This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Bernardi, L. Articles by Sleight, P. Search for Related Content PubMed PubMed Citation Articles by Bernardi, L. Articles by Sleight, P.

(Circulation. 1995;92:2895-2903.)
© 1995 American Heart Association, Inc.

Articles

Demonstrable Cardiac Reinnervation After Human Heart Transplantation by Carotid Baroreflex Modulation of RR Interval

Presented in part at the 67th Scientific Sessions of the American Heart Association, Dallas, Tex, November 14-17, 1994, and the 55th Congress of the Italian Society of Cardiology, Rome, Italy, December 13-16, 1994, and published in abstract form (Circulation. 1994;90[pt 2]:I-540).

Luciano Bernardi, MD; Beatrice Bianchini, MD; Giammario Spadacini, MD; Stefano Leuzzi, MD; Felice Valle, MD; Eugenia Marchesi, MD; Claudio Passino, MD; Alessandro Calciati, MD; Mario Viganó, MD; Mauro Rinaldi, MD; Luigi Martinelli, MD; Giorgio Finardi, MD; Peter Sleight, MD, FRCP

From the Dipartimento di Medicina Interna and Cattedra di Cardiochirurgia (M.V., M.R., L.M.), IRCCS S. Matteo and University of Pavia, Italy, and the Department of Cardiovascular Medicine (P.S.), John Radcliffe Hospital, Oxford, England.

Correspondence to Dr Luciano Bernardi, Clinica Medica 1, Universitá di Pavia, 27100 Pavia, Italy.

	Abstract

Top Abstract Introduction Methods Results Discussion References

 
Background After heart transplantation, respiration-synchronous fluctuations (0.18 to 0.35 Hz, high frequency [HF]) in RR interval may result from atrial stretch caused by changes in venous return, but slower fluctuations (0.03 to 0.15 Hz, low frequency [LF]) not due to respiration suggest reinnervation. In normal subjects, sinusoidal neck suction selectively stimulates carotid baroreceptors and causes reflex oscillations of RR interval.

Methods and Results To evaluate the presence of reinnervation, we measured the power of RR-LF and RR-HF in 26 heart transplant recipients and 16 control subjects before and during sinusoidal neck suction at 0.1 Hz and 0.20 Hz (similar to but distinct from that of controlled respiration, 0.25 Hz) and before and during administration of atropine or ß-blocker (esmolol hydrochloride) by spectral analysis. All transplant recipients showed small respiratory HF fluctuations. Nonrespiratory LF fluctuations were present in 13 of 26 transplant recipients and increased with months since transplantation (r=.53, P<.01). HF neck suction induced a 0.20-Hz component in all 16 control subjects and none of the 26 transplant subjects. LF neck suction increased RR-LF (from 0.73±0.20 to 1.30±0.26 ln ms², P<.001), similar to but less than in control subjects (from 6.12±0.21 to 8.27±0.21 ln ms², P<.001). Atropine reduced all fluctuations in control subjects and blocked the HF increase caused by 0.20-Hz neck suction but not the LF increase during 0.10-Hz stimulation. Neck suction–induced changes in LF fluctuations persisted after administration of atropine in transplant recipients but were attenuated by esmolol hydrochloride, suggesting sympathetic rather than vagal reinnervation.

Conclusions The presence of baroreceptor-induced RR oscillations is evidence of functional, although incomplete, autonomic reinnervation.

Key Words: heart rate • nervous system • autonomic • transplantation • baroreceptors • reflex

	Introduction

Top Abstract Introduction Methods Results Discussion References

 
In the immediate posttransplant period, the heart is thought to be denervated. During the second year after transplantation, several reports found no evidence of reinnervation^{1 2 3 4 5} or only partial sympathetic reinnervation, the latter based on measurement of plasma catecholamines,^{6 7} occurrence of ischemic chest pain,⁸ or inducible vasovagal reactions.^{9 10} These clinical findings are not very useful as markers of reinnervation in the transplanted heart because they are present in only a minority of transplant patients.

In the normal nontransplanted subject, spontaneous changes in heart rate (RR interval) are due to autonomic activity (sinus arrhythmia); hence, the presence of such arrhythmia in the transplanted donor heart might (mistakenly) be regarded as a simple method of identifying reinnervation. However, we have previously shown that changes in the heart period synchronous with respiration are common in the human transplanted heart, independent of autonomic tone, and are most likely due to stretch of the donor atrium caused by the inspiratory increase in venous return.^{11 12 13} But 20 months after transplantation, nonrespiratory RR interval fluctuations of LF (0.03 to 0.15 Hz) have been described,¹⁴ further suggesting the possibility of reinnervation, although changes could not be induced by maneuvers known to modify autonomic nervous discharge reflexly, such as passive tilting.

These results call for a "positive" proof of reinnervation. It is well known that a large proportion of sinus changes in the RR interval in conscious humans are related to the activity of arterial baroreceptors, and this baroreceptor input influences variability in both the HF and LF range.^{15 16} Arterial baroreceptors can be selectively stimulated by pressure changes in an applied neck chamber, initially without any change in hemodynamics or respiration,¹⁷ and thus can cause reflex RR interval changes (either rhythmic or impulsive, depending on the type of stimulation^{18 19 20} ). In the present study, we have used the neck chamber to test for reflex heart rate responses in the transplanted heart. We stimulated the carotid baroreceptors at both LF (0.10 Hz) and HF (0.21 Hz, close to but different from the respiratory frequency of 0.25 Hz). These tests were repeated after administration of atropine and, in a selected group of subjects, after a short-acting ß-blocker was given to block vagal or sympathetic efferents, if any. Of course, neck suction is still able to modulate sympathetic tone to noncardiac areas; hence, blood pressure modulation is largely intact after transplantation.

