Shortening of the QT interval of the EKG is associated primarily with increased ventricular contractility rather than heart rate

The objective of this study was to gather direct evidence on whether the duration of the QT interval relates primarily to heart rate or to ventricular contractility. The electrocardiographic and cardiodynamic consequences of electrical stimulation (15 V, 5 ms, 10Hz) of various intrathoracic sympathetic efferent neuronal structures were studied in 10 anesthetized mongrel dogs. Stimulation of efferent sympathetic axons in the right intraganglionic nerve, which innervates the sinoatrial node, induced tachycardia (110±5 - 133±6 bpm; p<0.01) without significantly altering right or left ventricular intramyocardial ventricular chamber pressures. The QT interval, as determined by leads I, II and III of the EKG and a transthoracic lead, was not affected by this intervention (310±8 - 302±ms). Increasing heart rate to a similar degree (111±3 - 131±3 bpm) by right atrial pacing did not induce changes in the QT interval. When right (23±3 - 49±8 mm Hg; p<0.01) and left (81±10 - 127±19 mm Hg; p<0.01) ventricular forces were augmented without concomitant increases in heart rate by stimulating efferent sympathetic axons in the left caudal pole cardiopulmonary nerve the QT interval shortened (322±11 - 290±12 ms; p<0.01). Only when an efferent sympathetic nerve, that contains fibers destined for both the sinoatrial node and the ventricles was stimulated did both heart rate and ventricular contractility augment and QT shorten (318±10 - 290±11 ms; p<0.01). These results indicate that QT interval shortening induced by efferent sympathetic neurons is primarily associated with augmentation of ventricular contractility rather than heart rate.     

Heart rate—QT relationships during baroreceptor stimulation with diminished autonomic influence on the ventricles

The shortening of the QT interval of the electrocardiogram coincident with acceleration of heart rate and vice versa has been accepted for many years as evidence that the action potential duration and hence QT are necessarily dependant on heart rate. Exceptions to this rule have been attributed to the intervention of counteracting autonomic effects. In order to test this assumption, 26 conscious dogs divided into three groups were tested during baroreceptor stimulation by a bolus injection of phenylephrine. Seventeen dogs had been used in earlier studies in which they had undergone an experimental anterior myocardial infarction with apparent full recovery. A group of those dogs underwent beta-adrenergic blockade by intravenous atenolol 30 min prior to the baroreceptor activation. To test the intactness of the baroreceptor responses in the previously infarcted dogs, a third group of nine dogs that had had no prior myocardial infarction was included. All dogs were adapted to the laboratory environment and were not sedated during experiments. Simultaneous recordings of RR, QT interval, and phasic arterial pressure were made in all dogs before and during baroreceptor stimulation. In the normal group, and the previously infarcted group that received no atenolol, baroreceptor stimulation elicited a small (8/msec), but significant prolongation of the QT associated with a nearly 50% reduction in heart rate. The QT interval of the atenolol-treated dogs, although significantly more prolonged before stimulation, remained unchanged during the reflex. The data indicate that withdrawal of ventricular sympathetic tone may prolong the QT interval, thereby confirming the role of sympathetic innervation in the control of QT. This inference was strengthened by the complete absence of change of QT after atenolol block and the absence of any detectable influence of vagal activity on ventricular repolarization during the reflex. Therefore, in what appeared to be the absence of evidence of autonomic influence, no inverse relationship between heart rate and QT was observed.     

Heart rate—QT interval relationship during postural change and exercise

The QT interval of the electrocardiogram, although often coordinated in an inverse relationship to heart rate, appears from the existing evidence to be regulated by mechanisms separate from those that govern heart rate. The purpose of this inquiry was to explore further the relationship of the two and to test other factors that may contribute to the regulation of QT. Heart rate and QT duration were measured in healthy human subjects during postural change and exercise. The data showed that, while coordinated in an inverse relationship under some circumstances, under others heart rate and QT were discordant, leading to the inference that there is no fixed linkage between the two. Previous work by the authors and other published studies have suggested that ventricular contractility may have more predictable association with QT shortening than do increases in heart rate. Observations of force of ventricular contraction (HI and IJ velocity) as reflected in ballisto-cardiographic tracings were made using each of 12 human subjects as the authors’ own control. The data revealed a highly significant correlation between the velocity measures and QT duration (p<0.01) but not heart rate. These findings led to the conclusion that there is clearly not a fixed relationship between heart rate and QT interval and to the hypothesis that QT may more closely reflect the degree of cardiac contractility.     

