Is heart rate variability a reliable method to assess autonomic modulation in left ventricular dysfunction and heart failure?

International Journal of Cardiology 67 (1998) 9–17

Is heart rate variability a reliable method to assess autonomic modulation in left ventricular dysfunction and heart failure? Assessment of autonomic modulation with heart rate variability Simonetta Scalvini a , Maurizio Volterrani a , Emanuela Zanelli a , Marco Pagani a , Giorgio Mazzuero b , c a, Andrew J. Coats , Amerigo Giordano * a

Division of Cardiology, Rehabilitation Institute of Gussago, Salvatore Maugeri Foundation, IRCCS Gussago, Brescia, Italy b Division of Cardiology, Rehabilitation Institute of Veruno, Salvatore Maugeri Foundation, IRCCS Veruno, Novara, Italy c Cardiac Medicine Department NHLI Brompton Hospital, London, UK Received 9 February 1998; received in revised form 26 August 1998; accepted 26 August 1998

Abstract Autonomic dysfunction seems to be involved in the progression and prognosis of congestive heart failure. Measurement of heart rate variability (HRV) provides a noninvasive method to obtain reliable and reproducible information on autonomic modulation of heart rate, but there is a difficulty in using HRV as a quantitative estimate of autonomic dysfunction in heart failure. This study was aimed at testing the hypothesis that abnormal modulation of heart rate assessed by power spectrum analysis may be present also in asymptomatic patients with left ventricular dysfunction and progress in patients with overt symptoms of congestive heart failure. HRV was measured in three groups of subjects: Group 1: 30 patients with chronic heart failure; Group 2: 21 patients with asymptomatic left ventricular dysfunction; and Group 3: 25 healthy volunteers as control group. HRV was evaluated by autoregressive spectral analysis with 600-beat ECG samples, while subjects were quietly recumbent (BS1), in conditions of controlled breathing (15 acts / min) (RSC) and passive orthostatism after tilting (808) (TLT). Patients in group 1 showed a reduction in the standard deviation of the R–R intervals (SDRR) ( p,0.0003) and in the low frequency component (LF) ( p,0.0001) compared to normal subjects. Low frequency component was not detectable in 11 patients of group 1 ( p,0.0008). On RSC and TLT, group 1 failed to show any modification in the low frequency and high frequency components (HF) under any stimulation. Group 2 showed no modification at baseline evaluation, no increase in the high frequency component on RSC and in LF during TLT compared to controls ( p,0.01 and p,0.0001 respectively). At baseline, group 1 had a lower SDRR ( p,0.03) and LF ( p,0.0001) vs. group 2, whereas during stimulation the two groups exhibited the same behaviour. In conclusion, reduced heart rate variability is specific for both asymptomatic and symptomatic post-ischemic left ventricular dysfunction. Our results suggest that frequency domain analysis of heart rate variability during a stimulation test allows a more accurate definition of the degree of autonomic control of heart rate.  1998 Published by Elsevier Science Ireland Ltd. All rights reserved. Keywords: Heart rate variability; Left ventricular dysfunction; Chronic heart failure

1. Introduction The clinical syndrome of congestive heart failure is associated with neurohumoral excitation resulting in an abnormal autonomic modulation of the car*Corresponding author. Tel.: 139-30-252-860; fax: 139-30-252-118.

diovascular system. Increased sympathetic activity, parasympathetic withdrawal and impaired baroreflex gain have been described [1–4]. Analysis of heart rate variability provides a noninvasive method to obtain reliable and reproducible information on the sympatho-vagal balance of the heart [5–8]. The problem is the difficulty in using power spectral

0167-5273 / 98 / $ – see front matter  1998 Published by Elsevier Science Ireland Ltd. All rights reserved. PII: S0167-5273( 98 )00252-6

S. Scalvini et al. / International Journal of Cardiology 67 (1998) 9 – 17

10

analysis measures as a quantitative estimate of autonomic dysfunction in heart failure or of the severity of heart failure. A reduction in heart rate variability has been reported in patients with congestive heart failure, particularly those in the advanced stages of the disease [9–12] and in patients with recent myocardial infarction [13–18]. To our knowledge, there is little information on spectral analysis of heart rate variability in asymptomatic patients with chronic left ventricular dysfunction. We investigated a group of controls, a group with asymptomatic left ventricular dysfunction and a group with symptomatic heart failure and calculated the individual spectral parameters to test, with heart rate variability, the presence and the possible progression of an abnormal autonomic modulation.

