Journal of Applied Physiology
Vol. 81, No. 4, pp. 1516-1521, October 1996
CONTROL OF BREATHING, CIRCULATION, AND TEMPERATURE

Failure of prostaglandins to modulate the time course of blood flow during dynamic forearm exercise in humans

J. K. Shoemaker, H. L. Naylor, Z. I. Pozeg, and R. L. Hughson

Department of Kinesiology, University of Waterloo, Waterloo, Ontario, N2L 3G1, Canada

ABSTRACT

INTRODUCTION

METHODS

RESULTS

DISCUSSION

ACKNOWLEDGEMENTS

FOOTNOTES

REFERENCES

ABSTRACT

Shoemaker, J. K., H. L. Naylor, Z. I. Pozeg, and R. L. Hughson. Failure of prostaglandins to modulate the time course of blood flow during dynamic forearm exercise in humans. J. Appl. Physiol. 81(4): 1516-1521, 1996.The time course and magnitude of increases in brachial artery mean blood velocity (MBV; pulsed Doppler), diameter (D; echo Doppler), mean perfusion pressure (MPP; Finapres), shear rate ( = 8 ^· MBV/D), and forearm blood flow (FBF = MBV ^· r²) were assessed to investigate the effect that prostaglandins (PGs) have on the hyperemic response on going from rest to rhythmic exercise in humans. While supine, eight healthy men performed 5 min of dynamic handgrip exercise by alternately raising and lowering a 4.4-kg weight (~10% maximal voluntary contraction) with a work-to-rest cycle of 1:1 (s/s). When the exercise was performed with the arm positioned below the heart, the rate of increase in MBV and was faster compared with the same exercise performed above the heart. Ibuprofen (Ibu; 1,200 mg/day, to reduce PG-induced vasodilation) and placebo were administered orally for 2 days before two separate testing sessions in a double-blind manner. Resting heart rate was reduced in Ibu (52 ± 3 beats/min) compared with placebo (57 ± 3 beats/min) (P < 0.05) without change to MPP. With placebo, D increased in both arm positions from ~4.3 mm at rest to ~4.5 mm at 5 min of exercise (P < 0.05). This response was not altered with Ibu (P > 0.05). Ibu did not alter the time course of MBV or forearm blood flow (P > 0.05) in either arm position. The was significantly greater in Ibu vs. placebo at 30 and 40 s of above the heart exercise and for all time points after 25 s of below the heart exercise (P < 0.05). Because PG inhibition altered the time course of at the brachial artery, but not FBF, it was concluded that PGs are not essential in regulating the blood flow responses to dynamic exercise in humans.

arterial diameter; pulsed Doppler ultrasound; echo Doppler ultrasound

INTRODUCTION

PROSTAGLANDINS (PGs) are released from endothelial cells of the vasculature during exercise (16, 29, 32, 34, 36) and can account for much of the flow-induced vasodilation observed in isolated rat cremaster arterioles (6, 17, 18) as well as dog carotid arteries (14). Furthermore, inhibition of PG synthesis has attenuated the flow-induced dilation of rat arterioles following passive increases in blood velocity (17). These results have led to the speculation that PG release may contribute to the exercise-induced vasodilation and hyperemia observed during rhythmic exercise.

PGs appear to have a role in modulating postexercise, or postocclusion, hyperemia (3, 4, 16, 35). However, the role of prostanoid compounds in blood flow control during rhythmic exercise is less certain as some (16, 34), but not all (4, 35, 36), researchers have observed a reduction in blood flow with inhibition of PG synthesis. The role of PGs in blood flow control may be greater during postexercise hyperemia because shear rate () is elevated to higher levels than during steady-state rhythmic exercise because of an increase in mean blood velocity (MBV) (9, 17). PG release occurs rapidly after the initiation of an increase in flow velocity (9) and has been associated with flow-induced vasodilation that occurs within 5-10 s in rat arterioles (17). Because the increase in MBV during human rhythmic exercise is rapid, with a doubling of flow velocity over rest by 5 s of exercise (5, 19, 27), we hypothesized that the effects of PGs on exercise hyperemia might be exerted more during the transition between rest and exercise rather than during steady-state exercise when muscle metabolites also contribute to the vasodilatory response (11, 28). The amount of PG released from cultured human endothelial cells and the subsequent decay rate are proportional to the magnitude of the increase in (9). Therefore, we also hypothesized that, if PGs are a major regulator of limb blood flow during exercise, their effect would be graded with the rate at which MBV adapted to the exercise challenge. The kinetics of the MBV response to exercise are faster when the same work is performed with the exercising arm positioned below, compared with above, the heart.

