Distribution Of Muscle Blood Flow Research Articles

Respiratory and locomotor muscle blood flow measurements using near-infrared spectroscopy and indocyanine green dye in health and disease.

Measuring respiratory and locomotor muscle blood flow during exercise is pivotal for understanding the factors limiting exercise tolerance in health and disease. Traditional methods to measure muscle blood flow present limitations for exercise testing. This article reviews a method utilising near-infrared spectroscopy (NIRS) in combination with the light-absorbing tracer indocyanine green dye (ICG) to simultaneously assess respiratory and locomotor muscle blood flow during exercise in health and disease. NIRS provides high spatiotemporal resolution and can detect chromophore concentrations. Intravenously administered ICG binds to albumin and undergoes rapid metabolism, making it suitable for repeated measurements. NIRS-ICG allows calculation of local muscle blood flow based on the rate of ICG accumulation in the muscle over time. Studies presented in this review provide evidence of the technical and clinical validity of the NIRS-ICG method in quantifying respiratory and locomotor muscle blood flow. Over the past decade, use of this method during exercise has provided insights into respiratory and locomotor muscle blood flow competition theory and the effect of ergogenic aids and pharmacological agents on local muscle blood flow distribution in COPD. Originally, arterial blood sampling was required via a photodensitometer, though the method has subsequently been adapted to provide a local muscle blood flow index using venous cannulation. In summary, the significance of the NIRS-ICG method is that it provides a minimally invasive tool to simultaneously assess respiratory and locomotor muscle blood flow at rest and during exercise in health and disease to better appreciate the impact of ergogenic aids or pharmacological treatments.

Heterogeneous ageing of skeletal muscle microvascular function.

The distribution of blood flow to skeletal muscle during exercise is altered with advancing age. Changes in arteriolar function that are muscle specific underlie age-induced changes in blood flow distribution. With advancing age, functional adaptations that occur in resistance arterioles from oxidative muscles differ from those that occur in glycolytic muscles. Age-related adaptations of morphology, as well as changes in both endothelial and vascular smooth muscle signalling, differ in muscle of diverse fibre type. Age-induced endothelial dysfunction has been reported in most skeletal muscle arterioles; however, unique alterations in signalling contribute to the dysfunction in arterioles from oxidative muscles as compared with those from glycolytic muscles. In resistance arterioles from oxidative muscle, loss of nitric oxide signalling contributes significantly to endothelial dysfunction, whereas in resistance arterioles from glycolytic muscle, alterations in both nitric oxide and prostanoid signalling underlie endothelial dysfunction. Similarly, adaptations of the vascular smooth muscle that occur with advancing age are heterogeneous between arterioles from oxidative and glycolytic muscles. In both oxidative and glycolytic muscle, late-life exercise training reverses age-related microvascular dysfunction, and exercise training appears to be particularly effective in reversing endothelial dysfunction. Patterns of microvascular ageing that develop among muscles of diverse fibre type and function may be attributable to changing patterns of physical activity with ageing. Importantly, aerobic exercise training, initiated even at an advanced age, restores muscle blood flow distribution patterns and vascular function in old animals to those seen in their young counterparts.

Functional sympatholysis in hypertension.

Sympathetic vasoconstriction is normally attenuated in exercising muscle by local changes in muscle metabolites and other substances that reduce vascular responsiveness to α-adrenergic receptor activation. Termed functional sympatholysis, this protective mechanism is thought to optimize muscle blood flow distribution to match perfusion with metabolic demand. Emerging evidence from both animal and human studies indicate that functional sympatholysis is impaired in hypertension and may constitute an important underlying cause of skeletal muscle malperfusion during exercise in this common cardiovascular condition. Findings from studies of animal models of hypertension and patients with essential hypertension will be integrated in this review to provide insight into the underlying mechanisms responsible for inappropriate sympathetic vasoconstriction in exercising muscle and the treatment options that may restore functional sympatholysis and improve muscle perfusion during exercise.

