Contribution of Muscle Afferents to the Hemodynamic, Autonomic, and Ventilatory Responses to Exercise in Patients With Chronic Heart Failure
Effects of Physical Training
Background A neural linkage between peripheral abnormalities and the exaggerated exercise responses in chronic heart failure (CHF) was postulated. We studied the ergoreceptors (afferents sensitive to skeletal muscle work) in CHF and whether training can affect their activity.
Methods and Results In 12 stable CHF patients (ejection fraction [EF]=26.4%) and 10 control subjects (EF=55.3%), we compared the responses to dynamic handgrip and during a 3-minute period of posthandgrip regional circulatory occlusion (PH-RCO). The ergoreflex contribution was quantified as the percentage responses to exercise maintained by PH-RCO compared with recovery without PH-RCO. Patients showed ergoreflex overactivation compared with control subjects in terms of ventilation (86.5% versus 54.5%), diastolic pressure (97.8% versus 53.5%), and leg vascular resistance (108.1% versus 48.9%) (all P<.05). The contribution of the ergoreflex to vagal withdrawal (high frequency of RR variability) and sympathetic activation (low frequency of RR, pressure variability) was evident in both groups. Nine control subjects and nine CHF patients participated in 6 weeks of forearm training. Training reduced the ergoreflex contributions more in CHF than in control subjects: diastolic pressure (−33.2% versus −4.6%), ventilation (−57.6% versus −24.6%), and leg vascular resistance (−59.9% versus −8.0%) (all P<.05).
Conclusions (1) The ergoreflex role has a larger effect on the responses to exercise in CHF than in control subjects. (2) Training may reduce this exaggerated ergoreflex activity, thereby improving the responses to exercise.
Reduced exercise capacity with increased breathlessness and muscle fatigue is a major cause of morbidity in patients with stable CHF who limit activity to avoid these symptoms. This results in deconditioning and hence possibly leads to a vicious circle of progressively worsening exercise tolerance.
Ventilatory responses to exercise are increased in CHF.1 2 Peripheral abnormalities have been demonstrated in skeletal muscle histology,3 mitochondria,4 oxidative enzyme activities,5 and high-energy phosphate handling6 together with early muscle fatigue,7 sympathetic overactivity, and parasympathetic withdrawal.8 9 The exact mechanisms causing these modifications after the development of left ventricular dysfunction have not been fully characterized.10
Several observations suggest that changes in the periphery are important determinants of exercise performance in CHF. First, left ventricular function and systemic hemodynamics correlate poorly with exercise capacity.11 Second, central hemodynamic improvements with drug therapy, including angiotensin-converting enzyme inhibitors, are rapid, but increases in exercise capacity are delayed for weeks or months.12 Neither pulmonary capillary wedge pressure13 nor blood lactate14 correlates well with exertional breathlessness, suggesting a crucial missing link in the mechanisms of exercise limitation in CHF.
Physical training in CHF has been shown to partially reverse the muscle metabolic abnormalities15 in addition to those of the autonomic tone (a reduction of sympathetic and an increase in vagal activity9 16 ) and the ventilation (reductions in ventilatory equivalents for CO2 [V̇/V̇co2], tidal volume, and respiratory rate during submaximal exercise) but without any detectable effect on central hemodynamics.13 16 Thus, the major adaptations to training may be peripheral. This may clarify the link between the abnormal skeletal muscle metabolism and the heightened respiratory, circulatory, and autonomic responses seen in CHF.17
Unmyelinated and small myelinated afferents in muscle exist that are sensitive to metabolic changes related to work (ergoreceptors).18 They have been shown to be responsible for the early circulatory response to exercise, including activation of the sympathetic vasoconstrictor drive.19 20 This reflex might be sensitized by the muscle acidosis seen in CHF during exercise. Therefore, we investigated this reflex in patients with CHF before and after physical conditioning.
The study was approved by the local ethics committees. All subjects gave written informed consent.
We studied 12 patients with stable CHF due to ischemic heart disease and 10 control subjects with normal left ventricular function (Table 1⇓). All subjects were clinically stable, physically inactive, in sinus rhythm, and on unchanged medication for at least 3 months before the study. None were participating in any exercise training program or had any significant chronic lung or valvular heart disease, diabetes, or neuromuscular disorder.
The CHF patients were all symptomatic (mean duration of heart failure, 18±3 months; range, 8 to 38 months). They were selected only if they were limited by breathlessness or muscle fatigue, without evidence of exercise-induced myocardial ischemia, arrhythmias, or claudication; patients with diabetes, peripheral vascular disease, myopathy, or thyroid dysfunction or those taking β-blockers, β-stimulants, or digoxin were excluded because these interfere with the autonomic indexes and potassium homeostasis. Left ventricular ejection fraction was <30% at cardiac catheterization (9 cases) or radionuclide ventriculography (3 cases) in patients and >50% in the control subjects who had previously undergone cardiac catheterization for ischemic heart disease assessment. Exercise capacity and peak O2 consumption (V̇o2max) were tested on a Tunturi upright bicycle ergometer in 5-minute, 25-W incremental stages to exhaustion; the tests were repeated at least twice in the laboratory until reproducible exercise performances were determined.
After clinical screening, the subjects performed the bicycle exercise tests to determine their exercise capacity and to become familiar with the laboratory environment.