2. Methods

2.1. Study population Three groups of subjects were identified: Group 1: Thirty patients with congestive heart failure, in stable condition with no change in signs and symptoms within two weeks of measurements of heart rate variability. They were 28 men and two women, mean age 6269. Symptoms had been present for 6–18 months in 16 patients, for over 18 months in the others. Fifteen subjects were in NYHA class II, fifteen in NYHA class III–IV. All subjects were taking diuretics; 24 were on ACE-inhibitors (80%); 24 were receiving digitalis (80%), and ten (33%) nitrates. Group 2: Twenty-one patients with asymptomatic left ventricular dysfunction. Nineteen men, two women, mean age 5769. Four were taking diuretics (19%), six were on digitalis (28%), and 15 on ACEinhibitors (71%). In both groups left ventricular dysfunction (ejection fraction ,40%) was secondary only to ischemic heart disease. All of the patients were in sinus rhythm, none had any history of acute myocardial infarction or had undergone cardiac surgery in the previous six months. Patients with diabetes and neurological problems were excluded. None of the subjects were taking beta-blockers in either group.

Group 3. Our control group consisted of 25 healthy volunteers without significant cardiac disorders or other pathology that could affect heart rate. It comprised 20 men and five women, mean age 5469. Subjects with more than 10 / min supraventricular and / or ventricular extrasystoles at holter monitoring were excluded. The study protocol was approved by the local Ethics Committee. All of the subjects gave their informed consent.

2.2. Experimental procedures Two-dimensional Echo-Doppler (Vingmed CFM 750 CV) and 24-h Holter monitoring were performed the day before the test of heart rate variability. The ejection fraction (EF) was calculated by the Simpson’s method [19]. Patients were examined in the morning, after a light meal, in a room with constant temperature and humidity. The subjects were trained to breathe in synchrony with a metronome at 15 breath / min (0.25 Hz) to ensure that respiratorylinked variations in heart rate did not overlap with LF heart rate fluctuations (below 0.12 Hz) due to other sources. The study protocol comprised, after adaptation to the environment, three periods, lasting 600 heart cycles each: (i) resting, with subject quietly recumbent; (ii) controlled breathing (RSC), 15 cycles / min, to enhance the vagal-mediated respiratory component of HRV; and (iii) passive orthostatism, produced by a head-up tilt manoeuvre to upright position (808), as a sympathetic stimulus.

2.3. Data acquisition The electrocardiographic signal (single channel, two leads) and the respiratory signal (chest impedance variability) were recorded through an ECG respiratory monitor (Kontron). A dedicated computer programme acquired the electrocardiographic signal sampled at a frequency of 1000 Hz throughout each electrocardiographic R wave.

2.4. Data analysis The recorded time series were analyzed to obtain variability indices in both the time and frequency

S. Scalvini et al. / International Journal of Cardiology 67 (1998) 9 – 17

domains following the method previously described [5,20–22].