Idioventricular low frequency oscillation in QT interval responds univocally to RR confusing kinds of mental stress.

Seventeen male subjects, aged nineteen to twenty, went through a protocol including, while supine, relaxation at rest (10 min) and mental stress (MS) by a Kraepelin (arithmetic) test (5 min), as part of a larger study. With a polygraphic analog recording set-up we also collected a 1 ms - digital facsimile of a lead II-like thoracic ECG with maximum T-wave (Codas, Dataq Instr). Twenty-three stress responses were assigned to three classes according to known cardiotacho-, plethysmo-, and pneumo-graphic marks of "concentrated attention mainly," "emotion," or still "high emotion." During each setting the most stationary 3 min RR epoch in cardiotachogram was selected for joint RR & QT beat-by-beat variability study. RR and QT intervals were detected using a published algorithm. Conventional RR and QT Fourier autospectra were computed, while using RR*QT mean square coherence spectrum we detached the RR-independent, idioventricular (IV) fraction of QT low frequency (LF: 0.04-0.15 Hz) power of variability (IV QT-LF). IV QT-LF responded consistently to varieties of mental stress that confuse RR-LF or let QT-LF unchanged, best witnessing the cortically-issued ventricular adrenergic strain. Indeed, while emotion propels the same way all spectral variables above, concentrated attention increased (Wilcoxon) significantly IV QT-LF only (0.54-0.80 ms2) and decreased RR-LF (715-463 ms2). Findings hold promise of a non-invasive, high resolution Holter based monitoring of sympathetic status of myocardium, robust vis-à-vis of confusion caused by the autonomic interplay at sino-atrial node.     

Heart rate-Qt interval relationship during postural change and exercise. A possible connection to cardiac contractility

The QT interval of the electrocardiogram, although often coordinated in an inverse relationship to heart rate, appears from the existing evidence to be regulated by mechanisms separate from those that govern heart rate. The purpose of this inquiry was to explore further the relationship of the two and to test other factors that may contribute to the regulation of QT. Heart rate and QT duration were measured in healthy human subjects during postural change and exercise. The data showed that, while coordinated in an inverse relationship under some circumstances, under others heart rate and QT were discordant, leading to the inference that there is no fixed linkage between the two. Previous work by the authors and other published studies have suggested that ventricular contractility may have more predictable association with QT shortening than do increases in heart rate. Observations of force of ventricular contraction (HI and IJ velocity) as reflected in ballisto-cardiographic tracings were made using each of 12 human subjects as the authors' own control. The data revealed a highly significant correlation between the velocity measures and QT duration (p less than 0.01) but not heart rate. These findings led to the conclusion that there is clearly not a fixed relationship between heart rate and QT interval and to the hypothesis that QT may more closely reflect the degree of cardiac contractility.     

Heart rate--QT relationships during baroreceptor stimulation with diminished autonomic influence on the ventricles. Ventricular autonomic tone and QT interval.

The shortening of the QT interval of the electrocardiogram coincident with acceleration of heart rate and vice versa has been accepted for many years as evidence that the action potential duration and hence QT are necessarily dependent on heart rate. Exceptions to this rule have been attributed to the intervention of counteracting autonomic effects. In order to test this assumption, 26 conscious dogs divided into three groups were tested during baroreceptor stimulation by a bolus injection of phenylephrine. Seventeen dogs had been used in earlier studies in which they had undergone an experimental anterior myocardial infarction with apparent full recovery. A group of those dogs underwent beta-adrenergic blockade by intravenous atenolol 30 min prior to the baroreceptor activation. To test the intactness of the baroreceptor responses in the previously infarcted dogs, a third group of nine dogs that had had no prior myocardial infarction was included. All dogs were adapted to the laboratory environment and were not sedated during experiments. Simultaneous recordings of RR, QT interval, and phasic arterial pressure were made in all dogs before and during baroreceptor stimulation. In the normal group, and the previously infarcted group that received no atenolol, baroreceptor stimulation elicited a small (8/msec), but significant prolongation of the QT associated with a nearly 50% reduction in heart rate. The QT interval of the atenolol-treated dogs, although significantly more prolonged before stimulation, remained unchanged during the reflex. The data indicate that withdrawal of ventricular sympathetic tone may prolong the QT interval, thereby confirming the role of sympathetic innervation in the control of QT.(ABSTRACT TRUNCATED AT 250 WORDS)     

Idioventricular low frequency oscillation in QT interval responds univocally to RR confusing kinds of mental stress