A first assessment test was performed to evaluate the muscle ergoreflex activity (assessment 1); thereafter, the subjects were enrolled for the training program (Fig 1⇓). The study was designed as a random-order, crossover, controlled comparison21 of 6 weeks of localized training and 6 weeks of restricted activity (detraining). The physician supervising the patient during the exercise test was unaware of the clinical conditions and the training status of the subject.
The subjects were asked to avoid strenuous physical activity for at least 24 hours before each assessment test and to refrain from eating and smoking for 3 hours before. All exercise tests were performed at the same time of day on constant medication with the patient supine with the arm extended. First, a maximal voluntary handgrip contraction was measured as the greatest of the peak forces produced by three brief maximal handgrip contractions.
The evaluation of the ergoreflex activity before and after training or detraining (Fig 1⇑) included two exercises, performed in random order: (1) a control handgrip exercise using repetitive finger flexion by pulling a lever of a dynamometer at 50% of the predetermined maximal contraction (this was done with the nondominant arm at the rate of 40 pulls per minute until exhaustion); (2) the same protocol followed by, from 10 seconds before the end of exercise, 3 minutes of circulatory (venous and arterial) occlusion by forearm tourniquet inflation to +30 mm Hg above systolic pressure (PH-RCO). After the cuff was inflated, the subject was instructed to relax. Thus, the contribution of the muscle ergoreceptors was evaluated by trapping of the metabolites in the exercising muscle after exercise. This protocol has been shown to fix the metabolic state of the muscle and to prolong the activation of the ergoreceptors.20 22
Thirty minutes separated each bout of handgrip exercise. A control RCO by tourniquet inflation at 200 mm Hg at rest without previous exercise was also performed to exclude the possibility that occlusion alone was producing any reflex response.
Handgrip tests were performed on a conventional handgrip dynamometer (Jamar) modified and connected to a display positioned in front of the subject to show him or her the amount of effort of each contraction (in watts) to help the subjects maintain the same level of exercise.
The subjects were randomly allocated either to a nondominant forearm training program (training), previously described,23 or to avoid any exercise (detraining) (Fig 1⇑). Training consisted of two or three sessions daily for 6 weeks. Subjects gently stretched the forearm muscles first and then squeezed a hand-held dynamometer device (the Gripper) as tightly as possible for 10 seconds, followed by 20 seconds of rest, for a total of 10 repetitions per session. After a 5-minute rest period, they repeatedly squeezed the handgrip 40 times per minute for 5 minutes. Compliance was monitored by daily calendar; compliance exceeded 95%.
After the 6 weeks of training or detraining periods, maximal voluntary handgrip contraction was measured (the reported values are the means of the three determinations), and the ergoreflex assessment tests were repeated (assessment 2 and assessment 3) (Fig 1⇑).
The subject first rested for 30 minutes in a quiet, darkened environment. For assessment 1 only, a 20-gauge brachial artery catheter was previously inserted into the nonexercising arm under local anesthesia for the later withdrawal of blood samples. Catheter patency was maintained by repeated flushing with heparinized saline (2 IU/mL). Fig 2⇓ is a schematic representation of a subject with the study measurements.
Heart Rate, Blood Pressure, and Power Spectral Analysis
Heart period was derived from the ECG. Blood pressure was recorded beat to beat with the noninvasive Ohmeda 2300 Finapres device (Ohmeda Monitoring System), which has been shown to provide a reliable index of blood pressure changes over a short period.24 For assessment 1 only, when a brachial artery catheter was inserted into the nonexercising arm, the blood pressure was also assessed with the arterial line; however, since the line was also used for blood sampling, continuous blood pressure monitoring was not possible for all phases of the study.
The nonquantitative respiratory signal for spectral analysis was obtained by an impedance device.25 Data were digitized on-line by a 12-bit analog-to-digital converter (NB-MIO-16 board, National Instruments) at a sampling rate of 500 samples per second and connected to a Macintosh IIcx computer (Apple Inc) equipped with 5-MB RAM memory and a 60-MB hard disk.25 Stable sections of data of 256 beats were selected by visual examination and analyzed: relatively short data records were chosen because the physiological changes that occurred during and after exercise were changing significantly on a longer time scale.
Periodic variations in the RR interval and systolic and diastolic blood pressures provide indexes to follow changes in autonomic neural activity induced by changes in physical activity, posture, or other sources of stimulation or blockade. This variability was presented as SDs and spectral power density in defining HF and LF regions. The absolute powers of the harmonic components were presented as natural logarithms (ln ms2 or ln mm Hg2) to minimize skewness. The LF range of periodicity was defined as that between 0.03 and 0.14 Hz, and the HF component was defined as lying between 0.18 and 0.4 Hz. The LF component of RR interval and blood pressure variability has been considered predominantly an index of sympathetic activity and the HF component in the RR variability an index of vagal tone.25 26 To estimate the relative predominance of each frequency component, the relative power of each component was computed as percentage of the total variability (excluding very-low-frequency fluctuations, <0.03 Hz).
Peripheral Blood Flow
Right leg (nonexercising limb) blood flow was measured at the same time by standard techniques27 with a mercury-in-Silastic strain-gauge plethysmograph (Hokanson) and expressed in mL·100 mL tissue−1·min−1. The corresponding leg vascular resistance (expressed in arbitrary units) was computed from these data.
