Skip Navigation

Cardiovascular Research 1998 40(3):591-599; doi:10.1016/S0008-6363(98)00190-4
© 1998 by European Society of Cardiology
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow E-letters: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Kamath, M. V.
Right arrow Articles by Tougas, G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kamath, M. V.
Right arrow Articles by Tougas, G.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Copyright © 1998, European Society of Cardiology

Neurocardiac and cerebral responses evoked by esophageal vago-afferent stimulation in humans: effect of varying intensities

Markad V. Kamatha, Stephan Hollerbachb, Absar Bajwaa, Ernest L. Fallena,*, Adrian R.M. Uptona and Gervais Tougasa

aDepartment of Medicine, McMaster University, 1200 Main St. West, Room 3U4, Hamilton, Ontario, Canada L8N 3Z5
bDepartment of Internal Medicine (SH), University of Regensburg, D-93042, Regensburg, Germany

* Corresponding author. Tel.: +1-(905)-521-2100, Ext. 6222; fax: +1-(905)-521-5053; e-mail: fallene@fhs.csu.mcmaster.ca

Received 2 December 1997; accepted 24 April 1998


   Abstract
Top
 Abstract
1 Introduction
 2 Methods
 3 Results
 4 Discussion
 References
 
Objective: This study was designed to determine whether esophageal vago-afferent electrostimulation, over a wide range of stimulus intensities, can sustain a cardiac vago-efferent effect by way of central nervous system processing. Methods: Studies were performed in ten healthy male subjects (23.9±6.3 years). Esophageal electrostimulation was carried out using a stimulating electrode placed in the distal esophagus. Stimulation of esophageal vago-afferent fibres was employed using electrical impulses (200 µs at 0.2 Hz x 128 s) varying from 2.7 to 20 mA. Respiratory frequencies, beat-to-beat heart rate autospectra and cerebral evoked potentials were recorded at baseline and at each stimulus intensity in random order. Results: With esophageal electrical stimulation, we observed a small non-significant decrease in heart rate. There was a dramatic shift of the instantaneous heart rate power spectra towards enhanced cardiac vagal modulation with intensities as low as 5 mA. This effect was sustained throughout all intensities with no further change in either the low frequency or high frequency power. Conversely, there was a linear dose response relationship between cerebral evoked potential amplitude and stimulus intensity mainly occurring above perception threshold (10 mA). Esophageal stimulation had no significant effect on heart rate or respiratory frequency at any stimulus intensity. Conclusions: These results indicate that electrical stimulation of the distal esophagus across a wide range of current intensities elicits a reproducible shift in the heart rate power spectrum towards enhanced vagal modulation. The data suggest a closed loop afferent/efferent circuitry wherein tonic visceral afferent impulses appear to elicit a phasic or modulatory vago-efferent cardiac response in healthy subjects.

KEYWORDS Autonomic nervous system; Heart rate variability; Sinus node; ECG; Signal transduction; Discipline; Clinical


   1 Introduction
Top
 Abstract
 1 Introduction
2 Methods
 3 Results
 4 Discussion
 References
 
The power spectrum of heart rate variability (HRV) provides a convenient window through which the efferent pathways governing autonomic regulation of sino-atrial function can be studied [1–3]. The high frequency (HF) component, ranging from 0.15 to 0.35 Hz, corresponds to respiratory sinus arrhythmia and thus constitutes a modulatory signal of respiratory driven cardiac vago-efferent drive [4, 5]. The low frequency (LF) peak (0.05–0.15 Hz) is sensitive to manoeuvres that augment sympathetic efferent output to the sinus node [6]. Its physiologic correlate under quiescent conditions remains poorly understood but is influenced by the existing balance between central sympathetic processing and baroreceptor activity [7]. Conceptually, one can model central processing of afferent signals with a systems approach involving both linear and nonlinear dynamics [8]. Baroreceptor, respiratory and visceral neural afferent impulses represent some of those tonic signals which impinge on central oscillators resulting in modulatory or phasic output responses characterized by instantaneous beat-to-beat changes in heart rate. In the dog, transfer function models of sinus node activity have been developed using electrical stimulation of the efferent branch of the cardiac vagus nerve [9], while models of autonomic transfer function in humans have employed random respiratory signals as the input and HRV as the output signal [10]. Direct access to afferent input pathways is difficult in humans. Investigators therefore have had to rely on manoeuvres such as orthostasis [11], mental stress [12]and selective autonomic blocking drugs [5, 13]as interventions for examining the frequency composition of successive interbeat signals.

