SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. References
  7. Appendix

At the onset of both electrically evoked (STIM) and voluntary (VOL) isometric calf exercise there is an increase in vascular conductance of the contralateral lower limb, suggesting withdrawal of muscle sympathetic nerve activity (MSNA). Seven subjects performed STIM or VOL ischaemic calf exercise at 30% maximum voluntary contraction in a seated position. Blood pressure, heart rate and peroneal MSNA in the resting contralateral lower limb were recorded. During both STIM and VOL exercise blood pressure increased (P < 0.05). Blood flow increased by 40 ± 3 and 35 ± 3% and conductance increased by 37 ± 3 and 31 ± 4% (P < 0.05) after 10 s of STIM and VOL, respectively, and thereafter declined. The time course and direction of these changes persisted with subjects in a semisupine position, confirming that the transient conductance changes were not an artefact of the dependent leg position. Thigh cuff inflation for 1 min without exercise caused a 47 ± 7.5% (P < 0.05) reduction in MSNA, which recovered when the circulation was restored. However, when cuff inflation was followed by STIM or VOL exercise, MSNA did not fall further. These data suggest that the transient increase in vascular conductance at the onset of contralateral electrically evoked or voluntary lower limb exercise is unrelated to MSNA.

The mechanisms underpinning the increase in vascular conductance of the resting limb at the onset of contralateral isometric exercise in humans are unclear (Fisher & White, 2003). Saunders et al. (1989) reported vasodilatation in the contralateral resting forearm that was cholinergically mediated, since it was abolished by atropine administration. However, there is no histological or anatomical evidence for sympathetic cholinergic vasodilator fibres in human muscle (Bolme & Fuxe, 1970). Although an overflow of acetylcholine from the neuromuscular junction (or endothelial cells) might explain an atropine-sensitive vasodilatation in an active limb, it cannot explain vasodilatation in a quiescent limb (Joyner & Dietz, 2003). Alternatively, pharmacological manipulations have highlighted the role of locally produced nitric oxide and circulating catecholamines (Joyner & Halliwill, 2000; Joyner & Dietz, 2003). In addition, Cotzias & Marshall (1993) reported that the vasodilatation only occurs in the inactive limb when inadvertent muscle activity is detected by EMG, thus implicating local metabolites as important. However, our findings do not support this contention, since a transient increase in lower limb vascular conductance occurs in the absence of EMG activity at the onset of voluntary (VOL) and electrically evoked (STIM) contralateral isometric calf exercise and sham stimulation, but not prior to the expected exercise (Fisher & White, 2003). Circulating catecholamines and a flow-induced nitric oxide release were ruled out as mechanisms for this transient vasodilatation (Fisher & White, 2003). However, other pathways, e.g. arousal and the alerting response, could cause withdrawal of vasoconstrictor sympathetic nerve activity at the onset of exercise and subsequently increase vascular conductance in the resting limb (Halliwill et al. 1997; Marshall, 1997; Donadio et al. 2002).

The present study resolved whether muscle sympathetic nerve activity (MSNA) in the resting limb decreases at the onset of contralateral calf plantar flexor exercise. The study design had three main features. Firstly, VOL and STIM exercise were compared to investigate the contribution made by central command and muscle afferents. Since STIM exercise activates the muscle directly, central command is bypassed and the cardiovascular response to muscle afferent activation is isolated. Comparison of the responses to STIM exercise with those to VOL exercise reveals the contribution made by central command. Secondly, to ensure that any metabolites produced in the active leg did not enter the circulation and influence the vasculature in the contralateral leg, VOL and STIM exercise were performed ischaemically. Thirdly, supplementary VOL experiments were performed with subjects in the supine position to ensure that transient vasodilatation at the onset of exercise could not be attributed to an artefact of using plethysmography with the leg in the dependent position (Gamble et al. 1997; Fisher & White, 2003).

Methods

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. References
  7. Appendix

Healthy subjects (6 male, 1 female) were recruited, with a mean (±s.d.) age 24 ± 3.7 years, weight 80 ± 12.3 kg and height 181 ± 5.4 cm. Informed written consent was obtained from participants prior to experimentation as approved by the South Birmingham Research Ethics Committee in accordance with the Declaration of Helsinki (2000).

