Journal Logo


The Mechanoreflex and Hemodynamic Response to Passive Leg Movement in Heart Failure


Author Information
Medicine & Science in Sports & Exercise: March 2016 - Volume 48 - Issue 3 - p 368–376
doi: 10.1249/MSS.0000000000000782
  • Free


Heart failure (HF) is typically the result of a myocardial insult or exaggerated afterload, both of which compromise myocardial performance and, subsequently, systemic hemodynamics (26). Although compromised central hemodynamics is the hallmark of HF, the notion that peripheral dysfunction exacerbates these central hemodynamic abnormalities, the “muscle hypothesis” (7), has been increasingly recognized (9,18,26,37). Specifically, the exercise pressor reflex (EPR) has been reported to be upregulated in both animal models of HF (30) and in patients with HF (24,25), leading to excessive increases in HR, ventilation (V˙E), sympathetic nerve activity, and mean arterial pressure (MAP). The sensitization of this reflex has the potential to result in greater exertional dyspnea and increased cardiac afterload, exaggerating myocardial work, reducing exercise tolerance, and subsequently enhancing disease progression because of inactivity.

The hypersensitivity of the EPR in patients with HF is now recognized to be mediated by exaggerated afferent signaling via the thinly myelinated “mechanosensitive” group III afferents and/or the unmyelinated “metabosensitive” group IV afferents (3). Indeed, our group (3) recently demonstrated that partial afferent blockade with the μ-opioid agonist, fentanyl, reduced central hemodynamic responses, MAP, and norepinephrine spillover, resulting in an improvement in leg blood flow and oxygen delivery during knee extensor exercise in patients with HF. Unfortunately, with this integrative physiological model, it was not possible to determine the contribution of mechanosensitive and metabosensitive afferent signals to the overall cardiovascular response. However, given the experimental design, we were able to exclude the role of feed-forward signaling (3). Previous studies that have selectively activated the metaboreflex suggest that this reflex is normal (6), or even reduced (18,33), in those with HF. Moreover, Middlekauff et al. (19) used low-intensity handgrip and passive wrist flexion as a means to selectively activate and study the mechanoreflex and revealed an exaggerated renal vasoconstriction (19) paralleled by an excessive rise in muscle sympathetic nerve activity in those with HF (17). Although, in combination, these findings and other work in animal models (26) imply that an enhanced mechanoreflex is the likely factor mediating the sensitization of the EPR in this population, a comprehensive assessment of the role of the mechanoreflex and the central and peripheral hemodynamic consequences in humans with HF has yet to be performed.

Previously, our group (34,39) and others (10,21) have used dynamic passive leg movement (PLM) as a model to study exercise-induced hyperemia and the contribution of the mechanoreflex in mediating this response. Importantly, Trinity et al. (34) revealed that the partial blockade of afferent signals with fentanyl significantly reduced both the central and peripheral hemodynamic responses to PLM in young healthy individuals, highlighting an important role of mechanosensitive afferents in facilitating the hemodynamic response to movement. Witman et al. (39) subsequently demonstrated a significant attenuation of PLM-induced hyperemia in patients with HF compared with that in healthy controls. However, to date, the pharmacological blockade of afferent signals during PLM has not yet been used to determine the role of the mechanosensitive afferents in the EPR of patients with HF.

Accordingly, this study sought to determine the contribution of mechanosensitive afferents in patients with HF using a reductionist approach using PLM and the intrathecal injection of fentanyl to blunt neural feedback. We hypothesized that the attenuated afferent feedback would suppress the central hemodynamic but peripheral hemodynamics would be significantly improved, suggestive of a negative role of the mechanoreflex in HF. This, if proven to be correct, would provide considerable insight into the mechanisms responsible for the exercise intolerance, such as excess sympathetic vasoconstriction, in this patient population.


Subjects And General Procedures

Ten patients diagnosed with New York Heart Association (NYHA) class II HF (reduced ejection fraction) participated in this study. The protocol was approved by the institutional review boards of the University of Utah and the Salt Lake City VA Medical Center. A written informed consent was obtained from all subjects before their participation in the study. All studies were performed in a thermoneutral environment (22°C) at an elevation of approximately 1419 m. Subjects reported to the laboratory in a fasted state and without caffeine or alcohol use for 12 and 24 h, respectively. They also had not performed any exercise within the past 24 h. For safety and ethical reasons, all subjects maintained their medication regimen on the day of the study.

