- Email: [email protected]
Autonomic modulation and clinical outcome in patients with chronic heart failure
International Journal of Cardiology 100 (2005) 247 – 251 www.elsevier.com/locate/ijcard
Autonomic modulation and clinical outcome in patients with chronic heart failure Antonio Franco Folinoa,T, Barbara Tokajuk a, Alberto Portab, Silvia Romanoa, Sergio Cerutti c, Sergio Dalla Voltaa b
a Department of Cardiology, University of Padua, Italy Department of Preclinic Sciences, LITA di Vialba, University of Milan, Italy c Department o Bioengineering, Polytechnic of Milan, Italy
Received 17 May 2004; received in revised form 3 August 2004; accepted 7 August 2004 Available online 19 February 2005
Abstract Objective: Several studies documented the relevance of autonomic activity in the pathophysiology of heart failure. In our study we evaluated the adjustment of this activity under different stimuli, by means of heart rate variability (HRV), and correlated these findings with long-term mortality and sustained VT occurrence. Patients and method: Fifty-three patients (mean age 54F9 years) with heart failure were submitted to time and frequency domain HRV analysis. This latter analysis was performed at rest, during paced breathing and during passive tilt. Results: Lower standard deviation of RR intervals (76.76F24 versus 107.70F43, p=0.02), mean of the 5-min standard deviations of RR intervals (35.14F15 versus 62.39, pb0.01), standard deviation of the 5-min average RR intervals (69.42F19 versus 91.79F30, p=0.02), and baseline Low-Frequency (LF) power (15.15F12 versus 40.39F24 nu, p=0.001) characterized patients who died. Paced breathing induced a significant reduction of LF (40.39F24 to 20.12F18 nu, pb0.0001) and increase of High Frequency power (HF) (47.31F23 to 70.63F16 nu, pb0.0001) in survivors, while tilting induced a reduction of HF (47.31F23 to 29.80F16 nu, pb0.0001). Patients who died did not show significant variation of HRV neither during paced breathing nor during tilt. Reduced time domain indexes were significantly correlated to sustained VT occurrence. Conclusions: Patients with heart failure with a better prognosis are characterized by a responsiveness of autonomic modulation. Simple maneuvers, such as tilting and paced breathing, seem to provide more useful information, than the baseline evaluation of autonomic status, in identifying patients with a higher mortality. Time domain analysis was more helpful to estimate arrhythmic risk. D 2005 Elsevier Ireland Ltd. All rights reserved. Keywords: Heart failure; Autonomic nervous system; Heart rate variability; Ventricular arrhythmias; Mortality
1. Introduction Several studies stressed the relevance of heart rate variability (HRV) in the risk stratification after myocardial infarction [1,2]. This method has been also used in other cardiovascular diseases with similar results [3,4]. Particularly, HRV was analyzed in patients with heart failure [5–7], T Corresponding author. Clinica Cardiologica-Universita` di Padova, Via Giustiniani, 2 35128 Padova PD, Italy. Tel.: +39 49 8212426; fax: +39 49 8761764. E-mail address: [email protected] (A.F. Folino). 0167-5273/$ - see front matter D 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ijcard.2004.08.057
a condition where autonomic nervous system plays a relevant role both in the pathophysiology of the disease and in the regulation of myocardial electrical stability [8,9]. Besides evaluation of baseline autonomic activity, the study of the modulation of such activity can provide more relevant prognostic information. Baroreflex sensitivity, for instance, tests the autonomic reactivity to an acute blood pressure increase, induced by means of a pharmacological stimulation, providing a more accurate prognostic stratification after myocardial infarction [10]. More recently, the modification of RR intervals after ventricular premature beats was recognized as a postinfarction risk stratifier
248
A.F. Folino et al. / International Journal of Cardiology 100 (2005) 247–251
independent and stronger than other risk predictors [11]. Also this method is aimed to test the reciprocal reactivity of sympathetic and parasympathetic systems, in response to hemodynamic perturbations induced by a premature ventricular beat. Therefore, considering the importance of a complete autonomic evaluation, including baseline profile and elicited activity, in this prospective study we analyzed HRV in patients with congestive heart failure evaluating autonomic activity at rest, and after two simple maneuvers inducing respectively an increase of sympathetic and vagal output: passive orthostatism, and paced breathing [12]. Moreover, these data have been correlated to long-term mortality and sustained ventricular tachycardia occurrence.
2. Methods 2.1. Study population The study cohort comprised 53 patients (41 males, 12 females) with a mean age of 54F9 years (range 33–73 years). They were affected by dilated cardiomyopathy: idiopathic in 14 patients, and secondary to coronary artery disease in 39. Sixteen patients were included in the New York Heart Association Class II, 27 in Class III, and 10 in Class IV. All of them underwent coronary arteriography. At the moment of the evaluation none of them was taking medications potentially interfering with autonomic activity; permitted therapies were diuretics, nitrates, digoxin. Also ACE-inhibitors were consented, considering the limited influence on sympathetic activity [13]. Treatment with hblocking agents was considered an exclusion criteria.