Measured variables

The force produced by the calf plantar flexors was measured according to Fisher & White (2003). Subjects were seated in a dynamometer with the lumbar spine against the back of the bench, the thigh of the preferred leg horizontal and the ankle at 85 deg (1.46 rad). A curved metal plate was clamped proximal to the knee joint, preventing heel lift during contraction. The upward force generated by the triceps surae was amplified and transmitted to an analog-to-digital converter (Cambridge Electronic Design 1401plus, CED, Cambridge, UK). The force produced was displayed on a chart recorder and recorded on Spike 2 computer software (Cambridge Electronic Design) at a frequency of 250 Hz. Heart rate (HR) was recorded using a three-lead ECG (Cardiorater CR7, Cardiac Records Ltd, London, UK), whilst blood pressure (BP) was monitored from the middle finger of the right hand using a Finapres system (Ohmeda 2300, Louisville, CO, USA). Whole-limb blood flow of the non-contracting, passive contralateral calf was measured by mercury in Silastic strain gauge plethysmography (Whitney, 1953; (AG101 Air Source, E20 Rapid cuff inflator, EC6 strain gauge plethysmograph, Hokanson, Bellevue, WA, USA). The strain gauge was positioned around the widest portion of the calf, and the leg supported in a dependent position at the foot and the knee to eliminate muscle tension. A cuff (CC17, Hokanson, Bellevue, WA, USA) was placed around the passive thigh and inflated to a pressure sufficient to prevent venous outflow. Venous occlusion pressure was calculated as 50 mmHg plus the sum of the distance from the centre of the thigh to the heart, multiplied by 0.779, accounting for the specific gravity of the blood (1.055; Gamble et al. 1997). The thigh cuff was rapidly inflated for 5 s every 10 s throughout the protocols. During the protocol, the blood flow was maintained to the foot (Siggaard-Anderson, 1970). Vascular conductance (in ml−1 min−1 (100 ml)−1 mmHg−1) was calculated by dividing blood flow by mean arterial blood pressure (MAP). In addition, respiration was measured using a Pneumotrace (model 1132, UFI, Moro Bay, CA, USA).

Post-ganglionic MSNA was measured from the non-active lower limb whilst the subject was seated (Vallbo et al. 1979; Wallin & Eckberg, 1982). Multi-unit recordings of MSNA were obtained with unipolar tungsten microelectrodes (Department of Clinical Neurophysiology, Karolinska Hospital, University of Stockholm, Sweden). The peroneal nerve was located using a probe, which discharged short-lasting (2 ms) stimulations (40–100 V) at 1 Hz (Grass-Telefactor, West Warick, RI, USA), to induce foot dorsiflexion and eversion. This was repeated using the recording needle at a much lower voltage (3 V) for precise location of the peroneal nerve. A reference needle electrode was also inserted near to the fibular head. The neural signals were amplified (95.5 × 103), filtered (bandwidth 700– 2000 Hz), rectified and integrated (time constant 0.1 s) to obtain a mean voltage neurogram (Nerve Traffic Analyser, model 662C-3, University of Iowa, Department of Bioengineering, Iowa City, IA, USA). A recording was considered acceptable when the neurogram revealed spontaneous pulse-synchronous bursts, with a minimum signal-to-noise ratio of 3:1 that increased during breath-holding manoeuvres. The absence of a response to arousal (loud noises) or skin stroking was used to discriminate between muscle and skin sympathetic nerve activity (Sundlof & Wallin, 1977). The recording was allowed to stabilize for at least 10 min before data were recorded for analysis. One researcher randomised the neurograms and another, who was blinded to the subject identity and experimental condition, performed scoring. MSNA was expressed as burst frequency (in bursts min−1), burst height and as total activity. Total activity was calculated as the product of average burst height and burst frequency, and is expressed in arbitrary units (a.u.).

Protocol

The protocol included a 2 min rest period followed by 2 min of VOL or STIM exercise at 30% maximum voluntary contraction (MVC; Fisher & White, 2003). Prior to exercise a cuff was inflated around the thigh to 200 mmHg, thus occluding the circulation to the leg during exercise; cuff inflation took 4–5 s and was completed 10 s before exercise. The protocol, during which HR and BP were measured, was conducted three times on each subject. In addition, on the first two sessions, blood flow was measured in the contralateral calf. On the third experimental session, MSNA was recorded from the same limb. During this session the MVC was only measured prior to MSNA recording. During the first two sessions the order of VOL and STIM was randomised and separated by a 20 min rest period. However, during the MSNA recording session STIM followed VOL.

