AMERXCAN
JOURNAL
OF PHYSIOLOGY
Vol. 229, No. 1, July 1975. Printed in U.S.A.
Prolonged vasodilation following exercise of dog skeletal muscle MELVIN Department
L. MORGANROTH, DAVID E. MOHRMAN, AND HARVEY of Physiology, University of Michigan, Ann Arbor, Michigan 48104
MURGANROTH,MELVIN L., DAVID E. MOHRMAN, AND HARVEY SPARKS. Prolonged uasodilation following fatiguing exercise of dog skeletal muscle. Am. J, Physiol. 229(l) : 38-43. 1975.-Prolonged vasodilation follows 20- and 60-min periods of stimulation (4 twitches/s) of dog skeletal muscle with blood flow held constant at 23 + 2 ml/100 g per min. During stimulation isometric tension development fell 61 h 3%, and we have operationally defined this as fatiguing exercise. During stimulation vasodilation was maximum. Following stimulation vascular resistance returned to control with average half-times of 6.9 + 0.6 and 5.8 =I= 0.4 min after 20 and 60 min of stimulation, respectively. This prolonged vascular recovery following fatiguing exercise is an order of magnitude slower than vascular recovery following less severe exercise. The pressor response to an intra-arterial bolus of angiotensin (0.5 pg) is not depressed during prolonged vasodilation. Prolonged vasodilation does not appear to be closely linked to changes in tissue oxygen consumption. In addition, the changes in resistance can be dissociated from changes in K+, osmolality, and lactate production following fatiguing exercise. We conclude that prolonged vasodilation following fatiguing exercise is caused by a metabolic vasodilator substance which is yet to be identified.
V
fatiguing
l
exercise hyperemia; metabolic 00,; lactic acid; vascular resistance
vasodilator;
Kc;
osmolality;
SEVERAL INVESTIGATORS (1, 2, 5) have shown that a vasodilation lasting approximately 30 min follows fatiguing exercise of skeletal muscle. This vasodilation lasts far beyond the 3-min period of increased oxygen consumption during recovery (2, 5). In the present study we attempted to further characterize this long-lasting vasodilation and determine its cause. We tested the ability of vascular smooth muscle to contract during prolonged vasodilation and examined the possible role of several metabolic vasodilator mechanisms which have been implicated in exercise hyperemia, including elevated K+ (4), oxmolality (8), oxygen consumption (3), and lactic acid (11). METHODS
All of our experiments were performed on isolated, in situ, anterior calf muscles (tibialis cranialis and extensor digitorum longus) of ZO- to 25-kg male dogs, using a preparation identical to one described by Mohrman and Sparks (10). Animals were anesthetized with sodium pentobarbital (30 mg/kg, iv) supplemented as required. After complete isolation of the muscles’ blood flow, cannulas were placed in the sole arterial and venous supply of the muscles. A
V. SPARKS
perfusion
pump was then placed in the arterial line to blood from the contralateral femoral artery at a constant rate. Vascular resistance (mmHg/ml blood/ 100 g per min) was calculated by dividing values of perfusion pressure by the constant flow values. Perfusion and systemic pressures were measured continuously using Statham pressure transducers. Isometric tension was monitored by attaching the tendons of the two muscles to force transducers. All experiments were performed with the muscles at in situ length. Twitch contractions were produced by supramaximal square-current pulse stimulation of the fibular nerve using parameters (0.5-1.0 mA at 0.1 ms duration) that will not stimulate nonsomatic fibers as confirmed by the absence of a vascular response to nerve stimulation after somatic neuromuscular blockade. Venous and arterial blood samples were taken directly from the perfusion lines to determine the concentrations of plasma K+ by flame photometry, plasma osmolality by freezing-point depression, and lactic acid by an enzymatic technique (Sigma Chemical Carp,). 02 content of the venous blood was monitored continuously by passing a portion of the muscle effluent through a Gilford densitometer used as an oximeter, and arterial blood samples were taken periodically for measurement of 02 content by Van Slyke manometric analysis. Muscle 02 consumption (vow) was calculated as flow times arterial-venous 02 content difference. Four series of experiments were performed. The first series of experiments (n = 4) consisted of a comparison of the resistance vessel responsiveness to angiotensin during the prolonged vasodilation with the resistance vessel response to the same dose of angiotensin at the same initial resistances during mild exercise (0.5-Z twitches/s) The mild exercise control observations were obtained before and after the heavy exercise and ensuing prolonged vasodilation Flow was held constant throughout. Each trial consisted of a close intra-arterial injection of 0.5 pg of angiotensin in a .Ol-ml volume of saline. The vasoconstrictor response to angiotensin was normalized within each animal by dividing the peak magnitude of all pressor responses by the average magnitude of those control responses for which the initial perfusion pressure was greater than or equal to 40 mmHg. The second set of experiments consisted of stimulating the muscle at 4 twitches/s for 20 min while holding blood flow constant throughout. During and following the stimulation period, venous and arterial blood samples were pump
l
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on January 10, 2019.
