Ischemia of active skeletal muscle elicits a pressor response termed the muscle metaboreflex. We tested the hypothesis that in normal dogs during dynamic exercise, graded muscle metaboreflex activation (MMA) would progressively attenuate spontaneous heart rate baroreflex sensitivity (SBRS). The animals were chronically instrumented to measure heart rate (HR), cardiac output (CO), mean and systolic arterial pressure (MAP and SAP), and left ventricular systolic pressures (LVSP) at rest and during mild or moderate treadmill exercise before and during progressive MMA [via graded reductions of hindlimb blood flow (HLBF)]. SBRS [slopes of the linear relationships (LRs) between HR and LVSP or SAP during spontaneous sequences of ≥3 consecutive beats when HR changed inversely vs. pressure] decreased during mild exercise from the resting values (−5.56 ± 0.86 vs. −2.67 ± 0.50 beats·min−1·mmHg−1, P <0.05), and in addition, these LRs were shifted upward. Progressive MMA gradually and linearly increased MAP, CO, and HR; linearly decreased SBRS; and shifted LRs upward and rightward to higher HR and pressures denoting baroreflex resetting. Moderate exercise caused a substantial reduction in SBRS (−1.57 ± 0.38 beats·min−1·mmHg−1, P <0.05) and both an upward and rightward resetting. Gradual MMA at this higher workload also caused significant progressive increases in MAP, CO, and HR and progressive decreases in SBRS, and the LRs were shifted to higher MAP and HR. Our results demonstrate that gradual MMA during mild and moderate dynamic exercise progressively decreases SBRS. In addition, baroreflex control of HR is progressively reset to higher blood pressure and HR in proportion to the extent of MMA.
- exercise reflexes
- pressor response
- arterial baroreflex sensitivity
whole body dynamic exercise can elicit profound changes in autonomic activity. Two negative feedback reflexes implicated in mediating these responses are the arterial baroreflex and the muscle metaboreflex, which negate perturbations in arterial pressure and blood flow to active skeletal muscle, respectively (34–36). The arterial baroreflex operates via modulation of both cardiac and peripheral vascular function, but it does so with markedly different time courses. The buffering of rapid (seconds) changes in arterial pressure occurs via rapid, parasympathetically induced changes in heart rate (HR) that usually elicits proportional changes in cardiac output (CO). Ogoh et al. (25) showed that virtually all of the initial compensatory responses to transient activation of the carotid sinus baroreceptors are due to reflex changes in CO. In contrast, as the baroreflex perturbation is maintained past the initial few seconds, most of the reflex changes in arterial pressure now occur via changes in the peripheral vasculature (25, 27).
This time-dependent difference in baroreflex mechanisms extends from rest through heavy whole body dynamic exercise as the baroreflex is progressively reset to a higher pressure as workload rises (5, 6, 20, 32). Activation of skeletal muscle afferents may be a key factor involved in this resetting (8, 11, 13, 17, 28, 30, 33). Recently, several groups have investigated baroreflex control of HR using the spontaneous baroreflex technique (3, 4, 15, 26, 37, 48). This technique takes advantage of spontaneously occurring fluctuations in arterial pressure and the resultant baroreflex-mediated changes in HR. These rapid baroreflex responses occur solely via changes in parasympathetic activity (10, 12). Employing this spontaneous HR baroreflex sensitivity (HR-SBRS) technique, our group has recently observed that during dynamic exercise, muscle metaboreflex activation (MMA) causes not only resetting of the arterial baroreflex but also a decrease in HR-SBRS (40). In our study we activated the muscle metaboreflex by imposing a set decrease in skeletal muscle blood flow to the hindlimbs in running dogs (∼50% during mild exercise and ∼30% during moderate exercise). Although these imposed decreases in skeletal muscle blood flow were clearly sufficient to activate the muscle metaboreflex, previous studies using this or similar preparations have shown that the magnitude of the metaboreflex pressor response is proportional to the extent of decrease in skeletal muscle blood flow (14, 43, 49). When the metaboreflex is activated, quite linear increases in arterial pressure, HR, and CO occur with progressive further decreases in skeletal muscle blood flow. Whether resetting of the baroreflex and decreases in HR-SBRS occur proportionately with progressive MMA is unknown. Also unknown is whether there is a threshold level of MMA for effects on the baroreflex, whether the effects on the baroreflex saturate at a certain level of MMA, and whether there are differential effects of MMA on baroreflex resetting vs. the effects on reduced HR-SBRS. The purpose of the present study was to directly address these questions in normal dogs during mild and moderate dynamic exercise. We hypothesized that graded MMA would progressively reset the arterial baroreflex and attenuate HR-SBRS in direct proportion to the extent of MMA.