PLM Protocol

All subjects underwent a noninvasive familiarization trial on the day before the catheter-based experiment to become acquainted with the instrumentation and PLM protocol. On the study day, before the experimental protocols, arterial and venous catheters were placed under local anesthesia (1% lidocaine) in the right common femoral artery and vein using sterile technique. After catheter placement, the patients rested for 30 min. Once they have recovered, subjects were moved to a comfortable chair for the hypercapnic ventilatory response test (HCVR) to establish a baseline hyperventilatory response to 3% and 6% CO2, a process repeated in the fentanyl condition to ensure the drug had not migrated cephalically (1). Subjects were then escorted to the knee extensor ergometer and rested in an upright seated position for 20 min before the start of data collection.

The lower leg of the subject was fitted into a boot attached to a custom-built single-leg knee extension ergometer, as described elsewhere (4,9). The protocol consisted of a 60-s resting baseline followed by a 3-min bout of single-leg PLM performed in an upright seated position, as described previously (34). After completion of the control trial, subjects rested for 1 h. After the rest period, subjects were positioned in the upright seated position, and under sterile conditions, subjects were administered local anesthesia (1% lidocaine) at the level of the L3–L4 intervertebral space. As described previously (2), 1 mL of fentanyl (0.05 mg·mL−1) was injected into the intrathecal space to pharmacologically blunt afferent feedback, which was confirmed in each subject with the assessment of cutaneous hypoesthesia to a pinprick and cold perception as well as paresthesia to hand strokes on the torso and lower limb dermatomes. Subjects remained in the upright seated position for the remainder of the protocol to minimize potential cephalic migration of the fentanyl. The HCVR test was repeated, followed by the baseline measurements and passive movement protocol, as described previously.


Central hemodynamic variables

HR, stroke volume (SV), and cardiac output (CO) were determined with the Finometer (Finapress Medical Systems BV, Amsterdam, The Netherlands). SV was calculated using the modelflow method (36), which includes age, sex, height, and weight in its algorithm (Beatscope version 1.1; Finapres Medical Systems, Amsterdam, The Netherlands), and has previously been documented to accurately track SV at rest and during exercise (8,13,20,27) including those with cardiovascular disease (5). CO was then calculated as the product of HR and SV. Arterial and venous pressure transducers placed at the level of the catheter and measurement site (common femoral artery and vein) were used for real-time determination of mean arterial and venous pressure and then used to calculate perfusion pressure across the leg.

Femoral blood flow

Measurements of femoral arterial blood velocity and vessel diameter were performed in the passively moved leg distal to the inguinal ligament and proximal to the bifurcation of the superficial and deep femoral arteries with a Logic 7 Doppler ultrasound system (General Electric (GE) Medical Systems, Milwaukee, WI) operated by a trained technician. The Logic 7 system was equipped with a linear array transducer operating at an imaging frequency of 12 MHz. Vessel diameter was determined at a perpendicular angle along the central axis of the scanned area. Blood velocity was obtained using the same transducer with a Doppler frequency of 5 MHz. All blood velocity measurements were obtained with the probe appropriately positioned to maintain an insonation angle of 60° or less. The sample volume was maximized according to vessel size and was centered within the vessel based on real-time ultrasound visualization. Arterial diameter was measured, and angle was corrected. Intensity weighted mean velocity (Vmean) values were then calculated using commercially available software (GE Medical Systems, Milwaukee, WI). Using arterial diameter and Vmean, Femoral blood flow (FBF) was calculated as: Vmeanπ (vessel diameter/2)2 × 60, where FBF is in milliliters per minute (mL·min−1). To account for potential differences in MAP, FVC was calculated as follows: FBF/MAP.

Tissue oxygenation

Frequency-domain, multidistance near-infrared spectroscopy (NIRS) of the vastus lateralis muscle (Oxiplex TS; ISS, Champaign, IL) in the passively moved leg allowed the absolute quantitation of the absorption and scattering coefficients of the chromophores, hemoglobin (Hb), and myoglobin (Mb) (11). While subsequent calculations allowed the determination of tissue oxygen saturation (StO2%), oxy-Hb and oxy-Mb (Hb + MbO2), deoxy-Hb and deoxy-Mb (HHb + Mb), and total Hb + Mb concentrations (all expressed in μM). Before use, in the control and fentanyl conditions, the probe was calibrated using a phantom block with known absorption and scattering coefficients. Before placement, the site of assessment over the vastus lateralis was extensively cleaned, the diode was attached with double-sided adhesive tape, and the area was covered and further secured with coban wrap (3 M, St. Paul, MN). The data were acquired at 0.5 Hz averaged over 30 s at baseline and during the passive movement. Although frequency-domain, multidistance NIRS in combination with skinfold measurements can allow a correction for overlying fat, because of the nature of this investigation, PLM in a thermoneutral environment, this approach was not used in the current study.