The acquisition of the ECG signal to perform frequency domain analysis of HRV was performed in the morning, in a quite room, using an acquisition module with a sampling frequency of 350 Hz. A simultaneous recording of the respiratory rate was obtained by mean of a thoracic belt with strain sensor. Three periods of 15 min each of ECG were recorded: in supine position, during paced breath (12 breaths/min, controlled by a metronome), and in upright tilt position (Tilt table at 608). Inside these periods a stable segment, without artifact, of 300 beats was selected. For the upright position the segment was selected after at least 5 min of tilting, when the tachogram showed a stationary state. This series length was kept constant in the analysis. The power spectrum was estimated using autoregressive technique [14]. Briefly, the coefficients of the autoregressive model were calculated via Levinson–Durbin recursion and the model order was selected according to the Akaike figure of merit in the range from 8 to 14. The power spectrum was decomposed in components labeled as low frequency (LF) if the central frequency was between 0.04 and 0.15 Hz and high frequency (HF) if centered around the breathing rate (F0.05 Hz). The power of LF and HF components was expressed in absolute and normalized units (i.e. the absolute power divided by the variance minus the power below 0.04 Hz); however, for the analysis of the results the normalized units were considered. The recordings, collections and elaboration of the results were made according to the guidelines of the Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology [15]. The procedures followed were in accordance with our institutional guidelines.
2.2. Baseline evaluation 2.3. Statistical analysis All patients were studied with 2-D echocardiography, and the parameters considered were left ventricle enddiastolic volume (LVEDV) and left ventricle ejection fraction (LVEF). The time domain heart rate variability was analyzed during normal unrestricted in-hospital activity, by a 24-h Holter monitoring (Del Mar tape recorders). Ambulatory ECGs were analyzed by a Del Mar analyzer (563/A-operator controlled analysis); normal and aberrant complex were discriminated and all the adjacent intervals between normal beats (NN) during the 24 h were computed. The NN intervals were then analyzed by a time domain method for heart rate variability and the following indexes were calculated: the standard deviation of all NN intervals (SDNN), the mean of the 5-min standard deviation of NN intervals calculated over 24 h (SDNN index), the standard deviation of the average NN intervals calculated over 5-min intervals (SDANN), the square root of the mean squared differences of successive NN intervals (RMSSD), the percentages of the number of interval differences of successive NN intervals greater than 50 ms (pNN50).
The data are expressed as meanFstandard deviation. The differences of indexes in patients who died and survivors were analyzed by between-groups ANOVA multivariate analysis of variance. Correlation between two variables was evaluated by means of linear regression analysis and
Table 1 Clinical features and time domain heart rate variability in all the patients MeanFSD Age (yrs) LVEDV (ml/m2) LVEF (%) SDNN (ms) RMSSD (ms) SDNN index (ms) SDANN (ms) pNN50 (%)
54F9 134F32 30F9 101F41 77F65 56F31 87F29 10F7
LVEDV: left ventricle end-diastolic volume; LVEF: left ventricle ejection fraction.
A.F. Folino et al. / International Journal of Cardiology 100 (2005) 247–251 Table 2 Time and frequency domain heart rate variability at rest in survivors and nonsurvivors Patients (n) SDNN (ms) RMSSD (ms) SDNN index (ms) SDANN (ms) pNN50 (%) LF (ms2) LF (nu) HF (ms2) HF (nu) L/H
Survivors
Nonsurvivors
p
41 108F43 87F69 62F32 92F30 11F8 84F96 40F24 103F83 47F23 1.62F3
12 77F24 45F39 35F15 69F19 6F4 11F17 15F12 58F109 53F34 0.8F1
=0.02 NS b0.01 =0.02 NS =0.01 =0.001 NS NS NS
See explanation on text.
expressed as the correlation coefficient (r). Statistical significance was assumed with a level of pb0.05.
3. Results The baseline characteristics and time domain indexes of HRV for all patients are detailed in Table 1. The mean duration of the follow-up was 1585F156 days. Twelve patients died (22.6%), with an annual mortality rate of 5.13%. The causes of death were progressive heart failure with multiorgan dysfunction in 7 patients and sudden cardiac death in 5. Seven patients who died were classified in NYHA class III, and 5 in class IV. The survivors evidenced a higher LVEF (31.90% versus 22.42%, p=0.001). Twelve patients experienced episodes of sustained ventricular tachycardia, and six of them died in the following period. Arrhythmic patients were treated by Amiodarone, and an ICD was implanted in 3 subjects. The analysis of time domain HRV showed significantly reduced values of SDNN, SDNN index, and SDANN in nonsurvivors. Also RMSSD showed reduction, but without reaching the statistical significance, probably because of the large variability of its values (Table 2). The analysis of frequency domain parameters, obtained at rest, evidenced a Table 3 Time and frequency domain heart rate variability at rest in patients with or without episodes of sustained ventricular tachycardia Patients (n) SDNN (ms) RMSSD (ms) SDNN index (ms) SDANN (ms) pNN50 (%) LF (ms2) LF (nu) HF (ms2) HF (nu) L/H
Sustained VT
No sustained VT
p
12 61F25 32F11 28F11 63F19 7F6 18F44 29F29 56F87 48F35 1.19F2
41 1122F38 91F69 65F30 94F28 10F7 82F95 36F23 104F89 49F23 1.51F3
b0.0001 =0.005 =0.0001 b0.001 NS b0.05 NS NS NS NS
See explanation on text.
