Regulation of human skeletal muscle perfusion and its heterogeneity ...

Am J Physiol Regul Integr Comp Physiol 299: R72–R79, 2010. First published April 28, 2010; doi:10.1152/ajpregu.00056.2010.

Regulation of human skeletal muscle perfusion and its heterogeneity during exercise in moderate hypoxia Ilkka H. Heinonen,1,2 Jukka Kemppainen,1,2 Kimmo Kaskinoro,1,3 Juha E. Peltonen,6 Ronald Borra,1,5 Markus Lindroos,1 Vesa Oikonen,1 Pirjo Nuutila,1,4 Juhani Knuuti,1 Robert Boushel,7 and Kari K. Kalliokoski1 1

Turku PET, Departments of 2Clinical Physiology and Nuclear Medicine, 3Anesthesiology and Intensive Care, 4Medicine, and Radiology, Turku University Hospital, University of Turku, Turku, Finland; 6Unit for Sports and Exercise Medicine, Institute of Clinical Medicine, University of Helsinki, Helsinki, Finland; and 7Centre for Healthy Aging, Department of Biomedical Sciences, University of Copenhagen, Copenhagen, Denmark 5

Submitted 22 January 2010; accepted in final form 19 April 2010

Heinonen IH, Kemppainen J, Kaskinoro K, Peltonen JE, Borra R, Lindroos M, Oikonen V, Nuutila P, Knuuti J, Boushel R, Kalliokoski KK. Regulation of human skeletal muscle perfusion and its heterogeneity during exercise in moderate hypoxia. Am J Physiol Regul Integr Comp Physiol 299: R72–R79, 2010. First published April 28, 2010; doi:10.1152/ajpregu.00056.2010.—Although many effects of both acute and chronic hypoxia on the circulation are well characterized, the distribution and regulation of blood flow (BF) heterogeneity in skeletal muscle during systemic hypoxia is not well understood in humans. We measured muscle BF within the thigh muscles of nine healthy young men using positron emission tomography during one-leg dynamic knee extension exercise in normoxia and moderate physiological systemic hypoxia (14% O2 corresponding to ⬃3,400 m of altitude) without and with local adenosine receptor inhibition with femoral artery infusion of aminophylline. Systemic hypoxia reduced oxygen extraction of the limb but increased muscle BF, and this flow increment was confined solely to the exercising quadriceps femoris muscle. Exercising muscle BF heterogeneity was reduced from rest (P ⫽ 0.055) but was not affected by hypoxia. Adenosine receptor inhibition had no effect on capillary BF during exercise in either normoxia or hypoxia. Finally, one-leg exercise increased muscle BF heterogeneity both in the resting posterior hamstring part of the exercising leg and in the resting contralateral leg, whereas mean BF was unchanged. In conclusion, the results show that increased BF during one-leg exercise in moderate hypoxia is confined only to the contracting muscles, and the working muscle hyperemia appears not to be directly mediated by adenosine. Increased flow heterogeneity in noncontracting muscles likely reflects sympathetic nervous constraints to curtail BF increments in areas other than working skeletal muscles, but this effect is not potentiated in moderate systemic hypoxia during small muscle mass exercise. positron emission tomography; exercise; hypoxia

reported that, in addition to the importance of capillary density for tissue oxygenation, it is not sufficient to simply supply an adequate amount of oxygen to the organ as a whole, but oxygen has to be distributed within the organ precisely where it is needed (27, 28). Thus it must be appreciated that not only bulk blood flow but especially its appropriate distribution between and within working skeletal muscles is of importance for precise matching of oxygen supply and metabolism (11), especially during exercise. In this regard, we recently reported that blood flow heterogeneity

IN 1919, AUGUST KROGH

Address for reprint requests and other correspondence: I. Heinonen, Turku PET Centre, P.O. Box 52, FIN-20521 Turku, Finland (e-mail: ilkka.heinonen @tyks.fi). R72

decreases with increasing exercise intensity within contracting quadriceps femoris (QF) muscle (21). It is, however, not known how blood flow heterogeneity within active and nonactive thigh muscles changes when metabolism is challenged by reduced oxygen content of the blood, as is the case during exercise in systemic hypoxia. Although many effects of both acute and chronic hypoxia on the circulation are well characterized, relatively little is known about how muscle blood flow is regulated in acute hypoxia in humans. In the working limb, mean blood flow is usually enhanced with systemic hypoxia (10, 26), yet vasoconstrictor activity is preserved or enhanced (6, 43). In nonworking muscles, it is generally considered that already during normoxic exercise increasing sympathetic nervous activation prevents increases in muscle blood flow, and, thus, cardiac output is redirected effectively to active skeletal muscles. Sympathetic restraint of muscle blood flow is augmented during hypoxic exercise (43), but this pattern has been studied only sparsely in humans. Only a few previous human studies have addressed the effect of increased sympathetic activation in acute and chronic hypoxia on whole limb blood flow (6, 30), but flow distribution and its regulation specifically within muscle has never been determined. Metabolic vasodilation is considered an important signaling mechanism controlling muscle blood flow in exercising muscle in both normoxia and hypoxia. Endogenous adenosine can be one of the signals mediating metabolic vasodilation in skeletal muscle. The adenosine hypothesis proposes that adenosine functions as a distress signal to match oxygen supply to cellular metabolism in hypoxic and/or ischemic tissue (3). Evidence to support that increased blood flow in the exercising limb under moderate hypoxia is indeed the result of enhanced metabolic signals is inferred, since symphatholysis does not contribute to the enhanced blood flow in hypoxic forearm exercise (47). A major role of adenosine for enhancing limb blood flow during hypoxic forearm exercise was not substantiated by a recent study, which found no difference between flow in normoxia and hypoxia with adenosine receptor blockade (8). To the best of our knowledge, the contribution of adenosine to the muscle nutritive blood flow response during exercise with decreased arterial oxygen supply in hypoxia has never been studied in humans. Taken together, there were several purposes in the present study. First, we aimed to elucidate the patterns of muscle blood flow redistribution in active and nonactive muscles during dynamic one-leg knee extension exercise during physiological

