J Appl Physiol 99: 45–52, 2005.
First published March 3, 2005; doi:10.1152/japplphysiol.01289.2004.
Contribution of prostaglandins to the dilation that follows isometric forearm
contraction in human subjects: effects of aspirin and hyperoxia
Thet Su Win and Janice M. Marshall
Department of Physiology, Division of Medical Sciences, The Medical School, Birmingham, United Kingdom
Submitted 16 November 2004; accepted in final form 26 February 2005
the extent to which prostaglandins
(PGs) contribute to the hyperemia and vasodilation that are
associated with muscle contraction in human subjects. Consistent with a proposed dilator role, Nowak and Wennmalm (26)
showed that the venous efflux of PGE increased during dynamic leg exercise, whereas Wilson and Kapoor (38) showed
the venous efflux of 6-keto-PGF1a, the stable metabolite of
PGI2 (prostacyclin), and PGE2 increased during dynamic exercise of the wrist. More recently, others showed by using
microdialysis that muscle interstitial concentration of 6-ketoPGF1␣ (10) and PGE2 (2, 20) increased during dynamic leg
exercise.
The first functional evidence that PGs may contribute to
exercise hyperemia was provided by Kilbom and Wennmalm
(21), who reported that the cyclooxygenase (COX) inhibitor
indomethacin reduced the hyperemia that followed low-level
dynamic exercise of the forearm and that occurred during and
after isometric contraction of the forearm by ⬃50%. However,
these results are difficult to evaluate because recordings were
made before and after indomethacin in the same experimental
session with no reported time controls. The conclusion drawn
for isometric contraction is particularly unsafe because blood
flow was recorded by venous occlusion plethysmography,
which is unreliable when applied during isometric contraction
(see Ref. 38).
More recently, both indomethacin and aspirin substantially
reduced the increase in calf blood flow and decrease in vascular
resistance that occurred following a 30-min period of treadmill
exercise (4). In this study, a placebo or one of the COX
inhibitors was given on different experimental days in randomized order. However, these indomethacin also potentiated the
increase in systolic pressure that occurred during exercise, and
both COX inhibitors facilitated an increase in blood flow that
occurred in the nonexercised forearm. Thus interpretation of
the effects of COX inhibition on the local response to muscle
contraction was complicated by its effect on the systemic
cardiovascular response to exercise.
In a further study, infusion of indomethacin into the forearm
reduced the increase in forearm blood flow (FBF) and the
decrease in forearm vascular resistance (FVR) measured by
plethysmography in transient breaks during 5-min periods of
dynamic wrist exercise (38). However, indomethacin produced
a decrease in baseline FBF and an increase in baseline FVR
that were similar in magnitudes to the effects it had on FBF and
FVR values recorded during exercise. Thus there was no clear
evidence that indomethacin affected the changes in FBF or
FVR induced by exercise. The study of Duffy et al. (6) is more
conclusive in that infusion of aspirin into forearm reduced the
increase in FBF measured by plethysmography and the decrease in FVR that followed a 2-min period of dynamic wrist
exercise: they corrected for the change in baseline FBF and
FVR when calculating the effects on the postexercise hyperemia but acknowledged that the change in baseline per se may
have limited the hyperemia.
By contrast, Nowak and Wennmalm (27), who used plethysmography, reported that indomethacin had no effect on the
hyperemia that occurred in the calf during dynamic leg exercise. Moreover, Shoemaker et al. (34), who used the ultrasonic
Doppler technique, which can be used effectively during muscle contraction, showed that COX inhibition had no effect on
the increase in FBF recorded during dynamic wrist exercise.
To summarize, the one study so far performed on the role of
PGs in the hyperemia associated with isometric contraction
(21) produced inconclusive results, and the various studies
performed on dynamic exercise allow only the proposal that
Address for reprint requests and other correspondence: J. M. Marshall, Dept.
of Physiology, Division of Medical Sciences, The Medical School, Birmingham B15 2TT, UK (E-mail: [email protected]).
The costs of publication of this article were defrayed in part by the payment
of page charges. The article must therefore be hereby marked “advertisement”
in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
vasodilation; hyperemia; hypoxia; hyperoxia
THERE IS CONTROVERSY OVER
http://www. jap.org
8750-7587/05 $8.00 Copyright © 2005 the American Physiological Society
45
Downloaded from http://jap.physiology.org/ by 10.220.32.246 on July 28, 2017
Win, Thet Su, and Janice M. Marshall. Contribution of
prostaglandins to the dilation that follows isometric forearm contraction in human subjects: effects of aspirin and hyperoxia. J Appl
Physiol 99: 45–52, 2005. First published March 3, 2005;
doi:10.1152/japplphysiol.01289.2004.—In 11 healthy volunteers, we
evaluated, in a double-blind crossover study, whether the vasodilation
that follows isometric contraction is mediated by prostaglandins (PGs)
and/or is O2 dependent. Subjects performed isometric handgrip for 2
min at 60% maximal voluntary contraction (MVC), after pretreatment
with placebo or aspirin (600 mg orally), when breathing air or 40%
O2. Forearm blood flow was measured in the dominant forearm by
venous occlusion plethysmography. Arterial blood pressure was also
recorded, allowing calculation of forearm vascular conductance
(FVC; forearm blood flow/arterial blood pressure). During air breathing, aspirin significantly reduced the increase in FVC that followed
contraction at 60% MVC: from a baseline of 0.09 ⫾ 0.011 [mean ⫾
SE, conductance units (CU)], the peak value was reduced from 0.24 ⫾
0.03 to 0.14 ⫾ 0.01 CU. Breathing 40% O2 similarly reduced the
increase in FVC relative to that evoked when breathing air; the peak
value was 0.24 ⫾ 0.03 vs. 0.15 ⫾ 0.02 CU. However, after aspirin,
breathing 40% O2 had no further effect on the contraction-evoked
increase in FVC (the peak value was 0.15 ⫾ 0.02 vs. 0.16 ⫾ 0.02
CU). Thus the present study indicates that prostaglandins make a
substantial contribution to the peak of the vasodilation that follows
isometric contraction of forearm muscles at 60% MVC. Given that
hyperoxia similarly reduced the vasodilation and attenuated the effect
of aspirin, we propose that the stimulus for prostaglandin synthesis
and release is hypoxia of the endothelium.
