|

Received: 28 November 2019    Accepted: 16 December 2019

DOI: 10.1111/aas.13533

REVIEW ARTICLE

Understanding volume kinetics

Robert G. Hahn1,2

1
Research Unit, Södertälje Hospital,
Södertälje, Sweden The distribution and elimination kinetics of the water volume in infusion fluids can be
2
Karolinska Institutet at Danderyds Hospital studied by volume kinetics. The approach is a modification of drug pharmacokinetics
(KIDS), Stockholm, Sweden
and uses repeated measurements of blood hemoglobin and urinary excretion as input
Correspondence variables in (usually) a two-compartment model with expandable walls. Study results
Robert G. Hahn, Research Unit, Södertälje
show that crystalloid fluid has a distribution phase that gives these fluids a plasma vol-
Hospital, 152 86 Södertälje, Sweden.
Email: r.hahn@telia.com and robert.hahn@ ume expansion amounting to 50%-60% of the infused volume as long as the infusion
sll.se
lasts, while the fraction is reduced to 15%-20% within 30 minutes after the infusion
ends. Small volumes of crystalloid barely distribute to the interstitium, whereas rapid
infusions tend to cause edema. Fluid elimination is very slow during general anes-
thesia due to the vasodilatation-induced reduction of the arterial pressure, whereas
elimination is less affected by hemorrhage. The half-life is twice as long for saline
than for Ringer solutions. Elimination is slower in conscious males than conscious
females, and high red blood cell and thrombocyte counts retard both distribution and
re-distribution. Children have faster turnover than adults. Plasma volume expansions
are similar for glucose solutions and Ringer's, but the expansion duration is shorter
for glucose. Concentrated urine before and during infusion slows down the elimina-
tion of crystalloid fluid. Colloid fluids have no distribution phase, an intravascular
persistence half-life of 2-3 hours, and—at least for hydroxyethyl starch—the ability to
reduce the effect of subsequently infused crystalloids. Accelerated distribution due
to degradation of the endothelial glycocalyx layer has not yet been demonstrated.

Volume kinetics is pharmacokinetics for infusion fluids and describes Therefore, a strong inverse correlation exists between the blood
how an infused volume is distributed and eliminated. Volume kinetics water and the blood Hb concentration at baseline1-3 (Figure 1A) as
opens up possibilities for simulation of the outcome of infusions that well as during infusions of fluid (Figure 1B). The Hb molecule has
have not yet been performed and, within certain limits, to study basic a slow metabolic turnover and does not undergo capillary leakage,
fluid physiology. A benefit of the approach is that it allows analysis which makes it a suitable index of the water volume kinetics. Using
of dynamic events, which is difficult with radioactive tracer methods. the dilution of Hb as an input variable in a volume kinetic analysis
yields information about the kinetics of the infused water volume.
By contrast, the volume of the circulating blood does not matter.
1 | TH E D I LU TI O N CO N C E P T

Pharmacokinetics uses mathematical fitting of repeated measure- 2 | TH E K I N E TI C M O D E L

ments of the plasma concentration of a drug to a kinetic model that
aims to correspond reasonably well to body physiology. Infusion flu- Volume kinetics detects a “wall” separating a water volume that
ids are devoid of an easily grasped concentration concept because equilibrates quickly with the site of infusion (venous blood) and
almost all the infused fluid consists of water. The blood contains 80% a more remote water volume with which the equilibration occurs
water, with the remainder being almost exclusively hemoglobin (Hb). more slowly. For this purpose, serial measurements of the blood Hb

|
570     © 2019 The Acta Anaesthesiologica Scandinavica wileyonlinelibrary.com/journal/aas Acta Anaesthesiol Scand. 2020;64:570–578.
Foundation. Published by John Wiley & Sons Ltd
HAHN |
      571

concentration and the urinary excretion are used as input variables

in (usually) a kinetic model that consists of two volume compart- Editorial Comment
ments. Both a clearance model and a micro-constant model have
been used (Figure 1C). In these models, fluid is infused in a central In this narrative review concerning intravascular fluid ki-

volume (Vc) and distributes and re-distributes to a peripheral vol- netics, perioperative theoretical and practical perspectives

ume (Vt). Elimination occurs from the central volume. The major dif- are presented and explained. Avenues for future research

ference between traditional pharmacokinetics and volume kinetics in this subject area are also presented.

