Sympathovagal Balance

HomeCirculationVol. 96, No. 9Sympathovagal Balance Free AccessResearch ArticleDownload EPUBAboutView EPUBSections ToolsAdd to favoritesDownload citationsTrack citationsPermissions ShareShare onFacebookTwitterLinked InMendeleyReddit Jump toFree AccessResearch ArticleDownload EPUBSympathovagal Balance A Critical Appraisal Dwain L. Eckberg Dwain L. EckbergDwain L. Eckberg From the Departments of Medicine and Physiology, Hunter Holmes McGuire Department of Veterans Affairs Medical Center and Medical College of Virginia, Virginia Commonwealth University, Richmond. Search for more papers by this author Originally published4 Nov 1997https://doi.org/10.1161/01.CIR.96.9.3224Circulation. 1997;96:3224–3232Given the importance of the autonomic nervous system to cardiovascular health, it is not surprising that there is and has been great interest in measurements of human sympathetic and vagus nerve traffic as tools that might inform physiological and pathophysiological mechanisms. Pagani and coworkers1 advanced the provocative notion that the instantaneous balance between sympathetic and vagal nerve activities can be captured by a single number, obtained by dividing RR-interval spectral power centered at ≈0.1 Hz by spectral power centered at higher, primarily respiratory frequencies. This ratio, or sympathovagal balance, has been embraced with great enthusiasm2 because it offers new possibilities for understanding dynamic, critically important autonomic interrelations in humans by the use of totally noninvasive, unobtrusive means.3The broad bases for this mathematical treatment are as follows: (1) 0.1-Hz RR intervals are importantly mediated by fluctuations of sympathetic nerve activity; (2) higher-frequency RR-interval rhythms are mediated almost exclusively by fluctuations of vagal-cardiac nerve activity; and (3) physiological interventions tend to provoke reciprocal changes of sympathetic and vagal neural outflows. Sympathovagal balance, the ratio of these periodicities, is taken to reflect the balance between the opposing neural mechanisms. This review examines the physiological foundations of sympathovagal balance.The Method and Its MathematicsThe ECG is recorded with the subject in a steady state (when rhythms are stationary) for a period sufficiently long to define events occurring over frequencies of interest. RR-interval spectral power is calculated from this series of intervals with an autoregressive algorithm, which yields center frequencies and absolute power of component fluctuations, based on a model whose order (the number of parameters) is selected automatically to minimize Akaike’s information criterion statistic.4 (The statistical uncertainty and consequences of the automatic selection of the autoregressive model have not been defined fully; however, it is clear that the model order importantly determines both center frequency and magnitude of spectral peaks.5)Autoregressive analyses of short (≤5 minutes) RR-interval time series typically yield three peaks in very low (0 to 0.04 Hz; VLF), low (0.04 to 0.15 Hz; LF), and respiratory or high (0.15 to 0.40 Hz; HF) frequency ranges.6 Published examples (Fig 1 in Reference 11 and Figs 5c and 5d in Reference 66 ) suggest that in such short-term recordings, these three frequencies account for almost all spectral power (in the examples, 94.6%, 96.1%, and 98.5%). Therefore, in practice, power in normalized (also called relative or fractional) units (nu) is power centered at the frequency of interest (LF or HF) divided by total power less power at VLFs. Thus, LF nu=LF/(LF+HF) and HF nu=HF/(LF+HF). Sympathovagal balance is the ratio between LF and respiratory-frequency powers. Sympathovagal balance (in dimensionless units) is simply the ratio of absolute LF to absolute HF power, or LF/HF.Several consequences issue from the mathematics of sympathovagal balance. First, a change of either the numerator or the denominator must change the fraction. Thus, a reduction of LF with no change of HF will reduce the fraction, and a reduction of HF with no change of LF will increase the fraction. According to published precedents,1,3 the former change would be interpreted as a shift of sympathovagal balance toward vagal predominance and the latter as a shift of sympathovagal balance toward sympathetic predominance. Second, because calculations of normalized LF and HF powers involve merely rearrangement of the same LF and HF terms, a change of normalized LF power must necessarily be associated with a change of normalized HF power. For the same reason, relations among the different calculated values can be predicted simply on the basis of mathematics. For example, if LF and HF are distributed nearly normally, sympathovagal balance, plotted as a function of normalized LF power, will be almost exponential.The remaining discussion touches on some of the physiological and pathophysiological processes that contribute to sympathovagal balance. The discussion is