Acute ascorbic acid and hindlimb skeletal muscle blood flow distribution in old rats: rest and exercise

Excess reactive oxygen species are implicated in the impaired peripheral vascular function evident during exercise in older individuals. We tested the hypothesis that an acute infusion of the antioxidant ascorbic acid (AA) in old rats would improve antioxidant capacity and reduce oxidative stress and, therefore, elevate hindlimb muscle blood flow at rest and during treadmill exercise in muscles containing principally type I and IIa muscle fibers. Total and individual hindlimb skeletal muscle blood flow was measured (radiolabeled microspheres) in old rats (26-28 months) at rest (n = 8) and during treadmill exercise (n = 8; 20 m·min⁻¹, 5% grade) before and after AA treatment (76 mg·(kg body mass)⁻¹ intra-arterial (i.a.) injection). AA elevated total antioxidant capacity (rest, ~37%; and exercise, 31%) and reduced oxidative stress (~26%, exercise only). AA reduced resting total hindlimb muscle blood flow (control, 25 ± 3; AA, 16 ± 2 mL·min⁻¹·(100 g)⁻¹; p < 0.05) and blood flow to 8 of 28 individual muscles with no fiber-type correlation (p > 0.05). During exercise there was no effect of AA on total hindlimb muscle blood flow (control, 154 ± 14; AA, 162 ± 13 mL·min⁻¹·(100 g)⁻¹; p > 0.05) or blood flow to any individual muscle. This disconnect between whole-body antioxidant status and skeletal muscle blood flow in old rats mandates consideration when pursuing antioxidant treatments experimentally or clinically in older populations.

The power–duration relationship of high‐intensity exercise: from mathematical parameters to physiological mechanisms

The power–duration relationship of high‐intensity exercise: from mathematical parameters to physiological mechanisms

Comfortable at just below your critical speed: how is blood flow distribution coupled to muscle fibre recruitment during exercise?

Comfortable at just below your critical speed: how is blood flow distribution coupled to muscle fibre recruitment during exercise?

Critical speed in the rat: implications for hindlimb muscle blood flow distribution and fibre recruitment

Critical speed (CS) constitutes an important metabolic and performance demarcator. However, active skeletal muscle blood flow distribution specifically surrounding CS remains unknown. We tested the hypotheses that CS could be accurately determined in the running rat and that measurement of hindlimb inter- and intramuscular blood flow below and above CS would support that the greatest muscle fibre recruitment above, relative to below, CS occurs in the predominantly glycolytic muscles. Seven male Sprague-Dawley rats performed five constant-speed tests to exhaustion at speeds between 95 and 115% of the speed that elicited to determine CS. Subsequent constant-speed tests were performed at speeds incrementally surrounding CS to determine time to exhaustion, V(O2), and hindlimb muscle blood flow distribution. Speed and time to exhaustion conformed to a hyperbolic relationship (r(2) = 0.92 ± 0.03) which corresponded to a linear 1/time function (r(2) = 0.93 ± 0.02) with a CS of 48.6 ± 1.0 m min(-1). Time to exhaustion below CS was ∼ 5× greater (P < 0.01) than that above. Below CS V(O2) stabilized at a submaximal value (58.5 ± 2.5 ml kg(-1) min(-1)) whereas above CS (81.7 ± 2.5 ml kg(-1) min(-1)) increased to (84.0 ± 1.8 ml kg(-1) min(-1), P > 0.05 vs. above CS). The 11 individual muscles or muscle parts that evidenced the greatest blood flow increases above, relative to below, CS were composed of ≥ 69% Type IIb/d/x muscle fibres. Moreover, there was a significant correlation (P < 0.05, r = 0.42) between the increased blood flow above expressed relative to below CS and the percentage Type IIb/d/x fibres found in the individual muscles or muscle parts. These data validate the powerful CS construct in the rat and identify that running above CS, relative to below CS, incurs disproportionate blood flow increases (indicative of recruitment) in predominantly highly glycolytic muscle fibres.

Regional heterogeneity of α-adrenoreceptor subtypes in arteriolar networks of mouse skeletal muscle