0363-6119/10 Copyright © 2010 the American Physiological Society

http://www.ajpregu.org

SKELETAL MUSCLE BLOOD FLOW REGULATION

R73

systemic hypoxia corresponding to ⬃3,400 meters of altitude (inspired O2 14%). Second, we wanted to study whether blood flow heterogeneity changes within the active and inactive muscles during exercise in hypoxia. Third, we studied whether adenosine is involved in the regulation of muscle capillary blood flow in hypoxic exercise by applying similar antagonism procedures as in several recent studies (8, 29, 32). We hypothesized that, in hypoxia, 1) mean blood flow would increase along with a decrease in flow heterogeneity in contracting muscles; 2) the opposite response would occur in inactive muscle; and 3) if adenosine indeed plays a role in matching oxygen supply to metabolic demands, capillary blood flow in exercising muscle would be reduced and flow heterogeneity increased during adenosine receptor blockade and the reduced blood flow would be compensated for by increased oxygen extraction to assure similar muscle oxygen consumption and work performance. METHODS

Subjects. Nine healthy fit young men (25 ⫾ 5 yr, 184 ⫾ 6 cm, 76 ⫾ 9 kg) volunteered to participate in the study. The purpose, nature, and potential risks of the study were explained to the subjects before they gave their written informed consent to participate. The subjects were requested to abstain from caffeine-containing beverages for at least 24 h before the experiments and to avoid strenuous exercise within 48 h before the study. The subjects were not taking any regular medication. The study was performed at least 3 h after the subjects had eaten a light breakfast. The study was performed according to the Declaration of Helsinki and was approved by the Ethical Committee of the Hospital District of South-Western Finland and the National Agency for Medicines. Study design. Skeletal muscle blood flow in the femoral region (Fig. 1) was measured using positron emission tomography (PET) with [15O]H2O, as described below. Muscle blood flow was measured first under normal resting conditions and then during systemic hypoxia (14% inspired O2 in N2; equivalent to altitude of ⬃3,400 m). After these measurements at rest, blood flow was measured during one-leg dynamic exercise (Fig. 1) in a counterbalanced setting without and with locally administered adenosine receptor antagonism by aminophylline with the subject breathing either normal room air or hypoxic gas. Additionally, radial artery and femoral vein blood samples for blood gas parameters were drawn for analysis in each occasion mentioned above. We recently reported muscle blood flow values at rest and during normoxic exercise (20). A portion of the data gathered from that study in the same subjects is included in the present paper. In this study, we present mostly new data on the effect of hypoxia and adenosine receptor blockade with aminophylline on muscle blood flow during exercise. Other procedures before and after PET measurements. Before the PET experiments, the antecubital vein was cannulated for tracer administration. For blood sampling, a radial artery cannula was placed under local anesthesia in the contralateral arm. Additionally, cannulas were placed under local anesthesia in the femoral artery and vein for local drug infusion (aminophylline) and blood sampling, respectively. Subjects were then moved to the PET scanner with the femoral region in the gantry, and the right leg was fastened to a house-designed dynamometer (Fig. 1B). Systemic mean arterial pressure (MAP) was measured (Omron, M5–1; Omron Healthcare Europe, Hoofddorf, The Netherlands), and local vascular resistance and conductance were calculated by dividing systemic MAP with blood flow and blood flow with MAP, respectively. Oxygen consumption of the muscle was calculated by Fick’s principle. To calculate exercising muscle oxygen consumption, the blood flow value from the QF muscle was used. This

Fig. 1. Positron emission tomography imaging. A: the subject was positioned in the gantry so that the middle thigh was in the middle of the imaging area (12 cm). B and C: one-legged dynamic exercise was performed in the range of almost 90 degrees. Dynamic muscular work for the quadriceps femoris (QF) muscle consisted of a lifting phase and subsequent braking phase back to the starting position. D: regions of interest for blood flow analysis were drawn to represent whole thigh muscle, QF muscle, and the posterior muscle of the thigh. At rest, thigh muscle blood flow was fairly uniform (D), but, during exercise, the increment of blood flow was solely concentrated on exercising QF muscle (E). Color scaling for the respective blood flow values in images are provided on left. Thus the more green, yellow, or red is seen, the higher is the local blood flow. In general, increased heterogeneity in resting contralateral leg [left leg (L)] during one-leg exercise is readily seen in E. R, right leg.

was reasoned to be the best estimate for oxygen consumption, since it is assumed that the knee extensions are performed almost solely by QF and that oxygen extraction, thus the arterial-venous difference of oxygen, is higher in exercising QF than is the value obtained over the leg, recognizing that the latter value represents the oxygen extraction of the whole leg. Blood flow measurements and analysis. Radiowater positron-emitting tracer [15O]H2O was produced as previously described in detail (41), and the ECAT EXACT HR⫹ scanner (Siemens/CTI, Knoxville, TN) was used in three-dimensional mode for image acquisition to measure muscle blood flow. Photon attenuation was corrected by 5-min transmission scans performed both at the beginning of the