Recently we have introduced a novel approach for examining both visceral sensory afferent and efferent autonomic neural signals in humans [14]. The human esophagus is richly innervated with afferent vagal fibres. By introducing into the distal esophagus a catheter to which stainless steel electrodes have been attached, one can direct pulsed electrical stimuli along the vago-afferent neural pathways [15]. We have shown that both electrical and mechanical stimulation of the esophagus not only activates vagal afferent pathways, but also reproducibly increases the high frequency (vagal) component of the heart rate autospectrum with a corresponding decrease in the LF:HF ratio [14, 16]. While esophageal stimulation had no effects on respiratory frequency, the cardioautonomic effects were observed over a wide range of esophageal stimulation frequencies, ranging from 0.1 to 1 Hz with a maximal effect at 0.2 Hz [16]. In previous studies, we determined that reproducible cerebral evoked responses can only be obtained using esophageal stimuli of sufficient intensity as to be consciously perceived by the subject, suggesting that the afferent neural response is intensity dependent [17]. However, the effects of varying stimulation intensity on vago-efferent modulation of sinus node activity has not been examined, nor has it been determined if the cardioautonomic response to esophageal stimulation necessarily requires the involvement of cortical or sub-cortical neural pathways. In the present study we examined the autospectra of the HRV signal and cerebral evoked potentials in response to esophageal electrical stimulation using current intensities ranging from low stimuli; 2.7 mA and 5 mA (well below the threshold of sensory perception) to current intensities up to 20 mA.


   2 Methods
Top
 Abstract
 1 Introduction
 2 Methods
3 Results
 4 Discussion
 References
 
2.1 Subjects
Ten healthy male volunteers (23.9±6.3 years), free of medication and with no respiratory, gastrointestinal, cardiac or neurological abnormalities, were enrolled following written informed consent. This study was reviewed and approved by the Committee on Ethics for Research, McMaster University Medical Centre.

2.2 Esophageal stimulation
A stainless steel electrode was fixed to the distal end of a standard polyvinyl esophageal manometric catheter using 4-0 silk suture. After application of a water soluble lubricant, the tube was inserted nasally and the electrode positioned in the distal esophagus, 35 cm from the nostrils. The stainless steel electrode was connected to a stimulator by a shielded cable attached to the manometric assembly. The positive reference electrode was placed on the abdominal wall 7 cm below the xiphoid process, over the linea alba. Esophageal stimulation was delivered as a series of 25 repetitive single electrical stimuli using a 200-µs square wave pulse for a total stimulation period of 128 s. All stimuli were delivered at a frequency of 0.2 Hz which was shown to be most effective in a previous study [16]. Perception thresholds were individually determined for each subject. The stimulation intensity level was varied from 2.7 to a maximum of 20 mA. A detailed description of the technique has been published elsewhere [14, 15].

2.3 Respiratory recordings
Respiratory activity was recorded using a piezo-electric belt, (RESP-EZ, EPM Canada, Kanata, Ontario, Canada), attached to the abdomen of the subject. The signal was amplified through a Grass EEG amplifier Model 8-20D (Astromed, Grass Instruments Division, W. Warwick, USA). This produced a clearly visible respiratory signal which could be quantified with respect to respiratory rate and the frequency autospectrum. Changes in ventilatory volume could be estimated by visual control of the individual respiratory signal during rest and electrical esophageal stimulation. The respiratory signal was then fed into a personal computer using a CODAS data acquisition software (Dataq, Akron, OH, USA). From the continuous respiratory waveforms a power spectrum of the respiratory frequency was generated using the Blackman–Tukey method [18]. No attempt was made to regulate either the breathing frequency or the tidal volume.

2.4 Heart rate variability and power spectral analysis
Three ECG electrodes were attached to the chest and the lead II ECG signal was fed through a Hewlett-Packard 7087C amplifier. These signals were digitized using a 12-bit analog-to-digital converter at a sampling frequency of 1 kHz and processed on a PC (Gateway 486/33). A QRS detection algorithm was implemented in the software to locate a stable and noise independent fiducial point on the R wave. An R–R interval series was then generated from the continuous ECG data. A beat-to-beat heart rate variability signal was computed, and then resampled at 2 Hz using linear interpolation to obtain an equally sampled time series. A record length of 256 points from the resampled signal (128 s) was selected for power spectral analysis. No ectopic beats were observed in any of the subjects during electrical stimulation, as well as during resting conditions. The mean value of the signal was removed and the equally sampled HRV signals were fed through a second order high pass Butterworth filter with a cut-off of 0.02 Hz. A 9th order autoregressive model was then applied to the demeaned, filtered heart rate variability data [19]. The information contained within the power spectrum was analyzed in the following manner. The maximum peak power of the LF and HF bands was identified and expressed as [(beats/min)2/Hz]. The frequencies at which these peaks occurred (central frequencies) were obtained and the area subtended by each spectral band was then computed by numerically integrating the power therein contained. This band area was expressed in absolute units as (beats/min)2. In addition, the normalized areas within both LF and HF bands were derived by dividing the integrative power within each band by the total power contained in the entire spectrum. There was no need to subtract the DC component as this was already removed by filtering. The LF:HF ratio was computed as the ratio of these normalized areas.