To assess the role of PG in the time course of hemodynamic responses to rhythmic exercise in humans, pulsed Doppler and echo Doppler sonography were used to obtain continuous measurements of MBV and arterial diameter, respectively, in the brachial artery during the transition from rest to steady-state handgrip exercise. With these measurements, the time course of change in forearm blood flow (FBF) and arterial could be calculated allowing a more complete analysis of the role that PGs play in regulating blood flow and/or vascular tone during exercise.

METHODS

Subjects. Eight healthy males volunteered for this study. The average (±SE) physical characteristics of the subjects were 23.1 ± 0.4 yr of age, 180 ± 3 cm in height, and 76.2 ± 2.0 kg in weight. The volunteers indicated that they were free of any form of cardiovascular or musculoskeletal disease, and none was currently on any medication, as assessed by a medical history. After receiving a complete description of the experimental protocol and potential risks, each subject provided signed consent to the testing procedures on a form approved by the Office of Human Research at the University of Waterloo.

Experimental design. The experiment required that subjects perform dynamic handgrip exercise in the supine position. The handgrip exercise involved raising and lowering a weight of 4.4 kg a distance of 5 cm in a work-to-rest ratio of 1:1 (s/s). The exercise was performed with the right arm, which was positioned at an angle of 50° above (Above) horizontal or 50° below (Below) horizontal. This manipulation allowed modification of arterial perfusion pressure and, therefore, the kinetics of the MBV, FBF, and responses. Brachial arterial pressure in the Below position was elevated by ~30 mmHg, compared with the arm positioned Above the heart.

Before data collection, subjects underwent an orientation session where they were familiarized with the experimental design and data-collection procedures. On a subsequent day, subjects performed a mock data-collection session to more rigorously familiarize themselves with the exercise and data-gathering procedures outlined in detail below.

Four trials of the exercise, separated by at least 10 min of rest, were performed in each arm position while under the influence of either placebo or ibuprofen (Ibu). In each exercise trial, data were collected during 1 min of rest followed by a step increase in workload, which lasted 5 min. To minimize the anticipatory effects of exercise, the subjects were not aware of the time in any trial but were told when to begin and to end the forearm contractions. The subjects rested in the supine position for 15-20 min before data collection. They had not eaten for 2-3 h and had abstained from consumption of alcohol and caffeine for 24 h before any test.

The placebo and Ibu tests were performed in a double-blind manner separated by at least 5 days. Ibu, placed in a gelatin capsule, was consumed orally at 1,200 mg/day in regularly spaced aliquots of 200 mg for 2 days before the test. To enhance the Ibu effects, these levels were increased by having each subject consume an additional 200 mg at 2 and 1 h before the test. A further 200 mg were consumed on arrival at the laboratory. Because each data-collection session lasted ~3 h, a final 200 mg dose was consumed 1.5 h into the test in an attempt to sustain blood Ibu levels. Placebo capsules were consumed in the same manner.

Data collection. The echo Doppler (Toshiba model SSH-140A) image of the brachial artery was collected continuously in real time during the first trial of each experimental condition. A hand-held 7.5-MHz linear probe operating in B-mode (brightness mode) was positioned over the brachial artery between the biceps aponeurosis and muscle belly. The imaged data were stored on VHS tape for analysis.

In the three subsequent trials of each experimental session, beat-to-beat heart rate [Cambridge model VS4 electrocardiograph (ECG)], arterial blood pressure (Finapres finger cuff, Ohmeda 2300) (13), and brachial artery MBV (pulsed Doppler velocimetry, Multigon model 500V) were collected continuously on a computer-based system at 100 Hz. Forearm mean perfusion pressure (MPP) was estimated by holding the left arm and hand, from which blood pressure was collected, at the level of the pulsed Doppler probe.