46
PROSTAGLANDINS AND POSTCONTRACTION DILATION
METHODS
Subjects
Eleven healthy, recreationally active male subjects of age 21–28 yr
(21.45 ⫾ 0.8 yr, mean ⫾ SE) participated in the present study. After
full details of the experimental protocol, which was approved by the
University of Birmingham Ethics Review Committee, consent forms
were signed by the subjects. All subjects denied taking aspirin or any
other nonsteroidal anti-inflammatory drugs for 10 days before the
experiment. The subjects also refrained from smoking or drinking
alcohol within 24 h of the experiment and were asked not to eat a
heavy meal, consume caffeine, or take vigorous exercise for at least
2 h before the experiment. In practice, none of the subjects consumed
caffeine or undertook any exercise other than that required to walk to
the laboratory on the day of the experiment. Each subject attended a
temperature-controlled laboratory (18 –20°C) on six occasions, each
on a different day. The first visit served to familiarize the subjects with
the study protocol. They filled in a questionnaire concerning their age,
weight, height, exercise status, smoking and dietary habits, alcohol
intake, and history of any cardiovascular disease. In addition, the
MVC of each subject was determined.
MVC
Each subject was asked to perform a maximal strength hand grip
with his dominant hand, using a hand-grip dynamometer (TKK 5001,
J Appl Physiol • VOL
Takei, Japan) and was told before he began that he would have to
maintain the contraction for 10 s. Each subject was given strong
verbal encouragement during the contraction. The force reading at the
end of the 10 s was taken as 100% MVC: in each subject, it was no
more than 2% lower than the force measured in the first 1 s of the
contraction. The 100% MVC was measured in this way to avoid the
subject using muscle in addition to those of the forearm to generate an
unrealistically large “snap” contraction.
Forearm Exercise
During the protocols (see below), subjects performed an isometric
contraction of forearm at 60% MVC with the dominant hand for 2
min. The force reading was displayed by the deflection of an arrow on
the dynamometer so that the force exerted was accurately monitored
by the subject and checked by the experimenter.
Recordings
FBF was measured from the dominant forearm, which was arranged on foam cushions at heart level, by venous occlusion plethysmography (37) by using a dual strain-gauge plethysmograph (Lectromed, Letchworth Garden City, UK). The strain gauge was connected via a Wheatstone bridge circuit and amplifier to an Apple
Macintosh computer (Power Macintosh 7100/66, Apple Computer).
The output of the strain gauge was displayed on the computer screen
using MacLab hardware (MacLab/400; AD Instruments, Hastings,
UK). For each recording of FBF, the wrist cuff was inflated to ⬃200
mmHg, and then, when the output of the strain gauge was stable, the
arm cuff was inflated to ⬃40 mmHg. The pressure in the arm cuff was
maintained for ⬃4 –5 s and that in the wrist cuff for no more than 10 s
for each recording of FBF. Throughout the experiment, a continuous
recording of cutaneous red cell flux (cRCF) of dominant forearm was
made by using the laser Doppler perfusion monitor (DRT-4, Moor
Instruments, Axminster, UK), which was connected to a probe (PF308). The probe was positioned just distal to the plethysmograph at
the medial, hair-free aspect of the half-pronated forearm. The laser
Doppler flowmeter was also connected to the computer. Measurements of cRCF for analysis were taken immediately before each
measurement of FBF since inflating the arm cuff affected forearm
cRCF.
Arterial blood pressure (ABP) was measured from the nondominant arm, using a semiautomatic sphygmomanometer (Omro, M4,
Omron Healthcare, Hamburg, Germany) that also provided a measure
of heart rate. Mean ABP (MABP) was calculated as one-third the
pulse pressure plus diastolic pressure. Vascular conductance in the
forearm and cutaneous circulations (FVC and CVC, respectively)
were calculated as FBF or cRCF divided by MABP.
Delivery of Hyperoxic Gas or Atmospheric Air
Delivery of hyperoxic gas was achieved by the use of an O2
concentrator (O2 Live Machine, Oxygen Leisure Products, London,
UK) that delivered atmospheric air concentrated to 95–100% O2. An
intersurgical face mask (Intersurgical Wokingham) was fastened on
the subject, with the mask covering his nose and mouth. The mask
was connected via a Venturi valve (Intersurgical Wokingham) and
tubing to the O2 concentrator. This allowed 95–100% of O2 to be
forced at 2.5 l/min through the Venturi valve, which entrained air
from the atmosphere and reduced the final concentration to ⬃40% O2.
The gas delivered to the subject was within the range of 39 to 41% O2.
The O2 concentration of the gas delivered by the O2 concentrator and
by the mask was checked at regular intervals by passing a sample
through a gas analyzer (IL 1640). In other experiments, delivery of
atmospheric air through the mask was achieved by the use of a
Reciprocator Pump (Stanhope Seta), which was connected via tubing
to the mask via the Venturi valve. This provided a control for the
experience of breathing via a mask through which gas was delivered.