249
Table 4 Modifications of frequency domain heart rate variability during rhythmic breathing and tilt in survivors and nonsurvivors
Survivors LF (ms2) LF (nu) HF (ms2) HF (nu) L/H
Baseline
Tilt
p*
Paced breathing
p**
84F96 40F24 103F83 47F23 1.62F3
75F70 49F26 42F30 30F16 2.51F3
NS NS b0.00001 b0.00001 NS
53F61 20F18 177F131 71F16 0.49F1
NS b0.00002 b0.005 =0.000001 b0.05
34F41 26F22 54F104 38F30 2.14F2
NS NS NS NS =0.01
13F24 18F25 106F75 61F27 0.87F1
NS NS NS NS NS
Nonsurvivors LF (ms2) 11F17 LF (nu) 15F12 58F109 HF (ms2) HF (nu) 53F34 L/H 0.80F1
*p value comparing baseline versus tilt; **p value comparing baseline versus paced breathing.
significant lower value of LF, both in normalized and absolute units in patients who died (Table 2). In the 12 patients with episodes of sustained ventricular tachycardia all but one time domain indexes of HRV were markedly depressed in comparison of patients without severe ventricular arrhythmias. In arrhythmic patients also the LF component at rest was reduced, but only if considered in absolute units (Table 3). Considering the autonomic modification induced by the provocative maneuvers, survivors had a significant reduction of LF in normalized units, and a significant increase of HF, both in absolute and normalized units, during paced breathing, while tilting induced only a significant reduction of the HF component, both in absolute and normalized units. The L/H ratio changed significantly in this group only with paced breathing, confirming a shift of the sympathovagal balance toward an increase of parasympathetic activity. On the contrary, patients who died did not show Table 5 Modifications of frequency domain heart rate variability during rhythmic breathing and tilt in patients with or without episodes of sustained ventricular tachycardia Baseline
Tilt
p*
Paced breathing
p**
Sustained VT LF (ms2) 18F44 LF (nu) 29F29 HF (ms2) 56F87 HF (nu) 48F35 L/H 1.19F2
61F35 52F30 27F18 23F16 3.65F3
=0.01 b0.0005 NS =0.01 NS
6F14 8F16 114F101 74F16 0.17F0.4
NS =0.01 b0.005 =0.01 NS
No sustained VT LF (ms2) 82F95 LF (nu) 36F23 HF (ms2) 104F89 HF (nu) 49F23 L/H 1.51F3
67F74 41F26 50F61 34F20 2.07F2
NS NS =0.00002 b0.00005 NS
55F60 23F19 175F127 67F20 0.69F1
NS b0.005 =0.005 b0.0001 NS
*p value comparing baseline versus tilt; **p value comparing baseline versus paced breathing.
250
A.F. Folino et al. / International Journal of Cardiology 100 (2005) 247–251
significant modifications of both indexes, neither during paced breathing nor during tilt, in spite of a significant modification of L/H induced by tilt reflecting an increase of sympathetic activity (Table 4). Considering patients with or without episodes of sustained ventricular tachycardia, the same maneuvers, did not provide useful results to identify patients at higher risk for severe ventricular arrhythmias (Table 5).
4. Discussion Autonomic nervous system plays a crucial role in the mechanisms of heart failure, being involved from the early stages of the disease and acting on different cardiovascular structures. The characteristic increase of sympathetic activity, as well as the blunted vagal tone, has their main clinical relevance on the electrical instability and therefore on mortality. The more reliable non-invasive method to assess autonomic activity is the study of HRV. It provides useful information on parasympathetic and sympathetic tone, and if depressed is a powerful predictor of mortality in patients after acute myocardial infarction. However, besides the evaluation of baseline autonomic activity, in the last years, other tests demonstrated an even stronger correlation with mortality, through the analysis of autonomic activity modifications induced by different stimuli. One of the most important is the study of baroreflex sensitivity, where a pharmacological induced increase of sympathetic activity provides information about the reflex vagal response [10]. Similarly, analysis of RR intervals changes after a premature ventricular beat, known as heart rate turbulence, assess the modifications of autonomic activity induced by an endogenous stimulus [16]. More recently, a reduced short-term low-frequency power during controlled breathing was demonstrated, a powerful predictor of sudden death in patients with chronic heart failure [17]. In our study, we investigated the modifications of autonomic activity in patients with heart failure, induced by means of simple maneuvers such as tilting and paced breathing, and its correlations with mortality. Baseline evaluation of time domain HRV analysis showed a global reduction of all indexes in patients who died, that reached the statistical significance for SDNN, SDNN index, and SDANN. Similarly, the study of baseline frequency domain analysis evidenced that the most important characteristic of the patients who died was a significantly lower LF components. As reported in previous studies, patients with heart failure present different patterns of heart rate variability [18–20], often based on the severity of their clinical condition. However, in heart failure, the interpretation of spectral components of HRV is very difficult, and a specific correlation with parasympathetic or sympathetic activity, cannot be made. The association of reduced LF components and a worse prognosis in patients with heart failure is, however, similar
to that reported in previous studies, in which HRV was evaluated in basal conditions [18,21,22]. Even if a conventional classification of spectral components is compromised, their variation under different stimuli can be evaluated, and the provocative maneuvers adopted in our protocol seem to add crucial information about the autonomic activity in failing hearts. In comparison with the study of La Rovere et al. [17] we explored also autonomic modulation induced by passive orthostatism, obtained by mean of a tilting table, as similarly tested by Guzzetti [18], and the results obtained provided useful information to stratify the risk in our cohort of patients. The LF component showed, during tilt, a minimal, nonsignificant increase in both survivors and nonsurvivors. Nevertheless, this index decreased significantly in survivors during the paced breathing, whereas remained almost unchanged in nonsurvivors. The modifications of HF component were more distinct: it remains unchanged either during tilting or during paced breathing in patients who died, whereas survivors evidenced a reduction during tilt, and an increase during paced breathing. It seems evident that survivors maintain some degree of responsiveness of autonomic modulation, particularly, generating signals included in the HF band. Also, the LF components showed a residual responsiveness induced by paced breathing. On the contrary, in patients who died, these mechanisms were unable to react to the two maneuvers tested. The analysis of HRV in patients with or without episodes of sustained ventricular tachycardia provided discouraging results. However, it is noteworthy that time domain indexes provided stronger correlations between arrhythmic and nonarrhythmic patients, than between survivors and nonsurvivors. On the other hand, the modulating maneuvers associated with HRV analysis were not useful to stratify the arrhythmic risk, providing similar results in patients with or without sustained ventricular tachycardia. In conclusion, our findings seem to indicate that patients with heart failure and a better prognosis are characterized by the presence of still active mechanisms of autonomic modulation. On the contrary, patients who died demonstrated a rigidity of autonomic nervous system, unable to react to some evocative stimuli. Therefore, simple maneuvers, as tilting and paced breathing, associated with HRV analysis, seem to provide more useful information, than simple baseline evaluation of autonomic status, in identifying patients with heart failure at higher risk. Time domain indexes of HRV seem more effective than frequency domain parameters, at rest and during evocative maneuvers, in arrhythmic risk stratification. References [1] Kleiger RE, MIller JP, Bigger JT, Moss AJ, Multicenter Postinfarction Research Group. Decreased heart rate variability and its association with increased mortality after acute myocardial infarction. Am J Cardiol 1987;59:256 – 62.