Subjects breathed air through a mouthpiece with a nose clip in place, and continuous on-line ventilation and expiratory gas data were collected. The analysis was performed with two different systems but with the same equipment throughout the different stages of the study for a given subject. One method evaluated ventilation (V̇, inspiratory flow, by Harvard Instruments Dry Gas Meter) and expiratory CO2 and O2 concentrations (Servomex 570 and PA404 meters) to derive O2 consumption (V̇o2) and CO2 production (V̇co2) from standard formulas. The gas meters were calibrated against gases of known concentrations before each test.16 The other equipment calculated V̇, V̇co2, and V̇o2 on line every 10 seconds by an inert gas dilution technique using a respiratory mass spectrometer (Innovision).28 29
Arterial Potassium and Gases
Arterial blood was sampled before each exercise, at peak effort, and at 3 and 8 minutes of recovery. The plasma was centrifuged within 15 minutes for potassium (K+) and analyzed by flame photometry using a dedicated machine (CIBA-Corning) calibrated before each run. Arterial blood samples were immediately analyzed for partial pressure of carbon dioxide (Paco2), oxygen (Pao2), and pH with a Ciba-Corning Blood Gas System (after a two-point calibration before each run).
Within each group, we assessed (1) whether there was any difference between exercise runs with or without PH-RCO, (2) the effect of these two exercises in comparison with respective resting values, and (3) the ergoreflex contribution by comparing 3-minute PH-RCO with the 3-minute recovery run without RCO, evaluated as absolute values. For the ventilatory and hemodynamic data, we also compared the percentage values of the peak exercise responses still persisting at 3 minutes.
Between the two groups (CHF and control subjects), we compared (1) the difference in the resting values and in the exercise responses (at peak exercise and at matched work duration, ie, 3-minute handgrip); (2) the ergoreflex contribution to the exercise changes, estimated as the difference between the 3-minute PH-RCO and the 3-minute recovery run without RCO; and (3) (autonomic data excepted) the percentage values of the peak exercise responses still persisting at 3 minutes.
Assessments 2 and 3 (After Detraining Phase or Training Phase)
Within each group, the physical conditioning effect was assessed by comparison of the detraining with the training data for the resting values, the exercise responses, and the ergoreflex contribution to exercise changes.
Between the two groups, we compared the exercise responses (at peak exercise and at matched work duration) and the ergoreflex contribution to the exercise changes.
To compare the resting and exercise data after detraining and after training, we considered the means of the two runs (with and without PH-RCO), since preliminary analysis had shown no significant difference between the two runs. To compare the ergoreflex contribution, we considered the absolute differences and (autonomic data excepted) the percentage differences between the two recovery runs with and without PH-RCO (at 3 minutes of recovery).
ANOVA for repeated measures was used. Post hoc assessment of individual point comparisons was assessed by paired t test when the same group of subjects was evaluated, with a two-tailed level of significance less than the 5% level.
Unpaired t test or nonparametric tests, depending on the normality of the distributions, corrected when appropriate by Scheffé’s procedure for multiple comparisons, were used in the post hoc comparison between different groups. Results are quoted as mean±SEM.
At rest there was no significant change in any of the measured variables with RCO compared with no RCO.
All 12 CHF patients and the 10 normal subjects completed the three phases of the exercise protocol. Compared with control subjects, the CHF patients were sex- and age-matched but showed lower ejection fraction and V̇o2max (Table 1⇑).
Within each group, the resting values of all considered variables were not significantly different before the two exercise runs (ie, with and without PH-RCO).
Compared with control subjects, the CHF patients had significantly higher diastolic pressure and heart rate but reduced V̇co2 and V̇o2 values. When power spectral analysis data were used, the CHF patients showed depressed RR variability with reduced values of both LF and HF components and depressed systolic and diastolic blood pressure variability with reduced LF components. When normalized for the total variability, the LF component of RR interval variability (%) was more elevated in CHF patients than in control subjects (Table 2⇓).
Within each group of subjects, no differences in the exercise duration or for any variable at peak exercise were seen between the two handgrip exercises (ie, with and without PH-RCO). The CHF patients exercised less than control subjects (Table 2⇑) and had lower maximal strength (67.5±3.7 versus 75.4±5.5 kg, P<.01).
Increases of heart rate, systolic and diastolic pressures, V̇, respiratory rate, V̇co2, V̇o2, leg vascular resistance, and reductions in leg blood flow (all P<.001 versus baselines) were observed in both groups of subjects without significant changes in arterial K+ concentration, Paco2, Pao2, or pH.
Compared with control subjects, at peak exercise, patients presented higher V̇ and respiratory rate and lower V̇co2 and V̇o2 levels (all P<.001) (Table 2⇑).
For matched work duration (3-minute handgrip), in comparison with control subjects, patients also showed higher diastolic pressure (98.8±3.0 versus 90.1±3.0 mm Hg, P<.01), heart rate (92.6±2.5 versus 85.5±3.4 beats per minute, P<.01), and leg vascular resistance (96.5±6.2 versus 81.5±5.2 units, P<.01), with reduced leg blood flow (1.4±0.1 versus 1.7±0.1 mL·100 mL−1·min−1, P<.001).
Within each group of subjects, the changes in the spectral analysis parameters observed during the two handgrip runs were similar, with reductions of RR variability and its HF component, without significant changes in the LF component (Table 2⇑). The LF component rose as a percentage of the total variability, and the HF component fell in both groups. Systolic and diastolic pressure variability increased, with concomitant increases in LF but no changes in HF components. During exercise, CHF patients still showed lower RR, systolic and diastolic variability, LF, and HF compared with control subjects (Table 2⇑).