2.4.1. Time domain analysis The following time domain measures of heart rate variability were evaluated: the mean of all R–R intervals for each period (resting, controlled breathing, passive orthostatism), and the standard deviation of all R–R intervals (SDRR), which is an index of total HRV. 2.4.2. Frequency domain analysis Heart rate variability measures in the frequency domain were computed by autoregressive spectral techniques [5,20]. This technique calculates the model for data generation mechanism by a least square minimisation of the prediction error. Such a model allows the entire spectrum to be divided into single spectral components, one for each degree of freedom of the model itself. The optimal order of autoregressive model identification was chosen by minimisation of the Akaike Information criteria figure of merit, which provides the best spectral resolution enabling spectral decomposition with automatic identification of the low frequency and high frequency components [23,24]. Where the LF component was not found, the data were re-analyzed with a minimum model order of 12. The frequency ranges were subdivided into 0.03–0.15 Hz (LF) and 0.15– 0.35 Hz (HF), respectively [22–25]. A high frequency component has been used as a marker of vagal activity, whereas both vagal and sympathetic outflows modulate LF. Thus, the LF / HF ratio can be considered as a marker of sympatho-vagal balance [8]. LF / HF ratio increases during sympathetic stimulation and decreases during vagal stimulation. In this study each component has been expressed as the relative power (i.e. percentage of filtered total spectral power) in normalised units (nu) [8,26]. Normalised units are a standardised percentage index of the presence of high and low frequencies in the whole spectrum, calculated as follows: power of LF or HF nu% 2 ]]]]]]]]] total power 2 DC component Normalised units are obtained by dividing the power of a given component by the total variance (from which the component centred at 0.00 Hz, DC com-

11

ponent, has been subtracted) and multiplying this by 100.

3. Statistical analysis All data are expressed as mean6SE. ShapiroWilk’s W-Stat, Kurtosis and Skewness are applied to testing the normality distribution of each variable. Where the test of normality failed, appropriate transformations were utilised. Comparison of general data was performed by one way analysis of variance, T-test and chi-square test. Heart variability data among the three groups and within phases (basal, RSC and TLT) was performed by two-way analysis of variance for repeated measure (correction for degrees of freedom in simple effect of repeated measure was calculated by Greenhouse-Geisser). Where appropriate, additional post-hoc or T-tests with Bonferroni’s correction were applied. The presence of detectable LF component was tested by the Pearson chi-square with Yates’s correction, where appropriate. A probability value of 0.05 was regarded as statistically significant throughout the study.

4. Results Table 1 reports the general characteristics of patients with congestive heart failure (group 1) and asymptomatic left ventricular dysfunction (group 2). There were no differences in gender, mean age, sex, history of a previous myocardial infarction, and use

Table 1 CHF (Group 1)

LVD (Group 2)

30 6269 28 / 2 3464

21 5769 19 / 2 2868

NS NS ,0.01

Myocardial infarct location Anterior (%) Inferior (%)

73 27

58 47

NS NS

Medications Digitalis Diuretics Ace inhibitors Nitrates

80 100 80 33

28 19 71 47

,0.001 ,0.001 NS NS

No. pts Age Male / female LV ejection fraction

(n) (years) (n) (%)

(%) (%) (%) (%)

p

S. Scalvini et al. / International Journal of Cardiology 67 (1998) 9 – 17

12

of vasodilators between the two groups. Compared to group 2, in group 1 a higher portion of patients were on diuretics and digoxin; mean left ventricular ejection fraction was significantly lower in group 1 than in group 2 (2868% vs. 3466%, p,0.01, respectively). Heart rate variability measures in the two groups and in controls are reported in Table 2. Group 1. At baseline, patients with heart failure had a reduced SDRR compared to controls ( p, 0.0003). No differences were found in the R–R mean between group 1 (918624) and controls (916633). LF and LFnu were significantly lower compared to the control group ( p,0.0001); in particular, LF was undetectable in 11 patients. HF did not differ in either group, while HFnu was higher in patients ( p,0.01). As expected, the LF / HF ratio was lower in patients with heart failure than in healthy subjects. The differences in all baseline parameters recorded in patients with heart failure were maintained during parasympathetic and sympathetic stimulation. In contrast to healthy subjects, who had a normal modulation of LF and HF, with an expected HF increase and LF decrease during controlled breathing and a LF increase and a HF decrease during tilting, patients with heart failure failed to exhibit any change in both low and high frequency components under any stimulation. In Fig. 1 HRV power spectra are represented examples of decomposition of the entire