	Methods

Top Abstract Introduction Methods Results Discussion References

 
Subjects
We studied 26 orthotopic heart transplant recipients and 16 normal control subjects, matched for donor age (because reflex control of heart rate declines with age). Age of donors, control subjects, and transplant recipients was 36±3, 33±2, and 51±2 years, respectively (mean±SEM); time since transplantation was 21±4 months (mean±SEM, range 2 to 63 months). Twenty-three patients were treated with cyclosporine (282±25 mg) and 3 with Tacrolimus (4.3±3 mg), 24 with oral steroids (8.8±0.6 mg), and 24 with azathioprine (61±6 mg). Nine patients were taking ACE inhibitors and 8 were taking calcium antagonists; 10 patients were taking furosemide. No transplant recipients had signs or symptoms of active cardiorespiratory disease, other than controlled hypertension (16 cases). Endomyocardial biopsy did not show any evidence of tissue rejection at the time of the study. The protocol was approved by the local Institutional Review Board for Human Experimentation, and all subjects gave informed consent.

Study Protocol
After 20 to 30 minutes of supine rest and familiarization with the laboratory, we recorded the ECG (lead II), respiration, noninvasive blood pressure (Finapres model 2300, Ohmeda), and neck chamber pressure. The respiratory signal was obtained from ECG electrodes by means of an electrical impedance pneumograph designed in our laboratory (with a flat frequency response from 0 to 25 Hz).¹² Noninvasive recordings of blood pressure were also obtained by means of a conventional sphygmomanometer. Neck suction was applied to a flexible, molded lead collar connected to a vacuum cleaner whose power was modulated by a function generator (Krohn-Hite model 5200) through a phase-control power unit. By selecting the appropriate signal amplitude and frequency on the function generator, we could obtain sinusoidal suction with the desired characteristics. The neck chamber pressure was continuously monitored by use of a Statham P23d pressure transducer and set to oscillate sinusoidally from 0 to -30 mm Hg at either 0.1 or 0.2 Hz in each subject. Respiration was controlled at 0.25 Hz (15 breaths per minute). The latter was done to obtain a stimulation similar in frequency to that of respiration but without its hemodynamic effects (respiratory increases in venous return and hence in blood pressure). LF (0.1 Hz) sinusoidal changes can be followed normally by both sympathetic and vagal efferent reflex responses, whereas the faster 0.2-Hz stimulation can only be tracked by the vagus.²¹ After a 4-minute recording during controlled respiration, without neck suction, LF and HF stimulations were performed for periods of 4 minutes each or until 350 heart beats were recorded.

Atropine Study
The significance of LF fluctuations in the RR interval is not fully understood. Although it is clear that LF fluctuations are a "marker" of sympathetic activity, at least in relative terms, it is also probable that power in the LF range is influenced by parasympathetic activity, since its absolute power decreases after atropine^{22 23} and during maneuvers that cause sympathetic activation and decreased vagal tone, such as tilting^{14 24} and physical exercise.^{12 25} Furthermore, the effects of neck suction on RR interval fluctuation might be due only to the vagus, or to the vagus in addition to modulation of sympathetic activity. As a consequence, only after atropine administration could any LF fluctuations observed be considered as evidence of sympathetic activity. We then repeated the recordings after IV injection of 0.04 mg/kg atropine in 4 control subjects and 7 heart-transplant subjects.

Esmolol Hydrochloride Study
Four transplant recipients who showed good evidence of reinnervation on the above protocol consented to be restudied before and after IV injection of a short-acting ß-blocker (esmolol hydrochloride, Gensia Europe Ltd). After precautionary placement of a transvenous pacemaker wire (pacing was not needed), the protocol was repeated to reconfirm that the presence of LF oscillations seen in RR interval were due to sympathetic reinnervation. Esmolol hydrochloride was injected through a previously placed venous cannula by use of a loading dose of 500 µg, followed by a slow infusion over the next 20 minutes (total mean dose 1.8 mg/kg) to lower the resting heart rate by 15 to 20 beats per minute. It was not thought to be justifiable to attempt complete ß-blockade with higher doses. ß-Blockade caused no disability or symptoms. We used only 0.1-Hz neck suction; it was considered unnecessary to retest at 0.2 Hz because we had found no evidence of vagal activity on the prior test. Because the results were consistent, it was not considered justifiable to carry out ß-blockade in the remaining subjects.

Data Acquisition and Analysis
Data were digitized on-line by a 12-bit analog-to-digital converter (NB-MIO-16 board, National Instruments) at a sampling rate of 500 samples per second for each channel. The converter was connected to a Macintosh II computer (Apple Inc). A "C" language program identified all the QRS complexes in each sequence and then located the peak of each R wave. RR interval, SBP and DBP, neck suction, and respiration time series were obtained from these data. The respiratory time series was obtained from the respiration signal sampled at the peak of each R wave and was expressed in arbitrary values. For each step of the protocol, 250 to 350 RR intervals were analyzed. The very few premature beats were interactively identified and corrected by linear interpolation with the previous and following beats.

Power Spectrum Analysis
We applied power spectrum analysis to RR interval, respiratory, SBP and DBP, and neck suction signals using an autoregressive model. Unlike other methods of computing the power spectrum (for example, the fast Fourier transform), the autoregressive method has the advantage of giving reliable estimates of the power associated with the peaks at various frequencies with use of a relatively smaller amount of data. In addition, it is able to provide a better identification of the frequency of each significant peak.^{12 14} This analysis is only valid during steady state periods, and therefore we compared the mean±SD of RR intervals for the first and second half of each sequence, rejecting those in which the RR interval changed more than 10% and the SD more than 30%. The power (area below each spectral peak) at frequencies of interest was identified by a spectral decomposition method.^{12 14} Two frequency bands were considered: the so-called LF band (from 0.03 to 0.15 Hz, considered a predominant sympathetic marker) and the respiration frequency (HF) band (from 0.18 to 0.35 Hz, considered a vagal marker). HF fluctuations (an index of respiratory sinus arrhythmia) were more precisely defined by testing their coherence with oscillations in the respiratory spectrum (see below). During baseline recordings and during HF (near-respiratory rate, 0.20 Hz) neck suction, the HF band was further divided into two components: the respiratory component at 0.25 Hz, identified by coincidence in frequency and by significant coherence (see below) with respiration; and the neck suction component at 0.20 Hz, identified by (1) coincidence in frequency and significant coherence with the neck pressure signal and (2) absence of synchronicity and no significant coherence with the respiratory signal. To avoid the possibility that LF oscillations were due to occasionally slow breaths, LF fluctuations were considered "nonrespiratory" only when there was no coherence with respiration in the LF band.

