An α-adrenergic constrictor tone is present in the coronary vasculature and limits the extent of coronary dilation and oxygen delivery during exercise(2,6,10,12,13,16,18,20,21,28). Many investigations suggest that in the exercising dog, thisα-constrictor tone is due primarily to activation ofα1-adrenergic receptors(2,6,10,14,16,31). Studies suggest α2-receptor inhibition results in very little effect on exercise-induced coronary hyperemia in dogs with normal coronary arteries, since there is little, if any, difference in the effects ofα1- versus nonselective α1- plusα2-receptor blockade on coronary blood flow during exercise(2,6,10,31). However, whether this coronary constrictor tone is beneficial or deleterious to ventricular contractile function is controversial. On the one hand, Gwirtz and coworkers have reported that during strenuous exercise(14,16,31) and during intense sympathetic stimulation (15), the restriction of coronary vasodilation by an α-adrenergic constrictor tone is associated with a limitation of left ventricular subendocardial contractile function. Unfortunately, these studies examined the influence of anα1-adrenergic constrictor tone at only one level of exercise. In contrast, Huang and Feigl (24) reported that anα-coronary constrictor tone has a beneficial effect during exercise by redistributing blood flow toward the endocardium. These investigators found that the α-coronary constrictor tone preserves endocardial blood flow at various levels of myocardial oxygen consumption. Since the ventricular endocardium appears to be more prone to ischemia than epicardial layers due to greater compressive forces, greater degree of fiber shortening and tension development, and less coronary reserve in this region(5,22,30), it would seem that increased endocardial flow during exercise would result in increased ventricular contractile function. However, Huang and Feigl (24) did not measure regional or global contractile function to evaluate whether myocardial function is affected by the presence and/or removal of theα-constrictor tone.
One interpretation of previous data is that the effect of anα1-coronary constrictor tone on left ventricular contractile function varies with intensity of exercise. During physical stress, theα-constrictor tone may limit total coronary blood flow, but distributes a greater percentage of the remaining blood flow toward the subendocardium, where effects of ventricular compression of coronary vessels are greatest, thus preserving subendocardial contractile function without compromising subepicardial function (5,22,30). In contrast, an α1-coronary constrictor tone could restrict total coronary flow to the point that the transmural energy supply:demand ratio would be deleteriously affected. As a result, regional and global myocardial contractile function would be flow-restricted. Under these conditions, blockade of α-coronary constrictor tone would increase flow and, as a consequence, remove the flow-limitation on contractile function and improve performance of the left ventricle.
The present study in dogs was designed to examine the hypothesis that anα1-receptor mediated coronary constrictor tone exists transmurally in the left ventricle and imposes a significant limitation on myocardial blood flow and, as a result, myocardial contractile function. We also propose that the magnitude of this effect increases with increasing intensities of myocardial oxygen demand imposed by exercise. Left ventricular contractile function and coronary flow were measured during graded treadmill exercise during control conditions and after selectiveα1-coronary constrictor tone blockade with prazosin. To avoid the possibility of undetected changes in regional contractile performance unaccompanied by changes in global contractile performance, both regional and global contractile function were monitored. This study was limited to examination of a coronary α1-constrictor tone, since all previous studies during exercise from this laboratory indicate that the sympathetic constrictor tone in the coronary vasculature is mediated by postsynapticα1-receptors.
Mongrel dogs of either sex weighing between 25 and 35 kg were anesthetized using acepromazine (0.5 mg, s.c.) and sodium pentobarbital (30 mg·kg-1, i.v.). The trachea was intubated, and each dog was ventilated using a Harvard Model 614 Respirator. Using sterile technique, a thoracotomy was performed through the left fifth intercostal space. The aorta was exposed and cannulated distal to the aortic arch. This catheter was used to monitor arterial pressure (AP) and withdraw arterial blood samples. The heart was then exposed and suspended in a pericardial cradle. To measure left ventricular pressure (LVP), a Konigsberg P-6.5 solid-state micromanometer and a fluid-filled Tygon catheter (1.27-mm OD) was inserted into the left ventricle through a stab wound in the apex. At the beginning of each experiment, the Tygon catheter was connected to an Isotec® pressure transducer, and the Konigsberg micromanometer was calibrated against the catheter pressure. The Isotec pressure transducer was calibrated using a mercury manometer. The circumflex artery was dissected free of the surrounding tissue for approximately 3 cm beginning at the origin of the vessel. For measurement of coronary flow velocity (CFV), a 4-mm ID 10-MHz Doppler flow probe was placed around the circumflex artery. To check the zero flow reference, a pneumatic occluder was placed around the circumflex artery immediately distal to the Doppler flow transducer such that there was no vessel branch between the two. For injection of the solutions into the circumflex artery, a heparin-filled Silastic catheter (0.12-mm ID and 0.6-mm OD) was inserted into the circumflex artery distal to the occluder. This catheter has been shown not to interfere either with flow in this vessel or with its distribution during conditions of rest or exercise(12). To permit coronary venous blood sampling, a Silastic catheter (0.94 mm OD) was placed in the coronary sinus. For measurement of regional myocardial contractile function, two pairs of opposing 5 MHZ ultrasonic crystals were placed in the circumflex (posterior) perfusion territory, and one pair was placed in the left anterior descending (LAD, anterior) perfusion territory. Each pair of crystals was implanted 0.5-1.0 cm apart and 0.5-0.7 cm below the epicardial surface to monitor contractile function by measuring changes in segment length (SL) in the subendocardial layer (32). Placement of segment length transducers was confirmed by postmortem analysis.