Activation of vascular adrenoreceptors (ARs) governs the magnitude and distribution of muscle blood flow in accord with the distribution of AR subtypes. Functional studies in the rat cremaster muscle indicate that α1ARs predominate in proximal arterioles (first-order, 1A) while α2ARs predominate in distal arterioles (third-order, 3A). However, little is known of AR subtype distribution in arteriolar networks of locomotor skeletal muscles, particularly in the mouse. We tested the hypotheses that functional AR subtypes exhibit heterogeneity among branches of arteriolar networks in a locomotor muscle and that the nature of this heterogeneity can vary between muscles having diverse functions. In anaesthetized male C57BL/6J mice (3 months old), concentration-response curves (10(-9) m to 10(-5) m, 0.5 log increments) were evaluated in the gluteus maximus muscle superfused with physiological saline solution (35°C, pH 7.4; n ≥ 5 per group). Noradrenaline (NA, non-selective αAR agonist) constricted 1A, 2A and 3A with similar potency and efficacy. Phenylephrine (PE; α1AR agonist) evoked greater (P < 0.05) constriction in 3A (inhibited by 10(-8) m prazosin; α1AR antagonist) while UK 14304 (UK; α2AR agonist) evoked greater (P < 0.05) constriction in 1A (inhibited by 10(-7) m rauwolscine; α2AR antagonist). Isoproterenol (isoprenaline; βAR agonist) dilated 1A, 2A and 3A near-maximally with similar potency and efficacy; these dilatations were inhibited by 10(-7) m propranolol (βAR antagonist) which otherwise had no effect on responses to NA, PE, or UK. Complementary experiments in the mouse cremaster muscle revealed a pattern of αAR subtype distribution that, while distinct from the gluteus maximus muscle, was consistent with that reported for the rat cremaster muscle. We conclude that functional αAR subtype distribution in arteriolar networks of skeletal muscle varies with muscle function as well as vessel branch order.

Speeding of V̇o2 kinetics with endurance training in old and young men is associated with improved matching of local O2 delivery to muscle O2 utilization

The time course and mechanisms of adjustment of pulmonary oxygen uptake (V(O(2))) kinetics (time constant tauV(O(2p))) were examined during step transitions from 20 W to moderate-intensity cycling in eight older men (O; 68 +/- 7 yr) and eight young men (Y; 23 +/- 5 yr) before training and at 3, 6, 9, and 12 wk of endurance training. V(O(2p)) was measured breath by breath with a volume turbine and a mass spectrometer. Changes in deoxygenated hemoglobin concentration (Delta[HHb]) were measured by near-infrared spectroscopy. V(O(2p)) and Delta[HHb] were modeled with a monoexponential model. Training was performed on a cycle ergometer three times per week for 45 min at approximately 70% of peak V(O(2)). Pretraining tauV(O(2p)) was greater (P < 0.05) in O (43 +/- 10 s) than Y (34 +/- 8 s). tauV(O(2p)) decreased (P < 0.05) by 3 wk of training in both O (35 +/- 9 s) and Y (22 +/- 8 s), with no further changes thereafter. The pretraining overall adjustment of Delta[HHb] was faster than tauV(O(2p)) in both O and Y, resulting in Delta[HHb]/V(O(2p)) displaying an "overshoot" during the transient relative to the subsequent steady-state level. After 3 wk of training the Delta[HHb]/V(O(2p)) overshoot was attenuated in both O and Y. With further training, this overshoot persisted in O but was eliminated after 6 wk in Y. The training-induced speeding of V(O(2p)) kinetics in O and Y at 3 wk of training was associated with an improved matching of local O(2) delivery to muscle V(O(2)) (as represented by a lower Delta[HHb]/V(O(2p))). The continued overshoot in Delta[HHb]/V(O(2p)) in O may reflect a reduced vasodilatory responsiveness that may limit muscle blood flow distribution during the on-transient of exercise.

Changes in muscle blood flow distribution during hyperthermia

Blood flow is a critical parameter for obtaining satisfactory temperature distributions during clinical hyperthermia. This study examines the changes in blood flow distribution in normal porcine skeletal muscle before, during and after a period of regional microwave hyperthermia. The baseline blood flow distribution during general anaesthesia and after the insertion of the thermal probes was established independently in order to isolate the changes due to hyperthermia. General anaesthesia alone and thermocouple insertion during anesthesia had no significant effect on the muscle blood flow distribution. Regional microwave heating generated a non-uniform blood flow distribution which was a function of the tissue temperature distribution. Blood flow was greater in those tissues samples in which higher temperatures were recorded and less in those sampled further from the applicators peak SAR (Specific Absorption Rate). The increase in blood flow appears to be primarily a local phenomenon. Although muscle blood flow may be considered to be uniform prior to heating, this does not hold during hyperthermia treatment. Therefore, the non-uniform nature of the blood distribution during heating should be incorporated into any practical bioheat transfer model.

Blood flow distribution in working in situ canine muscle during blood flow reduction.