99 • JULY 2005 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.32.246 on July 28, 2017
PGs contribute to postexercise hyperemia. It has also been
reported that PGs contribute to reactive hyperemia, i.e., the
dilation that follows arterial occlusion: reactive hyperemia has
been attributed partly to a myogenic response and partly to the
vasodilator metabolites that accumulate during the period of
occlusion (8, 33).
We reasoned that, if PGs do contribute to exercise hyperemia, they are most likely to do so when there is relative
hypoxia, because O2 demand exceeds O2 supply and/or an
increase in endothelial shear stress. Hypoxia increased phospholipase A2 activity, the first step in PG synthesis, and
increased PGI2 production in cultured endothelial cells (25),
and COX inhibition attenuated hypoxia-induced dilation of
muscle arterioles and small arteries in vitro (3, 12, 24) and in
vivo (31). Moreover, shear stress stimulated PG synthesis from
cultured endothelial cells (11, 14), and COX inhibition attenuated shear stress-induced dilation of muscle arterioles in vivo
and in vitro (22, 23). We, therefore, hypothesized that PGs
make a substantial contribution to the hyperemia that follows a
period of isometric contraction, given that PO2 falls in muscle
during contraction (13) and that the increase in blood velocity
that occurs when muscle contraction ceases (19) is likely to
increase endothelial shear stress.
To test this hypothesis, we investigated the effect of aspirin
on the hyperemia that followed isometric contraction of the
forearm at 60% maximum voluntary contraction (MVC) by
using a double-blind crossover study design. To make an
indirect test of the role of hypoxia in this hyperemia, we used
a comparable study design in which subjects were given
supplementary O2 (40% O2) to breathe with the aim of alleviating the fall in PO2 that occurs during contraction. In a final
protocol, we tested the combined effect of aspirin and breathing 40% O2 on hyperemia. Our results allow the novel proposals that PGs are a main contributor to the hyperemia that
follows isometric contraction and that this contribution is O2
dependent.
47
PROSTAGLANDINS AND POSTCONTRACTION DILATION
The equipment was screened so that the subject could not see which
gas he was breathing.
Drug Administration
Aspirin (Aspro Clear, Laboratories Roche Nicholas) was given
orally (600 mg): this dose produces maximal inhibition of the synthesis of PGs and thromboxane A2 after 30 min, and PGI2 synthesis
is still inhibited by 70% after 90 min (17). To disguise the taste and
appearance of aspirin, it was dissolved in 450 ml of diluted orange
juice (200 ml of Sainsbury’s Pure Orange Juice diluted with 250 ml of
water). For placebo intervention, 450 ml of diluted orange juice was
given. Placebo or aspirin was given in a double-blind randomized
fashion.
Protocols
Statistical Analysis
Data are expressed as means ⫾ SE. Each subject served as his own
control. Repeated-measures ANOVA was used to compare values
recorded at different times after isometric contraction with those
recorded at rest within one condition. Factorial ANOVA was used to
detect differences between conditions. If a difference was found at
P ⬍ 0.05, Bonferroni/Dunnett’s post hoc tests were performed to
RESULTS
Protocol 1
The weight and height of the subjects were 71.09 ⫾ 1.8 kg
and 1.71 ⫾ 0.034 m, respectively. The forearm circumference
was 24 –29 cm (25.73 ⫾ 0.52 cm). The baseline values for
MABP, heart rate, FBF, and cRCF after placebo or aspirin are
shown in Table 1. There were no significant differences between the baselines after placebo or aspirin treatment.
After placebo treatment, isometric contraction produced the
expected increase in MABP: it reached a value that was
different from baseline immediately after the release of contraction and then swiftly returned toward baseline (Fig. 1).
There was also an increase in FBF, this being significantly
different from baseline at time 0 and throughout the 3-min
recovery period. Accordingly, FVC was significantly increased, indicating vasodilation, at all time points during the
recovery period (Fig. 1). There was no significant increase in
cRCF. Furthermore, CVC did not change significantly from the
baseline values (Fig. 1). In fact, cRCF and CVC showed no
significant change after contraction in any of the protocols.
Moreover, baseline rCRF and CVC were not affected by
aspirin or by breathing 40% O2 or air via the mask. Thus these
variables are not given any further consideration below.
After aspirin treatment, isometric contraction again increased MABP, with the values being comparable to those
recorded after placebo treatment (Table 2). However, after
aspirin treatment, the increase in FBF from baseline was
significant only at time 0 and 0.5 min after the isometric
contraction, and by ANOVA, FBF was lower after aspirin than
placebo treatment over the full 3-min recovery period. The
increase in FVC was also substantially reduced after aspirin
treatment over the whole time period after contraction, with
this effect being particularly pronounced at time 0 (Fig. 2A).
Because changes in the effect of isometric contraction on FVC
demonstrate changes in the effect on vascular tone, we have
concentrated on FVC when describing the results in the text
below and in Figs. 1 and 2. The change in MABP evoked by
isometric contraction was not altered by any of the experimental conditions. Thus the changes evoked in FBF reflected the
changes in FVC (see Table 2).
Table 1. Baseline values after placebo and after aspirin when subjects breathed air and when they
breathed 40% O2 with or without a mask
Placebo
MABP, mmHg
HR, beats/min
FBF, ml䡠100 ml tissue⫺1䡠min⫺1
cRCF, Pu
Aspirin
Air
40% O2
Air
40% O2
Air
Without mask
With mask
Without mask
With mask
With mask
79.6⫾1.6
64.2⫾2.0
6.8⫾0.6
10.3⫾1.3
80.1⫾2.3
64.7⫾2.0
5.6⫾0.6
9.4⫾0.7
82.5⫾1.8
64.2⫾1.7
5.4⫾0.5
10.7⫾1.6
80.3⫾2.1
64.6⫾2.3
6.0⫾0.5
9.9⫾0.9
81.0⫾2.7
63.6⫾1.1
6.5⫾1.2
10.4⫾0.9
Values are means ⫾ SE. Pu, perfusion units; MABP, mean arterial blood pressure; HR, heart rate; FBF, forearm blood flow; cRCF, cutaneous red cell flux.
