European Journal of
Eur J Appl Physiol (1986) 55:450--456
The effects of body position and exercise on plasma volume
dynamics*
J. M. Pivarnik, M. P. Goetting, and L. C. Senay, Jr.
Department of Physiology, St. Louis University School of Medicine, St. Louis, Missouri 63104, USA
Summary. We examined the plasma volume
changes associated with a protocol of either exer-
cise or controlled rest under identical positional
and ambient conditions. Nine healthy adult males
rode (E) and on another occasion sat quietly (C)
on a cycle ergometer for 30 min. Ten minutes of
cycle exercise immediately followed the resting
C protocol. Ambient temperature was 30~
(rh= 3.5%) and exercise load was equal to 50% of
peak Vo2. Venous blood samples were obtained
with subjects both in the supine and seated posi-
tions prior to all experiments. Additional blood
was drawn during minutes 1, 5, 10, and 30 in both
experimental conditions. A final sample was
taken during C after the 10 rain exercise. Moving
from the supine to a seated position resulted in an
average loss of 162 ml of plasma across all experi-
ments. During the E condition a further reduction
in plasma volume (76 ml) occurred by one minute
of exercise. Plasma volume stabilized by 5 rain of
exercise under the E protocol. During the C con-
dition, subsequent fluid loss (98 ml) was not ap-
parent until 10 min after the first seated sample
and totalled 176 ml at the end of 30 min of rest.
Ten minutes of cycling at the end of the C experi-
ment resulted in a further plasma volume reduc-
tion of 137 ml. Plasma protein and albumin con-
tents decreased by 5 min of exercise in E and by
30rain of rest in C. [Na+] and [C1-] did not
change in either condition but a rapid increase in
[K+] during exercise indicated an addition of po-
tassium to the vascular volume. An hypothesis
Offprint requests to: J. M. Pivarnik, Department of Health,
Physical Education and Recreation, University of Houston,
University Park, Houston, Texas 77004, USA
* This work was supported in part by the National Institutes
of Health, under Grant H207050-09
Applied
Physiology
and Occupational Physiology
9 Springar-Verlag 1986
concerning the factors involved in postural and
exercise body fluid shifts is presented.
Key words: Plasma proteins -- Human -- Cycle
ergometer exercise -- Posture -- Blood volume
Introduction
The hemoconcentration accompanying endurance
exercise upon a cycle ergometer has been shown
to depend not only on body position prior to
exercise and time spent in this position but also
upon any changes in position with onset of exer-
cise (Harrison 1985). While studies involving
body position are not new (Thompson et al. 1928;
Waterfield 193 la, 1931b), the relevance of posture
to reported results of plasma volume shifts has
only recently been appreciated. Hagan et al.
(1978) found that hemoconcentration occurs
when changing from a supine to a standing posi-
tion and that some 20 rain need elapse before an
apparent equilibrium can be attained. Addition-
ally (Diaz et al. 1979; Hagan et al. 1980), when
both control position and exercise intensity were
varied, it was found that the greater the loss of
plasma during pre-exercise, the less the hemocon-
centration during the exercise.
While it now appears obvious that pre-exer-
cise body position (and length of time spent in the
particular position) affects the experimental re-
sults, it would be naive to assume that the control
of plasma volume during rest would be the same
during exercise. Between rest and exercise, there
are marked differences in venous hydrostatic
pressure, capillary filtration pressure, tissue os-
motic pressure, and protein oncotic pressure.
Thus, control of plasma volume at rest is primar-
J. M. Pivarnik et al.: Exercise, position, and plasma volume
ily a hydrostatic phenomenon while a number of
factors quickly intrude with the onset of exercise
and diminish the importance of static fluid pres-
sures. Although changes in plasma volume have
been reported in the first few minutes of exercise
(Harrison et al. 1975; van Beaumont et al. 1981),
it is possible that they may become apparent even
sooner.
The present investigation had two main objec-
tives: to compare plasma volume shifts in a group
of subjects during exercise and rest under identi-
cal postural and environmental conditions, and to
seek information as to vascular changes during
the first minute of exercise. We sought insight into
the separate and combined influence of position
and exercise upon fluid dynamics. An hypothesis
dealing with the possible location of vascular
fluid lost during both experimental conditions
will be presented.
Methods
Nine healthy but untrained adult male volunteers were used as
subjects in the study. Following a detailed explanation of the
experimental procedures and the signing of consent forms,
each underwent a Bruce Stress Test conducted by the Division