Four subjects out of 12 CHF and 2 control subjects out of 10 felt slight discomfort when the cuff was inflated after exercise, but none of them described this as pain; in all, the sensation immediately disappeared when the cuff was released.
In both CHF and control groups, the ergoreflex activation (by PH-RCO) prolonged the increased systolic and diastolic blood pressures, V̇, respiratory rate, and leg vascular resistance, but with no concomitant changes in heart rate: these contributions were evident both in absolute and also in percentage values (Table 3⇓). These differences could not be attributed to any significant differences in V̇co2, V̇o2, arterial K+ concentration, pH, Paco2, or Pao2 (Table 3⇓).
In both CHF and control subjects, ergoreceptor activation increased the power spectral indexes of sympathetic tone (elevation of LF in RR and systolic and diastolic variability), with a fall in the vagal index (HF of RR variability); systolic and diastolic variability also increased (Table 3⇑).
The ergoreflex contributions to exercise responses were higher in patients than in control subjects in both absolute and percentage values with regard to the systolic (42.4±8.2 versus 25.6±11.3 mm Hg, P<.01; 89.0±8.1% versus 53.5±9.7%, P<.001) and the diastolic (23.5±3.9 versus 15.9±4.3 mm Hg, P<.01; 97.8±10.3% versus 53.5±9.7%, P<.001) pressures, the ventilatory responses to exercise as V̇ (11.5±1.7 versus 3.2±1.0 L/min, P<.01; 86.5±7.1% versus 54.5±8.6%, P<.001) and respiratory rate (11.5±1.8 versus 3.0±1.0 breaths per minute, P<.01; 84.3±12.7% versus 48.6±24.7%, P<.01), leg vascular resistance (47.0±10.0 versus 32.1±6.4 units, P<.05; 108.1±6.5% versus 48.9±6.0%, P<.001), and blood flow (1.0±0.2 versus 0.5±0.1 mL·100 mL−1·min−1, P<.01; 101.2±13.1% versus 57.3±13.4%, P<.001). CHF patients did not show any differences in the changes of autonomic parameters compared with control subjects.
Fig 3⇓ compares the V̇, diastolic blood pressure, and leg vascular resistance responses in CHF patients and control subjects during the two handgrip runs without and with PH-RCO.
Assessments 2 and 3: Training Versus Detraining
Nine of 12 CHF and 9 of 10 control subjects completed the training regimen, with 95% compliance: 2 patients and 1 control subject did not complete it because of work commitments, and 1 patient found it too difficult to follow.
There were no changes in the resting values induced by either training or detraining in either group of subjects. No significant weight loss occurred in either group.
Within each group of subjects, no differences in the exercise duration were seen between the two exercise runs, with and without PH-RCO. After training compared with after detraining, the two groups showed longer durations of forearm exercise (CHF: 8.0±0.5 versus 4.8±0.4 minutes, P<.001; control subjects: 9.0±0.3 versus 6.5±0.1 minutes, P<.001) and increased peak work rates (CHF: 78.5±1.1 versus 64.9±1.7 kg, P<.05; control subjects: 85.0±2.6 versus 76.9±3.1 kg, P<.005). Thus, even after training, CHF patients exercised less than control subjects.
After training versus detraining, CHF and control subjects increased peak exercise V̇co2 (CHF: 422.0±31.3 versus 348.8±25.8 mL/min, P<.01; control subjects: 463.4±16.5 versus 404.6±11.0 mL/min, P<.05) and V̇o2 (450.0±25.2 versus 396.5±19.1 mL/min, P<.01; 480.8±15.4 versus 463.9±17.0 mL/min, P<.05).
Changes in matched work duration (3-minute handgrip) were compared after detraining and after training (Table 4⇓): in both groups, conditioning reduced systolic and diastolic pressures, V̇, V̇o2, V̇co2, and leg vascular resistance with increased blood flow. The reduction in heart rate and respiratory frequency did not reach statistical significance in the CHF group. In comparison with control subjects, after training, patients still showed higher hemodynamic and ventilatory responses to exercise at matched work duration (Table 4⇓).
After detraining, both groups of subjects showed the same ergoreflex-driven responses in all variables as the baseline test (assessment 1).
In the CHF group, training significantly reduced the ergoreflex contributions (by PH-RCO) to (1) the systolic and diastolic pressure changes, (2) the ventilatory response to exercise (both V̇ and respiratory rate), and (3) leg vascular resistance changes with an increased leg blood flow (Table 5⇓). Reduced sympathetic activation was also seen, as assessed by power spectral analysis of systolic and diastolic pressure variability LF components.
In the control subjects, training reduced the ergoreflex contributions to blood pressure, ventilatory, leg vascular resistance, and blood flow responses to exercise, although this effect was significant only for V̇ (as a percentage) and leg vascular resistance. Reduced sympathetic activation of the circulation was also evident in control subjects (LF components of systolic and diastolic pressure variability) (Table 5⇑).
After training, no significant differences were observed in V̇co2 and V̇o2 data in both groups. The differences between CHF patients and control subjects in the percentage contribution of the ergoreceptors to the responses to exercise were less marked after training than after detraining: systolic (23.5±12.3% versus 31.2±10.3%, P<.05) and diastolic (4.9±13.1% versus 33.5±14.5%, P<.05, Fig 4⇓) blood pressures, V̇ (16.7±11.3% versus 49.7±9.9%, P<.05), respiratory rate (4.8±15.8% versus 33.9±26.7%, P=NS), leg vascular resistance (12.2±8.9% versus 64.1±8.7%, P<.01), and leg blood flow (24.9±18.9% versus 61.9±17.8%, P<.01). The autonomic parameters showed no significant differences between control subjects and CHF patients (Table 5⇑).