The purpose of this study was to determine whether reduction in apparent muscle O2 diffusing capacity (Dmo2) calculated during reduced blood flow conditions in maximally working muscle is a reflection of alterations in blood flow distribution. Isolated dog gastrocnemius muscle (n = 6) was stimulated for 3 min to achieve peak O2 uptake (VO2) at two levels of blood flow (controlled by pump perfusion): control (C) conditions at normal perfusion pressure (blood flow = 111 +/- 10 ml.100 g-1.min-1) and reduced blood flow treatment [ischemia (I); 52 +/- 6 ml.100 g-1.min-1]. In addition, maximal vasodilation was achieved by adenosine (A) infusion (10(-2)M) at both levels of blood flow, so that each muscle was subjected randomly to a total of four conditions (C, CA, I, and IA; each separated by 45 min of rest). Muscle blood flow distribution was measured with 15-microns-diameter colored microspheres. A numerical integration technique was used to calculate Dmo2 for each treatment with use of a model that calculates O2 loss along a capillary on the basis of Fick's law of diffusion. Peak VO2 was reduced significantly (P < 0.01) with ischemia and was unchanged by adenosine infusion at either flow rate (10.6 +/- 0.9, 9.7 +/- 1.0, 6.7 +/- 0.2, and 5.9 +/- 0.8 ml.100 g-1.min-1 for C, CA, I, and IA, respectively). Dmo2 was significantly lower by 30-35% (P < 0.01) when flow was reduced (except for CA vs. I; 0.23 +/- 0.03, 0.20 +/- 0.02, 0.16 +/- 0.01, and 0.13 +/- 0.01 ml.100 g-1.min-1.Torr-1 for C, CA, I, and IA, respectively). As expressed by the coefficient of variation (0.45 +/- 0.04, 0.47 +/- 0.04, 0.55 +/- 0.03, and 0.53 +/- 0.04 for C, CA, I, and IA, respectively), blood flow heterogeneity per se was not significantly different among the four conditions when examined by analysis of variance. However, there was a strong negative correlation (r = 0.89, P < 0.05) between Dmo2 and blood flow heterogeneity among the four conditions, suggesting that blood flow redistribution (likely a result of a decrease in the number of perfused capillaries) becomes an increasingly important factor in the determination of Dmo2 as blood flow is diminished.

Blood flow and pressure relationships which determine VO2max

The role of O2 delivery in regulating VO2max has been studied in an isolated gastrocnemius-plantaris muscle preparation contracting in situ; recent data addressing this issue are presented. VO2 increases nonlinearly with stimulation frequency reaching a peak at 5 twitches.s-1 or 1 tet.s-1 (200 ms trains, 50 imp.s-1). Further increases in stimulation frequency result in a lower VO2. Measured VO2 maxima are less than predicted VO2 capacity, and peak VO2 during tetanic contractions is greater than that during twitches. Above 150 imp.min-1, VO2 is directly related to the level of blood flow attained as VO2/Q (arterial-venous O2 difference) is fixed by some unknown mechanism. Increasing blood flow, with a pump, during 1.s-1 tetanic contractions increases O2 diffusive conductance and peak VO2. When O2 delivery is reduced, ischemic hypoxia appears to result in more rapid reductions in muscle performance than hypoxic hypoxia because of decreases in perfusion pressure and Q. 31P-NMR studies reveal that reductions in creatine phosphate and energy charge are similar between ischemia and hypoxia suggesting a common regulator, O2. We conclude that VO2max is limited by O2 delivery as a result of a limited and uneven distribution of muscle blood flow. These limitations appear secondary to mechanical restraints imposed by contraction duty cycle and vascular compression.