organized as a series of postulates that I believe tacitly or explicitly underlie the concept.The NumeratorHuman cardiovascular rhythms occurring at ≈0.1 Hz are of great interest because they convey information regarding activity of the sympathetic nervous system. Healthy humans may have strong fluctuations of muscle sympathetic nerve activity about every 10 seconds,7 a period that is sufficiently long for vascular smooth muscle and sinoatrial node effectors to respond to released norepinephrine and to modulate arterial pressure and heart rate.8LF RR-Interval Rhythms Reflect Sympathetic Nerve Traffic to the HeartMalliani and coworkers3 published what may be the only power spectral analyses of sympathetic and vagus nerve traffic assumed but not proven to be destined for the heart. Fig 1 shows preganglionic thoracic sympathetic and vagus nerve discharges measured from one artificially ventilated decerebrate cat. My integration of the two LF spectral powers in this figure suggests that sympathetic spectral power is ≈15% greater than vagal spectral power.Kingwell and colleagues9 studied healthy supine subjects and found no significant correlation between myocardial norepinephrine spillover and absolute or relative 0.1-Hz RR-interval spectral power. Saul and coworkers10 also studied healthy supine subjects and found no significant correlation between baseline peroneal sympathetic nerve activity expressed as bursts/min (or antecubital vein plasma norepinephrine levels) and absolute or normalized LF heart rate spectral power.Hopf et al11 and Introna et al12 measured RR-interval spectral power before and after high spinal anesthesia. Sympathetic blockade in the supine position (but see responses to tilt below) did not alter absolute or relative 0.1-Hz RR-interval spectral power significantly.Pagani et al1 reported that propranolol 0.2 mg/kg IV does not reduce normalized 0.1-Hz RR-interval spectral power significantly. Moreover, Jokkel et al13 reported that short-term β-adrenergic blockade actually increases 0.1-Hz frequency RR-interval spectral power. Long-term β-adrenergic blockade reduces normalized 0.1-Hz RR-interval spectral power.1Low-dose cholinergic blocking drugs, such as atropine and scopolamine, paradoxically increase average RR intervals14 and LF RR-interval spectral power.15 Increased LF spectral power is unlikely to be due to increased sympathetic nerve traffic because atropine does not alter (muscle) sympathetic nerve activity or plasma norepinephrine levels.16 Conversely, large-dose atropine abolishes nearly all RR-interval spectral power in LF as well as HF ranges.17,18 The simplest interpretation of the effects of large-dose atropine is that it blocks sinoatrial responses to acetylcholine released by vagus nerve traffic; atropine does not alter heart rate in the absence of vagus nerve activity.19Thus, vagal contributions to baseline LF RR-interval fluctuations are great, and evidence that baseline LF RR-interval spectral power is related quantitatively to sympathetic-cardiac nerve traffic is nonexistent.Changes of 0.1-Hz RR-Interval Spectral Power Reflect Changes of Sympathetic-Cardiac Nerve TrafficTwo groups have sought correlations between muscle sympathetic nerve traffic and heart rate or RR-interval spectral power over a range of arterial pressures provoked by graded infusions of nitroprusside and phenylephrine. Saul and coworkers10 found significant correlations between sympathetic nerve activity, expressed as bursts/min, and normalized LF heart rate spectral power, but only in 4 of 10 subjects studied. Pagani and associates20 found no significant correlation between sympathetic nerve activity, expressed as bursts/min, and normalized LF RR-interval spectral power. However, Pagani did find a highly significant but probably weak correlation between sympathetic nerve activity, expressed as burst area per second, and normalized LF RR-interval spectral power.The studies by Saul et al10 and Pagani et al20 differ in several respects. First, Saul normalized sympathetic bursts; Pagani did not. Normalization is necessary because burst heights differ among individuals; if sympathetic activity is not normalized, subjects with large sympathetic bursts will dominate average results. Second, subjects in the study by Saul et al controlled their breathing frequencies; those in the study by Pagani et al did not. As discussed below, breathing frequency profoundly influences spectral power measurements. Third, Saul reported individual regressions; Pagani did not. (It cannot be determined from the article by Pagani et al if there is a significant correlation between muscle sympathetic burst area per second and normalized LF RR-interval spectral power.) Rather, Pagani used a nonparametric regression analysis (Theil21) of all his data, both within and across subjects. This may have been inappropriate, because the Theil method requires that data be stochastically independent. Repeated measures (seven in each subject) violate