of Cardiology. All prospective participants were pronounced
clinically fit for an exercise experiment and their anthropo-
metric characteristics are presented in Table 1.
On the day preceding his first experiment, the blood vol-
ume of each subject was measured using the carbon monoxide
(CO) rebreathing technique (Dahms and Horvath 1974). Red
cell and plasma volume estimates were based on this determi-
nation. The accuracy of this method for measuring CO in hu-
man blood samples was reported by Dahms and Horvath
(1974) as _+ 1.7%. Repeat estimates in the present study did not
differ by more than +2%. Additionally, aerobic capacities
(peak Vo2) were determined using a Monark cycle ergometer.
The pedal rate was 50 rev. min-~ and the load was increased
every two minutes. Beginning with 0 load, subjects then ped-
alled at 50 and 100 watts (W) followed by increments of 25 W.
Oxygen consumption was determined during the second min-
ute at each exercise level. Peak 12o2 was taken as the exercise
level at which subjects could no longer maintain pedal rhythm
and voluntarily stopped.
All subjects performed both an exercise (E) and a control
(C) experiment. Each C experiment was conducted within five
days after completion of the E. In order to insure adequate
and uniform hydration among subjects, each drank an amount
of tap water equal to 1% of body weight some 90 rain before
each experiment.
Upon arrival at the laboratory, each participant donned
athletic shorts and shoes and lay supine under a light blanket
for 30 min at room temperature (20 ~ C). At the end of the rest
period, an indwelling needle (Butterfly, Abbott) was inserted
into a cubital vein and a 9 ml blood sample (R) was taken. The
needle and attached tubing were back-flushed with a heparin-
in-saline solution (10 units 9ml - ~). After the blood sample was
obtained, the subject rose and walked into an adjacent envi-
ronmental chamber and sat on a cycle ergometer. After final
adjustments were made, a second blood sample (0) was ob-
451
Table 1. Anthropometric and other subject data
Age, Ht, Wt, Peak Vo2, Blood vol,
yr cm kg ml.kg-l.min-I ml-kg-I
X (n=9) 23.9 185.6 78.9 43.8 71.5
(sd) (0.6) (2.4) (2.8) (1.5) (1.7)
tained. Five minutes elapsed between the supine and seated
samples. The seated position required that both feet be on the
pedals equidistant from the floor, with hands resting loosely
on the handlebars. The ambient temperature was 30~
(rh=35%), with an air flow of 0.2 m . sec -~ at the subject's
waist level.
During the E condition, subjects began pedalling the er-
gometer immediately after the seated blood sample was taken.
The exercise load was set at 50% of peak Vo2 with a pedal rate
of 50 rev. min-~. Additional blood samples were obtained
during minutes 1, 5, 10, and 30 of exercise. The experimental
session was terminated after the 30 min sample was taken.
In contrast to the E sessions, subjects sat quietly on the
ergometer for 30 min in the C experiments. Blood samples
were taken at the same time intervals as during the E protocol.
At the conclusion of the resting protocol of the C experiment,
each subject pedalled for 10 min at 50% of peak Vo2. A final
blood sample was taken during the last minute of the exer-
cise.
Hematocrit was estimated in triplicate and corrected for
trapped plasma (Dill and Costill 1974). Whole blood hemoglo-
bin concentration was assayed in duplicate using the cyan-
methemoglobin method. Plasma was used to determine total
protein ([PP], biuret) and albumin ([Alb], bromocresol green)
concentrations. Plasma concentrations of sodium ([Na + 1) and
potassium ([K+]) were determined by flame photometry while
chloride ([CI-]) was measured amperometrically with a chlo-
ridometer. Mean (sd) values of the measured vascular consti-
tuents are presented in Table 2.
Relative changes in plasma volume were calculated from
hematocrit ratio and hemoglobin concentration using the Dill
and Costill model (1974). Absolute changes in plasma volume
were calculated by multiplying the relative change by the
plasma volume estimated from the CO technique. It was as-
sumed that red cell volume did not change during either ex-
perimental condition.
A repeated measures design analysis of variance
(ANOVA) was used to determine the statistical significance of
changes in plasma volume, electrolyte concentrations, and dif-
ferences between measured and expected protein and albumin
concentrations during the experimental sessions. Significant
main effects (p<0.05) were examined using the Tukey A test
(Winer 1971).
Results
Mean (with SE bars) absolute plasma volumes for
both the E and C conditions are presented in
Fig. 1. All subjects showed a hemoconcentration
as a result of the change from a supine to upright
seated position (p<0.01). Statistically, there was
no difference between the two conditions as aver-
452 J. M. Pivarnik et al.: Exercise, position, and plasma volume