Coherence Analysis
A simple way to assess the presence or absence of respiration- or neck suction–related fluctuations in RR interval is by comparison of RR interval and respiratory/neck suction spectra obtained simultaneously. For example, if a peak in the RR interval spectrum is not present at all in the spectrum of respiration, then this RR oscillation is certainly unrelated to respiration; however, the simultaneous presence of a peak with a similar frequency in both the RR interval and respiration spectrum does not necessarily imply a relation between these two oscillations. For example, these oscillations may occur at different times in the two signals or their phase relation might not be constant, as it should be if they were deterministically related. The squared coherence function is a mathematical bivariate spectral analysis method used to evaluate the phase stability of pairs of oscillations with the same frequency present in two signals. As with the correlation coefficient, this function spans from 0 (no relation between the two signals) to 1 (strong relation) and can be used as a statistical test for the relation between each pair of oscillations. The coherence between RR interval and respiration is close to 1 at the frequency of breathing in a normal subject. A similar high coherence is also normally found in heart-transplant recipients.¹⁴ The same technique can be used to confirm that LF oscillations on the RR interval spectrum are not artifacts caused by occasional slow breathing.¹⁴ With the pressure signal obtained in the neck collar as a reference, the coherence function was also used to assess whether neck suction generated a similar fluctuation in RR interval and blood pressure; if so, this could then be regarded as evidence of reinnervation.

The squared coherence function was evaluated by an autoregressive algorithm, as described by Baselli et al.²⁶ We assumed (as they did) that only spectral components with high squared coherence (>0.5) demonstrate a significantly stable phase relation between instantaneous RR interval and respiration; the squared coherence function between RR interval and respiration as well as neck pressure differentiated between respiratory- and nonrespiratory-related RR interval oscillations. We similarly assessed baroreflex-related effects on RR interval by relating RR interval fluctuations to the neck-suction pressure signal.

Statistical Analysis
Results are given as mean±SEM. Due to their skewed distribution, LF and HF oscillations were analyzed statistically only after natural logarithmic transformation. Student's t test for paired observations was used to evaluate differences within groups, and an unpaired t test was used for differences between groups. Simple linear regression analysis was used to assess the relation between the observed changes induced by neck suction and the time since transplantation.

	Results

Top Abstract Introduction Methods Results Discussion References

 
In the absence of atropine, there were highly significant (P<.001) differences in mean RR interval, RR interval variability (SD), and the power of LF and HF fluctuations, with or without neck suction, between control and transplant subjects (Table 1). Data for the four patients who underwent a repeat study before and after administration of a short-acting ß-blocker are given in Table 2.

View this table: [in this window] [in a new window]   Table 1. Effect of Neck Suction and Atropine on RR Interval in Heart-Transplant and Control Subjects

View this table: [in this window] [in a new window]   Table 2. Effect of Neck Suction and ß-Blockade by Esmolol Hydrochloride on RR Interval in Heart-Transplant Subjects (n=4)

Baseline Recordings
Compared with control subjects, all transplanted subjects showed abnormally low-amplitude RR interval HF related to respiration (Table 1). LF oscillations unrelated to the respiration rate were present in 13 of 26 subjects who underwent transplantation at least 14 months previously; the power of these oscillations was positively correlated with months since transplantation (r=.53, P<.01). During baseline rest, the power of the LF oscillations in transplant subjects was significantly lower than that in control subjects (0.73±0.20 versus 6.12±0.21 ln ms², P<.001, Table 1). The power of the LF oscillations in the 13 subjects with detectable LF was 1.42±0.35 ln ms².

LF power in SBP and DBP was similar in the transplant and control groups (SBP-LF: 0.96±0.16 versus 1.16±0.17 ln mm Hg², P=NS; DBP-LF: 0.66±0.11 versus 0.82±0.09 ln mm Hg², P=NS), whereas the HF power in DBP was greater in transplant subjects compared with control subjects (SBP-HF: 1.39±0.14 versus 1.32±0.12 ln mm Hg², P=NS; DBP-HF: 0.75±0.08 versus 0.29±0.03 ln mm Hg², P<.001), probably because of the lack of vagal, beat-to-beat control of RR interval.

Effect of HF Neck Suction
HF neck suction stimulation at 0.2 Hz caused the appearance of a second and distinct peak in the RR interval power spectrum of control subjects (Fig 1), such that the power at 0.20 Hz increased from 0.00±0.00 to 6.95±0.28 ln ms². The power in the 0.20-Hz peak was similar to that associated with the respiratory peak (6.58±0.22, P=NS). Conversely, HF neck suction did not generate any fluctuation in transplant subjects (Table 1; Fig 2).

View larger version (32K): [in this window] [in a new window]   Figure 1. Graphic example of the effects of sinusoidal neck suction at near-respiratory frequency in a control subject. Left graphs show the time series of RR interval, respiration, and neck chamber pressure; right, their autoregressive (AR) power spectra. Notice that the peaks of respiration (0.25 Hz) and neck chamber pressure (0.20 Hz) are close in frequency but distinct (middle right and bottom right). These two baroreceptor stimuli produce two distinct peaks on the RR interval power spectrum (top right). The RR interval time series (top left) shows typical beats due to the summation/subtraction of two input waves of close frequency. au indicates arbitrary units.

View larger version (31K): [in this window] [in a new window]   Figure 2. Graphic example of the effects of sinusoidal neck suction at near-respiratory frequency (0.20 Hz) compared with those of respiration (0.25 Hz) in a nonreinnervated heart-transplant subject who underwent transplantation 5 months earlier. Notice that the RR interval time series (top left) shows a steady but small monotonous oscillation without beats. As a consequence, only one peak is evident in the RR interval spectrum (top right). This peak is coincident with that of the respiration (middle right), which modifies the heart period through nonreflex inspiratory atrial stretch, and not with that of neck pressure (bottom right), which normally modifies the heart period only through reflex autonomic effects. As a consequence, this respiratory fluctuation in RR cannot be considered evidence of autonomic activity. AR indicates autoregressive; au, arbitrary units.