After instrumentation was completed, a chest tube was placed in the thoracic cavity to evacuate the pneumothorax and any post-surgical intrathoracic exudate accumulation. All wires and catheters were passed subcutaneously to exit between the scapula. The indwelling catheters were flushed daily with heparinized saline to maintain patency. Post-operative analgesics, antibiotics, and antipyretics were given as specified by the veterinarian.
At the conclusion of all experimentation, the dogs were sacrificed using an overdose of sodium pentobarbital, followed by potassium chloride (KCl). Injection of Evans blue dye was injected into the circumflex catheter (while the occluder was inflated to prevent back diffusion of dye) to delineate the posterior perfusion territory. The weight of the posterior perfusion territory was used to convert Doppler flow and myocardial oxygen consumption to a per gram basis and to verify by visual inspection that the ultrasonic crystals were indeed placed in the circumflex perfusion region. Post-mortem examination showed no evidence of thrombus formation, local inflammation, or scarring from implanted crystals. The cross-sectional area of the circumflex artery was measured and used to convert Doppler flow velocity data to a volume flow.
Experimentation began 10-14 d following surgery. In nine dogs coronary blood flow and global and regional myocardial function were evaluated while the dog was subjected to a standardized submaximal exercise protocol described by Tipton et al. (33). This protocol consists of six levels of exercise of increasing workload or intensity, beginning with a 3-min warm-up at 4.8 kph, 0% grade. The treadmill speed was increased to 6.4 kph, and the incline was increased at 3 min intervals to include 0, 4, 8, 12, and 16% grades. Each dog performed this exercise protocol twice on separate days. A minimum of 24 h elapsed between exercise protocols. During these tests an intracoronary arterial infusion of either prazosin or its vehicle was performed to assess the magnitude of the α1-coronary constrictor tone at each level of exercise intensity. Prior to exercise, the vasoconstrictor response to an intracoronary 20-μg phenylephrine challenge was performed. This dose of intracoronary phenylephrine consistently elicits a substantial coronary constriction (17).α1-Adrenoceptor blockade of the circumflex perfusion territory was accomplished with a 0.5-mg dose of prazosin infused directly into the circumflex artery over 3.5 min. This dose of intracoronary prazosin has previously been shown to abolish the vasoconstrictor response to an intracoronary phenylephrine challenge of 20 μg without secondary peripheral circulatory effects, such as changes in heart rate or blood pressure(17). Blockade of α1-adrenergic receptors was maintained during the exercise regimen by continuous infusion of prazosin at a rate that did not exceed 1 μg·kg-1·min-1. Infusion of this α1-antagonist at a similar rate was shown by Dai et al. (10) to maintain effective blockade until completion of this exercise protocol. In the present experiments, the effectiveness of α1-adrenergic blockade at the end of the exercise protocol was verified in each dog by loss of the constrictor response to the phenylephrine challenge. Before blockade, 20-μg, i.c. phenylephrine caused a 16% decrease in coronary flow from 0.95 ± 0.1 to 0.80 ± 0.1 ml·min-1·g-1. No vasoconstrictor response to phenylephrine was observed after prazosin. During the second exercise protocol, the vehicle for prazosin (sterile water) was infused at the same rate as described for prazosin above. To avoid volume effects on coronary blood flow, the infusion rate of prazosin or sterile water vehicle never exceeded 0.50 ml·min-1 at rest, and was maintained at 0.25 ml·min-1 during the exercise regimen. It is important to note that infusion of the vehicle alone did not cause any cardiovascular response at this rate. Data were collected during the control period and during the final 30 s of each of the six exercise stages. Aortic and coronary sinus venous blood samples were also obtained at rest and at each level of exercise.