MATERIALS AND METHODS
Experiments were performed on seven adult, mongrel dogs (weight ∼20–25 kg) of either sex (4 males, 3 females). The protocols employed in the present study were reviewed and approved by the Wayne State University Animal Investigation Committee and conform with the National Institutes of Health guidelines.
Surgical preparation and procedures.
On arrival at the laboratory, all animals were accustomed to human handling and during ∼10 sessions were individually trained to comfortably run freely on a motor-driven treadmill at different speeds. Once the animals were successfully trained to run on the treadmill, two surgical procedures were performed on each animal (left thoracotomy and left flank abdominal surgery separated by at least 10 days).
Before each surgery, for tranquilization, the animals received an intramuscular injection of acepromazine (0.2 mg/kg). The dogs were anesthetized with thiopental sodium (25 mg/kg iv). After endotracheal intubation, anesthesia was maintained with isoflurane gas (1–3%). Before the surgery, the animals received cefazolin (antibiotic, 500 mg iv), carprofen (analgesic, 2.0 mg/kg iv), and buprenorphine (analgesic, 0.1 mg/kg im), and a 72-h transdermal fentanyl patch was applied (analgesic, 125–175 μg/h). In addition, before the left thoracotomy, selective intercostal nerve blocks were performed with bupivacaine hydrochloride (2.0 mg/kg). After each surgical procedure, the dogs received a second intravenous dose of cefazolin (500 mg), and antibiotics were continued for the length of the experimental protocol at an oral dose of cephalexin (30 mg·kg−1·12 h−1) to prevent infections. Moreover, after each surgical procedure, for the following 12 h, buprenorphine and acepromazine were administered (0.05 and 0.5 mg/kg iv, respectively) as needed to control any type of discomfort. Thereafter, carprofen was administered orally (4 mg·kg−1·day−1) for 10 days.
In the first surgical procedure, under sterile conditions, a left thoracotomy (4th intercostal space) was performed. A fully implantable telemetered blood pressure transducer (model PAD-70; Data Sciences International) was placed subcutaneously 10 cm caudal to the thoracotomy incision. Its catheter was tunneled into the thoracic cavity through the seventh intercostal space and located inside the left ventricle for measuring left ventricular pressure (LVP). To measure CO, a 20-mm blood flow transducer (Transonic Systems) was placed around the ascending aorta. For studies unrelated to the present investigation, three stainless steel ventricular pacing electrodes (O-Flexon; Ethicon) were sutured to the right ventricular free wall, vascular occluders were placed on the superior and inferior venous cava, and two pairs of sonomicrometry crystals were placed on the left ventricular endocardium. The pericardium was reapproximated loosely, and the chest was closed in layers. After at least 10 days (recovery period), a second surgical procedure (left abdominal retroperitoneal surgery) was performed on each dog. A 10-mm blood flow transducer (Transonic Systems) was placed on the terminal aorta for measuring hindlimb blood flow (HLBF). All side branches between the iliac arteries and the flow probe were ligated and severed, and two 10-mm vascular occluders (DocXS Biomedical Products) were placed on the terminal aorta distal to the flow probe to allow us to reduce flow to the hindlimbs during the experiments (via partial external inflation) and elicit the muscle metaboreflex. In addition, a catheter was placed in a lumbar side branch of the aorta above the flow probe and occluders to monitor arterial pressure. All flow probe cables, pacing wires, vascular occluder tubings, and the aortic catheter were tunneled subcutaneously and exteriorized between the scapulae at the end of its corresponding surgical procedure.