The ventilatory responses (V˙E, tidal volume (VT), and breathing frequency (Bf)) to the HCVR test and passive movement were assessed using a nonrebreathing mouthpiece connected to a mixing chamber and gas analysis system (Parvo Medics, Salt Lake City, UT). Data were collected breath-by-breath and averaged over 10 s before analysis.

Blood sampling

Arterial and venous blood samples, obtained at rest and during passive movement, were drawn into heparinized syringes and used for blood gas analyses or saved for biochemical analyses. Blood gas samples were analyzed using standard clinical technique (GEM 4000; Instrumentation Laboratory, Bedford, MA). Remaining blood samples were analyzed for epinephrine and norepinephrine using a quantitative enzyme immunoassay (2-CAT ELISA; Labor Diagnostika Nord GmbH & Co. KG). Samples were measured in triplicate. From these measurements, norepinephrine spillover was determined using the following formula:

where, NE = norepinephrine and Epi = epinephrine.

Data acquisition

Throughout the protocol, signals reflecting HR, SV, CO, MAP, and knee joint angle underwent analog to digital conversion and were simultaneously acquired (200 Hz) using commercially available data acquisition software (AcqKnowledge; Biopac Systems Inc., Goleta, CA). In addition, the audio antegrade and retrograde signals from the Doppler ultrasound system were acquired (10,000 Hz) to serve as qualitative indicators of blood velocity changes and to ensure accurate temporal alignment of blood velocity measurements obtained from this system and the other variables collected (i.e., HR, SV, CO, and MAP).

Data Analysis

From the velocity and femoral artery diameter, net FBF was calculated on a second-by-second basis for the passively moved leg. Before analysis, all hemodynamic data were smoothed using a 3-s rolling average. To identify condition-specific responses across time, two-way (condition × time) repeated-measures ANOVA tests were used. The steady-state FBF responses (second to third minute of movement) were also assessed in all subjects (n = 10) to pair with the blood samples that were drawn during the last 30 s. As the responses to passive movement are transient and vary between individuals, a peak response was determined for all variables on an individual basis for a subset of patients (n = 6). This was then compared between the control and fentanyl conditions by paired t-tests as was the determination of individual baseline, maximal, absolute and relative changes, and time to maximal response in all measured variables. Alpha was set at 0.05 for all comparisons. All data are presented as mean ± SE.


Subject and clinical characteristics are listed in Table 1. All subjects produced a similar hyperventilatory response to the hypercapnic gas in the control and fentanyl conditions, indicating that the fentanyl had not migrated cephalically to the brainstem, leaving central chemoreceptor function intact. The efficacy of the blockade was verified in all subjects who uniformly exhibited sensory reductions at the level of T5–T6 and below, including the lower limbs, whereas no such changes were observed above this level and in the upper limbs.

Subject characteristics.

Central hemodynamic responses

Fentanyl had no significant effect on resting central hemodynamics, such that CO (6.5 ± 0.6 vs 5.9 ± 0.6 L·min−1), SV (108 ± 10 vs 94 ± 7 mL per beat), and HR (61 ± 2 vs 62 ± 3 bpm) were similar between control and fentanyl conditions, respectively (Fig. 1). The peak PLM-induced changes in CO (1.3 ± 0.2 vs 1.3 ± 0.2 ΔL·min−1), SV (19 ± 3 vs 20 ± 2 ΔmL per beat), and HR (6 ± 1 vs 9 ± 3 Δbpm) were not different after fentanyl injection (Fig. 1). The area under the curve (AUC) for CO tended to be reduced (control vs fentanyl, 0.59 ± 0.09 vs 0.39 ± 0.10 L), but this did not reach statistical significance (P = 0.10). There was no effect of fentanyl on the time to peak response for any of the central hemodynamic variables.

Central hemodynamic responses to PLM. CO (A), SV (B), HR (C), with and without (control) intrathecal fentanyl (n = 10).