Regional Perfusion Study
To measure transmural myocardial blood flow distribution, a separate group of four dogs were instrumented as described above, with the addition of a catheter placed into the left atrial appendage for injection of tracer microspheres. The aortic catheter was used for withdrawal of reference samples. Transmural flow distribution was determined using three different species of 15-μm diameter microspheres (57Co, 46Sc and113 Sn, New England Nuclear, Boston, MA) in 10% dextran with 0.014% Tween 80. Each microsphere dose (2-4 million, depending on specific activity) was diluted with 10-ml normal saline, sonicated for 30 min, and vortexed prior to infusion. After heparinization of the dog (500 U·kg-1), control recordings were made and microspheres injected. The dogs were then exercised as described above. A second species of microspheres was injected while the dogs were running at 6.4 kph, 16% incline. Prazosin (0.5 mg, i.c.) was injected while the dogs continued running and the third species of microspheres were injected 1 min after prazosin during exercise. Arterial reference blood samples were collected using the aortic catheter at a constant rate of 15.0 ml·min-1 for 2 min beginning 10 s before injection of the microspheres. At the conclusion of each experiment the left ventricle was excised and samples taken from the region perfused by the left anterior descending and circumflex coronary arteries. Each piece was divided into subendocardial and subepicardial layers of equal thickness. These tissue samples weighed 2.41 ± 0.3 g. Tissue samples and blood reference samples were analyzed for radioactivity using a Canberra Instruments gamma spectrometer. Isotope separation and blood flow computations were performed with an IBM computer. Blood flow was calculated from the equation of Heymann et al. (19) as follows: MBF = Fr × Rt/Rr, where MBF is myocardial blood flow (ml·min-1), Fr is the reference blood sampling rate (ml·min-1), and Rt and Rr are the radioactivites (counts·min-1) of the tissue and reference samples, respectively. MBF was divided by the tissue sample weight and is reported as ml·min-1 per gram of tissue. The subendocardial/subepicardial flow ratio was calculated for each of the transmural samples obtained from the anterior and posterior regions. Blood flow was also measured in left and right kidneys, left and right gastrocnemius muscles and left and right lower lobes of the lungs to verify adequate mixing and distribution of microspheres and absence of shunting.
Data Collection and Analysis
On-line variables were recorded on a Coulbourn eight-channel chart recorder and on an eight-channel Hewlett-Packard Model 3968A tape recorder for subsequent computer analysis. Recorded variables included LVP, SL in the posterior (PSL) and anterior (ASL) perfusion territories, AP, and CFV. Computer data analysis recorded data at 2-ms intervals over 10 consecutive beats. The following data were analyzed from the recorded variables: left ventricular systolic pressure (LVSP) and end-diastolic pressure (LVEDP), HR, mean AP, mean CFV, and SL. The program also determines dP/dt and dL/dt from the respective LVP and SL signals, and then derives the maximal rate of pressure generation (dP/dtmax) and segment length shortening(-dL/dtmax). End-diastolic (EDL) and end-systolic (ESL) segment lengths were derived by referring to the dP/dt signal as described by Theroux et al.(32). Percent segment length shortening (%SL) was calculated as 100 × (EDL - ESL)/EDL.
To convert CFV to volume rate flow, the circumflex artery diameter within the Doppler probe was measured post mortem. Volume rate was then calculated as: Equation
where D = measured diameter (cm), and CFV = coronary flow velocity(cm·s-1). Myocardial oxygen consumption (M˙VO2) was determined using the Fick equation: Equation
where CBF is circumflex blood flow(ml·min-1·g-1), AO2 is coronary arterial oxygen saturation, ˙VO2 is coronary venous oxygen saturation, Hgb is hemoglobin concentration (g·ml-1 blood), and 1.34 is the average O2 carrying capacity of Hgb. The arterial and coronary venous oxygen saturations and hemoglobin concentration were determined using a Radiometer OSM 2 Hemoximeter. Calculation of M˙VO2 using CBF to the posterior perfusion territory (as measured in this study) and coronary sinus venous oxygen saturation (which reflect oxygen extraction from both the anterior and posterior perfusion territories) results in an underestimation of M˙VO2 by the posterior region of the left ventricle. Nevertheless, an increased oxygen extraction would be expected if the increase in contractile performance were the primary event and precipitated an increase in circumflex blood flow. Likewise, a decrease in oxygen extraction would be expected if the increase in circumflex blood flow were the primary event, and the increase in contractile function were the secondary event.