A.F. Folino et al. / International Journal of Cardiology 100 (2005) 247–251 [2] Bigger JT, Fleiss JM, Steinman NC, Rolnitzkky LM, Kleiger RE, Rottman JN. Frequency domain measures of heart period variability and mortality after myocardial infarction. Circulation 1992;85:164 – 71. [3] Counihan PJ, Fei L, Bashir Y, Farrel TG, Haywood GA, McKenna WJ. Assessment of heart rate variability in hypertrophic cardiomyopathy. Association with clinical and prognostic features. Circulation 1993;88:1682 – 90. [4] Fei L, Keeling J, Gill JS, Bashir Y, Statters DJ, Poloniecki J, et al. Heart rate variability and its relations to ventricular arrhythmias in congestive heart failure. Br Heart J 1994;71:322 – 8. [5] 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. [6] Kingwell BA, Thompson JM, Kaye DM, McPherson GA, Jennings MD, Esler MD. Heart rate spectral analysis, cardiac norepinephrine spillover, and muscle sympathetic nerve activity during human sympathetic nervous activation and failure. Circulation 1994;90: 234 – 40. [7] Tygesen H, Rundqvist B, Waagstein F, Wennerblom B. Heart rate variability measurement correlates with cardiac norepinephrine spillover in congestive heart failure. Am J Cardiol 2001;87:1308 – 11. [8] Nolan J, Batin PD, Andrews R, Lindsay SJ, Brooksby P, Mullen M, et al. Prospective study of heart rate variability and mortality in chronic heart failure. Results of the United Kingdom heart failure evaluation and assessment of risk trial. Circulation 1998;98:1510 – 6. [9] Bilchick KC, Fetics B, Djoukeng R, Gross Fisher S, Fletcher RD, Singh SN, et al. Prognostic value of heart rate variability in chronic congestive heart failure (Veteran affairs’ survival trial of antiarrhythmic therapy in congestive heart failure). Am J Cardiol 2002;90:24 – 8. [10] La Rovere MT, Bigger JT, Marcus FI, Mortara A, Schwartz PJ, for the ATRAMI. Baroreflex sensitivity and heart rate variability in predictor of total cardiac mortality after myocardial infarction. Lancet 1998;351:478 – 84. [11] Schmidt G, Malik M, Barthel P, Schneider R, Ulm K, Rolnitzky L, et al. Heart rate turbulence after ventricular premature beats as a predictor of mortality after acute myocardial infarction. Lancet 1999; 353:1390 – 6. [12] Malliani A, Pagani M, Lombardi F, Cerutti S. Cardiovascular neural regulation explored in the frequency domain. Circulation 1991;84: 482 – 92.
251
[13] Rimoldi O, Pagani MR, Piazza S, Pagani M, Malliani A. Restraining effects of captopril on sympathetic excitatory responses in dogs: a spectral analysis approach. Am J Physiol 1994;267:1608 – 18. [14] 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. [15] Task force of the European Society of Cardiology, and the North American Society of Pacing and Electrophysiology. Heart rate variability. Standard of measurement, physiological interpretation, and clinical use. Circulation 1996;93:1043 – 65. [16] Schmidt G, Malik M, Barthel P, Schneider R, Ulm K, Rolnitzky L, et al. Heart-rate turbulence after ventricular premature beats as predictor of mortality after acute myocardial infarction. Lancet 1999; 353:1390 – 9. [17] La Rovere MT, Pinna GD, Maestri R, Mortara A, Capomolla S, Febo O, et al. Short-term heart rate variability strongly predicts sudden cardiac death in chronic heart failure patients. Circulation 2003;107:565 – 70. [18] Guzzetti S, Cogliati C, Turiel 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. [19] Van De Borne P, Montano N, Pagani M, Oren RM, Somers VK. Absence of low frequency variability of sympathetic nerve activity in severe heart failure. Circulation 1997;95:1449 – 14455. [20] Notarius CF, Butler GC, Ando S, Pollard MJ, Senn BL, Floras JS. Dissociation between microneurographic and heart rate variability estimates of sympathetic tone in normal subjects and in patients with heart failure. Clin Sci 1999;96:557 – 656. [21] Mortara A, La Rovere MT, Signorini MG, Pantaleo P, Pinna G, Martinelli L, et al. Can power spectral analysis of heart rate variability identify a high risk subgroup of congestive heart failure patients with excessive sympathetic activation? A pilot study before and after heart transplantation. Br Heart J 1994;71:422 – 30. [22] Galinier M, Pathak A, Fourcade J, Androdias C, Curnier D, Varnous S, et al. Depressed low-frequency power of heart rate variability as an independent predictor of sudden death in chronic heart failure. Eur Heart J 2000;21:475 – 82.