The effects of thigh cuff inflation around the passive limb on MSNA were investigated in three subjects. After 2 min of baseline data collection a thigh cuff was inflated around the limb contralateral to that from which MSNA was being measured for 1 min and followed by a 2 min recovery period.

The effects of posture on the lower limb vascular responses to VOL were examined in one female and six male subjects (four of whom had completed the described protocol). Subjects were seated in a Biodex System 3 isokinetic dynamometer (Biodex Medical Systems, Shirley, NY, USA) in a semi-supine position with the right knee flexed at 150 deg and the right foot firmly attached to the footplate for measurement of the torque produced by the ankle plantar flexors. The left leg was supported at ∼10 cm below the heart and blood flow was measured from the calf every 15 s using strain gauge plethysmography. Following a 2 min rest period a cuff was inflated around the right thigh to 200 mmHg. Subjects then performed 2 min isometric exercise at 30% of maximum voluntary torque (MVT).

Data reduction and analysis

MAP, HR and MSNA are expressed as averages over 15 s. Blood flow was measured every 15 s and vascular conductance calculated by dividing this value by the average MAP at the time of measurement. Conductance values for each subject for the two experimental trials were averaged before calculation of a group mean. Values are presented as means ±s.e.m. Repeated measures analysis of variance (ANOVA), adjusted using the Greenhouse–Geisser correction, was conducted to investigate time and condition effects between rest and exercise phases (Ludbrook, 1994), with significance taken at the P < 0.05 level. Post hoc analysis was performed using two-tailed Student's paired t tests with Bonferroni adjustment for multiple comparisons.

Results

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. References
  7. Appendix

Contractile characteristics

The mean time to peak tension (TPT) for the triceps surae was 118 ± 3.2 ms and the half-time of relaxation (RT50) was 85 ± 5.1 ms. MVC was not significantly different between VOL and STIM conditions during the blood flow experiments (1092 ± 136 and 1116 ± 129 N, respectively) and prior to the MSNA recording protocols (1093 ± 98 N). In the supine position subjects elicited a MVT of 89.9 ± 6 N m.

No significant difference was found between resting values for MAP, HR, blood flow, vascular conductance and MSNA measured during the rest period of the VOL and STIM protocols (Table 1). However, in the 15 s prior to exercise, after the thigh cuff was inflated, total MSNA decreased in the seven subjects, compared to that observed over the previous 100 s (Fig. 1B). Furthermore, inflation of a thigh cuff around the resting limb for 1 min, in three subjects, reduced MSNA by 48 ± 7.5%. These findings meant that the changes in MSNA are taken from resting MSNA measured over the 15 s prior to exercise.

Table 1.  Cardiovascular parameters during STIM and VOL protocols at rest, 15 and 60 s of exercise
 STIMVOL
Rest15 s exercise60 s exerciseRest15 s exercise60 s exercise
  1. Note that conductance was measured on separate sessions from all other values in this table. MSNA values were recorded during rest with thigh cuff inflation around the contralateral limb. STIM, stimulated protocol; VOL, voluntary protocol; MAP, mean arterial blood pressure; HR, heart rate; a.u., arbitrary units. Values are means ±s.e.m.* Significant difference from rest (P < 0.05).

MAP (mmHg)90.1 ± 2.7 89.1 ± 2.5 96.9 ± 2.5*83.5 ± 3.7 89.1 ± 4.3  97.1 ± 4.3*
HR (beats min−1)66.8 ± 3.8 69.9 ± 3.5 67.3 ± 5.0 68.8 ± 3.0775.1 ± 3.4  70.2 ± 3.7 
Blood flow (ml−1 (100 ml)−1 min−1)1.9 ± 0.2 3.3 ± 0.5*2.1 ± 0.32.4 ± 0.33.8 ± 0.6*2.4 ± 0.4
Conductance (ml−1 (100 ml)−1 min−1 mmHg−1)0.016 ± 0.002 0.027 ± 0.004*0.017 ± 0.0030.022 ± 0.0030.033 ± 0.006*0.019 ± 0.003
Burst frequency (bursts min−1)18.9 ± 1.9 26.9 ± 2.4 23.4 ± 3.4 22.9 ± 3.1 19.4 ± 3.4  22.3 ± 4.6 
Mean burst height (V)0.24 ± 0.020.26 ± 0.03 0.3 ± 0.040.28 ± 0.040.27 ± 0.04 0.27 ± 0.05
Total activity (a.u.)4.3 ± 0.46.9 ± 0.87.3 ± 1.86.0 ± 0.95.5 ± 1.7 6.0 ± 1.6
image