To determine the effect of an α1-adrenergic constrictor tone on coronary flow and left ventricular contractile function, values for LVSP, dP/dtmax,% SL, -dL/dtmax, AP, HR, and CBF were compared at rest and during each level of exercise using two-way analysis of variance (ANOVA) with repeated measures. In these analyses, factors were A, with or without prazosin, and B, exercise level. If the ANOVA detected significant differences within factor means, then these differences were identified with the Student paired t-test with repeated measures for Factor A and with the Student Neuman-Keuls posthoc test for Factor B. A least-squares regression line to fit the observations of CBF, dP/dtmax, and dl/dtmax from each dog version M˙VO2 was performed. The individual regression lines from each dog were summarized by multiple linear regression. The difference in regression slopes were compared by pairedt-tests. Regional blood flow data obtained from the posterior left ventricular region were compared with those obtained from the anterior region using Student's paired t-test. Each dog served as its own control. All values are expressed as mean ± standard error, and statistical significance was accepted at P < 0.05.
These studies were conducted in conformance with the policy statement of the American College of Sports Medicine on research with experimental animals.
The effects of intracoronary prazosin on HR, mean AP, and left ventricular global contractile function at standing rest and at each level of exercise are shown in Table 1. There were no significant differences between the vehicle and prazosin tests in the responses of HR, mean AP, LVSP, and LVEDP, all showing typical responses to graded submaximal exercise. dP/dtmax increased significantly at each level of exercise in both groups, but significantly greater increases at the two highest exercise intensities were evident when prazosin was given (P < 0.05).
The effects of intracoronary prazosin on CBF, left ventricular oxygen extraction, and M˙VO2 during exercise are presented inTable 2. It should be noted that there was no difference in oxygen extraction during prazosin administration compared with vehicle administration at any level of exercise. Both showed similar increases at each level of exercise. Exercise was associated with an increase in mean CBF and M˙VO2 above resting values in both tests at all levels of exercise. Intracoronary administration of prazosin resulted in a significantly greater increase in CBF of 15 ± 7%, 24 ± 9%, and 35 ± 10%, respectively, at the three most strenuous levels of exercise (P < 0.05). These data are consistent with an α1-coronary constrictor tone, which increases in magnitude with increasing levels of exercise intensity. Resting M˙VO2 and mean CBF were similar before and afterα1-receptor blockade. M˙VO2 increased 2.4 times during submaximal exercise with vehicle infusion, and 3.3-fold during submaximal exercise with prazosin infusion. Furthermore, increased CBF without a change in oxygen extraction allowed for a greater M˙VO2 at the two most strenuous intensities of exercise during α1-adrenergic blockade as compared with during vehicle infusion (P < 0.05). A linear regression line was generated to fit the observations from each individual dog using M˙VO2 and CBF. The mean data ± SE from all dogs is depicted in Figure 1. At the mid-level and upper range of oxygen demands during exercise (when M˙VO2 was 250% of the resting value or greater), CBF was significantly greater during prazosin infusion versus vehicle infusion (P < 0.05). The slopes of this relationship were significantly different (F = 9.667; P = 0.0264), indicating a greater CBF at equivalent levels of M˙VO2.
Table 3 shows the effects of intracircumflex infusion of prazosin on regional left ventricular contractile function in the anterior and posterior regions. Exercise was never associated with a change in EDL compared with resting values in either region. Compared with resting values, exercise was always associated with increases in contractile function in both the anterior and posterior regions, as indicated by increases in%SL and in-dL/dtmax. Compared with vehicle infusion, regional prazosin infusion into the posterior left ventricle was associated with greater increases in contractile function in the posterior region, as indicated by greater increases -dL/dtmax of 36 ± 12%, and 42 ± 13% at exercise intensities of 6.4 kph, 12%, and 16% inclines, respectively. It should also be noted that regional infusion of prazosin into the posterior left ventricle did not alter contractile function in the left ventricular anterior region. These data suggest that the α1-adrenergic blocking effects of prazosin infused into the left circumflex artery were limited to the left circumflex artery perfusion territory and that little or no recirculation of prazosin occurred. A linear regression line was also generated to fit the observations from each dog, using M˙VO2 and dP/dt max and posterior regional-dL/dtmax. As shown in Figures 2 and 3, at the mid-level and upper range of oxygen demands during exercise, dP/dtmax and -dL/dtmax were significantly greater during prazosin infusion versus vehicle infusion (P < 0.05). Similar to the response found for CBF in Figure 1, the slopes of these relationships were significantly different, indicating a greater dP/dtmax (F = 9.200; P = 0.0287) and-dL/dtmax (F = 4.341; P = 0.0544) at equivalent levels of M˙VO2.
