INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

THE ABILITY OF SKELETAL MUSCLE to sustain repeated contractions during rhythmic exercise is dependent on an adequate blood supply. A failure to increase blood flow in proportion to the metabolic demands of exercising muscle may result in early fatigue. Studies in humans performing single-leg knee extension exercise have shown that, when a small muscle mass is recruited, such as the quadriceps (~2.5 kg), leg blood flow increases in proportion to increases in power output without any evidence of a blood flow limitation at higher work rates (1, 24). However, previous studies using isometric (static) knee extension exercise have indicated that, when the quadriceps contract at 10% maximal voluntary contraction (MVC), intramuscular pressures are increased sufficiently to occlude blood flow in some regions of the contracting muscle (10, 33, 36). As the percentage of MVC was increased further, blood flow during contraction became progressively compromised because of further increases in intramuscular pressure.

The growing application of nuclear magnetic resonance spectroscopy to examine mechanisms controlling cellular respiration has led researchers to utilize knee extension exercise. With the exception of Richardson et al. (26), who used a large bore magnet, most knee extension exercise in a magnetic resonance spectroscopy magnet is characterized by a limited range of motion because of the physical constraints of the magnet (for example, Ref. 28). Unlike the cyclic motion of the Krogh ergometer typically used to examine muscle blood flow during knee extension exercise (1), the exercise mode in a magnet results in greater muscle tension development for the same power output over the entire range of motion, which may lead to greater impedance to blood flow. In addition to this mode-specific exercise, subjects are required to exercise while lying supine, which has been shown to affect the non-steady-state blood flow response (17) and, consequently, affect the metabolic response to the exercise (37).

It is readily apparent from the studies of Walloe and Wesche (35) that large oscillations in blood flow occur consequent to the mechanical interference caused by intermittent muscular contractions, which are followed by a transient hyperemia before the onset of the next muscular contraction. Studies in isolated muscle preparations have shown that the time available for flow between rhythmic contractions (relaxation duration) also affects the mean blood flow (MBF) response (9). There is evidence (2, 4, 13) in the diaphragm that a critical duty cycle (i.e., ratio of contraction time to total cycle duration) of ~20-50% exists, above which blood flow may become mechanically impeded. Additionally, Dodd et al. (6) examined blood flow in a dog hind-leg preparation and concluded that muscle tension development and contraction duty cycle were the primary determinants regulating blood flow. Previous exercise studies in humans (32, 35) have used contraction-relaxation phases that are relatively longer than those observed during natural locomotor activities, such as running or cycling, where contraction frequencies may be as high as 80 contractions/min (cpm). Interestingly, Kagaya (14) observed a plateau in MBF in the calf muscle of humans as running velocity increased from 60 to 220 m/min, suggesting that blood flow may have been impeded as contraction frequency (i.e., strides per minute) increased. However, blood flow was determined using strain-gauge plethysmography, which precluded obtaining blood flow measurements during exercise.

Therefore, the purpose of the present investigation was to characterize the parameters of rhythmic muscle contraction (frequency, relaxation duration, tension), which may affect the mean and instantaneous leg blood flow during exercise in humans. We tested the hypothesis that, at the same power output, increases in contraction frequency (or decreases in relaxation duration and time available for blood flow) would be associated with an attenuation in MBF during incremental exercise.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects. Nine healthy male volunteers participated in the study. The average physical characteristics of the subjects were 23.6 ± 3.7 yr (range: 20-45 yr) of age, 177.0 ± 2.1 cm (168-188 cm) in height, 77.0 ± 3.2 kg (68-93 kg) in weight, and 3.43 ± 0.57 l/min (2.5-4.4 l/min) peak O₂ uptake. Subjects were informed of all possible risks and discomforts associated with the experiment protocol before providing written consent. This study was approved by the Human Subjects Committee at Kansas State University where all exercise tests were conducted.

Experimental design. For this experiment, subjects performed incremental one-leg knee extension exercise in the supine position. Each incremental test was performed at a contraction frequency of 80 cpm (fast), 60 cpm (medium), or 40 cpm (slow). A metronome and verbal encouragement were used to assist the subjects in maintaining the appropriate kicking frequency. The leg ergometer used in this study was specifically designed for use in a magnetic resonance scanner; therefore, the distance traveled by the lower limb during knee extension was limited to a range of motion of ~20°. When the leg ergometer is used, the right knee is placed on a brace, which resulted in an angle about the knee of ~90° at rest. A strap was then placed around the ankle of the subject and attached to a pneumatic cylinder by means of a cable-pulley system. Work was accomplished by compressing the air in the cylinder as the lower leg was extended. The displacement (d) of the lower leg was fixed at 10.9 cm. The amount of work performed (per stroke) was calculated as [P_INT + (P_F + P_INT)/2] · d, where P_INT is the initial pressure in the cylinder and P_F is the final pressure at the end of the work stroke. This in turn was divided by the duration of the contraction cycle to yield power (W). The continuous changes in pressure and piston position that occurred with each knee extension were recorded and stored on computer disk for further analysis of upstroke duration (knee extension or contraction phase), downstroke duration (passive knee flexion or relaxation phase), relaxation duration (passive knee flexion and rest phase), and duty cycle (upstroke duration/cycle duration). The electrocardiogram was digitized and stored simultaneously with the Doppler velocity data for subsequent off-line analysis.