Fig 5⇓ compares the training effect on V̇, diastolic pressure, and leg vascular resistance responses in CHF patients and control subjects during the two handgrips without and with PH-RCO after training.
A determinant role for the skeletal muscle abnormalities in the limited exercise performance in established CHF has been hypothesized. Initially, a reduction in blood flow to exercising limbs during both submaximal and maximal exertion was demonstrated.30 Subsequently, altered skeletal muscle morphology and metabolism were described. Biopsy studies reported decreased oxidative enzyme activity and muscle fiber atrophy.3 4 5 Using 31P nuclear magnetic resonance spectroscopy, Wilson et al31 demonstrated excessive depletion of PCr and greater acidosis in the forearm muscles during submaximal exercise in CHF patients compared with normal subjects. These findings were later confirmed by Massie et al,6 who also found a relation between the degree of skeletal muscle alterations in the forearm during exercise and the clinical severity of CHF. Furthermore, these changes were unrelated to limb blood flow and were present during ischemic as well as aerobic exercise.32 Therefore, there is little evidence that an impairment of blood flow is the primary cause of these metabolic changes.33 Instead, leg muscle volume correlates with exercise limitation in CHF.34
These studies suggested that an alteration in the muscle itself might be a major cause of exercise intolerance, with consequent exertional fatigue and dyspnea. The frequent coexistence of dyspnea and fatigue and the delay in resolution of dyspnea after hemodynamic improvement raise the possibility that structural skeletal muscle changes may be involved in the generation of dyspnea as well as fatigue. Dyspnea and increased ventilation could also be manifestations of changes in respiratory musculature similar to those seen in skeletal muscle.35 36 An alternative explanation is that skeletal muscle signals contribute directly to the perception of both muscle fatigue and dyspnea via an exaggerated neural signal from the muscle. Indeed, we have shown no objective difference in characteristics of patients limited by fatigue compared with those limited by dyspnea.37
Deconditioning may be responsible for a part of the skeletal muscle abnormalities: several studies on the beneficial effects of exercise training on muscle function in heart failure have suggested that inactivity may play a role in the peripheral abnormalities. Training not only improves V̇o2max (even though central hemodynamics do not improve) but also causes a number of significant changes in the exercising limbs, including increases in O2 uptake and arteriovenous O2 difference at maximal exercise and a decrease in lactate accumulation at submaximal exercise with an increase in blood flow.29 Surprisingly, a strictly localized forearm training, with no systemic effects, improved both metabolic and functional capacity in CHF patients:23 38 the trained forearm demonstrated a greater submaximal endurance and a lower ratio of inorganic phosphate to PCr, without associated changes in muscle blood flow or systemic hemodynamic response. This result reflects an improved oxidative capacity of the muscle. Physical training in CHF induces reductions in PCr depletion and increases in ADP during exercise, with enhanced rates of PCr resynthesis in recovery, indicating a substantial correction of the impaired oxidative capacity of skeletal muscle in heart failure.15
Afferents sensitive to skeletal muscle work appear to be responsible for hemodynamic, autonomic, and ventilatory changes during exercise in animal models and in humans.19 20 21 23 We hypothesize that the beneficial effect of physical training may be mediated by a reduction of the activity of the muscle ergoreflex, either directly or via reduction of the metabolic signal activating the ergoreceptors.
The major finding of this controlled study was that the contribution of the ergoreceptors to the autonomic, hemodynamic, and respiratory responses to exercise is enhanced in CHF patients: therefore, the heightened sympathetic, vasoconstrictor, and ventilatory drives characteristic of heart failure may be partially explained by an exaggerated ergoreflex contribution during exercise. Physical training not only improved the exercise capacity of the trained forearm but also partially reversed the abnormal responses to exercise via a reduction of this reflex excitation. This benefit was seen in both CHF patients and control subjects, but with training, the responses to the enhanced ergoreflex of CHF patients were brought closer to those of the normal subjects.
We used three different assessment phases: baseline, detraining, and training. Both untrained periods showed similar values for respiratory, hemodynamic, and autonomic variables. In both physiological and pathological conditions, the muscle ergoreflex, together with the baroreflex,39 seems to be a determinant of the exercise-induced rises in systolic and diastolic blood pressures and in the redistribution of the blood flow by increasing the vascular resistance in the nonexercising limbs. These responses could also have been mediated by sympathetic overactivity to the heart and circulation and concomitant vagal withdrawal as a result of the known baroreflex abnormalities in CHF.40 These effects were shown by spectral analysis data of RR-interval and blood pressure variability: increases in the LF components of RR and blood pressure variability and falls in the HF components of RR-interval variability. However, the improvement seen with training of such a small muscle mass leads us to believe that the abnormality in the muscle reflex secondary to the metabolic changes in skeletal muscle in CHF gives the dominant role to the ergoreflex.
Ventilatory responses to exercise are also mediated by the muscle afferents: increases in respiratory frequency and ventilation correlated with ergoreflex activity; these responses could not be explained by any difference in V̇co2, V̇o2, arterial blood gases, or potassium concentrations, although one would not have expected those to change, since the trained, exercising muscle was small (Table 2⇑).