All experiments were performed individually and after the animals had fully recovered from the surgeries and were alert, active, afebrile, and of good appetite. After the animals had fully recovered from the instrumentation and before an experiment for data collection was performed, every animal was refamiliarized (∼5 times) to run on the motor-driven treadmill. Before each experiment, one animal was brought to the laboratory and allowed to roam freely for ∼20 min. The animal was then directed to the treadmill. The CO and HLBF probes were connected to a flow meter (Transonic Systems). The arterial catheter was connected to a pressure transducer (Transpac IV; Abbott Laboratories), the LVP telemetered signal was calibrated, and HR was computed by a cardiotachometer triggered by the CO signal. All data were recorded on analog-to-digital recording systems for subsequent off-line analyses. For a given experimental session, data were collected at rest and then at a randomly selected workload [mild exercise: 3.2 km/h, 0% grade elevation; moderate exercise: 6.4 km/h, 10% grade elevation, which causes CO to increase to ∼40 and 70% of maximal levels (1)]. Every animal successfully performed both experimental protocols (mild exercise and moderate exercise) on different days, since only one workload was performed per experimental day. All animals ran freely with only positive verbal encouragement. Steady-state data were recorded at rest while the animal was standing on the treadmill, during exercise (at either mild or moderate workload) with unrestricted blood flow to the hindlimbs, and after graded reductions of HLBF (via partial inflations of the terminal aortic occluders) to elicit gradual metaboreflex activation. Each level of reduction in hindlimb perfusion was maintained until all parameters reached steady state (3–5 min).
Beat-to-beat CO, HLBF, HR, mean arterial pressure (MAP), and LVP were continuously collected during each experiment. Stroke volume (SV) was calculated as CO/HR. As previously stated, data were recorded for 3–5 min of steady state at standing rest, free-flow exercise (mild or moderate), and each level of HLBF (each reduction) so that each period spanned multiple respiratory cycles. Since left ventricular systolic pressure (LVSP) is virtually identical to systolic pressure in the aortic arch, we used LVSP as the input to the arterial baroreflex. Spontaneous baroreflex control of HR was dynamically assessed by analyzing the beat-to-beat relationship between HR and LVSP as previously described (40). Briefly, the beat-to-beat time series of LVSP and HR were searched for three or more consecutive beats in which the LVSP and HR of the following beat changed in opposite direction (i.e., −HR/+LVSP and +HR/−LVSP). These sequences were identified as baroreflex sequences. A linear regression was applied to each individual sequence, and only those sequences in which r2 was >0.85 were accepted, and subsequently, a slope was calculated. The mean slope of the LVSP-HR relationship, obtained by averaging all slopes computed within a given test period, was calculated and taken as a measure of spontaneous baroreflex sensitivity for that period. Sinoaortic baroreflex denervation virtually abolishes baroreflex sensitivity assessed via this method, indicating that these spontaneous reciprocal HR changes that occur as a result of changes in arterial pressure are mediated by the baroreflex (3, 16).
The nonlinear patterns of the hemodynamic and HR-SBRS responses to graded reductions in HLBF were analyzed by plotting the variable (e.g., MAP) vs. HLBF during free-flow exercise and at each level of partial vascular occlusion as shown in Fig. 1. As described in detail previously (44, 45, 49), during mild exercise, initial reductions in hindlimb perfusion do not elicit metaboreflex responses; however, once hindlimb perfusion is reduced below a threshold level, a pressor response occurs. The threshold was approximated as the intersection between two regression lines, the initial response line in which no reflex responses occurred during the initial reductions in hindlimb perfusion and the pressor response line in which further reductions in hindlimb perfusion elicited a reflex pressor response. During moderate exercise, often no apparent threshold exists and the initial reduction in hindlimb perfusion elicits reflex responses. If no threshold was apparent, then the threshold was ascribed as the free-flow value of hindlimb perfusion.