spectrum into the single spectral component from a normal subject, a patient with left ventricular dysfunction and a patient with chronic heart failure, in baseline condition (a), controlled breathing (b) and tilting (c). Group 2: At baseline no significant differences were evidenced both in the time and frequency domain analysis between patients with asymptomatic left ventricular dysfunction and controls. No differences either were present in the R–R mean between group 2 (945640) and controls (918624). Low frequency component was undetectable in one patient. Only during sympathetic stimulation, did the differences between the two groups become evident. In the time domain, compared to controls, SDRR was significantly lower during tilting ( p,0.01), and LF during tilting does not increase ( p,0.0001) as in controls. Group 1 vs group 2: LF was undetectable in 11 patients of group 1 and in one patient of group 2 ( p,0.008). At baseline SDRR ( p,0.03), LF ( p, 0.0001), and LFnu ( p,0.0001) were significantly lower, whereas HFnu was significantly higher ( p, 0.001) in patients with heart failure compared to patients with asymptomatic left ventricular dysfunction. On the contrary, during sympathetic stimulation, group 1 and group 2 exhibited the same behaviour with no modification of any component.

Table 2 Patients and control subjects heart rate variability LF (ms 2 )

LF (nu)

330671 275655 4636132

5464 4664 7067

194634 275654 82612

LVD Rest RSC TLT

228674 221679 86640†

4965 4266 2667†

CHF Rest RSC TLT

42611*§ 56615*§ 43624*§

1964*§ 2363*§ 2565*§

CTRL Rest RSC TLT

HF (ms 2 )

HF (nu)

LF / HF

RR (ms)

RRSD (ms)

3964 5363 2664

2.460.4 1.360.2 7.462

918624 919624 833618

3463 3363 3563

96626 144636 49610†

3365 4365 4265†

2.260.2 1.660.4 4.663†

945640 965639 866638

3063 2762 2562†

144654 161658 3668

5764*§ 5463 4064

1.160.4*§ 0.660.1*§ 1.760.4*

916633 920633 832629

2263*§ 2061*§ 1662*§

* p,0.0001 CHF vs CTRL; † p,0.001 LVD vs CTRL; § p,0.0001 CHF vs LVD. CTR5controls, LVD5patients with left ventrical dysfunction, CHF5patients with chronic heart failure, RSC5controlled breathing, TLT5tilting, LF5low frequency power, HF5high frequency power, nu5normalized unit, RR5mean RR interval, RRSD5RR standard deviation.

S. Scalvini et al. / International Journal of Cardiology 67 (1998) 9 – 17

13

Fig. 1. Examples of decomposition of the entire spectrum into the single spectral components from a normal subject, a patient with left ventricular disfunction (LVD) and a patient with chronic heart failure (CHF), at rest (1a), during control breathing (RSC) (1b) and during tilting (TLT) (1c). In patients with LVD and CHF, LF and HF show a markedly abnormal power in response to various stimuli (RSC, TLT).