Mean blood velocity measurements. Continuous measurement of blood velocity in the right femoral artery was obtained by a Doppler ultrasound velocimetry system (model 500-V, Multigon Industries) operating in pulsed mode. The pulsed-wave Doppler transducer, with an operating frequency of 4 MHz and fixed transducer crystal and sound beam angle of 45° relative to the skin, was placed flat on the thigh 2-3 cm below the inguinal ligament, above and parallel to the common femoral artery. This position was selected to minimize turbulent flow arising from the bifurcation of the common femoral artery into the superficial and profundus branches. The gate was set at full width to ensure complete femoral artery insonation. The frequency spectrum of Doppler audio signals was converted to an instantaneous mean blood velocity by using a quadrature audio demodulator that was calibrated according to the specifications of the manufacturer (Hokanson). Software developed in our laboratory was used to calculate three indexes of femoral artery blood velocity averaged over one cardiac cycle between the R-wave-R-wave interval: 1) net mean blood velocity, calculated as the integrated area under the average velocity profile; 2) antegrade blood velocity (i.e., caudal direction velocities); and 3) retrograde blood velocity (i.e., rostral direction velocities). The blood velocities were expressed per minute (i.e., cm/min) by multiplying the cardiac cycle-by-cycle values by the corresponding heart rate (HR).

Femoral artery diameter measurement. On a separate day, the femoral artery cross-sectional area (CSA) was determined by using a duplex Doppler computed sonography system (model 128XP, Acuson) in two-dimensional echo mode. The vessel diameter was determined during supine rest over 10-15 cardiac cycles at peak systole and end diastole from a cross-sectional view of the artery at the level used to measure blood velocity. The measurement of arterial diameter was performed on cardiac cycles that provided optimal resolution of the arterial borders. The mean of the peak systolic and end-diastolic diameters over 10-15 beats was used to calculate an average diameter (D) using the estimate, D = (systole/3) + 2 · (diastole/3) (22). The average diameter measurements were used for computing the CSA (CSA = r²) of the artery, which in turn was multiplied by the appropriate blood velocity (cm/min) to obtain the relevant flows (ml/min): net MBF (MBF_NET), antegrade MBF (MBF_ANT), and retrograde MBF (MBF_RET).

Statistical analysis. During the incremental tests, each subject fatigued at different stages of the tests, and, therefore, comparisons of the exercise responses were only made up to the last power output completed by all subjects (i.e., the first four exercise stages) and at exhaustion. A two-way repeated-measures ANOVA design was used to test for differences for the cardiovascular and leg ergometer variables of interest with power output and kicking frequency as the main effects. All values are presented as means ± SE, unless indicated otherwise. Statistical significance was accepted at P < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Ergometer data. The group mean values for the average ergometer response over the entire exercise test (excluding the last 30 s of exercise) are listed in Table 1. The slope of the power output-time relationship for each kicking frequency is presented in Fig. 1. Power output increased linearly for all contraction frequencies with no significant differences among increases for each of the kicking protocols. Table 2 shows that end-exercise power output was lower as contraction frequency increased, such that 40, 60, and 80 cpm elicited peak power outputs of 16.2 ± 1.4, 14.8 ± 1.4, and 13.2 ± 1.3 W, respectively (P < 0.05).

View larger version (15K): [in this window] [in a new window]   Fig. 1.   Comparison of power outputs during submaximal exercise and at exhaustion (Exh) for incremental knee extension exercise. Note that the values for power output at fatigue appear lower than the mean values at lower work rates because of the dropout of subjects in the 600- to 840-s range. cpm, Contractions per minute. * Significantly different from corresponding value at 80 cpm, P < 0.05. + Significantly different from corresponding value at 60 cpm, P < 0.05.

