Contribution to Heart Rate Variability by Mechanoelectric Feedback
Stretch of the Sinoatrial Node Reduces Heart Rate Variability
Background Heart rate variability is an important prognostic indicator for sudden death. An increased risk of sudden death and arrhythmia is associated with reduced heart rate variability in heart failure. In heart failure, there is also dilatation of the atria, which raises the prospect that there could be some physiological basis to possibly link heart rate variability with atrial dilatation. We therefore investigated whether sustained atrial stretch could modulate heart rate variability directly.
Methods and Results Pigs were anesthetized and their hearts exposed. A specially built device stretched the sinoatrial node before and after vagal section and then after administration of propranolol. Stretch of the sinoatrial node decreases heart rate variability in the following ways: The standard deviation of the beat-to-beat interval decreases (4.2 to 2.6 ms; P=.004), and the high-frequency components are reduced (control, 6.5±2.2 ms2; during stretch, 1.4±0.3 ms2; P=.003). After section of both vagi, the high-frequency components are reduced by stretch of the sinoatrial node (2.8±0.9 ms2 for control versus 1.2±0.3 ms2 during stretch; P=.05). Similarly, after both vagal section and β-blockade, stretch of the sinoatrial node reduces the high-frequency components (10.6±3.5 ms2 for control versus 3.0±1.5 ms2 during stretch; P=.01).
Conclusions We conclude that stretch of the sinoatrial node reduces high-frequency heart rate variability. This may account in part for the reduced heart rate variability seen in clinical conditions in which the right atrium is dilated, such as congestive cardiac failure.
Heart rate variability is one of the most important prognostic indexes for sudden cardiac death1 2 3 4 and cardiac arrhythmias.5 6 However, its pathophysiological mechanism has not been conclusively delineated. Increased vagal tone increases heart rate variability, especially the low-frequency components.7 Reduced vagal tone is thought to be responsible for both the decrease in heart rate variability and the associated increase in arrhythmias and mortality.8 9 However, there is evidence from heart transplant patients that there is an intrinsic mechanism independent of autonomic reflexes involved in the genesis of heart rate variability.10
Heart rate variability has been found to be reduced in patients with impaired ventricular function,11 12 although this finding is not universal.3 Impaired ventricular function is often associated with high atrial pressures and atrial dilatation. Dilatation of the atrium is associated with a higher degree of mechanical strain in the atrial myocardium. This could occur via mechanoelectric feedback, that is, the effect of the mechanical activity of the heart on its electric activity,13 which could cause an electrophysiological effect on the sinus node that would modulate heart rate variability. We investigated the influence of stretch of the sinoatrial node on heart rate variability in the in situ heart of the anesthetized pig.
Fifteen large, white, Landrace pigs of either sex (weight, 18 to 26 kg) were premedicated with ketamine (10 to 20 mg/kg) and anesthetized with intravenous methohexital sodium via the marginal ear vein. The animals were intubated and ventilated while anesthesia was maintained with 1% to 1.5% halothane in an equal-volume mixture of nitrous oxide and oxygen. Tidal volume was set at levels that maintained physiological values of arterial pH, Po2, and Pco2, and this was checked by regular blood gas analysis. Arterial pressure was recorded via the left common carotid artery with a 7F Gaeltec micromanometer catheter that was advanced into the aorta. An intravenous cannula was introduced into the right jugular vein for infusion of 0.9% saline solution. The anterior chest wall was removed at the costochondral junctions. The pericardium was opened and the heart supported in a pericardial cradle.
Monophasic action potentials were recorded from the anterior surface of the right ventricle by use of a suction electrode14 (Fig 1A⇓). These electrodes allow ease of placement and recordings for up to 30 minutes from a single placement. The RR interval from the ventricle was chosen so that the results would be as relevant as possible to the already large body of information on clinical heart rate variability. In a series of 10 experiments, monophasic action potentials were simultaneously recorded from the right atrium and right ventricle. All the signals were amplified by high-impedance DC amplifiers (Lectromed MT8P) and displayed on a Tektronic model 5103 N oscilloscope during the experiment. The signals were stored on magnetic tape (TEAC XR-50l) for later analysis and also digitized at 1000 Hz by use of a Cambridge Electronic Design (CED 1401) and IBM PC-AT–compatible computer.
