Heart Rate Variability



Heart Rate Variability


Diana E. McMillan

Robert L. Burr



Heart rate variability (HRV) is the beat-to-beat variation of the cardiac cycle that results, in large part, from the interaction of sympathetic and parasympathetic inputs to the sinus node. The term “arrhythmia” often carries negative connotations, and many serious disturbances of heart rhythm and waveform morphology are malignant. However, some variation in the time between successive beats is normal, reflecting a healthy heart and healthy autonomic nervous system (ANS). A moderate amount of respiratory sinus arrhythmia (RSA), for example, is viewed as evidence of good cardiovascular health. In addition to short-term or beat-to-beat variation, a healthy individual also exhibits a marked circadian or 24-hour variation in heart rate.

Measures of HRV provide clinicians and researchers with a noninvasive, practical, reproducible, sensitive, and dynamic insight into the autonomic neural regulation of the heart. These measures are increasingly popular in cardiac care and are recognized as important diagnostic tools for risk identification in a wide range of cardiovascular conditions and health conditions that predispose cardiac complications.

This chapter provides a basic overview of the mechanisms of HRV, the approaches used in measuring HRV, and guidance for the interpretations of these measurements. Current research related to HRV patterns in common cardiovascular conditions and in health conditions predisposing cardiac complications is presented. General health history factors that can influence HRV patterns are discussed. The chapter concludes with a brief review of pharmacological and nonpharmacological interventions and their impact on HRV patterns.


MECHANISMS OF HRV

The beat-to-beat variation of the cardiac electrical signal expressed in normal sinus rhythm is termed HRV and is considered to be an index of ANS balance and imbalance. The time between successive beats is governed by the intrinsic firing rate of the sinoatrial (SA) node and the modulation of the SA node firing rate by input from the ANS. The input of the ANS is based on the relative contributions of the two ANS branches: the sympathetic nervous system (SNS) and the parasympathetic nervous system (PSNS). Thus, HRV does not reflect absolute sympathovagal input, but rather the relative dominance and interaction of these two ANS branches. PSNS activity normally dominates under conditions of rest and restoration. SNS activity predominance is associated with increased physiological arousal.

Complicating the interpretation of HRV indices aimed at identification of these respective ANS inputs are neural and nonneural factors that can modify the SA node firing rate. These factors include the central nervous system integration of cardiac neural input, positive feedback from sympathetic afferents, and negative feedback from baroreceptors and vagal afferents.1 Despite our incomplete understanding of the physiological mechanisms associated with specific HRV measures, analysis of HRV provides significant clinical predictive usefulness within cardiac care.


HRV MEASUREMENT

HRV measures are statistical or mathematical summaries of within-subject variation in beat-to-beat heart period or instantaneous heart rate.2 This section summarizes the principles behind some of the more common HRV measures.


General Considerations

The current diversity of HRV measures and nomenclature is partially caused by the relative novelty and rapid proliferation of these methods. It also reflects simultaneous independent development in several distinct disciplines by clinical researchers with very different purposes and very different typical sources of heart rhythm information. Most HRV measures are so strongly correlated with each other that they are nearly redundant. However, no one subset of HRV measures so consistently outperforms all the others, in all circumstances, that clear choices can be made. There have been several attempts at standardization of HRV measures and nomenclature,3 but the recommendations have not been universally accepted, particularly in the interdisciplinary literature relevant to nursing.

Practically, the instantaneous heart period must be defined from a series of discrete events corresponding to the beating of the heart. This discrete event series itself is usually derived from fiducial features of the raw electrocardiograph (ECG) waveform. The arrival time of a beat, in particular the time interval from the previous beat, provides us with somewhat irregularly spaced information about short-term fluctuations in heart rhythm, and by inference, the dynamic autonomic control of the heart.4,5

Despite the variety of purposes motivating HRV analysis, the primary goal is usually to compute some within-subject or withincondition indices of heart rhythm variation to make some qualified inferences, not about the heart organ itself, but about the sympathetic and parasympathetic neural traffic impinging on the SA node of the heart. Thus it would be ideal to base the definition of the interbeat heart period on the interval between adjacent P waves to reflect as closely as possible the statistics of the firing of the SA pacemaker node.6,7 However, the P-P interval is much harder to empirically define than the R-R interval, particularly from noisy low-frequency Holter recordings of ambulatory subjects. Most HRV studies, and essentially all of those performed using ambulatory ECG monitoring technology, use the R-R interval as the fundamental metric. Although it is conventional to speak of heart period as specific to a particular beat, an R-R interval is actually a measurement of the time interval between the R waves of
two successive normal beats, sometimes called an NN doublet (Fig. 17-1).