Peripheral hemodynamic responses

Fentanyl had no effect on resting FBF (278 ± 97 vs 246 ± 49 mL·min−1), MAP (108 ± 2 vs 108 ± 6 mm Hg), FVC (2.6 ± 0.9 vs 2.6 ± 0.7 mL·min−1·mm Hg−1), nor the oscillatory nature of FBF; antegrade (416 ± 81 vs 369 ± 57 mL·min−1) and retrograde (138 ± 52 vs 101 ± 29 mL·min−1). The peak PLM-induced hyperemia was significantly greater after fentanyl (493 ± 155 vs 804 ± 198 ΔmL·min−1) (Fig. 2). As the MAP response was not different between control and fentanyl trials (7 ± 1 vs 10 ± 3 Δmm Hg), FVC was also augmented with fentanyl blockade (4.7 ± 2 vs 8.5 ± 3 ΔmL·min−1·mm Hg−1) (Fig. 3). AUC analysis of the first minute of passive movement also revealed that net FBF AUC was significantly increased after fentanyl (153 ± 30 vs 314 ± 70 mL), although neither antegrade FBF AUC (254 ± 25 vs 401 ± 62 mL) nor retrograde FBF AUC (143 ± 43 vs 110 ± 49 mL) was statistically different (Fig. 2). Assessment of the peak microvascular responses using NIRS revealed a significant effect of PLM resulting in increased StO2%, Hb + MbO2, and total Hb and Mb (THb + Mb), whereas HHb + Mb was decreased (Table 2). Intrathecal fentanyl enhanced the PLM-induced increase in StO2% but had no effect on Hb + Mb O2, HHb + Mb, or THb + Mb. There was no effect of fentanyl on the time to peak response for any of the peripheral variables.

NIRS-derived variables assessed in the vastus lateralis at rest and during PLM (peak and steady-state responses) with and without (control) intrathecal fentanyl.
FBF responses to PLM. Net FBF (A), antegrade FBF (B), and retrograde FBF (C), with and without (control) intrathecal fentanyl. *P < 0.05 vs control, n = 6.
Peripheral hemodynamic response to PLM. Femoral MAP (n = 10) (A) and FVC (B), with and without (control) intrathecal fentanyl (n = 6). *P < 0.05 vs control.

Analysis of the steady-state peripheral hemodynamic responses, obtained during minutes 2–3 of PLM, revealed a significantly elevated net blood flow (481 ± 85 vs 583 ± 86 mL·min−1, P < 0.05), no difference in antegrade FBF (779 ± 68 vs 810 ± 57 mL·min−1, P > 0.05), and a tendency for a reduction in retrograde FBF (303 ± 67 vs 232 ± 46 mL·min−1, P = 0.07) in the presence of fentanyl. In parallel, the NIRS data revealed a significant PLM-induced increase in StO2%, which was enhanced in the fentanyl condition (Table 2). There was a significant main effect of passive movement on HHb + Mb; however, there was no effect of fentanyl (Table 2).


Analysis of the individual peak responses in the control trial revealed that PLM significantly increased V˙E, Bf, and VT, (Table 3). Fentanyl had no effect on resting V˙E, Bf, or VT (Table 3). Fentanyl had no effect on the peak V˙E response to PLM, as V˙E, Bf, and VT responses were similar between conditions (control vs fentanyl, respectively).

E and blood gas variables at rest and in response to PLM with and without intrathecal fentanyl.

Analysis of the steady-state responses during minutes 2–3 of PLM in the control condition revealed a small but significant effect of PLM on V˙E but did not significantly alter respiratory rate or VT. Fentanyl also had no effect on the PLM-induced change in V˙E, Bf, or VT (all P > 0.05, control vs fentanyl, respectively).

Blood analyses

The analyses of blood variables are presented in Table 3. PLM reduced venous oxygen content, oxyhemoglobin, increased oxygen delivery, fractional oxygen extraction, and leg V˙O2 (Table 3). Fentanyl resulted in a greater PLM-induced increase in oxygen delivery and reduction in venous saturation, oxygen content, and oxyhemoglobin (Table 3). Biochemical analysis of the blood samples revealed a clear increase in norepinephrine spillover as a result of PLM (P < 0.05), an effect that tended to be attenuated after intrathecal fentanyl injection (103% ± 19% vs 58% ± 17%Δ, P = 0.06).


This study sought to elucidate the role of mechanosensitive muscle afferents on cardiovascular responses to PLM in patients with HF. During PLM, pharmacological blunting of afferent feedback with fentanyl had no effect on central hemodynamics or pulmonary V˙E but significantly improved the peripheral hemodynamic response. Indeed, the partial blockade of afferent signaling tended to reduce both the PLM-induced increase in norepinephrine spillover and retrograde FBF, which significantly improved FBF, oxygen delivery, and StO2%. These findings suggest that, under normal conditions, in patients with HF, a heightened mechanoreflex augments peripheral sympathetic vasoconstriction, a phenomenon that may confound existing central hemodynamic abnormalities, further contributing to exercise intolerance in this patient population.

