Jerry S. Weiss

Power Spectral Analysis of Heart Rate Varability in Sudden Cardiac Death: Comparison to Other Methods

Power spectrum analysis of heart rate variability is described and compared to four other reported methods, with respect to their efficacy as predictors of risk of sudden cardiac death (SCD). Approximate frequency domain representations were obtained for each. The underlying physiologic processes which may give rise to spectral components are considered. These methods were employed to analyze 24-h ambulatory ECGs of patient populations at different degrees of risk of SCD. Heart rate variability was found to be reduced in cardiac patients known to be at increased risk of SCD, when compared to those not at increased risk. These differences were greatest in power spectral methods. Thus, power spectrum analysis appears to be more effective than the other methods in segregating these populations, suggesting that this method may be useful in categorizing cardiac patients according to risk of sudden cardiac death.

Diurnal pattern of QTc interval: how long is prolonged? Possible relation to circadian triggers of cardiovascular events.

OBJECTIVES This study sought to evaluate the range and variability of the QT and corrected QT (QTc) intervals over 24 h and to assess their pattern and relation to heart rate variability. BACKGROUND Recent Holter monitoring data have revealed a high degree of daily variability in the QTc interval. The pattern of this variability and its relation to heart rate variability remain poorly characterized. METHODS We developed and validated a new method for continuous measurement of QT intervals from three-channel, 24-h Holter recordings. Average RR, QT, QTc and heart rate variability were measured from 5-min segments of data from 21 healthy subjects. RESULTS Measurement of 6,048 segments showed mean (+/- SD) RR, QT and QTc intervals of 830 +/- 100, 407 +/- 23 and 445 +/- 16 ms, respectively (mean QTc interval for men 434 +/- 12 ms, 457 +/- 10 ms for women, p < 0.0001). The average maximal QTc interval was 495 +/- 21 ms and the average QTc range 95 +/- 20 ms. The maximal QTc interval was > or = 500 ms in 6 subjects and > or = 490 ms in 13. The 95% upper confidence limit for the mean 24-h QTc interval was 452 ms (men 439 ms, women 461 ms). The RR, QT and QTc intervals and the high frequency component of heart rate variability were greater during sleep. Both the QTc interval and the variability between hourly minimal and maximal QTc intervals reached their circadian peak shortly after awakening, before declining to daytime levels. CONCLUSIONS The maximal QTc interval over 24 h in normal subjects is longer than heretofore thought. Both QT and QTc intervals are longer during sleep. The QTc interval and QTc variability reach a peak shortly after awakening, which may reflect increased autonomic instability during early waking hours, and the time of the peak value corresponds in time to the period of reported increased vulnerability to ventricular tachycardia and sudden cardiac death. These findings have implications regarding the definition of QT prolongation and its use in predicting arrhythmias and sudden death.

Heart Rate Variability: Frequency Domain Analysis

Experience with frequency domain analysis over the past two decades strongly suggests that it represents a unique, noninvasive tool for achieving a more precise assessment of autonomic function in both the experimental and clinical settings. Available studies indicate that the significance of the HF component is far better understood than that of the lower frequency components. In general, it is considered to reflect vagal activity, and because it is readily manipulated pharmacologically, is used as a an index of that activity. However, some caution is required because this parameter also is strongly influenced by the degree of coupling between respiration and heart rate, which, in turn, reflects the intensity of the respiratory effort as well as of parasympathetic activity. Respiratory pattern also can significantly influence HF power. The use of controlled breathing minimizes these problems, improves reproducibility of test findings, and also facilitates quantitative comparisons. The situation with respect to LF power is more complicated because it is modulated by both sympathetic and parasympathetic outflows (see previous discussion) as well as by other factors, including baroreceptor activity. Therefore, LF analysis per se cannot afford a precise delineation of the state of sympathetic activation. Determinations of the LF/HF ratio, an index of sympathovagal balance both under control conditions and in conjunction with interventions that maximize sympathetic and parasympathetic activity, provide additional insights, as do correlations between spectral activity and direct nerve recordings, plasma norepinephrine concentrations, and radionuclide imaging of adrenergic nerves. Renewed interest has recently been evinced in frequencies lower than 0.04 Hz in view of reports that the VLF portion of the spectrum (0.01-0.04 Hz) reflects a purer form of sympathetic activity than does the LF band. Despite the potential applicability to clinical problems, only very little is known about the physiologic basis of the VLF and ULF bands. Further study is required. However, it is important to note that meaningful determinations of VLF and ULF power may be difficult because decreases in frequency to such low levels are associated with an increasing propensity to violate the rules governing power spectral determinations (see previous discussion and appendix), violations that diminish reliability despite the most sophisticated preprocessing. It is also noteworthy that the reliability of spectral power determinations diminishes with decreases in the power of the signal and of the signal-to-noise ratio.(ABSTRACT TRUNCATED AT 400 WORDS)