AJP-Regul Integr Comp Physiol • VOL

299 • JULY 2010 •

www.ajpregu.org

R74


resting and exercise PET studies. All data were corrected for dead time, decay, and measured photon attenuation, and the images were reconstructed into a 256 ⫻ 256 matrix, producing 2.57 ⫻ 2.57 mm in-plane dimensions of voxels with 2.43 mm plane thickness. For the measurement of blood flow at rest, scanning began 20 s after the bolus infusion of the tracer and during exercise simultaneously with the infusion, and scanning consisted of the following frames: 6 ⫻ 5, 12 ⫻ 10, and 7 ⫻ 30 s at rest and 6 ⫻ 5 and 12 ⫻ 10 s during exercise. During systemic hypoxia, breathing 14% oxygen gas began 5 min before imaging. During exercise, scanning commenced 3 min after exercise onset to obtain a metabolic steady-state situation and continued until the end of the exercise bout, which every time lasted 2.5 half-minutes, thus, altogether 5.5 min. Arterial blood radioactivity was also sampled continuously with a detector during imaging for blood flow quantification. Exercise consisted of dynamic one-leg exercise at 40 rpm with individually chosen workloads (4.3 ⫾ 2.1 kg) with a knee angle range of motion of ⬃70 – 80 degrees (Fig. 1B). During pretesting before the actual experiments, individually appropriate workloads for each subject were chosen so that they could maintain exercise for at least ⬃10 min without fatigue or discomfort. The data analysis was performed using the standard models (24) and methods (37). Heterogeneity of blood flow (relative dispersion) was calculated as the coefficient of variation (SD divided by mean blood flow) of voxel blood flow values within the region of interest. Blood flow and its heterogeneity were analyzed from QF muscle and posterior hamstring muscles of the exercising leg, and from whole thigh muscles of the nonworking leg (Fig. 1C). The measurement error of heterogeneity with PET (which depends mostly on the radioactivity level in the tissue, PET scanner, and reconstruction method that is used for image reconstruction) has been determined previously (25). The measurements were performed using uniformity phantom, a cylindrical tube containing a uniform amount of a tracer ([18F]FDG), scanned in the PET scanner. In the reconstructed images from that phantom, the heterogeneity (measurement error) was found to be ⬃7–10% at the corresponding tissue radioactivity levels as in the present study. Adenosine receptor blockade with aminophylline. To elucidate the role of adenosine for regulating muscle blood flow, similar procedures for adenosine receptor blockade with aminophylline were used as previously described (8, 32). In the study by Martin and colleagues (32), a 15% reduction in blood flow was observed in mild handgrip exercise, but Casey et al. (8) did not find evidence for adenosine

mediation explaining augmented exercise hyperemia in systemic hypoxia. Aminophylline (2 mg·min⫺1·l⫺1 thigh volume) was infused in the femoral artery to induce local competitive inhibition of endogenous adenosine receptor binding and to minimize any confounding systemic influence of the drug. To maximize its delivery to the site of action, the intra-arterial infusion of aminophylline was initiated simultaneously with one-leg exercise, thus 3 min before scanning, and continued until the end of the respective PET scan (8 min totally). The infusion of aminophylline was not possible to perform for two subjects for technical reasons, and the aminophylline results are therefore from seven subjects. Magnetic resonance imaging. Structural Magnetic Resonance Imaging was performed ⬃1 wk before the PET study to obtain an accurate volume measurement of the thigh, which was needed for calculation of the infused aminophylline drug dose. The total mean thigh volume of the subjects was 7.3 ⫾ 1.2 liters. Blood gases. Blood samples for blood gases were drawn in the middle of each PET scan from the femoral vein and radial artery and analyzed with standardized hospital practices (Radiometer ABL 835 blood gas analyzer). Statistical analysis. Statistical analyses were performed with SAS 8.2 and SAS Enterprise 4.2 programs (SAS Institute, Cary, NC). The distribution of the parameter was tested, and statistical analyses were performed using two-way ANOVA for repeated measures. If a significant main effect(s) was found, pairwise differences were identified using the Tukey-Kramer post hoc procedure. Results are expressed as means ⫾ SD, and P ⱕ 0.05 was considered statistically significant. RESULTS

The effects of hypoxia at rest. At rest, hypoxia tended to increase heart rate (P ⫽ 0.07) and both systolic arterial pressure and MAP (P ⫽ 0.051), but diastolic blood pressure was not changed (P ⫽ 0.19). Hypoxia did not change mean thigh muscle blood flow at rest (2.9 ⫾ 1.6 vs. 3.3 ⫾ 1.4 ml·100 g⫺1·min⫺1, P ⫽ 0.62, normoxia vs. hypoxia) and did not have an effect on vascular resistance (P ⫽ 0.86) or conductance (P ⫽ 0.71). Hypoxia decreased arterial oxygen content (Table 1), with a concomitant decrease in venous oxygen content while oxygen extraction remained unchanged (P ⫽ 0.6). Hypoxia did

Table 1. Heart rates, blood pressures, leg vascular resistance and conductance, and blood oxygen in radial artery and femoral vein at rest and at baseline hypoxia, as well as during exercise in NORMO and HYPO without and with intra-arterial aminophylline infusion Rest

HR, beats/min MAP, mmHg BPs, mmHg BPd, mmHg Leg VR, mmHg·(ml·100 g⫺1·min⫺1) Leg VC, (ml·100 g⫺1·min⫺1)·mmHg⫺1 Oxygen saturation-A, % Oxygen content-A, ml/l Oxygen saturation-V, % Oxygen content-V, ml/l Oxygen extraction, ml/l OEF, % ˙ O2, ml·100 g⫺1·min⫺1 Leg V

Exercise Without Aminophylline

Exercise With Aminophylline

Baseline

Hypoxia

NORMO

HYPO

NORMO

HYPO

61 ⫾ 10 91 ⫾ 7 125 ⫾ 9 74 ⫾ 6 39 ⫾ 19 0.03 ⫾ 0.02 98 ⫾ 1 199 ⫾ 9 76 ⫾ 9 152 ⫾ 23 48 ⫾ 16 24 ⫾ 9 0.1 ⫾ 0.1

69 ⫾ 10 98 ⫾ 12 137 ⫾ 18 79 ⫾ 10 37 ⫾ 22 0.04 ⫾ 0.02 91 ⫾ 5e 186 ⫾ 13e 71 ⫾ 10f 141 ⫾ 18f 45 ⫾ 14 24 ⫾ 8 0.1 ⫾ 0.1

92 ⫾ 12a 108 ⫾ 6a 146 ⫾ 7a 90 ⫾ 9a 3.2 ⫾ 1.0a 0.3 ⫾ 0.1a 98 ⫾ 1 205 ⫾ 9 43 ⫾ 16a 88 ⫾ 32a 117 ⫾ 33a 57 ⫾ 16a 4.1 ⫾ 1.0a

102 ⫾ 10a,b 112 ⫾ 10a 152 ⫾ 11a 92 ⫾ 12a 2.8 ⫾ 0.4a 0.4 ⫾ 0.1a 88 ⫾ 5c 182 ⫾ 11d 37 ⫾ 13a,b 78 ⫾ 26a,b 101 ⫾ 29a,b 56 ⫾ 16a 4.0 ⫾ 1.3a

97 ⫾ 20a 105 ⫾ 7a 147 ⫾ 12a 83 ⫾ 7a 2.8 ⫾ 0.4a 0.4 ⫾ 0.1a 99 ⫾ 0.4 208 ⫾ 24 48 ⫾ 17a 96 ⫾ 31a 114 ⫾ 44a 53 ⫾ 18a 4.4 ⫾ 1.9a