Brachial artery MBV was determined from the spectra of the pulsed Doppler ultrasound signal. A flat probe with an operating frequency of 4 MHz was fixed to the skin over the brachial artery in the antecubital fossa region of the right elbow. The angle of the transducer crystal relative to the skin was 45°, and echo ultrasound imaging confirmed that the brachial artery ran parallel to the skin in this location. The ultrasound gate was set to insonate the total width of the artery lumen. With this apparatus, we were able to maintain a clear Doppler signal both at rest and during the handgrip exercise. The Doppler shift frequency spectra were processed by a quadrature audio demodulator (21), which provided the instantaneous MBV in real time. Beat-by-beat MBV was calculated as the average of the instantaneous MBV values over each cardiac cycle using the QRS complex of the ECG tracing to signal the end of one heart beat and the beginning of the next.

Data analysis. An estimate of arterial diameter was made from the average of three measurements each at 20-s intervals during 1 min of rest, at 5-s intervals between 0 and 30 s of exercise, at 10-s intervals between 30 and 60 s, and also at 90, 120, 180, 240, and 300 s of exercise. All diameter measurements were made during diastole in a muscle-relaxation phase. The diameter measurement markers of the Toshiba imaging system moved in increments of 0.1 mm. We have previously observed that, with these methods, the day-to-day coefficients of variation for these data range from 2-4% (unpublished observations).

For each subject, the three trials of MBV and MPP data from a given experimental session were time aligned and averaged into a single data set with data points at each 2 s. Each 2-s time bin incorporated a contraction and relaxation phase of the exercise. From these averaged data sets, MBV and MPP values were obtained at the same times used for the diameter estimations, as indicated above. FBF at each time was calculated as FBF = MBV ^· r². From the simultaneous MBV and diameter (D) measures, was calculated (22) as

To describe the rate of increase in MBV, an exponential curve-fitting procedure was applied to all the averaged data sets. This procedure has been described in detail elsewhere (27) and allows calculation of the mean response time, which is the time required to achieve 63% of the increase in MBV between rest and the steady-state exercise level.

To determine the effectiveness of Ibu treatment on cyclooxygenase inhibition, the time to the onset, and to the peak, of platelet aggregation in whole blood (Chrono-Loc whole blood aggregometer, Havertown, PA) was evaluated in two subjects at rest and after 3 min of occluded forearm exercise with and without Ibu. Five milliliters of forearm venous blood were obtained from an antecubital fossa vein into cuvettes containing 0.5 ml of 0.105 M sodium citrate. Aggregation was stimulated by the addition of 1 mM arachidonic acid.

Statistics. Initial comparisons of the individual main effects of Ibu, arm position, and time were analyzed for each time point by repeated-measures three-way analysis of variance (ANOVA; Statistical Analysis System), with arm position, drug, and time as the independent variables. Where appropriate, interaction analysis was performed by one-way ANOVA procedures using the three-way ANOVA mean square error of the interaction as the error term (15). The level of significance was set at P < 0.05, and any differences were further analyzed with Student-Newman-Keuls post hoc test. Data are presented as means ± SE.

RESULTS

Heart rate. Ibu resulted in significant reductions in resting heart rate compared with placebo (P < 0.05) (Table 1). Exercise heart rates also tended to be lower with Ibu treatment, with the differences being significant (P < 0.05) at 2 min. Resting heart rates were not different with the arm positioned Above vs. Below the heart. However, exercise heart rates tended to be lower with the arm positioned Below the heart, with significant reductions observed at 3-5 min (P < 0.05) (Table 1).

Table 1. Heart rate responses to dynamic handgrip exercise with and without ibuprofen Rest Time, min 1 2 3 4 5 Placebo   Above 58 ± 3  62 ± 2  65 ± 4  66 ± 3  66 ± 4  65 ± 4    Below 57 ± 3  62 ± 5  62 ± 4  62 ± 4* 62 ± 4* 63 ± 4  Ibu   Above 55 ± 3 60 ± 3  60 ± 3 64 ± 4  68 ± 5  66 ± 5    Below 52 ± 3 57 ± 3  58 ± 3 57 ± 4* 58 ± 4* 58 ± 4* Values (in beats/min) are means ± SE. Above and Below are arm positions; Ibu, ibuprofen. * Significantly different from Above position (P < 0.05); significantly different from placebo for that arm position (P < 0.05).