Figure 17-1 Example strip of ambulatory Holter ECG recorded during sleep. The beats are coded as Normal (“N”), and the R-R intervals (in milliseconds) for each overlapping pair of beats are displayed above and slightly to the left of the R wave that terminates the interval. Two respiratory cycles of probable RSA are visually apparent in the strip. Note that the measured sequential R-R intervals vary considerably within a few seconds for this high HRV subject.

The quality of HRV indices is ultimately dependent on the consistency of the basic measurement of each R-R interval. In modern digital applications, the R-R interval is partially determined by the sampling rate of the raw ECG, and also by characteristics of the R-wave location finding algorithm. Typical sampling rates may vary from approximately 100 Hz (samples per second), still common in long time scale ambulatory monitoring, to 1,000 Hz or faster in laboratory studies. In general, a higher digital waveform sampling rate allows proportionally more precision in the estimation of the location of the R waves at the cost of greater processing and memory requirements. However, the resulting apparent gain in precision may be illusory in ambulatory ECG recordings that contain noise and morphologies that vary slightly with posture and activity.








Table 17-1 ▪ DESCRIPTION OF TIME DOMAIN MEASURES OF HRV





































Measure


Units


Description


Mean RR


ms


Mean of all NN intervals


SDNN


ms


Standard deviation of all NN intervals


CoV



Coefficient of variation, equal to 100 × SDNN/(mean RR)


SDANN


ms


Standard deviation of the averages of NN intervals in all 5-minute segments of the recording


SDNN index


ms


Mean of the standard deviations of all NN intervals in all 5-minute segments of the recording


rmsSD


ms


Square root of the mean squared differences between successive NN intervals


pNN50


%


Number of successive NN intervals differing by more than 50 ms divided by the total number of successive NN intervals, expressed as a percentage


ms, milliseconds.


The R-R intervals thus decoded from the raw ECG can be placed into an ordered temporal sequence to form a time series, in which the continuous length of each cardiac cycle is an interval measure of time, usually reported in units of milliseconds(ms). Each R-R interval may be inverted to a beat-specific instantaneous equivalent heart rate, which can be considered the heart rate in beats per minute that would have been observed if all the heart beats in a 60-second period had exactly the length of that specific individual interval.5


HRV Measures

Although there are a variety of approaches used to analyze HRV, the two major procedures are time domain analysis and frequency domain analysis. Definitions for HRV measures based on these approaches are presented in Tables 17-1 and 17-2, respectively.








Table 17-2 ▪ DESCRIPTION OF FREQUENCY DOMAIN MEASURES OF HRV





































Measure


Units


Description


PSD plot


ms2/Hz


Plot of power spectral density (PSD) versus frequency; frequency range is generally less than 0.4 Hz


Total power


ms2


Area under PSD curve, equal to the variance of the segment; segment length can be short (5 minutes) or entire recording


LF


ms2


Power in the LF band between 0.04 and 0.15 Hz; it reflects both sympathetic and parasympathetic activity


HF


ms2


Power in the HF band between 0.15 and 0.4 Hz; it predominantly reflects parasympathetic activity


LF:HF



Ratio of LF power to HF power; a higher number indicates increased sympathetic activity or reduced parasympathetic activity


LFnu


%


Low-frequency power in normalized units, LF/(LF + HF), expressed as a percentage


HFnu


%


High-frequency power in normalized units, HF/(LF + HF), expressed as a percentage


LF, low frequency; HF, high frequency.




Time Domain Analysis

Time domain analysis is based on the statistical interpretation of R-R time interval values. Time domain measures of HRV (Table 17-1) are closely related to the total variance of the heart signal.3,8 The most common index of overall HRV is the standard deviation of all R-R intervals (SDNN), typically involving 80,000 to 150,000 heart period values in a 24-hour recording. Long-term variability, such as that reflecting normal circadian influence over a 24-hour period, is best reflected by two measures based on partitioning the full recording into sequential 5-minute segments. Each segment typically contains 300 to 500 R-R intervals, and there would be 288 such segments in a 24-hour recording. The SDANN is defined as the standard deviation of the means of the R-R intervals in each 5-minute segment, whereas the complementary SDNN index is the mean of the standard deviations of the R-R intervals in each 5-minute segment.