110 ⫾ 13a,b 107 ⫾ 8a 149 ⫾ 13a 86 ⫾ 11a 2.9 ⫾ 0.9a 0.4 ⫾ 0.1a 89 ⫾ 4c 182 ⫾ 8d 41 ⫾ 18a,b 83 ⫾ 35a,b 96 ⫾ 37a,b 53 ⫾ 20a 3.9 ⫾ 1.9a

The results are means ⫾ SD. NORMO, normoxia; HYPO, hypoxia; HR, heart rate; MAP, mean arterial pressure; BPs, systolic blood pressure; BPd, diastolic blood pressure; VR, vascular resistance; VC, vascular conductance; oxygen saturation and content-A, arterial oxygen saturation and content; oxygen saturation ˙ O2, oxygen consumption. eP ⬍ 0.01 and fP ⬍ 0.05 compared with rest. and content-V, venous oxygen saturation and content; OEF, oxygen extraction fraction; V b P ⬍ 0.05, cP ⬍ 0.01, and dP ⬍ 0.001 compared with NORMO and resting baseline. aP ⬍ 0.01 compared with baseline rest and hypoxia. AJP-Regul Integr Comp Physiol • VOL

299 • JULY 2010 •

www.ajpregu.org

R75


Fig. 2. The effect of hypoxia (HYPO) and aminophylline on QF muscle blood flow (A) and its heterogeneity (B) during one-legged exercise. Hypoxia increased blood flow in exercising QF muscle, but aminophylline did not affect it either in normoxia or hypoxia. Despite increased blood flow in hypoxia, exercising muscle flow heterogeneity was similar to normoxic exercise. NORMO, normoxia; amp, under aminophylline blockade. The results are means ⫾ SE. *P ⬍ 0.05.

not affect blood flow heterogeneity at rest (63 ⫾ 16% in normoxia and 60 ⫾ 13% in hypoxia, P ⫽ 0.72). Exercise measurements. Blood pressure and heart rate increased from rest to exercise. Systolic, diastolic, and mean arterial blood pressures were all similar during exercise in normoxia, hypoxia, and under aminophylline infusion. Heart rate was increased during hypoxic compared with normoxic exercise (P ⫽ 0.02), and there was a tendency for aminophylline to increase heart rate in the hypoxia compared with normoxic exercise (P ⫽ 0.08) (Table 1 and Figs. 1 and 2). Mean blood flow and oxygen consumption in working QF muscle was higher during exercise than at rest (P ⬍ 0.001). There was also a significant redistribution of blood flow in thigh muscles from rest to exercise (Fig. 1) and in absolute heterogeneity values over all thigh muscles (63 ⫾ 16% at rest and 95 ⫾ 14% during exercise, P ⬍ 0.001). Whole thigh blood

flow heterogeneity was not different between the exercise interventions studied (P ⬎ 0.45, data not shown), whereas blood flow was higher in QF in hypoxic compared with normoxic exercise (P ⫽ 0.02), and aminophylline had no effect in either normoxia or hypoxia (P ⫽ 0.5) (Fig. 2A). Oxygen saturation (⬃88%) and oxygen content of arterial blood was significantly lower during one-leg exercise in hypoxia compared with normoxia (Table 1). Arterial-venous oxygen extraction was reduced under hypoxic exercise (P ⫽ 0.02), and aminophylline had no effect on extraction (P ⫽ 0.92). Muscle oxygen consumption was similar in all four exercise interventions (P ⬎ 0.15; Table 2). Blood flow heterogeneity in working QF muscle tended to be significantly reduced compared with rest (60 ⫾ 19% normoxia at rest and 46 ⫾ 6% normoxia during exercise; P ⫽ 0.055), but neither hypoxia nor aminophylline changed this heterogeneity during exercise (P ⬎ 0.61; Fig. 2B). There were no differences in vascular resistance or conductance between different exercise interventions (Table 1), but vascular resistance decreased and conductance increased from rest. Of importance, the hypoxia-induced increase in blood flow was confined only to working QF muscle, since blood flow did not change significantly in the posterior part of the thigh muscles from rest for any of the exercise conditions (P ⬎ 0.4; Fig. 3A). Posterior hamstring muscle blood flow during normoxic control exercise was comparable to flow at rest (Fig. 3A). Blood flow heterogeneity was, however, increased significantly from rest to exercise in posterior muscles of the thigh, but there were no changes in posterior muscle blood flow heterogeneity between the different exercise interventions (Fig. 3B). Vascular conductance or resistance did not change significantly in posterior muscles in any condition studied (Table 2). Mean blood flow and vascular conductance or resistance in whole thigh musculature of the resting contralateral leg were not changed from rest to one-leg normoxic control exercise (Table 2), but heterogeneity of blood flow was markedly increased (Table 2 and Fig. 1E). There were no changes between different exercise conditions in blood flow or its heterogeneity during exercise (Table 2). DISCUSSION

We examined the skeletal muscle blood flow response directly within the thigh muscles of the exercising and resting contralateral limb in normoxia and systemic hypoxia (14% O2) and under nonspecific adenosine receptor inhibition. The main findings were that 1) limb oxygen extraction was reduced and

Table 2. Vascular conductance, resistance as well as mean blood flow and flow heterogeneity in noncontracting muscles at resting baseline, at rest in hypoxia, and during exercise in different conditions studied Rest

VC-Post, (ml·100 g⫺1·min⫺1)·mmHg⫺1 VR-Post, mmHg·(ml·100 g⫺1·min⫺1) VC-CL, (ml·100 g⫺1·min⫺1)·mmHg⫺1 VR-CL, mmHg·(ml·100 g⫺1·min⫺1) Muscle BF-CL, ml·100 g⫺1·min⫺1 Muscle BF heterogeneity-CL, %