Identification and Stretch of Sinoatrial Node
A cooling probe was made that consisted of a brass head connected to two tubes. Ice-cold water was pumped into the probe through one tube and returned down the other tube to a bucket that contained a mixture of ice and water. The cooling probe was used to identify the site of maximal heart rate decrease on application of the probe to the wall of the atrium. This site was found to be ≈1.5 cm inferior to the junction of the atrium with the superior vena cava on the posterolateral aspect of the right atrium. The probe was never observed to have any effect on heart rate at other sites. This procedure was repeated at least three times and the site accepted only if a reproducible fall in heart rate was obtained. The site so identified was taken to be the site governing the heart rate and is hereafter referred to as the site of the sinoatrial node.
A specially built device was used to stretch the sinoatrial node. This consisted of three legs, each of which had a suction foot to allow attachment to the heart. The legs could be moved apart by a known amount by the action of a screw (Fig 1B⇑). This device was attached by suction feet, one on each leg. The device was placed on the atrium with the sinoatrial node at the center of the area subtended by the feet.
A control recording was taken with the device in place between the pericardium and the heart but without the feet attached by suction to the atrium. A recording was then taken with the device attached to the atrium, and the distance between the feet was increased so that the sinoatrial node was stretched by 10% of its initial measured length. All recordings were for 7 minutes. The control and intervention recordings were taken with the ventilator on the same settings. The RR interval was measured from the upstroke of one monophasic action potential to the upstroke of the next (Fig 1A⇑). The vagi that had been identified previously were divided and recordings repeated. The cardiac sympathetic nervous system was blocked with 40 μg/kg propranolol, and recordings were repeated. This dose in this preparation reduces heart rate and also gives excellent attenuation (≈80%) of the heart rate increase in response to a test injection of 10 μg/kg adrenaline.
The RR interval was measured from one monophasic action potential upstroke to the next. Sections of data were taken where possible without premature beats, but if premature beats were present, the short RR interval and the subsequent long RR interval were replaced by the mean RR interval. An autoregressive frequency spectrum was calculated, and the frequency components were defined as follows: low frequency, 0.04 to 0.15 Hz; high frequency, 0.15 to 0.4 Hz. We did not consider the length of the data recording to be long enough to comment on the ultralow-frequency band.
The Wilcoxon signed rank test was used to compare groups and Spearman's rank correlation to investigate the relation between variables.
Validation of Study System
Although heart rate variability is almost classically described by use of the ventricular ECG, we had the opportunity to compare ventricularly derived signals with those from the atrium. Fig 2⇓ shows that the same degree of variability applied whether the records came from the atrium (PP intervals) or ventricle (RR intervals). There was no evidence of modulation of heart rate variability by the atrioventricular node; that is, the SDs of the RR intervals or low- or high-frequency components of heart rate variability were not significantly different in the atrial recordings compared with the ventricular recordings either before or during stretch of the sinoatrial node.
Recordings Before Section of the Vagi or β-Blockade
We made recordings before and during sinoatrial node stretch in 15 experiments. The variation in beat-to-beat interval before and during sinoatrial node stretch is shown in Fig 3⇓, which plots a time series. Stretching the sinoatrial node decreased the mean RR interval (581 to 569 ms), but this did not reach statistical significance. Stretch of the sinoatrial node reduced the beat-to-beat excursion in heart rate. It significantly reduced the SD of the RR intervals (4.2 to 2.6 ms; P=.004).