Figure 1. Original records and summary of MSNA data A, an original MSNA record showing transition to voluntary (VOL) and evoked (STIM) contraction. Bold arrow indicates exercise onset. B, total MSNA at rest, during cuff inflation prior to contraction and during exercise at 15 s (Ex 15), 30 s (Ex 30), 45 s (Ex 45) and 60 s (Ex 60), for the VOL and STIM conditions. * Significant difference from rest (P < 0.05).

Download figure to PowerPoint

No significant difference was found between resting values for the cardiovascular parameters recorded during the MSNA experiments (Table 1) and during the conductance measurement sessions (VOL, 82.1 ± 2.8 mmHg and 67.2 ± 3.2 beats min−1; STIM, 88.6 ± 3.0 mmHg and 67.1 ± 3.0 beats min−1, for MAP and HR, respectively). At rest in the semi-supine condition MAP was 83.7 ± 3.2 mmHg and HR was 52.7 ± 2.6 beats min−1.

Onset of exercise

Over the first 15 s of exercise, HR increased by (3.1 ± 1.6 and 6.3 ± 1.8 beats min−1 for STIM and VOL, respectively) whilst in the semi-supine position it increased by 8.3 ± 2.8 beats min−1 (P < 0.05). In all protocols, exercise MAP increased (P < 0.05; Table 1) for upright exercise and there was a comparable 7.6 ± 1.6 mmHg increase during VOL in the semi-supine position. Blood flow increased by 1.37 ± 0.29 and 1.39 ± 0.29 ml−1 (100 ml−1) min−1 in STIM and VOL, respectively (P < 0.05). Conductance increased significantly after 10 s of STIM and VOL exercise by 0.012 ± 0.002 (37 ± 3%) and 0.011 ± 0.003 ml−1 (100 ml−1) min−1 mmHg−1 (31 ± 4%), respectively, declining thereafter. When VOL exercise was performed with subjects in a semi-supine position, there was also an increase in vascular conductance by 0.016 ± 0.006 ml−1 (100 ml−1) min−1 mmHg−1 (26 ± 10%).

The MAP and HR responses derived from the MSNA measurement trial were similar to those found during baseline sessions when conductance was calculated. During the baseline trial, at 15 s of exercise HR increased by 4.6 ± 2.8 and 10.9 ± 2.7 beats min−1 in STIM and VOL, respectively. During the 60 s of exercise MAP increased by 5.8 ± 2.1 and 13.2 ± 1.3 mmHg for STIM and VOL, respectively. At the onset of both STIM and VOL exercise MSNA was unchanged from resting values with the thigh cuff inflated (Fig. 1A and Table 1). However, during STIM the change in MSNA exceeded that recorded during VOL.

Discussion

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. References
  7. Appendix

The transient increase in conductance in the resting limb during contralateral voluntary and electrically evoked human calf muscle exercise was confirmed in the present study. This conductance change is independent of muscle force and metabolite generation (Fisher & White, 2003). The stability of MSNA at the onset of both STIM and VOL exercise suggests that the increase in conductance at this time is unrelated to withdrawal of sympathetic drive to the human calf muscles.

Thigh cuff inflation

Thigh cuff inflation prior to exercise reduced MSNA, as confirmed and extended in a small group of subjects in whom the effects of 1 min of thigh cuff inflation were studied (decreased by ∼50%). The explanation for this baseline shift may be that the mechanical forces created by rapid thigh cuff inflation caused an increase in blood pressure through an increase in total peripheral resistance. The increase in blood pressure could stimulate the baroreceptors and subsequently inhibit MSNA. Mark et al. (1985) and McClain et al. (1994) did not find a change in MSNA when an occlusion cuff was inflated around the upper arm. When compared to the effect of occluding the smaller vascular bed of the upper limb, occlusion of the leg vasculature must have a relatively larger effect on total peripheral resistance. This would be expected to create a greater error signal (increase in blood pressure) to activate the baroreceptors and withdraw MSNA. Indeed, in the present study, when blood flow measurement was made (and conductance calculated) immediately before exercise, whilst the thigh cuff was inflated around the contralateral limb, systolic blood pressure was 4.2 and 3.4 mmHg higher than mean resting levels for VOL and STIM, respectively. Finally, in the study of Mark et al. (1985), it is unlikely that arm cuff inflation performed on supine subjects with the arm supported above the level of the heart would cause a marked further increase in the already augmented venous return.