2.5 Cerebral evoked potential recordings
To ascertain the CNS responses to afferent nerve signals activated by esophageal stimulation, the cerebral potentials evoked by esophageal electrostimulation were simultaneously measured using 21 gold-plated EEG recording scalp electrodes positioned according to the International 10–20 method of electrode placement [20]. This approach, which has been validated in clinical as well as in experimental neurology, is based on a fixed distance between the recording electrodes as measured from the occiput, the mastoid, and the base of nose. Reference and ground electrodes were placed on both mastoids and the forehead (Fpz). For recording and analysis of data, the Bio-Logic brain atlas (BA) recording system and the BA analysis software program (model 173, version 2.311) were used (Bio-Logic Systems, Mundelein, IL, USA). The amplifier gain was set to x 10 000 and the recording sensitivity was 20 µV. The bandpass filter settings were 0.1 and 100 Hz. The recording epoch was set to 512 ms to incorporate all reproducible peaks of the evoked potential response to esophageal stimulation, as previously reported [21]. Electrode impedance was maintained below 4 k{Omega}. Electrical activity was averaged on-line with automatic control of stimulus presentation and artifact rejection. No series had less than 5% artifact rejection, and series with more than 20% artifacts were rejected and then repeated. Cerebral evoked potential amplitudes were measured from the Cz electrode which is centrally positioned on the scalp. This electrode provides the largest amplitudes of evoked potential recordings as shown in previous studies [15, 23].

2.6 Experimental protocol
All studies were performed between 0900 and 1100 h, after a 12-h fast. Recordings were obtained in a quiet room with the lights dimmed. The fasting supine subject remained awake for the entire study with eyes open and fixed on a stable target to avoid visual stimuli artifacts during esophageal stimulation. No masking noise was used. Fifteen minutes following insertion of the esophageal probe, baseline ECG and respiratory waveforms were continuously recorded for a period of 10 min while the subject rested comfortably. Esophageal stimulation (0.2 Hz) was applied with successive stimulation intensities of 2.7, 5, 10, 15 or 20 mA. The different current intensities were selected according to an individually randomized order generated a priori for each subject. A rest period of 5 min was inserted between stimulation periods. The subject was blinded to the order at which different intensities were used. Continuous ECG, and respiratory waveforms were recorded during all stimulation and rest periods. The cerebral evoked responses were recorded separately for each stimulation period. Each stimulation intensity was applied for two separate periods, and each period separately recorded.

2.7 Statistical analysis
A one-way anova was used to compare the effect of each electrical stimulation intensity with baseline values for the following power spectral parameters; LF and HF absolute power; the normalized LF and HF components expressed as percent of total power; LF:HF ratio: LF and HF central frequencies as well as heart rate and respiratory frequency. A 95% confidence interval with p<0.05 was considered statistically significant. Linear regression analysis was used to test the correlation between the HF central frequency and the respiratory rate at each stimulation intensity. Cerebral evoked potentials (amplitudes) were obtained from the Cz electrode, as previously reported [15]. All data are reported as means±S.E.M., unless stated otherwise. Peak-to-peak amplitudes were measured from the maximal positive to maximal negative potential deflection. Evoked potential data are reported as absolute peak amplitudes for each stimulation intensity. For statistical analysis, calculations were done using univariate and multivariate analysis as appropriate (ANOVA, MANOVA). p values <0.05 were considered to indicate a statistically significant difference between each intensity and baseline values. Correlation analysis was done by using the Pearson correlation test.


   3 Results
Top
 Abstract
 1 Introduction
 2 Methods
 3 Results
4 Discussion
 References
 
With electrical stimulation of the esophagus, a small but consistent decrease in heart rate (3–6 beats/min) was observed at all electrical stimulation intensities, but these changes were not significantly different from baseline values (Table 1). However, there was still a trend towards a decrease in heart rate with esophageal stimulation. On average, the subjects first perceived the electrical stimuli at a current intensity of 8.87 mA (range 5–10 mA). The stimulus was felt as a light, sharp, non-painful sensation behind the xiphoid process. Increasing the stimulus intensity resulted in a more pronounced sensation which never became painful within the range of intensities used. There were no arrhythmias observed or recorded during electrical stimulation of the esophagus at any of the current intensities employed.


View this table:
[in this window]
[in a new window]

  		Table 1 Cardiac autospectral responses to esophageal electrostimulation of varying current amplitudes

 
3.1 Cerebral evoked potentials
There were no detectable cerebral potentials obtained during sham stimulation as shown in Fig. 1a, whereas electrical stimulation at or above the threshold for perception of electrical stimuli (8–10 mA) always resulted in reproducible cerebral responses, especially at higher (15–20 mA) stimulus intensities (Fig. 1b). As demonstrated in Fig. 1b, the cerebral evoked responses to esophageal electrical stimulation had their maximum amplitudes in the central region of the scalp near the location of the Cz electrode. These evoked potentials consisted of three negative peaks (N1, N2, N3) with three intercalated positive potential deflections (P1, P2, P3). The respective amplitudes for each of these peaks are to be found in Table 2.