Table 2. Selected variables in plasma

R 0 1 5 10 30 10X

Hct (%)
E 41.9 43.3 44.0 45.0 44.6 45.0
(1.4) (2.0) (1.7) (1.4) (1.5) (1.6)
C 42.4 43.3 43.1 43.6 44.0 44.4 45.6
(1.7) (1.8) (1.7) (2.0) (1.9) (2.1) (2.2)
Hb (gin- dl - 1)
E 14.5 15.0 15.2 15.6 15.7 15.8
(0.5) (0.8) (0.8) (0.7) (0.6) (0.6)
C 14.6 15.0 15.0 15.2 15.3 15.6 16.0
(0.9) (0.8) (0.8) (0.8) (0.7) (0.8) (0.9)
[PP] (gm. d l - ' )
E 6.7 7.0 7.2 7.4 7.3 7.4
(0.4) (0.5) (0.5) (0.5) (0.5) (0.4)
C 6.6 6.9 6.9 7.0 7.1 7.2 7.4
(0.5) (0.5) (0.5) (0.6) (0.5) (0.5) (0.5)
[Alb] (gm- dl- 1)
E 4.3 4.6 4.7 4.8 4.8 4.9
(0.2) (0.2) (0.2) (0.2) (0.2) (0.2)
C 4.5 4.6 4.6 4.7 4.8 4.8 5.0
(0.2) (0.2) (0.2) (0.2) (0.2) (0.1) (0.1)
[Na+] (mmol 9 1-1)
E 137.6 138.6 139.0 139.3 139.3 139.0
(1.4) (0.9) (1.6) (0.9) (1.1) (1.4)
C 137.1 137.6 137.8 137.5 137.8 137.5 138.6
(2.9) (3.0) (2.4) (2.5) (2.6) (1.5) (2.3)
[C1-] (mmol. 1 t)
E 99.7 99.4 98.8 99.4 100.4 100.6
(1.5) (2.1) (1.5) (1.5) (2.4) (2.1)
C 99.3 98.4 99.7 99.9 99.7 99.9 101.2
(2.0) (1.4) (1.8) (1.1) (1.5) (1.6) (1.3)
[K+] (mmol 9 1-1)
E 3.9 4.0 4.3 4.4 4.4 4.6
(0.2) (0.2) (0.3) (0.3) (0.2) (0.2)
C 4.0 4.1 4.1 4.1 4.1 4.1 4.6
(0.2) (0.2) (0.2) (0.2) (0.2) (0.2) (0.2)

Values are means (sd) of Exercise (E) and Control (C) condition. R indicates supine rest and 0, 1, 5, 10, 30 correspond to minutes
of the experimental condition. 10X refers to the end of the exercise which followed the rest period of C

age f l u i d loss a m o u n t e d to 162 ml o f p l a s m a w i t h T

<>
change of position. 3300-

U s i n g t h e initial s e a t e d b l o o d s a m p l e as a ref-
3200- / T
erence, plasma volume decreased by an average
o f 76 ml (p < 0.01) d u r i n g t h e first m i n u t e o f exer- o 3100-
cise in t h e E c o n d i t i o n . A f u r t h e r d e c r e a s e
(132 ml, p < 0 . 0 1 ) o c c u r r e d b y t h e fifth m i n u t e o f ooo