Net blood flow (MBF_NET). The cardiac cycle-by-cycle blood flow for one subject performing incremental exercise at 40, 60, and 80 cpm is shown in Fig. 2. At rest, MBF_NET was similar among kicking frequencies (369.2 ± 148.6, 424.3 ± 206.4, and 322.2 ± 135.8 ml/min for 40, 60, and 80 cpm, respectively). With exercise, MBF_NET increased (P < 0.05) from rest for all kicking frequencies. However, for all three kicking frequencies, MBF_NET did not increase linearly throughout the range of exercise intensities but rather exhibited an attenuated rise at the higher power outputs. Figure 3A illustrates that, beyond a power output of 5 W, MBF_NET was higher (P < 0.05) for 40 than for 80 cpm. At a power output of 9 W, MBF_NET was also higher (P < 0.05) during exercise at 40 than 60 cpm. However, at exhaustion, no difference in MBF_NET was observed among the kicking frequencies because of large intersubject variability in both the peak power output achieved and the MBF_NET response (Table 2).

View larger version (32K): [in this window] [in a new window]   Fig. 2.   Net mean blood flow (MBFNET) averaged over a cardiac cycle in 1 subject during incremental knee extension exercise performed in the supine position at 40 (A), 60 (B), and 80 cpm (C).

View larger version (16K): [in this window] [in a new window]   Fig. 3.   The group mean response for MBFNET (A), antegrade blood flow (MBFANT; B), and retrograde blood flow (MBFRET; C) to incremental single-leg knee extension exercise to exhaustion. All subjects contributed to the group mean response during the first 4 stages of exercise and at exhaustion. * Significantly different from corresponding value at 80 cpm, P < 0.05. + Significantly different from corresponding value at 60 cpm, P < 0.05. Note that statistics were only performed on the first 4 power outputs and at exhaustion, where all 9 subjects contributed data to the group mean.

Antegrade blood flow (MBF_ANT). Figure 3B shows that MBF_ANT increased from resting levels for all kicking frequencies in a pattern similar to that for MBF_NET. Only at a power output of 9 W was there a difference (P < 0.05) between 40 and 80 cpm. End-exercise values for MBF_ANT were not significantly different among exercise protocols.

Retrograde blood flow (MBF_RET). At rest, ~40% of the forward blood flow in the femoral artery (MBF_RET) "rebounded" during each cardiac cycle, resulting in significant retrograde blood flow (MBF_RET) within each cardiac cycle (Fig. 3C). Resting MBF_RET was not significantly different among kicking protocols. At the lightest work intensity (3 W), there was a significant reduction in MBF_RET compared with rest for all of the kicking frequencies. Above 3 W, MBF_RET increased (P < 0.05) for both 60 and 80 cpm but remained constant for the 40-cpm trial. Figure 3C shows that, at >5 W, MBF_RET was less (P < 0.05) in 40 compared with 80 cpm and that, at >7 W, MBF_RET was also less (P < 0.05) in 40 than 60 cpm.

HR and blood pressure. HR increased from resting values to similar peak values at exhaustion (Table 2). For all submaximal work rates, HR at 80 cpm was greater than at 40 cpm (P < 0.05), although this difference was <10 beats/min on average. At a power output >5 W, HR was higher (P < 0.05) during exercise at 60 than at 40 cpm.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The present study was designed to quantify the effect of contraction frequency and muscle tension on leg blood flow in humans during exercise in the supine position. Essential to the interpretation of our findings was our ability to alter the relaxation duration and resulting duty cycle for each of the kicking protocols. This was successfully accomplished in each subject by maintenance of a constant contraction duration while kicking frequency was varied. Consistent with our hypothesis, we found that, at the same power output, an increase in kicking frequency was associated with an attenuated increase in the MBF response. Moreover, blood flow for each of the kicking frequencies did not rise linearly as a function of power output but rather showed a blunted response at higher power outputs.

Limitations. One assumption that we made in the present experiment was that resting femoral artery diameters were representative of exercise values. This assumption was based on preliminary studies performed in our laboratory (12) that showed that the femoral artery diameter did not significantly change from rest to exercise during knee extension exercise identical to that performed in the present study. The observation of a constant femoral artery diameter for rest and upright knee extension exercise was also reported by Rådegran (22). In contrast, MacDonald et al. (17) have shown an increase in femoral artery diameter during the rest-to-exercise transition when knee extension exercise was performed in the supine but not the upright position. If our assumption that femoral artery diameter remained unchanged from rest values was incorrect, then, based on the change in femoral artery diameter reported by MacDonald et al., we may have underestimated MBF by up to 16% in the present study. However, we are not aware of any reports that would suggest that a significant difference in conduit artery diameter may be expected simply by changing kicking frequency while maintaining the same power output and, therefore, metabolic requirements. Indeed, the difference in MBF between 40 and 80 cpm cannot be explained by assuming that an increase in arterial diameter occurred during exercise at 80 but not 40 cpm. The diameter of the femoral artery will be determined, in part, by the distending pressure. An increase in distending pressure, determined as the MAP, is associated with an increase in the femoral artery diameter (30). Blood pressure was recorded noninvasively by means of arterial tonometry in the present study, but data for only five subjects were reliable for 40 and 80 cpm because of the sensitivity of the blood pressure monitoring system to slight movements of the arm during exercise. Results from these five subjects did indicate a significant increase in MAP with increasing power output, but no difference was observed among contraction frequencies. This suggests that femoral artery distending pressure, and thus diameter, was not significantly different among kicking conditions.