Seventeen male subjects, aged nineteen to twenty, went through a protocol including, while supine, relaxation at rest (10 min) and mental stress (MS) by a Kraepelin (arithmetic) test (5 min), as part of a larger study. With a polygraphic analog recording set-up we also collected a 1 ms - digital facsimile of a lead II-like thoracic ECG with maximum T-wave (Codas, Dataq Instr). Twenty-three stress responses were assigned to three classes according to known cardiotacho-, plethysmo-, and pneumo-graphic marks of “concentrated attention mainly,” “emotion,” or still “high emotion.” During each setting the most stationary 3 min RR epoch in cardiotachogram was selected for joint RR & QT beat-by-beat variability study. RR and QT intervals were detected using a published algorithm. Conventional RR and QT Fourier autospectra were computed while using RR^*QT mean square coherence spectrum we detached the RR-independent, idioventricular (IV) fraction of QT low frequency (LF: 0.04–0.15 Hz) power of variability (IV QT-LF). IV QT-LF responded consistently to varieties of mental stress that confuse RR-LF or let QT-LF unchanged, best witnessing the cortically-issued ventricular adrenergic strain. Indeed, while emotion propels the same way all spectral variables above, concentrated attention increased (Wilcoxon) significantly IV QT-LF only (0.54–0.80 ms²) and decreased RR-LF (715–465 ms²). Findings hold promise of a non-invasive, high resolution Holter based monitoring of sympathetic status of myocardium, robust vis-à-vis of confusion caused by the autonomic interplay at sino-atrial node.     

Spectral indices of cardiovascular adaptations to short-term simulated microgravity exposure

We investigated the effects of exposure to microgravity on the baseline autonomic balance in cardiovascular regulation using spectral analysis of cardiovascular variables measured during supine rest. Heart rate, arterial pressure, radial flow, thoracic fluid impedance and central venous pressure were recorded from nine volunteers before and after simulated microgravity, produced by 20 hours of 6° head down bedrest plus furosemide. Spectral powers increased after simulated microgravity in the low frequency region (centered at about 0.03 Hz) in arterial pressure, heart rate and radial flow, and decreased in the respiratory frequency region (centered at about 0.25 Hz) in heart rate. Reduced heart rate power in the respiratory frequency region indicates reduced parasympathetic influence on the heart. A concurrent increase in the low frequency power in arterial pressure, heart rate, and radial flow indicates increased sympathetic influence. These results suggest that the baseline autonomic balance in cardiovascular regulation is shifted towards increased sympathetic and decreased parasympathetic influence after exposure to short-term simulated microgravity.     

QT & RR variability spots the earliest autonomic deregulation in diabetes. Fading of vagal sino-atrial drive but not of sympathetic ventricular responsiveness to life challenges.

27 consecutive insulin-dependent diabetic patients (pts), under 50 years, with blood glucose controlled within normal limits and no significant or multiple cardiovascular/neurological complications in the lights of clinical tests, went through a protocol as follows: laiddown at relaxed rest for 10 min, then stood-up quietly for 7 min, and finally experienced a stress-interview for 10 min while supine. A thoracic ECG lead was digitized at I ms (Codas, Dataq Instr.), RR and QT intervals were software-detected, resampled at 500 ms, and Fourier-transformed over 3 min epochs to get auto-or cross-spectra. RR-by-QT mean square coherence detached the RR-independent fraction of QT low fequency (LF) spectral power, called idioventricular QT-LF. We detected autonomic impairment of three types (discriminant score = 92.31%), presumably differentiated upon the locus of lesion, using RR's basal variance and mean RR shortening when standing as follows: (I) RR shortening > 200 ms in 10 pts; (II) normal RR shortening but no RR variance in 4 pts; (III) stiff RR around 600 ms and no RR variance in 2 pts. The above pts have been excluded from further analysis. The remaining 11 pts with no such impairments (5M and 6F, 36.4 y +/- 4.4 SD, history of 6.0 y +/- 5.2) have been compared with 11 normal subjects in an age and gender-paired control group in two steps. Step 1: Preliminary MANOVA/ANOVA showed significant effects on the ensemble of spectral variables of every single factor (status: normal or patient group; intervention; gender) with no significant factor interactions. Significant effects of intervention or status on main RR spectral variables and on a few QT spectral variables were also documented. Step 2: Non-parametric tests showed that diabetics had (mildly to moderately) shorter mean RR, while their RR-LF was always significantly lower than those found in normals--a difference propagated to QT-LF but not to idioventricular QT-LF. In the intra-group there were similar responses to interventions except stress with respect to mean RR. Consistent reduction in RR-LF under moderate or no change in mean RR suggests vagal down- regulation that, judging by idioventricular QT-LF showing, goes perhaps before a similar process with sympathetic control of ventricles. This phase delay may introduce an early arrhythmic risk worth dealing with in secondary prevention.     