exercise w i t h n o s i g n i f i c a n t c h a n g e s o c c u r r i n g "'-
thereafter. • "'.,~
I n the C e x p e r i m e n t , p l a s m a v o l u m e d i d n o t 2800-
d e c r e a s e s i g n i f i c a n t l y f r o m t h e t i m e o f t h e initial
I I I I L 3I I
s e a t e d s a m p l e until 10 m i n u t e s i n t o t h e c o n t r o l l e d R 0 1 5 10 0 1OK
TIME (rain)
rest (98 ml, p < 0.01). I n c o n t r a s t to t h e E c o n d i -
tion, p l a s m a v o l u m e d e c r e a s e d b y 78 ml (p < 0.01) Fig. 1. Absolute plasma volumes in the Exercise (E) and Con-
d u r i n g t h e r e m a i n d e r o f t h e rest p e r i o d . By the trol (C) conditions. R refers to supine rest. 0, 1, 5, 10, 30 are
min during both experimental conditions. 10X refers to
e n d ot t h e 10 m i n u t e exercise s e s s i o n w h i c h fol- the end of the exercise which followed controlled rest in C.
l o w e d the rest, the s u b j e c t s lost a n a d d i t i o n a l Values not covered by the same line are significantly different.
137 ml o f p l a s m a (p < 0.01). (E) R 0 1 5-i0-3t3 (C) R t3-i-5 10 30 10X
J. M. Pivarnik et al.: Exercise, position, and plasma volume 453

140-
Figures 2 and 3 represent the total plasma pro- ,,/a~ a
.... ........
tein and albumin dynamics during the two experi- [Na+] 138-
mental conditions. Differences between measured (mmo[. (1) 1 3 6 -

and expected (based on plasma volume changes) 101 -

concentrations are plotted as a function of time.
FC ' I ~ 100-
Positive values would represent actual gains in to- (mmob I-~)
99-
tal protein or albumin content and negative val-
ues would indicate losses. 98-

In the E condition, a decrease in plasma pro-

tein content (p < 0.01) was seen by 5 min of exer- [K+I o:
(mmol.l -~)
cise with no further change occurring for the re-
mainder of the session. In the C condition, a pro-
I
tein loss was not apparent until the end of the R 0
I I
1
I
5
[
10
I
310 1Ix
~.,.~ ( ..,. )
seated rest (p < 0.05). Protein content was further
reduced (p<0.01) during the 10min exercise Fig. 4. Electrolyte concentrations during Exercise (E) and
which followed the rest. Control (C) conditions. R refers to supine rest. 0, 1, 5, 10, 30
are min during both experimental conditions. 10X refers to the
The albumin fraction of the plasma proteins end of the exercise which followed controlled rest in C. Differ-
followed the same pattern as did total protein. Al- ences in [Na+] and [C1-] were not statistically significant.
bumin was lost from the circulation by five min- [ K + ] values no_t_covered by the same line are significantly dif-
utes (p<0.01) in the E condition with no changes ferent. (E) R 0 5 10 30 (C) 1~-0-1-5-10-30 10X