PROLONGED
39
VASODILATION
taken at intervals to determine the concentration of the various metabolites studiedThe third series of experiments were entirely analogous to the ZO-min stimulation experiment except that the stimulation period was extended we to 60 min (n = 3). In the fourth series of experiments determined the relationship between changes in oxygen consumption and vascular resistance during constant-flow perfusion (n = 6). voz was varied by inducing mild exercise (O-3 twitches/s). Resistance was normalized for each preparation by dividing all resistance values by the resistance value at resting oxygen consumption (% control resistance). Statistical treatment of data followed standard procedures (17). A P value < .05 for the f statistic (or paired-t statistic where appropriate) was accepted as an indication of a statistically significant difference between means of populations. Rate constants were obtained by linear regression analysis of the natural log of alterations of a measured variable against time. RESULTS
Figure 1 illustrates the differences in the vascular responses associated with mild and fatiguing exercise. Plotted are muscle tension and vascular resistance versus time during and following 1 h of stimulation at 2 and 4 twitches/s with the flow rate, in both cases, constant throughout at 20 ml/ 100 g per min. When flow is restricted to 20 ml/ IO0 g per min, tension development is fairly well maintained throughout a 60-min period of Z/s stimulation, whereas with 4/s stimulation tension declines more than 50 % in the first 20 min of the exercise period. The latter case has been operationally defined as fatiguing exercise. Figure I also illustrates the divergent patterns of vascular recovery following these two types of exercise. Following the mild exercise period (2 twitches/s), vascular resistance recovers very rapidly and reaches or even exceeds the control value within 1 min+ In contrast, following a 4 twitches/s stimu-
DURING
PROLONGED
VASOUIL
ATION
I
.
I
0
tNITtA1
1
40 PERFUStUN
I
80 PRESSURE
120 (mmHg)
FIG. 2. Comparison of normalized increase in perfusion pressure in response to a 0.5~pg intra-arterial injection of angiotensin vs. initial perfusion pressure during control conditions (broken) and during prolonged vasodilation (solid). For control trials, initial perfusion pressure was varied by mild exercise (O-Z twitches/s). Points consist of averages of 12 cases for control and 4 cases for prolonged vasodilation obtained using same 4 animals. Flow was held constant throughout and averaged 36,6 =t 3.6 ml/l00 g per min. Bars indicate & SE. See METHODS for response normahzation procedure and Table 1 for raw data.