Table 4 presents values for regional myocardial perfusion and transmural flow distribution during control conditions, during submaximal exercise at 6.4 kph, 16% incline, and exercise after intracoronary administration of prazosin. It can be seen that in both the anterior and posterior regions, exercise increased flow equally in both the subepicardial and subendocardial layers of the left ventricle such that the subendocardial/subepicardial flow ratios were not altered. It can also be seen that prazosin administration during exercise further increased perfusion in the posterior region. The increase in myocardial flow was approximately equal in both the subepicardial and subendocardial layers such that the flow ratio was not altered. Therefore, these data suggest that the constrictor tone was uniform through the left ventricular wall at the highest level of submaximal exercise tested.
This study found that as an α1-coronary constrictor tone associated with graded treadmill exercise became increased, there was a progressively greater flow-restriction on left ventricular function. Thisα1-constrictor tone was uniform through the left ventricular wall during strenuous exercise. Increased left ventricular contractile function after α-blockade was observed regionally and globally at more strenuous intensities of exercise. Thus, in addition to the existence of a coronary reserve, there appears to be a myocardial contractile reserve. It is also unknown whether both are called upon or still exist at maximal levels of exercise. It is unknown under what circumstances we fully use a coronary (and perhaps a contractile) reserve-certainly not during conditions of myocardial ischemia(1,3,5,7,22,27).
Previous evidence from this laboratory indicate that anα1-adrenergic coronary vasoconstriction opposes metabolic coronary vasodilator mechanisms and, as a result, minimizes or restricts increases in ventricular contractile function associated with strenuous exercise (13-16,31). However, these studies did not evaluate the effects of coronaryα1-blockade at less strenuous levels of exercise. Dai et al.(10) also demonstrated that blockade of a coronaryα1-constrictor tone was associated with improved global left ventricular contractile function (dP/dtmax) in dogs running at 6.4 kph, 15% grade. However, at greater or lesser intensities of exercise, the observed increase in global left ventricular contractile function after removal of coronary constrictor tone was not statistically significant. The reason for this difference in global function between this study and the study by Dai et al. is unknown. It is likely that global measurements in ventricular contractile function may not accurately reflect regional changes in contractile function associated with regional blockade ofα1-coronary constriction. That is, increases in regional segmental work can be seen in the setting of global decreases in external work(26). Increases in regional contractile function consequent to blockade of α1-coronary constriction can be missed if global function is analyzed without any measure of regional contractile performance. For this reason, our measurements of regional contractile function may more accurately reflect the effects of regional changes in ventricular perfusion than do measurements of global contractile function. The present study found a significant increase in regional contractile function(dL/dt) after α1-blockade at higher exercise intensities. No significant increase in percent segment shortening occurred due to greater variability between dogs in this measurement.
An increase in subendocardial contractile function that occurred without a concomitant change in subepicardial contractile function afterα1-adrenergic receptor blockade during exercise may be explained by one of two mechanisms. First, an α1-adrenergic constrictor tone could be uniformly distributed across the ventricular wall such that abolition of this tone causes an equal increase in both subepicardial and subendocardial perfusion. Functional limitation would then be limited to the sub-endocardium only if this layer had relatively less vasodilatory reserve than the subepicardium. Second, a non-uniform sympathetic coronary constrictor tone across the left ventricular wall such that the constrictor influence is greater in the subendocardial layer could limit contractile function in this layer. The results of the regional perfusion experiments demonstrate a uniform effect of α1-adrenergic blockade on myocardial perfusion across the left ventricular wall. As a result, it is likely that the increase in contractile function caused by prazosin in only the subendocardial layers may be due to the fact that even though an α-constrictor tone was uniform across the entire left ventricular wall, it imposed a limitation on contractile function only in the deeper, more ischemiaprone layers. The combined effects of a greater degree of shortening and tension development by the subendocardial myocytes compared with the subepicardial myocytes(30), a smaller vasodilator reserve in the subendocardium (5), and a shorter duration of perfusion during each cardiac cycle in the subendocardium versus the subepicardium(5) make the subendocardial myocytes more prone to ischemia than are the subepicardial myocytes.