The mechanoreflex and central hemodynamics in HF

In terms of the contribution of CO to the EPR in HF and the role of mechanoreflex in mediating this response, previous work in animal models of HF using passive muscle stretch has revealed either a significant augmentation of the HR response (30,31) or no change at all (14). Subsequent animal studies have revealed that passive static stretch increased group III but not group IV afferent nerve discharge, an effect exaggerated by HF (35). Previous studies in humans with HF documented either no change in HR as a result of passive wrist flexion (17), or the authors did not report HR, SV, and CO during passive knee extension (28), the latter making any comparison between the current work and these investigations impossible. Of note, the authors who reported no change in HR during passive wrist flexion (17) did not indicate the frequency of the passive movement, and previous studies suggest that mechanoreceptors are likely frequency dependent (15). Therefore, unlike the current study (Fig. 1), the movement in this previous study (17) may not have been performed at a high enough frequency to elicit a significant increase in HR. In support of this interpretation, our group has demonstrated in young healthy individuals that slow passive movement (1°·s−1) produced no cardioacceleration (16) unlike the 1 Hz used in the current study. In addition, mechanoreflex activation of CO may be muscle mass dependent, with the small muscle mass perturbed during passive wrist flexion not providing enough afferent feedback to stimulate an increase in HR.

In the current study, using passive knee extension, in agreement with our previous work (39), we observed a significant increase in HR and CO in response to PLM. Although there was not a statistically significant effect of the fentanyl on these responses, CO tended to be lower (P = 0.10) after partial afferent blockade (Fig. 1). Our findings contrast with those of Smith et al. (31), who found that afferent neural blockade with gadolinium significantly reduced, but did not abolish, the tachycardia as a result of tendon stretch in rodents with dilated cardiomyopathy. However, dynamic PLM likely elicits lower levels of tendon stretch compared with the static approach used by Smith et al. (31) and others (32), and, furthermore, fentanyl (a targeted μ-opioid agonist that reduces afferent feedback) and gadolinium (a paramagnetic rare earth metal that alters stretch-activated channels) are very different pharmacological approaches. In summary, using the current PLM method, humans with HF seem to exhibit only a modest afferent-sensitive CO response.

The mechanoreflex and peripheral hemodynamics in HF

During exercise, patients with HF often exhibit reduced muscle blood flow and convective O2 transport (3,9,26,37) as well as exaggerated renal vasoconstriction (18), implicating exaggerated total peripheral resistance (TPR) as a contributor to the EPR in HF (38). Recent work performed by our group revealed that during knee extensor exercise, patients with HF exhibit reduced FBF (3), which was somewhat restored to the level of controls after partial blockade of lower limb muscle afferents. This fentanyl-induced increase in FBF in the patients with HF was paralleled by a blunted MAP response, augmenting femoral vascular conductance (FVC) during knee extensor exercise over a wide range of exercise intensities (3). These observations suggest a role of afferent neurons in altered peripheral hemodynamics in these patients. However, the cardiovascular response to exercise is a composite of central command as well as metabo- and mechanosensitive reflexes, confounding the ability to ascertain the factor responsible for the greater TPR and reduced muscle blood flow during exercise. Interestingly, there has been little work investigating the peripheral hemodynamic consequences of selective mechanoreflex activation in animals or humans with HF.

Middlekauff et al. (19) used a human model of low-intensity involuntary muscle contractions to selectively activate the mechanoreflex in the absence of central command and found that patients with HF displayed a disproportionate rise in renal vascular resistance compared with age-matched controls, suggestive of an enhanced mechanoreflex. Subsequently, the same group (17) demonstrated that the contribution of the metaboreflex to muscle sympathetic nerve activity during handgrip exercise in patients with HF was minimal and that neither pH nor lactate seemed to sensitize the mechanoreflex. In addition, using passive wrist flexion–extension, the authors documented an increase in muscle sympathetic nerve activity in patients with HF, but not controls, confirming enhanced mechanoreflex in this population (17). However, in this study (17), the peripheral vascular response was not measured; thus, they were unable to determine whether the rise in muscle sympathetic nerve activity did in fact affect muscle blood flow. Subsequently, our group (39) used PLM to assess the peripheral hemodynamic response in patients with HF and age-matched healthy controls, revealing a significantly blunted FBF response to PLM in those with HF. Although a blunted hyperemia was observed, the exact mechanism responsible could not be identified in this previous study but was suggested to be peripheral in nature.