this condition.Koh and coworkers18 also measured RR-interval spectral power during infusions of nitroprusside and phenylephrine. Fig 2 shows RR-interval spectral power of healthy subjects and patients with complete high cervical spinal cord injuries during saline infusion and two or three levels of nitroprusside and phenylephrine. Arterial pressures (not shown) are highest toward the back and left and lowest toward the front and right of both graphs. Changes of spectral power in the 0.05- to 0.15-Hz range (enlarged in the insets) were different in the two groups. Healthy subjects (right) did not alter their LF RR-interval spectral power systematically. Tetraplegic patients, on the other hand, altered their LF RR-interval spectral power systematically but in direct, not inverse, proportion to arterial pressure changes.Koh et al18 concluded that vagal mechanisms contribute importantly to baroreflex-mediated 0.1-Hz RR-interval spectral power in both healthy subjects and tetraplegic patients. In healthy subjects, LF spectral power does not change systematically, because arterial pressure changes provoke reciprocal changes of sympathetic and vagal nerve fluctuations, which cancel one another. Conversely, in tetraplegic patients, who lack the ability to modulate sympathetic nerve activity according to changes of baroreceptor input, LF RR-interval spectral power changes in proportion to arterial pressure. Koh’s conclusion that baseline 0.1-Hz RR-interval rhythms importantly reflect fluctuations of vagus nerve traffic was supported by responses of both healthy subjects and tetraplegic patients to large-dose atropine, which nearly abolished LF RR-interval spectral power.Thus, most evidence refutes the notion that LF RR-interval spectral power tracks baroreflex-mediated changes of sympathetic nerve activity; the published documentation of a weak association20 may have resulted from use of inappropriate statistical methods.The DenominatorRespiratory-Frequency RR-Interval Fluctuations Reflect Vagus Nerve Traffic to the HeartThe denominator in the sympathovagal balance equation comprises higher, primarily respiratory-frequency RR-interval spectral power. HF RR-interval fluctuations are thought to be mediated by fluctuations of vagal activity because in anesthetized but spontaneously breathing dogs, they are proportional to vagus nerve traffic,22 and in humans, they are nearly abolished by large-dose atropine.14,17,23 Eckberg,24 Fouad et al,23 and Hayano et al25 promoted respiratory-frequency RR-interval fluctuations as noninvasive indexes of vagal-cardiac nerve traffic in humans. Kollai and Mizsei26 qualified this recommendation. They compared respiratory peak minus valley RR-interval changes with RR-interval shortening provoked by large-dose atropine after β-adrenergic blockade. Although their study supported the use of respiratory RR-interval fluctuations as indexes of vagal-cardiac nerve traffic, it showed that this measure is not a perfect index.Thus, baseline respiratory-frequency RR-interval fluctuations are related significantly but imperfectly to the level of human vagal-cardiac nerve traffic.Changes of HF RR-Interval Fluctuations Reflect Changes of Vagus Nerve Traffic to the HeartStudies by Eckberg et al27 and Goldberger et al28 in humans extended observations made much earlier by Anrep et al in dogs,29 which challenged the notion that respiratory-frequency RR-interval fluctuations are proportional to vagal-cardiac nerve activity during changes of arterial pressure. Fig 3, adapted from the study by Eckberg et al,27 shows average diastolic arterial pressure (range, 70 to 86 mm Hg), RR intervals, and peak minus valley RR-interval changes during graded infusions of nitroprusside and phenylephrine in healthy supine subjects breathing at a fixed frequency of 0.2 Hz. Over this arterial pressure range, RR intervals but not peak minus valley RR-interval changes were proportional to diastolic arterial pressure. (In the studies by both Eckberg et al and Anrep et al, however, larger arterial pressure reductions provoked parallel reductions of RR intervals and peak minus valley RR-interval changes.)There are also circumstances in which changes of HF RR-interval fluctuations may not reflect changes in vagal-cardiac nerve traffic at all. Brown and coworkers2 measured RR-interval spectral power in healthy supine subjects who breathed to a constant depth at frequencies between 0.4 and 0.1 Hz (24 to 6 breaths/min). Fig 4 indicates that maximum spectral power was ≈10-fold greater at the slowest breathing frequency (to the left and back) than at the fastest breathing frequency (to the right and front). Notwithstanding the huge differences in spectral power at different breathing frequencies, mean RR intervals (right) were nearly constant across all breathing frequencies. In that study, the constancy of RR intervals was taken as inferential evidence for the constancy of vagal-cardiac nerve traffic.30Brown’s data document a