In all subjects, there was a high coherence (0.95 to 0.99) in the HF band at the frequency of respiration (0.25 Hz) between RR interval and respiration (since the neck suction was at 0.20 Hz). As expected, no coherence was present between RR interval and neck suction pressure at 0.25 Hz. In all control subjects, we found a high (0.90 to 0.99) coherence between RR interval and neck suction pressure coincident with the frequency of neck suction (0.20 Hz), whereas again no coherence was expected or found at this frequency between RR interval and respiration (0.25 Hz). No coherence at 0.20 Hz was found in transplant subjects between RR interval and neck pressure or respiration (ie, no evidence for vagal reinnervation).

HF neck suction induced similar SBP and DBP changes in the two groups. Because of a slight increase in breathing depth during the HF neck stimulation, the SBP respiratory component (ie, the power corresponding with the 0.25-Hz peak) slightly increased in control subjects to 1.52±0.13 ln mm Hg² (P=NS) and to a greater extent in transplant subjects (to 1.70±0.11 ln mm Hg², P<.01). The DBP respiratory component (ie, the power corresponding with the 0.25-Hz peak) increased in transplant subjects (to 1.05±0.14 ln mm Hg², P<.01) but not in control subjects (to 0.34±0.06 ln mm Hg², P=NS). Again, this difference may be because control subjects with intact vagi can rapidly stabilize DBP, which is more readily regulated by beat-to-beat RR interval change. There was no evident SBP or DBP power corresponding to the 0.20-Hz peak (ie, the frequency of neck suction stimulation) in either control or transplant subjects, with or without HF neck suction.

Effect of LF Neck Suction
In control subjects, LF neck suction at 0.1 Hz increased the power in the LF band from 6.12±0.21 to 8.27±0.21 ln ms², P<.001 (Table 1; Fig 3).

View larger version (29K): [in this window] [in a new window]   Figure 3. Graphic example of the effects of sinusoidal neck suction at LF (0.1 Hz) in a control subject. Notice the marked RR interval variability induced by this stimulation, causing a very high peak at the same frequency as that of the neck chamber pressure. The respiratory component on RR interval spectrum is not evident due to the very large scale necessary to show the LF peak, but it can be seen on greater amplification (not shown). Abbreviations as in Fig 2.

LF neck suction stimulation similarly increased the power in the LF band in the 13 transplant subjects in whom (reduced) LF oscillations were present at rest (from 1.42±0.35 to 2.18±0.34 ln ms², P<.01) and generated LF oscillations in 4 more heart transplant subjects (who underwent transplantation 10 to 21 months previously and in whom spontaneous LF oscillations were not evident at baseline) from 0.00±0.00 to 1.45±0.12 ln ms² (P<.001). Fig 4 shows an example of the RR interval LF fluctuation during neck suction at 0.1 Hz in 1 transplant subject. Table 1 reports the average results of all 26 transplant subjects; the average increase in LF was still significant.

View larger version (26K): [in this window] [in a new window]   Figure 4. Graphic example of the effects of sinusoidal neck suction at LF (0.1 Hz) in a heart-transplant subject who underwent transplantation 20 months earlier. The RR interval time series (top left) shows clear, slow fluctuations synchronous with the slow fluctuations present in the neck chamber (bottom left). The RR interval power spectrum (top right) also shows a clear peak synchronous with that of the neck, much greater than that due to respiration (barely visible). This pattern was observed in more than 50% of our subjects, but thus far only more than 14 months after transplantation. Notice that respiration has a clearly different pattern and spectrum; therefore, it is not possible that it could have generated such a marked fluctuation in RR. Abbreviations as in Fig 2.

We found a high coherence between RR interval and neck suction pressure coincident with the frequency of neck suction (0.1 Hz) in all control subjects (0.97 to 0.99) and in all transplant subjects who exhibited LF during neck suction (0.67 to 0.98). No coherence was found with the respiratory signal in the LF band, in either control or transplant subjects, confirming that the LF oscillations observed were not due to the effect of occasional slow breathing. Conversely, in all subjects, there was a significant coherence between RR interval and respiration in the HF band (0.91 to 0.99) at the frequency of respiration (0.25 Hz).

As expected because peripheral sympathetic efferents were intact, LF neck suction induced similar SBP and DBP changes in control and transplant subjects: SBP-LF increased to 1.88±0.20 ln mm Hg² (P<.001) in control subjects and to 1.70±0.22 ln mm Hg² (P<.01) in transplant subjects; DBP-LF increased to 1.19±0.14 ln mm Hg² (P<.05) in control subjects and to 1.11±0.16 ln mm Hg² (P<.025) in transplant subjects.

Effect of Removal of Vagal Efferent Activity
The effects of atropine were compared with baseline only in the same subgroup of subjects who took the drug. No significant differences at baseline were found between subjects who underwent the atropine test compared with those who did not.

Intravenous atropine greatly reduced all the RR interval fluctuations in control subjects (Table 1) and completely abolished the effect of HF neck suction (Fig 5). As a consequence, no significant differences were observed between control and transplant subjects after atropine. LF neck suction still significantly increased LF power (0.88±0.54 to 2.60±0.09 ln ms², P<.05). Fig 6 shows one example of the RR interval spectra obtained before and after atropine, with and without LF neck suction.

View larger version (21K): [in this window] [in a new window]   Figure 5. Graphic example of the effects of removal of vagal efferent activity by atropine in a control subject, during near–respiratory rate neck suction. The spectra of RR interval (top), respiration (middle), and neck chamber pressure (bottom) are shown before (left) and after (right) 0.04 mg/kg atropine IV. Notice the two peaks seen at rest (same as in Fig 1), one synchronous with respiration (0.25 Hz) and one synchronous with neck chamber pressure (0.20 Hz) before atropine. After atropine, the RR interval spectral power is markedly reduced (compare the two different scales), but whereas the neck pressure synchronous peak disappeared completely, the respiratory synchronous peak is still present, although markedly reduced. We believe this is due to inspiratory venous return causing atrial stretch and hence mechanical modulation of heart rate. Since the RR interval was not influenced by neck suction at 0.20 Hz, this 0.25-Hz fluctuation is not autonomic (see text, Fig 6, and References 11-13). Notice that the power of this respiratory peak is comparable with that seen in heart-transplant subjects. The extra graph on the right shows the same RR interval spectrum on a 10-fold reduction in the y-axis amplification, to show the persistence of a reduced LF component at 0.25 Hz after atropine. au indicates arbitrary units.