By utilizing the averaged responses for each animal, statistical analyses were performed with Systat software (Systat 11.0). An α level of P < 0.05 was set to determine statistical significance. One-way analysis of variance for repeated measures was used for comparing hemodynamic data obtained at rest and during exercise under free-flow conditions at threshold and at maximal levels of HLBF reduction during mild and moderate workloads. If a significant interaction term was found, a test for simple effects post hoc analysis (C-Matrix) was performed to determine significant group mean differences. We compared the slope of the linear regression line between HR-SBRS and HLBF after threshold between mild and moderate exercise using a paired t-test. Data are means ± SE.
Figure 2 shows data from one animal during both protocols (mild and moderate exercise). From rest to mild exercise, the relationship between HR and LVSP was shifted upward and the linear relationship was less steep, which represents a decrease in SBRS. With the initial partial occlusion of the terminal aorta and imposed reductions in HLBF, no metaboreflex pressor response was engaged and there was little change in the HR-LVSP relationship, and thus HR-SBRS remained essentially constant. However, once HLBF was reduced below the metaboreflex threshold, HR and blood pressure increased. With the generation of the pressor response, there was a progressive shifting of the HR-LVSP relationship upward and to the right, and the slope progressively flattened. This resulted in a quite linear relationship between HR-SBRS and HLBF as the metaboreflex became progressively more engaged. Similar responses were observed during moderate exercise with even more pronounced falls in HR-SBRS to ∼20% of resting levels at maximal metaboreflex activation. At both workloads, the decrease in the strength of HR-SBRS was linearly related to the reduction in HLBF over the entire range of metaboreflex activation. As we have previously shown (40), neither exercise nor metaboreflex activation affected the number of SBRS occurrences observed per minute (rest, −8.3 ± 1.8; mild exercise, −8.5 ± 1.3; mild exercise + MMA, −10.3 ± 1.8; moderate exercise, −8.3 ± 1.0; moderate exercise + MMA, −9.5 ± 2.2; ANOVA, P > 0.05).
Figure 3 shows HR, SV, CO, MAP, LVSP, and HR-SBRS at rest and during mild and moderate exercise without any imposed reductions in HLBF (free flow) at threshold and at approximately maximum activation of the muscle metaboreflex. As expected and in agreement with previous studies (41, 42) from rest to mild exercise, although MAP and LVSP did not change significantly, we observed increases in HR and SV (as a result in CO) and also a rise in HLBF. Thus the increase in CO is offset by a rise in total vascular conductance due to the active vasodilation in the exercising skeletal muscles, resulting in little change in MAP. In addition, a substantial decrease in HR-SBRS occurred compared with standing rest. MMA at this workload generated a progressive rise in MAP and LVSP, a significant tachycardia, and a small rise in SV, and thus CO substantially increased. Moreover, MMA during mild exercise caused considerable changes in HR-SBRS with a pattern very similar to the changes induced in most other hemodynamic parameters; that is, a clear threshold existed for the effects on HR-SBRS, and the changes progressed linearly with progressive reductions in HLBF and did not saturate despite marked reductions in HLBF that were the maximum we could impose and obtain steady-state data. Similar results were observed during moderate exercise with the exception that the prevailing level of HLBF was much closer to metaboreflex threshold (and no threshold was observed in some experiments). Again, the pattern of the changes HR-SBRS with metaboreflex activation mirrored those that occurred in most hemodynamic parameters (e.g., HR, CO, MAP, LVSP). Table 1 shows the average slope (of all animals) of the linear regression line during mild and moderate exercise between HLBF and HR-SBRS. The average slope after muscle metaboreflex threshold during mild exercise was about −4.15, meaning that for every liter per minute decrease in HLBF beyond metaboreflex threshold, HR-SBRS decreased 4.15 beats·min−1·mmHg−1. The high value of the correlation coefficient indicates a very close linear relationship between these two variables. Moreover, although the slope was significantly reduced during moderate exercise, there was still a very close relationship between HLBF and HR-SBRS as shown by a high correlation coefficient value in this condition.