is that Vc and Vt can expand. In fact, their expansion exerts the most
important clinical effect of the therapy. Findings made with the
clearance model were summarized 10 years ago.4 Therefore, this re- group, followed by a search for individual-specific factors that can
view focuses on findings generated with the micro-constant model. mathematically affect these estimates. This procedure allows a
more precise prediction of the input parameters. The most com-
monly studied covariates in drug research are age, gender, and
3 |  PH YS I O LO G I C A L CO R R E L ATE S body weight. In anesthesia, the arterial pressure is also an impor-
tant covariate. 8,9
Warnings are often raised regarding physiological interpretations of
pharmacokinetic analyses, but volume kinetics undoubtedly tempts
the reader to make interpretations in physiological terms. The size 5 | R E P O RTI N G TH E R E S U LT S
of Vc is usually very close to the plasma volume, as indicated by an-
thropometric measures that are, in turn, based on radioactive tracer The result of a volume kinetic analysis might be illustrated in several
measurements. The rate constant k12, multiplied by the expansion of ways. The optimal parameter estimates can be reported together
Vc at any time, can be interpreted as representing the capillary leak- with their confidence intervals, giving the reader the possibility to
age of water. The rate constant k 21 represents re-distribution, and compare results between studies10 (Table 1). However, the com-
actually yields flow patterns that agree well with direct measure- bined effect of fixed parameters and covariates can be difficult for
ments of the lymphatic flow in the thoracic duct5 (Figure 2). Apart the reader to interpret. A good choice is instead to illustrate the out-
from urinary excretion, a second elimination function is often statis- put graphically by comparing relevant situations, either by plotting
tically justified, and this then represents filtered fluid that does not parameter estimates or by simulating infusions.
return to the circulation during the period of the experiment. The The first option, plotting parameter estimates, is illustrated in
symbol used for this flow is kb. Figure 3. The box-plots show the parameter estimates in a popula-
tion kinetic analysis of eight infusions of crystalloid fluid in women
with mild to moderately severe toxicosis of pregnancy; their val-
4 |  P O PU L ATI O N K I N E TI C A N A LYS I S ues are compared with those of eight gestation-week-matched
controls undergoing a normal pregnancy.11 Toxicosis is associated
A modern form of pharmacokinetics implies that a series of experi- with a slightly faster elimination (higher k10) and a greatly retarded
ments is not analyzed one by one but in a single run. The average re-distribution (low k 21), which both reduce the plasma volume
values of the basic (fixed) parameters are first obtained for the (smaller Vc).

F I G U R E 1   Inverse relationship between the blood water and blood hemoglobin concentrations. A, At rest. Data from in 12 healthy
volunteers before and after eating a meal2 and in 18 renal failure patients before and after hemodialysis.3 B, During fluid infusion. Sampling
made in the course of two infusion experiments performed in 1992, when 1 and 1.5 L of Ringer's acetate was infused over 15 minutes.
These were the first cases ever to be analyzed by volume kinetics. C, The micro-constant model most often used for volume kinetic studies
today
 |
572       HAHN

6.1 | Volume of distribution

Infusion of 10-25 mL/kg over 30 minutes expands the entire plasma

volume but only 2/3 of the physiological interstitial fluid space.
Hence, the expanded part of the interstitium is twice as large as the
plasma volume. This ratio is evidenced by the ratio k12/k 21 having
a value close to 2 (Table 1). By contrast, tracer molecules without
notable volume equilibrate across the entire interstitial fluid space.12

6.2 | Distribution phase

The distribution half-life is approximately 8 minutes, which means that

F I G U R E 2   Lymphatic flow in the thoracic duct when the distribution requires 30 minutes for completion. As long as the
infusing 20 mL/kg of isotonic crystalloid fluid over 5 min infusion is ongoing, the clinician rides on a “wave” of plasma volume
in dogs[5] (dotted line) compared to the corresponding re-
expansion that makes crystalloid fluid a much better acute volume ex-
distribution of fluid as predicted from the product of k 21 and
pander than is usually believed8,13 (Figure 5). The distribution phase is
(v t–Vt) based on volume kinetic data from two studies in humans
(solid blue 6 and red7 lines). The re-distribution is 10 times greater more pronounced for short infusions, which means that bolus infusions
than the thoracic duct flow. The faster reduction of the flow in may cause re-bleeding in cases of uncontrolled hemorrhage.14
the dogs is probably due to faster fluid turnover depending on
their smaller body size.