The rate and distribution of muscle blood flow after prolonged ischemia

The magnitude and distribution of muscle blood flow in the lower extremity after relief of an acute arterial occlusion may influence the extent of the resulting necrosis. The object of this study was to document the distribution of blood flow in the resting state and after prolonged periods of complete ischemia, and to assess the relationship between the degree of reactive hyperemia and subsequent necrosis. The isolated bilateral canine gracilis muscle preparation that we have previously characterized was used for microsphere studies. Total blood flow was measured by means of timed venous collections, and the distribution of flow was determined by means of a multiple microsphere injection technique. Measurements of microsphere distribution and blood flow were made before ischemia and during the initial 48 minutes of reperfusion after both 4 and 5 hours of normothermic ischemia, which resulted in 46.7% ± 6% and 71.2% ± 7% necrosis, respectively. The muscle was harvested and sectioned transversely into six slices approximately 1.5 cm thick, and the extent of necrosis was quantified by means of nitroblue tetrazolium staining 48 hours after reperfusion. Blood flow distribution during the early reperfusion phase was determined in each muscle slice and in both the alive and dead portions of each slice by use of the microsphere injection technique. Preischemic blood flow was distributed homogeneously throughout the muscle and was 4.5 ± 0.8 ml/100 gm/min (mean ± SEM, n = 8). On reperfusion total flow was 6 to 10 times higher than it was before ischemia and was distributed predominantly to the middle slices (p < 0.05, n = 12). In both the 4- and 5-hour ischemia groups there was no correlation between blood flow in each slice and the extent of necrosis (p > 0.1). The degree of reactive hyperemia was greater after 4 hours of ischemia than after 5 hours (20.3 ± 2.5 ml/100 gm/min vs 12.5 ± 1.6 ml/100 gm/min, p < 0.05). On reperfusion, blood flow is not distributed homogeneously throughout the muscle and the extent of reactive hyperemia is inversely related to the length of ischemia. Hyperemic blood flow reaches the maximum rate in the center portions of the muscle belly closest to the inflow and does not seem to be related to the distribution of necrosis.

The rate and distribution of muscle blood flow after prolonged ischemia

The rate and distribution of muscle blood flow after prolonged ischemia

Spatial and temporal variability of blood flow in stimulated dog gastrocnemius muscle.

The distribution of blood flow in skeletal muscle stimulated to rhythmic isotonic contractions was studied by injections of radioactive microspheres into the arterial supply of gastrocnemius muscles (mean weight 88 g) subsequently cut into 0.5 g pieces for determination of radioactivity. The coefficient of variation (CV = SD/mean) of the ratio of simultaneously injected 10 microns and 15 microns microspheres, 0.12, was taken as the inherent scatter of the method. The average spatial distribution inequality of 10-15 microns microspheres corresponded to a CV of 0.45 and the specific local blood flow inhomogeneity to a CV = 0.43 (= square root of 0.45(2) - 0.12(2)), but there were marked differences between muscles. The temporal variability of blood flow in individual muscle pieces was obtained from the comparison of fractional trapping of 4 to 5 differently labeled microspheres injected at intervals of 2 minutes into steadily stimulated muscles. The mean CV for the variations in time was 0.23 and that corrected for methodological scatter, 0.19. There were large differences between muscle pieces within a muscle and between muscles. The presence of considerable spatial and temporal variations of blood flow in exercising muscle during apparent steady state may be important in limiting and/or modulating tissue O2 supply.

Magnitude and distribution of muscle blood flow in conscious animals during locomotory exercise.

A large number of papers in the literature have reported peak blood flows in electrically stimulated isolated skeletal muscles in the range of 40-140 ml.min-1.100 g-1. Peak muscle blood flows of similar magnitude have been measured in human subjects with 133Xe clearance and plethysmography. In contrast, measurements of average blood flow in conscious animals at VO2max with radiolabeled microspheres have ranged from about 150-250 ml.min-1.100 g-1. Furthermore, flows in active high oxidative muscles in conscious animals during maximal exercise may attain 500 ml.min-1.100 g-1. Reasons for these apparent discrepancies are unknown, but possible explanations including technical artifacts and differences in mechanical influences on contracting muscles in isolated preparations and exercising conscious animals are discussed. It is concluded that the primary reason for the differences in magnitude of muscle blood flow is the efficient muscle pumping action that occurs in vivo during locomotory exercise.

Sympathetic neural influences on muscle blood flow in rats during submaximal exercise.

These experiments were designed to estimate the involvement of the sympathetic innervation in regulation of hindlimb muscle blood flow distribution among and within muscles during submaximal locomotory exercise in rats. Blood flows to 32 hindlimb muscles and 13 other selected tissues were measured using the radiolabeled microsphere technique, before exercise and at 0.5, 2, 5, and 15 min of treadmill exercise at 15 m/min. The two groups of rats studied were 1) intact control, and 2) acutely sympathectomized (hindlimb sympathectomy accomplished by bilateral section of the lumbar sympathetic chain and its connections to the spinal cord at L2-L3). There were no differences in total hindlimb muscle blood flow among the two groups during preexercise or at 30 s or 2 min of exercise. However, flow was higher in eight individual muscles at 2 min of exercise in the sympathectomized rats. At 5 and 15 min of exercise there was higher total hindlimb muscle blood flow in the denervated group compared with control. These differences were also present in many individual muscles. Our results suggest that 1) sympathetic nerves do not exert a net influence on the initial elevations in muscle blood flow at the beginning of exercise, 2) sympathetic nerves are involved in regulating muscle blood flow during steady-state submaximal exercise in conscious rats, and 3) these changes are seen in muscles of all fiber types.