There was less high-frequency variation apparent during sinoatrial node stretch (Fig 3⇑), and although the RR interval decreased slightly in the recording taken during stretch, the change was marginal, and it seems unlikely that it could account for the changes in heart rate variability seen. The frequency spectrum of RR intervals before and during sinoatrial node stretch is shown in Fig 4⇓. The two peaks at 1.9 and 3.75 Hz are both within the conventional definition of the high-frequency band of frequencies. Stretch of the sinoatrial node markedly attenuated the amplitude of both of these peaks. The amplitude of peaks in the frequency spectrum decreased during stretch of the sinoatrial node, with peaks at higher frequencies being affected more. This is seen in Fig 5A⇓, which shows the frequency spectrum before and during stretch of the sinoatrial node. Three peaks were seen in the frequency spectrum in 4 of the 15 experiments; these were seen in the Fourier spectrum as well as the autoregressive spectrum. As Fig 5⇓ shows, they were seen under all three different physiological conditions.
In the entire group, the low-frequency components were not significantly affected (6.6±4.8 ms2 for control versus 1.0±0.3 ms2 during stretch; P=.21), but the high-frequency components were significantly reduced (6.5±2.2 ms2 for control versus 1.4±0.3 ms2 during stretch; P=.003) (Fig 6A⇓; Table⇓).
Recordings After Vagal Section
Recordings were made before and after section of the vagus nerve in eight experiments. Compared with control (Fig 5A⇑), vagal section altered the frequency spectrum of the RR intervals, with a decrease in the amplitude of all frequencies (Fig 5B⇑). Stretching the sinoatrial node in eight experiments after section of the vagi did not change the mean RR interval (538 versus 540 ms, control versus stretch); however, the SD of the RR intervals was reduced (3.3±0.5 ms for control versus 2.1±0.1 ms during stretch; P=.02). Stretch of the sinoatrial node more markedly attenuated the amplitude of the peaks in the frequency spectrum (Fig 5B⇑). Stretch did not significantly affect the low-frequency components in the group as a whole (0.5±0.2 ms2 for control versus 0.4±0.1 ms2 during stretch; P=.49) (Fig 6B⇑; Table). Stretch of the sinoatrial node did significantly reduce the high-frequency components of the frequency spectrum (2.8±0.9 versus 1.2±0.3 ms2, control versus stretch; P=.05) (Fig 6B⇑; Table).
Recordings After Vagal Section and β-Blockade
In eight experiments after vagal section, recordings were made during β-blockade. Stretching the sinoatrial node did not affect the mean RR interval (583 versus 591 ms, control versus stretch). It significantly reduced the SD of the RR intervals (3.6±0.6 versus 2.3±0.4 ms, control versus stretch; P=.03). Sinoatrial node stretch reduced all the peaks in the frequency spectrum in this experiment (Fig 5C⇑). In the entire group, the low-frequency components were not significantly affected by sinoatrial node stretch (0.6±0.1 versus 0.8±0.3 ms2, control versus stretch; P=.67) (Fig 6C⇑; Table), but the high-frequency components were significantly reduced (10.6±3.5 versus 3.0±1.5 ms2, control versus stretch; P=.01) (Fig 6C⇑; Table).
Recordings During Stretch of the Right Atrial Wall Distant From the Sinoatrial Node Before Vagal Section or β-Blockade
In nine experiments, we stretched a part of the atrial wall more anterior than the sinoatrial node before any other intervention. This decreased the RR interval (605 versus 574 ms, control versus stretch; P=.015). Stretch significantly reduced the SD of the RR intervals (3.7 versus 2.7 ms, control versus stretch; P=.04). The low-frequency components were not significantly affected (0.9±0.3 versus 0.6±0.3 ms2, control versus stretch; P=.14). Similarly, the high-frequency components were not significantly reduced (1.9±0.3 versus 1.9±0.3 ms2, control versus stretch; P=.51).
Interaction of Indexes of Heart Rate Variability With Heart Rate
In the recordings made before section of the vagi or pharmacological autonomic blockade, there was a positive correlation between RR interval and the high-frequency components of heart rate variability before the application of stretch (r=.56; P=.03). However, during stretch of the sinoatrial node, there was no significant correlation of the high-frequency components with RR interval (r=.20; P=.47). There was no significant correlation between the low-frequency components and RR interval either before (r=.49; P=.06) or during stretch of the sinoatrial node (r=.34; P=.21).
