 150% between studies, and the coefficient of
variation for isoflurane tissue–gas partition coefficient
is up to 150% (for muscle–gas partition coefficient)
between patients.15 In addition, regional blood flows
used in the theoretical models may not be accurate.
Finally, the theoretical models may not account for
uptake by anatomically or functionally less well
defined “tissue groups”. Also, inter-tissue diffusion
may be a factor affecting uptake and distribution of
inhaled anaesthetics, but it is unclear how this affects
total body uptake.16–18 These arguments indicate that
Figure 3 Measured cardiac output (CO) in a subgroup of 12
patients (CO group) (mean, SD).
Figure 4 Correlation between (A) cardiac output (CO) and
(B) cardiac index (CI) and b1 in a subgroup of 12 patients
(CO group). b1 represents the slope of the linear fit of sevoflurane
uptake (Vsevo) between 3 and 60 min.
Table 6 Correlations (r2) between b1 and patient characteristics,
CO, and CI. b1 represents the slope of the linearly fitted curve of
sevoflurane uptake between 3 and 60 min. BSABody surface
area, COcardiac output, CIcardiac index
Age
Height
Weight
BSA
CO
CI
All patient
CO group
0.01
0.01
0.08
0.05
—
—
0.35
0.01
0.01
0.01
0.36
0.35
Figure 5 Cumulative sevoflurane dose (mean, SD ml liquid) in a
subgroup of 12 patients (CO group), and cumulative dose, as
predicted by the square root of time (SqRT) and four-compartment
(4C) models using measured cardiac output in the formulae of
these models.
500
many steps along the partial pressure cascade of
potent inhaled anaesthetics are poorly studied. Even
though more sophisticated uptake models exist, they
lack sufficient experimental validation.19–22
Our observations have several implications. First,
quantification of whole body uptake is important
because it forms the basis for the general anaesthetic
equation which determines vaporizer settings over
time with different fresh gas flows to maintain a constant end-expired concentration.1 The relatively long
time constant (112 min) of the second exponential
function of the two-exponential curve fit of the average Vsevo implies that after the desired end-expired
concentration has been obtained and maintained for
approximately 5 min, the changes in vaporizer setting
are minimal for the remainder of the anaesthetic,
even if low-flow or closed-circuit anaesthesia is used.
The increasing availability of anaesthetic gas monitoring and introduction of agents with a low
blood–gas partition coefficient may further facilitate
low-flow anaesthesia. Second, current computer
programs simulating anaesthetic uptake may not
be correct because they are based on theoretical
models, although these programs may still be
useful to teach some basic principles of anaesthetic
uptake. Pharmaco-economical studies comparing
the costs of different agents and evaluating the
cost–effectiveness of future automated low-flow and
closed-circuit anaesthesia machines should use real
uptake data. Third, we could not find any correlation
between Vsevo and patient weight, height or BSA.
Even though some authors found weak correlations,23 most authors did not.5 6 8 24 At the tissue level,
no correlation was found between tissue–gas solubilities and age (except for a correlation between age and
desflurane muscle–gas partition coefficient, with
r20.59).15 Automated feedback-controlled administration of inhaled anaesthetics in future anaesthesia
delivery systems based on low-flow and closedcircuit anaesthesia may provide a clinical solution for
inter-individual variation in anaesthetic uptake.
Fourth, Eger has described that when the inspired
concentration is kept constant, the alveolar concentration of an inhaled anaesthetic increases more
slowly when CO is greater because anaesthetic
uptake is greater.3 We also found that anaesthetic
uptake is greater with a greater CO, but in our study
design the end-expired concentration was kept constant, and hence the rate of increase in alveolar
concentration was dependent on the speed of injection and vaporization of the agent only. While a
greater CO may slow induction (defined as the rate
of rise of brain concentration) with the former
technique (constant inspired concentration) and
hasten induction with the latter (constant endexpired concentration), this has yet to be proved
in vivo. When using the closed-loop feedback technique to maintain the end-expired concentration
constant, the rate of increase in alveolar concentration of an inhaled anaesthetic becomes independent
of both cardiac output and blood–gas solubility of
the agent.
