INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

EXERCISE SERVES AS A POTENT stimulus for the activation of the autonomic nervous system. These autonomic responses lead to increases in cardiac output, ventilation, and vascular sympathetic tone. Two separate neural mechanisms are thought to be key players in eliciting these responses: 1) the central neural drive associated with the volitional component of exercise, termed as "central command" (5, 26), and 2) a reflex arising from activation of mechanically and chemically sensitive afferents in contracting muscles, termed "the exercise pressor reflex" (17). The afferent arm of this "pressor reflex" consists of thinly myelinated, slowly conducting group III and unmyelinated group IV fibers (1, 2, 14, 15, 23).

Previous studies, performed using a static contraction paradigm on barbiturate-anesthetized cats, set forth the notion that the majority of group III muscle afferents responded to mechanical deformation, whereas the majority of group IV afferents appeared to be chemoreceptors responding to the accumulation of the metabolic products of muscle contraction. Furthermore, group III mechanoreceptors appeared to show a substantial burst of activity at the onset of tetanic contraction followed by rapid adaptation. Group IV chemoreceptors, on the other hand, appeared to have a relatively longer latency of afferent activation from the onset of muscle contraction with greater responses when contracting muscles were made ischemic by vascular occlusion. The augmentation of this afferent response was further maintained during postcontraction vascular occlusion in which muscle relaxation eliminated mechanical stimulus to muscle afferents and circulatory arrest maintained the concentrations of metabolites in the vicinity of muscle afferent endings (10, 11, 16).

Such traditional thinking about the discharge properties of group III and IV muscle afferents on the exercise pressor reflex has come under question with recent works by Adreani and Kaufman, prompting novel conclusions about the discharge characteristics of group III and IV skeletal muscle afferents (1, 2). These authors conclude that mild rhythmic exercise is a sufficient stimulus to activate both group III and IV muscle afferents. In addition, these authors' findings demonstrate that group III and IV afferents tend to be more similar than different in their response to both freely perfused and ischemic exercise (2).

In a previous study from our lab, McClain et al. (13) suggested that forearm venous congestion (FVC) increased muscle sympathetic nerve activity (MSNA) responses to sustained static forearm exercise by sensitizing mechanically sensitive muscle afferents. Whereas direct FVC augmented reflex responses to static handgrip exercise, the MSNA responses during posthandgrip circulatory arrest (PHG-CA) remained unaffected, suggesting that FVC increased MSNA responses primarily by sensitizing mechanoreceptors.

In the present study, our interest has turned to the effect of FVC on MSNA during short, repetitive, nonfatiguing rhythmic contractions at 40% maximal voluntary contraction (MVC). We used relatively short bouts of work (2 s) with relatively longer rest periods (3 s). This work was continued for 5 min. We speculated that limb venous congestion would sensitize mechanoreceptor activity and in the process alter sympathetic nerve discharge characteristics. With the use of signal-averaging techniques (7), the data generated from the present study suggest that limb congestion enhances synchronization between contraction and sympathetic discharge.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

The following parameters were measured throughout the experiments: heart rate [HR; electrocardiogram (ECG)], mean arterial blood pressure (MAP; Finapres; Ohmeda, Madison, WI), respiration (pneumograph), and MSNA (peroneal nerve microneurography).

Protocols

Effects of FVC on nerve traffic during rhythmic forearm exercise. Subjects were placed supine, and ECG leads and a pneumobelt were placed to measure HR and respiration. MVC of the nondominant forearm was determined by using a handgrip dynamometer placed in the hand. A mercury-in-Silastic strain gauge was placed around the circumference of the midforearm, and an occlusion cuff was placed on the upper arm. A Finapres device was positioned on the middle forefinger of the opposite hand to measure blood pressure.

View larger version (20K): [in this window] [in a new window]   Fig. 1.   This schematic represents the experimental paradigm. The sequence of the 2 trials was varied. RHG, rhythmic handgrip; PHG-CA, posthandgrip circulatory arrest; Rec, recovery.

Microneurography

MSNA. This technique allows direct measurements of sympathetic nerve traffic directed to blood vessels in the skin or skeletal muscle (12, 25). This procedure, as used in our laboratory, has been previously described in detail (13, 21). Multiunit recordings of MSNA were obtained using a tungsten electrode placed in the right peroneal nerve just below the fibular head. A reference electrode was placed in the skin a few centimeters from the active electrode. The signal then was amplified, filtered, rectified, and integrated. On completion of the study, each neurogram was analyzed by manually counting the number of sympathetic bursts per minute as well as the amplitude of each burst (in mm) to calculate the total activity per minute. Each record was reviewed by one of the other investigators.