Exercise Without Aminophylline

Exercise With Aminophylline

Baseline

Hypoxia

NORMO

HYPO

NORMO

HYPO

0.03 ⫾ 0.02 40 ⫾ 22 0.03 ⫾ 0.01 38 ⫾ 19 3.1 ⫾ 1.9 56 ⫾ 12

0.04 ⫾ 0.02 39 ⫾ 23 0.04 ⫾ 0.02 39 ⫾ 22 3.4 ⫾ 1.9 52 ⫾ 10

0.04 ⫾ 0.01 31 ⫾ 13 0.03 ⫾ 0.02 47 ⫾ 39 3.7 ⫾ 2.6 113 ⫾ 14*

0.04 ⫾ 0.02 35 ⫾ 18 0.04 ⫾ 0.03 37 ⫾ 33 4.0 ⫾ 4.0 112 ⫾ 28*

0.04 ⫾ 0.02 32 ⫾ 17 0.04 ⫾ 0.04 44 ⫾ 32 4.5 ⫾ 4.2 107 ⫾ 21*

0.04 ⫾ 0.02 30 ⫾ 9 0.04 ⫾ 0.03 53 ⫾ 41 4.1 ⫾ 3.9 107 ⫾ 9*

The results are means ⫾ SD. Post, posterior muscles; CL, (resting) contralateral leg; BF, blood flow. Blood flow and its heterogeneity are reported here only form resting contralateral whole thigh muscles, since data from posterior muscles from working leg are shown in Fig. 3. *P ⬍ 0.001 compared with resting baseline and hypoxia. AJP-Regul Integr Comp Physiol • VOL

299 • JULY 2010 •

www.ajpregu.org

R76


Fig. 3. Mean blood flow (A) and its heterogeneity (B) in nonactive posterior hamstring muscles of the working leg at resting baseline, in hypoxia at rest, and during exercise in four different study conditions. Although posterior muscle blood flow did not change significantly from rest to any of the exercise conditions (P ⬎ 0.3), flow heterogeneity in posterior muscles was increased from rest when exercise was performed by QF muscles. Similar but more pronouncedly enhanced flow heterogeneity was also observed in resting contralateral musculature (see the end of RESULTS for details). ***P ⬍ 0.001 compared with resting baseline and hypoxia at rest. The results are means ⫾ SD.

muscle blood flow increased, which was confined only to the exercising muscles in moderate hypoxic exercise; 2) in the working muscles, blood flow heterogeneity tended to be reduced from rest (P ⫽ 0.055), which was not affected by hypoxia; 3) the hypoxia-induced increase in exercising muscle blood flow was not mediated via adenosine; and 4) increased blood flow heterogeneity but unchanged mean blood flow was observed in noncontracting muscles (posterior hamstring in working limb and resting contralateral leg), which likely reflects sympathetic nervous constraints to muscle vasculature to blunt blood flow increments elsewhere than in working skeletal muscles. Muscle blood flow was not significantly affected by the moderate hypoxic exposure at rest in the present study. This is in line with recent studies showing that even more severe systemic hypoxia does not change resting leg blood flow (5, 36), which can be attributed to a profound increase in leg muscle sympathetic nervous activation (35, 38). In contrast, in the forearm, net flow increases are often observed (45, 47), possibly because blood flow increases in tissues other than muscle, such as in skin (40, 45). The reduced arterial oxygen content was compensated for by a concomitant reduction in venous oxygen content such that oxygen extraction rate remained the same. During exercise, however, the drop in blood oxygen content was larger in the arterial than venous circula-

tion; thus, limb oxygen extraction was reduced in hypoxic exercise, and oxygen supply to working muscles was provided by increased blood flow. In this respect, our study adds novel information to this well-established hemodynamic hypoxic response (10, 26, 36) by showing for the first time in humans that hypoxia-induced blood flow enhancement is confined only to working muscle while blood flow in other muscles (posterior hamstring part, and also resting contralateral leg) remained similar to that in normoxic exercise. Because heart rate was higher in hypoxia, the absence of flow increase in posterior muscles suggests increased sympathetic nervous system activation (18, 22, 42), which likely restrained local muscle vasodilation. Furthermore, that blood flow heterogeneity was increased significantly from rest to exercise in posterior muscles of the thigh even without any change in mean blood flow supports the likelihood of increased sympathetic nervous activity. Hypoxia did not evoke a further increase in flow heterogeneity in posterior muscles during exercise compared with normoxia (39), which suggests that sympathetic activity was not markedly enhanced by this moderate level of hypoxia during small muscle mass knee extension exercise. These data support the conclusion that, rather than primarily reducing inactive muscle blood flow as we hypothesized, neural restraints on blood flow in hypoxia with a small muscle mass occur also in contracting muscles, as recent animal and human studies suggest (6, 43). Muscle blood flow heterogeneity in exercising muscle and the effect of hypoxia. Changes in muscle blood flow heterogeneity from rest to exercise depend largely on exercise intensity. We previously showed that blood flow heterogeneity in working QF actually increases from rest to mild exercise intensity and starts to diminish when exercise intensity is increased to a moderate level (21). Blood flow heterogeneity in working QF in the present study tended to be reduced compared with the resting state (P ⫽ 0.055), which can be explained by relatively high exercise intensity. Hypoxia, however, did not change blood flow heterogeneity in exercising muscle, despite a 10% increase in mean blood flow. Structural and functional properties likely play an important role in changes in blood flow heterogeneity in exercising muscle. At rest, the diameter of arterioles controlling blood flow in a group of capillaries changes rhythmically (vasomotion). During muscle contraction, capillary units are likely to be recruited (9, 23), and the time frequency of diminished diameter of arteriole is reduced (17), contributing to the often-noted decrease in voxel-by-voxel heterogeneity. Regarding heterogeneity in hypoxia, we hypothesized that, if more capillaries are recruited and/or the basal contraction frequency of smooth muscle cells is reduced in hypoxic exercise, muscle blood flow would be even more uniform. In contrast to our hypothesis, heterogeneity remained essentially the same during exercise in hypoxia. It is highly likely that the capillary recruitment was already maximal during this level of exercise, or hypoxia per se simply does not induce greater capillary recruitment within the contracting muscle. This pattern may be altered by more pronounced hypoxia and/or intensive exercise (4). However, although it is commonly known that the arterial blood oxygen saturation is tightly coupled to inspired O2 fraction, even with the combination of more severe hypoxic exposure (12% O2) and exercise intensity (maximal one-legged exercise), arterial oxygen saturation does not necessarily decrease lower than the