Stretching the sinoatrial node decreased the SD of RR intervals and high-frequency components of heart rate variability. Low-frequency components of heart rate variability were not affected. After section of the vagi and administration of a β-blocker, stretch of the sinoatrial node still reduced high-frequency heart rate variability.
Sinoatrial Node Stretch and Interaction With Heart Rate
Increasing right atrial pressure, thus stretching the sinoatrial node, increases the heart rate.15 However, if the original rate is high, there may be a decrease in heart rate, and if the rise in right atrial pressure is slight, there will be a minimal rise in heart rate.15 Also, there will be a slow decrease to nearly the original rate.15 16 17 The reduction in heart rate with time after an increase in filling pressure is not seen when the sinoatrial node is rendered nonfunctional.15 In our experiments in the present study, we let the heart equilibrate over several minutes before recording; thus, the effect of stretch on the sinoatrial node to tend to reduce the heart rate toward the original value would have been fully operational in our experiments. Consequently, in keeping with previous reports, we found an increase in heart rate when stretch was applied locally to the atrium but not to the sinoatrial node. The heart rate increase was not thought to be responsible for the decrease in heart rate variability because there was no significant increase in heart rate when the sinoatrial node was stretched, and we saw the decrease in heart rate variability in experiments in which the heart rate did not increase.
It is possible that in stretching the sinoatrial node, we stretched nerve endings, which increases the frequency of impulses in afferent nerves to change the efferent activity of the autonomic nervous system. This would be responsible for the changes that we saw. Stretching an adjacent part of the atrium did not change heart rate variability. Of course, it may be that an adjacent part of the atrium does not have afferent nerve endings, although we did see a significant increase in heart rate when we stretched the atrium. Afferent sensory endings may be more concentrated in the sinoatrial node, however. Cyclic tensioning could still reflexly produce the high-frequency heart rate variability. Vagal section and pharmacological autonomic blockade did not entirely abolish heart rate variability, and tensioning of the node produced a final reduction in heart rate variability. Moreover, heart transplant recipients continue to show heart rate variability,10 although others18 have found little variability after transplantation. These observations strongly suggest that the tension-induced reduction in heart rate variability is not entirely a reflex mechanism. Other, more direct mechanisms are involved.
The observations support the existence, as in ventricular myocardium,19 20 21 22 of a type of mechanoelectric feedback in sinoatrial tissue in the intact heart. Mechanoelectric feedback operates when a mechanical change such as stretch changes membrane potential. In the present case, cyclic stretch directly changes the cyclic firing rate of the sinoatrial node. Stretching or tensioning the sinoatrial node increases overall mechanoelectric feedback but reduces the cyclic variation in length of the atrial myocardium and reduces cyclic mechanoelectric feedback. However, the link between mechanical stretch and heart rate variability could potentially be through the release of atrial natriuretic factor.23
Possible Clinical Relevance
In the failing heart, the atria dilate, and on the basis of our results, the degree of heart rate variability would be expected to decrease because of sustained stretch of the sinoatrial node. This is because the sustained stretch reduces the amplitude of the cyclic stretch. The variation in the effect of stretch on the sinoatrial node could potentially act as a confounding variable that masks the relation between decreased heart rate variability and hemodynamic or clinical parameters in those with cardiac failure.
Heart rate variability is reduced in congestive cardiac failure,12 24 25 26 but the pathophysiological mechanism for this is unclear.27 A relation between ejection fraction and heart rate variability has been found by some11 28 but not all investigators.3 In a study by Kienzle et al,29 clinical variables were not found to be related to heart rate variability, although in that same study, both high- and low-frequency components of heart rate variability were related to cardiac output. A complicated pattern of reduction of heart rate variability in patients with congestive cardiac failure has been found by others,30 with the total power of heart rate variability being reduced but with two distinct patterns of reduction. Similarly, in patients with severe mitral regurgitation, only ultralow-frequency components are related to ejection fraction of both the left and right ventricles.31 Thus, the overall evidence points to a reduction in heart rate variability in conditions in which the atria are subject to a constant increased load.