Autonomic modulation and clinical outcome in patients with chronic heart failure Antonio Franco Folinoa,T, Barbara Tokajuk a, Alberto Portab, Silvia Romanoa, Sergio Cerutti c, Sergio Dalla Voltaa b
a Department of Cardiology, University of Padua, Italy Department of Preclinic Sciences, LITA di Vialba, University of Milan, Italy c Department o Bioengineering, Polytechnic of Milan, Italy
Received 17 May 2004; received in revised form 3 August 2004; accepted 7 August 2004 Available online 19 February 2005
Abstract Objective: Several studies documented the relevance of autonomic activity in the pathophysiology of heart failure. In our study we evaluated the adjustment of this activity under different stimuli, by means of heart rate variability (HRV), and correlated these findings with long-term mortality and sustained VT occurrence. Patients and method: Fifty-three patients (mean age 54F9 years) with heart failure were submitted to time and frequency domain HRV analysis. This latter analysis was performed at rest, during paced breathing and during passive tilt. Results: Lower standard deviation of RR intervals (76.76F24 versus 107.70F43, p=0.02), mean of the 5-min standard deviations of RR intervals (35.14F15 versus 62.39, pb0.01), standard deviation of the 5-min average RR intervals (69.42F19 versus 91.79F30, p=0.02), and baseline Low-Frequency (LF) power (15.15F12 versus 40.39F24 nu, p=0.001) characterized patients who died. Paced breathing induced a significant reduction of LF (40.39F24 to 20.12F18 nu, pb0.0001) and increase of High Frequency power (HF) (47.31F23 to 70.63F16 nu, pb0.0001) in survivors, while tilting induced a reduction of HF (47.31F23 to 29.80F16 nu, pb0.0001). Patients who died did not show significant variation of HRV neither during paced breathing nor during tilt. Reduced time domain indexes were significantly correlated to sustained VT occurrence. Conclusions: Patients with heart failure with a better prognosis are characterized by a responsiveness of autonomic modulation. Simple maneuvers, such as tilting and paced breathing, seem to provide more useful information, than the baseline evaluation of autonomic status, in identifying patients with a higher mortality. Time domain analysis was more helpful to estimate arrhythmic risk. D 2005 Elsevier Ireland Ltd. All rights reserved. Keywords: Heart failure; Autonomic nervous system; Heart rate variability; Ventricular arrhythmias; Mortality
1. Introduction Several studies stressed the relevance of heart rate variability (HRV) in the risk stratification after myocardial infarction [1,2]. This method has been also used in other cardiovascular diseases with similar results [3,4]. Particularly, HRV was analyzed in patients with heart failure [5–7], T Corresponding author. Clinica Cardiologica-Universita` di Padova, Via Giustiniani, 2 35128 Padova PD, Italy. Tel.: +39 49 8212426; fax: +39 49 8761764. E-mail address: [email protected] (A.F. Folino). 0167-5273/$ - see front matter D 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ijcard.2004.08.057
a condition where autonomic nervous system plays a relevant role both in the pathophysiology of the disease and in the regulation of myocardial electrical stability [8,9]. Besides evaluation of baseline autonomic activity, the study of the modulation of such activity can provide more relevant prognostic information. Baroreflex sensitivity, for instance, tests the autonomic reactivity to an acute blood pressure increase, induced by means of a pharmacological stimulation, providing a more accurate prognostic stratification after myocardial infarction [10]. More recently, the modification of RR intervals after ventricular premature beats was recognized as a postinfarction risk stratifier
248
A.F. Folino et al. / International Journal of Cardiology 100 (2005) 247–251
independent and stronger than other risk predictors [11]. Also this method is aimed to test the reciprocal reactivity of sympathetic and parasympathetic systems, in response to hemodynamic perturbations induced by a premature ventricular beat. Therefore, considering the importance of a complete autonomic evaluation, including baseline profile and elicited activity, in this prospective study we analyzed HRV in patients with congestive heart failure evaluating autonomic activity at rest, and after two simple maneuvers inducing respectively an increase of sympathetic and vagal output: passive orthostatism, and paced breathing [12]. Moreover, these data have been correlated to long-term mortality and sustained ventricular tachycardia occurrence.
2. Methods 2.1. Study population The study cohort comprised 53 patients (41 males, 12 females) with a mean age of 54F9 years (range 33–73 years). They were affected by dilated cardiomyopathy: idiopathic in 14 patients, and secondary to coronary artery disease in 39. Sixteen patients were included in the New York Heart Association Class II, 27 in Class III, and 10 in Class IV. All of them underwent coronary arteriography. At the moment of the evaluation none of them was taking medications potentially interfering with autonomic activity; permitted therapies were diuretics, nitrates, digoxin. Also ACE-inhibitors were consented, considering the limited influence on sympathetic activity [13]. Treatment with hblocking agents was considered an exclusion criteria.
The acquisition of the ECG signal to perform frequency domain analysis of HRV was performed in the morning, in a quite room, using an acquisition module with a sampling frequency of 350 Hz. A simultaneous recording of the respiratory rate was obtained by mean of a thoracic belt with strain sensor. Three periods of 15 min each of ECG were recorded: in supine position, during paced breath (12 breaths/min, controlled by a metronome), and in upright tilt position (Tilt table at 608). Inside these periods a stable segment, without artifact, of 300 beats was selected. For the upright position the segment was selected after at least 5 min of tilting, when the tachogram showed a stationary state. This series length was kept constant in the analysis. The power spectrum was estimated using autoregressive technique [14]. Briefly, the coefficients of the autoregressive model were calculated via Levinson–Durbin recursion and the model order was selected according to the Akaike figure of merit in the range from 8 to 14. The power spectrum was decomposed in components labeled as low frequency (LF) if the central frequency was between 0.04 and 0.15 Hz and high frequency (HF) if centered around the breathing rate (F0.05 Hz). The power of LF and HF components was expressed in absolute and normalized units (i.e. the absolute power divided by the variance minus the power below 0.04 Hz); however, for the analysis of the results the normalized units were considered. The recordings, collections and elaboration of the results were made according to the guidelines of the Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology [15]. The procedures followed were in accordance with our institutional guidelines.