14

S. Scalvini et al. / International Journal of Cardiology 67 (1998) 9 – 17

5. Discussion Our data show that autonomic modulation of cardiac rate is abnormal both in asymptomatic and symptomatic patients with post-ischemic left ventricular dysfunction. In particular we see the same changes in severe heart failure and in asymptomatic left ventricular dysfunction with the addition of sympathetic stress. Studies on heart rate variability in ischemic heart disease are focused mainly on patients with acute myocardial infarction using time domain analysis during long-term recording [15–17,27–29]. Heart rate variability has been shown to be reduced within few days and this correlates well with left ventricular dysfunction, peak creatinine kinase, killip class, and poor prognosis [14]. It has been hypothesised that impairment of heart rate variability may predict development of heart failure and unfavourable left ventricular remodelling [30]. However, since heart rate variability may recover within 3 to 6 months after myocardial infarction [31,32], patients with documented myocardial infarction in the previous six months were excluded from our study in order to focus our attention on a group of patients representative of a general population with left ventricular dysfunction. We showed that in controls no subject had un undetectable LF rhythm and hence absolute LF, absolute HF and LF / HF ratio could be considered indices of autonomic control in the normal population. In asymptomatic left ventricular dysfunction only one patient demonstrated an absent LF rhythm and therefore overall in this group of patients spectral analysis at rest can be considered reasonably useful in a quantitative sense in assessing autonomic control. The resting measures of heart rate variability did not differ significantly from normals however the addition of stress testing by controlled breathing or tilting demonstrates significant differences between asymptomatic left ventricular dysfunction and controls, showing the importance of incorporating stress testing to maximise the clinical information available from heart rate variability. In particular in a percentage of these patients with a detectable LF rhythm at rest, the LF rhythm disappears with tilting so that they demonstrate the feature of the absence of an LF rhythm, that of increasing sympathetic discharge (Fig. 1). This can be produced by either increasing severity of heart failure or taking a patient with

asymptomatic left ventricular dysfunction and putting in an additional sympathetic stress. Thus in this group of patients heart rate variability spectral measures at rest are not reliable given that they occasionally lose their LF rhythm in the presence of a slight increase in sympathetic tone. In this group therefore more complicated analysis of heart rate variability are necessary, incorporating the transition to absent LF as indicative of increased sympathetic tone. As confirmed by other investigators, we found heart rate variability to be reduced in chronic heart failure patients [3,10–12]. As expected, indices of heart rate variability recorded both in the time and frequency domain were significantly lower in chronic heart failure patients than in control subjects. In these patients, significant heart rate variability abnormalities are detected at rest, but given that a high proportion of the patients have an absent LF rhythm conventional estimators such as relative LF power and the LF / HF ratio then become unreliable as descriptive terms for autonomic function. With tilt tests the percentage of patients with absent LF rhythm goes up even further so that paradoxically it would appear that the LF / HF ratio decreases with increasing sympathetic tone in heart failure and becomes even less than in controls. The most likely explanation for this is that spectral analysis identifies autonomic modulation for heart rate and not autonomic control, which is lost in these patients. Sympathetic modulation of heart rate is maximal when sympathetic tone is moderately increased so that there is the possibility of both a reduction and an increase in sympathetic tone affecting heart rate. When sympathetic output is maximal there is no possibility for modulation of heart rate in that the sinus node is maximally stimulated so that the heart rate is fixed and no longer fluctuates in the LF or HF band to any appreciable extent. Similar results are reported in previous studies in chronic heart failure patients [1,2,33–35]. Mortara and colleagues [36], in a series of 30 patients with advanced stages of heart failure, observed that patients with undetectable LF component were more severely affected, with more depressed left ventricular function and a higher degree of sympatho-excitation, reflected by higher concentrations of noradrenaline in the plasma. In a recent paper of Van De Borne [37], they found that the complete absence of the LF component in RR