The prognostic information in heart rate variability in those with heart failure is less clear cut than in other conditions. Some researchers have found no useful prognostic information in heart rate variability analysis in congestive cardiac failure,28 while others32 have found useful information in measurements of both low- and high-frequencies. Conversely, it is clear that heart rate variability is reduced after myocardial infarction and that the reduction in low-1 4 33 and high-frequency4 34 components contains important prognostic information. Therefore, it is possible that in heart failure, there is a confounding variable that obscures the relationship and that the confounding variable is the degree of right atrial stretch.
General anesthetic is known to decrease heart rate variability35 and could have masked any effect of stretch. However, we still saw changes in heart rate variability after the sinoatrial node was stretched. Anesthetic may have altered the relative importance of reflex and mechanoelectric mechanisms by an effect on either. Another limitation is that the duration of our recordings was not long enough to allow us to comment on ultralow-frequency elements of heart rate variability. The animal model that we used is well established as a leading mammalian model of the human heart, and although much electrophysiological work has been done in both humans and the porcine heart, no significant difference has been reported. The acute stretch of the sinoatrial node was necessary to provide controls that were closely related in time to the intervention. However, in disease states in humans, the stretch is chronic; the use of chronic stretch may form the basis for further study but would necessitate the design of a stretching device that could be implanted. We also did not address the issue of stretch of the whole atrium, which is present in disease states. However, had we accomplished this, for instance, by inducing cardiac failure, many other physiological variables would have changed; using localized stretch of the sinoatrial node allowed us to dissect out physiological mechanisms to some extent. It could be argued that in the clinical situation in congestive cardiac failure, the central abnormality is that of left ventricular failure with stretch of the left atrium. However, there is also an element of fluid retention and right atrial dilatation, and it is not the left atrium but the sinoatrial node that governs heart rate and hence heart rate variability.
Stretch of the sinoatrial node reduces high-frequency components of heart rate variability. This may account in part for the reduced heart rate variability seen in clinical conditions in which the right atrium is dilated, such as congestive cardiac failure.
This work was supported by a grant from the British Heart Foundation.
- Received December 7, 1995.
- Revision received March 25, 1996.
- Accepted April 11, 1996.
- Copyright © 1996 by American Heart Association
Bigger JT Jr, Fleiss JL, Steinman RC, Rolnitzky LM, Kleiger RE, Rottman JN. Frequency domain measures of heart period variability and mortality after myocardial infarction. Circulation. 1992;85:164-171.
Farrell TG, Paul V, Cripps TR, Malik M, Bennett ED, Ward D, Camm AJ. Baroreflex sensitivity and electrophysiological correlates in patients after acute myocardial infarction. Circulation. 1991;83:945-952.
Fei L, Anderson MH, Katritsis D, Sneddon J, Statters DJ, Malik M, Camm AJ. Decreased heart rate variability in survivors of sudden cardiac death not associated with coronary artery disease. Br Heart J. 1994;71:16-21.
Schwartz PJ, La Rovere MT, Vanoli E. Autonomic nervous system and sudden cardiac death: experimental basis and clinical observations for post-myocardial infarction risk stratification. Circulation. 1992;85:I-77-I-91.
Bernardi L, Salvucci F, Suardi R, Solda 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.
Nolan J, Flapan AD, Capewell S, MacDonald TM, Neilson JM, Ewing DJ. Decreased cardiac parasympathetic activity in chronic heart failure and its relation to left ventricular function. Br Heart J. 1992;67:482-485.
Lab MJ. Contraction-excitation feedback in myocardium: physiological basis and clinical relevance. Circ Res. 1982;50:757-766.