6.3 | Small volumes
This combination of parameter estimates agrees well with the
clinical findings of hypovolemia and peripheral edema in toxico- Infusion of small volumes in conscious humans, such as 2-5 mL/kg over
sis. It also provides an avenue for computer simulation of the out- 15-30 minutes, are rapidly excreted and only a limited fraction of the
come of different fluid infusions that are not performed in reality. fluid undergoes distribution.15 Hence, they are effective plasma volume
Examples of such simulations are given in Figure 4, which illustrate expanders (Figure 6). The reason is probably that the increased filtra-
the influence of mean arterial pressure and age on the kinetics of tion pressure is not sufficient to expand the gel-like interstitial matrix.
crystalloid fluid in 78 humans, some of which were under general
anesthesia. 8
6.4 | Rapid infusions

6 |  C RYS TA LLO I D FLU I D High-rate infusions (>40-50 mL/min) are associated with a slower re-
distribution (lower k 21); therefore, they have an inherent tendency
Approximately 60 published studies have reported volume kinetic to cause tissue edema.15 The reason is probably that the lymphatic
analyses. The following is a summary of the most important results flow rate is not sufficiently high to fully compensate for the capillary
with isotonic crystalloid fluid. fluid filtration.

TA B L E 1   Population kinetic
  Covariate Best estimate 2.5% CI 97.5% CI CV%
parameters for 29 patients undergoing
Fixed parameters thyroid surgery. Re-analysis of the data
Vc (L) — 2.67 2.26 3.08 7.8 published by Ewaldsson.10

k12 (10 −3 min−1) — 64.8 48.7 81.5 12.8

k21 (10 −3 min−1) — 32.0 23.1 40.9 14.2
k10 (10 −3 min−1) — 2.4 1.6 3.2 16.4
k b (10 −3 min−1) — 5.3 1.7 8.9 34.7
Covariate effects
k10 Mean arterial 3.85 1.84 5.75 23.8
pressure
Vc Body weight 1.18 0.71 1.66 20.3
3.85
Note: The individual (ind) k10 has the form: k10 ind = 2.67 [(MAPind/ population mean MAP)^ ].
Abbreviations: CI, confidence interval; CV, between-subject coefficient of variation.
HAHN |
      573

urinary excretion during surgery,17 but it is still far from being as ef-
fective as elimination in the conscious state. The patient's capacity to
handle infused fluid in the post-operative setting appears to be much
better.18

6.6 | Slow distribution during onset of anesthesia

Infusion of fluid during the induction of epidural, spinal, and general

anesthesia causes excessive hemodilution that correlates strongly
with the accompanying decrease in arterial pressure.19-22 By con-
trast, the induction of spinal anesthesia without hypotension shows
a normal distribution profile.19,23 The transfer of fluid between Vc
and Vt even becomes arrested when the mean arterial pressure be-
comes established at 20 mmHg below baseline.16 The lowered arte-
rial pressure probably reduces the capillary fluid filtration until a new
Starling equilibrium is established.

6.7 | Hemorrhage

Loss of blood up to a near-fainting state is followed by a 50% reduction in

the rate of elimination of infused fluid, whereas the distribution function
is only marginally affected. No excessive intravascular retention of the
F I G U R E 3   Box-plots showing the distribution of the three infused fluid occurs as a result of the hemorrhage, which implies a risk
kinetic parameters governing the distribution of infused Ringer's of rebound hypovolemia.24 The volume kinetics in hypotensive hemor-
acetate during normal pregnancy and in mild/moderate toxicosis rhage is not known, but as yet unpublished experiments in pigs suggest
of pregnancy. All three parameters differ significantly by at
that rebound hypovolemia occurs there as well. Hence, the body's han-
least P < .05. Population kinetic analysis based on data from
Reference 11 dling of infused fluid in hemorrhage differs greatly from its handling in
vasodilatation-induced hypovolemia. The effect of vasodilatation seems
to overshadow the effect of hemorrhage on fluid kinetics.25
6.5 | Slow elimination during anesthesia

The half-life of a crystalloid fluid is between 20 and 50 minutes in 6.8 | Saline

conscious volunteers16 and is usually longer in males.6 Based on the
urine flow rate, the elimination is approximately 10 times longer dur- The kinetics between lactated and acetated Ringer's show no differ-
ing general anesthesia.16 The main cause appears to be the lowered ences, but 0.9% saline has a half-life that is almost twice as long as for
8,9
mean arterial pressure. Stimulating the alpha-1-receptors with phe- Ringer's.7 Therefore, 0.9% saline has a slightly stronger plasma volume
nylephrine and blocking the beta-1-receptors with esmolol triples the expanding effect (+10%) than Ringer's.26 However, the difference in

F I G U R E 4   The volume expansion when 1.5 L of Ringer's lactate is infused over 45 minutes in conscious and anesthetized adults of
both genders. A, central body fluid space (the plasma) and B, the peripheral space (interstitium) depending on two different mean arterial
pressures but in subjects of the same age. C and D, the same plots but with different subject ages. Simulations based on data from
Reference 8
 |
574       HAHN