We limited this study to 1 h because more data are
needed to define the clinically safe combinations of
fresh gas flow, sevoflurane concentration and duration of administration. While closed-circuit anaesthesia with sevoflurane up to 3–5 h was found previously
British Journal of Anaesthesia
to be safe,25 more recent data urge caution because of
the accumulation of possible nephrotoxic sevoflurane
degradation products.26 Studies using closed-circuit
anaesthesia to determine Vsevo during prolonged procedures will have to be delayed until this issue has
been clarified.
In summary, the SqRT and 4C models were inaccurate in their prediction of Vsevo, and their accuracy
did not improve when measured CO was used in the
calculations. Vsevo cannot be predicted by patient
characteristics but was greater in patients with a
higher cardiac output.
Appendix A
SEVOFLURANE UPTAKE ACCORDING TO THE SQUARE ROOT
1
OF TIME MODEL OF LOWE AND ERNST
During closed-circuit anaesthesia, uptake of a potent inhaled
anaesthetic has been predicted by the SqRT model.1 A “unit dose”
is taken up by the body during the first minute and during each
subsequent time interval (3 min, 5 min, 7 min, 9 min, 11 min,
13 min, etc.). In addition, the “prime dose” is required to saturate
the circuit, functional residual capacity (FRC) and arterial delivery system. The addition of the prime dose and unit dose over time
result in the “cumulative dose”.
(1) Unit dose (litre of vapour) 2 × f × MAC × λB/G × Q
(2) Prime dose (litre of vapour) f × MAC× λB/G ×Q + Vol × f × MAC
(3) Cumulative dose (litre of vapour)prime doseunit
dose × t = f × MAC× λB/G ×Q + Vol × f × MAC+2× f × MAC× λB/G ×Q× t
(4) Conversion of vapour into liquid: 1 litre vapour
MW/(24D) litre liquid
where ffraction of MAC administered (in this study 1.3); MAC
minimal alveolar concentration (for sevoflurane 2%); ␭B/G
blood–gas partition coefficient (for sevoflurane 0.6); Qcardiac
output (CO, litre min1), calculated, based on Brody:
CO0.2weight (kg)0.75: 4.68 (0.44) litre min1 (n34), and
measured, in the CO group: 4.36 (0.18) litre min1 (n12); ttime
(min); tsquare root of time; Volvolume of circuit and FRC
(litre); MWmolecular weight (for sevoflurane 200); Ddensity
(for sevoflurane 1.505103 g litre1).
Appendix B
SEVOFLURANE UPTAKE ACCORDING TO THE FOURCOMPARTMENT MODEL OF EGER
3
Anaesthetic uptake by an organ at constant arterial concentration
can be calculated when the size of the organ, solubility of the
anaesthetic vapour in the organ and organ perfusion are known.1 2
Organs and tissues are grouped into VRG, MG, FG and VPG
(vessel-rich group, muscle group, fat group and vessel poor group,
respectively). VPG is ignored in the calculations because its contribution to uptake during the first hour is small and because sevoflurane solubility in these tissues has not been determined. The
amount of liquid anaesthetic needed to load the anaesthesia
breathing system and an estimated functional residual capacity
(FRC) of 2 litre with 2.6% sevoflurane was 1.3 ml of liquid
sevoflurane. This dose is included in the model prediction.