TAI~LE 1, Increase in perfusion pressure f do w ing angio tensin injec tiun Dog
Situation
m-39
Initial
Perfusion
Pressure
4G59
m-79
8G99
Ranges 10&l 19
12G139
mmHg I I
2 2 3 3 4 4
Control Prolonged vasodilation Control Prolonged vasodilation Control Prolonged vasodilation Control Prolonged vasodilation
19.7 17.0
24.0 46.0
30.0
33.0 31 .o
26.2
20.0
40,5 48.5
46.3 55.0
47.5 62.0
58.0
19.0 18.0
36.8 36.0
42.5 40.0
44.0 37.0
32.3 31.0
28.0 15.0
38,5 32.7
38.0
35.0 44*5
50.0
21.7 zt2.1 16.7 H.5
35.0 h3.7 40.8 ~t3,8
39.2 zlz3.5 47,5 &7.5
39.9 h3.5 43.6 h6.7
41.6 h7.3
28.0
33.0
‘ENS/ON
Mean *SE:
Control Prolonged vasodilation
6
PRU 4 P value
2
1
0
'
' 30
'
'
1
'
'
60
1 90
b
PRU ; OL’
0
’
’
’
i
30
’
’ 60
’
’
1 90
MINUTES
1. TypicaI responses of isometric tension deveIopment and vascular resistance of dog calf muscle to mild exercise (2 twitches/s) and fatiguing exercise (4 twitches/s) when flow is restricted. Flow is constant in both cases at 20 ml/100 g per min. FIG,
.28
.41
.86
.51
Control angiotensin responses were obtained muscles between 0 and 2 twitches/s. Responses are ing to initial perfusion pressure. Each control entry mean of two or more trials. Prolonged vasodilation sent a single observation. Flow was held constant at in both situations for each dog and averaged 36.6 g per min in all dogs.
by exercising classed accordrepresents the entries reprethe same value A 3.6 ml/100
lation period at the same flow rate vascular resistance returns to control slowly, reaching control values only after at least 20 min of recovery. We will refer to this phenomenon as prolonged vasodilation following fatiguing exercise, and the remainder of the experiments were done to characterize this prolonged vasodilation and to attempt to determine the mechanism responsible for it. Resfionse to angio tensin. Figure 2 and Table 1 show the
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on January 10, 2019.
40
MORGANROTH,
results from four experiments designed to test the ability of vessels to contract in response to an intra-arterial injection of 0.5 pg of angiotensin during the period of prolonged vasodilation following fatiguing exercise. Administration of angiotensin produced vasoconstriction which peaked within 30 s and lasted approximately 2 min. Such injei=tions repeated at 3-min intervals gave reproducible responses with no evidence of tachyphylaxis. Angiotensin responses are classed according to initial perfusion pressure because of the dependence of changes in resistance on that variable (12). In Table 1, raw data from these experiments are presented. Average control and prolonged vasodilation responses are shown for each ZO-mmHg range of initial perfusion value. In Fig. 2 the normalized (see METHODS) peak perfusion pressure increase in response to 0.5 pg of angiotensin is plotted against initial perfusion pressure which increased gradually during the recovery period following a ZO-min stimulation period at 4 twitches/s. Once again, the responses to angiotensin observed during prolonged vasodilation are compared to control responses to the same dose of angiotensin at the same level of initial perfusion pressure. No systematic differences were observed in the pre- and postcontrol response, and they have been averaged together in Fig. 2. The vasoconstrictor response to angiotensin is not attenuated during prolonged vasodilation compared to conM8Un
fsEM
n=4 KGllOOG
4 -
TENSION
o-
PRU
6 4 *
0 5r
7-T
f
T
T
ML O,/
100G/MlN
-
0
r _
320
OSMOLALITY
nOSM/l3O0 280 6 MMOlESl L 3 5
MOLES/L
3
t
0
1
1
I
I
I
20 40 MINUTES FIG. 3. Average alterations in isometric tension development, vascular resistance, vo2, venous osmoIality, venous K+, and venous lactate during and following 20 min of stimulation of calf muscle preparation at 4 twitches/s with flow held constant at 23 & 2 ml/100 g per min. Horizontal reference lines in top 3 traces represent an extension of control values and in bottom 3 traces arterial concentrations. Bars indicate + SE.