However, considerable controversy exists in the literature regarding the transmural distribution of a coronary adrenergic constrictor tone. The uniform transmural nature of a coronary α1-adrenergic tone found in this study is consistent with studies by others(4,9,15,23,25,29). In contrast, several other investigators have favored either a subepicardial or a subendocardial predominance of an α-adrenergic coronary constriction. Huang and Feigl (24) proposed that the adrenergic constrictor tone may be more intense in the subepicardial muscle layers, such that it diverts blood to the more vulnerable deeper layers during exercise. They suggested that an α-adrenergic coronary constrictor tone results in a redistribution of blood flow transmurally across the ventricular wall, which tends to shunt blood toward the subendocardium as exercise intensity increases. In their study, regional blockade of coronary α-adrenergic receptors was produced by intracoronary administration of the nonselectiveα-blocker phenoxybenzamine. Although mean coronary blood flow during exercise was less in the unblocked region than in the blocked region, the ratio of subendocardial-to-subepicardial blood flow was better maintained in the α1-blocked region than in the unblocked region. They concluded that an α1-adrenoceptor-mediated coronary constrictor tone maintains blood flow to the subendocardial layers in spite of limiting mean transmural coronary blood flow. A similar nonuniform adrenergic constriction across the ventricular wall has also been reported by Chilian and Ackell (7) during exercise after coronary stenosis. However, we found no evidence for a nonuniform distribution ofα1-receptors in left ventricular coronary vasculature. The precise basis of the differences between our results and the results of Huang and Feigl (24) and Chilian and Ackell(7) is at present unclear and will require further study. It is important to note, however, that none of these studies addressed the possibility of a differential effect of a coronary α-constrictor tone on left ventricular contractile function.
In an open-chest preparation, Baumgart et al. (3) mimicked physiologic sympathetic stimulation with cardiac pacing and intravenous infusion of norepinephrine. They demonstrated that administration of phentolamine increased coronary blood flow and M˙VO2. However, prior coronary dilation with dypridamole altered the phentolamine-related changes in coronary blood flow and abolished the increase in myocardial oxygen consumption associated with phentolamine administration in this model. These findings, in conjunction with our data, suggest that the flow-related alterations on myocardial oxygen consumption we observed in running dogs receiving prazosin are likely to be achieved by other mechanisms of increasing transmural perfusion. Once some perfusion threshold is surpassed, additional increases in blood flow do not further raise M˙VO2.
As previously shown, a coronary α-constrictor tone is nonexistent in the resting conscious dog (8,16). In the present study, coronary blood flow increases linearly with exercise intensity and with myocardial oxygen demand (M˙VO2). The increase in coronary flow was substantially greater when prazosin was infused to blockα1-adrenergic receptors and remove a vasoconstrictor tone. As exercise intensity increased (with the concomitant increase in sympathetic stimulation of the heart), the magnitude of the α1-constrictor tone also increased. In the current study, both coronary flow and M˙VO2 were significantly elevated at the highest three levels of submaximal exercise after α1-blockade. Was the increase in coronary flow due to an increase in oxygen demand, or did the increase in flow permit a further increase in contractile function (16)? Examination of the relationship between coronary flow and M˙VO2 indicated that some other factor in addition to metabolic demand is playing a role in regulating coronary flow during exercise. If coronary flow was determined solely by oxygen demand, then the relationship between flow and M˙VO2 (i.e., slopes of the regression lines) would be similar afterα1-blockade. However, the slope of the relationship was significantly increased after prazosin, indicating that coronary flow increased more per unit myocardial oxygen demand compared to vehicle. Thus, a neurogenic factor limited coronary vasodilation during exercise, with the degree of limitation increasing with increased exercise intensity.
Arterial-venous oxygen difference did not change significantly with infusion of prazosin and M˙VO2 increased along with coronary blood flow. These data imply that coronary flow limited M˙VO2. Hence, the increase in coronary flow led to an increased oxygen delivery which enables the myocardium to increase contractile function, thereby increasing M˙VO2. Removal of the α1-mediated vasoconstrictor tone during exercise was associated with an increase in regional left ventricular contractile function in the circumflex perfusion territory, as well as in global LV contractile function. Gwirtz et al. (14) showed similar results when coronary dilation with intracoronary adenosine infusion during exercise caused an increase in contractile function of the same magnitude to that seen with prazosin infusion. Furthermore, prior ventricular sympathectomy abolishes the responses to prazosin on coronary flow and contractile function during exercise (13).