It is unlikely that a thigh cuff-mediated withdrawal of MSNA is responsible for the observed transient vasodilatation at the onset of exercise. Previously, we have shown that thigh cuff inflation per se had no significant effect on leg vascular conductance (Fisher & White, 2003). In the present study, leg vascular conductance tended to decrease during thigh cuff inflation prior to exercise onset in both STIM and VOL. This occurred at a time when MSNA had fallen (a disassociation between MSNA and vascular conductance has been shown by Shoemaker et al. 2000). Furthermore, the transient increase in vascular conductance observed at the onset of exercise is also present during isometric exercise performed without thigh cuff inflation (Fisher & White, 2003).

Donadio et al. (2002) demonstrated that MSNA is attenuated by sensory stimuli via an arousal effect. However, this effect was only evident on the burst immediately following arousal. Consequently, arousal is unlikely to account for the fall in MSNA for the duration of thigh cuff inflation during this study.

Onset of exercise

Vascular conductance increased in the contralateral lower limb at the onset of exercise with no difference between VOL and STIM. Since MSNA remained unchanged during VOL and reached a slightly higher level during STIM, MSNA withdrawal cannot be responsible for the transient increase in vascular conductance at the onset of both forms of exercise. This conclusion is supported by the findings of Herr et al. (1999), who reported an increase in total MSNA with a latency of only 4–6 s during quadriceps contractions.

Owing to the rapidity of the observed vasodilatation in the resting lower limb during contralateral isometric exercise, flow-mediated nitric oxide release may be involved, and the small blood flow increases we measured at the onset of exercise would support this. The time course of the transient vasodilatation may be too fast for circulating catecholamines to be responsible (Kozlowski et al. 1973).

Effects of posture

Resting lower limb blood flow assessed using the plethysmography is lower when the leg is in the dependent position (Jorfeldt et al. 2003). This is arguably due to incomplete venous drainage leading to a decrease in the venous capacity of the limb because of venous pooling. The lower venous drainage observed with the leg in the dependent position could mean that at the onset of exercise an increase in cardiac output would clear the venous pooling, leading to an increase in venous capacity. Therefore, with plethysmography, an apparent increase in lower limb blood flow would be seen. Although a possibility, this is unlikely to be an explanation for the findings of the present study. In the supine position, there is still a transient vasodilatation observed at the onset of isometric exercise of the contralateral lower limb. In the supine position venous drainage would be enhanced and pooling would be minimized.

Limitations

A possible limitation of the present study is that, during the MSNA recording session, we were unable to randomise the trial order, with VOL preceding STIM. This was because of the increased likelihood of losing a recording site whilst preparing to carry out STIM. Additionally, plethysmography is a reliable technique, but we have reached the limit of its time resolution. Furthermore, the plethysmography technique cannot be used at the same time as MSNA measurement. However, we do not believe that combination of these factors materially alters the value of our data.

Inadvertent muscle activity in the resting limb as detected by EMG has been associated with vasodilatation during isometric exercise in the contralateral limb (Cotzias & Marshall, 1993). Subjects were instructed to avoid inadvertent muscle contractions and in our previous study we detected no EMG activity in the resting calf during isometric exercise of the other calf (Fisher & White, 2003).

We conclude that decreased MSNA is not responsible for the transient increase in vascular conductance observed in the contralateral lower limb at the onset of voluntary and stimulated isometric calf muscle exercise.