View larger version (16K): [in this window] [in a new window]   Figure 6. Graphic comparisons of the changes vs baseline (top left) induced by LF neck suction alone (top right), by atropine alone (bottom left), and by the combination of atropine and LF neck suction (bottom right) in one control subject on the RR interval power spectrum. Notice the marked increase in the LF component induced by neck suction. The marked reduction in overall power induced by atropine (the scales have been markedly reduced in both bottom graphs) does not prevent the increase in LF power during LF neck suction.

In transplant subjects, atropine did not reduce the inspiration-induced (mechanical) fluctuations observed at rest in HF (Fig 7) and did not block the increase in LF power induced by LF neck suction stimulation (0.50±0.34 to 1.46±0.49 ln ms², P<.05). Fig 8 shows one example of the RR interval spectra obtained before and after atropine, with and without LF neck suction, in one heart-transplant subject.

View larger version (26K): [in this window] [in a new window]   Figure 7. Graphic example of the effects of removal of vagal activity by atropine in a heart-transplant subject who underwent transplantation 63 months earlier, during near-respiratory neck suction. Notice the presence of clear LF and HF components in the RR interval spectrum. However, only one HF peak is present in the spectrum: the one synchronous with the respiratory component, at 0.25 Hz (as in Fig 3 and unlike Figs 1 and 5). This indicates that the observed 0.25-Hz HF component cannot be due to autonomic activity, despite the fact that the presence of a nonrespiratory LF component suggests evidence of reinnervation. After atropine, the 0.25-Hz HF component is not reduced (as it would be if it were due to vagal activity), and neither is the LF component, suggesting a solely sympathetic origin for the latter. There is no activity at the neck suction frequency (0.2 Hz) since it is too fast for sympathetic activity to be able to follow. au indicates arbitrary units.

View larger version (21K): [in this window] [in a new window]   Figure 8. Graphic comparisons of changes versus baseline (top left) induced by LF neck suction alone (top right), by atropine alone (bottom left), and by the combination of atropine and LF neck suction (bottom right) on the RR interval power spectra in one heart-transplant subject (same subject as in Fig 7). Notice that no changes in either x or y scale occurred after atropine, indicating no changes in heart rate or in overall spectral power. Without neck suction, the LF component is not reduced after atropine. The LF component (0.1 Hz) is increased by neck suction both before and after atropine to a similar extent, hence suggesting that this LF component is due to sympathetic activation, not vagal activation. The absence of vagal activity is further confirmed by the lack of decrease in the HF component after atropine, as in Fig 7.

The coherence between RR interval and respiration remained high at the frequency of respiration (0.25 Hz) in all conditions (with or without neck suction) and in all subjects (0.81 to 0.98). The coherence between RR interval and neck suction pressure at 0.1 Hz remained present both in control subjects and in those transplant subjects who had LF fluctuations before atropine (0.67 to 0.99). Conversely, the coherence between RR interval and neck suction pressure at 0.20 Hz disappeared during HF neck suction in all control subjects after atropine.

Atropine induced only minor and overall insignificant changes in both SBP and DBP fluctuations in both groups, and the effects of LF or HF neck suction stimulations were not significantly influenced.

Effect of ß-Blockade
In the four subjects tested, short-acting ß-blockade by esmolol hydrochloride (Table 2) increased the resting RR interval from 604±23 to 689±23 ms (P<.01). There was also a reduction in spontaneous LF power from 3.23±0.42 to 2.43±23 ln ms² (P<.05). During 0.1-Hz neck suction before ß-blockade, the expected increase in LF occurred (from 3.23±0.42 to 3.81±0.30 ln ms², P<.05). After ß-blockade, an increase in LF power was still seen during neck suction (from 2.43±0.23 to 2.75±0.11 ln ms², P<.05), but the latter value during neck suction and ß-blockade was significantly less than during neck suction before ß-blockade (2.75±0.11 versus 3.81±0.30 ln ms², respectively, P<.05). Fig 9 shows an example of the results obtained in one subject.

View larger version (18K): [in this window] [in a new window]   Figure 9. Graphic comparisons of power spectral analysis during controlled breathing before and after ß-blockade in a 60-year-old woman who underwent heart transplantation 52 months previously. Bottom, Note the marked diminution of the power at 0.1 Hz (LF) resulting from ß-blockade. Whereas LF power was markedly augmented by neck suction at 0.1 Hz (top right), this reflex increase in LF power was markedly diminished by ß-blockade (bottom right).

	Discussion

Top Abstract Introduction Methods Results Discussion References

 
Previous Reports on Reinnervation After Human Cardiac Transplantation
Several recent studies suggest the possibility of reinnervation of the transplanted human heart. Stark et al⁸ reported the occurrence of anginal pain in a few transplanted subjects, which they interpret as due to a possible sympathetic reinnervation. This observation, although interesting, has the obvious limitation that the majority of transplant subjects do not have angina. Similar limitations also apply to reports of infrequent vasovagal reactions. In addition, these can occur in subjects without donor heart reinnervation.^{9 27} In summary, these observations cannot be used as practical tests of reinnervation or of the time of onset of reinnervation in any group of transplant subjects. Because the origin of the anginal pain and of the vasovagal reaction are still unclear, they cannot be used as sure proof of reinnervation.

Wilson et al⁶ demonstrated a significant, although reduced, release of norepinephrine in response to various physiological or pharmacological stimuli at least 1 year after transplantation. These results were recently confirmed by Kaye et al.⁷ Such methods are difficult to apply and to repeat on a large number of subjects and, despite their efficacy, can test only the hypothesis of sympathetic reinnervation.