pivotal, largely ignored2 role for respiration as a determinant of HF RR-interval spectral power. Grossman et al31 studied the usefulness of respiratory-frequency RR-interval fluctuations as indexes of tonic vagal-cardiac neural outflow, as reflected by mean RR intervals, in healthy subjects after β-adrenergic blockade. They documented a reasonable correspondence between the two measures during interventions that raised or lowered tonic vagal-cardiac outflow, but only when respiration was controlled (and therefore accounted for). When respiration was not controlled, respiratory-frequency RR-interval fluctuations bore no significant relation to tonic vagal-cardiac nerve activity.Saul and coworkers32 made a convincing case that there is no need to postulate changes of vagal-cardiac nerve traffic to explain the RR-interval variability shown in Fig 4 (left); such variability can be explained simply on the basis of the kinetics of sinoatrial node responses to acetylcholine. During slow breathing, responses to acetylcholine are expressed more fully than during rapid breathing, when responses to one mainly expiratory bolus of acetylcholine merge with responses to the next, and fluctuations of RR intervals are minimized. Data shown in Fig 4 (left) may apply to the numerator as well as the denominator of the sympathovagal equation; during periods of slow breathing (which healthy people have33), LF and respiratory-frequency center frequencies may be the same.Thus, moderate changes of arterial pressure, which alter vagal-cardiac nerve activity, do not change HF RR-interval fluctuations, and changes of breathing frequency and depth, which profoundly alter HF RR-interval fluctuations, may not change vagal-cardiac nerve activity at all.The QuotientThe foregoing discussion treats the numerator and denominator of the sympathovagal fraction separately, with the unstated belief that these measures should correspond with actual levels of nerve traffic. I now discuss the quotient itself.Sympathetic and Vagal Nerve Activities Change ReciprocallyImmersion of the face in cold water, a “diving reflex,” provokes simultaneous increases of muscle sympathetic nerve activity34 and bradycardia,35 which can be prevented by large-dose atropine. Reductions of carbon dioxide chemoreceptor stimulation appear to lead to parallel reductions of sympathetic and vagal nerve activities (D.L.E., MD, et al, unpublished data, 1997).There may be no physiological intervention more likely to change sympathetic and vagal outflows reciprocally than the arterial baroreflex. However, examination of baroreflex physiology exposes layers of complexity that qualify the notion that changes of baroreceptor activity trigger simple reciprocal changes of sympathetic and vagal nerve outflow.All evidence points to the fact that humans lie above threshold on their arterial pressure-sympathetic and -vagal response relations. Fig 5 shows average changes of muscle sympathetic nerve and RR-interval (vagal) responses to abrupt increases and decreases of carotid baroreceptor input provoked by changes of external neck pressure.36 The resting position of the subjects studied on their arterial pressure–sympathetic response relations was eccentric, being close to threshold. The consequence of this is that although small increases of arterial pressure provoked reciprocal changes of sympathetic and vagal neural outflows, larger increases of pressure provoked exclusively increases of vagal-cardiac nerve activity. Fig 5 (right) indicates that these subjects lie in the middle their vagally mediated arterial pressure–RR-interval response relations. However, other studies indicate that some healthy subjects lie in the threshold37 or saturation38 regions of this relation.Thus, some physiological interventions provoke parallel, not reciprocal, changes of vagal and sympathetic nerve activity, and other interventions, such as baroreceptor stimulation, provoke reciprocal changes, but only over a very limited range of arterial pressure.Measurements of Sympathovagal Balance Are Applicable to Patients as Well as Healthy SubjectsThere are at least two problems with the use of sympathovagal balance in patients with heart failure. First, responses of heart failure patients to baroreflex forcings may be qualitatively different from those of healthy people. Heart failure patients may not increase their plasma norepinephrine levels during upright tilt,39 and they may reduce rather than increase their plasma norepinephrine levels39 and muscle sympathetic nerve traffic40 during pharmacological arterial pressure reductions. Second, although heart failure patients may have extraordinarily high levels of muscle sympathetic nerve activity,16,41 plasma norepinephrine,42 and cardiac norepinephrine spillover43,44 and low RR-interval SDs (Fig 6),16 their calculated sympathovagal balance is extremely low.3,45There are other patients in whom the concept of sympathovagal balance may not apply. Although hypertensive