299 • JULY 2010 •

www.ajpregu.org


88% value we also observed in the present study (2). This is because lower inspired oxygen usually also elicits marked ventilatory compensation trying to maintain arterial O2. Thus, during exercise with a small muscle mass, pulmonary gas exchange and muscle O2 delivery are well preserved in hypoxia by virtue of the capacity to increase blood flow and O2 delivery to a given recruited muscle mass (6). In addition, oxygen extraction fraction values (⬃57% in control exercise) in general suggest that the one-leg exercise indeed was of moderate to high intensity, since they are close to the wellestablished maximal value of 70% in comparable (exercise duration, etc.) one-leg exercise (1). Thus severity of hypoxia or exercise intensity is unlikely to explain similar blood flow heterogeneity during hypoxic and normoxic exercise in the one-leg knee extension model. No evidence for adenosine mediation in hypoxia-induced muscle capillary blood flow increase. The importance of adenosine in blood flow control during exercise in normoxia and hypoxia was examined in the present study by femoral arterial infusion of the nonspecific adenosine receptor blocker aminophylline as previously applied in the studies of forearm blood flow in systemic hypoxia at rest (29) and in normoxic (32) and hypoxic (8) forearm exercise. These studies showed that aminophylline is effectively delivered from luminal to muscle interstitial spaces and that it not only inhibited the normal flow increase with contractions (32) but also effectively inhibited intraluminal adenosine acting on the endothelium, since resting hypoxic vasodilation was abolished (29). As mentioned, in the present study, blood flow was enhanced substantially from rest and was significantly higher in hypoxic exercise compared with normoxia. Aminophylline did not affect blood flow in exercising QF, suggesting no contribution of adenosine to capillary blood flow in normoxic or hypoxic exercise in the conditions studied. Additionally, adenosine receptor blockade did not change blood flow heterogeneity either during normoxic or hypoxic exercise. The fact that flow heterogeneity was not changed during hypoxia under adenosine receptor blockade is logical, since hypoxia per se did not affect exercising muscle blood flow heterogeneity. Some animal studies suggest that adenosine could account for up to 40% of exercise hyperemia, but there are some disparate findings on the significance of adenosine that relate to species differences, measurement methods, the inhibitors used, and the exercise conditions, such as intensity (31). Also, some (33, 34), but not all (8, 21), human studies suggest that adenosine would play a role in exercise hyperemia. In the present study, we did not find any support for adenosine in the regulation of muscle capillary blood flow during exercise. Clearly, methodological factors account for these different outcomes. While Doppler ultrasound or thermodilution methods measure bulk limb blood flow, PET allows determination of nutritive capillary blood flow within the muscles. Thus it seems that capillary blood flow is not as easily disturbed by adenosine blockade as bulk blood flow. Supporting this, it is known from the coronary circulation that, when stenosis limits blood flow in large coronary arteries, there is a substantial compensatory downstream vasodilation (12). Because it is obvious that QF is at least slightly overperfused during one-leg knee extension (20), it may well be that blockade affects resistance vessels, and overall conduit blood flow may be slightly reduced (33, 34). Simultaneously, there is a compen-

R77

satory vasodilation in more distal resistance vessels close to the capillaries preserving more critical working muscle capillary nutritive blood flow. In this respect, it is noteworthy that blood flow in conduit arteries does not always represent blood flow in capillaries, especially in the onset of exercise (19), but possibly also in other conditions. This mainly unexplored issue warrants future experimentation. It is also highly likely that compensatory vasodilators, such as ATP derived from red blood cells, nitric oxide, and/or prostanoids or ␤-aderenergic vasodilation, appeared to preserve exercising muscle blood flow during inhibition (7, 15, 16, 45, 46). Although the data in the present and our previous study (21) clearly suggest that adenosine is not obligatory for the increase in muscle capillary blood flow in exercising muscle, it does not, however, exclude the possibility that adenosine is one of the many tonic regulators of skeletal muscle blood flow as shown by Duncker and colleagues (13) in swine myocardium and Edlund & Sollevi (14) in the human heart. Normally adenosine emanates mainly from extracellular pathways, and intracellular adenosine formation and its subsequent release to muscle interstitial space to the vicinity of smooth muscle cells of resistance vessels is triggered only when muscle metabolic demands exceed oxygen delivery (12). Thus the physiological contribution of adenosine to muscle hyperemia seems not to be substantially activated in steady-state exercise with a small muscle mass in moderate systemic hypoxia but may require more severe ischemic conditions or higher exercise intensities (32) to be activated. There were no signs of this in the present study since, for instance, blood lactate remained essentially similar during all conditions (data not shown). Therefore, hyperemic mediation by adenosine needs to be further examined in maximal whole body exercise where a mismatch between metabolism and oxygen delivery can be assumed to exist. The conclusion that adenosine is not mandatory for the regulation of skeletal muscle nutritive capillary blood flow during exercise with a small muscle mass in normoxia or physiological systemic hypoxia is thus well in accordance with cardiac studies showing no indication for adenosine mediation in normal coronary exercise hyperemia in dogs, swine, or humans (12). Only when stenosis or severe ischemia affects cardiac perfusion does adenosine contribute importantly to coronary vasodilation (12). Other groups, having studied the contraction-induced skeletal muscle hyperemia and reviewed the evidence for adenosine mediation, conclude that, despite long-standing interest in adenosine as a mediator of metabolic vasodilator, there is strong evidence against its involvement in exercise hyperemia (44). Finally, although we (20) and others (32) have found substantial variation in blood flow responses with adenosine infusion, no subdivision into “responders and nonresponders” was observed in the present study regarding the actions of endogenous adenosine. Perspectives and Significance Although many effects of both acute and chronic hypoxia on the circulation are well characterized, the distribution and regulation of blood flow heterogeneity in skeletal muscle during systemic hypoxia is not well elucidated in humans. In this study, we report several major findings. First, during small muscle mass exercise in moderate systemic hypoxia, muscle capillary blood flow is increased to a larger extent than oxygen