MOHRMAN,
AND
SPARKS
trol responses at the same initial perfusion pressure level. No statistical difference exists between the two groups of data shown in Fig. 2 or between the two groups of raw data in Table 1, within any range of initial perfusion pressures tested, which span the entire range of resistances observed during prolonged vasodilation. Twenty-minute stimulation fieriod. Figure 2 shows values for the venous concentration of metabolites measured during and following a ZO-min stimulation period of the muscle at 4 twitches/s with flow held constant at an average of 23 ZII 2 ml/ 100 g per min (n = 4). In the top trace it can be seen that tension declined from 6.4 to 1.9 kg/ 100 g within the stimulation period. A prolonged vasodilation was observed following this exercise as shown in the second trace. Also shown in Fig. 3 are the accompanying changes observed in muscle oxygen consumption, venous plasma K+ venous plasma osmolality, and lactate concentration. HoLontaI reference lines in the top three traces indicate control values, and the horizontal lines in the lower three traces indicate measurements of arterial concentration. Since the arterial concentrations of all the metabolites measured remained essentially constant during both stimulation and recovery periods and since flow was constant throughout, the venous concentrations illustrated are directly related to metabolite release or uptake. At the onset of stimulation there was a marked decrease in resistance, and there were significant changes in the production of all the chemical variables we measured. The vasodilation remained constant throughout the stimulation period, whereas the initial large chinges of some of the measured metabolites were not maintained. This is especially true of venous osmolality, which returned to values not significantly different from control during the stimulation period. This is not true, however, for venous 02 content which fell to 1.1 & .3 vol % at the onset of stimulation and remained at this level for the remainder of the stimulation period. Following cessation of the fatiguing exercise, resistance recovery is monoexponential with an average half-time of 6.9 & 0.6 (r = .97 & 0.1, P = .Ol) min and thus returns to control only after more than 25 min. As shown in the third trace oxygen consumption’s return to control following the stimulation period consists of two phases. Dur’ing the first phase, oxygen consumption recovers much more rapidly than re&tance and returns to near control level with a half-time of 60 s. During the second phase, oxygen consumption returns to control much more slowly, but it is only elevated by less than +5 ml 0 2/ 100 g per min over resting level. In Fig. 4, a curve is shown relating changes in ~OZ during mild exercise to changes in vascular resistance. Averages of percent control resistance are plotted against average oxygen consumption for six dogs. Figure 4 shows that changes in oxygen consumption from resting level up to 1.O ml O,/ 100 g per min during mild exercise are not associated with lowered vascular resistance. Within 2 min after cessation of stimulation causing fatiguing exercise, the vo 2 level is less than 1.O ml 0 e/ 100 g per min. Thus prolonged vasodilation remains for at least 18 min after oxygen consumption is at a level that is not associated with resistance changes during mild exercise. As shown in the fourth trace (Fig. 3) venous osmolality
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on January 10, 2019.
PROLONGED
41
VASODILATION
is not statistically different (P = .79) from arterial levels after 20 min of stimulation. When stimulation is stopped, it quickly falls 5 mosM below the arterial value and remains below control throughout the recovery period. The fifth trace shows that when the stimulation is stopped, venous K+ concentration returns to the arterial value within 30 s and at 1 min is .33 mM below arterial and then returns slowly toward arterial concentration during the remainder of the recovery period. As shown in the lowest trace, lactic acid remains elevated in the venous effluent from the muscle long after exercise has ceased. It declines from a venous concentration of 4.5 mM at the end of the stimulation period to the venous control level of 1.7 mM after 30 min of recovery in a nearly monoexponentia1 manner with a half-time of 6.3 mm. Sixty-minutes stimulation period. The protocol for these experiments wds analogous to the previous set of experiments except the exercise period was extended to 60 min n = 3). The results are presented in Fig+ 5. We observed ( decreasing muscle tension development (upper trace), maintained vasodilation during the exercise period, and prolonged vasodilation following it (second trace) as in the ZOmin stimulation period. As in the ZO-min experiments (Fig* 3) oxygen consumption returns to a level not associated with resistance changes during mild exercise within 2 min. Both venous osmolality and K+ are below control during the recovery period. During this longer stimulation period venous lactate concentration declines from 3.5 mM at 20 min to a level not statistically different from venous control by the 60th min of exercise (P = -98). This is in marked contrasti to the previous set of experiments when venous lactate concentration was at a high level at the end of stimulation (see Fig. 3). Following 60 min of stimulation venous lactate concentration is not statistically different from venous control at any point during the recovery. The absence of lactate output during the recovery phase had very little efTect on the prolonged vasodilation. The half-time of resistance recovery was decreased from 6.9 & .6 tin following the ZO-min stimulation experiments in which lactate production was high to 5.8 & .4 min (r = .99, P = .0002) after the 60-min stimulation exDeriments
0
2
4
6
FIG. 4. ReIationship between changes in muscle oxygen consumption and normalized resistance change observed in 6 animals. vo, was varied by mild exercise (0-3 twitches/s). Percent control resistance was calculated by dividing resistances at all oxygen consumptions by resistance at resting oxygen consumption. Fiow averaged 26 + 3 ml/100 g per min. All points consist of averages of 9 cases except 2 initial points which consist of averages of 5 cases. Bars indicate Z/I SE.