F I G U R E 5   The volume expansion of the central body fluid space (the plasma) when 1.5 L of crystalloid fluid is infused at different rates.
A. Plot based on mean kinetic data from 30 experiments in 10 conscious males receiving acetated or lactated Ringer's and isotonic saline on
different occasions.7 B. Plot based on kinetic data from 25 females undergoing hysterectomy under general anesthesia.13

urinary excretion rate is only apparent in subjects who have concen- children weighing 15 kg than in adults. 28 In general, the elimina-
27
trated urine before the infusion; this is a sign of renal fluid retention. tion of crystalloid fluid slows down with advanced age (Figure 4),
but no difference in kinetics was found during induction of epi-
dural or general anesthesia when comparing patients younger or
6.9 | Age older than 65 years. 22

Just before anesthesia induction, the rate of elimination of lac-

tated Ringer's was several times higher per kg body weight in 6.10 | Gender

The plasma volume expanding effect is slightly larger in males than

females when fluid is infused in mL/kg; this is due to a slower rate of
excretion (half-lives 38 versus 24 minutes).6 Because females have a
lower plasma volume at baseline (Vc), Nyberg et al recently reported
a greater plasma dilution in response to 0.9% saline in women than
in men. 25 Whether these differences are due to the phase of the
menstrual cycle is unclear.

6.11 | Blood composition

Red blood cell and platelet counts in the upper end of the normal range
are associated with retarded distribution and re-distribution.7 The sum-
mary effect over time is a greater expansion of the peripheral space,
with only a minimal effect on the central space. A leukocyte count in
the upper part of the normal range does not change the fluid kinetics.
Extreme values, as found in leukemia, have not yet been studied.

6.12 | Post-operatively

Awakening from general anesthesia is associated with a rise in ar-

terial pressure and urinary excretion. 29,30 The plasma volume de-
creases by 10% over a 30-minute-period. In kinetic terms, k12 is
markedly reduced while k 21 and k10 increase, the latter by 10 times
to reach nearly normal values for the conscious state.
F I G U R E 6   The volume of fluid residing in the central (plasma)
The diuretic response to Ringer's is even higher 4 hours later
and peripheral (interstitial) fluid space after infusing acetated
Ringer's at two different rates. Peripheral accumulation is more than before the surgery. This response is unaffected by the
profound with the high-rate infusion. Simulation based on data amount of intraoperative crystalloid fluid given18 but might be
from conscious volunteers.15 further boosted if hydroxyethyl starch is also given during the
HAHN |
      575

surgery. 31 Although a colloid stimulates diuresis during surgery because some of the recruited fluid is subject to urinary excretion.42
and shortly thereafter, the urinary excretion might be lowered on Although debated, the traditional view is that the interstitium is the
the first post-operative day once the colloid has been excreted/ source of the recruited non-circulating fluid. The maximum plasma
metabolized. 32 volume expansion occurs 20-30 minutes after an infusion ends, and
the half-life is very long (between 6 and 10 hours) both in volunteers
and in patients who have undergone major surgery.42
7 |  OTH E R FLU I DS

7.1 | Colloid fluids 8 | M I S C E LL A N EO U S

Albumin 5% and hydroxyethyl starch (Voluven) expand the plasma vol- 8.1 | Interactions
ume by almost the same volume as is infused. These colloids have no
distribution phase, and elimination occurs with a half-life of approxi- If a crystalloid fluid is infused after hydroxyethyl starch (Voluven),
33,34
mately 2 hours. They have a slight dehydrating effect on volun- the crystalloid is rapidly extravasated and adds very little to the
teers.16 The kinetics of these colloids during surgery is poorly studied, plasma volume expansion.34 The elimination of the crystalloid is very
but a dramatic reduction in their turnover, similar to that observed for slow, which correlates with the increase in plasma oncotic pressure
crystalloid fluid, seems unlikely because their intravascular persistence that occurs from the starch.43 The water-binding properties of col-
is mainly explained by capillary leakage of the infused macromolecules. loid macromolecules that have leaked across the capillary wall prob-
ably contribute to the slow elimination.

7.2 | Glucose solutions
8.2 | Bolus versus continuous infusion
Plain 5% glucose and 2.5% glucose with electrolytes expand the
plasma volume to a degree similar to that seen with Ringer's lac- The efficacy of a fluid infusion is always better with a continu-
tate. 35 However, the expansion disappears faster post-infusion due ous infusion than with a bolus infusion. This is a consequence of
to uptake of glucose into the body's cells; this uptake carries along the exponential nature of both the distribution and elimination
water by means of osmosis. During surgery, this uptake occurs functions, which is also the case for nearly all biologic processes
more slowly due to insulin resistance, making 2.5% glucose an even (Figure 6).
more effective plasma volume expander. 29 Due to improved glu-
cose turnover, the efficacy of glucose 2.5% as a plasma volume ex-
pander is intermediately good during the post-operative phase. 36 8.3 | Dehydration