In the current study, activation of the mechanoreflex with PLM resulted in an increase in MAP (Fig. 3), FBF (Fig. 2), and FVC (Fig. 3). However, this vasodilatory response seems to be restrained by muscle sympathetic nerve activity, as norepinephrine spillover tended to increase concomitantly. Furthermore, pharmacological blunting of afferent feedback with fentanyl revealed a tendency for less of an increase in norepinephrine spillover and lower retrograde FBF, resulting in significantly greater peak and steady-state net FBF (Fig. 2) and FVC (Fig. 3). Padilla et al. (23), using multiple sympathoexcitatory maneuvers (e.g., cold pressor test) have documented that increases in muscle sympathetic nerve activity result in increased retrograde FBF, highlighting retrograde FBF as an index of vascular resistance downstream from the conduit artery (12). In this context, we propose that PLM stimulates mechanosensitive afferent nerves, leading to a reflex increase in muscle sympathetic nerve activity-mediated norepinephrine release, which increases resistance downstream, restraining FBF, and FVC (Fig. 3). This sympathetic restraint of FBF ultimately results in less tissue oxygenation (StO2%), as detected by NIRS, such that fentanyl improved the PLM-induced peak and steady-state StO2% (Table 2). Based on the previously observed mechanoreceptor-induced increase in muscle sympathetic nerve activity in patients with HF, but not controls (17), and the current tendency for an increase in norepinephrine as a result of PLM, it is likely that such enhanced afferent feedback, specific to the mechanoreflex is, at least partly, responsible for the reduced leg blood flow and muscle perfusion often associated with this population during exercise (26). In addition, although small in magnitude because of the passive experimental model, the documented improvement in oxygen delivery in the leg being moved, the reduced venous oxygen content, and the resultant greater arterial–venous oxygen difference and leg oxygen uptake (Table 3) after fentanyl-induced blockade support this contention.

The mechanoreflex and V˙E in HF

Previous work (25,28) suggests that patients with HF exhibit an enhanced ergoreflex, as measured by the hyperventilatory response to exercise, and that this disproportionately elevated hyperpnea contributes to their exercise intolerance, as suggested in the “muscle hypothesis” (7). These authors, through the use of postexercise circulatory occlusion to isolate the metaboreflex, suggest that ergoreflex activation is likely mediated by metabosensitive afferents (28), which can be ameliorated by reducing H+ concentration with infusion of sodium bicarbonate (29). To further determine whether mechanosensitive afferents were contributing to the exercise-induced hyperpnea, Scott et al. (28) used passive knee extension, as in the current study, and found an increase in V˙E, V˙O2, and the ratio of V˙O2 to V˙E, which was similar in both patients with HF and age-matched healthy controls. Our data are in agreement with those of Scott et al. (28), revealing a small but significant increase in peak V˙E in parallel with increased leg V˙O2 (Table 3). Interestingly, the increased V˙O2 observed with fentanyl may be interpreted, as others have (26), that HF shifts metabolic control, at least partly, from the mitochondria to the vasculature. Somewhat surprisingly, blunting of afferent feedback via intrathecal fentanyl had no effect on V˙E in current patients with HF, a finding that contrasts with those of Olson et al. (22) that revealed a fentanyl-induced reduction in V˙E during cycling exercise. The results from the current study suggest little or no role of mechanosensitive afferents alone in mediating the exaggerated hyperpnea often observed in patients with HF during exercise.

Experimental considerations

The current study and findings are not without experimental considerations. First, the patients enroled in this study were all designated as NYHA class II HF and, therefore, it is unclear how variations in HF severity would affect the results of this study; thus, the current findings cannot be confidently extrapolated to all classes or types of HF. Second, although a clinically relevant dose of fentanyl was administered and used in routine clinical practice and in previous research studies in health (1,2) and HF (3,22), which resulted in significant physiological changes in this and previous studies likely through partial blockade of afferent feedback, the actual degree of blockade remains unknown. In addition, in terms of the effect of fentanyl blockade, while the HCVR test to rule out the cephalic migration of fentanyl was performed before the PLM, this assessment was not repeated after PLM to again assess potential fentanyl migration. However, previous work (22), also in HF, measured the HCVR after exercise and found no evidence of cephalic migration in patients with HF who were maintained upright postfentanyl delivery, as in the current study. Third, although we contend that the mechanosensitive afferents are being activated with this passive movement model, there is a possible contribution from other afferent nerves, and thus, the exact mechanism (e.g., extramuscular/joint or metabosensitive afferents) responsible for the fentanyl-induced changes observed in this study cannot be definitively identified and remains an area to be further explored. Finally, to minimize subject burden, the current study did not include age-matched controls. Such an assessment, using exactly the same protocol, would be a useful future endeavor.


Using a comprehensive and integrative approach to understand the role of the mechanosensitive component of the EPR in patients with HF, the current study provides evidence that feedback from mechanosensitive muscle afferents may play a negative role in patients with HF. Specifically, likely as a consequence of peripheral sympathetic vasoconstriction, mechanosensitive afferent feedback attenuates the hyperemic response to PLM and tissue oxygenation. This phenomenon may contribute to exercise intolerance in patients with HF and may help guide future pharmacological therapy or exercise interventions to minimize the effect of mechanosensitive muscle afferents in this population.

The authors would like to thank the patients for their gracious participation as well as the staff at the cardiology clinic at the Salt Lake City VA Hospital, Dr. Josef Stehlik, and nurse practitioners Mary Beth Hagan and Robin Waxman. We also would like to thank Van Reese for performing the blood assays.