At peak exercise, CHF patients demonstrated higher ventilatory responses than control subjects, with more marked abnormality of the autonomic variables (Table 2⇑). At similar exercise durations (3-minute handgrip), CHF patients presented greater hemodynamic (diastolic pressure, leg vascular resistance) and ventilatory (minute V̇ and respiratory rate) responses and responses to exercise, despite working at the same percentage of maximal handgrip strength. The ergoreflex was more evident in the CHF group with regard to the control of nonexercising limb blood flow and vascular resistance, systolic and diastolic blood pressures, and ventilation, suggesting a major contribution of these receptors in the abnormal responses to exercise (Table 3⇑).
Perhaps surprisingly, we found no differences between CHF and control subjects in the ergoreflex contribution to the autonomic response to exercise. Spectral analysis techniques showed only very low values of variability at baseline, during exercise, and at recovery in the CHF group, and thus they may have been unable to identify any differences in autonomic balance with respect to control subjects.
After training, exercise tolerance improved in both CHF and control subjects: prolonged exercise duration with higher workloads were accompanied by higher peak V̇o2. At matched work duration, however (3-minute handgrip), local conditioning reduced systolic and diastolic blood pressures, V̇, V̇co2, V̇o2, and nonexercising vascular resistance and consequently raised the blood flow, suggesting a more efficient response to exercise (Table 4⇑). These beneficial changes of training were associated with reduced ergoreflex responses in blood pressure, ventilation, and leg blood flow changes, despite a greater workload achieved (and increased work duration). Reduced sympathetic activation was also seen (LF components of power spectral analysis of systolic and diastolic pressure variability).
A localized physical training period of only the muscles of one forearm brought the responses to exercise of the patient group closer to those of the subjects with normal left ventricular function: the differences between CHF patients and control subjects in the ergoreflex contribution to the responses to exercise were less marked after training in regard to blood pressure (Fig 5⇑), leg vascular resistance, leg blood flow, V̇, and the autonomic responses.
Because the ergoreceptors are sensitive to the metabolic changes related to skeletal muscle work, it may be hypothesized that in conditions of higher workloads or durations, the production of metabolites may be greater, with a consequent greater ergoreflex activation. In agreement with this idea, Sterns et al41 reported a higher contribution of ergoreflex to muscle sympathetic nerve activity during PH-RCO after exercise at 40% than at 30% maximal voluntary contraction in small groups of CHF and control subjects. Gandevia and Hobbs,42 evaluating the pressor responses to handgrip and to 3- to 4-minute PH-RCO in normal humans, observed that the responses increased in parallel as the duration (from 45 to 120 seconds) and the intensity (from 33% to 50% of maximal voluntary contraction) of exercise increased.
In our study, however, the control subjects compared with CHF patients and also both groups of subjects after training showed higher V̇o2max and exercise duration but lower ergoreflex contribution compared with the detraining phase (Table 5⇑, Figs 4⇑ and 5⇑), suggesting a nondirect relation between exercise level and ergoreflex contribution. The link between workload and ergoreflex seems to be mediated by the metabolic products of muscle work. After training, the improved muscle metabolism, by delaying the onset of anaerobic metabolism,23 38 could have reduced the production of those metabolites that activate the ergoreceptors, with a resultant lower activity of the reflex effects.
Previous reports of the ergoreceptor reflex control of sympathetic efferent activity in CHF have been conflicting. Sterns et al,41 using a lower exercise load than our study (30% maximal voluntary capacity performed for 2 minutes), reported that during PH-RCO, mean blood pressure and muscle sympathetic nerve activity remained above baseline values for both CHF and control subjects, but only for control subjects were these increases significant. They concluded that the sympathetic nervous response to ergoreceptor activation in CHF was blunted; however, the vastly different baseline levels of sympathetic tone in the two patient groups in that study made it difficult to interpret the incremental effect of metabolic reflex activation. More recently, the same group of workers43 evaluated the muscle ergoreflex at a lower workload (25% maximal voluntary capacity) but for a longer period of time (20 minutes), which was not tolerated by all CHF patients: an increased metabolic reflex activation in CHF was observed. The longer time of exercise performed with perhaps more exercise metabolites produced may have allowed a greater stimulation of the ergoreflex sufficient to differentiate their activity in the two groups of subjects studied.
On the basis of the reported observations and our own experience,21 we chose handgrip exercise at 50% maximal voluntary contraction until exhaustion to highlight the differences of the ergoreflex contribution to exercise in physiological and pathological conditions, such as CHF.
Limitations of the Study
No direct measurements of muscle metabolic activity, muscle mass, or muscle circumference were performed that could have given information regarding the triggers of the ergoreflex activation. However, the metabolic improvements induced by a physical training program similar to the program used in our study suggest that reduced PCr depletion and the acidification of the muscle during exercise24 may be responsible for the reversed ergoreflex overactivation.
Although the method of using the same maximal voluntary contraction to normalize the workload in different group of subjects is well established,41 43 we cannot exclude the possibility that CHF patients were using more muscle than control subjects, which could be responsible for the enhanced ergoreflex activation, or that a smaller mass of muscle was working harder per unit muscle. Thus, after training, the partially reversed overactivity of this reflex could be related to more efficient muscle work or to increased working muscle mass.