Short-term time domain measures of HRV are derived from the differences of successive normal R-R intervals. They are highly correlated and are considered to provide good estimates of PSNS activity.3 Short-term measures include the square root of the mean squared difference of successive normal R-R intervals (rmsSD) and the percentage of successive normal R-R intervals that change by more than 50 milliseconds compared with the total number of R-R intervals (pNN50).


Frequency Domain Analysis

Frequency domain analysis, or spectral analysis, is an elegant method for studying the rhythmic components in an R-R interval sequence and presents intriguing possibilities for disentangling PSNS and SNS influences on the heart.9 A plot of the power spectral density of HRV versus frequency describes how the variances of the frequency components of the heart signal are distributed.3

Both parametric and nonparametric methods common to time series analysis have been used to estimate the power spectral density. The most common methods are the discrete Fourier transform (DFT) (nonparametric) and autoregressive (AR) (parametric) time series models. The AR model-based spectrum is usually less computationally efficient than the DFT, but it can be applied to data sequences of arbitrary length, including very short segments. The AR approach tends to produce a spectrum that is statistically more stable than that produced by the DFT but requires assumptions about the time series model.3

The total area under the curve of the power spectral density versus frequency plot is equal to the total statistical variance, or the power of the signal. These power (variance) distributions are calculated for defined frequency bands and are interpreted as an estimate of the variance of the HRV signal within that band (Table 17-2). There are two major spectral components seen in HRV data: the high-frequency (HF) (0.15-0.40 Hz) component and the low-frequency (LF) (0.04-0.15 Hz) component (Fig. 17-2). The HF component is associated with respiration8,10 and is considered to reflect the relative input of the PSNS. The basis of the LF component is more controversial and may be the result of both SNS and PSNS activity input.11 The LF to HF ratio (LF:HF) has been regarded as reflecting the balance between the mixed PSNS and SNS activity input to the PSNS activity input.3 The spectral HF and the LF:HF are often reported together in nursing research studies seeking to explore the joint contribution of the SNS and the PSNS branches to HRV phenomena. Studies of very low-frequency and ultra-low-frequency ranges have also been conducted but require long uninterrupted sampling periods and specialized methods of analysis. In addition, the clinical interpretation of findings in these frequency ranges remains controversial.3

In common with other variance-like measures, the within-subject HRV band power estimates are often reexpressed using the natural logarithm transform to reduce distributional skewness before use in statistical procedures. HRV quantitative band power summary indices (LF, HF, etc.) computed using the AR or DFT methods should be virtually identical.12

Several derived measures can easily be computed from these spectral band summaries (see Table 17-2). Normalized variants of the LF and HF indices are often defined by dividing the power in each band by the total power, with the result expressed as a percentage.13

It should be pointed out that the HRV spectrum and spectrum-based band power (variance) summary statistics, like all HRV measures, are defined over blocks of R-R intervals; thus, their meaning is not localized to a particular instant in time or to a particular beat. Typical block window lengths in clinical and research applications range from 2 minutes to 24 hours. Spectra derived from shorter blocks are more localized in time and are more likely to be internally stationary but may have less frequency resolution, especially with respect to slower rhythm patterns. HRV spectra based on very long individual blocks (e.g., 24 hours) will have the ability to resolve very slow rhythmic patterns but will almost certainly span nonstationary data segments and heterogeneous latent autonomic states.


HRV PATTERNS IN COMMON CARDIOVASCULAR CONDITIONS

The following section provides a summary of HRV research findings in myocardial infarction (MI); arrhythmias and sudden death; angina; hypertension (HTN); heart failure; and cardiac surgery, heart transplant, and other invasive procedures. The reader should refer to related chapters for more detailed descriptions of these conditions and interventions.


Myocardial Infarction

It is well established that HRV patterns are disturbed in patients who have experienced MI.14, 15, 16, 17 In post-MI patients, those with restrictive left ventricular filling have been reported to have especially reduced HRV patterns compared with those without this disorder.18