299 • JULY 2010 •

www.ajpregu.org

R78


extraction to compensate for the lower arterial oxygen content. It is shown here for the first time in humans that this increased blood flow is confined only to the exercising muscle, and flow heterogeneity per se is not affected by hypoxia. The hypoxiainduced increase in muscle capillary blood flow appeared not to be mediated by adenosine, which supports recent whole limb blood flow findings, and exercise intensity rather than the oxygen content of the blood determines the level of capillary recruitment of the muscle in this exercise model. Finally, the results also suggest that the observed increase in blood flow heterogeneity in noncontracting human muscles reflects sympathetic nervous constraints to the muscular vasculature to blunt blood flow increments elsewhere than in working skeletal muscles, but which are not yet potentiated in this moderate level of hypoxia. We propose that, when oxygen delivery to a given mass of activated muscle can be maintained by an increase in muscle blood flow, no alterations in muscle recruitment occur, the muscle is not in a situation of ischemia, and adenosine is not a mediator of the flow increase. Of interest is how flow distribution and its regulation may be altered during exercise with a larger muscle mass where a central flow limitation may occur at high intensities of exercise. ACKNOWLEDGMENTS This study could not have been performed without the contribution of the personnel of the Turku PET, and we thank them for excellent assistance during the study. GRANTS The study was conducted within the Centre of Excellence in Molecular Imaging in Cardiovascular and Metabolic Research, supported by the Academy of Finland, University of Turku, Turku University Hospital, and Abo Academy. The present study was financially supported by The Ministry of Education of the State of Finland (Grants 74/627/2006, 58/627/2007, and 45/627/2008), Academy of Finland (Grants 108539 and 214329 and Centre of Excellence funding), The Finnish Cultural Foundation and its South-Western Fund, The Finnish Sport Research Foundation, and Turku University Hospital (EVO funding). DISCLOSURES No conflicts of interest are declared by the authors. REFERENCES 1. Andersen P, Saltin B. Maximal perfusion of skeletal muscle in man. J Physiol (Lond) 366: 233–249, 1985. 2. Barden J, Lawrenson L, Poole JG, Kim J, Wray DW, Bailey DM, ˙ O2max in Richardson RS. Limitations to vasodilatory capacity and V trained human skeletal muscle. Am J Physiol Heart Circ Physiol 292: H2491–H2497, 2007. 3. Berne RM. Cardiac nucleotides in hypoxia: possible role in regulation of coronary blood flow. Am J Physiol 204: 317–322, 1963. 4. Bourdillon N, Mollard P, Letournel M, Beaudry M, Richalet JP. Non-invasive evaluation of the capillary recruitment in the human muscle during exercise in hypoxia. Respir Physiol Neurobiol 165: 237–244, 2009. 5. Brooks GA, Wolfel EE, Butterfield GE, Cymerman A, Roberts AC, Mazzeo RS, Reeves JT. Poor relationship between arterial [lactate] and leg net release during exercise at 4,300 m altitude. Am J Physiol Regul Integr Comp Physiol 275: R1192–R1201, 1998. 6. Calbet JA, Radegran G, Boushel R, Saltin B. On the mechanisms that limit oxygen uptake during exercise in acute and chronic hypoxia: role of muscle mass. J Physiol 587: 477–490, 2009. 7. Casey DP, Madery BD, Curry TB, Eisenach JH, Wilkins BW, Joyner MJ. Nitric oxide contributes to the augmented vasodilatation during hypoxic exercise. J Physiol 588: 373–385, 2010. 8. Casey DP, Madery BD, Pike TL, Eisenach JH, Dietz NM, Joyner MJ, Wilkins BW. Adenosine receptor antagonist and augmented vasodilation during hypoxic exercise. J Appl Physiol 107: 1128 –1137, 2009.

9. Clark MG, Rattigan S, Barrett EJ, Vincent MA. Point: There is capillary recruitment in active skeletal muscle during exercise. J Appl Physiol 104: 889 –891, 2008. 10. DeLorey DS, Shaw CN, Shoemaker JK, Kowalchuk JM, Paterson DH. The effect of hypoxia on pulmonary O2 uptake, leg blood flow and muscle deoxygenation during single-leg knee-extension exercise. Exp Physiol 89: 293–302, 2004. 11. Duling BR, Damon DH. An examination of the measurement of flow heterogeneity in striated muscle. Circ Res 60: 1–13, 1987. 12. Duncker DJ, Bache RJ. Regulation of coronary blood flow during exercise. Physiol Rev 88: 1009 –1086, 2008. 13. Duncker DJ, Stubenitsky R, Verdouw PD. Role of adenosine in the regulation of coronary blood flow in swine at rest and during treadmill exercise. Am J Physiol Heart Circ Physiol 275: H1663–H1672, 1998. 14. Edlund A, Sollevi A. Theophylline increases coronary vascular tone in humans: evidence for a role of endogenous adenosine in flow regulation. Acta Physiol Scand 155: 303–311, 1995. 15. Gonzalez-Alonso J, Mortensen SP, Dawson EA, Secher NH, Damsgaard R. Erythrocytes and the regulation of human skeletal muscle blood flow and oxygen delivery: role of erythrocyte count and oxygenation state of haemoglobin. J Physiol 572: 295–305, 2006. 16. Gonzalez-Alonso J, Olsen DB, Saltin B. Erythrocyte and the regulation of human skeletal muscle blood flow and oxygen delivery: role of circulating ATP. Circ Res 91: 1046 –1055, 2002. 17. Haddock RE, Hill CE. Rhythmicity in arterial smooth muscle. J Physiol 566: 645–656, 2005. 18. Halliwill JR, Minson CT. Effect of hypoxia on arterial baroreflex control of heart rate and muscle sympathetic nerve activity in humans. J Appl Physiol 93: 857–864, 2002. 19. Harper AJ, Ferreira LF, Lutjemeier BJ, Townsend DK, Barstow TJ. Human femoral artery and estimated muscle capillary blood flow kinetics following the onset of exercise. Exp Physiol 91: 661–671, 2006. 20. Heinonen I, Kemppainen J, Kaskinoro K, Peltonen JE, Borra R, Lindroos MM, Oikonen V, Nuutila P, Knuuti J, Hellsten Y, Boushel R, Kalliokoski KK. Comparison of exogenous adenosine and voluntary exercise on human skeletal muscle perfusion and perfusion heterogeneity. J Appl Physiol 108: 378 –386, 2010. 21. Heinonen IH, Nesterov SV, Kemppainen JT, Nuutila P, Knuuti J, Laitio R, Kjaer M, Boushel R, Kalliokoski KK. Role of adenosine in regulating the heterogeneity of skeletal muscle blood flow during exercise in humans. J Appl Physiol 103: 2042–2048, 2007. 22. Heistad DD, Abboud Dickinson W FM. Richards Lecture: Circulatory adjustments to hypoxia. Circulation 61: 463–470, 1980. 23. Honig CR, Odoroff CL, Frierson JL. Capillary recruitment in exercise: rate, extent, uniformity, and relation to blood flow. Am J Physiol Heart Circ Physiol 238: H31–H42, 1980. 24. Iida H, Kanno I, Miura S, Murakami M, Takahashi K, Uemura K. Error analysis of a quantitative cerebral blood flow measurement using H2(15)O autoradiography and positron emission tomography, with respect to the dispersion of the input function. J Cereb Blood Flow Metab 6: 536 –545, 1986. 25. Kalliokoski KK, Kuusela TA, Nuutila P, Tolvanen T, Oikonen V, Teras M, Takala TE, Knuuti J. Perfusion heterogeneity in human skeletal muscle: fractal analysis of PET data. Eur J Nucl Med 28: 450 –456, 2001. 26. Koskolou MD, Calbet JA, Radegran G, Roach RC. Hypoxia and the cardiovascular response to dynamic knee-extensor exercise. Am J Physiol Heart Circ Physiol 272: H2655–H2663, 1997. 27. Krogh A. The number and distribution of capillaries in muscles with calculations of the oxygen pressure head necessary for supplying the tissue. J Physiol 52: 409 –415, 1919. 28. Krogh A. The supply of oxygen to the tissues and the regulation of the capillary circulation. J Physiol 52: 457–474, 1919. 29. Leuenberger UA, Gray K, Herr MD. Adenosine contributes to hypoxiainduced forearm vasodilation in humans. J Appl Physiol 87: 2218 –2224, 1999. 30. Lundby C, Boushel R, Robach P, Moller K, Saltin B, Calbet JA. During hypoxic exercise some vasoconstriction is needed to match O2 delivery with O2 demand at the microcirculatory level. J Physiol 586: 123–130, 2008. 31. Marshall JM. The roles of adenosine and related substances in exercise hyperaemia. J Physiol 583: 835–845, 2007. 32. Martin EA, Nicholson WT, Eisenach JH, Charkoudian N, Joyner MJ. Influences of adenosine receptor antagonism on vasodilator responses to