Figure 1
View larger version (19K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1 Cortical evoked potential responses under conditions of sham stimulation (panel a) and esophageal electrical stimulation at 20 mA (panel b). All recordings were obtained from the Cz scalp electrode, as outlined in Section 2. Panel b represents an averaged response to 25 repetitive stimuli. Two averaged responses are shown in (b) to illustrate reproducibility. N1, N2, N3 are the three negative peaks with intercalated positive peaks at P1, P2 and P3. An average of two repeated trials in the same subjects is shown here (n=1).

 

View this table:
[in this window]
[in a new window]

  		Table 2 Dose–response relationship of cerebral evoked potentials in response to esophageal electrical stimulation at different current intensities

 
A dose–response increase in the amplitudes of the evoked potential response was observed for the first two peaks N2,P2 (Table 2, Fig. 2). We could not detect any cerebral evoked potential response until the intensity (>5 mA) of the esophageal stimulus was perceived (Fig. 2). This intensity threshold was between 5 and 10 mA for all subjects. The linear relationship between stimulus intensity and the amplitude of the first two peaks shows that a four-fold increase in stimulus intensity is associated with a three-fold augmentation in the amplitude of the cerebral evoked potential response (Table 2).


Figure 2
View larger version (22K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2 Effect of stimulus intensity on both cerebral evoked peak to peak amplitude (N2/P2 peak) and the LF:HF area ratio of the HR autospectrum. Note the linear increase in N2/P2 amplitude with increasing intensities compared to the plateau effect of the LF:HF ratio beyond 5 mA. Values are mean±S.E.M. Values of p are compared to values at baseline (0 mA).

 
3.2 Heart rate autospectrum analysis
Electrical stimulation of the distal esophagus elicited a significant shift in the baseline HR autospectrum. There was a significant shift in the LF area, HF area, LF:HF area ratio, and percentage of both LF and HF normalized areas towards a dominant HF component with esophageal stimulation ranging from 5 to 20 mA (Table 1). These changes in HR autospectra were present even with the smallest intensity of esophageal stimulation (2.7 mA), well below the intensity associated with conscious perception of the stimulus. This cardiac autonomic effect of esophageal stimulation reached its maximum between stimulation intensities of 10 to 20 mA with a 29 to 41% decrease in the LF:HF area (Table 1, Fig. 2). These changes in neurocardiac modulation occurred as soon as electrical stimulation was set on, were present throughout the stimulation period, and returned to baseline values as soon as the esophageal stimulation ceased. The effects of electrical esophageal stimulation were observed in all subjects, although at considerable variability (Table 1). Changes in the normalized area under the curve for both HF and LF peaks are demonstrated in Fig. 3. Despite a four-fold increase in current intensities, there were no further linear changes in either the power of the LF and HF components or the LF:HF area ratio (Fig. 2). This is in contrast to the dose response relationship observed with the evoked potentials (N2/P2) with increasing current intensities (Fig. 2).


Figure 3
View larger version (18K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3 Percentage change in the normalized area under the curve of LF and HF components in response to increasing to stimulus intensity. Values are expressed as mean±S.E.M. HF=high frequency; LF=low frequency; *p<0.05 compared to baseline (no stimulation).

 
Despite reversal of the LF:HF ratio, there was no shift in the central frequency of either the LF or HF band with increasing current intensities (Table 1). Similarly, the spontaneous breathing rate was unaltered from baseline values throughout all stimulation intensities. A representative example of the effect of different stimulation intensities on the heart rate autospectrum is illustrated in Fig. 4. A high correlation (R2=0.821) was observed between the HF central frequency and the breathing rate regardless of stimulation intensity (Fig. 5). A comparison of the autospectral values at baseline and post-electrical stimulation showed that all power spectral values and heart rate returned to their baseline state within 1 to 2 min.


Figure 4
View larger version (18K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4 Representative example of the changes in the HR autospectrum at different stimulus intensities. Note the early reversal of the LF:HF ratio at low stimulus intensities. The scale is identical for each power density spectrum.

 

Figure 5
View larger version (23K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 5 Linear regression of respiratory frequency (as measured by the piezo-electric belt) against the central frequency of the HF peak derived from autospectral analysis; p=<0.001.

 

   4 Discussion
Top
 Abstract
 1 Introduction
 2 Methods
 3 Results
 4 Discussion
References
 
This study shows that electrical stimulation of the distal esophagus enhances cardiac vagal modulation, as determined by a small yet reproducible drop in heart rate and a significant shift in the heart rate power spectrum. This effect is seen across current intensities ranging from those that are not consciously perceived to those above perception threshold. It supports and further extends our previous observation wherein electrical esophageal stimulation augments cardioautonomic modulation across a broad range of stimulation frequencies, independent of respiratory driven vago-efferent fluctuations [16]. The time course and the stationarity of the heart rate variability signal have been reported in recent studies [14]and show that the effects of mechanical and electrical stimulation of esophageal afferent neural pathways exhibit a mostly transient phenomenon which is gradually reversed after terminating the stimulation event.