MPP. When the arm was moved from the Above to the Below position, perfusion pressure was increased by ~30 mmHg (P < 0.05) (Table 2), and this difference was maintained throughout the exercise. Compared with placebo, the difference in MPP between arm positions was not affected by Ibu treatment (P > 0.05).

Table 2. Mean perfusion pressure responses to dynamic handgrip exercise with and without ibuprofen Rest Time, min 1 2 3 4 5 Placebo   Above 77.9 ± 3.1  82.7 ± 3.0  87.3 ± 3.0  91.8 ± 4.3  93.6 ± 4.5  94.8 ± 4.0    Below 105 ± 4* 109 ± 4* 113 ± 4* 113 ± 4* 112 ± 4* 115 ± 5* Ibu   Above 77.8 ± 5.7  83.2 ± 6.2  90.1 ± 6.3  92.9 ± 7.2  93.0 ± 7.4  94.9 ± 7.6    Below 108 ± 4* 112 ± 4* 114 ± 5* 114 ± 4* 113 ± 5* 114 ± 5* Values (in mmHg) are means ± SE. * Significantly greater than Above (P < 0.05).

MBV. Resting MBV ranged from 8.08 ± 0.86 to 9.25 ± 1.38 cm/s across the experimental conditions (P > 0.05) (Fig. 1). MBV increased rapidly after the onset of exercise, and the mean response time for MBV with placebo was faster in Below (24.2 ± 4.1 s) than in Above (75.3 ± 42.3 s) exercise (P < 0.05). Ibu treatment did not alter these MBV kinetics. Therefore, by 5 s, MBV was 13.7 ± 1.2 and 12.1 ± 1.6 cm/s (placebo and Ibu, respectively) in Above (P > 0.05) and 17.9 ± 1.8 and 17.6 ± 1.2 cm/s (placebo and Ibu, respectively) in Below tests (P > 0.05). These 5-s MBV values were all greater than at rest (P < 0.05). At all measured time points during the first 60 s of exercise, the increase in MBV during the Below tests was greater than in the Above tests (P < 0.05). However, the steady-state MBV levels were not different between arm position or experimental conditions. For example, at 5 min of exercise, MBV was 29.2 ± 2.3 and 30.1 ± 2.7 cm/s during placebo and Ibu tests, respectively, in the Above position (P > 0.05). In the Below position, MBV at 5 min of exercise was 31.0 ± 1.5 cm/s in placebo and 35.3 ± 2.7 cm/s in Ibu (P > 0.05).

Brachial artery diameter. Resting diameters were 4.3 ± 0.2 mm in the Above position to 4.4 ± 0.2 with the arm Below the heart (P > 0.05) and were not different between drug conditions (P > 0.05) (Fig. 1). In all tests, diameter was reduced during the first 15 s of exercise (P < 0.05). However, the overall response was vasodilation where diameters became significantly greater than rest by 90 s of exercise (main effect, P < 0.05). There was a tendency for exercise diameters to be smaller during Below exercise with Ibu, but this failed to reach statistical significance (P > 0.05).

FBF. Compared with the placebo tests, Ibu treatment did not alter the rate or magnitude of increase in FBF during Above or Below exercise (P > 0.05). The resting FBF was ~70-80 ml/min. By 5 s of exercise, FBF had increased significantly above rest levels (P < 0.05) in both arm positions, with greater flows observed during Below exercise (160 ± 19 ml/min) compared with Above (122 ± 11 ml/min) (P < 0.05; Fig. 1). The greater FBF in the Below, compared with Above, tests continued throughout the exercise period so that FBF had increased to 326 ± 20 and 279 ± 26 ml/min in the Below and Above tests, respectively (P < 0.05) by 5 min.

The values. At rest, was similar across all test conditions. Exercise in the Below position resulted in greater values compared with Above (main effect, P < 0.05), and this difference tended to be dependent on the drug condition (P = 0.05) (Fig. 2). In Above exercise, Ibu treatment was associated with greater at 30 and 40 s of exercise, compared with placebo. In the Below position, was greater with Ibu than with placebo, beginning at 25 s of exercise and continuing for the remainder of the exercise period.