(1) Organ capacity (litre vapour) f × MAC × λ T/G ×VO
(2) Organ time constant (t, min) (VO× λ T/B )/ QO
(3) Organ uptake (litre vapour min1)amount delivered
fraction absorbed (Ca ×10 × QO ) × e -t/τ
(4) Cumulative organ uptake (litre vapour)(fMAC
λ T/G × VO ) × (1 − e t/τ )
where VOorgan volume (litre); QOorgan blood flow (litre min1);
τorgan time constant (min); ␭T/Gtissue–gas partition coefficient
(see Laster and colleagues10); ␭T/Btissue–blood partition coefficient (see Laster and colleagues10); Caarterial concentration
(fMAC␭B/G).
In table 1A, we present the information that was used to calculate Vsevo according to the four compartment model for all 34
patients combined and for the subgroup of 12 patients in which
CO was measured invasively.1 3 15 ␭B/G0.6; 3τ gives 95% organ
saturation.
Sevoflurane pharmacokinetics
501
Table 1A Vsevo according to the four-compartment model for all 34 patients combined and for the subgroup of 12 patients in which cardiac
output (CO) was measured invasively.1 3 15 λB/G0.6; 3 gives 95% organ saturation. MGMuscle group, FGfat group
Brain
Liver
Heart
Kidney
Blood
MG
FG
␭T/B
␭T/G
1.70
1.15
1.85
1.25
1.78
1.21
1.15
0.78
—
0.6
3.13
2.38
47.5
34.0
All patients combined
V0 (litre)
Organ capacity (ml liquid)
Organ blood flow (Q0, litre min1)
τ (min)
1.41
0.23
0.74
3.23
3.82
0.69
1.41
5.02
0.27
0.05
0.22
2.18
0.27
0.03
1.19
0.26
4.70
0.38
—
—
28.58
9.77
0.59
151
10.07
49.16
0.22
2174
CO group
V0 (litre)
Organ capacity (ml liquid)
Organ blood flow (Q0, litre min1)
τ (min)
1.49
0.25
0.69
3.66
4.04
0.72
1.31
5.70
0.28
0.05
0.21
2.37
0.28
0.03
1.10
0.30
4.96
0.40
—
—
30.16
10.31
0.55
171
10.62
51.87
0.21
2402
References
1. Lowe HJ, Ernst EA. The Quantitative Practice of Anesthesia—
Use of Closed Circuit. Baltimore: Williams and Wilkins, 1981.
2. Lowe HJ. Flame ionization detection of volatile organic anesthetics in blood, gases and tissues. Anesthesiology 1964; 25:
808–814.
3. Eger EI II. Anesthetic Uptake and Action. Baltimore/London:
Williams and Wilkins, 1974.
4. Lockhart SH, Cohen Y, Yasuda N, Freire B, Taheri S, Litt L,
Eger EI II. Cerebral uptake and elimination of desflurane,
isoflurane, and halothane from rabbit brain: an in vivo NMR
study. Anesthesiology 1991; 74: 575–580.
5. Lockwood GG, Chakrabarti MK, Whitwam JG. The uptake of
isoflurane during anaesthesia. Anaesthesia 1993; 48: 748–752.
6. Hendrickx JFA, Soetens M, Van der Donck A, Meeuwis H,
Smolders F, De Wolf AM. Uptake of desflurane and isoflurane
during closed-circuit anesthesia with spontaneous and controlled mechanical ventilation. Anesthesia and Analgesia 1997;
84: 413–418.
7. Walker TJ, Chakrabarti MK, Lockwood GG. Uptake of desflurane during anaesthesia. Anaesthesia 1996; 51: 33–36.
8. Hendrickx JFA, De Wolf AM. Desflurane uptake during liver
transplantation and liver resection using closed-circuit anesthesia. Anesthesia and Analgesia 1997; 84: S497.
9. Eger EI II. Complexities overlooked: things may not be what
they seem. Anesthesia and Analgesia 1997; 84: 239–240.
10. Laster MJ, Taheri S, Eger EI II, Liu J, Rampil IJ, Dwyer R.
Visceral losses of desflurane, isoflurane, and halothane in
swine. Anesthesia and Analgesia 1991; 73: 209–212.