-~" 0-
occurring thereafter. A decrease in albumin con-
tent was not found in the C condition until the
-1.0 -
30 min blood sample at the end of rest (p < 0.05).
0 -2.0- A further decrease occurred during the 10 minute
-3.0 -
exercise (p<0.01). The albumin to globulin ratio
did not change in either condition in the course of
< -4.0-
the experimental sessions.
I I I I ~ / I
0 1 5 10 3 IOX Figure 4 presents the plasma electrolyte con-
.... (.ni.) centrations in both experimental conditions as a
Fig. 2. Differences between measured and expected (based on function of time. There were no significant
plasma volume changes) total protein concentrations during changes in [Na + ] or [ C I - ] in either E or C. This
Exercise (E) and Control (C) conditions. 0, 1, 5, 10, 30 are rain
during both experimental conditions. 10X refers to the end was not true in the case of [K+]. In the E condi-
of exercise which followed controlled rest in C. Values tion, increases in [ K + ] were apparent by one
not c_ov_erefl___b_y_the same line are significantly different. minute of exercise (p<0.01). Further increases
(E) 0 1 5 10 30 (C) ()-]-5-1() 30 10X (p < 0.05) were seen at 30 min. In the C condition,
[ K + ] remain unchanged during the controlled
rest but increased during the 10rain exercise
='~ 1.0
(p<0.01).
oZ o
O -1.0
_z
z~
-2.0- Discussion
< -3.0-
The purpose of the experiment was to examine
-4.0- the nature and rapidity of the plasma volume
; 1 ; 1'0 3; I()X changes associated with a protocol of either cycle
T,ME (m~n) exercise or controlled rest under identical posi-
Fig. 3. Differences between measured and expected (based on tional and ambient conditions. The manipulation
plasma volume changes) albumin concentrations during Exer- of changing from a supine to a seated position did
cise (E) and Control (C) conditions. 0, 1, 5, 10, 30 are min. not differ between the experimental conditions.
during both experimental conditions. 10X refers to the end
of exercise which followed the controlled rest in C. Values
The initial five minute positional change resulted
not c ov_er_ed__b_y_the same
. . . . . .line
.. are significantly different. in an average plasma volume loss of 4.8% for all
(E) 0 1 5 10 30 (C) 0 1 5 10 30 10X subjects which is in agreement with other work
454
(Diaz et al. 1979; Hagan et al. 1978, 1980). The
reduction in plasma volume with change in posi-
tion was probably due to an increase in venous
hydrostatic pressure in the legs (Pollack and
Wood 1949).
With the~ onset of exercise in the E condition,
the rate of plasma volume loss increased (Fig. 1).
Approximately 76 ml of plasma was lost during
the first minute of cycle exercise. Plasma volume
eventually stabilized by the fifth minute of exer-
cise as the decrease in fluid totalled 208 ml. In
contrast, during the C condition, plasma volume
recuction after the 0 sample did not occur until
10 min into rest and amounted to a loss of only
176 ml after 30 min. The 10 min of exercise after
the rest period caused a further reduction in
plasma volume of 137 ml. It has been previously
reported that plasma volume loss appears to be
related to the intensity of exercise (Costill et al.
1974) and that the amount lost during the exercise
bout is affected by pre-exercise position (Diaz et
al. 1979; Hagan et al. 1980). Although the exercise
intensity in both conditions was identical, the
plasma volume reduction in the E condition was
greater than that during exercise in C. Apparent-
ly, the time spent in the preexercise position did
affect exercise hemoconcentration. However,
since plasma volume loss occurred during exer-
cise in both conditions, it would appear that the
compartment into which fluid is sequestered dur-
ing rest does not include the volume made availa-
ble after exercise begins.
It is possible that any fluid leaving the vascu-
lar volume of the legs might be entering both the
muscular and extramuscular spaces depending on
the activity (or lack thereof) at the time of the
hemoconcentration. Hargens et al. (1977) studied
the muscular compartments of the lower leg and
showed a large increase in intra-compartmental
pressure due to isometric contractions. This
points toward a limited compliance of the muscle
fiber connective tissue. Pollack and Wood (1949)
have shown that saphenous vein pressure reaches
a steady state very quickly upon positional
change. If it takes little time and fluid to saturate
an inactive muscle compartment, the large
amount of plasma which leaves the vascular vol-
ume when changing from supine to an upright
posture would travel the path of least resistance
into the extramuscular spaces of the legs. Wells et
al. (1938) found a minimal increase in calf pres-
sure even after 100 min of standing. The addi-