The rate of elimination of crystalloid fluid is retarded if a volunteer is

7.3 | Hypertonic solutions made dehydrated with furosemide44 or if a patient has concentrated
urine as evidence of renal fluid retention.45 In the elderly, the plasma
In volunteers, the plasma volume is expanded 4.0 times more by hy- volume expansion is considerably greater if the voided urine is concen-
pertonic (7.5%) saline than by 0.9% saline.37 Experiments in sheep trated (urine osmolality >700 mosmol/kg) before the infusion starts.27
show that allocation of intracellular fluid to the plasma requires ap-
proximately 15 minutes for completion and that the elimination is
governed by the effectiveness of the sodium excretion.38 Hypertonic 8.4 | Shock states
(7.5) saline in 6% dextran 70 supersedes the plasma volume expan-
sion from 0.9% saline over time by a factor of 7.2.37 Sepsis in sheep,46 the transurethral resection syndrome pigs,47 and
14 39
Computer simulations and animal experiments show that toxicosis of pregnancy in humans (Figure 3) are unique in the sense
the recommended dose of hypertonic saline in dextran to combat that they are all characterized by arrested re-distribution (low or
hemorrhage (4 mL/kg) is too large. Hypervolemia is produced, and negative k 21). Hence, very little of the filtered fluid returns to the
re-bleeding occurs in nearly all animals with uncontrolled hemor- plasma, thereby promoting peripheral edema, hypovolemia, and a
rhage. The best survival is obtained with 1 mL/kg over 5 minutes.40 low cardiac output. In sepsis, the elimination rate is also very low.46

7.4 | Hyperoncotic solutions 8.5 | The glycocalyx

Infusions of 20% albumin recruit three times the infused volume to Degradation of the endothelial glycocalyx layer is believed to increase
the plasma.41 However, the plasma volume expansion is only doubled the capillary leakage of infusion fluids, which can be indicated by an
 |
576       HAHN

increased k12. This degradation has not been possible to demonstrate conversion factor between dilution and volume. Vc might still be in-
by volume kinetics in patients undergoing surgery for cholecystitis or terpreted to represent the plasma volume, and only Vc is dependent
appendicitis48 or for 20% albumin in patients who have undergone on body size.
major surgery,42 probably because elevations of the plasma concen- Both kinetic models are independent of the blood volume and
trations of degradation products have been quite small. can indicate distributions of fluid that are much larger than the vol-
ume of the circulating blood. The reason is that the Hb changes are
governed by the distribution of the infused water volume and not
8.6 | “Third-space” losses by the distribution of the amount of Hb in the blood. However, the
model is sensitive to changes in the Hb mass during an experiment;
Fifteen years ago, two studies in sheep detected a discrepancy therefore, correction for surgical blood loss, if any, as well as loss due
between the urinary excretion and the model-predicted elimi- to blood sampling, is recommended.4,10 Differences in transport time
nation of fluid, and the researchers attributed the excessive ex- between plasma and the erythrocytes are not relevant. Variations
travasation to isoflurane anesthesia.49,50 Similar extravasation in the Hb content between vascular beds affect Vc, but not the
was also found in surgical patients, during isoflurane and propofol fluid distribution, as long as the variations remain stable during an
anesthesia.10 experiment.
The greater stability of the population kinetic approach has made
possible the routine estimation of kb, which describes a first-order
rather than a zero-order process.15 The value of kb appears to be 1/5 10 | C LI N I C A L I M PLI C ATI O N S
to 1/3 of the k10 in volunteers,6,7,15 while kb attains a value similar
to, or even higher than, k10 during general anesthesia8 (Table 1). This 1. The distribution gives crystalloids a much better acute plas-
flow is believed to be filtered fluid that is not in balance with the ma-volume-expanding effect than is commonly believed. The
circulating plasma. downside is that a bolus infusion causes a distinct peak in
the volume expansion, which might stimulate hormonal re-
sponses of fluid overload and cause re-bleeding in cases of
9 |  M O D E L A S S U M P TI O N S uncontrolled hemorrhage.14
2. Crystalloid fluid is a particularly effective plasma volume ex-
In the clearance model, the flows of fluid are governed by differ- pander (up to 100%) during anesthesia-induced hypotension
ences in dilution of the water volumes in Vc and Vt. The output in a until a new Starling equilibrium is established, which might re-
clearance model is dependent on body size. Vc is a key parameter and quire 20-30 minutes.13
4
is interpreted to represent the plasma volume. 3. The rebound hypovolemia occurring when crystalloid fluid is
In the micro-constant model, the flows are governed by the vol- used to combat hemorrhage-induced hypovolemia should be
ume expansion of Vc and Vt. The fluid distribution is given only by the handled with by a gradual step-down of the infusion rate.24
micro-constants (k12, k21, k10, and kb), which are independent of body 4. The severe depression of the urinary excretion during general
size and infused volume. Therefore, the calculated fluid distribution anesthesia invalidates urine volume as a measure of fluid over-
will be the same, even if all data on plasma dilution is multiplied by load. The urine flow rate instead reflects the arterial pressure,
a fixed value. Only Vc would change, as this parameter serves as a the catecholamine pattern, and the patient's age.8,17