Cerebral evoked potential responses can only be observed in response to esophageal stimulation if the stimuli are above the perception threshold, suggesting that these evoked responses do not simply reflect viscerosensory function from brainstem nuclei, but also require the involvement of higher brain centers (i.e., thalamic and cortical neural projections) associated with the perception of visceral sensations. Within the range of stimulation intensities used in this study there was a linear dose–response relationship between the intensity of esophageal electrostimulation at stimulus intensities above the level at which they could be actually perceived (5–10 mA), and the amplitude of the cerebral evoked potential. In contrast, the significant reversal of baseline cardiac sympathovagal balance was clearly visible even at sub-threshold intensities and reached its maximum at stimulation intensities between 5 and 10 mA. This suggests that the cardiac autonomic response to increasing intensities of visceral vago-afferent stimulation, at least at low current intensities, is largely independent of cortical neural projections, whereas activation of cortical brain centers during perceived stimuli appears to somewhat augment the sino-atrial cardiac response. Rather, there appears to be a non-linear systems effect, namely a modulatory or phasic response to repetitive tonic visceral afferent impulses involving integrative central yet subcortical processing. By directly accessing and stimulating esophageal vago-afferent fibres and measuring both cerebral evoked potentials and heart rate variability one is able to describe some of the properties of a closed loop afferent/efferent feedback control system.

The evidence for the feedback loop observed is several-fold. First, the effect on heart rate is negligible in comparison to the effect of direct stimulation of vago-efferent fibres to the heart [22]. More persuasive is the evidence obtained from cerebral evoked potentials which represent the brain's electrical response to activation of afferent sensory pathways. The triphasic configuration of the evoked potential and the peak-to-peak amplitudes are similar to those previously reported by ourselves and by others using esophageal stimulation [23, 24]as well as those reported using direct vagal electrical stimulation in patients treated for intractable partial seizures [25]. These characteristics suggest that vago-afferent sensory inputs from the esophagus are processed within different brain centers ranging from autonomic brainstem nuclei over synaptic hypothalamic and thalamic projections to cortical association fields. However, the precise central projections of the afferent pathways involved with esophageal stimulation remain incompletely understood in humans. Using positron emission tomography and O15-labeled water, we found, in humans, that esophageal electrical stimulation was associated with increased blood flow and metabolic activity in the anterior thalamic nuclei as well as in the non-dominant (right) parietal (Brodman regions 39 and 40) and temporal (region 31) sensory cortex [40]. Using topographic brain mapping techniques applied to evoked potential responses, we found highest electrical activity in deep midbrain structures such as the thalamus and highest evoked potential amplitudes in different cortical association fields [17]. The early cerebral peaks (N1, P1, N2) are said to involve deep central structures including the nucleus tractus solitarius as well as thalamic sites whereas the later peaks (P2, P3) appear to be localized bilaterally in cortical areas of the parietocentral and temporal associative cortex [31]. More recent studies by Aziz et al. [26]suggest that mechanical stimulation of the esophagus is associated with increased activity in the right anterior insular cortex and cingulate gyrus using positron emission tomography and O15-labeled water, which is in accordance with our findings [40].

As to the efferent loop of the closed-loop feedback system, different models exist to study the effects of visceral stimulation on cardiovascular reflex responses. From animal studies there is some divergence of opinion as to whether electrical stimulation of visceral structures leads to a vagal or sympathetic efferent response [27, 28]. Gieroba et al. recently reported a sympathoexcitatory response to stimulation of abdominal vagal afferents in the anaesthetized rabbit [29]. They postulated that input from abdominal vago-afferent fibres were projected to bulbospinal barosensitive neurons in the rostral central lateral medulla. However, we have previously shown in the rat that gastric distention is associated with decreased heart rate, a response that can be prevented either by prior bilateral subdiaphragmatic vagotomy (showing the involvement of vagal afferent pathways), or by prior administration of atropine indicating cholinergic/vagal efferent modulation of the response [30].

Of particular interest was the observation that increases in stimulus intensity were associated with a proportional augmentation in amplitude of the evoked potential response once the intensity of the esophageal stimulus was above the perception threshold. Conversely, the slowing effect on baseline heart rate was small but consistently observed at all electrical stimulation intensities, and the power spectral response was different, with the effect on the LF:HF area ratio first occurring with esophageal stimuli below conscious perception levels. Further increases in current intensity caused an augmentation of the effect on the LF:HF ratio until a plateau between stimulation intensities above 10 mA was reached. Since the effects of esophageal stimulation on cerebral evoked responses are not observed until the stimulus is perceived while its effects on heart rate variability occur with sub-threshold stimuli, it suggests that the modulation of instantaneous heart rate by esophageal stimulation occurs as a result of subcortical or even subthalamic processing of the afferent input rather than through a response primarily involving cortical pathways. The increased amplitudes of the cerebral evoked potentials with increasing stimulus intensity, once the perception threshold is reached, likely reflect the fact that more afferent fibres are recruited, and/or that a greater cortical surface area is activated.