To focus on a small muscle group, we studied only forearm exercise rather than exercise that might be more relevant to the effort intolerance of the CHF syndrome. However, a similar protocol (PH-RCO) after leg exercise has been used for studies in normal subjects and also demonstrated the contribution of the ergoreceptors.44 Complete circulatory occlusion would be less certain during leg exercise because of the larger and deeper vessels of the lower limbs and the shape of the thigh, and it was for this reason that we chose the forearm in this study.
Four patients and two control subjects felt slight discomfort during PH-RCO, but there was no correlation between pain and the ergoreflex activity to the hemodynamic, autonomic, or ventilatory responses. Some of the most striking increases in blood pressure, ventilation, and leg vascular resistances occurred in three control subjects and four CHF subjects who did not experience any discomfort. However, a contribution from pain receptors could not be excluded. Small, unmyelinated pain afferents with C fibers may also be similar to or even the same as the skeletal muscle afferents evaluated in this research. This does not preclude their participation in a physiologically relevant reflex response.
The overactivation of this receptor may not be specific to CHF but rather may be common to all conditions of altered muscle metabolism associated with muscle wasting, such as detraining, chronic lung diseases, chronic liver diseases, or neoplastic cachexia. In particular, the cachexia of neoplastic disease has some similarities with cardiac cachexia, and this syndrome too can be associated with unexplained dyspnea.45 In all these situations, effort intolerance could be partially related to an enhanced ergoreflex activation due to an abnormality in the peripheral muscle.
As a result of these experiments and other studies,20 44 a “muscle hypothesis” is proposed as the basis of the generation of the exercise intolerance in CHF. The metabolic state of skeletal muscle is centrally monitored by the activation of ergoreceptors, whose fibers, traveling in the lateral spinothalamic tract, increase ventilation and sympathetic outflow, producing vasoconstriction in distant nonexercising vascular beds, with consequent effects on blood pressure and possibly a small increase in heart rate. They are sensitive to the metabolic state of the muscle, but their triggers are still unclear. They have the properties necessary to link the skeletal muscle abnormality to the fatigue, dyspnea, hyperpnea, and sympathoexcitation characteristic of CHF. This hypothesis proposes another cycle of deterioration similar to those of neuroendocrine activation (Fig 6⇓). A reduction in left ventricular function sets in motion a series of metabolic events that leads to wasting of skeletal muscle and resultant abnormalities of muscular metabolism and function. In response to early metabolic distress in exercising muscle, an exaggerated ergoreflex activation occurs that is perceived by the patient as both muscle fatigue and dyspnea and that leads reflexly to excessive sympathetic vasoconstrictor drive to nonexercising beds and an excessive ventilatory response to exercise.
Symptomatic improvements after treatments for CHF and the exaggerated reflex responses will depend on the resolution of skeletal muscle abnormalities; therefore, they will be delayed after hemodynamic correction and would be most marked in therapies associated with a specific improvement in muscle function or exercise responses.
Selected Abbreviations and Acronyms
|CHF||=||chronic heart failure|
|PH-RCO||=||posthandgrip regional circulatory occlusion|
|RCO||=||regional circulatory occlusion|
Dr Piepoli is a European Society of Cardiology Research Fellow. Dr Clark is a Robert Luff Research Fellow. Dr Coats is the British Heart Foundation and Viscount Royston Senior Lecturer in Cardiomyopathy. We gratefully acknowledge the advice and support of Dr James Conway and Dr Theo E. Meyer and the expertise of Dr Luciano Bernardi.
- Received June 13, 1995.
- Revision received September 20, 1995.
- Accepted October 4, 1995.
- Copyright © 1996 by American Heart Association
Buller NB, Poole-Wilson PA. Mechanism of the increased ventilatory response to exercise in patients with chronic heart failure. Br Heart J. 1990;63:281-283.
Lipkin DP, Perrins J, Poole-Wilson PA. Respiratory gas exchange in the assessment of patients with impaired ventricular function. Br Heart J. 1985;54:321-328.
Drexler H, Riede U, Münzel T, König H, Funke E, Just H. Alterations of skeletal muscle in chronic heart failure. Circulation. 1992;85:1751-1759.
Sullivan MJ, Green HJ, Cobb FR. Skeletal muscle biochemistry and histology in ambulatory patients with long-term heart failure. Circulation. 1990;81:518-527.
Massie BM, Conway M, Yonge R, Frostick S, Sleight P, Ledingham J, Radda G, Rajagopalan B. Skeletal muscle metabolism in patients with congestive heart failure: relation to clinical severity and blood flow. Circulation. 1987;76:1009-1019.
Buller NP, Jones D, Poole-Wilson PA. Direct measurement of skeletal muscle fatigue in patients with chronic heart failure. Br Heart J. 1991;65:20-24.
Francis GS, Goldsmith SR, Olivari MT, Levine TB, Cohn JN. The neurohumoral axis in congestive heart failure. Ann Intern Med. 1984;101:370-377.
Adamopoulos S, Piepoli M, McCance A, Bernardi L, Radaelli A, Ormerod O, Forfar C, Sleight P, Coats AJS. Comparison of different methods for assessing sympathovagal balance in chronic congestive heart failure secondary to coronary artery disease. Am J Cardiol. 1992;70:1576-1582.
Floras JS. Clinical aspects of sympathetic activation and parasympathetic withdrawal in heart failure. J Am Coll Cardiol. 1993;22:72A-84A.
Sullivan MJ, Higginbotham MB, Cobb FR. Exercise training in patients with chronic heart failure delays ventilatory anaerobic threshold and improves submaximal exercise performance. Circulation. 1989;79:324-329.