Regional muscle blood flow capacity and exercise hyperemia in high-intensity trained rats.

The purpose of this study was to determine the effects of high-intensity treadmill exercise training on 1) the regional distribution of muscle blood flow within and among muscles in rats during high-intensity treadmill exercise (phase I) and 2) on the total and regional hindlimb skeletal muscle blood flow capacities as measured in isolated perfused rat hindquarters during maximal papaverine vasodilation (phase II). Two groups of male Sprague-Dawley rats were trained 5 days/wk for 6 wk with a program consisting of 6 bouts/day of 2.5-min runs at 60 m/min up a 15% grade with 4.5-min rest periods between bouts. After training, blood flows were measured with the radiolabeled microsphere technique (phase I) in pair-weighted sedentary control and exercise-trained rats while they ran at 60 m/min (0% grade). In phase II of the study, regional vascular flow capacities were determined at three perfusion pressures (30, 40, and 50 mmHg) in isolated perfused hindquarters of control and trained rats maximally vasodilated with papaverine. The results indicate that this exercise training program produces increases in the vascular flow capacity of fast-twitch glycolytic muscle tissue of rats. However, these changes were not apparent in the magnitude or distribution of muscle blood flow in conscious rats running at 60 m/min, since blood flows within and among muscles during exercise were the same in trained and control rats.

Progressive elevations in muscle blood flow during prolonged exercise in swine.

Distribution of muscle blood flow has not been measured in man during prolonged exercise, but progressive elevations in skin flow coupled with constant cardiac output (QT) have suggested muscle blood flow may be compromised. However, previous experiments with rats demonstrated progressive increases in muscle blood flow over time during prolonged submaximal exercise. The present study was performed to study muscle blood flow in miniature swine during long-term exercise to shed light on this apparent anomaly. QT and distribution of QT were studied with radiolabeled microspheres while pigs ran on a level treadmill at a speed (10.5 km/h) requiring 71 +/- 4% of maximal O2 consumption (VO2 max). QT increased 23% from the 5th to the 30th min of exercise, whereas total skeletal muscle flow increased by 49%. Increases in flow in the muscles resulted from decreased resistance, since mean arterial pressure declined over this time period (-7%). In addition, the proportional increases in muscle flow were similar within synergistic muscle groups independent of fiber type composition (e.g., elbow extensors: 59-78%; elbow flexors: 26-40%). The factor that limited continued exercise appeared to be body temperature. Colonic temperature rose in linear fashion over time; the animals became exhausted at approximately 42 degrees C. These flow data are similar to previous findings in rats and indicate that during prolonged treadmill locomotion in quadrupedal animals muscle blood flow increases over time to near maximal levels.

Influence of thigh blood flow upon the arterial pressure gradient over the collateral arteries in patients with occlusion of the superficial femoral artery.

The effect of circulatory arrest of the calf and foot and head-up tilt upon pressure gradient over the thigh and distribution of muscle blood flow in the thigh was studied in 14 patients with occlusion of the superficial femoral artery. In the supine position circulatory arrest just below the knee initially reduced the directly measured mean arterial pressure gradient over the thigh by about 30% indicating that a similar proportion of total leg blood flow supplied the calf and foot. Local muscle blood flow was measured by the local 133Xe washout technique proximally and distally in the thigh muscle. Total leg blood flow was measured by an indicator dilution technique. During 10 min of circulatory arrest of the calf and foot local muscle blood flow in the thigh increased by approximately 35% proximally and distally. Total leg blood flow remained unchanged and the pressure gradient was reduced by about 10% as compared with that before tourniquet. During head-up tilt local muscle blood flow in the thigh decreased by about 30% proximally and distally. Total leg blood flow also decreased by about 30%. Circulatory arrest below the knee did not alter this response. These findings indicate that blood flow change uniformly in the thigh during head-up tilt and during tourniquet on the calf. Furthermore blood flow to the calf appears only to account for 30% of the arterial pressure drop over the collateral arteries in the thigh, indicating that thigh blood flow is of major importance for the pressure gradient over the collateral arteries.