Platelet aggregation. In the two subjects tested, Ibu slowed the platelet aggregation times (P < 0.05) (Table 3). The delaying effects of Ibu on platelet aggregation were observed in both the resting and postocclusion samples.

Table 3. Platelet aggregation times with and without ibuprofen Placebo Ibu Rest   Onset 0.5 ± 0  7.3 ± 1*   Peak 7.5 ± 1  15 ± 0* Postocclusion   Onset 0.5 ± 0.1  4.3 ± 1*   Peak 10.5 ± 6  12 ± 1 Values (in min) are means ± SE; n = 2 subjects. * Significantly different from placebo (P < 0.05).

DISCUSSION

In these experiments, we were able to achieve greater values at the onset of identical external work rates by positioning the arm Below, compared with Above, the heart. As elevated is a potent stimulus for PG release (9, 17), it was anticipated that any effect of PG inhibition by Ibu might be unmasked by this exercise protocol. The FBF response was clearly greater during the first 60 s of exercise when the arm was positioned Below, compared with Above, the heart, but there was no effect of Ibu. Therefore, it was concluded that, for the exercise protocols employed in the present study, PGs were not essential in regulating FBF at any time during the on-transient or steady-state phases of submaximal dynamic exercise. However, PGs did attenuate arterial in a manner that appeared dependent on the rate of increase in MBV. These observations are in contrast to earlier reports (16, 34) and must be evaluated critically.

Background and methodological concerns. The role that PGs play in the regulation of blood flow responses to exercise has received considerable interest, but their contribution remains unclear. PGs have been shown to be effective modulators of microvascular tone in rats (6, 12, 17, 18, 20) and appear to contribute to resting blood flow levels in dogs (36). In humans, PGs do not appear to exert any measurable influence over resting blood flow (34). With dynamic exercise, PGs are released (16, 34), but the evidence supporting a role for PGs in regulating the exercise hyperemia (16, 34) is questionable. Kilbom and Wennmalm (16) used strain-gauge plethysmography to measure blood flow responses during isometric exercise. However, it has been clearly shown with Doppler techniques that blood flow is impeded even during low-intensity contractions (31). Therefore, it is uncertain what is being indicated by measuring flow during isometric contractions. In the study of Wilson and Kapoor (34) strain-gauge plethysmography was also used to measure the effects of infused indomethacin on FBF responses to dynamic exercise. Although these authors concluded that cyclooxygenase inhibition resulted in a reduction in exercise FBF, closer analysis of their data suggests that the variability in the FBF responses in the control group could account for almost all of the effects of indomethacin. Also, these authors (34) acknowledged that strain-gauge plethysmographic measures of FBF require the cessation of exercise so that the observed flows reflect a combination of active and reactive hyperemia. These concerns, in conjunction with recent evidence that venous occluding pressures may interfere with limb blood flow (30), make the conclusions of these earlier reports (16, 34) uncertain.

In the present study, we have attempted to avoid these earlier methodological difficulties by using Doppler ultrasound blood flow measurements of both arterial MBV and diameter. These measurements are noninvasive and allow continuous collection of both MBV and diameter without stopping the exercise. Also, the continuous nature of Doppler measurements allows investigation of the time-dependent effects, if any, of PGs on vascular dynamics. An additional advantage of Doppler ultrasound methods is that the posture of the exercising arm can be manipulated. Strain-gauge plethysmography requires the arm be positioned somewhat above the level of the heart to facilitate venous drainage. Such an arm position attenuates the rate of increase in MBV and, therefore, of compared with placing the arm below the heart. The appears to be an important stimulant for PG release from arterial endothelial tissue (9, 17, 18), and we reasoned that the ability to measure blood flow responses under conditions of altered kinetics might help clarify the role of PGs during exercise. An important advantage of Doppler ultrasound, then, is the ability to calculate from the relationship between simultaneously determined MBV and arterial diameter.

Our methods of collecting pulsed Doppler MBV data have been highly correlated with strain-gauge plethysmographic estimates of total FBF (30). Furthermore, we have demonstrated a strong linear relationship (r = 0.99) between the measured MBV and actual blood velocities through plastic tubes (Doppler MBV = 0.9 velocity + 2.4). Together with the low degree of variability of our diameter measurements between days, we are confident that our blood flow estimates by ultrasonography were representative of true flow rates through the brachial artery.