The possibility that the increased contractile function observed in the deeper myocardial layers after α1-adrenergic blockade was due to“Gregg effect,” rather than to an increased perfusion of flow-deficient myocardium, must also be considered (11). It has been reported that myocardial oxygen consumption and mechanical function can be increased by increases in blood flow, even in well-perfused hearts operating within the coronary auto-regulatory range(1,4). However, it should be noted that in the present experiments, α1-adrenergic blockade during exercise caused an equal increase in perfusion of subepicardial and subendocardial muscle, whereas an increase in contractile function was seen only in the deeper layers. These results suggest that the increased mechanical function was probably not due to the Gregg effect. The results are compatible with the proposal that, although the increased perfusion was transmural, O2 delivery and contractile function only in the deeper layers were flow limited during stellate stimulation. In this regard, it is important to recognize that the coronary vasodilatory reserve in left ventricular subendocardium is generally felt to be less than in subepicardium (5). Nevertheless, it cannot be discounted that the Gregg effect itself is not uniform across the ventricular wall, although a transmural distribution of the effect has never been explored.
In summary, this study demonstrates that the magnitude of theα1-vasoconstrictor tone, which modulates oxygen delivery to the myocardium and limits coronary vasodilation during exercise, increases with increasing exercise intensity. At higher exercise intensities it imposes a limitation on cardiac contractile function. Thus, it appears that a transmural coronary vasoconstrictor tone is a physiological response to a stress such as exercise. The precise physiological function of the presence of anα1-adrenergic constrictor tone on the coronary vasculature is not clear at the present time. It may be that the vasoconstrictor tone has no true physiological function, but is a consequence of the increased sympathetic outflow, which occurs during exercise. However, other investigators believe that the vasoconstrictor tone is an important mechanism regulating the oxygen supply/demand relationship across the left ventricular wall(7,24).
1. Able, R. M. and R. L. Reis. Effects of coronary blood flow and perfusion pressure on left ventricular contractility in dogs.Circ. Res.
2. Bache, R. J., X-Z. Dai, C. A. Hrzog, and J. S. Schwartz. Effects of non-selective and selective α1
-adrenergic blockade on coronary blood flow during exercise. Circ. Res.
61(Suppl. 2):II36-II41, 1987.
3. Baumgart, D., T. Ehring, P. Kowallik, B. O. Guth, M. Krajcar, and G. Heusch. Impact of α-adrenergic coronary vasoconstriction on the transmural myocardial blood flow distribution during humoral and neuronal adrenergic activation. Circ. Res.
4. Buffington, C. W. and E. O. Feigl. Effect of coronary artery pressure on transmural distribution of adrenergic coronary vasoconstriction in the dog. Circ. Res.
5. Braunwald, E., B. E. Sobel. Coronary blood flow and myocardial ischemia. In: Heart Disease: A Textbook of Cardiovascular Medicine
, 4th Ed., E. Braunwald (Ed.). Philadelphia: W. B. Saunders, 1992, pp. 1175-1192.
6. Chen, D. G., E. Z. Dai, B. G. Zimmerman, and R. J. Bache. Postsynaptic α1
- and α2
-adrenergic mechanisms in coronary vasoconstriction. Cardiovasc. Pharmacol.
7. Chilian, W. M. and P. H. Ackell. Transmural differences in sympathetic coronary constriction during exercise in the presence of coronary stenosis. Cric. Res.
8. Chilian, W. M., R. B. Boatwright, T. Shoji, and D. M. Griggs, Jr. Evidence against significant resting sympathetic coronary vasoconstrictor tone in the conscious dog. Circ. Res.
9. Downey, H. F., D. P. Grice, and C. E. Jones. Systemic hypoxia activates a coronary vasoconstrictor reflex response that is blocked by prazosin. J. Cardiovasc. Pharmacol.
10. Dai, X-Z, E. Sublett, P. Lindstrom, J. S. Schwartz, D. C. Homans, and R. J. Bache. Coronary flow during exercise α1
-adrenergic blockade. Am. J. Physiol. (Heart Circ. Physiol. 25)
11. Gregg, D. E. Regulation of the collateral and coronary circulation of the heart. In: Proceedings of the Harvey Tercentenary Congress. The Coronary Circulation
, J. McMichael (Ed.). Oxford, UK: Blackwell, 1958, pp. 163-186.