Meon
? SEM n=3
6
PRU 4 2
5 ML 02/ lOOG/MIN 0
320r TT
280
L 6
MMOLWL 3 4
MMOlESIL
LACTATE I I 0
1
1
1
I
I
30
I
60
1
I
90
MINUTES 5. Average alterations in isometric vascular resistance, Tjo2, venous osmolality, lactate during and following 60 min of preparation at 4 twitches/s with flow held g per min. Horizontal reference lines in extension of control values and indicate bottom 3 traces. Bars indicate & SE. FIG.
tension development, venous K+, and venous stimulation of calf muscle constant at 23 + 2 ml/100 top 3 traces represent an arterial concentrations in
in which there was not excess lactate production. This is still 20 times slower than the resistance recovery observed after nonfatiguing exercise. The decrease in the mean half-time of resistance recovery from 6.9 to 5.8 min was not statistically significant (P = JO). DISCUSSION
A prolonged vasodilation lasting for periods up to 30 min followed the fatiguing muscle exercise used in this study. Similar prolonged vasodilations have been reported by others following 5-6 min of exercise with flow completely arrested (1, 2). The recovery of vascular resistance following exercise with flow restricted or arrested is more than an order of magnitude slower than following more normal exercise situations in which flow is allowed to vary (13) or when the exercise is mild in relation to a constant flow (9, 10). Since we did not find evidence of a decrease in the ability of the vacular bed to vasoconstrict to angiotensin during prolonged vasodilation, we believe it is unlikely that prolonged vasodilation is a manifestation of an impairment of the vascular smooth muscle itself (e.g., lack of energy supply for contraction) caused by some long-lasting vasodilator influence. During the stimulation period steady-state J&Z was 4. I
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on January 10, 2019.
42 & .5 mI/lOO g per min for the 20-min period and 3-6 & 4 ml/ 100 g per min for the 60-min period. In four dogs we allowed Aow to vary, and during stimulation at 4 impulses/s AOW was 84 & 7 ml/ 100 g per min and Vo 2 was 14 & 3 ml/100 g per min. With constant flow at 23 =t 2 ml/l00 g per min Voz is limited and the prolonged vasodilation could be viewed as reactive hyperemia in response to a period of partial occlusion. Our egorts to determine the mechanism of the prolonged vasodilation resulted from the assumption that the low constant flow relative to demand might result in prolonged activation of one of the vasodiIator mechanisms associated with exercise hyperemia. We approached the problem of identification of the mechanism by using three of the criteria suggested by Mellander and Johannson (7): 1) the substance should be released from skeletal muscle, 2) it should be a vasodilator in appropriate concentrations, and 3) the substance should be released with a time course which corresponds to the time course of vascular resistance changes. Muscle oxygen consumption returned to control much more rapidly (Z+ = 1 min) than did vascular resistance CT 112= 6.9 min). The same result was observed by Cerretelli et al. (12) when flow was released after a 6- to IO-min period of exercise without flow. Although %%z initially returns toward control much more quickly than vascular resistance, ii02 does remain slightly above control for almost as long as vascular resistance is decreased (Figs. 3 and 5). If some factor related to the slight elevation in ~OZ (e.g., vessel wall POT) is responsible for prolonged vasodilation, this should be reflected in the relationship between VOZ and vascular resistance in other situations. That is, a similar “dose-response curve” should apply. For this reason we examined the relationship between vo2 and vascular resistance during exercise not resulting in prolonged vasodilation (Fig. 4). During the last 18 min of prolonged vasodilation, vo2 is at a level which is not associated with vasodilation during mild exercise; therefore, we conclude that voz cannot be causally associated with prolonged vasodilation in any direct way. It is possible that the minimally elevated vo 2 enhances the vasodilation resulting from some other agent released during prolonged vasodilation, but this possibility can only be tested after the agent is identified. In the study of Cerretelli et al. (Z), tissue POZ was probably well above control during the prolonged vasodilation because muscle oxygen consumption is at the resting level while flow is above the resting value for 20 min. I3erne et al. (1) f ound evidence for the release of adenosine from the muscle 5 s after flow was reinstated following a 5-min period of exercise with flow arrested. Although we did not measure adenosine in this study, we believe it is unlikely that it plays a role in prolonged vasodilation because the tissue is not hypoxic during the recovery period and adenosine production is thought to be closely linked to tissue hypoxia (1). In addition, this substance is not a long-acting vasodilator, presumably as a result of well-developed enzymatic pathways for its destruction in the interstitium, capillary wall, and blood (1). Thus we would not expect adenosine vasodilation to continue 20 min after its release during the stimulation period. A final conclusion on this possibility must await measurements of tissue adenosine during prolonged vasodilation.