J Appl Physiol • VOL
99 • JULY 2005 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.32.246 on July 28, 2017
Protocol 1. On the second visit, when the subject was seated
comfortably in the laboratory, he was given either placebo or aspirin.
The subject then rested for at least 35 min, and during this period he
was instrumented to record FBF, ABP, and cRCF (see above). During
the last 5 min of the resting period, FBF and ABP were measured four
times to ensure that a stable baseline had been achieved. The subject
then performed an isometric contraction at 60% MVC with his
dominant hand for 2 min. Immediately after the contraction, FBF and
ABP were measured (time 0), and these measurements were repeated
at 30 s, 1 min, 2 min, and 3 min after the contraction: to achieve the
FBF measurements at these times, the cuff on the wrist was inflated
3– 4 s before the end of the contraction and before the 1-, 2-, and
3-min readings (see above). On the third visit, the subject received the
alternative intervention (i.e., either aspirin or placebo). The protocol
described above was then repeated.
Protocol 2. This protocol was conducted on the fourth and fifth
visits. It was similar to protocol 1, with the subjects being given
aspirin or placebo at the beginning of each session, except that from
the end of the 35-min rest period (see above), the subject breathed
40% O2 for 5 min before, during, and after the 2-min period of
contraction at 60% MVC.
Protocol 3. In this protocol, which was undertaken on the sixth
visit, the subject was given aspirin on arrival at the laboratory. Then,
the first day’s session of protocol 2 was performed, but with the
subject breathing atmospheric air via the intersurgical mask instead of
hyperoxic gas. The subject was unaware of which gas he was
breathing and did not know whether he had taken aspirin. This
protocol was not performed at a second session.
locate the differences, with significance again being set at P ⬍ 0.05.
Data were also analyzed as described above using change from
baseline rather than absolute values. The outcome of these analyses
were the same, and thus we have not considered them separately in the
description below.
48
PROSTAGLANDINS AND POSTCONTRACTION DILATION
The changes in FVC seen after placebo and while breathing
40% O2 were comparable to those seen after aspirin treatment
when breathing 40% O2 (Fig. 2C). Thus, by contrast to the
results obtained when subjects breathed air, when they
breathed 40% O2 there were no differences between the
changes evoked in FVC after aspirin and after placebo (Fig.
2C, cf. Fig. 2A).
Protocol 3
There were no differences between the baseline values
recorded when breathing air without a mask after aspirin
treatment (protocol 1) and those recorded when breathing air
via the mask after aspirin (Table 1). Furthermore, the values of
FVC recorded after isometric contraction when the subjects
breathed air via the mask after aspirin were fully comparable to
those recorded when they breathed air without a mask after
aspirin treatment in protocol 1 (Fig. 2D).
In the present double-blind crossover study, the hyperemia
and vasodilation that occurred in forearm after a 2-min period
of isometric contraction at 60% MVC was substantially reduced by the COX inhibitor aspirin, and a similar reduction
occurred when the subjects breathed 40% O2 rather than air.
By contrast, in the presence of aspirin, breathing 40% O2 had
no effect on the postcontraction vasodilation.
The pattern of response evoked by 60% MVC was as
expected from many previous studies. There was an increase in
MABP that waned over the 3 min after the end of the contraction, and this was associated with an increase in FBF that
waned from the cessation of contraction but was still significantly above control level at 3 min. FVC showed an increase of
⬃3.5-fold at the end of contraction that persisted at ⬃3-fold
above control at 3 min, indicating maintained vasodilation in
the forearm. Because cRCF in the forearm and CVC showed
no significant change, we can assume, as others have done (see
Refs. 18, 35), that the increase in FVC primarily reflected
postcontraction vasodilation in skeletal muscle.
Fig. 1. Responses evoked by isometric contraction after placebo in protocol 1.
Values are means ⫾ SE at rest (R) and at intervals after contraction; the period
of contraction is shown as a solid bar. FBF, forearm blood flow; MABP, mean
arterial blood pressure; FVC, forearm vascular conductance; cRCF, cutaneous
red cell flux; CVC, cutaneous vascular conductance; Pu, perfusion units. The
changes evoked over time in FBF, MABP, and FVC, but not cRCF or CVC,
were significant: P ⬍ 0.05 by ANOVA. *Significant increase from R by post
hoc test.
Protocol 2
There were no significant differences between the baseline
values for MABP, FBF, and FVC after placebo and aspirin
treatment when the subjects breathed 40% O2 (Table 1 and
Fig. 2).
After placebo treatment and while 40% O2 was breathed,
FVC was increased from baseline at similar time points after
isometric contraction as in subjects who breathed air without a
mask after placebo treatment in protocol 1 (see Fig. 2B).
However, FVC was lower over the whole time period when
subjects breathed 40% O2 than when they breathed air; post
hoc tests suggesting this effect were particularly pronounced
immediately after control at time 0 (Fig. 2B).
J Appl Physiol • VOL
Effects of Aspirin
Aspirin had no effect on the baseline levels of any of the
recorded variables, including FBF and FVC. This is consistent
with the findings of other studies in which a COX inhibitor
(aspirin, indomethacin, or ibuprofen) was given systemically
(4, 21, 34). By contrast, when indomethacin or aspirin was
directly infused into the forearm circulation, baseline FVC
decreased and venous efflux of PGI2 and PGE2 decreased,
indicating a tonic dilator influence of PGs (6, 38). The reason
for this disparity is not clear: it may be that the local effect of
PG inhibition that is evident after infusion of a COX inhibitor
into the forearm (6, 38) is obscured by other systemic and
compensatory effects when recordings are made after a single
systemic dose of the antagonist or when repeated doses have
been given for 24 h (Refs. 4, 21, 34, and present study).