Decreased HRV after MI is viewed as a significant risk factor for cardiac death16,17 or subsequent nonfatal MI within 12 months.19 HRV measures of total variability, such as SDNN and SDANN, are viewed as the most useful predictors of mortality.3,16 Erratic sinus rhythms (sinus arrhythmia of nonrespiratory origin) as identified by abnormal Poincaré plots in post-MI patients are significant risk markers for increased mortality at follow-up.17 Results from a landmark study, the Multicenter Post-Infarction Project, indicate that patients with an SDNN of less than 50 milliseconds (24-hour recording), measured within 11 days of the MI, have a risk of mortality at 1 year that is 5.3 times higher than do patients with an SDNN greater than 100 milliseconds.20 Predicted risk is also increased for patients with below-normal SDNN and SDANN values in the chronic phase after MI. Reported normal lower limits measured in patients, at least 3 months after MI,
range between 63 and 89 milliseconds for SDNN and between 57 and 79 milliseconds for SDANN.16,21 Reduced circadian variation in the cardiac signal, reduced total power, and a shift toward sympathetic predominance as reflected by an increased LF:HF are also seen after MI.3






Figure 17-2 Two 5-minute HRV power/variance spectra based on data collected on a single male subject with diabetes during sleep between 3 and 4 a.m. The top figure is based on data collected in 1992 when the subject was 42 years of age. The average heart rate of the analysis segment is 61 beats per minute, and the HRV spectrum demonstrates high total power/variance with a well-developed mid-frequency peak at 0.25 Hz, probably reflective of vagally mediated RSA. The lower figure is based on data collected from the same subject a decade later in 2002, when he was 52 years old. The average heart rate of the nocturnal analysis segment is 70 beats per minute. The total power of the HRV spectrum is an order of magnitude less, and while the low frequency peak has diminished, the high frequency peak is significantly attenuated. The quantitative band power and derived measure summaries appear in the table to the right of each figure, and document that the HF power is lower and the LF:HF much increased in the lower figure. The changes over time may reflect the joint influence of aging and diabetes. VLF, very low frequency.

While HRV measures can provide independent risk prediction, the combination of HRV indices with other cardiovascular risk factors can enhance prediction of cardiac events. In a prospective study of 304 patients with acute coronary syndrome without ST-segment elevation, a combination of risk factors including clinical data, troponin T concentrations, ST-segment monitoring, and HRV provided a prediction of risk for ischemic death or nonfatal MI of 40% in the first 30 days and of 46.9% in the first 12 months.19


Arrhythmias and Sudden Death

HRV studies indicate autonomic disturbance before arrhythmia, although the pattern of disturbance varies. Clinical research supports that reduced parasympathetic tone and increased sympathetic tone predisposes ventricular fibrillation and ventricular tachycardia.22 The onset of paroxysmal atrial fibrillation has been reported to follow autonomic changes characterized by an initial steady increase in SNS activity and a subsequent sharp predominance in PSNS activity.23

Reduced HRV is a consistent finding in a review of studies monitoring sudden death or sudden death and malignant arrhythmias.24 In a prospective study (N = 1071) of post-MI patients, reduced HRV independently contributed to the risk of sudden death and or sustained ventricular tachycardia.22 The combination of
low HRV, nonsustained ventricular tachycardia, and baroreflex sensitivity led to a 22-fold increase in the risk for sudden death or sustained VT. Findings from the Women’s Health Initiative, a prospective, population-based study of postmenopausal women, identify a reduction in HRV as one of the five major ECG abnormality predictors for mortality.25

Sudden unexpected death syndrome is a leading cause of death in young men of Southeast Asian descent. Although nighttime ventricular fibrillation typically precedes the cardiac arrest, the pathophysiology of this disorder is not known. HRV analysis indicates reduced 24-hour HRV and reduced circadian variation in HRV, with very low-nighttime HRV in survivors of this syndrome compared with controls.26


Angina

Low HRV is associated with poor prognosis in stable angina27 and unstable angina.19,28 Low HRV, including reduced total power and reduced HF, LF, and very low-frequency components, provide strong and independent predictors of cardiac death but not nonfatal MI in stable angina pectoris.27 Low HRV in patients presenting with unstable angina increases the risk of either cardiac death or nonfatal MI within 12 months,19 with patients having SDNN, HF, and LF power values in the lowest quartile at increased risk for in-hospital death.28


Hypertension

Individuals with HTN29 and those at high risk for HTN30,31 exhibit abnormal HRV patterns. HRV is reduced in individuals with essential HTN compared with healthy control subjects, as reflected in significantly reduced values for SDNN, SDANN, pNN50, and rmsSD.32 These findings are consistent with results from the Framingham Heart Study (N = 1,919), a prospective epidemiological study of coronary risk factors. Singh et al.31 found significant reductions in time domain measures of HRV in men and women with HTN compared with normotensive subjects. Hypertensive patients also exhibit a lower than normal vagal tone (low HF).31,32

Some HTN studies indicate a significant increase in mixed SNS and PSNS activity, reflected by increased resting LF power.32,33 However, results from the Framingham Heart Study indicate reduced LF activity in patients with HTN and suggest that low vagal tone is a strong risk factor for the development of HTN in men.31 Methodological differences in covariate adjustment may have contributed to inconsistent study findings.