F I G U R E 7   A, Plasma volume expansion when 1 L of isotonic saline (blue line) and acetated or lactated Ringer's (red line) is infused over
30 minutes in male volunteers.7 B, Plasma volume expansion when 1 L of Ringer's acetate (blue line) or colloid fluid (red and green lines) is
infused over 30 minutes in male volunteers. 24,33,34 C, Ratio between the plasma volume expansion created by infusing 1 L of hydroxyethyl
starch (Voluven) and Ringer's acetate in male volunteers at different rates (30, 180, and 400 minutes). 24,34
HAHN |
      577

10. The most widely recommended scheme for intravenous sup-

plementation of fluid during surgery in adults is to infuse
3-5 mL kg−1 h−1 of Ringer's. This program expands the plasma
volume by 150-350 mL and must be preceded by a bolus of 300-
500 mL to achieve a stable plasma volume expansion (Figure 8).

11 | CO N C LU S I O N S

Volume kinetics allows the analysis of the behavior of infused fluid

volumes in the body. In contrast to isotope methods, the kinetics
method provides information about time courses of the fluid distri-
bution between (usually) two compartments, presumably the plasma
and the interstitial fluid volumes. Volume kinetics has mostly been
used to study crystalloids in volunteers and during surgery, while the
colloid fluids have been less well-investigated. Several findings made
in the volume kinetic studies enhance our understanding of the ef-
fects and physiology of infusion fluids in humans.

C O N FL I C T O F I N T E R E S T
The author holds a research grant from Grifols for studies of 20%
albumin.

ORCID
Robert G. Hahn  https://orcid.org/0000-0002-1528-3803
F I G U R E 8   The volume of Ringer's lactate/acetate residing in
the central (plasma) fluid space for two commonly used schemes
REFERENCES
of intraoperative fluid administration in surgeries with small
hemorrhage (up to 150 mL). No vasopressor was used. Shown are 1. Lijnema TH, Huizenga JR, Jager J, Mackor AJ, Gips CH. Gravimetric
data-based population kinetic analyses of 20 patients undergoing determination of the water concentration in whole blood, plasma
and erythrocytes and correlations with hematological and clini-
laparoscopic gynecological surgery (top line),17 of 25 patients
co-chemical parameters. Clin Chim Acta. 1993;214:129-138.
subjected to open hysterectomy (middle line),13 and of 29 patients
2. Hahn RG, Norberg Å, Gabrielsson J, Danielsson A, Jones AW. Eating
during thyroid surgery (bottom line).10
a meal increases the clearance of ethanol given by intravenous infu-
sion. Alcohol Alcohol. 1994;29:673-677.
3. Jones AW, Hahn RG. Pharmacokinetics of ethanol in patients
5. The slow elimination during surgery severely impairs the body's
with renal failure before and after hemodialysis. Forensic Sci Int.
capacity to deal with fluid overload and enhances the impor- 1997;90:175-183.
tance of correct timing when providing fluid in this setting. 4. Hahn RG. Volume kinetics for infusion fluids (review). Anesthesiology.
6. The kinetics of crystalloid fluid is context sensitive. In addition to 2010;113:470-481.
5. Brace RA, Power GG. Effects of hypotonic, isotonic and hy-
the situations listed above, the turnover of crystalloid is affected
pertonic fluids on thoracic duct lymph flow. Am J Physiol.
by gender, age, and glucose content. Isotonic saline has twice as 1983;245:R785-R791.
long a half-life than acetated and lactated Ringer's26 but the dif- 6. Hahn RG. The elimination half-life of crystalloid fluid is shorter in
ference in plasma volume expansion is limited (Figure 7A). female than in male volunteers; a retrospective population kinetic
7. Colloid fluid has no distribution phase but has an intravascular analysis. Biol Sex Diff. 2016;7:54.
7. Hahn RG. Influences of the red blood cell count on the distribution
persistence half-life between 2 and 3 hours; this might be longer
and elimination of crystalloid fluid. Medicina. 2017;53:233-241.
for 20% albumin.42 8. Hahn RG. Arterial pressure and the elimination of crystalloid fluid: a
8. The difference in potency between crystalloid and colloid fluid population-based study. Anesth Analg. 2017;124:1824-1833.
is not a fixed value but changes over time. The difference is usu- 9. Lee J-H, Choo Y-J, Lee Y-H, et al. Population-based volume kinet-
ics of Ringer's lacate solution in patients undergoing open gastrec-
ally 1:2, with a difference of 1:3 during the distribution phase
tomy. Acta Pharmacol Sin. 2019;40:710-716.
of the crystalloid (Figure 7B). The potencies of crystalloids and 10. Ewaldsson C-A, Hahn RG. Kinetics and extravascular retention of
iso-oncotic colloids do not differ 8-10 hours later (Figure 7C). acetated Ringer's solution during isoflurane and propofol anesthe-
9. Isotonic glucose solution has a plasma-volume-expanding effect sia for thyroid surgery. Anesthesiology. 2005;103:460-469.
11. Drobin RG, Hahn RG. Distribution and elimination of crystalloid
that is similar to electrolyte crystalloid fluid, but it disappears
fluid in pre-eclampsia. Clin Sci. 2004;106:307-313.
faster.
 |
578       HAHN