Exercise MBV and FBF. In contrast to the earlier conclusions of Kilbom and Wennmalm (16) and Wilson and Kapoor (34), the results of the present study do not support the hypothesis that PGs contribute to the magnitude or time course of blood flow responses during dynamic exercise. Although FBF increased at a faster rate during Below exercise, compared with Above, these responses were not altered by Ibu. Notably, cyclooxygenase inhibition did not modulate the large and rapid increase in FBF at the onset of exercise. With Ibu, FBF was maintained in the face of modest reductions in diameter due to slightly elevated MBV. These data suggest either that blood flow regulatory factors other than PG, acting peripherally to the brachial artery, must have compensated for the reduction in PG or that PGs do not play an essential role in the hyperemia with exercise. In this aspect, microvessels embedded within parenchymal tissue are highly reactive to various vasoactive stimuli associated with skeletal muscle contraction (26). These signals are integrated and conducted upstream to cause vasodilation, first in arterioles and then in the feed arteries (24, 33). Feed arteries, and not conduit arteries, are believed to regulate the magnitude of blood flow perfusing the vascular beds of active skeletal muscle (25). In these small arteries, stimuli conducted upstream from arterioles and flow-induced signals are integrated to cause a substantial increase in vascular conductance in proportion to the metabolic stress (25, 33). Therefore, despite PG inhibition, the compensating influence of other vasodilatory stimuli acting on microvessels and feed arteries could facilitate adequate muscle perfusion with a subsequent augmentation of MBV through the slightly constricted conduit artery.

It is believed that the exercise chosen for the current study was of sufficient intensity to cause an increase in the rate of PG release (34). Furthermore, the oral dosage used in the current study was clinically relevant for treatment of musculoskeletal inflammation (1) and is believed to be more than adequate for suppressing PG synthesis (2). Importantly, the dosage of Ibu was effective in inhibiting platelet cyclooxygenase activity. Some nonsteroidal anti-inflammatory drugs, indomethacin in particular, may have direct vasoconstrictor properties independent of their effect on cyclooxygenase when consumed in dosages that are greater than those required for control of inflammation (7, 8, 23). Whether these effects extend to Ibu is not currently known. However, it is clear that at clinically prescribed dosages, such as those used in the current study, the preferential action of anti-inflammatory drugs is on cyclooxygenase inhibition (3, 8).

The values. Although PGs did not appear to be essential in the regulation of the exercise hyperemia in the current study, evidence that was greater with Ibu suggests that PGs may alter vascular tone during exercise. The is proportional to the quotient of MBV divided by diameter. Therefore, small directionally opposite changes in MBV and diameter with Ibu, compared with placebo, can result in significant changes in , as observed in this study. With Ibu, brachial artery was greater at 30 and 40 s of Above exercise and for all measured time points from 25 s to 5 min in the Below tests. Therefore, PGs appeared to be released early in the exercise, which is consistent with an earlier report that PGs released from the endothelial tissue of the isolated rat microcirculation exert a strong vasodilatory influence with a time delay of ~20 s (18).

The observation that the effects of Ibu on during steady-state exercise were dependent on the arm position may be explained by previous observations that PG release from cultured human endothelial cells is biphasic, with an initial peak at the onset of the shearing stimulus and a fast decay (9, 10). In these cultured cells, the initial peak production rate of PGs was directly proportional, and the subsequent decay rate was inversely proportional, to the magnitude of increase in shear stress (10).

Summary. Continuous measures of both brachial artery diameter and MBV during a placebo or Ibu condition have shown that PGs provide little or no contribution to the exercise hyperemia associated with dynamic forearm exercise. Although blood flow responses were not different between the placebo and Ibu conditions, the levels of were augmented with cyclooxygenase inhibition. The augmented levels were observed in both arm positions at 30-40 s of exercise; only during the Below tests was elevated with Ibu during steady-state exercise. It was concluded, therefore, that PGs are not essential to achieve a normal blood flow response to dynamic handgrip exercise performed either Above or Below the heart. However, with Ibu, the normal blood flow response was achieved with blood flowing faster through a somewhat narrower brachial artery.