The fact that aspirin had no effect on baseline FBF or FVC
in the present study makes the interpretation of our results
straightforward (cf. Refs. 6, 38). The dose of aspirin we used
produces near-maximal inhibition of COX for 0.5–1.5 h (17).
Thus our finding that aspirin attenuated the whole period of
postcontraction hyperemia and vasodilation (increase in FVC),
99 • JULY 2005 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.32.246 on July 28, 2017
DISCUSSION
49
PROSTAGLANDINS AND POSTCONTRACTION DILATION
Table 2. FBF and MABP values recorded at rest and at intervals after isometric contraction under different conditions
Time After Contraction, min
Rest
0
0.5
⫺1
FBF, ml䡠100 ml tissue
Placebo
Aspirin
Placebo ⫹ 40% O2
Aspirin ⫹ 40% O2
Aspirin ⫹ air with mask
6.8⫾0.6
5.4⫾0.5
7.0⫾0.6
6.0⫾0.5
6.5⫾1.2
20.7⫾3.0*
13.1⫾1.1*†
16.5⫾1.2*†
13.6⫾1.4
12.6⫾1.2*
1
2
3
16.2⫾2.9*
11.6⫾0.8*
13.8⫾1.3*
11.4⫾1.4*
11.6⫾1.3*
14.4⫾2.1*
10.4⫾1.0
12.7⫾1.4*
11.0⫾1.4*
11.1⫾1.3*
15.1⫾2.7*
10.9⫾1.0
11.4⫾1.3
10.0⫾1.4
10.2⫾1.3
81.5⫾2.2
81.6⫾1.9
82.3⫾2.5
81.2⫾2.3
80.7⫾2.8
80.2⫾2.1
81.6⫾1.9
82.5⫾3.0
80.0⫾2.6
81.3⫾3.2
81.5⫾1.9
80.8⫾2.0
78.9⫾2.2
79.0⫾2.5
79.7⫾3.2
䡠min
⫺1
16.6⫾2.6*
11.8⫾0.9*
14.4⫾1.3*
11.9⫾1.4*
12.1⫾1.2*
MABP, mmHg
Placebo
Aspirin
Placebo ⫹ 40% O2
Aspirin ⫹ 40% O2
Aspirin ⫹ air with mask
79.6⫾1.6
82.5⫾1.8
80.1⫾2.3
80.3⫾2.1
81.0⫾2.7
87.2⫾2.3*
92.3⫾1.6*
90.2⫾2.3*
88.2⫾2.4*
89.8⫾2.3*
82.5⫾2.0
82.6⫾1.6
83.7⫾2.7
82.0⫾2.7
82.8⫾3.2
with a particularly pronounced effect on the peak hyperemia
and peak increase in FVC, which were reduced by ⬃35%,
strongly suggests that PGs make a major contribution to this
vasodilation, especially to the early part of the response. As far
as we are aware, the only previous study on the effect of a
COX inhibitor on the hyperemia that follows isometric contraction is that of Kilbom and Wennmalm (21). Our results are
consistent with their findings. However, we suggest that our
experimental design was better, and thus our conclusion is
more secure. The particularly pronounced effect of aspirin on
the peak vasodilation immediately after contraction is reminiscent of the finding that COX inhibition decreased the peak
reactive hyperemia recorded in the forearm after a period of
arterial occlusion (8). Given that the peak of reactive hyperemia is attributed at least in part to mediators that accumulate
during the period of ischemia (33), our finding is consistent
with the proposal that PGs accumulate in muscle when PO2
falls as a consequence of the increase in O2 demand caused by
muscle contraction (see Combined Effects of Aspirin and Hyperoxia).
Effects of Hyperoxia
It is reasonable to assume that when 40% O2 is breathed, the
PO2 of the arterial blood increased to ⬃240 Torr from ⬃ 98
Torr (see Ref. 7). There must have been a consequent increase
in the dissolved O2 in plasma and a small increase in the
arterial saturation of hemoglobin from ⬃98 to ⬃100%. In
other words, we must have produced a state of hyperoxia. This
condition produced no change in baseline FBF or FVC. In
previous studies, subjects breathing 100 or 200% O2 showed a
decrease in baseline forearm or calf blood flow (1, 16). However, consistent with our finding, Pedersen et al. (28) showed
no change in baseline FBF when subjects breathed 60% O2.
Thus it seems that inspired O2 concentrations of ⬎60% may be
required to reveal the recognized vasoconstrictor effect of a
rise in PO2 (29). Given that baseline FBF did not change in our
study, we can conclude that, under resting conditions, any
increase in O2 delivery to the forearm was small and mainly the
consequence of an increase in the O2 content of the plasma.
J Appl Physiol • VOL
The second new finding of the present study was that the
postcontraction hyperemia and increase in FVC were reduced
when breathing 40% O2, with the peak increase in FVC being
reduced by ⬃35%. The experiments in which comparisons
were made between responses evoked during air breathing with
and without the mask (see Protocol 3, RESULTS) indicated that
the mask per se did not affect the response. Therefore, the
reduction in the vasodilation was due to breathing 40% O2
rather than to the mask. Others have shown that breathing 100
or 60% O2 reduced the increase in leg blood flow that occurred
during dynamic leg exercise (28, 36). As far as we are aware,
our study is the first to show that hyperoxia can reduce the
vasodilation that follows isometric contraction of the forearm.