To compare whether the effects of MMA on spontaneous baroreflex control of HR occurred concurrently with the effects of the muscle metaboreflex on the other hemodynamic parameters, we compared the HLBF at metaboreflex threshold, which is separately calculated for each of the variables. Figure 4 shows that at either workload, there was no significant difference in HLBF at the calculated MMA threshold for HR-SBRS, HR, CO, LVSP, and MAP, indicating that the muscle metaboreflex modulates the heart, blood pressure, and the baroreflex in concert. In addition, for each experiment, the relationship between HR-SBRS and LVSP beyond metaboreflex threshold was also quite linear at both workloads (Fig. 5). On average, HR-SBRS decreased ∼10% for every 10 mmHg increase in LVSP during steady-state metaboreflex activation at both workloads.
To our knowledge, this is the first study to show that 1) graded MMA during dynamic exercise not only progressively resets the arterial baroreflex operating point but also gradually decreases HR-SBRS in direct proportion to the extent of muscle metaboreflex activation; 2) the metaboreflex threshold levels of HLBF for effects on HR-SBRS were not different between the hemodynamic parameters and the arterial baroreflex, and the effects of the muscle metaboreflex on HR-SBRS occurred proportionately with the reflex effects on hemodynamic parameters such as HR, CO, and MAP; 3) the effects of MMA on HR-SBRS occur progressively with graded metaboreflex activation and do not saturate over a wide range of MMA; and 4) the effects of MMA on baroreflex resetting were coincident with the effects on reduced HR-SBRS.
Feedback reflexes responsible for autonomic modulation during whole body dynamic exercise.
Overall, in addition to the feedforward role of central command, two negative feedback reflexes likely responsible for the autonomic modulation during dynamic exercise are the arterial baroreflex and reflexes arising from activation of skeletal muscle afferents (mechanosensitive and metabosensitive). A fall in skeletal muscle oxygen delivery and flow leads to accumulation of metabolic by-products within the active muscle that stimulate group III and IV afferent neurons, which evokes reflex changes in autonomic nerve activity and release of vasoactive hormones, termed the muscle metaboreflex. This reflex can trigger a significant increase in CO and/or also cause peripheral vasoconstriction, which in turn raises MAP. The arterial baroreflex is the primary short-term regulator of arterial blood pressure by altering peripheral vasoconstriction and cardiac output, via adjustments of sympathetic and parasympathetic nerve activity (38). In addition, it is well known that the baroreflex control of HR and blood pressure is reset during exercise. One known mechanism responsible for this baroreflex resetting during exercise is the feedback from skeletal muscle afferents. In animal and human studies, it has been shown that neural feedback from the skeletal muscles modulates the arterial baroreflex function (19, 28, 46). Employing the HR-SBRS technique to study the interaction between these two reflexes, our group recently observed that during mild and moderate dynamic exercise, MMA causes not only resetting of the arterial baroreflex but also a decrease in HR-SBRS (40). In that study, MMA was obtained via a one-step reduction in HLBF (reduced to ∼50% and ∼70% of exercising blood flow level for mild and moderate workload, respectively). However, it remained unknown whether the muscle metaboreflex affects the arterial baroreflex in fashion similar to the cardiovascular effects of the metaboreflex on hemodynamic parameters. Previous studies from our and other laboratories have shown that the stimulus-response relationship between O2 delivery or blood flow to active skeletal muscle is quite linear once beyond threshold and that no saturation occurs at submaximal workloads over a wide range of metaboreflex activation (14, 43, 49). The question remained whether the effects of MMA on HR-SBRS would show a similar trend, and we found that this was the case. Progressive MMA caused progressive increases in HR and MAP and progressively reset the HR-LVSP relationship upward and to the right with a decrease in baroreflex HR sensitivity as indexed by the spontaneous method. The effects of MMA on HR-SBRS were quite linear when analyzed as both the relationship between the imposed reductions in HLBF and HR-SBRS and the metaboreflex-induced increases in LVSP vs. HR-SBRS. The effects of the muscle metaboreflex became apparent at the same threshold level of HLBF as all other hemodynamic parameters, and no saturation of the effects of MMA on HR-SBRS was evident over the range of MMA employed (which reflected the largest activation of the muscle metaboreflex we could obtain in which steady state could be achieved with the animal freely exercising on the treadmill). Thus the present study shows that the muscle metaboreflex control of cardiac function at either mild or moderate exercise is induced and progresses linearly in concert with baroreflex modulation, since HLBF values at the threshold for HR-SRBRS, CO, MAP, LVSP, and HR were not different and all responded linearly with further reductions in HLBF.