Heart rate variability and sudden death secondary to coronary artery disease during ambulatory electrocardiographic monitoring

Data are analyzed from 5 patients who died suddenly during ambulatory electrocardiographic monitoring. Three of the patients were also assessed in terms of 2 recently developed indexes of heart rate (HR) variability. One of these, the standard deviation of RR intervals during successive 5-minute segments averaged over 24 hours, has been reported to be a putative index of vagal tone. Comparisons were made with HR variability findings in 20 normal volunteers. Sudden death was due to ventricular tachycardia degenerating into ventricular fibrillation in all cases. Both early (3 patients) and late cycle (2 patients) ventricular premature complexes initiated the terminal dysrhythmia. An increased density of ventricular ectopic activity was noted in the hour before onset of ventricular fibrillation. HR variability as measured by the standard deviation was significantly (p less than 0.01) lower in the patients who died suddenly (30 +/- 10 ms) than in the normal subjects (76 +/- 14 ms). These findings support suggestions that HR variability analysis may be useful in identifying patients at a higher risk of sudden death.

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

Before heart rate (HR) variability can be used for predictive purposes in the clinical setting, day-to-day variation and reproducibility need to be defined as do relations to mean HR. HR variability and mean HR were therefore determined in 2 successive 24-hour ambulatory electrocardiograms obtained from 33 normal subjects (age 34 +/- 7 years, group I), and 22 patients with coronary disease and stable congestive heart failure (CHF) (age 59 +/- 7 years, group II). Three measures were used: (1) SDANN (standard deviation of all mean 5-minute normal sinus RR intervals in successive 5-minute recording periods over 24 hours); (2) SD (the mean of the standard deviation of all normal sinus RR intervals in successive 5-minute recording periods over 24 hours); and (3) CV (coefficient of variation of the SD measure), a new measure that compensates for HR effects. Group mean HR was higher and HR variability lower in group II than in group I (80 +/- 10 vs 74 +/- 9 beats/min, p less than 0.04). Mean group values for HR and HR variability showed good correlations between days 1 and 2 (mean RR, r = 0.89, 0.97; SDANN, r = 0.87, 0.87; SD, r = 0.93, 0.97; CV, r = 0.95, 0.97 in groups I and II, respectively). In contrast, considerable individual day-to-day variation occurred (group I, 0 to 46%; group II, 0 to 51%). Low HR variability values were more consistent than high values. SDANN and SD correlated moderately with HR in both groups (r = 0.50 to 0.64). The CV measure minimizes HR effects on HR variability.(ABSTRACT TRUNCATED AT 250 WORDS)

Evaluation of Five QT Correction Formulas Using a Software-Assisted Method of Continuous QT Measurement from 24-Hour Holter Recordings

To evaluate and compare QT correction formulas in healthy subjects, we used 24-hour Holter monitoring because it allows the assessment of QT intervals over a large range of rates. Computer-assisted QT-interval measurements were obtained from 21 subjects. QT-RR relations for individuals and the group were fitted by regression analysis to 5 QT prediction formulas: simple Bazetts, modified Bazetts, linear (Framingham), modified Fridericias and exponential (Sarmas). There were no significant differences in mean squared residuals between formulas. When using individually calculated regression parameters, each formula gave good or acceptable QT correction over the entire range of RR intervals. Simple Bazetts formula (which uses no regression parameters) was unreliable at high rates. Akaike information criteria rank was: Sarmas, Framingham, modified Bazetts, Fridericias, and simple Bazetts. When group-based regression parameters were applied to individuals, no formula had a clear advantage over simple Bazetts. We conclude that any formula that invokes regression parameters unique to each individual provides satisfactory QT correction. Determination of these parameters requires long-term recording to obtain an adequate range of rates. Group-based regression parameters give poor correction. When individual parameters cannot be determined, as in a 12-lead electrocardiogram, no formula provides an advantage over the familiar simple Bazetts.