2.2. Baseline evaluation 2.3. Statistical analysis All patients were studied with 2-D echocardiography, and the parameters considered were left ventricle enddiastolic volume (LVEDV) and left ventricle ejection fraction (LVEF). The time domain heart rate variability was analyzed during normal unrestricted in-hospital activity, by a 24-h Holter monitoring (Del Mar tape recorders). Ambulatory ECGs were analyzed by a Del Mar analyzer (563/A-operator controlled analysis); normal and aberrant complex were discriminated and all the adjacent intervals between normal beats (NN) during the 24 h were computed. The NN intervals were then analyzed by a time domain method for heart rate variability and the following indexes were calculated: the standard deviation of all NN intervals (SDNN), the mean of the 5-min standard deviation of NN intervals calculated over 24 h (SDNN index), the standard deviation of the average NN intervals calculated over 5-min intervals (SDANN), the square root of the mean squared differences of successive NN intervals (RMSSD), the percentages of the number of interval differences of successive NN intervals greater than 50 ms (pNN50).
The data are expressed as meanFstandard deviation. The differences of indexes in patients who died and survivors were analyzed by between-groups ANOVA multivariate analysis of variance. Correlation between two variables was evaluated by means of linear regression analysis and
Table 1 Clinical features and time domain heart rate variability in all the patients MeanFSD Age (yrs) LVEDV (ml/m2) LVEF (%) SDNN (ms) RMSSD (ms) SDNN index (ms) SDANN (ms) pNN50 (%)
54F9 134F32 30F9 101F41 77F65 56F31 87F29 10F7
LVEDV: left ventricle end-diastolic volume; LVEF: left ventricle ejection fraction.
A.F. Folino et al. / International Journal of Cardiology 100 (2005) 247–251 Table 2 Time and frequency domain heart rate variability at rest in survivors and nonsurvivors Patients (n) SDNN (ms) RMSSD (ms) SDNN index (ms) SDANN (ms) pNN50 (%) LF (ms2) LF (nu) HF (ms2) HF (nu) L/H
Survivors
Nonsurvivors
p
41 108F43 87F69 62F32 92F30 11F8 84F96 40F24 103F83 47F23 1.62F3
12 77F24 45F39 35F15 69F19 6F4 11F17 15F12 58F109 53F34 0.8F1
=0.02 NS b0.01 =0.02 NS =0.01 =0.001 NS NS NS
See explanation on text.
expressed as the correlation coefficient (r). Statistical significance was assumed with a level of pb0.05.
3. Results The baseline characteristics and time domain indexes of HRV for all patients are detailed in Table 1. The mean duration of the follow-up was 1585F156 days. Twelve patients died (22.6%), with an annual mortality rate of 5.13%. The causes of death were progressive heart failure with multiorgan dysfunction in 7 patients and sudden cardiac death in 5. Seven patients who died were classified in NYHA class III, and 5 in class IV. The survivors evidenced a higher LVEF (31.90% versus 22.42%, p=0.001). Twelve patients experienced episodes of sustained ventricular tachycardia, and six of them died in the following period. Arrhythmic patients were treated by Amiodarone, and an ICD was implanted in 3 subjects. The analysis of time domain HRV showed significantly reduced values of SDNN, SDNN index, and SDANN in nonsurvivors. Also RMSSD showed reduction, but without reaching the statistical significance, probably because of the large variability of its values (Table 2). The analysis of frequency domain parameters, obtained at rest, evidenced a Table 3 Time and frequency domain heart rate variability at rest in patients with or without episodes of sustained ventricular tachycardia Patients (n) SDNN (ms) RMSSD (ms) SDNN index (ms) SDANN (ms) pNN50 (%) LF (ms2) LF (nu) HF (ms2) HF (nu) L/H
Sustained VT
No sustained VT
p
12 61F25 32F11 28F11 63F19 7F6 18F44 29F29 56F87 48F35 1.19F2
41 1122F38 91F69 65F30 94F28 10F7 82F95 36F23 104F89 49F23 1.51F3
b0.0001 =0.005 =0.0001 b0.001 NS b0.05 NS NS NS NS
See explanation on text.
249
Table 4 Modifications of frequency domain heart rate variability during rhythmic breathing and tilt in survivors and nonsurvivors
Survivors LF (ms2) LF (nu) HF (ms2) HF (nu) L/H
Baseline
Tilt
p*
Paced breathing
p**
84F96 40F24 103F83 47F23 1.62F3
75F70 49F26 42F30 30F16 2.51F3
NS NS b0.00001 b0.00001 NS
53F61 20F18 177F131 71F16 0.49F1
NS b0.00002 b0.005 =0.000001 b0.05
34F41 26F22 54F104 38F30 2.14F2
NS NS NS NS =0.01
13F24 18F25 106F75 61F27 0.87F1
NS NS NS NS NS
Nonsurvivors LF (ms2) 11F17 LF (nu) 15F12 58F109 HF (ms2) HF (nu) 53F34 L/H 0.80F1
*p value comparing baseline versus tilt; **p value comparing baseline versus paced breathing.