Keatinge WR. The effect of increased filling pressure on rhythmicity and atrioventricular conduction in isolated hearts. J Physiol. 1959;149:193-208.
Pathak CL. Effect of stretch on formation and conduction of electrical impulses in the isolated sinoauricular chamber of the frog's heart. Am J Physiol. 1958;192:111-113.
Sands KE, Appel ML, Lilly LS, Schoen FJ, Mudge GH Jr, Cohen RJ. Power spectrum analysis of heart rate variability in human cardiac transplant recipients. Circulation. 1989;79:76-82.
Franz MR, Cima R, Wang D, Proffitt D, Kurtz R. Electrophysiological effects of myocardial stretch and mechanical determinants of stretch activated arrhythmias. Circulation. 1992;86:968-978.
Hansen DE. Mechanoelectrical feedback effects of altering preload, afterload, and ventricular shortening. Am J Physiol. 1993;264:H423-H432.
Taggart P, Sutton PM, Treasure T, Lab M, O'Brien W, Runnalls M, Swanton RH, Emanuel RW. Monophasic action potentials at discontinuation of cardiopulmonary bypass: evidence for contraction-excitation feedback in man. Circulation. 1988;77:1266-1275.
Levine JH, Guarnieri T, Kadish AH, White RI, Calkins H, Kan JS. Changes in myocardial repolarization in patients undergoing balloon valvuloplasty for congenital pulmonary stenosis: evidence for contraction-excitation feedback in humans. Circulation. 1988;77:70-77.
Butler GC, Senn BL, Floras JS. Influence of atrial natriuretic factor on heart rate variability in normal men. Am J Physiol. 1994;267:H500-H505.
Coumel P, Hermida JS, Wennerblom B, Leenhardt A, Maison-Blanche P, Cauchemez B. Heart rate variability in left ventricular hypertrophy and heart failure, and the effects of beta-blockade: a non-spectral analysis of heart rate variability in the frequency domain and in the time domain. Eur Heart J. 1991;12:412-422.
Van Hoogenhuyze D, Weinstein N, Martin GJ, Weiss JS, Schaad JW, Sahyouni XN, Fintel D, Remme WJ, Singer DH. Reproducibility and relation to mean heart rate of heart rate variability in normal subjects and in patients with congestive heart failure secondary to coronary artery disease. Am J Cardiol. 1991;68:1668-1676.
Malik M, Camm AJ. Heart rate variability and clinical cardiology. Br Heart J. 1994;71:3-6.
Fei L, Keeling PJ, Gill JS, Bashir Y, Statters DJ, Poloniecki J, McKenna WJ, Camm AJ. Heart rate variability and its relation to ventricular arrhythmias in congestive heart failure. Br Heart J. 1994;71:322-328.
Mortara A, La Rovere MT, Signorini MG, Pantaleo P, Pinna G, Martinelli L, Ceconi C, Cerutti S, Tavazzi L. Can power spectral analysis of heart rate variability identify a high risk subgroup of congestive heart failure patients with excessive sympathetic activation? A pilot study before and after heart transplantation. Br Heart J. 1994;71:422-430.
Stein KM, Borer JS, Hochreiter C, Okin PM, Herrold EM, Devereux RB, Kligfield P. Prognostic value and physiological correlates of heart rate variability in chronic severe mitral regurgitation. Circulation. 1993;88:127-135.
Cripps TR, Malik M, Farrell TG, Camm AJ. Prognostic value of reduced heart rate variability after myocardial infarction: clinical evaluation of a new analysis method. Br Heart J. 1991;65:14-19.
Algra A, Tijssen JG, Roelandt JR, Pool J, Lubsen J. Heart rate variability from 24-hour electrocardiography and the 2-year risk for sudden death. Circulation. 1993;88:180-185.
Galletly DC, Westenberg AM, Robinson BJ, Corfiatis T. Effect of halothane, isoflurane and fentanyl on spectral components of heart rate variability. Br J Anaesth. 1994;72:177-180.