The effects of electrical esophageal stimulation on the power spectrum parameters of heart rate variability raise the question as to which neurons are involved in this modulation of cardioautonomic tone. Our results show that there was no significant change in respiratory sinus arrhythmia associated with electrical stimulation. There was no alteration in the spontaneous breathing rates, which averaged 0.24±0.2 Hz, with increasing esophageal stimulation intensities. This indicates that there was little or no entrainment of spontaneous breathing frequencies by esophageal stimulation. In our previous study on the effect of varying stimulation frequencies, the maximum effect on HF power occurred at 0.22 Hz, a frequency that was almost identical to the spontaneous breathing rate of the subjects in that study [16]. However, when the power spectral HRV response was separately examined, the augmented vagal modulatory effect on sino-atrial function persisted across a broad range of stimulation frequencies in the absence of any change in spontaneous breathing frequencies [16]. It is conceivable that inhibitory inputs from both respiratory and esophageal afferents share similar pathways but to what extent esophageal vago-afferent impulses facilitate, enhance or entrain normal respiratory driven cardiac vagal output could not be determined by this study. We did not observe significant changes in spontaneous ventilatory volume during electrical esophageal stimulation, as assessed using visual control of respiratory signals derived from the piezoceramic belt device. Small fluctuations in respiratory volume, however, cannot be entirely excluded in our study.

The small but consistent effects of esophageal stimulation on heart rate in combination with significant changes observed in the power spectrum parameters of heart rate variability could represent an effect of esophageal electrical stimulation through afferent pathways affecting cardiac vagal motor neurons located in the brainstem. Whether this effect consists of direct trans-synaptic activation of vagal motor neurons in the brainstem via excitatory nerve fibers, or is indirectly mediated by subsequent modulation of the activity of other neural pathways, such as those associated with baroreceptor activities, is not known and remains to be addressed in future studies.

4.1 Study limitations
Electrical stimulation was applied at only one site in the distal esophagus precluding the possibility of an accurate estimation of the conduction velocity of the afferent impulse and of the type of afferent fibres involved. However, using identical esophageal stimulation parameters and a comparison of the latencies of the cerebral evoked responses obtained at different esophageal locations, we have previously measured the conduction velocities of the afferent input to range from 7.5 to 10 m/s, in keeping with involvement of myelinated A{delta} fibres rather than the slower conduction velocities (<2.5 m/s) associated with unmyelinated C-fibres [32, 33]. We have yet to determine whether other types of afferent nerve fibres with different conduction velocities can be recruited when different stimulation intensities are applied.

Power spectral analysis of HRV requires steady state conditions [34]. Although the duration of stimulation (128 s) was within the acceptable interval for estimation of autospectral data [34], it is doubtful that absolute steady state conditions prevailed during the period of electrical stimulation. However, there was remarkable reproducibility with repetitive stimuli at each intensity which indicates stationarity of the considered heart rate variability signal.

To describe the distribution of power in spectral components of the heart rate variability signal (PSHRV), we adopted the autoregressive method and Levinson–Durbin recursion for its stable estimation of the power spectra rather than the maximum entropy method using Burg's algorithm to calculate the spectral density estimator. The latter produces accurate autoregressive estimates primarily for data which are truly autoregressive. However, for signals which may contain sinusoids, Burg's algorithm is subject to line splitting. Another advantage of using the Levinson–Durbin recursion to calculate the autoregressive parameters is its minimum phase property and the fact that all poles of the filter lie within the unit cycle. Since heart rate variability data may contain significant sinusoidal components, we believe that the autoregressive method employed in this paper is reliable and provides robust estimators of the power spectra.

4.2 Clinical relevance
Sino-atrial node activity, a manifestation of cardiac autonomic function, is exquisitely sensitive to shifts in centrally mediated sympathovagal balance. There is increasing evidence that restraint or prevention of sympathetic dominance within this modulatory balance is a protective mechanism in a variety of cardiac disorders [22]. Not only does the LF:HF ratio confer predictive power for outcome events especially in patients recovering from acute myocardial infarction [35, 36]but there is some suggestion that interventions which shift the balance to vagal dominance, vis-a-vis beta blockers [37], exercise training [38]and transdermal scopolamine [39], may be clinically beneficial. Accessing and controlling electrical or mechanical stimulation of visceral vago-afferent fibres may ultimately prove to be a feasible alternative to current methods of enhancing vagal tone. From a clinical physiologic viewpoint, this technique offers a unique opportunity to study the afferent–efferent loop of autonomic control through systems analysis and transfer function methodologies. Further understanding of this very complex circuitry should shed light on our understanding of the central processing of autonomic control mechanisms in both health and disease.

Time for primary review 43 days.


   Acknowledgements
 
We thank Glenn Shine and Debbie Fitzpatrick for their expert technical assistance. Our group acknowledges funding from the following sources: the Heart and Stroke Foundation of Ontario (M.K. and E.F.); the Medical Research Council of Canada (G.T.); the Novartis Pharma in Nuremberg, Germany, by supporting Stephan Hollerbach; the Natural Sciences and Engineering Research Council of Canada (M.K.); and the DeGroote Foundation (A.U.).