S. Scalvini et al. / International Journal of Cardiology 67 (1998) 9 – 17

interval and in the muscle sympathetic nerve activity (MSNA) characterized those patients with the highest level of MSNA and the most severe heart failure and may have important prognostic implications. The different behaviour of heart rate variability in chronic heart failure patients and healthy subjects was particularly evident during sympathetic and parasympathetic stimulation. While healthy subjects showed a normal modulation of heart rate during tilting and control breathing, this was completely lost in chronic heart failure patients, as confirmed by the lack of changes in the low and high frequency components with respect to baseline values (Fig. 1). Our data are consistent with the results of previous studies in chronic heart failure patients showing an abnormal modulation of neurohumoral activity in response to various stimuli and a reduction in vagal activity [2,9,38–40]. A sympathetic predominance followed by functional denervation with an abnormal response during tilting and controlled breathing was found in 30 patients in different clinical classes of chronic heart failure [41]. The results of this study [41] suggest that autonomic neural modulation and cardiovascular response to neural activity differ at different stages of the disease and stress the importance of stimulation manoeuvres. In conclusion, our preliminary results suggest the hypothesis that progression from asymptomatic to symptomatic left ventricular dysfunction might be associated with a progressive loss of autonomic modulation of cardiac rate, which becomes particularly evident during sympathetic stimulation. If neural derangement is involved in the progression from asymptomatic to overt heart failure, patients with a more pronounced sympatho-vagal imbalance may be at higher risk of cardiac events. Continued follow-up of these patients could shed light on the natural course of autonomic dysfunction and could be of prognostic relevance. Moreover, we stress the difficulty in the use of power spectral analysis measures as a quantitative estimate of autonomic dysfunction in heart failure or of the severity of heart failure, because they do not respond in a linear way to increasing severity, but this can lead to paradoxical changes in any of these parameters. So the absence of LF rhythm can be produced by either increasing severity of heart failure or applying a sympathetic stress to a patient with asymptomatic ventricular

15

dysfunction. In this case, more complicated analysis of heart rate variability are necessary.

6. Condensed abstract We tested the hypothesis that abnormal modulation of heart rate variability may be present in asymptomatic patients with left ventricular dysfunction (LVD) and progress in patients with congestive heart failure (CHF). HRV was measured in 30 patients with CHF; 21 patients with LVD; and 25 healthy volunteers. HRV was evaluated, while subjects were quietly recumbent (BS1), with controlled breathing (RSC) and after tilting (TLT). Group 1 showed a reduction in the SDRR ( p,0.0003) and in the LF ( p,0.0001) compared to normal subjects. LF was not detectable in 11 patients of group 1 ( p,0.0008). On RSC and TLT, group 1 failed to show any modification in the LF and HF under any stimulation. Group 2 showed no modification at baseline evaluation, no increase in the HF on RSC and in LF during TLT compared to controls ( p,0.01 and p,0.0001 respectively). At baseline, group 1 had a lower SDRR ( p,0.03) and LF ( p,0.0001) vs. group 2, whereas during stimulation, the two groups exhibited the same behaviour. In conclusion, reduced HRV is specific for both asymptomatic and symptomatic left ventricular dysfunction and frequency domain analysis allows a more accurate definition of the degree of autonomic control of heart rate.

Acknowledgements Supported by current research grant of ‘Salvatore Maugeri’ Foundation

References [1] Leimbach WH, Wallin G, Victor RG, Aylward PE, Sundlof E, Mark A. Direct evidence from intraneural recordings for central sympathetic outflow in patients with heart failure. Circulation 1986;73:913–9. [2] Eckberg DL, Drabinsky M, Braunwald E. Defective cardiac parasympathetic control in patients with heart disease. N Engl J Med 1971;285:877–83. [3] Saul JP, Arai Y, Berger RD, Lilly LS, Colucci WS, Cohen TJ.

16

[4]

[5]

[6]

[7]

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17]

[18]

[19]

[20]