12. Zdolsek J, Lisander B, Hahn RG. Measuring the size of the extracel- 31. Borup T, Hahn RG, Holte K, Ravn L, Kehlet H. Intraoperative colloid
lular space using bromide, iohexol and sodium dilution. Anest Analg. administration increases the clearance of a postoperative fluid load.
2005;101:1770-1777. Acta Anaesthesiol Scand. 2009;53:311-317.
13. Nemme J, Krizhanovskii C, Ntikia SO, Vanags I, Hahn RG. 32. Hahn RG. Renal water conservation determines the increase in
Hypervolemia does not cause degradation of the endothelial glyco- body weight after surgery; a randomized controlled trial. Saudi J
calyx layer during open hysterectomy performed under sevoflurane Anaesth. 2017;11:144-151.
or propofol anesthesia. Acta Anaesthesiol Scand. 2020. https​://doi. 33. Hedin A, Hahn RG. Volume expansion and plasma protein clearance
org/10.1111/aas.13511​. during intravenous infusion of 5% albumin and autologous plasma.
14. Hahn RG. Fluid therapy in uncontrolled hemorrhage - what exper- Clin Sci. 2005;106:217-224.
imental models have taught us (review). Acta Anaesthesiol Scand. 3 4. Hahn RG, Bergek C, Gebäck T, Zdolsek J. Interactions between the
2013;57:16-28. volume effects of hydroxyethyl starch 130/0.4 and Ringer's ace-
15. Hahn RG, Drobin D, Zdolsek J. Distribution of crystalloid fluid tate. Crit Care. 2013;17:R104.
changes with the rate of infusion: a population-based study. Acta 35. Sjöstrand F, Edsberg L, Hahn RG. Volume kinetics of glucose solu-
Anaesthesiol Scand. 2016;60:569-578. tions given by intravenous infusion. Br J Anaesth. 2001;87:834-843.
16. Hahn RG, Lyons G. The half-life of infusion fluids: an educational 36. Strandberg P, Hahn RG. Volume kinetics of glucose 2.5% solution
review. Eur J Anaesthesiol. 2016;33:475-482. and insulin resistance after abdominal hysterectomy. Br J Anaesth.
17. Li Y, Zhu HB, Zheng X, Chen HJ, Shao L, Hahn RG. Low doses of 2005;94:30-38.
esmolol and phenylephrine act as diuretics during intravenous an- 37. Drobin D, Hahn RG. Kinetics of isotonic and hypertonic plasma vol-
esthesia. Crit Care. 2012;16:R18. ume expanders. Anesthesiology. 2002;96:1371-1380.
18. Holte K, Hahn RG, Ravn L, Bertelsen KG, Hansen S, Kehlet H. 38. Svensén CH, Waldrop KS, Edsberg L, Hahn RG. Natriuresis and the
Influence of liberal vs. restrictive fluid management on the elim- extracellular volume expansion by hypertonic saline. J Surg Res.
ination of a postoperative intravenous fluid load. Anesthesiology. 2003;113:6-12.
2007;106:75-79. 39. Riddez L, Hahn RG, Suneson A, Hjelmqvist H. Central and regional
19. Hahn RG. Haemoglobin dilution from epidural-induced hypo- hemodynamics during uncontrolled bleeding using hypertonic sa-
tension with and without fluid loading. Acta Anaesthesiol Scand. line dextran for resuscitation. Shock. 1998;10:176-181.
1992;36:241-244. 4 0. Riddez L, Drobin D, Sjöstrand F, Svensén C, Hahn RG. Lower dose
20. Drobin D, Hahn RG. Time course of increased haemodilution in of hypertonic-saline dextran reduces the risk of lethal rebleeding in
hypotension induced by extradural anaesthesia. Br J Anaesth. uncontrolled hemorrhage. Shock. 2002;17:377-382.
1996;77:223-226. 41. Zdolsek M, Hahn RG, Zdolsek JH. Recruitment of extravascular fluid by