Operant conditioning of large magnitude, 12-hour duration,heart rate elevations in the baboon

Two baboons were prepared with arterial and venous catheters and their heart rate and blood pressure were monitored continuously thereafter. Following a 2 to 3 week interval during which baseline cardiovascular levels were determined, the animals were exposed to daily 12 hr conditioning sessions (alternating with 12-hr “rest” or “Conditioning Off’ sessions) during which food reward and shockavoidance were programmed as contingent consequences of pre-specified increases in heart rate. Initially, the criterion heart rate was set at 10–15 bpm above the animal’s pre-experimental resting baseline level, with progressive increases programmed to occur at a rate approximating 7 bpm per week over a period of 8–10 weeks. Within this 2–3 month interval, heart rate doubled, reaching levels maintained above 160 bpm for more than 95 per cent of each daily 12-hour “Conditioning On” period. Propranolol selectively eliminated the conditioned heart rate increase but not the blood pressure elevation, indicating the contribution of sympathetic nervous activity to these operantly conditioned cardiovascular changes.     

Effectiveness of atrial pacing inMacaca Mulatta as a function of pacing pulse intensity: Modification by atropine and propranolol

The functional relationship between heart rate, under atrial pacing, and the intensity of pacing pulses, was determined for intensities ranging from well below the complete heart rate capture threshold to suprathreshold intensities. A separate capture-intensity function was determined for each of the pacing frequencies 180, 240, 300 and 360 pulses per min, and then used to evaluate the effects of propranolol hydrochloride and atropine sulfate on atrial pacing. Propranolol resulted in an increase in the amount of current (mA) required to achieve a given degree of cardiac capture, whereas atropine resulted in a decrease. These shifts in the capture-intensity functions along the intensity axis were possibly caused by drug antagonism of betaadrenergic and cholinergic receptors, respectively, although the effects might also have been due to modifications of the electrical responsivity of the cardiac muscle, independently of the neural blockade.     

Effect of Transcendental Meditation versus resting on physiological and subjective arousal

On four successive days, 10 highly trained and experienced meditators were asked to relax for 5 minutes, meditate for 20 minutes, and then relax for 5 minutes. In contrast, 10 other subjects who had no training or experience with meditation were asked to relax for 5 minutes, rest for 20 minutes, and then relax for 5 minutes. Physiological arousal (heart rate, skin resistance, respiration rate, systolic blood pressure, diastolic blood pressure) and subjective arousal (cognitive, somatic, relaxation) were measured throughout the experiment. Results indicated that (a) prior to meditating or resting, meditators tended to have higher heart rates and diastolic blood pressure than did nonmeditators, (b) meditation was associated with generally reduced arousal, but (c) while meditating, meditators did not evidence lower levels of arousal than nonmeditators did while resting. This investigation employed controls, which were not used in previous investigations, and the results place qualifications on previously reported results. The results have implications for the study of personality functioning, stress management, and psychotherapy.     

Individual differences in imagery vividness and voluntary heart rate control

Dispositional differences in imagery vividness may account partially for the large individual differences in learning heart rate control. Based upon scores on the Betts QMI Vividness of Imagery Scale, 10 subjects were defined as a high imagery group and 10 subjects were defined as a low imagery group. All subjects were instructed to try to alter their heart rate as indicated by appropriately labeled lights. Six increasing, six decreasing and six rest, 45-sec trials were presented. During the trials, visual heart rate feedback was provided by a digital voltmeter. The dependent measure was computed by subtracting the mean of the last 5 beats of a 15-sec intertrial interval from the mean rate per trial. Results indicated that subjects were able to significantly raise and lower heart rate relative to rest trials. Furthermore, high imagery subjects demonstrated significantly larger changes than low imagery subjects on increase trials but not on decrease trials. Subjects in both groups typically reported using more specific images for increasing heart rate than for decreasing heart rate. These data suggest that dispositional differences in imagery vividness may have contributed to the differential performance of the two groups during the increase task.     