References

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. References
  7. Appendix
  • Bolme P & Fuxe K (1970). Adrenergic and cholinergic nerve terminals in skeletal muscle vessels. Acta Physiol Scand 78, 5259.
  • Cotzias C & Marshall JM (1993). Vascular and electromyographic responses evoked in forearm muscle by isometric contraction of the contralateral forearm. Clin Auton Res 3, 2130.
  • Donadio V, Kallio M, Karlsson T, Nordin M & Wallin BG (2002). Inhibition of human muscle sympathetic activity by sensory stimulation. J Physiol 544, 285292.DOI: 10.1113/jphysiol.2002.019596
  • Fisher JP & White MJ (2003). The time course and direction of lower limb vascular conductance changes during voluntary and electrically evoked isometric exercise of the contralateral calf muscle in man. J Physiol 546, 315323.DOI: 10.1113/jphysiol.2002.031476
  • Gamble J, Christ F & Gartside IB (1997). The effect of passive tilting on microvascular parameters in the human calf: a strain gauge plethysmography study. J Physiol 498, 541552.
  • Halliwill JR, Lawer LA, Eickhoff TJ, Dietz NM, Nauss LA & Joyner MJ (1997). Forearm sympathetic withdrawal and vasodilatation during mental stress in humans. J Physiol 504, 211220.DOI: 10.1111/j.1469-7793.1997.211bf.x
  • Herr MD, Imadojemu V, Kunselman AR & Sinoway LI (1999). Characteristics of the muscle mechanoreflex during quadriceps contractions in humans. J Appl Physiol 86, 767772.
  • Jorfeldt L, Vedung T, Forsstrom E & Henriksson J (2003). Influence of leg position and environmental temperature on segmental volume expansion during venous occlusion plethysmography. Clin Sci 104, 599605.DOI: 10.1042/CS20020257
  • Joyner MJ & Dietz NM (2003). Sympathetic vasodilation in human muscle. Acta Physiol Scand 177, 329336.DOI: 10.1046/j.1365-201X.2003.01090.x
  • Joyner MJ & Halliwill JR (2000). Sympathetic vasodilatation in human limbs. J Physiol 526, 471480.
  • Kozlowski S, Brzeinska Z, Nazar K, Kowalski W & Franczyk M (1973). Plasma catecholamines during sustained isometric exercise. Clin Sci Mol Med 45, 723731.
  • Ludbrook J (1994). Repeated measurements and multiple comparisons in cardiovascular research. Cardiovasc Res 28, 303311.
  • McClain J, Hardy C & Sinoway L (1994). Forearm compression during exercise increases sympathetic nerve traffic. J Appl Physiol 77, 26122617.
  • Mark AL, Victor RG, Nerhed C & Wallin GB (1985). Microneurographic studies of the mechanisms of sympathetic nerve responses to static exercise in humans. Circ Res 57, 461469.
  • Marshall JM (1997). Cardiovascular changes associated with behavioural learning. In Cardiovascular Regulation (Studies in Physiology), 1st edn, ed. JordanD & MarshallJM, chap. 3, pp. 3759. Portland Press Ltd, London .
  • Saunders JS, Mark AL & Ferguson DW (1989). Evidence for cholinergically mediated vasodilation at the beginning of isometric exercise in humans. Circ 79, 815824.
  • Shoemaker JK, Herr MD & Sinoway LI (2000). Dissociation of muscle sympathetic nerve activity and leg vascular resistance in humans. Am J Physiol 279, H1215H1219.
  • Siggaard-Anderson J (1970). Venous occlusion plethysmography on the calf. Evaluation of diagnosis and results in vascular surgery. Dan Med Bull 17, 166.
  • Sundlof G & Wallin BG (1977). The variability of muscle nerve sympathetic activity in resting recumbent man. J Physiol 272, 383397.
  • Vallbo AB, Hagbarth KE, Torebjork HE & Wallin BG (1979). Somatosensory, proprioceptive, and human sympathetic activity in human peripheral nerves. Physiol Rev 59, 919957.
  • Wallin BG & Eckberg DL (1982). Sympathetic transients caused by abrupt alterations of carotid baroreceptor activity in humans. Am J Physiol 242, H185H190.
  • Whitney RJ (1953). The measurement of volume changes in human limbs. J Physiol 121, 127.

Appendix

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. References
  7. Appendix

Acknowledgements

Dr Fisher is supported by British Heart Foundation grant PG/03/148/16352. Dr Sander is supported by the Danish Heart Foundation, The Danish Medical Research Council and the Michaelsen Foundation. The authors thank Liz Simpson for her excellent technical assistance with the MSNA measurements.