299 • JULY 2010 •

www.ajpregu.org


33.

34. 35.

36.

37.

38.

39.

adenosine and exercise in adenosine responders and nonresponders. J Appl Physiol 101: 1678 –1684, 2006. Mortensen SP, Nyberg M, Thaning P, Saltin B, Hellsten Y. Adenosine contributes to blood flow regulation in the exercising human leg by increasing prostaglandin and nitric oxide formation. Hypertension 53: 993–999, 2009. Radegran G, Calbet JA. Role of adenosine in exercise-induced human skeletal muscle vasodilatation. Acta Physiol Scand 171: 177–185, 2001. Rowell LB, Johnson DG, Chase PB, Comess KA, Seals DR. Hypoxemia raises muscle sympathetic activity but not norepinephrine in resting humans. J Appl Physiol 66: 1736 –1743, 1989. Rowell LB, Saltin B, Kiens B, Christensen NJ. Is peak quadriceps blood flow in humans even higher during exercise with hypoxemia? Am J Physiol Heart Circ Physiol 251: H1038 –H1044, 1986. Ruotsalainen U, Raitakari M, Nuutila P, Oikonen V, Sipila H, Teras M, Knuuti MJ, Bloomfield PM, Iida H. Quantitative blood flow measurement of skeletal muscle using oxygen-15- water and PET. J Nucl Med 38: 314 –319, 1997. Saito M, Mano T, Iwase S, Koga K, Abe H, Yamazaki Y. Responses in muscle sympathetic activity to acute hypoxia in humans. J Appl Physiol 65: 1548 –1552, 1988. Seals DR, Johnson DG, Fregosi RF. Hypoxia potentiates exerciseinduced sympathetic neural activation in humans. J Appl Physiol 71: 1032–1040, 1991.

R79

40. Simmons GH, Minson CT, Cracowski JL, Halliwill JR. Systemic hypoxia causes cutaneous vasodilation in healthy humans. J Appl Physiol 103: 608 –615, 2007. 41. Sipilä HT, Clark JC, Peltola O, Teräs M. An automatic [15O]-H2O production system for heart and brain studies. J Lab Compound Radiopharm 44: S1066 –S1068, 2001. 42. Somers VK, Mark AL, Zavala DC, Abboud FM. Influence of ventilation and hypocapnia on sympathetic nerve responses to hypoxia in normal humans. J Appl Physiol 67: 2095–2100, 1989. 43. Stickland MK, Smith CA, Soriano BJ, Dempsey JA. Sympathetic restraint of muscle blood flow during hypoxic exercise. Am J Physiol Regul Integr Comp Physiol 296: R1538 –R1546, 2009. 44. Valic Z, Naik JS, Ruble SB, Buckwalter JB, Clifford PS. Elevation in resting blood flow attenuates exercise hyperemia. J Appl Physiol 93: 134 –140, 2002. 45. Weisbrod CJ, Minson CT, Joyner MJ, Halliwill JR. Effects of regional phentolamine on hypoxic vasodilatation in healthy humans. J Physiol 537: 613–621, 2001. 46. Wilkins BW, Pike TL, Martin EA, Curry TB, Ceridon ML, Joyner MJ. Exercise intensity-dependent contribution of beta-adrenergic receptor-mediated vasodilatation in hypoxic humans. J Physiol 586: 1195– 1205, 2008. 47. Wilkins BW, Schrage WG, Liu Z, Hancock KC, Joyner MJ. Systemic hypoxia and vasoconstrictor responsiveness in exercising human muscle. J Appl Physiol 101: 1343–1350, 2006.


299 • JULY 2010 •

www.ajpregu.org