12. Gwirtz, P. A. Construction and evaluation of a coronary catheter for chronic implantation in dogs. J. Appl. Physiol.
13. Gwirtz, P. A., M. A. Brandt, A. F. Meintjes, and C. E. Jones. Coronary α1
-constrictor tone in exercise is caused by neuronal catecholamines. Coronary Artery Dis.
14. Gwirtz, P. A., J. M. Dodd-o, M. A. Brandt, and C. E. Jones. Augmentation of coronary flow improves myocardial function in exercise.J. Cardiovasc. Pharmacol.
15. Gwirtz, P. A., J. M. Dodd-o, H. F. Downey, et al. Effects of a coronary α1
-constriction on transmural left ventricular flow and contractile function. Am. J. Physiol.
16. Gwirtz, P. A., S. P. Overn, H. J. Mass, and C. E. Jones. α1
-Adrenergic constriction limits coronary flow and cardiac function in running dogs. Am. J. Physiol. (Heart Circ. Physiol. 19)
17. Gwirtz, P. A. and H. L. Stone. Coronary blood flow changes following activation of adrenergic receptors in the conscious dog.Am. J. Physiol. (Heart Circ. Physiol. 12)
18. Gwirtz, P. A. and H. L. Stone. Coronary blood flow and myocardial oxygen consumption after α-adrenergic blockade during submaximal exercise. J. Pharmacol. Exp. Ther.
19. Heymann, M. A., B. D. Payne, J. I. E. Hoffman, and A. M. Rudolph. Blood flow measurements with radionuclide-labeled particles.Prog. Cardiovasc. Dis.
20. Heyndrickx, G. R., P. Muylaert, and J. L. Pannier.α-Adrenergic control of oxygen delivery to myocardium during exercise in conscious dogs. Am. J. Physiol. (Heart Circ. Physiol. 11)
21. Heyndrickx, G. R., J. P. Vilaine, E. J. Moerman, and I. Luesen. Role of prejunctional α2
-adrenergic receptors in the regulation of myocardial performance during exercise in conscious dogs.Circ. Res.
22. Hoffman, J. I. E. and G. D. Buckburg. Transmural variations in myocardial perfusion. In: Progress in Cardiology
, Chapter 3, P.N. Yu and J. F. Goodwin (Eds.). Philadelphia: Lea and Febiger, 1976, pp. 37-89.
23. Holtz, J., E. Mayer, and E. Bassenge. Demonstration ofα-adrenergic coronary control in different layers of canine myocardium by regional myocardial sympathectomy. Pfluegers Arch.
24. Huang, A. H. and E. O. Feigl. Adrenergic coronary vasoconstriction helps maintain uniform transmural blood flow distribution during exercise. Circ. Res.
25. Johannsen, U. J., A. L. Mark, and M. L. Marcus. Responsiveness to cardiac sympathetic nerve stimulation during maximal coronary dilation produced by adenosine. Circ. Res.
26. Kedem, J., J. Sonn, M. Scheinowitz, and H. S. Weiss. Relationship between local oxygen consumption and local and external cardiac work: effect of tachycardia. Cardiovasc. Res.
27. Laxson, D. D., X-Z. Dai, D. C. Homans, and R. J. Bache. The role of α1
- and α2
-adrenergic receptors in mediation of coronary vasoconstriction in hypoperfused ischemic myocardium during exercise. Circ. Res.
28. Murray, P. A. and S. F. Vatner. α-adrenoceptor attenuation of the coronary vascular response to severe exercise in the conscious dog. Circ. Res.
29. Rinkema, L. E., J. X. Thomas, Jr., and W. C. Randall. Regional coronary vasoconstriction in response to stimulation of stellate ganglia. Am. J. Physiol. (Heart Circ. Physiol. 12)
30. Streeter, D. D., N. Varishnav, D. J. Patel, H. M. Spotnitz, J. Ross, and E. H. Sonneblick. Stress distribution in the canine left ventricle during diastole and systole. Biophys. J.
31. Strader, J. R., P. A. Gwirtz, and C. E. Jones. Comparative effects of α1
- and α2
-adrenoceptors in modulation of coronary flow during exercise. J. Pharmacol. Exp. Ther.
32. Theroux, P., J. Ross, Jr., D. Franklin, W. S. Demper, and S. Sasayama. Regional myocardial function in the conscious dog during acute coronary occlusion and responses to morphine, propranolol, nitroglycerin, and lidocaine. Circulation
33. Tipton, C. M., R. A. Carey, W. C. Eastern, and H. H. Erickson. A submaximaltest for dogs: evaluation of effects of training, detraining, and cage refinement. J. Appl. Physiol.