Classical conditioning of blood pressure inMacaca mulatta with cardiac rate controlled

The interaction of heart rate and blood pressure responses was studied in four male rhesus monkeys (M. mulatta) during classical delay conditioning and extinction. During initial conditioning sessions, heart rate was held constant by means of an external cardiac pacemaker; in follow-up conditioning sessions, the heart was free of constraint. Observations were made after these conditioning sessions, (a) during several sessions given over to continued training while different pacing rates were in effect; (b) during a series of extinction sessions in which the heart was paced and unpaced; and, (c) when a heart rate CR was simulated by manipulation of the pacer with no conditioned or unconditioned stimuli present. Throughout all sessions, systolic and diastolic pressures were measured at each heart beat. It was found that blood pressure conditioning was largely unaffected when heart rate changes were prevented during acquisition. Extinction of blood pressure responses was also seen to proceed relatively unimpeded when the heart was paced at a fixed frequency. During the postacquisition conditioning sessions, the several different pacing rates that were tested did not produce any significant differential effects in the blood pressure CRs. Simulated heart rate CRs were accompanied by small changes in pressures, sometimes opposite in direction from the pressure CRs observed during unpaced conditioning trials.     

Cardiac pacing and the law of initial value in rhesus monkeys

A cardiac pacemaker was used to produce a wide range of pre-CS heart rates in Pavlovian conditioning, and correlations were computed between pre-CS and CS heart rates. Two rhesus monkeys were presented trials in which 30 sec of light (CS) was followed by a brief electric shock (UCS). Pacing rates during the inter-trial intervals were varied from 150 to 350 bpm. Pacing was discontinued during CS. Unpaced control sessions showed that the maximum heart rate in CS on each trial was proportional to the pre-CS heart rate. Following pacing, however, heart rate in CS did not depend on the pacing rate; rather, the rates were not different from those of the unpaced control sessions. Pacing at high rates shifted the maximum heart rate toward the end of CS.     

Heart rate variability in the human transplanted heart: Nonlinear dynamics and QT vs RR-QT alterations during exercise suggest a return of neurocardiac regulation in long-term recovery

Rationale. Functional reinnervation of the transplanted human heart by the autonomic nervous system has not been demonstrated. A lack of autonomic control of the transplanted allograft is reflected by an increased resting heart rate, a sluggish heart rate response to dynamical exercise and a reduced heart rate variability. Recent evidence suggests that a measure of deterministic chaos in the heartbeat interval series (point correlation dimension, PD2i) is superior to the conventional power spectrum or other stochastic measures in detecting changes in the mechanism underlying heartbeat generation.Methods. The PD2i is based on the presumption that the variability is determined and patterned, whereas the stochastic measures all assume that the variability is around a stationary mean and is noise. The PD2i reconstructs the degrees of freedom (number of independent variables) in the system that generates the time series examined, and does this irrespective of whether the system is stochastic or deterministic and is stationary in time.Results. PD2i was determined for heartbeat intervals (RR, ECG digitized at 1200 Hz; supine posture) of 23 heart transplant recipients (HTR: 9 adults, 14 children; 0.04-7.7 years after transplantation) and 21 healthy control subjects (CTL; 13 adults, 8 children). The PD2i (+/? SD) averaged 5.4 +/? 0.7 for the CTL adults and 5.4 +/? 0.6 for the CTL children. Mean PD2i was reduced after transplantation to 1.1 +/? 0.1 in 6 HTRs recorded within 1 year after surgery; in one HTR recorded 2 weeks after surgery the mean PD2i was 3.7. Between 1 to 2 years PD2i was found increased in 2 of 3 subjects and between 2 to 8 years it was increased in 13 of 13, but not to control levels. In normal hearts the QT subinterval of each heartbeat cycle is associated with inotropy and the RR-QT remainder with chronotropy (i.e., the dyastolic interval during which RR is primarily regulated). To examine more closely the residual and returning heartbeat dynamics of the HTR subjects, these subinterval series were examined during mild exercise (40 to 90 Watts) and its recovery. In recent HTRs, resting QT and RR-QT were moderately reduced and modulated by exercise and recovery, but with an approximate 100 beat latency. In long-term (7-8 years) HTR subjects there was a rapid and larger response to exercise/recovery, but compared to normal the range was smaller and the complexity of the subinterval trajectories in time was simpler.Conclusions. Recurrence of low-dimensional deterministic dynamics after transplantation suggests recovery of neurocardiac control attributable to 1) reorganization of the viable intrinsic cardiac nervous system, 2) reinnervation by the extrinsic autonomic nervous system, or 3) both.     