QT Interval Dispersion in Healthy Subjects and Survivors of Sudden Cardiac Death: Circadian Variation in a Twenty- Four-Hour Assessment

Twenty-four-hour acquisition of QT dispersion (QTd) from the Holter and the circadian variation of QTd were evaluated in 20 survivors of sudden cardiac death (SCD), in 20 healthy subjects, and in 14 control patients without a history of cardiac arrest who were age, sex, diagnosis and therapy matched to 14 SCD patients. Computer-assisted QT measurements were performed on 24-hour Holter recordings; each recording was divided into 288 5-minute segments and templates representing the average QRST were generated. QTd was calculated as the difference between QT intervals in leads V1 and V5 for each template on Holter. The 24-hour mean QTd was significantly greater in SCD patients (40 +/- 28 ms) than in healthy subjects (20 +/- 10 ms) and control patients (15 +/- 5 ms) (p <0.05). There was a circadian variation in QTd with greater values at night (0 to 6 A.M.) than at daytime (10 A.M. to 4 P.M.) in healthy subjects (25 +/- 13 vs 15 +/- 8 ms, p <0.001) and control patients (18 +/- 10 vs 12 +/- 4 ms p <0.05), whereas in SCD patients there was no significant difference between night and day values (45 +/- 31 vs 37 +/- 28 ms, p = NS). It is concluded that QTd measured by Holter was greater in SCD patients than in healthy subjects and matched control patients during the entire day. QTd has a clear circadian variation in normal subjects, whereas this variation is blunted in SCD patients. QTd measured on Holter differentiates survivors of cardiac arrest and may be a useful tool for risk stratification.

The missing second: What is the correct unit for the Bazett corrected Qt interval?

Abstract When Taran and Szilagyi 6 reexpressed the Bazett formula in its commonly used form, they eliminated the seemingly useless step of dividing by 1 second by requiring that RR be expressed in seconds. However, this produced the confusion with regard to the correct unit. Incorporation of the 1-second expression allows the QTc to be expressed in the unit that is both logical and arithmetically correct (namely that of the original QT), and eliminates any requirement regarding the units in which the original QT and RR are measured.

Does heart rate identify sudden death survivors? Assessment of heart rate, QT interval, and heart rate variability

The objective was to test whether the circadian variability of several electrocardiographic variables distinguishes sudden cardiac death survivors from heart disease patients without a history of cardiac arrest and from normal subjects. Heart rate, heart rate variability, and QT interval have been reported to identify survivors of sudden cardiac death. Computer-assisted continuous QT measurement and heart rate variability analysis were performed on 24-hour Holter records for three groups: (1) 14 sudden death survivors; (2) 14 control patients with diagnosis and therapy matched to survivors; and (3) 14 healthy subjects. There were no significant differences in 24-hour mean RR and QT intervals between groups. However, heart rate was significantly different between the three groups at night but not during the day because the expected nighttime decline was markedly blunted in survivors and somewhat blunted in control patients. The QT interval and frequency domain heart rate variability measures followed a similar circadian pattern. The mean QTc was significantly longer in control patients. The QTc had a wide range in all groups, but less in sudden death survivors. Of ten common time and frequency domain heart rate variability indices, only SDANN and SDNN were significantly lower in sudden death survivors. Reduced circadian variation of heart rate, with marked blunting of the nighttime heart rate decline, identifies sudden cardiac death survivors as well as does SDANN and SDNN, and, in contrast to heart rate variability measures, can easily be obtained from a Holter report without complex calculations.

Spectrum of heart rate variability

Summary form only given. To test the hypothesis that patients with advanced structural disease and congestive failure have intermediate heart rate variability (HRV) values, the authors compared HRV in 33 young healthy subjects without evidence of heart disease, 23 patients with ischemic heart disease complicated by stable NYHA class II-III heart failure (CHF), and 13 patients who died during Holter monitoring (sudden cardiac death-SCD). Five CHF patients died after 9+or-5 months (CHF/D). Two HRV measures, SDANN (standard deviation (SD) of the mean of sinus R-R intervals (R-R) for successive 5-minute periods over 24 h) and SD (mean of SD of R-R for successive 5-min periods over 24 h) were determined from 24-h Holter records. Group differences were tested by analysis of variance. The data suggest that there is a spectrum of HRV that is highest in normal young subjects, lowest in SCD patients, and intermediate in the CHF group. However, HRV values in all CHF/D patients approximated those for SCD patients, suggesting that progression of disease and increasing risk of dying are associated with a decline in HRV. HRV measures may thus provide a tool for monitoring mortality increase in patients with heart disease.<<ETX>>