The most obvious explanation for this finding is that breathing
40% O2 increased the O2 supply during the period of contraction and thereby reduced the accumulation of vasodilator
substances that caused vasodilation when contraction ceased.
This proposal obviously requires that blood flow through the
forearm continued during isometric contraction at 60% MVC.
The most comprehensive study on blood flow in the forearm
during isometric contraction was performed by Humphreys and
Lind (18), who measured FBF by venous occlusion plethysmography. They argued, from the effects of arterial occlusion
on time to physical exhaustion on forearm contractions at
different percentages of MVC, from measurements of intramuscular temperature in active and inactive muscles of the
forearm when the forearm had been immersed in warm or cool
water before contraction and from measurements of FBF made
during isometric contraction, that FBF increases above control
during isometric contractions to ⬎60% MVC. Their results
indicated that the increase in blood flow was through the
contracting muscles and not through inactive muscles or skin.
Moreover, their findings suggested that ⬎70% MVC is required to produce total occlusion. Consistent with these conclusions, more recent ultrasonic Doppler recordings from the
brachial artery indicated that FBF increased to similar levels,
by 85–95% from control, during isometric forearm contraction
at all intensities from 10 to 70% MVC: higher percent MVC
values were not tested (19).
99 • JULY 2005 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.32.246 on July 28, 2017
Values are means ⫾ SE at times corresponding to the FVC values shown in Figs. 1 and 2. Within each condition, the change in FVC and MABP over time
was significant (P ⬍ 0.05) by ANOVA. *Significant increase from resting value within condition. Values obtained after aspirin and after placebo ⫹ 40% O2
were significantly different by ANOVA from those obtained after placebo alone. †Significant difference between one of these conditions and placebo at a given
time point, by post hoc test. There was no significant difference by ANOVA between placebo ⫹ 40% O2 and aspirin ⫹ 40% O2 or between aspirin alone (with
mask) and aspirin ⫹ air with mask; see text and cf. Fig. 2.
50
PROSTAGLANDINS AND POSTCONTRACTION DILATION
than by increasing the bulk delivery of O2 and that the
increase in PO2 reduced the accumulation of a vasodilator
substance(s) whose formation is O2 dependent. This accords
with our recent finding that, when 40% O2 was breathed
only during the period of isometric contraction, the postcontraction increase in FBF was 18.64 ⫾ 1.95 ml 䡠 100 ml
tissue⫺1 䡠 min⫺1 at peak rather than 24.11 ⫾ 2.13 ml 䡠 100 ml
tissue⫺1 䡠 min⫺1 (Ref. 9; Fordy G and Marshall JM, unpublished observations). It is also consistent with evidence that
PO2 is an important signal for the regulation of skeletal
muscle blood flow (29).
Combined Effects of Aspirin and Hyperoxia
Thus it is very unlikely that FBF was fully occluded in the
present study during isometric contraction at 60% MVC; indeed, blood flow to the contracting muscles probably increased
to some extent. This proposal is consistent with the fact that the
circumference of the forearm was relatively small for male
subjects (see RESULTS), reflecting their lack of regular participation in sport involving the use of forearm muscles and by the
fact that they were able to sustain contraction at 60% MVC for
as long as 2 min: muscle mass determines the intramuscular
pressure developed at a given percent MVC and, in turn, time
to exhaustion.
Thus we can conclude that breathing 40% O2 produced its
effect on postcontraction dilation by increasing the PO2 of
the blood supplied to the muscle during contraction rather
J Appl Physiol • VOL
99 • JULY 2005 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.32.246 on July 28, 2017
Fig. 2. FVC recorded at R and at intervals after isometric contraction under
different conditions. A: placebo vs. aspirin treatment. B: placebo vs. placebo ⫹
40% O2 treatment. C: placebo ⫹ 40% O2 vs. aspirin ⫹ 40% O2 treatment. D:
aspirin ⫹ air without mask vs. aspirin ⫹air with mask treatment. Values are
means ⫾ SE at R and at intervals after contraction as in Fig. 1. Within each
condition, the change in FVC over time was significant by ANOVA. *Significant increase from R within condition (P ⬍ 0.05). §Significant difference
between conditions by ANOVA (P ⬍ 0.05). †Significant difference between
the 2 experimental conditions by post hoc test (P ⬍ 0.05).
Our third new finding was that, in the presence of COX
blockade with aspirin, hyperoxia had no further effect on the
postcontraction vasodilation, not even on the peak of the
response. This argues against endothelial shear stress being
a major stimulus for PG synthesis. Indeed, the more likely
explanation for this finding is that hyperoxia and aspirin
acted on the same dilator pathway(s). In other words,
hyperoxia prevented the accumulation of vasodilator PGs
that contributed to the postcontraction vasodilation. This
interpretation is fully consistent with the substantial evidence that vasodilator PGs are released from the endothelium of muscle arterioles in response to hypoxia, a fall in
PO2, and cause arteriolar dilation (3, 12, 24, 31). It is also
consistent with evidence that, during muscle contraction, the
decrease in venular PO2 causes the release of PGs from
venular endothelial cells, possibly via the action of ATP
released from red blood cells, and that these PGs then cause
the dilation of nearby arterioles (15, 31). If PGs are released
into the interstitial fluid during isometric contraction as they
are in dynamic exercise (see Refs. 2, 10, 20) and if this is
due to a relative hypoxia of the muscle fibers, it seems very
unlikely that this was counteracted by the small increase in
O2 delivery to muscle produced by breathing 40% O2. In
other words, supplementary O2 breathing is much more
likely to have affected PG production by endothelium than
by muscle fibers.