ACKNOWLEDGEMENTS

This study was supported by the Natural Sciences and Engineering Research Council of Canada. J. K. Shoemaker was a recipient of a Natural Sciences and Engineering Research Council of Canada Graduate Student Scholarship.

FOOTNOTES

Address for reprint requests: R. L. Hughson, Dept. of Kinesiology, Univ. of Waterloo, 200 University Ave. W, Waterloo, ON, N2L 3G1, Canada.

Received 16 January 1996; accepted in final form 4 June 1996.

REFERENCES

1.	Abramson, S. B. Non-steroidal anti-inflammatory drugs: mechanisms of action and therapeutic considerations. In: Sports-Induced Inflammation, edited by W. B. Leadbetter, J. A. Buckwalter, and S. L. Gordon. Park Ridge, IL: Am. Acad. of Orthopaedic Surgeons, 1990, p. 421-430.
2.	Canadian Pharmaceutical Association Compendium of Pharmaceuticals and Specialties (ed. 29). Ottawa, Canada: Can. Pharmaceut. Assoc., 1994, p. 812-814.
3.	Carlsson, I., and A. Wennmalm. Effect of different prostaglandin synthesis inhibitors on post-occlusive blood flow in human forearm. Prostaglandins 26: 241-251, 1983.
4.	Cowley, A. J., K. Stainer, J. M. Rowley, and R. G. Wilcox. 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.	Eriksen, M., B. A. Waaler, L. Walloe, and J. Wesche. Dynamics and dimensions of cardiac output changes in humans at the onset and at the end of moderate rhythmic exercise. J. Physiol. Lond. 426: 423-437, 1990.
6.	Faber, J. E., P. D. Harris, and I. G. Joshua. Microvascular response to blockade of prostaglandin synthesis in rat skeletal muscle. Am. J. Physiol. 243 (Heart Circ. Physiol. 12): H51-H60, 1982.
7.	Flores, A. G., and G. W. G. Sharp. Endogenous prostaglandins and osmotic water flow in the toad bladder. Am. J. Physiol. 223: 1392-1397, 1972.
8.	Flower, R. J. Drugs which inhibit prostaglandin synthesis. Pharmacol. Rev. 26: 33-67, 1974.
9.	Francos, J. A., S. G. Eskin, L. V. McIntire, and C. L. Ives. Flow effects on prostacyclin production by cultured human endothelial cells. Science Wash. DC 227: 1477-1479, 1985.
10.	Grabowski, E. F., E. A. Jaffe, and B. B. Weksler. Prostacyclin production by cultured endothelial cell monolayers exposed to step increases to shear stress. J. Lab. Clin. Med. 105: 36-43, 1985.
11.	Haddy, F. J., and J. B. Scott. Metabolic factors in peripheral circulatory regulation. Federation Proc. 34: 2006-2011, 1975.
12.	Hill, M. A., M. J. Davis, and G. A. Meininger. Cyclooxygenase inhibition potentiates myogenic activity in skeletal muscle arterioles. Am. J. Physiol. 258 (Heart Circ. Physiol. 27): H127-H133, 1990.
13.	Imholz, B. P. M., J. J. Settels, A. H. van der Meiracker, K. H. Wesseling, and W. Wieling. Non-invasive continuous finger blood pressure measurement during orthostatic stress compared to intra-arterial pressure. Cardiovasc. Res. 24: 214-221, 1990.
14.	Kamiya, A., and T. Togawa. Adaptive regulation of wall shear stress to flow change in the canine carotid artery. Am. J. Physiol. 239 (Heart Circ. Physiol. 8): H14-H21, 1980.
15.	Keppel, G. Design and Analysis: A Researcher's Handbook (2nd ed.). Englewood Cliffs, NJ: Prentice-Hall, 1982.
16.	Kilbom, A., and A. Wennmalm. Endogenous prostaglandins as local regulators of blood flow in man: effect of indomethacin on reactive and functional hyperaemia. J. Physiol. Lond. 257: 109-121, 1976.
17.	Koller, A., and G. Kaley. Prostaglandins mediate arteriolar dilation to increased blood flow velocity in skeletal muscle microcirculation. Circ. Res. 67: 529-634, 1990.