   References
Top
 Abstract
 1 Introduction
 2 Methods
 3 Results
 4 Discussion
 References
 

  1. Malliani A., Pagani M., Lombardi F., Cerutti S. Cardiovascular neural regulation explored in frequency domain. Circulation (1991) 84:482–492.[Abstract/Free Full Text]
  2. Akselrod S., Gordon D., Ubel F.A., Shannon D.C., Berger A.C., Cohen R.J. Power spectrum analysis of heart rate fluctuation: a quantitative probe of beat-to-beat cardiovascular control. Science (1981) 213:220–222.[Abstract/Free Full Text]
  3. Kamath M.V., Fallen E.L. Power spectral analysis of heart rate variability: a noninvasive signature of cardiac autonomic function. Crit Rev Biomed Eng (1993) 21:245–311.[Web of Science][Medline]
  4. Hyndman B.W., Kitney R.I., Sayers B.M. Spontaneous rhythms in physiological control systems. Nature (1971) 233:339–341.[CrossRef][Medline]
  5. Pomeranz B., Macaulay R.B., Caudill M.A., et al. Assessment of autonomic function in humans by heart rate spectral analysis. Am J Physiol (1985) 248:H151–H153.[Web of Science][Medline]
  6. Pagani M., Lombardi F., Guzzetti S., et al. Power spectral analysis of heart rate and arterial pressure variabilities as a marker of sympathovagal interaction in man and conscious dog. Circ Res (1986) 59:178–193.[Abstract/Free Full Text]
  7. Bernardi L., Lauzzi S., Radelli A., et al. Low frequency spontaneous fluctuations of R–R interval and blood pressure in conscious humans: a baroreceptor or central phenomenon? Clin Sci (1994) 87:649–654.[Web of Science][Medline]
  8. DeBoer R.W., Karemaker J.M., Strackee J. Relationships between short-term blood pressure fluctuations and heart rate variability in resting subjects. II: A simple model. Med Biol Eng Comput (1985) 23:359–364.[CrossRef][Web of Science][Medline]
  9. Berger R.D., Saul J.P., Cohen R.J. Transfer function analysis of autonomic regulationI. Canine Atrial rate response. Am J Physiol (1989) 256:H142–H152.[Web of Science][Medline]
  10. Saul J.P., Berger R.D., Chen M.H., Cohen R.J. Transfer function analysis of autonomic regulation II Respiratory sinus arrhythmia. Am J Physiol (1989) 256:H153–H161.[Web of Science][Medline]
  11. Lindquist A., Jalonen J., Parvianen P., Antila K., Laitinen A.L. Effect of posture on spontaneous and thermally stimulated cardiovascular oscillations. Cardiovasc Res (1990) 24:373–380.[Abstract/Free Full Text]
  12. Pagani M., Furlan R., Pizzinelli P., Crivellaro W., Cerutti S., Malliani A. Spectral analysis of R–R and arterial pressure variabilities to assess sympatho-vagal intervention during mental stress in humans. J Hypertens (1989) 7(Suppl_6):S14–S15.[Web of Science]
  13. Cook J.R., Bigger J.T., Kleiger R.E., et al. Effect of atenolol and diltiazem on heart rate variability in normal persons. J Am Coll Cardiol (1991) 17:480–484.[Abstract]
  14. Tougas G., Kamath M.V., Watteel G., et al. Modulation of neurocardiac function by esophageal stimulation in humans. Clin Sci (1997) 92:167–174.[Web of Science][Medline]
  15. Tougas G., Hudoba P., Fitzpatrick D., Hunt R.H., Upton A.R.M. Cerebral evoked potential responses following direct vagal and esophageal electrical stimulation in humans. Am J Physiol (1993) 264:G486–491.[Web of Science][Medline]
  16. Bajwa A., Hollerbach S., Kamath M.V., et al. Neurocardiac response to esophageal electric stimulation in humans: effect of varying stimulation frequencies. Am J Physiol (1997) 272:R896–R901.[Web of Science][Medline]
  17. Hollerbach S., Kamath M.V., Chen Y., et al. The magnitude of the central response to esophageal electrical stimulation is intensity dependent. Gastroenterology (1997) 112:1137–1146.[CrossRef][Medline]
  18. Kay SM. Modern spectral estimation. Prentice Hall, Englewood Cliffs, NJ, 1988:77.
  19. Kay S., Marple S.L. Spectrum analysis. A modern perspective. Proc IEEE (1981) 69:1380–1419.[CrossRef]
  20. Jasper H.H. Report of the committee on methods of clinical examination in electroencephalography. Electroencephal Clin Neurophysiol (1958) 10:370–375.[CrossRef]
  21. Tougas G., Fitzpatrick D., Hunt R.H., Upton A.R.M. Evidence of impaired afferent vagal junction in patients with diabetes gastroparesis. Pacing Clin Neurophysiol (1992) 15:1597–1602.[CrossRef]
  22. Schwartz PJ, Cerati D. Cardiac vagal efferent nerve activity. In: Levy MN, Schwartz PJ, Armonk NY, editors. Vagal control of the heart: experimental basis and clinical implications. Futura, 1994:77–87.