Hagberg JM, King ID, Rogers MA, Montain SJ, Jilka SM, Kohrt WM, Heller SL. Exercise hyperventilation and recovery in patients with McArdle’s disease. J Appl Physiol. 1989;52:991-994.
Coats AJS, Adamopoulos S, Radaelli A, McCance A, Meyer TE, Bernardi L, Solda PL, Davey P, Ormerod O, Forfar C, Conway J, Sleight P. Controlled trial of physical training in chronic heart failure: exercise performance, hemodynamics, ventilation and autonomic function. Circulation. 1992;85:2119-2131.
Coats AJS, Clark AL, Piepoli M, Volterrani M, Poole-Wilson PA. Symptoms and quality of life in heart failure: the muscle hypothesis. Br Heart J. 1994;72:S36-S39.
Kao FF. An experimental study of the pathways involved in exercise hyperpnea employing cross-circulation techniques. In: Cunningham DC, Lloyd BB, eds. The Regulation of Human Respiration. Oxford, UK: Blackwell; 1963:461-502.
Tibes U. Reflex inputs to the cardiovascular and respiratory centers from dynamically working canine muscles: some evidence for involvement of group III or IV fibers. Circ Res. 1977;41:332-341.
Piepoli M, Clark AL, Coats AJS. Muscle metaboreceptor in the hemodynamic, autonomic and ventilatory responses to exercise in men. Am J Physiol. 1995;(Heart Circ Physiol 38):H1428-H1436.
Stratton JR, Dunn JF, Adamopoulos S, Kemp GJ, Coats AJS, Rajagopalan B. Training partially reverses skeletal muscle metabolic abnormalities during exercise in heart failure. J Appl Physiol. 1994;76:1575-1582.
Piepoli M, Coats AJS, Adamopoulos S, Bernardi L, Feng YH, Conway J, Sleight P. Persistent peripheral vasodilatation and sympathetic activity in hypotension after maximal exercise. J Appl Physiol. 1993;75:1807-1814.
Malliani A, Pagani M, Lombardi F, Cerutti S. Cardiovascular neural regulation explored in the frequency domain. Circulation. 1991;84:482-492.
Greenfield A, Whitney R, Mowbray J. Methods for the investigation of peripheral blood flow. Br Med Bull. 1963;19:101-109.
Sullivan MJ, Higginbotham MB, Cobb FR. Exercise training in patients with severe left ventricular dysfunction: hemodynamic and metabolic effects. Circulation. 1988;78:506-515.
Zelis R, Longhurst J, Capone RJ, Mason DT. A comparison of regional blood flow and oxygen utilization during dynamic forearm exercise in normal subjects and patients with congestive heart failure. Circulation. 1974;50:137-143.
Wilson JR, Fink L, Maris J, Ferraro N, Power-Vanwart J, Eleff S, Chance B. Evaluation of energy metabolism in skeletal muscle of patients with heart failure with gated phosphorus-31 nuclear magnetic resonance. Circulation. 1985;71:57-62.
Massie BM, Conway M, Rajagopalan B, Yonge R, Frostick S, Ledingham J, Sleight P, Radda G. Skeletal muscle metabolism during exercise under ischemic conditions in congestive heart failure: evidence for abnormalities unrelated to blood flow. Circulation. 1988;78:320-326.
Weiner DH, Fink LI, Maris J, Jones RA, Chance B, Wilson JR. Abnormal skeletal muscle bioenergetics during exercise in patients with heart failure: role of the reduced muscle blood flow. Circulation. 1986;73:1127-1136.
Volterrani M, Clark AL, Ludman PF, Swan JW, Adamopoulos A, Piepoli M, Coats AJS. Predictors of exercise capacity in chronic heart failure. Eur Heart J. 1994;14:801-809.
Mancini DM, Henson D, LaManca J, Levine S. Respiratory muscle function and dyspnea in patients with chronic congestive heart failure. Circulation. 1992;86:909-918.
Clark AL, Sparrow JL, Coats AJS. Muscle fatigue and dyspnoea in chronic heart failure: two sides of the same coin? Eur Heart J. 1995;16:49-52.
Minotti JR, Johnson EC, Hudson TL, Zuroske G, Murata G, Fukushima E, Cagle TG, Chick TW, Massie BM, Icenogle MV. Skeletal muscle response to exercise training in congestive heart failure. J Clin Invest. 1990;86:751-758.
Melcher A, Donald DE. Maintained ability of carotid baroreflex to regulate arterial pressure during exercise. Am J Physiol. 1981;241:H838-H849.
Thames MD, Kunigawa T, Smith ML, Dibner-Dunlap ME. Abnormalities of baroreflex control in heart failure. J Am Coll Cardiol. 1993;22(suppl A):56A-60A.
Sterns DA, Ettinger SM, Gray KS, Whisler SK, Mosher TJ, Smith MB, Sinoway LI. Skeletal muscle metaboreceptor exercise responses are attenuated in heart failure. Circulation. 1991;84:2034-2039.
Gandevia SC, Hobbs SF. Cardiovascular responses to static exercise in man: central and reflex contributions. J Physiol. 1990;430:104-117.
Sutliff GM, Sinoway LI, Leuenberger UA. Rhythmic exercise leads to muscle metaboreflex activation in heart failure but not in controls. Circulation. 1994;90(suppl I):I-539. Abstract.