MORGANROTH,
MOHRMAN,
AND
SPARKS
Both venous osmolality and K+ concentration are below control levels during the recovery period when prolonged vasodilation occurs. These changes are opposite those aherations that cause vasodilation (4, 6). We conclude that it is unlikely that prolonged vasodilation is the result of alterations in either K+ or osmolality. This, of course, does not exclude the participation of these factors in the vasodflation present during the exercise period. There is ample evidence that elevated Kf and osmolality (4, 6, 10) participate in exercise hyperemia. During the 60-min stimulation period (Fig. 4), the initial rapid fall in vascular resistance in the first minutes of exercise could be caused by some factor related to increased oxygen consumption, increased osmoiality, increased K+, or a combination of these influences. By the 60th min of stimulation, however, neither osmolality nor K + differs from control sufficiently to explain the persistent maximum vasodilation (4, 6, 15, 16). voa, however, remains at a plateau throughout the stimulation period and thus is not dissociated from the resistance changes during the stimulation period. Assuming elevated K+ and osmolality do cause vasodilation, their relative participation in causing the resistance decrease clearly must diminish during the period of stimulation. This, in addition to showing the Iack of participation of either K+ or osmolality in the prolonged recovery vasodilation, supports the idea that vasodilator influences may assume different relative roles in causing exercise hyperemia in different expe rimental situations or even at va rious times within a single experimental maneuver (9, 10). Our finding that lactate release from the muscle may continue long after exercise has ceased supports the findings of Cerretelli et al. (2) who found net lactate output for 30 min following a 5- to IO-min period of exercise with flow arrested. Muscle lactate production does not appear to be causally related to prolonged vasodilation, since in our experiments involving 60 min of muscle stimulation, prolonged vasodilation was present in the absence of lactate production. Our observation that lactate production declines or ceases during a long period of muscle stimulation is similar to the findings of Stainsby and Welch (i8) who report initial lactate production declining or even changing to uptake during the course of 60-min stimulation experiments. In summary, prolonged vasodilation lasting up to 30 min follows periods of heavy muscle exercise during which flow is either restricted or arrested. In such instances the recovery of vascular tone after exercise is 10 or more times slower than that normally observed with exercise hyperemia. We have been unable to identify the mechanism or mechanisms responsible for this prolonged vasodilation, but our experiments indicate that it is unlikely that impairment of vascular muscle contractile ability, K+ release, smooth or fa ctors related increased osmolali .ty, lactate production, to increased tissue oxygen consumption are causal agents in However, our data also indicate prolonged vasodilation. that the vasodilation probably does result from the presence of some unidentified vasodilator substance in the environment of the vascular smooth muscle, since the response to angiotensin is unimpaired when compared to control. A full understanding of the mechanism of prolonged vaso-
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on January 10, 2019.
PROLONGED
43
VASODILATION
dilation following fatiguing exercise may be useful in understanding other long-term vascular responses, such as the response to tissue injury and circulatory collapse during hemorrhagic shock.