21. Ewaldsson C-A, Hahn RG. Volume kinetics during induction of spi- hyperoncotic albumin. Acta Anaesthesiol Scand. 2018;62:1255-1260.
nal and general anaesthesia. Br J Anaesth. 2001;87:406-414. 42. Hahn RG, Zdolsek M, Hasselgren E, Zdolsek J, Björne H. Fluid volume
22. Li Y, Zhu S, Hahn RG. The kinetics of Ringer's solution in young and kinetics of 20% albumin. Br J Clin Pharmacol. 2019;85:1303-1311.
elderly patients during induction of general and epidural anesthe- 43. Hahn RG. Why crystalloids will do the job in the operating room.
sia. Acta Anaesth Scand. 2007;51:880-887. Anaesthesiol Intensive Ther. 2014;46:342-349.
23. Hahn RG, Lindahl C, Drobin D. Volume kinetics of acetated Ringer's 4 4. Zdolsek J, Li Y, Hahn RG. Detection of dehydration by using volume
solution during experimental spinal anesthesia. Acta Anaesthesiol kinetics. Anesth Analg. 2012;115:814-822.
Scand. 2011;55:987-994. 45. Hahn RG, Bahlmann H, Nilsson L. Dehydration and fluid volume
24. Hahn RG, Drobin D, Li Y, Zdolsek J. Kinetics of Ringer's solution in kinetics before major open abdominal surgery. Acta Anaesthesiol
extracellular dehydration and hemorrhage. Shock. 2019. https​://doi. Scand. 2014;58:1258-1266.
org/10.1097/SHK.00000​0 0000​0 01422. 46. Li Y, Xiaozhu Z, Guomei R, Qiannan D, Hahn RG. Effects of vaso-
25. Nyberg J, Li H, Wessmark P, et al. Population kinetics of 0.9% saline active drugs on crystalloid fluid kinetics in septic sheep. PLoS ONE
distribution in hemorrhaged awake and isoflurane-anesthetized ONE. 2017;12:e0172361.
volunteers. Anesthesiology. 2019;131:501-511. 47. Hahn RG, Gebäck T. Fluid volume kinetics of dilutional hyponatre-
26. Hahn RG. Crystalloid fluids. In: Hahn RG, ed. Clinical fluid therapy in mia; a shock syndrome revisited. Clinics. 2014;69:120-127.
the perioperative setting, 2nd edn. Cambridge: Cambridge University 48. Li Y, Yi S, Zhu Y, Hahn RG. Volume kinetics of Ringer's lactate in
Press; 2016:3-9. acute inflammatory disease. Br J Anaesth. 2018;121:574-580.
27. Hahn RG, Nyberg Isacson M, Fagerström T, Rosvall J, Nyman CR. 49. Brauer KI, Svensén C, Hahn RG, Traber L, Prough DS. Volume ki-
Isotonic saline in elderly men: an open-labelled controlled infu- netic analysis of the distribution of 0.9% saline in conscious versus
sion study of electrolyte balance, urine flow and kidney function. isoflurane-anesthetized sheep. Anesthesiology. 2002;96:442-449.
Anaesthesia. 2016;71:155-162. 50. Connolly CM, Kramer GC, Hahn RG, et al. Isoflurane but not mechan-
28. Li Y, Hahn RG, Hu Y, Xiang Y, Zhu S. Plasma and renal clearances of ical ventilation promotes extravascular fluid accumulation during
lactated Ringer's solution in pediatric and adult patients just before crystalloid volume loading. Anesthesiology. 2003;98:670-681.
anesthesia is induced. Pediatric Anesthesia. 2009;19:682-687.
29. Sjöstrand F, Hahn RG. Volume kinetics of 2.5% glucose solu-
tion during laparoscopic cholecystectomy. Br J Anaesth.
How to cite this article: Hahn RG. Understanding volume
2004;92:485-492.
kinetics. Acta Anaesthesiol Scand. 2020;64:570–578. https​://
3 0. Olsson J, Svensén CH, Hahn RG. The volume kinetics of acetated
Ringer's solution during laparoscopic cholecystectomy. Anesth doi.org/10.1111/aas.13533​
Analg. 2004;99:1854-1860.