significant lower value of LF, both in normalized and absolute units in patients who died (Table 2). In the 12 patients with episodes of sustained ventricular tachycardia all but one time domain indexes of HRV were markedly depressed in comparison of patients without severe ventricular arrhythmias. In arrhythmic patients also the LF component at rest was reduced, but only if considered in absolute units (Table 3). Considering the autonomic modification induced by the provocative maneuvers, survivors had a significant reduction of LF in normalized units, and a significant increase of HF, both in absolute and normalized units, during paced breathing, while tilting induced only a significant reduction of the HF component, both in absolute and normalized units. The L/H ratio changed significantly in this group only with paced breathing, confirming a shift of the sympathovagal balance toward an increase of parasympathetic activity. On the contrary, patients who died did not show Table 5 Modifications of frequency domain heart rate variability during rhythmic breathing and tilt in patients with or without episodes of sustained ventricular tachycardia Baseline
Tilt
p*
Paced breathing
p**
Sustained VT LF (ms2) 18F44 LF (nu) 29F29 HF (ms2) 56F87 HF (nu) 48F35 L/H 1.19F2
61F35 52F30 27F18 23F16 3.65F3
=0.01 b0.0005 NS =0.01 NS
6F14 8F16 114F101 74F16 0.17F0.4
NS =0.01 b0.005 =0.01 NS
No sustained VT LF (ms2) 82F95 LF (nu) 36F23 HF (ms2) 104F89 HF (nu) 49F23 L/H 1.51F3
67F74 41F26 50F61 34F20 2.07F2
NS NS =0.00002 b0.00005 NS
55F60 23F19 175F127 67F20 0.69F1
NS b0.005 =0.005 b0.0001 NS
*p value comparing baseline versus tilt; **p value comparing baseline versus paced breathing.
250
A.F. Folino et al. / International Journal of Cardiology 100 (2005) 247–251
significant modifications of both indexes, neither during paced breathing nor during tilt, in spite of a significant modification of L/H induced by tilt reflecting an increase of sympathetic activity (Table 4). Considering patients with or without episodes of sustained ventricular tachycardia, the same maneuvers, did not provide useful results to identify patients at higher risk for severe ventricular arrhythmias (Table 5).
4. Discussion Autonomic nervous system plays a crucial role in the mechanisms of heart failure, being involved from the early stages of the disease and acting on different cardiovascular structures. The characteristic increase of sympathetic activity, as well as the blunted vagal tone, has their main clinical relevance on the electrical instability and therefore on mortality. The more reliable non-invasive method to assess autonomic activity is the study of HRV. It provides useful information on parasympathetic and sympathetic tone, and if depressed is a powerful predictor of mortality in patients after acute myocardial infarction. However, besides the evaluation of baseline autonomic activity, in the last years, other tests demonstrated an even stronger correlation with mortality, through the analysis of autonomic activity modifications induced by different stimuli. One of the most important is the study of baroreflex sensitivity, where a pharmacological induced increase of sympathetic activity provides information about the reflex vagal response [10]. Similarly, analysis of RR intervals changes after a premature ventricular beat, known as heart rate turbulence, assess the modifications of autonomic activity induced by an endogenous stimulus [16]. More recently, a reduced short-term low-frequency power during controlled breathing was demonstrated, a powerful predictor of sudden death in patients with chronic heart failure [17]. In our study, we investigated the modifications of autonomic activity in patients with heart failure, induced by means of simple maneuvers such as tilting and paced breathing, and its correlations with mortality. Baseline evaluation of time domain HRV analysis showed a global reduction of all indexes in patients who died, that reached the statistical significance for SDNN, SDNN index, and SDANN. Similarly, the study of baseline frequency domain analysis evidenced that the most important characteristic of the patients who died was a significantly lower LF components. As reported in previous studies, patients with heart failure present different patterns of heart rate variability [18–20], often based on the severity of their clinical condition. However, in heart failure, the interpretation of spectral components of HRV is very difficult, and a specific correlation with parasympathetic or sympathetic activity, cannot be made. The association of reduced LF components and a worse prognosis in patients with heart failure is, however, similar
to that reported in previous studies, in which HRV was evaluated in basal conditions [18,21,22]. Even if a conventional classification of spectral components is compromised, their variation under different stimuli can be evaluated, and the provocative maneuvers adopted in our protocol seem to add crucial information about the autonomic activity in failing hearts. In comparison with the study of La Rovere et al. [17] we explored also autonomic modulation induced by passive orthostatism, obtained by mean of a tilting table, as similarly tested by Guzzetti [18], and the results obtained provided useful information to stratify the risk in our cohort of patients. The LF component showed, during tilt, a minimal, nonsignificant increase in both survivors and nonsurvivors. Nevertheless, this index decreased significantly in survivors during the paced breathing, whereas remained almost unchanged in nonsurvivors. The modifications of HF component were more distinct: it remains unchanged either during tilting or during paced breathing in patients who died, whereas survivors evidenced a reduction during tilt, and an increase during paced breathing. It seems evident that survivors maintain some degree of responsiveness of autonomic modulation, particularly, generating signals included in the HF band. Also, the LF components showed a residual responsiveness induced by paced breathing. On the contrary, in patients who died, these mechanisms were unable to react to the two maneuvers tested. The analysis of HRV in patients with or without episodes of sustained ventricular tachycardia provided discouraging results. However, it is noteworthy that time domain indexes provided stronger correlations between arrhythmic and nonarrhythmic patients, than between survivors and nonsurvivors. On the other hand, the modulating maneuvers associated with HRV analysis were not useful to stratify the arrhythmic risk, providing similar results in patients with or without sustained ventricular tachycardia. In conclusion, our findings seem to indicate that patients with heart failure and a better prognosis are characterized by the presence of still active mechanisms of autonomic modulation. On the contrary, patients who died demonstrated a rigidity of autonomic nervous system, unable to react to some evocative stimuli. Therefore, simple maneuvers, as tilting and paced breathing, associated with HRV analysis, seem to provide more useful information, than simple baseline evaluation of autonomic status, in identifying patients with heart failure at higher risk. Time domain indexes of HRV seem more effective than frequency domain parameters, at rest and during evocative maneuvers, in arrhythmic risk stratification. References [1] Kleiger RE, MIller JP, Bigger JT, Moss AJ, Multicenter Postinfarction Research Group. Decreased heart rate variability and its association with increased mortality after acute myocardial infarction. Am J Cardiol 1987;59:256 – 62.