Use of Neck Suction Technique to Study Reinnervation
Our results indicate that reinnervation develops after human heart transplantation and that power spectrum analysis of RR interval variability is a very practical way to test it. Moreover, this specific baroreceptor stimulation changes the heart rate by unequivocal reflex activity. It has been shown that the RR interval is highly modulated by baroreceptor activity.^{18 28} The neck suction technique allows noninvasive stimulation of the carotid arterial baroreceptors without any hemodynamic effect other than reflex.^{17 18 19 20 28} Any change in RR interval unequivocally related to neck chamber stimulation is good evidence of reinnervation. In addition, we and others^{16 18 19 20 28} have shown (by using sinusoidal neck suction of a given frequency and amplitude) that arterial baroreceptors are capable of inducing both LF and HF components in the RR interval variability. It has also been shown in the normal subject that whereas both LF and HF spontaneous RR interval fluctuations can be produced by vagal activity, only LF fluctuations could be transmitted by the more slowly responding sympathetic activity.²¹ We therefore used this approach to provide further information about the mechanism responsible for the RR interval changes in transplant subjects.

Effect of LF Neck Suction Stimulation
With this methodology, we have found that neck suction could induce reflex changes in the RR interval of the donor heart in 17 of 26 transplant subjects. In all of these subjects, only the LF fluctuations could be modified by neck suction. All 13 of the 26 transplant subjects who exhibited spontaneous, nonrespiratory LF fluctuations showed increased oscillation with neck suction. Furthermore, neck suction could induce LF fluctuation in the RR interval in 4 subjects who did not have evident spontaneous LF fluctuations at rest. These results indicate that signs of baroreflex activity, and hence of reinnervation, are present in a large number of subjects after at least 14 months since heart transplantation. Neck suction–induced LF RR oscillations were highly correlated with neck pressure fluctuations in both transplant and control subjects and were not correlated with the respiratory signal, thus excluding the possibility of respiratory artifacts (ie, occasional slow breaths causing LF oscillations). Although the presence of spontaneous LF oscillations in the RR interval power spectrum does indicate autonomic activity, its absence cannot be regarded as a sign of denervation since neck suction could induce reflex RR interval changes in some subjects who did not show spontaneous LF fluctuations.

Effect of HF Neck Suction Stimulation
With the HF stimulation, all control subjects showed two distinct peaks in the RR interval spectrum: one at 0.20 Hz due to the effect of neck suction via the baroreflex and one at 0.25 Hz due to the combined autonomic and mechanical effects of respiration. Unlike the control subjects, none of the transplant subjects showed any response at a neck suction frequency of 0.20 Hz, whereas all subjects showed clear respiratory fluctuations at 0.25 Hz. These results further confirm that the respiratory fluctuations observed in the transplant subjects (at 0.25 Hz) were not the result of autonomic (vagal or sympathetic) modulation but only the result of a direct mechanical effect of respiration^{11 12 13} on the donor sinus node. The lack of response to HF neck suction suggested that, assuming the conclusions of Saul et al²¹ in normal subjects are valid, the vagus was not active in the heart transplant subjects.

Effect of Intravenous Atropine on Neck Suction Stimulation
Despite the lack of effect of HF stimulation, the absence of vagal activity could not be ruled out since it could be argued that in heart transplant recipients, some rudimentary vagal activity might be observed only at the LF of neck suction stimulation. The presence of spontaneous and/or neck suction–induced LF fluctuations therefore theoretically could be due to sympathetic activity because the arterial baroreflex acts on the heart via both vagal and sympathetic arms, and both vagal and sympathetic activity are thought to contribute to spontaneous LF oscillations in RR interval variability.²² However, the lack of any difference in neck suction–induced LF before and after administration of atropine in the transplant subjects clearly indicated that vagal activity was not present in any of the subjects examined in the present study. In control subjects, atropine produced a marked decrease in LF power, but the remaining LF fluctuations could still be increased by neck suction, and the observed values fell in a range similar to that of transplant subjects. This again indicated that reinnervation in the subjects we examined was only by way of sympathetic activity. Atropine did not change the spontaneous HF observed in the transplant subjects, further confirming previous reports about the nonautonomic origin of respiratory-related variability.^{11 12 14} Finally, the present data in the control subjects support the hypothesis²⁹ that the LF component of heart rate variability behaves as a marker of sympathetic activity, or rather of the sympathetic response via the baroreflex.¹⁶ After atropine, the different response in control subjects to 0.10 Hz (reduced LF still present in the RR interval variability could be increased during LF neck stimulation) and to 0.20 Hz (0.20-Hz neck suction could not produce or increase HF oscillations in the RR interval after atropine) confirms previous findings²¹ that sympathetic activity to the heart is limited to or predominant in the LF band, whereas vagal activity to the heart can produce RR interval oscillations in both LF and HF bands.

Effect of ß-Blockade
In contrast to the above-mentioned lack of effects with atropine, partial ß-blockade significantly reduced both the power of the resting spontaneous LF fluctuations and the increase in power caused by cycles of baroreceptor stimulation produced by rhythmic neck suction at 0.1 Hz, confirming that both the "spontaneous" LF fluctuations and its increase by neck suction are mediated by the efferent sympathetic nerves to the donor heart.

Conclusions
The present study demonstrates the presence of some autonomic reinnervation in the human transplanted heart. This phenomenon is frequently observed after about 14 months since transplantation, but its amount is limited and appears in the present study to be due only to the sympathetic component. Although thus far we have not found any evidence of vagal reinnervation, future observations made with the use of this technique might eventually disclose its occurrence. This could be tested by neck suction–induced HF in the RR interval spectrum at near-respiratory frequency. The neck suction method used in the present study, combined with the power spectrum analysis of RR interval variability (and its coherence with respiration and neck chamber pressure), is fairly simple, noninvasive, and well tolerated by patients and can be easily standardized. We have confirmed that the presence of spontaneous LF components in RR interval variability, detected by power spectrum analysis without neck suction, is a marker of reinnervation (provided care is taken to exclude occasional slow breathing).¹⁴ Spontaneous LF may occasionally be absent in subjects in whom LF can still be reflexly induced by neck suction.

The present data show that neck suction can be used to test the presence of both the sympathetic and vagal components of autonomic activity; it thus appears ideal for studying the presence and type of reinnervation and its progression over time.