Muscle metaboreflex and baroreflex interaction.
It is highly likely that muscle metaboreflex and baroreflex interaction occurs centrally. Previous studies have shown that central modulation of the baroreflex circuitry may be responsible for mediating resetting of the baroreflex during exercise (7, 8, 11, 17, 19, 30). Among the different possible central sites that are part of the central baroreflex arc, the nucleus tractus solitarii (NTS) is a region where an interaction between these two reflexes can take place, for this nucleus, apart for being known for participating in many viscerosensory systems, receives inputs from both the baroreceptor afferents and spinal somatosensory input and contains a complex network of excitatory and inhibitory interneurons (2). Indeed, neural feedback from skeletal muscle afferents have been shown to activate a GABAergic mechanism within the NTS that reduces the rapid bradycardic responses to transient excitability of baroreceptor activation (18, 31). A different or additional possible central site for this reflexes interaction is the rostral ventrolateral medulla, for it has been shown that skeletal muscle afferents can cause direct excitation of sympathetic premotor neurons in this brain stem region (22, 29, 47). Another possibility for the muscle metaboreflex and baroreflex interaction is that muscle metaboreflex-induced increase in plasma norepinephrine attenuates baroreflex modulation of HR (21), since it has previously been shown that MMA elicits a rise in plasma norepinephrine and that high plasma norepinephrine concentration attenuates parasympathetic control of HR (9, 23). However, whether more central regions are involved, as well as the cellular mechanisms responsible for the interaction, are still uncertain.
Limitations of the study.
Our approach to evaluate the arterial baroreflex control of HR based on spontaneous fluctuations in blood pressure and HR has advantages and disadvantages, which have previously been described in detail (39, 40). Briefly, the spontaneous baroreflex technique enables a qualitative and quantitative estimate of the baroreceptor-cardiac response relationships during spontaneous blood pressure fluctuations without the necessity of any mechanical or pharmacological intervention. Sympathostimulatory reflexes by stretch of cardiac chambers after phenylephrine-induced increase of afterload or a direct β-adrenergic stimulation at the sinus node level by high doses of the drug may affect baroreflex sensitivity determination. Alternatively, the autonomic mechanisms mediating these rapid baroreflex-induced changes in HR are likely only parasympathetic in nature (24, 26), and in addition, this approach only examines the baroreflex sensitivity over a relatively modest range of pressure, which therefore does not allow the calculation of the entire sigmoidal baroreflex stimulus-response relationship. It is possible, if not likely, that at least a portion of the reduction in HR-SBRS with exercise and metaboreflex activation stemmed from a shift in the operating point of the baroreflex from the high slope section near the middle of the stimulus-response relationship toward a flatter part of the sigmoid curve (33). Previous studies have shown that HR-SBRS is virtually abolished after arterial baroreceptor denervation, which shows the reflex nature of the HR responses (3, 16).
In conclusion, MMA during mild and moderate dynamic exercise progressively resets the arterial baroreflex to higher blood pressure and HR in direct proportion to the extent of MMA. As the muscle metaboreflex is engaged, it simultaneously resets and progressively depresses HR-SBRS in concert with the increases in HR, CO, and arterial blood pressure. Thus the central interactions occur concurrently with the efferent responses. The fall in SBRS with metaboreflex activation indicates that the ability of the baroreflex to rapidly respond to perturbations in arterial pressure via changes in HR progressively decreases as muscle ischemia ensues. This may be particularly important during exercise in patients with claudication and in other patients with high sympathetic activity such as those with heart failure and hypertension.
This research was supported by National Heart, Lung, and Blood Institute Grant HL-55473.
We thank Jody Helme-Day and Erin Krengel for expert technical assistance and care of the animals.
- Copyright © 2010 the American Physiological Society