11. Fassoulaki A, Lockhart SH, Freire BA, Yasuda N, Eger EI II,
Weiskopf RB, Johnson BH. Percutaneous loss of desflurane,
isoflurane, and halothane in humans. Anesthesiology 1991; 74:
479–483.
12. Morio M, Fujii K, Satoh N, Imai M, Kawakami U, Mizuno T,
Kawai Y, Ogasawara Y, Tamura T, Negishi A, Kumagai Y,
Kawai T. Reaction of sevoflurane and its degradation products
with soda lime. Anesthesiology 1992; 77: 1155–1164.
13. Shiraishi Y, Ikeda K. Uptake and biotransformation of sevoflurane in humans: a comparative study of sevoflurane with
halothane, enflurane, and isoflurane. Journal of Clinical
Anesthesia 1990; 2: 381–386.
14. Landon MJ, Matson AM, Royston BD, Hewlett AM, White
DC, Nunn JF. Components of the inspiratory–arterial isoflu-
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
rane partial pressure difference. British Journal of Anaesthesia
1993; 70: 605–611.
Yasuda N, Targ AG, Eger EI II. Solubility of I-653, sevoflurane, isoflurane, and halothane in human tissues. Anesthesia
and Analgesia 1989; 69: 370–373.
Perl W, Rackow H, Salanitre E, Wolf G, Epstein RM.
Intertissue diffusion effect for inert fat-soluble gases. Journal
of Applied Physiology 1965; 20: 621–627.
Cohen EN, Chow KL, Mathers L. Autoradiographic distribution of volatile anesthetics within the brain. Anesthesiology
1972; 37: 324–331.
Strum DP, Eger EI II, Unadkat JD, Johnson BH, Carpenter
RL. Age affects the pharmacokinetics of inhaled anesthetics in
humans. Anesthesia and Analgesia 1991; 73: 310–318.
Lockwood GG, White DC. Effect of ventilation and cardiac
output on the uptake of anaesthetic agents from different
breathing systems: a theoretical study. British Journal of
Anaesthesia 1991; 66: 519–526.
Yasuda N, Lockhart SH, Eger EI II, Weiskopf RB, Johnson BH,
Freire BA, Fassoulaki A. Kinetics of deflurane, isoflurane, and
halothane in humans. Anesthesiology 1991; 74: 489–498.
Carpenter RL, Eger EI II, Johnson BH, Unadkat JD, Sheiner
LB. Pharmacokinetics of inhaled anesthetics in humans:
measurements during and after the simultaneous administration of enflurane, halothane, isoflurane, methoxyflurane, and
nitrous oxide. Anesthesia and Analgesia 1986; 65: 575–582.
Lerou JGC, Dirksen R, Beneken Kolmer HH, Booij LHD.
A system model for closed-circuit inhalation anesthesia.
I. Computer study. Anesthesiology 1991; 75: 345–355.
O’Callaghan AC, Hawes DW, Ross JAS, White DC, Wloch RT.
Uptake of isoflurane during clinical anaesthesia. Servo-control
of liquid anaesthetic injection into a closed-circuit breathing
system. British Journal of Anaesthesia 1983; 55: 1061–1064.
Westenskow DR, Jordan WS, Hayes JK. Uptake of enflurane: a
study of the variability between patients. British Journal of
Anaesthesia 1983; 55: 595–601.
Bito H, Ikeda K. Closed-circuit anesthesia with sevoflurane in
humans. Effects on renal and hepatic function and concentrations of breakdown products with soda lime in the circuit.
Anesthesiology 1994; 80: 71–76.
Eger EI II, Koblin DD, Bowland T, Ionescu P, Laster MJ, Fang
Z, Gong D, Sonner J, Weiskopf RB. Nephrotoxicity of sevoflurane versus desflurane anesthesia in volunteers. Anesthesia and
Analgesia 1997; 84: 160–168.