This work was supported by Public Health Service Grant HL-14516, a fellowship from the Michigan Heart Association, and the University of Michigan Medical School Fund for Computing. Received
for publication
15 October
1974.
REFERENCES 1. BERNE, R., R. RUBIO, AND J. DOB~ON. Adenosine and adenine nucleotides as possible mediators of cardiac and skeletal muscle blood flow regulation, Circulation Res. 28, Suppl. I : 115-l 19, 197 1 m 2. CERRETELLI, P., P. DIPRAMPERO, AND J, PIIPER. Energy balance of anaerobic work in the dog gastrocnemius muscle. Am. J. Physiol. 217: 581-585, 1969. 3. GUYTON, A.,J. Ross, 0. CARRIER, AND J. WALKER. Evidence for tissue oxygen demand as the major factor causing autoregulation. Circulation Res, 25, Suppl. I : 160-169, 1964. 4. KJELLMER, I. The potassium ion as a vasodilator during muscular exercise. Acta Physiol. &and. 63 : 460-466, 1965. 5. KRAMER, K., 1;‘. OBAL, AND W. QUENSEL. Untersuchungen iiber den Muskelstoffwechsel des Warmbliiters. III. Mitteilung. Die Sauerstoffaufnahme des Muskels wtihrend rhythmischer Tztigkeit. PcfIuegers Arch. 241 : 7 17-729, 1939. 6. LWNDVALL, J. Tissue hyperosmolality as a mediator of vasodilation and transcapillary fluid flux in exercise skeletal muscle. Acta Physiol. &and. 86 : l-142, 1972. 7. MELLANDER, S., AND B. JOHANSSON. Control of resistance, exchange and capacitance functions in the peripheral circulation. Pharmacol. Rev. 20 : 117-l 96, 1968. 8. MELLANDER, S., B. JOHANSSON, S. GRAY, 0. JONSSON, J. LUNDVALL, AND B. LJUNC. The effects of hyperosmolality on intact and isolated vascular smooth muscle; possible role in exercise hyperemia. Angiologica 4 : 3 10-322, 1967. 9. MOHRMAN, D., J. CANT, AND H, SPARKS. Time course of vascular
10, 11.
12.
13.
14. 15.
16 l
17* 18.
resistance and venous oxygen changes following brief tetanus of dog skeletal mr,rcNcle. Circulation Res. 33 : 323-336, 1973. MOHRMAN, D., AND H. SPARKS. Role of K+ in the vascular response to brief tetanus. Circulation Res. 35 : 384-390, 1974. MOLNAR, J., J. SCOTT, E. FROHLICK, AND F:. J* HADDY. Local effects of various anions and I-I+ on dog forelimb and coronary vascular resistance. Am. J. Physiol. 203 : 135-142, 1962. MYERS, H., AND C. HONE. Influence of initial resistance on magnitude of response to vasomotor stimuli. Am. J. Physiol, 216: 1429-1436, 1969. PIIPER, J., P. E, DLPRAMPERO, AND P. CERRETELLI, Oxygen debt and high energy phosphates in gastrocnemius muscle of the dog. Am. J. Physiol. 215 : 523-53 1, 1968. RUBIO, R., R. BERNE, AND J, DOBSON. Sites of adenosine production in cardiac and skeletal muscle. Am. J, Physiol. 225: 938-953, 1973. SCOTT, J., AND D. RADOWSKI. Role of hyperosmolality in the genesis of active and reactive hyperemia. Circulation Res. 28, Suppl. I: 26-32, 1971. SKINNER, N., AND W. POWELL. Action of oxygen and potassium ion on the vascular resistance of dog skeletal muscle. Am. J. Physiol. 2 19 : 533-540, 1967. SNEDECOR, G., AND W. COCHRAN. Statistical Methods. Ames: Iowa State College Press, 1967. STAINSBY, W., AND H. WELCH. Lactate metabolism of contracting dog skeletal muscle in situ. Am. J. Physiol. 2 11: 177-183, 1966.
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on January 10, 2019.