18.	Koller, A., D. Sun, and G. Kaley. Role of shear stress and endothelial prostaglandins in flow- and viscosity-induced dilation of arterioles in vitro. Circ. Res. 72: 1276-1284, 1993.
19.	Leyk, D., D. Essfeld, K. Baum, and J. Stegemann. Early leg blood flow adjustment during dynamic foot plantarflexions in upright and supine body position. Int. J. Sports Med. 15: 447-452, 1994.
20.	Messina, E. J., R. Weiner, and G. Kaley. Inhibition of bradykinin vasodilation and potentiation of norepinephrine and angiotensin vasoconstriction by inhibitors of prostaglandin synthesis in skeletal muscle of the rat. Circ. Res. 37: 430-437, 1975.
21.	Micco, A. J. (Inventor). CW and Pulse Doppler Diagnostic System. US Patent 4,819,622. April 22, 1989. Int. Cl. A61B 10/00. US Patent 4,819,652, 22 April 1989.
22.	Nichols, W. W., and M. F. O'Rourke. McDonald's Blood Flow in Arteries. Philadelphia, PA: Lea & Febiger, 1990.
23.	Northover, B. J. Indomethacina calcium antagonist. Gen. Pharmacol. 8: 293-296, 1977.
24.	Segal, S. S. Microvascular recruitment in hamster striated muscle: role for conducted vasodilation. Am. J. Physiol. 261 (Heart Circ. Physiol. 30): H181-H189, 1991.
25.	Segal, S. S. Communication among endothelial, and smooth muscles coordinates blood flow control during exercise. News Physiol. Sci. 7: 152-156, 1992.
26.	Segal, S. S. Cell-to-cell communication coordinates blood flow control. Hypertension Dallas 23: 1113-1120, 1994.
27.	Shoemaker, J. K., S. M. Phillips, H. J. Green, and R. L. Hughson. Faster femoral artery blood velocity kinetics at the onset of exercise following short-term training. Cardiovasc. Res. 31: 278-286, 1996.
28.	Sparks, H. V., Jr. Effect of local metabolic factors on vascular smooth muscle. In: Handbook of Physiology. The Cardiovascular System. Vascular Smooth Muscle. Bethesda, MD: Am. Physiol. Soc., 1980. sect. 2, vol. II, chapt. 17, p. 475-513.
29.	Staessen, J., A. Cattaert, R. Fagard, P. Lijnen, E. Moerman, A. de Schaepdryver, and A. Amery. Hemodynamic and humoral effects of prostaglandin inhibition in exercising humans. J. Appl. Physiol. 56: 39-45, 1984.
30.	Tschakovsky, M. E., J. K. Shoemaker, and R. L. Hughson. Beat-by-beat forearm blood flow with Doppler ultrasound and strain-gauge plethysmography. J. Appl. Physiol. 79: 713-719, 1995.
31.	Walloe, L., and J. Wesche. Time course and magnitude of blood flow changes in the human quadriceps muscles during and following rhythmic exercise. J. Physiol. Lond. 405: 257-273, 1988.
32.	Wennmalm, A., and G. A. Fitzgerald. Excretion of prostacyclin and thromboxane A2 metabolites during leg exercise in humans. Am. J. Physiol. 255: H15-H18, 1988.
33.	Williams, D. A., and S. S. Segal. Feed artery role in blood flow control to rat hindlimb skeletal muscles. J. Physiol. Lond. 463: 631-646, 1993.
34.	Wilson, J. R., and S. Kapoor. Contribution of prostaglandins to exercise-induced vasodilation in humans. Am. J. Physiol. 265 (Heart Circ. Physiol. 34): H171-H175, 1993.
35.	Young, E. W., and H. V. Sparks. Prostaglandin E release from dog skeletal muscle during restricted flow exercise. Am. J. Physiol. 236 (Heart Circ. Physiol. 5): H596-H599, 1979.
36.	Young, E. W., and H. V. Sparks. Prostaglandins and exercise hyperemia of dog skeletal muscle. Am. J. Physiol. 238 (Heart Circ. Physiol. 7): H190-H195, 1980.