The pathways that mediate vasodilation are known to be
interdependent. For example, inhibition of the cAMP pathway
that mediates the dilation to PGs can inhibit dilation via cGMP,
the second messenger for nitric oxide (NO) (5). Furthermore,
we recently showed that, when arterial PO2 is reduced during
systemic hypoxia, adenosine acts on endothelial A1 receptors
to cause a large part of the hypoxia-induced dilation by
stimulating the synthesis of PGI2, which then increases NO
synthesis (30). Thus it would not be surprising if blockade of
COX with aspirin or hyperoxia, either separately or combined,
attenuated not only the component of the postcontraction
dilation that is mediated by PGs but also that mediated by
adenosine and NO. Consistent with this idea, the forearm
dilation that followed dynamic exercise of the wrist was
reduced to similar extents when aspirin or a NO synthesis
inhibitor was administered separately or together (6).
However, it must be acknowledged that other studies indicate that there is redundancy among the mechanisms that cause
dilation during and after muscle contraction. For example, a
COX inhibitor given during dynamic exercise of the forearm
produced only transient inhibition of the hyperemia (32),
PROSTAGLANDINS AND POSTCONTRACTION DILATION
REFERENCES
1. Bird AD and Telfer AB. The effect of oxygen at 1 and 2 atmospheres
on resting forearm blood flow. Surg Gynecol Obstet 123: 260 –268,
1966.
2. Boushel R, Langberg H, Gemmer C, Olesen J, Crameri R, Scheede C,
Sander M, and Kjaer M. Combined inhibition of nitric oxide and
prostaglandins reduces human skeletal muscle blood flow during exercise.
J Physiol 534: 691– 698, 2002.
3. Busse R, Pohl U, Kellner C, and Klemm U. Endothelial cells are
involved in the vasodilatory response to hypoxia. Pflügers Arch 397:
78 – 80, 1983.
4. Cowley AJ, Stainer K, Rowley JM, and Wilcox RG. Effect of aspirin
and indomethacin on exercise-induced changes in blood pressure and limb
blood flow in normal volunteers. Cardiovasc Res 19: 177–180, 1985.
5. De Wit C, von Bismarck P, and Pohl U. Synergistic action of vasodilators that increase cGMP and cAMP in the hamster cremaster microcirculation. Cardiovasc Res 28: 1513–1518 1994.
6. Duffy SJ, New G, Tran BT, Harper RW, and Meredith IT. Relative
contribution of vasodilator prostanoids and NO to metabolic vasodilation
in the human forearm. Am J Physiol Heart Circ Physiol 276: H663–H670,
1999.
7. Ekblom B, Huot R, Stein EM, and Thorstensson AT. Effect of changes
in arterial oxygen content on circulation and physical performance. J Appl
Physiol 39: 71–75, 1975.
8. Engelke KA, Halliwell JR, Proctor DN, Dietz NM, and Joyner MJ.
Contribution of NO and prostaglandins to reactive hyperemia in the human
forearm. J Appl Physiol 81: 1807–1814, 1996.
9. Fordy G and Marshall JM. The influence of breathing 40% oxygen on
the response to maximal voluntary contraction of the forearm in human
subjects (Abstract). J Physiol 555: 12, 2004.
10. Frandsen U, Bangsbo J, Langberg H, Saltin B, and Hellsten Y.
Inhibition of nitric oxide synthesis by systemic NG-monomethyl-L-arginine administration in humans: effects on interstitial adenosine, prostacyclin and potassium concentrations in resting and contracting skeletal
muscle. J Vasc Res 37: 297–302, 2000.
J Appl Physiol • VOL
11. Frangos JA, Eskin SG, McIntire LV, and Ives CL. Flow effects on
prostacyclin production by cultured human endothelial cells. Science 227:
1477–1479, 1985.
12. Frisbee JC, Maier KG, and Falck JR. Integration of hypoxic dilation
signaling pathways for skeletal muscle resistance arteries. Am J Physiol
Regul Integr Comp Physiol 283: R309 –R319, 2002.
13. Gaffney FA, Sjogaard G, and Saltin B. Cardiovascular and metabolic
responses to static contraction in man. Acta Physiol Scand 138: 249 –258,
1980.
14. Grabowski EF, Jaffe EA, and Weksler BB. Prostacyclin production by
cultured endothelial cell monolayers exposed to step increases in shear
stress. J Lab Clin Med 105: 36 – 43, 1985.
15. Hammer LW, Ligon AL, and Hester RL. ATP-mediated release
arachidonic acid metabolites from venular endothelium causes arteriolar dilation. Am J Physiol Heart Circ Physiol 280: H2616 –H2622,
2001.
16. Hansen M and Madsen J. Estimation of relative changes in resting
muscle blood flow by 133Xe washout: the effect of oxygen. Scand J Clin
Lab Invest 31: 133–139, 1973.
17. Heavey DJ, Barrow SE, Hickling NE, and Ritter JM. Aspirin causes
short-lived inhibition of bradykinin-stimulated prostacyclin production in
man. Nature 318: 186 –188, 1985.
18. Humphreys PW and Lind AR. The blood flow through active and
inactive muscles of the forearm during sustained hand-grip contraction.
J Physiol 166: 120 –126, 1963.
19. Kagaya A and Homma S. Brachial arterial blood flow during static
handgrip exercise of short duration at varying intensities studied by a
Doppler ultrasound method. Acta Physiol Scand 160: 257–265, 1997.
20. Karamouzis M, Langberg H, Skovgaard D, Bulow J, Kjaer M, and
Saltin B. In situ microdialysis of intramuscular prostaglandin and thromboxane in contracting skeletal muscle in humans. Acta Physiol Scand 171:
71–76, 2001.
21. Kilbom A and Wennmalm A. Endogenous prostaglandins as local
regulators of blood flow in man: effect of indomethacin on reactive and
functional hyperemia. J Physiol 257: 109 –121, 1976.