In addition, we would like to acknowledge financial support from NIH P01 HL-091830 (R. S. R), NIH R01 HL118313 (D. W. W.), I21RX001572 (M. A.), HL116579 (M. A.), HL103786 (M. A.), AHA14-17770016 (M. A.), VA RR&D 1121RX001418-01 (D. W. W.), and VA Merit grant E6910R (R. S. R.). Advanced Fellowships in Geriatrics from the Department of Veterans Affairs supported S. J. I. and M. A. H. W.

The authors declare no conflicts of interest.

The results of the present study do not constitute endorsement by the ACSM.


1. Amann M, Blain GM, Proctor LT, Sebranek JJ, Pegelow DF, Dempsey JA. Group III and IV muscle afferents contribute to ventilatory and cardiovascular response to rhythmic exercise in humans. J Appl Physiol (1985). 2010; 109(4): 966–76.
2. Amann M, Proctor LT, Sebranek JJ, Pegelow DF, Dempsey JA. Opioid-mediated muscle afferents inhibit central motor drive and limit peripheral muscle fatigue development in humans. J Physiol. 2009; 587(Pt 1): 271–83.
3. Amann M, Venturelli M, Ives SJ, et al. Group III/IV muscle afferents impair limb blood in patients with chronic heart failure. Int J Cardiol. 2014; 174(2): 368–75.
4. Andersen P, Saltin B. Maximal perfusion of skeletal muscle in man. J Physiol. 1985; 366: 233–49.
5. Bos WJ, Imholz BP, van Goudoever J, Wesseling KH, van Montfrans GA. The reliability of noninvasive continuous finger blood pressure measurement in patients with both hypertension and vascular disease. Am J Hypertens. 1992; 5(8): 529–35.
6. Carrington CA, Fisher JP, Davies MK, White MJ. Muscle afferent inputs to cardiovascular control during isometric exercise vary with muscle group in patients with chronic heart failure. Clin Sci (Lond). 2004; 107(2): 197–204.
7. Coats AJ, 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(2 Suppl): S36–9.
8. Critchley LA, Lee A, Ho AM. A critical review of the ability of continuous cardiac output monitors to measure trends in cardiac output. Anesth Analg. 2010; 111(5): 1180–92.
9. Esposito F, Mathieu-Costello O, Shabetai R, Wagner PD, Richardson RS. Limited maximal exercise capacity in patients with chronic heart failure: partitioning the contributors. J Am Coll Cardiol. 2010; 55(18): 1945–54.
10. González-Alonso J, Mortensen SP, Jeppesen TD, et al. Haemodynamic responses to exercise, ATP infusion and thigh compression in humans: insight into the role of muscle mechanisms on cardiovascular function. J Physiol. 2008; 586(9): 2405–17.
11. Gratton E, Fantini S, Franceschini MA, Gratton G, Fabiani M. Measurements of scattering and absorption changes in muscle and brain. Philos Trans R Soc Lond B Biol Sci. 1997; 352(1354): 727–35.
12. Halliwill JR, Minson CT. Retrograde shear: backwards into the future? Am J Physiol Heart Circ Physiol. 2010; 298(4): H1126–7.
13. Harms MP, Wesseling KH, Pott F, et al. Continuous stroke volume monitoring by modelling flow from non-invasive measurement of arterial pressure in humans under orthostatic stress. Clin Sci (Lond). 1999; 97(3): 291–301.
14. Koba S, Xing J, Sinoway LI, Li J. Sympathetic nerve responses to muscle contraction and stretch in ischemic heart failure. Am J Physiol Heart Circ Physiol. 2008; 294(1): H311–21.
15. Macefield VG. Physiological characteristics of low-threshold mechanoreceptors in joints, muscle and skin in human subjects. Clin Exp Pharmacol Physiol. 2005; 32: 135–44.
16. McDaniel J, Ives SJ, Richardson RS. Human muscle length-dependent changes in blood flow. J Appl Physiol (1985). 2012; 112(4): 560–5.
17. Middlekauff HR, Chiu J, Hamilton MA, et al. Muscle mechanoreceptor sensitivity in heart failure. Am J Physiol Heart Circ Physiol. 2004; 287(5): H1937–43.
18. Middlekauff HR, Nitzsche EU, Hoh CK, et al. Exaggerated renal vasoconstriction during exercise in heart failure patients. Circulation. 2000; 101(7): 784–9.
19. Middlekauff HR, Nitzsche EU, Hoh CK, et al. Exaggerated muscle mechanoreflex control of reflex renal vasoconstriction in heart failure. J Appl Physiol (1985). 2001; 90(5): 1714–9.