The response of MBF_NET during incremental knee extension exercise. At rest, MBF_NET was ~370 ml/min, similar to values previously reported by others using Doppler (22, 23, 35) and thermodilution techniques (1). With exercise, MBF_NET increased to over three times the resting value by the end of the lightest workload (~3 W). Surprisingly, MBF_NET did not increase linearly with further increases in power output as expected; we have confirmed this observation in a single subject using the same exercise protocol but using the thermodilution technique developed by Andersen and Saltin (1) to measure blood flow (Fig. 4). This finding is in disagreement with earlier reports, which showed a linear rise in leg blood flow with increases in power output to maximal exercise (1, 25, 29). This discrepancy may be explained, in part, by differences in the exercise protocols and/or knee extension ergometer used. Knee extension exercise has typically been performed using the Krogh style of ergometer in which the knee is able to move through a full range of motion (i.e., 90-170°) (1, 25, 29). In the present study, the range of motion for the lower limb was ~25% of that of the Krogh ergometer (~90-110°). This implies that, to generate the same power output (at a given kicking frequency), the muscle tension would have to be on the order of four times greater on the present ergometer compared with the muscle tension developed on the Krogh-style ergometer.

View larger version (13K): [in this window] [in a new window]   Fig. 4.   The response of an individual subject using the thermodilution technique to measure muscle blood flow compared with the group mean response using the Doppler ultrasound technique. The knee extension exercise was performed on the same ergometer and at the same contraction frequency of 40 cpm. It is readily apparent that the blood flow responses are both qualitatively and quantitatively similar and independent of the methods used to determine blood flow.

The effect of contraction frequency on muscle blood flow. In partial agreement with our hypothesis, the increase in MBF_NET was greater during 40 than 80 cpm at power outputs >5 W, although no difference was observed among contraction frequencies at exhaustion. Previous studies have demonstrated that the net blood flow per contraction is dependent on both the relaxation and contraction duration (4, 11, 35). In the present study, the highest MBF_NET was observed at the lowest contraction frequency, which was also associated with the longest duration between muscle contractions. At higher frequencies (80 cpm), blood flow was lower than at 40 cpm and did not increase further with increasing power outputs >5 W, possibly because of the decrease in relaxation duration. The effect of reduced relaxation duration on blood flow is further depicted in Fig. 5. When muscle blood flow was expressed as the amount of flow per contraction (i.e., ml/contraction), MBF_NET during 80 cpm was less than one-half of that for 40 cpm. Interestingly, this difference in MBF_NET was associated with a lower MBF_ANT during 80 compared with 40 cpm because the amount of retrograde blood flow per contraction was similar across contraction frequencies (Fig. 5). This observation is consistent with the hypothesis that the time allowed between contractions may determine, in part, the overall perfusion of the exercising muscle.

View larger version (23K): [in this window] [in a new window]   Fig. 5.   The group MBFNET (A), MBFANT (B), and MBFRET (C) blood flow response expressed in milliliters per contraction during incremental knee extension exercise to exhaustion for 40, 60, and 80 cpm. * Significantly different from corresponding value at 80 cpm, P < 0.05. + Significantly different from corresponding value at 60 cpm, P < 0.05. Note that statistics were only performed on the first 4 power outputs and at exhaustion, where all 9 subjects contributed data to the group mean.

View larger version (18K): [in this window] [in a new window]   Fig. 6.   MBFRET expressed in both flow per unit of time (ml/min; A) and the amount of blood flow per muscular contraction-relaxation cycle (ml/contraction; B) as a function of tension developed by the muscle (i.e., same psi) for each of the exercise bouts.

The effect of kicking frequency on end-exercise blood flow responses. Previous studies have shown that increases in contraction frequency are associated with a decrease in the time to exhaustion or the total number of contractions completed during an exercise bout (3, 27). In the present study, power output was matched at each submaximal exercise stage for all three kicking frequencies. The highest kicking frequency was associated with a lower end-exercise power output, consistent with the findings of earlier studies. Although the difference in end-exercise MBF_NET did not reach statistical significance because of large inter- and intrasubject variability, MBF_NET was ~560 ml/min lower during exercise at 80 compared with 40 cpm, suggesting that MBF_NET may have been limited at the higher contraction frequencies and, as a consequence, may have contributed to the earlier fatigue.