S. Scalvini et al. / International Journal of Cardiology 67 (1998) 9 – 17 Assessment of autonomic regulation in chronic congestive heart failure. Am J Cardiol 1988;61:1292–9. Ferguson DW, Berg WJ, Roach PJ, Oren RM, Mark AL. Effects of heart failure on baroreflex control of sympathetic neural activity. Am J Cardiol 1992;69:523–31. Askelrod S, Gordon D, Madwed J, Snidman N, Shannon D, Cohen R. Haemodynamic regulation: investigation by spectral analysis. Am J Physiol 1985;18:H867–75. Akselrod S, Gordon D, Ubel FA, Shannon DC, Barger AC, Cohen RJ. Power spectrum analysis of heart rate fluctuations. A quantitative probe of beat-to-beat cardiovascular control. Science 1981;213:220–2. Pomeranz B, Macaulay RJB, Caudill MD, Kutz I, Adam D, Gordon D, et al. Assessment of autonomic function in humans by heart rate spectral analysis. Am J Physiol 1985;248:H151–3. Pagani M, Lombardi F, Guzzetti S, Rimoldi O, Furlan R, Pizzinelli P, et al. Power spectral analysis of heart rate and arterial pressure variabilities as a marker of sympatho-vagal interaction in man and conscious dog. Circ Res 1986;59:178–93. Saul JP, Aray Y, Berger RD, Lilly LS, Colucci WS, Cohen RJ. Assessment of autonomic regulation in chronic congestive heart failure by heart rate spectral analysis. Am J Cardiol 1988;61:1292– 9. Binkley PF, Nunziata E, Haas FJ, Nelson SD, Cody RJ. Mortara, La Rovere. Parasympathetic withdrawal is an integral component of autonomic imbalance in congestive heart failure: demonstration in human subjects and verification in a paced canine model. J Am Coll Cardiol 1991;18:464–72. Casolo G, Balli E, Taddei T, Amuhasi J, Gori C. Decreased spontaneous heart rate variability in congestive heart failure. Am J Cardiol 1989;64:1162–7. Kienzle MG, Ferguson DW, Birkett CL, Myers GA, Berg WJ, Mariano DJ. Clinical, hemodynamic and sympathetic neural correlates of heart rate variability in congestive heart failure. Am J Cardiol 1992;69:761–7. Kleiger RE, Miller JP, Bigger JT, Moss AJ. Decreased heart rate variability and its association with increased mortality after acute myocardial infarction. Am J Cardiol 1987;59:256–62. Casolo GC, Stroder P, Signorini C, et al. Heart rate variability during the acute phase of myocardial infarction. Circulation 1992;85:2073–9. Bigger JT, Fleiss JL, Steinman RC, Rolnitzky LM, Kleiger RE, Rottam JN. Correlation among time and frequency domain measures of heart period variabilitiy two weeks after acute myocardial infarction. Am J Cardiol 1992;69:891–8. Cripps TR, Malik M, Farrell TM, Camm AJ. Prognostic value of reduced heart rate variability after myocardial infarction: clinical evaluation of a new analysis method. Br Heart J 1991;65:14–9. Farrell TG, Bashir Y, Cripps T, et al. Risk stratification for arrthythmic events in postinfarction patients based on heart rate variability, ambulatory electrocardiografic variables and the signalaverage electrocardiogram. J Am Coll Cardiol 1991;18:687–97. Martin GJ, Magid NM, Myers G, et al. Heart rate variability and sudden death secondary to coronary artery disease during ambulatory electrocardiographic monitoring. Am J Cardiol 1987;60:86–9. Rogers EW, Feigenbaum H, Weiman AE. Echocardiography for quantitation of cardiac chambers. Progress in Cardiology 1979, vol. 8, Philadelphia: Lea and Febiger. Askelrod S, Gordon D, Ubel FA, Shannon DC, Barger AC, Cohen RJ. Power spectrum analysis of heart rate fluctuations: a quantitative probe for beat-to-beat cardiovascular control. Science 1981;213:200–22.