Although other studies have shown the probability of some late reinnervation of the heart by tyramine-induced release of catecholamines,^{6 7 30} the present study is the first to demonstrate clear evidence of baroreflex control of heart rate by reinnervation, which appears to be solely sympathetic.

	Selected Abbreviations and Acronyms

 

DBP = diastolic blood pressure HF = high frequency LF = low frequency SBP = systolic blood pressure

	Acknowledgments

 
Dr Bianchini is the recipient of a Young Investigator Award from the Italian Society of Cardiology.

Received December 29, 1994; revision received April 26, 1995; accepted July 5, 1995.

	References

Top Abstract Introduction Methods Results Discussion References

 

1. Hunt S, Billingham M. Long-term results of cardiac transplantation. Annu Rev Med. 1991;42:437-447. [Medline] [Order article via Infotrieve]
   
2. Hallmann GL, Leatherman LL, Leachman RD, Rochelle DG, Bricker DL, Bloodwell RD, Cooley DA. Function of the transplanted human heart. J Thorac Cardiovasc Surg. 1969;58:318-325. [Medline] [Order article via Infotrieve]
   
3. Stinson EB, Griepp RB, Schroeder JS, Dong E, Shumway NE. Hemodynamic observations one and two years after cardiac transplantation in man. Circulation. 1972;45:1183-1194. [Abstract/Free Full Text]
   
4. Kent KM, Cooper T. The denervated heart. N Engl J Med. 1974;291:1017-1021.
   
5. Schroeder JS. Hemodynamic performance of the human transplanted heart. Transplant Proc. 1979;11:304-308. [Medline] [Order article via Infotrieve]
   
6. Wilson RF, Christensen BV, Olivari MT, Simon A, White CW, Laxson DD. Evidence of structural sympathetic reinnervation after orthotopic cardiac transplantation in humans. Circulation. 1991;83:1210-1220. [Abstract/Free Full Text]
   
7. Kaye DM, Esler M, Kingwell B, McPherson G, Esmore D, Jennings G. Functional and neurochemical evidence for partial cardiac sympathetic reinnervation after cardiac transplantation in humans. Circulation. 1993;88:1110-1118. [Abstract/Free Full Text]
   
8. Stark RP, McGinn AL, Wilson RF. Chest pain in cardiac-transplant recipients: evidence of sensory reinnervation after cardiac transplantation. N Engl J Med. 1991;324:1791-1794. [Medline] [Order article via Infotrieve]
   
9. Fitzpatrick AP, Banner N, Cheng A, Yacoub M, Sutton R. Vasovagal reaction may occur after orthotopic heart transplantation. J Am Coll Cardiol. 1993;21:1132-1137. [Abstract]
   
10. Rudas L, Pflugfelder PW, Kostuk WJ. Comparison of hemodynamic responses during dynamic exercise in the upright and supine postures after orthotopic cardiac transplantation. J Am Coll Cardiol. 1990;16:1367-1373. [Abstract]
    
11. Bernardi L, Keller F, Sanders M, Reddy PS, Griffith B, Meno F, Pinsky MR. Respiratory sinus arrhythmia in the denervated human heart. J Appl Physiol. 1989;67:1447-1455. [Abstract/Free Full Text]
    
12. Bernardi L, Salvucci F, Suardi R, Soldá PL, Calciati A, Perlini S, Falcone C, Ricciardi L. Evidence for an intrinsic mechanism regulating heart rate variability in the transplanted and the intact heart during submaximal dynamic exercise? Cardiovasc Res. 1990;24:969-981. [Free Full Text]
    
13. Bernardi L, Leuzzi S, Valle F, Marchesi E, Radaelli A, Martinelli L, Viganó M, Finardi G, Sleight P. Respiratory sinus arrhythmia in recently heart transplanted subjects is caused by a non-autonomic mechanism. Eur Heart J. 1993;14(suppl):207. Abstract.
    
14. Bernardi L, Valle F, Leuzzi S, Rinaldi M, Marchesi E, Falcone C, Martinelli L, Viganó M, Finardi G, Radaelli A. Non-respiratory components of heart rate variability in heart transplant recipients: evidence of autonomic reinnervation? Clin Sci (Colch). 1994;86:537-545. [Medline] [Order article via Infotrieve]
    
15. De Boer RW, Karemaker JW, Strackee J. Hemodynamic fluctuations and baroreflex sensitivity in humans: a beat-to-beat model. Am J Physiol. 1987;253:680-689.
    
16. Sleight P, La Rovere MT, Mortara A, Pinna G, Maestri R, Leuzzi S, Bianchini B, Tavazzi L, Bernardi L. Physiology and pathophysiology of heart rate and blood pressure variability in humans: is power spectral analysis largely an index of baroreflex gain? Clin Sci (Colch). 1995;88:103-109.[Medline] [Order article via Infotrieve]
    
17. Eckberg DL, Cavanagh MS, Mark AL, Abboud FM. A simplified neck suction device for activation of carotid baroreceptor. J Lab Clin Med. 1975;85:167-173. [Medline] [Order article via Infotrieve]
    
18. Sleight P, Leuzzi S, Valle F, Giangregorio F, Spadacini GM, Johnston JA, Bernardi L. Evidence that respiratory sinus arrhythmia (RSA) in humans originates from arterial baroreceptors. Presented at the XXXII Congress of the International Union of Physiological Sciences; August 1-6, 1993; Glasgow.
    