A.F. Folino et al. / International Journal of Cardiology 100 (2005) 247–251 [2] Bigger JT, Fleiss JM, Steinman NC, Rolnitzkky LM, Kleiger RE, Rottman JN. Frequency domain measures of heart period variability and mortality after myocardial infarction. Circulation 1992;85:164 – 71. [3] Counihan PJ, Fei L, Bashir Y, Farrel TG, Haywood GA, McKenna WJ. Assessment of heart rate variability in hypertrophic cardiomyopathy. Association with clinical and prognostic features. Circulation 1993;88:1682 – 90. [4] Fei L, Keeling J, Gill JS, Bashir Y, Statters DJ, Poloniecki J, et al. Heart rate variability and its relations to ventricular arrhythmias in congestive heart failure. Br Heart J 1994;71:322 – 8. [5] 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. [6] Kingwell BA, Thompson JM, Kaye DM, McPherson GA, Jennings MD, Esler MD. Heart rate spectral analysis, cardiac norepinephrine spillover, and muscle sympathetic nerve activity during human sympathetic nervous activation and failure. Circulation 1994;90: 234 – 40. [7] Tygesen H, Rundqvist B, Waagstein F, Wennerblom B. Heart rate variability measurement correlates with cardiac norepinephrine spillover in congestive heart failure. Am J Cardiol 2001;87:1308 – 11. [8] Nolan J, Batin PD, Andrews R, Lindsay SJ, Brooksby P, Mullen M, et al. Prospective study of heart rate variability and mortality in chronic heart failure. Results of the United Kingdom heart failure evaluation and assessment of risk trial. Circulation 1998;98:1510 – 6. [9] Bilchick KC, Fetics B, Djoukeng R, Gross Fisher S, Fletcher RD, Singh SN, et al. Prognostic value of heart rate variability in chronic congestive heart failure (Veteran affairs’ survival trial of antiarrhythmic therapy in congestive heart failure). Am J Cardiol 2002;90:24 – 8. [10] La Rovere MT, Bigger JT, Marcus FI, Mortara A, Schwartz PJ, for the ATRAMI. Baroreflex sensitivity and heart rate variability in predictor of total cardiac mortality after myocardial infarction. Lancet 1998;351:478 – 84. [11] Schmidt G, Malik M, Barthel P, Schneider R, Ulm K, Rolnitzky L, et al. Heart rate turbulence after ventricular premature beats as a predictor of mortality after acute myocardial infarction. Lancet 1999; 353:1390 – 6. [12] Malliani A, Pagani M, Lombardi F, Cerutti S. Cardiovascular neural regulation explored in the frequency domain. Circulation 1991;84: 482 – 92.
251
[13] Rimoldi O, Pagani MR, Piazza S, Pagani M, Malliani A. Restraining effects of captopril on sympathetic excitatory responses in dogs: a spectral analysis approach. Am J Physiol 1994;267:1608 – 18. [14] 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. [15] Task force of the European Society of Cardiology, and the North American Society of Pacing and Electrophysiology. Heart rate variability. Standard of measurement, physiological interpretation, and clinical use. Circulation 1996;93:1043 – 65. [16] Schmidt G, Malik M, Barthel P, Schneider R, Ulm K, Rolnitzky L, et al. Heart-rate turbulence after ventricular premature beats as predictor of mortality after acute myocardial infarction. Lancet 1999; 353:1390 – 9. [17] La Rovere MT, Pinna GD, Maestri R, Mortara A, Capomolla S, Febo O, et al. Short-term heart rate variability strongly predicts sudden cardiac death in chronic heart failure patients. Circulation 2003;107:565 – 70. [18] Guzzetti S, Cogliati C, Turiel 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. [19] Van De Borne P, Montano N, Pagani M, Oren RM, Somers VK. Absence of low frequency variability of sympathetic nerve activity in severe heart failure. Circulation 1997;95:1449 – 14455. [20] Notarius CF, Butler GC, Ando S, Pollard MJ, Senn BL, Floras JS. Dissociation between microneurographic and heart rate variability estimates of sympathetic tone in normal subjects and in patients with heart failure. Clin Sci 1999;96:557 – 656. [21] Mortara A, La Rovere MT, Signorini MG, Pantaleo P, Pinna G, Martinelli L, et al. Can power spectral analysis of heart rate variability identify a high risk subgroup of congestive heart failure patients with excessive sympathetic activation? A pilot study before and after heart transplantation. Br Heart J 1994;71:422 – 30. [22] Galinier M, Pathak A, Fourcade J, Androdias C, Curnier D, Varnous S, et al. Depressed low-frequency power of heart rate variability as an independent predictor of sudden death in chronic heart failure. Eur Heart J 2000;21:475 – 82.