22. Koller A and Kaley G. Prostaglandins mediate arteriolar dilation to
increased blood flow velocity in skeletal muscle microcirculation. Circ
Res 67: 529 –534, 1990.
23. Koller A, Sun A, Huang A, and Kaley G. Co-release of nitric oxide
and prostaglandins mediates flow-dependent dilation of rat gracilis
muscle arterioles. Am J Physiol Heart Circ Physiol 267: H326 –H332,
1994.
24. Messina EJ, Sun D, Koller A, Wolin MS, and Kaley G. Role of
endothelium-derived prostaglandins in hypoxia-elicited arteriolar dilation
in rat skeletal muscle. Circ Res 71: 790 –796, 1992.
25. Michiels C, Arnould T, Knott I, Dieu M, and Remacle J. Stimulation
of prostaglandin synthesis by human endothelial cells exposed to hypoxia.
Am J Physiol Cell Physiol 264: C866 –C874, 1993.
26. Nowak J and Wennmalm A. Effect of exercise on human arterial and
regional venous plasma concentrations of prostaglandin E. Prostaglandins
Med 1: 489 – 497, 1978.
27. Nowak J and Wennmalm A. A study on the role of endogenous
prostaglandins in the development of exercise-induced and post-occlusive
hyperemia in human limbs. Acta Physiol Scand 106: 365–369, 1979.
28. Pedersen PK, Kiens B, and Saltin B. Hyperoxia does not increase peak
muscle oxygen uptake in small muscle group exercise. Acta Physiol Scand
166: 309 –318, 1999.
29. Pries AR, Heide J, Ley K, Klotz KF, and Gaehtgens P. Effect of
oxygen tension on regulation of arterial diameter in skeletal muscle in situ.
Microvasc Res 49: 289 –299, 1995.
30. Ray CJ, Abbas MR, Coney AM, and Marshall JM. Interactions of
adenosine, prostaglandins and nitric oxide in hypoxia-induced vasodilation: in vivo and in vitro studies. J Physiol 544: 195–209, 2002.
31. Saito Y, Eraslan A, Lockard V, and Hester RL. Role of venular
endothelium in control of arteriolar diameter during functional hyperemia.
Am J Physiol Heart Circ Physiol 267: H1227–H1231, 1994.
32. Schrage WG, Joyner MJ, and Dinneno FA. Local inhibition of nitric
oxide and prostaglandins independently reduces forearm exercise hyperemia in humans. J Physiol 557: 599 – 611, 2004.
33. Shepherd JT. Circulation in skeletal muscle. In: Handbook of Physiology.
The Cardiovascular System. Peripheral Circulation and Organ Blood
99 • JULY 2005 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.32.246 on July 28, 2017
whereas the hyperemia induced by dynamic exercise of knee
extensors was inhibited by combined blockade of PG and NO
synthesis but not by single blockade of either (2). Such results
have led to the idea that blockade of one pathway involved in
generating the dilation results in a compensatory increase in the
role of other factors. Thus the combined effect of COX inhibition and hyperoxia on the postcontraction dilation of the
present study may have been similar to their independent
effects, not because they acted on the same dilator pathway but
because, in the presence of hyperoxia and COX inhibition,
there was a compensatory increase in the contribution of
factors that are independent of COX and PO2. Such factors may
include those released by hypoxia of the muscle fibers, including K⫹, lactate, and H⫹ ions (13) and the myogenic response
to the fall in intravascular pressure that occurs during contraction (33). Assay of PGE2, PGI2, adenosine, NO, K⫹, etc. in
venous plasma and interstium, with and without aspirin and/or
supplementary O2, are required to investigate these possibilities (see Refs. 2, 10, 20).
In summary, the present study has provided compelling
evidence that PGs make a major contribution to the vasodilation that occurs in forearm muscles after isometric contraction
at 60% MVC, particularly to the peak of that dilation. Furthermore, because a similar reduction in this postcontraction vasodilation was produced by breathing 40% O2, the study also
provided strong evidence that this dilation is O2 dependent.
Because the effect of 40% O2 did not occur in the presence of
COX inhibition, it is likely that they interfere with the same
dilator pathway. We propose that the stimulus for the synthesis
of vasodilator PGs during isometric contraction is a fall in PO2
of the endothelium.
51
52
PROSTAGLANDINS AND POSTCONTRACTION DILATION
Flow. Bethesda, MD: Am. Physiol. Soc., 1983, sect. 2, vol. III, pt. 1, chapt.
11, p. 319 –370.
34. Shoemaker JK, Naylor HL, Pozeg RL, and Hughson ZI. Failure of
prostaglandins to modulate the time course of blood flow during dynamic
forearm exercise in humans. J Appl Physiol 81: 1516 –1521, 1996.
35. Taylor JF, Johnson JM, and Kosiba WA. Roles of absolute and relative
load in skin vasoconstrictor responses to exercise. J Appl Physiol 69:
1131–1136, 1990.
36. Welch HG, Bonde-Petersen F, Graham T, Klausen K, and Secher N.
Effects of hyperoxia on leg blood flow and metabolism during exercise.
J Appl Physiol 42: 385–390, 1977.
37. Whitney RJ. The measurement of volume changes in human limbs.
J Physiol 121: 1–27, 1953.
38. Wilson JR and Kapoor S. Contributions of prostaglandins to exerciseinduced vasodilation in humans. Am J Physiol Heart Circ Physiol 265:
H171–H175, 1993.
Downloaded from http://jap.physiology.org/ by 10.220.32.246 on July 28, 2017
J Appl Physiol • VOL
99 • JULY 2005 •
www.jap.org