[21] Colombo R, Mazzuero G, Soffiantino F, Ardizzoia M, Minuco G. A comprehensive PC solution to heart rate variability: analysis in mental stress. Comp Cardiol 1989:475–8. [22] Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology. Heart rate variability. Standards of measurement, physiological interpretation and clinical use. Circulation 1996;93:1043–65. [23] Baselli G, Cerruti S, Civardi S, Liberati D, Lombardi F, Malliani A, et al. Spectral and cross spectral analysis of heart rate and arterial blood pressure variability signals. Comput Biomed Res 1986;19:520–34. [24] Baselli G, Cerruti S, Civardi S, Lombardi F, Malliani A, Merri M. Heart rate variability signal processing: a quantitative approach as an aid to diagnosis in cardiovascular pathologies. Int J Biomed Comput 1987;20:51–70. [25] Carpeggiani C, Macerata A, Edmin M, Michelassi C, Balocchi R, L’Abbate A. Power spectral analysis of heart rate viability signal: a multicenter study. In: Cerruti S, Minuco G, editors. Spectral analysis of heart rate variability signal. Pavia, Italy: La Gogliardica Pavese, 1991:133–40. [26] Malliani A, Pagani M, Lombardi F, Cerruti S. Cardiovascular neural regulation explored in the frequency domain. Circulation 1991;84:482–92. [27] Lombardi F, Sandrone G, Pernpruner S. Heart rate variability as an index of sympathovagal interaction after acute myocardial infarction. Am J Cardiol 1987;60:1239–45. [28] Bigger JT, Fleiss JL, Steinman RC, Rolnitzky LM, Kleiger RE, Rottman JN. Frequencies domain measures of heart period variability and mortality after infarction. Circulation 1992;85:164–71. [29] Bigger JT, Fleiss JL, Rolnitzky LM, Steinman RC. Frequencies domain measures of heart rate variability to assess risk late after myocardial infarction. J Am Coll Cardiol 1993;21:729–36. [30] Dambrink JHE, Tuininga YS, van Gilst WH, Kingma JH. Left ventricular dilatation predicted by reduced heart rate variability in survivors of a first anterior myocardial infarction. Br Heart J 1994;72:514–20. [31] Lombardi F, Sandrone G, Mortara A, et al. Circardian variation of spectral indices of heart rate variability after myocardial infarction. Am Heart J 1992;123:1521–9. [32] Bigger JT, Kleiger RE, Fleiss JL, Rolnitzky LM, Steinman RC, Miller JP, the Multicenter Post-Infarction Research Group. Components of heart rate variability measured during healing of acute myocardial infarction. Am J Cardiol 1988;61:208–15. [33] Cohn JN, Levine TB, Olivari MT, et al. Plasma norepinephrine as a guide to prognosis in patients with chronic congestive heart failure. N Engl J Med 1984;311:819–23. [34] Ajikik L, Murakawa Y, Yanagisawa-Miwa A, Usui M, Yamashita T, Oikawa N, Inoue H. Autonomic nervous system activity in idiopathic dilated cardiomyopathy and in hypertrophic cardiomyopathy. Am J Cardiol 1993;71:1316–20. [35] Casolo G, Balli E, Fati A, Gori C, Freni A, Geussini G. Twentyfour-hour spectral analysis of heart rate variability in congestive heart failure secondary to coronary artery disease. Am J Cardiol 1991;67:1154–8. [36] Mortara A, La Rovere MT, Signorini MG, et al. Can power spectral analysis of heart rate variability identify a high risk subgroup of congestive heart failure patients with excessive sympathetic activation? Pilot study before and after heart transplantation. Br Heart J 1994;71:422–30. [37] Van De Borne P, Montano N, Pagani M, Oren R, Somers VK. Absence of low-frequency variability of sympathetic nerve activity in severe heart failure. Circulation 1997;95(6):1449–54.

S. Scalvini et al. / International Journal of Cardiology 67 (1998) 9 – 17 [38] Olivari MT, Levine TB, Cohn JN. Abnormal neurohumoral response to nitroprusside infusion in congestive heart failure. JACC 1983;2:411–7. [39] Goldsmith SR, Francis GS, Levine TB, Cohn JN. Regional blood flow response to orthostasis in patients with congestive heart failure. Am J Cardiol 1983;51:1391–5. [40] Kubo SH, Cody RJ. Circulatory autoregulation in chronic congestive

17

heart failure: response to head-up tilt in 41 patients. Am J Cardiol 1983;52:512–8. [41] Guzzetti S, Cogliati C, Turriel M, Crema C, Lombardi F, Malliani A. Sympathetic predominance followed by functional denervation in the progression of chronic heart failure. Eur Heart J 1995;16:1100– 7.