19. Båth E, Lindblad LE, Wallin BG. Effects of dynamic and static neck suction on muscle nerve sympathetic activity, heart rate and blood pressure in man. J Physiol (Lond). 1981;311:551-564. [Abstract/Free Full Text]
    
20. Cross SJ, Cowie MR, Rawles JM. Mutual interactions of respiratory sinus arrhythmia and the carotid baroreceptor–heart-rate reflex. Clin Sci (Colch). 1992;82:139-145. [Medline] [Order article via Infotrieve]
    
21. Saul JP, Berger RD, Albrecht P, Stein SP, Chen MH, Cohen RJ. Transfer function analysis of the circulation: unique insights into cardiovascular regulation. Am J Physiol. 1991;261:H1231-H1245. [Abstract/Free Full Text]
    
22. Pomeranz B, Macaulay RJB, Caudill MA, Kutz I, Adam D, Gordon D, Kilborn KM, Barger AC, Shannon DC, Cohen RJ, Benson H. Assessment of autonomic function in humans by heart spectral analysis. Am J Physiol. 1985;248:H151-H153. [Abstract/Free Full Text]
    
23. Hayano J, Sakakibara Y, Yamada A, Yamada M, Mukai S, Fujinami T, Yokoyama K, Watanabe Y, Takata K. Accuracy of assessment of cardiac vagal tone by heart rate variability in normal subjects. Am J Cardiol. 1991;67:199-204.[Medline] [Order article via Infotrieve]
    
24. Pagani M, Lombardi F, Guzzetti S, Rimoldi O, Furlan R, Pizzinelli P, Sandrone G, Malfatto G, Dell'Orto S, Piccaluga E, Turiel M, Baselli G, Cerutti S, Malliani A. 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]
    
25. Arai Y, Saul P, Albrecht P, Hartley H, Lilly L, Cohen R, Colucci W. Modulation of cardiac autonomic activity during and immediately after exercise. Am J Physiol. 1989;256:H131-H141.
    
26. Baselli G, Cerutti S, Civardi S, Liberati D, Lombardi F, Malliani A, Pagani M. Spectral and cross-spectral analysis of heart rate and arterial blood pressure variability signals. Comput Biomed Res. 1986;19:520-534. [Medline] [Order article via Infotrieve]
    
27. Sherrer U, Vissing S, Morgan BJ, Hanson P, Victor RG. Vasovagal syncope after infusion of vasodilator in a heart-transplant recipient. N Engl J Med. 1990;322:602-604. [Medline] [Order article via Infotrieve]
    
28. Bernardi L, Leuzzi S, Radaelli A, Passino C, Johnston JA, Sleight P. Low frequency spontaneous fluctuations of R-R interval and blood pressure in conscious humans: a baroreceptor or central phenomenon? Clin Sci (Colch). 1994;87:649-654. [Medline] [Order article via Infotrieve]
    
29. Malliani A, Pagani M, Lombardi F, Cerutti S. Research advances series: cardiovascular neural regulation explored in the frequency domain. Circulation. 1991;84:482-492. [Abstract/Free Full Text]
    
30. Burke NM, McGinn AL, Homans DC, Christensen BV, Kubo SH, Wilson RF. Evidence of functional sympathetic reinnervation of left ventricle and coronary arteries after orthotopic cardiac transplantation in humans. Circulation. 1995;91:72-78.[Abstract/Free Full Text]
    

This article has been cited by other articles:

				A. Malliani, C. Julien, G. E. Billman, S. Cerutti, M. F. Piepoli, L. Bernardi, P. Sleight, M. A. Cohen, C. O. Tan, D. Laude, et al. Cardiovascular variability is/is not an index of autonomic control of circulation J Appl Physiol, August 1, 2006; 101(2): 684 - 688. [Full Text] [PDF]

				B. Stemper, L. Bernardi, F. B. Axelrod, G. Welsch, C. Passino, and M. J. Hilz Sympathetic and parasympathetic baroreflex dysfunction in familial dysautonomia Neurology, October 26, 2004; 63(8): 1427 - 1431. [Abstract] [Full Text] [PDF]

				S. S. Lovric, V. Avbelj, R. Trobec, D. Zorman, P. Rakovec, S. Hojker, B. Gersak, and M. Milcinski Sympathetic reinnervation after heart transplantation, assessed by iodine-123 metaiodobenzylguanidine imaging, and heart rate variability Eur. J. Cardiothorac. Surg., October 1, 2004; 26(4): 736 - 741. [Abstract] [Full Text] [PDF]

				M. Brys, C. M. Brown, H. Marthol, R. Franta, and M. J. Hilz Dynamic cerebral autoregulation remains stable during physical challenge in healthy persons Am J Physiol Heart Circ Physiol, August 7, 2003; 285(3): H1048 - H1054. [Abstract] [Full Text] [PDF]

				L Bernardi, C Passino, C Porta, E Anesi, G Palladini, and G Merlini Widespread cardiovascular autonomic dysfunction in primary amyloidosis: does spontaneous hyperventilation have a compensatory role against postural hypotension? Heart, December 1, 2002; 88(6): 615 - 621. [Abstract] [Full Text] [PDF]

				P. van de Borne, J. Neubauer, M. Rahnama, J.-L. Jansens, N. Montano, A. Porta, V. K. Somers, and J. P. Degaute Differential Characteristics of Neural Circulatory Control: Early Versus Late After Cardiac Transplantation Circulation, October 9, 2001; 104(15): 1809 - 1813. [Abstract] [Full Text] [PDF]

				S. Hunt Reinnervation of the Transplanted Heart -- Why Is It Important? N. Engl. J. Med., September 6, 2001; 345(10): 762 - 764. [Full Text] [PDF]

				C. Keyl, A. Schneider, M. Dambacher, and L. Bernardi Time delay of vagally mediated cardiac baroreflex response varies with autonomic cardiovascular control J Appl Physiol, July 1, 2001; 91(1): 283 - 289. [Abstract] [Full Text] [PDF]

				P. Salmon, S. C. Stanford, G. Mikhail, S. Zielinski, and J. R. Pepper Hemodynamic and Emotional Responses to a Psychological Stressor After Cardiac Transplantation Psychosom Med, March 1, 2001; 63(2): 289 - 299. [Abstract] [Full Text] [PDF]

				N. Montano, C. Cogliati, V. J. D. da Silva, T. Gnecchi-Ruscone, M. Massimini, A. Porta, and A. Malliani Effects of Spinal Section and of Positive-Feedback Excitatory Reflex on Sympathetic and Heart Rate Variability Hypertension, December 1, 2000; 36(6): 1029 - 1034. [Abstract] [Full Text] [PDF]

				C. Keyl, A. Schneider, M. Dambacher, U. Wegenhorst, M. Ingenlath, M. Gruber, and L. Bernardi Dynamic Cardiocirculatory Control During Propofol Anesthesia in Mechanically Ventilated Patients Anesth. Analg., October 1, 2000; 91(5): 1188 - 1195. [Abstract] [Full Text] [PDF]