  23. Frieling T., Enck P., Wienbeck M. Cerebral responses evoked by electrical stimulation of the esophagus in normal subjects. Gastroenterology (1989) 97:475–478.[Web of Science][Medline]
  24. DeVault K., Beacham S., Streletz L., Castell D.O. Cerebral evoked potentials — a method of quantitation of central nervous system response to esophageal pain. Dig Dis Sci (1993) 38:2241–2246.[CrossRef][Web of Science][Medline]
  25. Tougas G., Fitzpatrick D., Hudoba P., et al. Effects of chronic left vagal stimulation on visceral vagal function in man. Pacing Clin Electrophysiol (1997) 15:235–243.[CrossRef]
  26. Aziz Q., Anderson J., Valind S., et al. Identification of cortical loci processing nociceptive and non-nociceptive human esophageal sensation using positron emission tomography (PET) (Abstract). Gastroenterology (1996) 110(4):A627.
  27. Grind D., Davison J.S. Cardiovascular changes elicited by vagal gastric afferents in the rat. QJ Exp Physiol (1981) 66:307–310.
  28. Longhurst J.C., Spilker H.L., Ordway G.A. Cardiovascular reflexes elicited by passive gastric distention in rats. Am J Physiol (1981) 240:H539–H545.[Web of Science][Medline]
  29. Gieroba Z.J., Messenger J.P., Blessing W.W. Abdominal vagal stimulation excites bulbospinal barosensitive neurons in the rostral ventrolateral medulla. Neuroscience (1995) 65:355–364.[CrossRef][Web of Science][Medline]
  30. Tougas G., Wang L. The cardiac effects of gastric distension in rats are mediated via vagal afferents and atropine sensitive pathways (Abstract). Clin Inv Med Suppl (1995) 4:B56.
  31. Franssen H., Weusten B.L.A.M., Wieneke G.H., Smout AJPM. Source modelling of esophageal evoked potentials. Electroencephalogr Clin Neurophysiol Evoked Potentials (1996) 100(2):85–89.[CrossRef]
  32. Tougas G., Fitzpatrick D., Upton A.R.M., Hunt R.H. The cortical evoked responses produced by balloon distension and electrical stimulation of the esophagus involve different afferent vagal fibres (Abstract). Gastroenterology (1993) 104:A592.
  33. Hollerbach S., Tougas G., Frieling T., et al. Cerebral evoked responses to gastrointestinal stimulation in humans. Crit Rev Biomed Eng (1997) 25:203–242.[Web of Science][Medline]
  34. Heart rate variability. Standard of measurement, physiological interpretation and clinical use. Task Force of the European Society of Cardiology and NASPE. Circulation 1996;93:1043–1065.
  35. Bigger J.T. Jr., Albrecht P., Steinman R.C., et al. Comparison of time and frequency domain based measures of cardiac parasympathetic activity in Holter recordings after myocardial infarction. Am J Cardiol (1989) 64:536–538.[CrossRef][Web of Science][Medline]
  36. Farrell T.G., Bashir Y., Cripps T., et al. Risk stratification for arrhythmic events in post infarction patients based on heart rate variability, ambulatory electrocardiographic variables and signal averaged electrocardiogram. J Am Coll Cardiol (1991) 18:687–697.[Abstract]
  37. Sandrone G., Mortara A., Torzilla, et al. Effects of beta blockers (atenolol or metoprolol) on heart rate variability after acute myocardial infarction. Am J Cardiol (1996) 74:340–345.
  38. Piepoli M., Clark A.L., Volterrani M., et al. Contribution of muscle afferents to the hemodynamic autonomic and ventilatory responses to exercise in patients with chronic heart failure. Effects of physical training. Circulation (1996) 93(5):940–952.[Abstract/Free Full Text]
  39. Venkatesh R., Fallen E.L., Kamath M.V., Connolly S., Yusuf S. Double blind placebo controlled trial of short-term transdermal scopolamine in heart rate variability in patients with congestive heart failure. Heart (1996) 76:136–143.[Free Full Text]
  40. Tougas G., Kamath M.V., Garnett S., et al. Mapping of cerebral response to vagal and esophageal stimulation using positron emission tomography (PET) and topographic EEG in humans. Gastroenterology (1994) 106:A846.

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 J. Appl. Physiol.Home page
D. S. Ditor, M. V. Kamath, M. J. MacDonald, J. Bugaresti, N. McCartney, and A. L. Hicks
Effects of body weight-supported treadmill training on heart rate variability and blood pressure variability in individuals with spinal cord injury
J Appl Physiol, April 1, 2005; 98(4): 1519 - 1525.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow E-letters: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Kamath, M. V.
Right arrow Articles by Tougas, G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kamath, M. V.
Right arrow Articles by Tougas, G.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?