20. Mukkamala R, Xu D. Continuous and less invasive central hemodynamic monitoring by blood pressure waveform analysis. Am J Physiol Heart Circ Physiol. 2010; 299(3): H584–99.
21. Nobrega AC, Williamson JW, Friedman DB, Araujo CG, Mitchell JH. Cardiovascular responses to active and passive cycling movements. Med Sci Sports Exerc. 1994; 26(6): 709–14.
22. Olson TP, Joyner MJ, Eisenach JH, Curry TB, Johnson BD. Influence of locomotor muscle afferent inhibition on the ventilatory response to exercise in heart failure. Exp Physiol. 2014; 99(2): 414–26.
23. Padilla J, Young CN, Simmons GH, et al. Increased muscle sympathetic nerve activity acutely alters conduit artery shear rate patterns. Am J Physiol Heart Circ Physiol. 2010; 298(4): H1128–35.
24. Piepoli MF, Kaczmarek A, Francis DP, et al. Reduced peripheral skeletal muscle mass and abnormal reflex physiology in chronic heart failure. Circulation. 2006; 114(2): 126–34.
25. Ponikowski PP, Chua TP, Francis DP, Capucci A, Coats AJ, Piepoli MF. Muscle ergoreceptor overactivity reflects deterioration in clinical status and cardiorespiratory reflex control in chronic heart failure. Circulation. 2001; 104(19): 2324–30.
26. Poole DC, Hirai DM, Copp SW, Musch TI. Muscle oxygen transport and utilization in heart failure: implications for exercise (in)tolerance. Am J Physiol Heart Circ Physiol. 2012; 302(5): H1050–63.
27. Reisner AT, Xu D, Ryan KL, Convertino VA, Rickards CA, Mukkamala R. Monitoring non-invasive cardiac output and stroke volume during experimental human hypovolaemia and resuscitation. Br J Anaesth. 2011; 106(1): 23–30.
28. Scott AC, Francis DP, Davies LC, Ponikowski P, Coats AJ, Piepoli MF. Contribution of skeletal muscle ‘ergoreceptors’ in the human leg to respiratory control in chronic heart failure. J Physiol. 2000; 529(3): 863–70.
29. Scott AC, Wensel R, Davos CH, et al. Skeletal muscle reflex in heart failure patients: role of hydrogen. Circulation. 2003; 107(2): 300–6.
30. Smith SA, Mammen PP, Mitchell JH, Garry MG. Role of the exercise pressor reflex in rats with dilated cardiomyopathy. Circulation. 2003; 108(9): 1126–32.
31. Smith SA, Mitchell JH, Naseem RH, Garry MG. Mechanoreflex mediates the exaggerated exercise pressor reflex in heart failure. Circulation. 2005; 112(15): 2293–300.
32. Stebbins CL, Brown B, Levin D, Longhurst JC. Reflex effect of skeletal muscle mechanoreceptor stimulation on the cardiovascular system. J Appl Physiol (1985). 1988; 65: 1539–47.
33. Sterns DA, Ettinger SM, Gray KS, et al. Skeletal muscle metaboreceptor exercise responses are attenuated in heart failure. Circulation. 1991; 84(5): 2034–9.
34. Trinity JD, Amann M, McDaniel J, et al. Limb movement-induced hyperemia has a central hemodynamic component: evidence from a neural blockade study. Am J Physiol Heart Circ Physiol. 2010; 299(5): H1693–700.
35. Wang HJ, Li YL, Gao L, Zucker IH, Wang W. Alteration in skeletal muscle afferents in rats with chronic heart failure. J Physiol. 2010; 588(Pt 24): 5033–47.
36. Wesseling KH, Jansen JR, Settels JJ, Schreuder JJ. Computation of aortic flow from pressure in humans using a nonlinear, three-element model. J Appl Physiol (1985). 1993; 74(5): 2566–73.
37. Wilson JR, Martin JL, Schwartz D, Ferraro N. Exercise intolerance in patients with chronic heart failure: role of impaired nutritive flow to skeletal muscle. Circulation. 1984; 69(6): 1079–87.
38. Witman MA, McDaniel J, Fjeldstad AS, et al. A differing role of oxidative stress in the regulation of central and peripheral hemodynamics during exercise in heart failure. Am J Physiol Heart Circ Physiol. 2012; 303(10): H1237–44.
39. Witman MA, Ives SJ, Trinity JD, Groot HJ, Stehlik J, Richardson RS. Heart failure and movement-induced hemodynamics: partitioning the impact of central and peripheral dysfunction. Int J Cardiol. 2015; 178: 232–8.


© 2016 American College of Sports Medicine