The normal circadian pattern of HRV is disturbed in patients with HTN. In normotensive individuals, the nighttime fall in blood pressure is paralleled by a corresponding reduction in the mixed PSNS and SNS activity marker LF. This nocturnal drop in LF power is not as great in subjects with HTN.29


Heart Failure

Patients with congestive heart failure (CHF) have reduced HRV.34 The most consistently reported finding is a reduction in SDNN.35, 36, 37, 38 Some researchers report a significant early increase in sympathetic predominance (high LF:HF) using a paced canine model of CHF.39 This initial SNS surge appears to be lost as the condition worsens.34 Reduced vagal activity (low HF) and reduced total power have also been reported in patients with CHF.34

Although the derangement of ANS function in heart failure is well recognized, the ability of HRV measures to aid in the risk assessment of patients with this disorder is mixed.40 One of the factors complicating the interpretation of HRV patterns in heart failure is the impact of respiratory patterns. Cheyne-Stokes respiration and oscillatory breathing pattern (characterized by cyclic changes in ventilation without apnea) are common in CHF and are associated with significant reductions in mixed SNS and PSNS activity (low LF power).41 Interestingly, this severe LF power decrease contrasts with the increase in SNS activity found in patients with obstructive sleep apnea, hypoxia, and hypercapnia. For patients with obstructive sleep apnea, the LF:HF is considered the best estimator of the apnea/hypopnea index, a measure of disorder severity.42

Despite these confounding factors, the use of HRV measures in heart failure is reported to be helpful in the evaluation of risk for malignant cardiac events. HRV is a significant predictor for sudden death.35, 36, 37, 38 Based on a multivariate survival model, risk of sudden death in patients with CHF was strongly predicted by HRV.37 Researchers collected ECG data during 8 minutes of controlled breathing and found that reduced sympathetic predominance (LF power ≤13 ms2) in patients with CHF was associated with a relative risk of 3.7 for sudden death compared to patients with sympathetic input above this level. Of patients with CHF, those patients presenting with values of SDNN less than 65.3 milliseconds were reported to be at significantly greater risk for sudden death.35 In another study, mortality and hospitalization caused by deterioration of CHF were predicted by HRV. In this case, SDNN (<75 milliseconds) provided significant and independent predictive value in addition to the standard risk indices of left ventricular ejection fraction and peak oxygen intake.36

Another cardiac condition exhibiting disturbed ANS functioning is aortic regurgitation. Low SDANN significantly predicted risk of death or progression to aortic valve repair in a study of 50 asymptomatic or minimally symptomatic patients with chronic severe aortic regurgitation.43


Cardiac Surgery, Heart Transplantation, and Other Invasive Procedures

Cardiovascular surgery has been shown to have an impact on HRV patterns. Results from the Cardiac Arrhythmia Suppression Trial indicate that HRV is significantly reduced in post-MI patients after coronary artery bypass graft (CABG) surgery and that this reduction in HRV is not associated with increased mortality.44 This finding contrasts with the increased risk of mortality seen in post-MI Cardiac Arrhythmia Suppression Trial patients who did not undergo CABG surgery but who did exhibit reduced HRV, specifically reduced SDANN. This important finding may help to explain the lower than predicted mortality rates seen in some post-CABG surgery patients. Similar reductions in HRV were reported in a Danish study of CABG patients without a recent MI and with ejection fractions of 0.36 ± 0.07.45 These researchers found significantly reduced HRV immediately post-CABG and at the 6-month follow-up. Furthermore, improvement in myocardial function seen after surgery was not associated with post-CABG measures of HRV.

Heart transplant patients exhibit significantly reduced HRV immediately after surgery.34 This reduction continues to be significantly reduced 2 years after surgery despite evidence of a return to
near-normal cardiac-specific sympathetic nerve firing. This finding suggests insufficient or dysfunctional reinnervation of the cardiac muscle, and particularly the SA node with respect to ANS nerve fiber communication.34

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Jan 10, 2021 | Posted by in NURSING | Comments Off on Heart Rate Variability

Full access? Get Clinical Tree

Get Clinical Tree app for offline access