MSNA Signal-Averaging Analysis

MSNA signal preprocessing. For each subject, four 1-min segments of MSNA and coincident hand dynamometer signals were selected from paper recordings for analysis. These segments, corresponding to minutes 1 and 5 for control RHG and minutes 1 and 5 for RHG after venous congestion, were marked by hand to indicate time endpoints and MSNA signal amplitude limits. Digital images of the MSNA and dynamometer signals were obtained by optical scanning and stored. Numerical values for the MSNA and hand dynamometer signals along with time base information were obtained from these digital images and stored using Un-Scan-It (Silk Scientific, Orem, UT), a waveform tracing and digitizing program.

MSNA signal averaging. Each digitized 1-min data set comprised 11 or 12 5-s RHG cycles of MSNA and handgrip dynamometer data (see Fig. 2). These data yielded averaged MSNA and hand dynamometer signals spanning one RHG cycle. To process the data, we selected 1-min "raw" MSNA and handgrip dynamometer data sets. These sets were copied into a Microsoft Excel processing template developed for signal averaging. Two 300-sample signal-averaging buffers were initialized, one for MSNA data and one for handgrip dynamometer data. Each buffer held a 5-s sequence of data, the nominal length of the RHG cycle. The hand dynamometer data were quantized so that for each segment of the RHG cycle, the 2-s "grip" and 3-s "relaxation" signals were represented by sequences of ones and zeroes, respectively. These quantized data were used to identify the initial time point of each RHG cycle and its coincident MSNA data point. These 11 or 12 initial MSNA data points were averaged and saved in the MSNA-averaging buffer. Similarly, MSNA data points were averaged and saved for each remaining time point in the RHG cycle (3). In other words, 12 contraction cycles (each with five 1-s intervals) were summed over each of the 1-min periods of the study. The identical process was performed on the quantized dynamometer data to generate an averaged signal in the second buffer. Finally, the arbitrary direct-current background level in the averaged MSNA was removed by subtracting its minimum value from all values in the signal. This baseline correction not only simplifies the following numerical integration process, but, because MSNA cannot be less than zero, also makes intuitive physical sense as well.

MSNA signal integration. The linkage between muscle contraction and sympathetic activity was examined via numerical integration of each of the four previously averaged and baseline-adjusted MSNA signals. Specifically, a simple trapezoidal integration algorithm (24) computed an integral over the 5-s RHG interval for each of the four averaged MSNA waveforms selected for each subject (minute 1 with and without limb congestion and minute 5 with and without limb congestion). As MSNA was expressed in arbitrary units and its amplitude varied widely between subjects, each MSNA integral was then normalized to the range of 0.0-1.0 in which the value of 1.0 is reached at the end of the RHG interval (i.e., 100% of the MSNA activity within a 5-s interval). After this, the fraction of this normalized integrated MSNA (normalized integrated MSNA) was determined for each 1-s interval of the 5-s RHG cycle by subtracting the integral at the end of each interval from the previous interval (see Fig. 2).

Statistics. We used a two-way analysis of variance with two within-subject comparisons to compare mean HR, blood pressure, and MSNA during each minute of the paradigm before and after limb congestion. Post hoc analyses were performed using a Tukey's test. Tukey's test is a conservative fixed-range test, i.e., it uses a single value (w) for determining the significance of all differences in mean values. The signal-averaged data were analyzed in a similar fashion. A P value of 0.05 was considered statistically significant, and all data are presented as means ± SE.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Mean HR, MAP, and MSNA responses during forearm exercise. A summary of the mean data is presented in Table 1. MAP, HR, and MSNA (n = 10) all demonstrated an exercise main effect. There was no main effect of limb congestion and no interaction for these three indices.

Effect of rhythmic forearm exercise (40% MVC) and FVC (90 mmHg) on the signal-averaged MSNA response. In Table 2, the percentage of signal-averaged nerve activity during each second of the 5-s paradigm is shown. As mentioned, the highest percentage of integrated MSNA (n = 10) was seen during second 2 of the 5-s interval (Fig. 2 and Table 2).