tional fact that leg volume continued to increase
for some two and one half hours while the subcu-"
taneous pressure remained constant indicated an
J.M. Pivarniket al.: Exercise,position, and plasma volume
extremely compliant extramuscular tissue com-
partment.
While a number of workers have found and
commented upon plasma volume reductions after
5--10 min of cycle exercise (Senay and Pivarnik
1985), our one minute values show for the first
time just how rapidly the plasma volume changes
after onset of exercise. It seems logical that the
initial muscle contractions were accomplished in
near-anaerobic conditions, thus setting up an os-
motic gradient which, in turn, was present when
the resistance vessels dilated within the exercising
muscle. The osmotic gradient, coupled with an in-
crease in capillary hydrostatic pressure (i. e., from
closed to open), caused the rapid outflow of fluid
from the systemic circulation (Smith et al. 1976).
The plasma would be maintained at the reduced
volume when the Starling forces reach a new
equilibrium.
In addition to plasma fluid volume, changes in
[PP] and [Alb] were examined. The difference be-
tween measured and expected concentrations as
predicted by plasma volume shifts are shown in
Figs. 2 and 3. The significant negative values in
both total protein and albumin by five minutes of
exercise in the C condition denote a loss of pro-
tein from the circulation. Some investigators have
found similar protein losses (Harrison et al. 1975,
1981 ; McMurray 1983) while others have not (Ek-
blom 1970; Melin et al. 1980). In contrast, de-
creases in total protein and albumin did not ap-
pear until 30 min of rest in the C condition. Addi-
tional losses occurred by the end of the 10 minute
exercise in C. Harrison et al. (1983) found de-
creases in protein after subjects rested for 30 min
in a sitting position. Diaz et al. (1979) found
losses of protein in subjects who rested in either
of two sitting positions but not in subjects who
rested supine. It is possible that posture may ef-
fect plasma protein in addition to fluid dynamics.
Changes in total protein were proportional to
those of albumin. Thus, the albumin to globulin
ratio was maintained at all times in both experi-
mental conditions.
The possibility exists than any apparent
changes in protein or albumin content were
simply an artifact. This would be true if the F-cell
ratio changed from rest to exercise, that is, an al-
teration in the ratio of total body hematocrit to
the peripheral hematocrit measured experimental-
ly. The issue has been thoroughly addressed by
Harrison (1985) in a recent review and it was felt
that this potentially confounding problem does
not occur. Costill and Saltin (1974) provide sup-
port for a constant F-cell value as they found no
J. M. Pivarnik et al.: Exercise, position, and plasma volume
changes in the whole body to venous hematocrit
ratio after subjects lost 4% of their body weights
by either thermal or exercise dehydration. In a
protocol similar to the present investigation, Har-
rison et al. (1975) used radioiodinated serum al-
bumin and found that protein left the vascular
volume during the first few minutes of cycle exer-
cise performed at 55% of max. I7o2 in an ambient
temperature of 30 ~ C.
Figure 4 presents the plasma concentrations of
various electrolytes during both the E and C con-
ditions. There were no statistically significant
changes in [Na + ] or [CI - ] in either of the experi-
mental conditions. Since these two ions account
for most of the osmotically active particles in the
plasma, the fluid that left the vascular spaces was
approximately isoosmotic in nature. A significant
increase in [ K + ] occurred in the E condition by
one minute of exercise. In the C condition, an in-
crease in [ K + ] did not occur during rest but only
after the 10 min exercise session. Kilburn (1966)
reported that the increase in [ K + ] was due to a
release of potassium from the active muscles. It
was postulated that cellular osmolality rises from
the anaerobiosis of exercise so water would in
turn diffuse into the muscle cells and possibly
into the surrounding interstitium. This clearly
supports the idea of vascular fluid moving into
the intramuscular compartment during exercise
hemoconcentration.
When comparing cycling and resting subjects
at the same times under identical conditions, the
vascular fluid shifts are both qualitatively and
quantitatively different. From the results pre-
sented here and past evidence, we propose a two
compartment model for vascular fluid displace-
ment due to postural and exercise effects. In the
present study, the vascular fluid that was lost dur-
ing the initial positional change in both experi-