Maturation of fetal responses to music

Maturation of fetal response to music was characterized over the last trimester of pregnancy using a 5-minute piano recording of Brahms' Lullaby, played at an average of 95, 100, 105 or 110 dB (A). Within 30 seconds of the onset of the music, the youngest fetuses (28-32 weeks GA) showed a heart rate increase limited to the two highest dB levels; over gestation, the threshold level decreased and a response shift from acceleration to deceleration was observed for the lower dB levels, indicating attention to the stimulus. Over 5 minutes of music, fetuses older than 33 weeks GA showed a sustained increase in heart rate; body movement changes occurred at 35 weeks GA. These findings suggest a change in processing of complex sounds at around 33 weeks GA, with responding limited to the acoustic properties of the signal in younger fetuses but attention playing a role in older fetuses.     

Anxiety, perception, and control of heart rate

8 women and 8 men took Cattell's IPAT-anxiety questionnaire and later McFarland's test of ability to perceive heart activity. The second test involved subjects' tracking their own heart rates, then they enrolled in an EKG biofeedback session to evaluate ability to increase and decrease heart rate from subjects' resting baselines. At the end of the session each subject completed Blanchard, et al.'s questionnaire to specify the cognitive strategies used for heart-rate control. Heart rate, abdominal respiration rate, respiration amplitude, EEG percent power in theta, alpha, and beta bands were evaluated. Success of heart-rate decrease seemed to depend mainly on activity levels: the subjects who achieved high scores on the activity test decreased heart rate significantly better than did low scorers. The relationship between scores for perception of heart and increases in heart rate was nonsignificant: increased heart-rate seemed to depend on differences in respiration between the rest and periods of increase. The significant, negative correlation between trait anxiety and perceptions of heart activity suggested that anxiety affected subjects' ability to perceive heart rate. The theta EEG power of the right hemisphere was significantly higher in subjects scoring high than for those low in perception of heart activity. During heart-rate increase tasks subjects mainly reported use of 'arousal responses,' similarly during heart-rate decrease tasks they reported use of relaxation responses.     

Habituation of cognitive and physiological arousal and social anxiety

This study examined differences in habituation between high and low socially anxious Ss. Participants gave three impromptu speeches, each separated by a brief rest interval. Skin conductance and heart rate were monitored during the speeches. Following each speech participants completed self-ratings of nervousness, heart rate, and palmar sweat activity as well as a modified Social Interaction Self Statement Test. Low anxious controls showed significant reduction of negative expectations and self-reported nervousness, heart rate, and sweat activity across the three trials. Actual heart rate of low-anxious subjects also decreased significantly across trials. In contrast, high anxious subjects did not evidence significant decreases in any of the above measures of anxiety and stress across the three trials. Skin conductance measures increased across trials for both groups, but increased more for the high anxious group than low-anxious controls. Results indicate that high anxiety participants are slow to decrease cognitive and autonomic responsiveness to stressful social situations.     

Rhythmic heart rate variability (sinus arrhythmia) related to stages of sleep

Evidence that the excursion of rhythmic heart rate variability (sinus arrhythmia) may be manipulated by altering the subject’s state of concentration, prompted a study of effects associated with different stages of sleep. There were 9 healthy subjects, ages 16–69, and one 55-year-old subject (C.P.) with a myocardial infarction history. Awake and during sleep the excursion of heart rate variability was greatest among the younger subjects, decreasing progressively with age. C.P. displayed rhythmic heart rate variability only when asleep. In light sleep (Stages 1 and 2) all subjects, with the exception of the 69-year-old, whose rate remained almost fixed throughout the night, displayed rhythmic heart rate variability, roughly synchronous with respiration. Neither the elderly subjeot nor C.P. entered into deep sleep (Stages 3 and 4). Among the other subjects, deep sleep was marked by a precise correlation between the frequency of respiration and heart rate variability. In REM sleep, however, in all subjects, there was often a total dissociation between respiration and rhythmic heart rate variability. There were periods of no heart rate variability with continuing regular respiration and of rhythmic heart rate variability with spontaneous breath holding. In the presence of heart rate variability during REM, C.P. displayed frequent ectopic beats, but ectopic beats were nearly absent in REM periods characterized by a steady, unvarying heart rate. The range of heart rate variability at rest and during sleep (sinus arrhythmia) reflects the manner in which autonomic activity is brought to bear in regulating the heart beat. The study of the behavior of sinus arrhythmia may offer a way to anticipate the occurrence of dangerous cardiac dysarrhythmias and sudden death, especially during sleep.