View larger version (27K): [in this window] [in a new window]   Fig. 2.   This figure describes the data processing sequence to obtain delta-normalized integrals. MSNA, muscle sympathetic nerve activity; A-E, integrals.

View larger version (28K): [in this window] [in a new window]   Fig. 3.   This graph describes the percentage of activity that took place during second 2 during the 4 conditions: control during the 1st min of grip (G1), control during the 5th min of grip (G5), venous congestion during G1, and venous congestion during G5. The progression of exercise augmented the percentage of activity during second 2, whereas venous congestion increased it further.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In this report, we found that a rhythmic contraction paradigm increased the percentage of the MSNA signal seen during second 2 of the 5-s contraction/rest cycles. Moreover, this indicator of synchronization was increased by both contraction (i.e., % of second 2 in minute 5 was greater than in minute 1) and venous congestion. In the discussion to follow, we discuss the relationship of these data to previously published work as well as potential explanations for these observations.

Adreani et al. (1) and Adreani and Kaufman (2) have used a "walking cat" preparation to examine muscle afferent discharge during locomotion. The pattern of locomotion with this model closely simulates the pattern of the -motor neuron recruitment seen with normal gait in conscious animals (9). These authors demonstrated that within 2 s of initiating locomotion, thin fiber muscle afferent discharge increased. Of note, afferent activity was synchronous with contraction, with discharges occurring in less than 1 s after triceps surae muscle tension began to rise. In an earlier report, Kaufman et al. (10) demonstrated in a cat model that group III afferents increase their discharge ~220 ms after the initiation of static contraction.

In an analogous fashion, we speculate that ~200 ms after beginning the 2-s bout of contraction, thin fiber muscle afferents were engaged in the human forearm. Because of the slow conduction velocities of postganglionic sympathetic fascicles, we speculate that this led to an ~800-ms further delay between afferent activity and MSNA (19). This would lead to an increase in MSNA during second 2 of the paradigm.

The explanation for the exercise effect on the power of the MSNA signal (i.e., why second 2 in minute 5 was greater than second 2 in minute 1) is unclear. However, based on prior work, we believe this effect was due to a contraction-induced muscle afferent sensitizing process (6, 20). This finding would be consistent with prior human work from our laboratory (7) as well as with a large body of animal work supporting the presence of contraction and chemical-induced muscle afferent sensitization. It is unlikely that muscle ischemia per se contributed to the time-dependent effect of exercise, because the muscle was freely perfused and each bout of contraction was followed by 3 s of rest. Whether this postulated sensitizing effect is due to the effects of prostaglandins (6), lactate (20), or some other substance is not clear.

A prior report from this laboratory suggested that 5 min of FVC at 90 mmHg increased the mean MSNA response during 2 min of sustained static handgrip at 40% MVC (13). We proposed that this effect was due to the sensitization of mechanically sensitive afferents. The findings of the present report expand on this prior report by suggesting that limb congestion enhances the degree of entrainment of the MSNA signal, and this effect can be seen before limb congestion raises the MSNA.

Recently, we have shown 20-s periods of static quadriceps exercise leads to an increase in sympathetic nerve activity with an onset latency of ~4-6 s. From these data, we speculated that leg exercise engages mechanoreceptors with a latency between contraction and MSNA response of between 4 and 6 s. It is not entirely clear why in the present report we have sympathoexcitation with a shorter onset latency. One possibility is that both arm (present report) and leg exercise (prior report) engage the muscle reflex in a similar fashion, but the magnitude of arterial baroreflex buffering seen during the bouts of forearm exercise is less than seen during static quadriceps contractions. Changes in baroreflex buffering can alter the observed onset latency of MSNA responses to muscle contraction (18).

Alternate explanations and limitations. A second potential explanation for our findings is that limb congestion and exercise altered the central command response to exercise. From the work of Hill et al. (8), it is known 1) that the onset latency for sympathetic responses during electrical stimulation of mesencephalic locomotor region (MLR) is shorter than that seen for muscle reflex engagement, and 2) MLR stimulation evokes sympathetic activation that is easily buffered by the baroreflex (8). Moreover, as shown by Waldrop and Stremel (27), muscle contraction can excite neurons in regions of the subthalamus thought to be involved in central command. Therefore, it is possible that in the present report, central command caused the synchronization seen during minute 1 without limb congestion (7). The further increase in second-2 power could then have been due to modulation of the central command signal by afferent feedback from the skeletal muscle, a phenomenon consistent with the findings of Waldrop and Stremel (27).