mental conditions was most likely sequestered in
the subcutaneous space of the legs. Very little
fluid would have entered the intramuscular com-
partment since only a limited number of muscle
capillaries would have been open in the resting
state. In the E condition, a further loss of plasma
immediately occurred with the onset of exercise.
This fluid most likely entered the muscular com-
partment and continued to do so for a few min-
utes until the osmotic and capillary filtration pres-
sures in the muscle were sufficiently opposed by
the tissue pressure and a new equilibrium was
reached. In the C condition, the fluid that contin-
uously left the vascular volume during the seated
rest remained in the extramuscular spaces. When
the 10 min exercise began at the end of the rest
455
period, additional fluid was lost to the muscular
compartment just as in the E condition.
The constructive criticism of Dr. C. J. Gaebelein was sin-
cerely appreciated.
References
Beaumont van WV, Underkofler S, van Beaumont S (1981)
Erythrocyte volume, plasma volume, and acid-base
changes in exercise and heat dehydration. J Appl Physiol
50:1255--1262
Costill DL, Branam L, Eddy D, Fink W (1974) Alterations in
red cell volume following exercise and dehydration. J Appl
Physiol 37:912--916
Costill DL, Saltin B (1974) Changes in the ratio of venous to
body hematocrit following dehydration. J Appl Physiol
36:608--610
Dahms TE, Horvath SM (1974) Rapid, accurate technique for
determination of carbon monoxide in blood. Clin Chem
20:533--537
Diaz FJ, Bransford DR, Kobayashi K, Horvath SM, McMur-
ray RG (1979) Plasma volume changes during rest and
exercise in different postures in a hot humid environment.
J Appl Physiol 47:798--803
Dill DB, Costill DL (1974) Calculation of percentage changes
in volume of blood, plasma, and red cells in dehydration.
J Appl Physiol 37:247--248
Ekblom B (1970) Effect of physical training on circulation
during prolonged severe exercise. Acta Physiol Scand
78:145--158
Hagan RD, Diaz FJ, Horvath SM (1978) Plasma volume
changes with movement to supine and standing positions.
J Appl Physiol 45:414--418
Hagan RD, Diaz FJ, McMurray RG, Horvath SM (1980)
Plasma volume changes related to posture and exercise.
Proc Soc Exp Biol Med 165:155--160
Hargens AR, Mubarak SJ, Owens CA, Garetto LP, Aheson
WH (1977) Interstitial fluid pressure in muscle and com-
partment syndrome in man. Microvasc Res 14:1--10
Harrison MH (1985) Effects of thermal stress and exercise on
blood volume in humans. Physiol Rev 65:149--209
Harrison MH, Edwards RJ, Cochrane LA, Graveney MJ
(1983) Blood volume and protein responses to skin heating
and cooling in resting subjects. J Appl Physiol 54:515--523
Harrison MH, Edwards RJ, Graveney MJ, Cochrane LA, Da-
vies JA (1981) Blood volume and plasma protein responses
to heat acclimation in humans. J Appl Physiol 50:597--604
Harrison MH, Edwards RJ, Leitch DR (1975) Effect of exer-
cise and thermal stress on plasma volume. J Appl Physiol
39:925--931
Kilburn KH (1966) Muscular origin of elevated plasma potas-
sium during exercise. J Appl Physiol 21:675--678
Melin B, Eclache JP, Geelen G, Annat G, Allevard AM, Jar-
saillon E, Zebidi A, Legros JJ, Gharib C (1980) Plasma
AVP, neurophysin, renin activity, and aldosterone during
submaximal exercise performed until exhaustion in trained
and untrained men. Eur J Appl Physiol 44:141--151
McMurray RG (1983) Plasma volume changes during sub-
maximal swimming. Eur J Appl Physiol 51:347--356
Pollack AA, Wood EH (1949) Venous pressure in the saphe-
nous vein at the ankle in man during exercise and changes
in posture. J Appl Physiol 1:649--662
456
Senay LC Jr, Pivarnik JM (1985) Fluid shifts during exercise.
In: Terjung RL (ed) Exercise and sport sciences reviews.
Macmillan, New York, pp 335--387
Smith EE, Guyton AC, Manning RD, White ILl (1976) Inte-
grated mechanisms of cardiovascular response and control
during exercise in the normal human. Prog Cardiovasc Dis
18:421--443
Thompson WO, Thompson PK, Dailey ME (1928) The effect
of posture upon the composition and volume of the blood
in man. J Clin Invest 5:573--604
Waterfield RL (1931a) The effect of posture on the circulating
blood volume. J Physiol (Lond) 72:100--120
J.M. Pivarnik et al.: Exercise, position, and plasma volume
Waterfield RL (1931b) The effect of posture on the volume of
the leg. J Physiol (Lond) 72:121--131
Wells HS, Youmans JB, Miller DG, Jr (1983) Tissue pressure
(intracutaneous, subcutaneous, and intramuscular) as re-
lated to venous pressure, capillary filtration and other fac-
tors. J Clin Invest 17:489--499
Winer BJ (1971) Statistical principles in experimental design.
McGraw, New York, 2 ed
Accepted February 24, 1986