patients may have average breathing rates comparable to those of normotensive people, they are much more likely to have episodic slow breathing and even apnea.46 Tetraplegic patients breathe more rapidly than healthy people.47 In such patient groups, profound differences in sympathovagal balance may arise simply on the basis of systematic differences in breathing.Thus, measures of sympathovagal balance are not valid in heart failure patients and may not be valid in hypertensive or sleep apnea patients.Changes of Sympathovagal Balance Reflect Changes of the Ratio of Sympathetic to Vagus Nerve Traffic to the HeartThe intervention that makes the most compelling case for sympathovagal balance is upright tilt. Iwase and colleagues48 showed that muscle sympathetic nerve activity increases linearly with the sine of the angle during passive head-up tilt. Hopf and coworkers11 reported that epidural anesthesia reduces sympathovagal balance in the upright tilt position. Montano and coworkers49 and Mukai and Hayano50 extended this observation by showing that sympathovagal balance increases linearly with the angle of tilt. However, both groups showed that this increase results from progressive reductions of HF RR-interval spectral power without progressive reciprocal increases of LF RR-interval spectral power.Burke and coworkers51 showed clearly that in some individuals, ≈0.1-Hz fluctuations of muscle sympathetic nerve activity increase in the upright position. However, they also showed that some subjects reduce their muscle sympathetic nerve traffic in the upright position. Wallin and coworkers52 showed that although upright tilt increases myocardial norepinephrine spillover (and presumably sympathetic-cardiac nerve activity), it does not increase 0.1-Hz RR-interval spectral power.Exercise is another physiological perturbation that affects autonomic neural outflow profoundly. During exercise, muscle sympathetic nerve activity53 and myocardial norepinephrine spillover43 increase and vagal-cardiac nerve activity (as reflected by instantaneous RR-interval shortening) decreases.54 Sympathovagal balance, however, decreases during intense exercise55 in proportion to the intensity of exercise.56 Calculations of sympathovagal balance may give specious results during light exercise as well; two groups57,58 showed that the RR-interval shortening that occurs during the transition from rest to light exercise is due almost exclusively to vagal withdrawal. Therefore, the apparent shift from vagal to sympathetic predominance at the beginning of exercise55 probably reflects vagal withdrawal without concurrent sympathetic excitation.Thus, neither upright tilt (arguably the flagship of sympathovagal balance) nor light or heavy exercise provokes the reciprocal changes of sympathetic and vagal nerve traffic predicted by calculations of sympathovagal balance.DiscussionThis review, which examines the physiological foundations for the concept of sympathovagal balance, was undertaken because of concern that this new measure has entered the mainstream of cardiovascular research without having been critically vetted. My review in no way disputes the value of heart rate variability in stratifying risk in patients with cardiovascular diseases59–61 or in better understanding autonomic mechanisms.62 Rather, it focuses on the manipulation of heart rate variability to quantify the putative balance that exists between sympathetic and vagal cardiovascular neural outflows.The concept of sympathovagal balance has been promoted in part on philosophical grounds. For example, sympathovagal balance has been compared with the action of the arm. Fine control of arm movements is made possible by simultaneous reciprocal changes of tension in flexor and extensor muscle groups.3 Although my review is of physiology, I also have philosophical problems with the concept. I am aware of no compelling physiological requirement that the levels of sympathetic and vagal nerve fluctuations be balanced. I am aware of no physiological evidence that fluctuations of sympathetic and vagus nerve activities “constantly interact.”6 To be sure, oscillators influence one another in complex ways, and the two branches of the autonomic nervous system are “controlled and balanced.” However, such control probably reflects the fact that sympathetic and vagal motoneurons respond to interrelated neural influences. My sense is that the construct of sympathovagal balance imposes attributes on physiological regulatory mechanisms that they do not possess.I also have major concerns regarding the language used to support the concept of sympathovagal balance. Qualifiers such as “mainly”1 and “mostly,”63 as in “0.1 Hz RR-interval rhythms are mainly sympathetic,” imply >50%. Is sympathovagal balance a robust tool if sympathetic contributions to 0.1-Hz spectral power are only marginally greater than vagal contributions, as Fig 1 indicates? The qualifiers “marker”3 and “quantitative markers,”64 as in “>0.1-Hz RR-interval spectral power is a marker of sympathetic traffic,” are vague. Can there be any doubt that sympathetic neural mechanisms contribute to LF rhythms? Such contributions could arise simply on the basis that sympathetically mediated65 fluctuations of arterial pressure provoke vagally mediated fluctuations of RR intervals.66 Does the fact that there is a sympathetic contribution mean that LF rhythms are quantitative probes for sympathetic traffic?The word “drive,” as in “..normalized LF [low-frequency], an indicator of sympathetic drive,”64 and “..maneuvers enhancing vagal and sympathetic drive,”3 connotes nerve traffic rather than variability of nerve traffic. Similarly, the word “activity,” as in “maneuvers enhancing sympathetic and vagal activities,” connotes nerve traffic rather than variability of nerve traffic. The word “predominance,” as in “vagal predominance,”1 is vague; in such a context, does predominance mean that vagal fluctuations are larger than sympathetic fluctuations or that absolute levels of vagus nerve traffic are larger than absolute levels of sympathetic nerve traffic? The recurring3,67 assertion that “..a reciprocal relation exists between these two rhythms [0.1 Hz and respiratory] that is similar to that characterizing the sympathovagal balance” is tautological. The word “modulation,” as in “enhance the vagal modulation of the heart rate,”3 is binary, not graded; it is something that is present or not. These concerns about language reflect a larger concern that because the words are so familiar that the concepts engendered by these words may also be regarded as familiar. As this review suggests, I find many of the concepts foreign. In the end, however, it is not philosophy or language that must justify sympathovagal balance; it is physiology, which I have reviewed here.The NumeratorIt is inarguable that fluctuations of vagus nerve traffic contribute to 0.1-Hz RR-interval spectral power; Fig 1 documents a major vagal contribution. (It is curious that the near disappearance of respiratory-frequency RR-interval spectral power after large-dose atropine is taken as evidence that this component is vagally mediated,1 but the near disappearance of 0.1-Hz RR-interval spectral power is ascribed to different mechanisms.)Myocardial norepinephrine spillover is the current gold standard for sympathetic traffic to the human heart; this measure correlates well with muscle sympathetic nerve traffic and plasma norepinephrine levels, at rest and during isometric exercise and mental arithmetic.9,52 Normalized 0.1-Hz RR-interval spectral power bears no significant relation to baseline levels or to baroreflex-mediated changes of myocardial norepinephrine spillover,9 muscle sympathetic nerve activity,10 or antecubital vein plasma norepinephrine concentrations.10The DenominatorRR-interval shortening after large-dose atropine is given to β-adrenergically blocked subjects is the current gold standard for vagal traffic to the human heart. Normalized respiratory-frequency RR-interval spectral power at rest bears no significant relation to RR-interval shortening after large-dose atropine.25Although there is a quantitative relation between respiratory-frequency RR-interval spectral power and vagal-cardiac nerve activity,22,26 moderate changes of vagal-cardiac nerve activity from baseline, such as those that accompany increases or decreases of arterial pressure, do not alter respiratory-frequency RR-interval spectral power (Fig 3). Conversely, huge changes of HF RR-interval spectral power provoked by changes of breathing (which in most sympathovagal balance research is neither controlled nor measured2) may not reflect changes of vagal-cardiac nerve traffic at all. It is unclear how sympathovagal balance should be calculated or interpreted when systematic changes of breathing frequency occur during the period of measurement, such as those reported in patients with sleep apnea.68These considerations betray what I consider a fundamental misconception of what RR-interval rhythms represent. At arterial pressures moderately above and below baseline levels, HF RR-interval rhythms reflect primarily respiratory gating of vagal-cardiac motoneuron responsiveness to baroreceptor and other sensory receptor stimulation.69 At baseline arterial pressures, variations of RR-interval fluctuations associated with variations of breathing frequency or depth reflect primarily the kinetics of sinoatrial node responses to acetylcholine.32 Neither mechanism requires the fine-tuning of rhythmicity that is invoked to explain sympathovagal balance.The QuotientThe intuitively attractive notion that changes of sympathetic and vagal nerve activities occur reciprocally does not occur with all interventions, including immersion of the face in iced water34,35 and carbon dioxide chemoreceptor stimulation (D.L.E., MD, et al, unpublished data, 1997). During these interventions, both neural outflows appear to change in parallel. More im